Handbook of

Olfaction and
Gustation
Second Edition
Revised and Expanded

edited by
Richard L. Doty
University of Pennsylvania
Philadelphia, Pennsylvania, U.S.A.

MARCEL DEKKER, INC. NEW YORK • BASEL

Library of Congress Cataloging-in-Publication Data
A catalog record for this book is available from the Library of Congress.

ISBN: 0-8247-0719-2

This book is printed on acid-free paper.

Headquarters
Marcel Dekker, Inc.
270 Madison Avenue, New York, NY 10016
tel: 212-696-9000; fax: 212-685-4540

Eastern Hemisphere Distribution

Marcel Dekker AG
Hutgasse 4, Postfach 812, CH-4001 Basel, Switzerland
tel: 41-61-260-6300; fax: 41-61-260-6333

World Wide Web

http://www.dekker.com

The publisher offers discounts on this book when ordered in bulk quantities. For more information, write to Special Sales/Professional
Marketing at the headquarters address above.

Copyright © 2003 by Marcel Dekker, Inc. All Rights Reserved.

Neither this book nor any part may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photo-
copying, microfilming, and recording, or by any information storage and retrieval system, without permission in writing from the
publisher.

Current printing (last digit):

10 9 8 7 6 5 4 3 2 1

PRINTED IN THE UNITED STATES OF AMERICA

 NEUROLOGICAL DISEASE AND THERAPY

Advisory Board

Louis R. Caplan, M.D. William C. Koller, M.D.

Professor of Neurology Mount Sinai School of Medicine
Harvard University School of Medicine New York, New York
Beth Israel Deaconess Medical Center
Boston, Massachusetts

John C. Morris, M.D. Bruce Ransom, M.D., Ph.D.

Fnedman Professor of Neurology Warren Magnuson Professor
Co-Director, Alzheimer's Disease Research Center Chair, Department of Neurology
Washington University School of Medicine University of Washington School of Medicine
St Louis, Missouri Seattle, Washington

Kapil Sethi, M.D. Mark Tuszynski, M.D., Ph.D.

Professor of Neurology Associate Professor of Neurosciences
Director, Movement Disorders Program Director, Center for Neural Repair
Medical College of Georgia University of California-San Diego
Augusta, Georgia La Jolla, California

1. Handbook of Parkinson's Disease, edited by William C. Koller

2. Medical Therapy of Acute Stroke, edited by Mark Fisher
3 Familial Alzheimer's Disease Molecular Genetics and Clinical Perspectives, edited by Gary D. Miner, Ralph W.
Richter, John P 8/ass, Jimmie L. Valentine, and Linda A. Winters-Miner
4 Alzheimer's Disease. Treatment and Long-Term Management, edited by Jeffrey L. Cummmgs and Bruce L Miller
5. Therapy of Parkinson's Disease, edited by William C Koller and George Paulson
6. Handbook of Sleep Disorders, edited by Michael J Thorpy
1 Epilepsy and Sudden Death, edited by Claire M. Lathers and Paul L Schraeder
8. Handbook of Multiple Sclerosis, edited by Stuart D Cook
9. Memory Disorders Research and Clinical Practice, edited by Takehiko Yanagihara and Ronald C. Petersen
10. The Medical Treatment of Epilepsy, edited by Stanley R. Resor, Jr, and Henn Kuti
11. Cognitive Disorders Pathophysiology and Treatment, edited by Leon J Thai, Walter H Moos, and Elkan R Gamzu
12. Handbook of Amyotrophic Lateral Sclerosis, edited by Richard Alan Smith
13. Handbook of Parkinson's Disease' Second Edition, Revised and Expanded, edited by William C Koller
14 Handbook of Pediatric Epilepsy, edited by Jerome V Murphy and Fereydoun Dehkhargham
15 Handbook of Tourette's Syndrome and Related Tic and Behavioral Disorders, edited by Roger Kurlan
16 Handbook of Cerebellar Diseases, edited by Richard Lechtenberg
17 Handbook of Cerebrovascular Diseases, edited by Harold P. Adams, Jr.
18 Parkmsonian Syndromes, edited by Matthew B Stem and William C Koller
19 Handbook of Head and Spine Trauma, edited by Jonathan Greenberg
20. Brain Tumors' A Comprehensive Text, edited by Robert A Morantz and John W. Walsh
21. Monoamine Oxidase Inhibitors in Neurological Diseases, edited by Abraham Lieberman, C. Warren Olanow,
Moussa B H Youdim, and Keith Tipton
22 Handbook of Dementing Illnesses, edited by John C Moms
23. Handbook of Myasthema Gravis and Myasthenic Syndromes, edited by Robert P Lisak
24. Handbook of Neurorehabilitation, edited by David C Good and James R Couch, Jr.
25. Therapy with Botulinum Toxin, edited by Joseph Jankovic and Mark Hallett
26. Principles of Neurotoxicology, edited by Louis W. Chang
27 Handbook of Neurovirology, edited by Robert R. McKendall and William G Stroop
28 Handbook of Neuro-Urology, edited by David N Rushton
29. Handbook of Neuroepidemiology, edited by Philip B. Gorelick and Milton Alter
30 Handbook of Tremor Disorders, edited by Leslie J Findley and William C. Koller
31. Neuro-Ophthalmological Disorders. Diagnostic Work-Up and Management, edited by Ronald J Tusa and Steven A.
Newman
32. Handbook of Olfaction and Gustation, edited by Richard L Doty
33. Handbook of Neurological Speech and Language Disorders, edited by Howard S. Kirshner
34. Therapy of Parkinson's Disease Second Edition, Revised and Expanded, edited by William C. Kollerand George
Paulson
35. Evaluation and Management of Gait Disorders, edited by Barney S. Spivack
36. Handbook of Neurotoxicology, edited by Louis W. Chang and Robert S. Dyer
37 Neurological Complications of Cancer, edited by Ronald G. Wiley
38. Handbook of Autonomic Nervous System Dysfunction, edited by Amos D. Korczyn
39. Handbook of Dystonia, edited by Joseph King Ching Tsui and Donald B. Calne
40. Etiology of Parkinson's Disease, edited by Jonas H. Ellenberg, William C. Koller, andJ. William Langston
41. Practical Neurology of the Elderly, edited by Jacob I. Sage and Margery H. Mark
42. Handbook of Muscle Disease, edited by Russell J. M. Lane
43. Handbook of Multiple Sclerosis: Second Edition, Revised and Expanded, edited by Stuart D. Cook
44. Central Nervous System Infectious Diseases and Therapy, edited by Karen L. Roos
45. Subarachnoid Hemorrhage: Clinical Management, edited by Takehiko Yanagihara, David G. Piepgras, and John
L. D. Atkmson
46. Neurology Practice Guidelines, edited by Richard Lechtenberg and Henry S. Schutta
47. Spinal Cord Diseases: Diagnosis and Treatment, edited by Gordon L Engler, Jonathan Cole, and W. Louis
Merlon
48. Management of Acute Stroke, edited by Ashfaq Shuaib and Larry B. Goldstem
49. Sleep Disorders and Neurological Disease, edited by Antonio Culebras
50. Handbook of Ataxia Disorders, edited by Thomas Klockgether
51. The Autonomic Nervous System in Health and Disease, David S. Goldstein
52. Axonal Regeneration in the Central Nervous System, edited by Nicholas A. Ingoglia and Marion Murray
53 Handbook of Multiple Sclerosis. Third Edition, edited by Stuart D Cook
54. Long-Term Effects of Stroke, edited by Julien Bogousslavsky
55. Handbook of the Autonomic Nervous System in Health and Disease, edited by C. Liana Bolts, Julio Licinio, and
Stefano Govoni
56. Dopamme Receptors and Transporters: Function, Imaging, and Clinical Implication, Second Edition, edited by
Anita Sidhu, Marc Laruelle, and Philippe Vernier
57. Handbook of Olfaction and Gustation: Second Edition, Revised and Expanded, edited by Richard L. Doty
58. Handbook of Stereotactic and Functional Neurosurgery, edited by Michael Schulder
59. Handbook of Parkinson's Disease: Third Edition, edited by Rajesh Pahwa, Kelly E. Lyons, and William C. Koller

Additional Volumes in Preparation

Clinical Neurovirology, edited byAvi Nath and Joseph R. Berger

Foreword

In the rise of modern neuroscience during the last century, of the brain. These systems are therefore on the cutting
the great sensory systems—vision above all, but also the edge of current research on stem cells and neurogenesis in
somatosensory systems and audition—played the leading the brain.
roles in the elucidation of principles underlying the neural New psychophysical studies challenge the traditional
mechanisms of perception. Work on the senses of taste and view of human olfaction as weak, and suggest instead that
smell lagged behind, hampered by the difficulties of con- our ability to perceive low levels of odorants may be as
trolling the stimuli in precise ways and by the belief that good or better than that of macrosmats such as rodents and
these senses were of minor importance to humans. carnivores. Moreover, such studies have expanded our un-
No more! That era of chemosensory darkness ended derstanding of the complexity of the chemical senses, and,
during the past two decades because of illumination from along with a plethora of basic science studies, have demon-
new studies at many levels of these systems. Gene fami- strated that these senses are intimately involved in a wide
lies that express receptors for chemical stimuli have been range of medical disorders. Indeed, the olfactory system
identified. Membrane mechanisms of stimulus transduc- may provide early indicators for disease states such as neu-
tion and second messenger signaling have been revealed. rodegeneration and schizophrenia. We now have a better
Topographic patterns of the convergence of axons from understanding of the significance of the olfactory and gus-
the sensory cells onto higher levels in the sensory path- tatory systems for such critical human behaviors as infant
ways have been mapped. A variety of methods have re- nutrition, the prevalence of obesity in developed countries,
vealed that different odors elicit different activity patterns, and the strong links between the chemical senses and emo-
which constitute virtual “odor images.” In the olfactory tion and memory.
system, as in the other great sensory systems, stimulus All of these developments and many more are covered
space (in this case, the multidimensional space of odor in the second edition of this widely recognized book, the
molecules) has now been mapped into two-dimensional largest compendium of data on the chemical senses pub-
neural space. lished to date. Richard Doty’s introduction provides a mas-
The synaptic microcircuits in the olfactory bulb have terly overview of the rapidly evolving events in these
attracted a new generation of electophysiologists from fields, and the ensuing chapters provide a wealth of infor-
other fields. Using patch recordings, calcium imaging and mation on topics ranging from basic anatomy, physiology,
advanced microscopy, they are analyzing the dendritic and clinical disorders of the chemical senses to advances
and synaptic properties of the microcircuits that process in functional imaging, molecular neurobiology, human
the odor images as the basis for perception. In addition to and animal psychophysics, and even olfactory system
this revitalization of electrophysiological studies of the cybernetics.
chemical senses, the neuroscience community has been As the fields of gustation and olfaction continue their
attracted to the extraordinary plasticity of these systems, strong growth, there will be an increasing need for a source
evidenced, in part, by the ongoing turnover of taste and to which one can go for orientation to the broad range of re-
olfactory cells, and the constant generation of new in- search involved and critical assessments of progress, prob-
terneurons from the anterior migratory stream at the base lems, and opportunities. This new edition fills those needs

iii
iv Foreword

superlatively for a wide range of readers: neuroscientists,

organic chemists, toxicologists, biomedical engineers, psy-
chologists, and a variety of clinicians, as well as the inter-
ested layperson.

Gordon M. Shepherd, M.D., D. Phil.

Professor of Neuroscience
Yale University
New Haven, Connecticut, U.S.A.
Preface

Since the publication of the first edition of the Handbook of postdoctoral fellows from numerous disciplines. The chap-
Olfaction and Gustation in 1995, advances in chemosen- ters are conveniently arranged into three major sections
sory science have been staggering. Indeed, during this pe- corresponding to olfaction, gustation, and other chemosen-
riod the chemical senses have become a central element of sory systems and, with the exception of the last section, are
the field of modern neuroscience, largely because of their subdivided into (A) anatomy and neurobiology, (B) func-
regenerative capacities, integral association with stem cell tional measurement, ontogeny, and genetics, and (C) clini-
research, and unique transduction processes. As a conse- cal applications and perspectives. Unlike the first edition,
quence of the proliferation of commercially available ol- this edition contains an author index that makes it possible
factory tests, olfaction is now routinely and quantitatively for researchers and others to quickly find references and
evaluated in most major medical centers, as well as within sections based on individual contributions. As in the first
the food, beverage, cosmetic, and energy (e.g., gas works) edition, historical perspective and clinical relevance have
industries. Of particular relevance to the physician is the been emphasized, but not at the expense of basic science.
fact that decreased smell function is likely the first clinical The book has been expanded from 38 to 48 chapters, so as
manifestation of Alzheimer’s disease (AD) and idiopathic to take into account major growth in a number of fields, in-
Parkinson’s disease (PD). Indeed, accurate assessment of cluding neuroscience, functional imaging, cybernetics,
olfaction can aid in the “preclinical” identification of indi- toxicology, structure–activity assessment, molecular biol-
viduals at risk for these disorders. Such assessment can also ogy, and animal behavior.
aid in differential diagnosis, since diseases often misdiag- I am grateful to the contributors, who have been a model
nosed as AD or PD (e.g., major affective disorder, progres- of objectivity and scholarship in the development of their
sive supranuclear palsy) are unaccompanied by meaningful chapters, and who have graciously taken into consideration
olfactory loss. my often extensive editorial suggestions and criticisms. I
The second edition of the Handbook represents the am also indebted to the staff of Marcel Dekker, Inc., par-
largest collection of basic, clinical, and applied knowledge ticularly Jinnie Kim, Assistant Acquisitions Editor, and
on the chemical senses ever compiled in one volume, with Ann Pulido, Production Editor, who have patiently and
contributions from over 80 of the world’s leading re- painstakingly worked with me to ensure a volume of the
searchers. The material in this up-to-date treatise has been highest quality. Without the support of the following grants
tailored to be of value to a wide range of medical special- from the National Institutes of Health, this work would
ists, as well as to basic scientists working in academics, in- have never been accomplished: PO1 DC 00161, RO1 DC
dustry, and government. Because the information is pre- 04278, RO1 DC 02974, and RO1 AG 27496.
sented in a straightforward manner, this volume can serve
as a textbook for graduate students, medical students, and Richard L. Doty

v
Contents

Foreword Gordon M. Shepherd iii

Preface Richard L. Doty v
Contributors xi
Introduction and Historical Perspective Richard L. Doty xv

I. OLFACTION

A. Anatomy and Neurobiology

1. Anatomy of the Human Nasal Passages 1

Dean M. Clerico, Wyatt C. To, and Donald C. Lanza

2. Morphology of the Mammalian Olfactory Epithelium: Form, Fine Structure, Function, and Pathology 17
Bert Ph. M. Menco and Edward E. Morrison

3. Olfactory Mucosa: Composition, Enzymatic Localization, and Metabolism 51

Xinxin Ding and Alan R. Dahl

4. Molecular Neurobiology of Olfactory Transduction 75

Cheil Moon and Gabriele V. Ronnett

5. Neurogenesis in the Adult Olfactory Neuroepithelium 93

Alan Mackay-Sim

6. Developmental Anatomy of the Olfactory System 115

Meng Inn Chuah, James E. Schwob, and Albert I. Farbman

7. Anatomy and Neurochemistry of the Olfactory Bulb 139

Igor L. Kratskin and Ottorino Belluzzi

8. Central Olfactory Structures 165

Thomas A. Cleland and Christiane Linster

9. Sensory Physiology of Central Olfactory Pathways 181

Donald A. Wilson and Regina M. Sullivan

vii
viii Contents

B. Functional Measurement, Ontogeny, and Genetics

10. Psychophysical Measurement of Human Olfactory Function, Including Odorant Mixture Assessment 203
Richard L. Doty and David G. Laing

11. Electrophysiological Measurement of Olfactory Function 229

Gerd Kobal

12. Functional Neuroimaging of Human Olfaction 251

Noam Sobel, Bradley N. Johnson, Joel Mainland, and David M. Yousem

13. Structure–Odor Relationships: A Modern Perspective 275

Luca Turin and Fumiko Yoshii

14. Olfactory System Cybernetics: Artificial Noses 295

Krishna C. Persaud

15. Olfaction and the Development of Social Behavior in Neonatal Mammals 309
Richard H. Porter and Benoist Schaal

16. Genetics of Olfactory Perception 329

Nancy L. Segal and Tari D. Topolski

17. Mammalian Pheromones: Fact or Fantasy? 345

Richard L. Doty

18. Psychophysical Evaluation of Olfaction in Nonhuman Mammals 385

Lloyd Hastings

19. Methods for Determining Odor Preferences in Nonhuman Mammals 403

Richard L. Doty

20. Olfactory Memory 409

Aras Petrulis and Howard Eichenbaum

C. Clinical Applications and Perspectives

21. Nasal Patency and the Aerodynamics of Nasal Airflow: Measurement by Rhinomanometry and Acoustic
Rhinometry, and the Influence of Pharmacological Agents 439
Richard E. Frye

22. Clinical Disorders of Olfaction 461

Claire Murphy, Richard L. Doty, and Heather J. Duncan

23. Odor Perception in Neurodegenerative Diseases 479

Richard L. Doty

24. Olfactory System Neuropathology in Alzheimer’s Disease, Parkinson’s Disease, and Schizophrenia 503
Gregory S. Smutzer, Richard L. Doty, Steven E. Arnold, and John Q. Trojanowski
Contents ix

25. Multiple Chemical Intolerance 525

Claudia S. Miller

26. The Olfactory System and the Nasal Mucosa as Portals of Entry of Viruses, Drugs, and Other Exogenous
Agents into the Brain 549
Harriet Baker and Mary Beth Genter

27. Influence of Environmental Toxicants on Olfactory Function 575

Lloyd Hastings and Marian L. Miller

28. Evaluation of Olfactory Deficits by Structural Medical Imaging 593

Cheng Li, Richard L. Doty, David W. Kennedy, and David M. Yousem

29. Plasticity Within the Olfactory Pathways: Influences of Trauma, Deprivation, Stem Cells,
and Other Factors 615
Joel Maruniak

30. Head Injury and Olfaction 629

Richard M. Costanzo, Laurence J. DiNardo, and Evan R. Reiter

II. GUSTATION

A. Anatomy and Neurobiology

31. Saliva: Its Role in Taste Function 639

Robert M. Bradley and Lloyd M. Beidler

32. Morphology of the Peripheral Taste System 651

Martin Witt, Klaus Reutter, and Inglis J. Miller, Jr.

33. Central Taste Anatomy and Neurophysiology 679

Edmund T. Rolls and Thomas R. Scott

34. Molecular Physiology of Gustatory Transduction 707

Timothy A. Gilbertson and Robert F. Margolskee

35. Gustatory Neural Coding 731

David V. Smith and Thomas R. Scott

36. Development of the Taste System: Basic Neurobiology 759

Charlotte M. Mistretta and David L. Hill

B. Functional Measurement, Ontogeny, and Genetics

37. Contemporary Measurement of Human Gustatory Function 783

Marion E. Frank, Thomas P. Hettinger, Michael A. Barry, Janneane F. Gent, and Richard L. Doty
x Contents

38. Human Perception of Taste Mixtures 805

Hendrik N. J. Schifferstein

39. The Ontogeny of Human Flavor Perception 823

Judith R. Ganchrow and Julie A. Mennella

40. Genetics of Human Taste Perception 847

Adam Drewnowski

41. Psychophysical Evaluation of Taste Function in Nonhuman Mammals 861

Alan C. Spector

C. Clinical Applications and Perspectives

42. Nutritional Implications of Taste and Smell 881

Richard D. Mattes

43. Conditioned Taste Aversions 905

Kathleen C. Chambers and Ilene L. Bernstein

44. Clinical Disorders Affecting Taste: Evaluation and Management 935

Steven M. Bromley and Richard L. Doty

45. Head Injury and Taste 959

Richard M. Costanzo, Laurence J. DiNardo, and Evan R. Reiter

III. OTHER CHEMOSENSORY SYSTEMS

46. The Vomeronasal Organ 967

Peter A. Brennan and Eric B. Keverne

47. Trigeminal Chemosensation 981

Richard L. Doty and J. Enrique Cometto-Muňiz

48. The Structure and Function of the Nervus Terminalis 1001

Marlene Schwanzel-Fukuda and Donald W. Pfaff

Author Index 1027

Subject Index 1103
Contributors

Steven E. Arnold, M.D. Smell and Taste Center and Depart- Meng Inn Chuah, Ph.D. Department of Anatomy and Physiol-
ments of Psychiatry and Neurology, University of Pennsylvania, ogy, University of Tasmania, Hobart, Australia
Philadelphia, Pennsylvania, U.S.A.
Thomas A. Cleland, Ph.D. Department of Neurobiology and
Harriet Baker, Ph.D. Department of Neurology and Neuro- Behavior, Cornell University, Ithaca, New York, U.S.A.
science, The Burke Medical Research Institute, Weill Medical
College, Cornell University, White Plains, New York, U.S.A. Dean M. Clerico, M.D. Valley ENT, Forty Fort, Pennsylvania,
U.S.A.
Michael A. Barry, Ph.D. Division of Neurosciences, Depart-
ment of Oral Diagnosis, School of Dental Medicine, University of J. Enrique Cometto-Muñiz, Ph.D. Chemosensory Perception
Connecticut Health Center, Farmington, Connecticut, U.S.A. Laboratory, Department of Surgery (Otolaryngology), University
of California, San Diego, La Jolla, California, U.S.A.
Lloyd M. Beidler, Ph.D. Department of Biological Science,
Florida State University, Tallahassee, Florida, U.S.A.
Richard M. Costanzo, Ph.D. Department of Physiology, Vir-
ginia Commonwealth University, Richmond, Virginia, U.S.A.
Ottorino Belluzzi, Ph.D. Department of Biology, University of
Ferrara, Ferrara, Italy
Alan R. Dahl, Ph.D. Battelle Memorial Institute, Columbus,
Ilene L. Bernstein, Ph.D. Department of Psychology, Univer- Ohio, U.S.A.
sity of Washington, Seattle, Washington, U.S.A.
Laurence J. DiNardo, M.D. Department of Otolaryngology
Robert M. Bradley, M.D.S., Ph.D. Department of Biologic Head and Neck Surgery, Virginia Commonwealth University,
and Materials Science, School of Dentistry, University of Michi- Richmond, Virginia, U.S.A.
gan, Ann Arbor, Michigan, U.S.A.
Xinxin Ding, Ph.D. Wadsworth Center, New York State De-
Peter A. Brennan, Ph.D. Department of Animal Behaviour, partment of Health, and School of Public Health, State University
University of Cambridge, Cambridge, United Kingdom of New York at Albany, Albany, New York, U.S.A.

Steven M. Bromley, M.D. Smell and Taste Center, University Richard L. Doty, Ph.D. Smell and Taste Center and Depart-
of Pennsylvania, and Department of Neurology, Thomas Jeffer- ment of Otorhinolaryngology: Head and Neck Surgery, Univer-
son University, Philadelphia, Pennsylvania, U.S.A. sity of Pennsylvania, Philadelphia, Pennsylvania, U.S.A.

Kathleen C. Chambers, Ph.D. Department of Psychology, Adam Drewnowski, Ph.D. Nutritional Sciences Program,
University of Southern California, Los Angeles, California, School of Public Health and Community Medicine, University of
U.S.A. Washington, Seattle, Washington, U.S.A.

xi
xii Contributors

Heather J. Duncan, Ph.D. Department of Internal Medicine, Gerd Kobal, M.D., Ph.D. Department of Pharmacology and
University of Cincinnati College of Medicine, Cincinnati, Ohio, Toxicology, University of Erlangen, Erlangen, Germany
U.S.A.
Igor L. Kratskin, M.D., Ph.D. Smell and Taste Center and De-
Howard Eichenbaum, Ph.D. Department of Psychology, partment of Otorhinolaryngology: Head and Neck Surgery, Uni-
Boston University, Boston, Massachusetts, U.S.A. versity of Pennsylvania, Philadelphia, Pennsylvania, U.S.A.

Albert I. Farbman, D.M.D., Ph.D. Department of Neurobiol- David G. Laing, Ph.D. Centre for Advanced Food Research,
ogy and Physiology, Northwestern University, Evanston, Illinois, University of Western Sydney, Sydney, Australia
U.S.A.
Donald C. Lanza, M.D. Department of Otolaryngology and
Communicative Disorders, The Cleveland Clinic Foundation,
Marion E. Frank, Ph.D. Division of Neurosciences, Depart-
Cleveland, Ohio, U.S.A.
ment of Oral Diagnosis, School of Dental Medicine, University of
Connecticut Health Center, Farmington, Connecticut, U.S.A.
Cheng Li, M.D. Smell and Taste Center, and Department of
Otorhinolaryngology, Head and Neck Surgery, University of
Richard E. Frye, M.D., Ph.D. Department of Neurology, Chil-
Pennsylvania, Philadelphia, Pennsylvania, U.S.A.
dren’s Hospital, Boston, Massachusetts, U.S.A.
Christiane Linster, Ph.D. Department of Neurobiology and
Judith R. Ganchrow, Ph.D. Institute of Dental Sciences, The Behavior, Cornell University, Ithaca, New York, U.S.A.
Hebrew University–Hadassah School of Dental Medicine
Founded by the Alpha Omega Fraternity, Jerusalem, Israel Alan Mackay-Sim, Ph.D. Centre for Molecular Neurobiology,
Griffith University, Brisbane, Queensland, Australia
Janneane F. Gent, Ph.D. Department of Epidemiology and
Public Health, Yale University, New Haven, Connecticut, U.S.A. Joel Mainland, Ph.D. Wills Neuroscience Institute and De-
partment of Psychology, University of California, Berkeley,
Mary Beth Genter, Ph.D. Department of Environmental California, U.S.A.
Health, University of Cincinnati, Cincinnati, Ohio, U.S.A.
Robert F. Margolskee, M.D., Ph.D. Department of Physiol-
Timothy A. Gilbertson, Ph.D. Department of Biology, Utah ogy and Biophysics, Howard Hughes Medical Institute, The
State University, Logan, Utah, U.S.A. Mount Sinai School of Medicine, New York, New York, U.S.A.

Lloyd Hastings, Ph.D. Smell and Taste Center and Department Joel Maruniak, Ph.D. Department of Biological Sciences,
of Otorhinolaryngology: Head and Neck Surgery, University of University of Missouri, Columbia, Missouri, U.S.A.
Pennsylvania, Philadelphia, Pennsylvania, U.S.A.
Richard D. Mattes, Ph.D., R.D. Department of Foods and Nu-
trition, Purdue University, West Lafayette, Indiana, U.S.A.
Thomas P. Hettinger, Ph.D. Division of Neurosciences, De-
partment of Oral Diagnosis, School of Dental Medicine, Univer-
Bert Ph. M. Menco, Ph.D. Department of Neurobiology and
sity of Connecticut Health Center, Farmington, Connecticut,
Physiology, Northwestern University, Evanston, Illinois, U.S.A.
U.S.A.
Julie A. Mennella, Ph.D. Monell Chemical Senses Center,
David L. Hill, Ph.D. Department of Psychology, University of Philadelphia, Pennsylvania, U.S.A.
Virginia, Charlottesville, Virginia, U.S.A.
Claudia S. Miller, M.D. Department of Family Practice and
Bradley N. Johnson, M.D. Department of Bioengineering, Community Medicine, University of Texas Health Science Cen-
University of California, Berkeley, California, U.S.A. ter, San Antonio, Texas, U.S.A.

David W. Kennedy, M.D. Department of Otorhinolaryngol- Inglis J. Miller, Jr., Ph.D. Department of Neurobiology and
ogy: Head and Neck Surgery, University of Pennsylvania, Anatomy, Wake Forest University School of Medicine, Winston-
Philadelphia, Pennsylvania, U.S.A. Salem, North Carolina, U.S.A.

Eric B. Keverne, M.A., Ph.D., D.Sc., F.R.S. Department of Marian L. Miller, Ph.D. Department of Environmental Health,
Animal Behaviour, University of Cambridge, Cambridge, United University of Cincinnati College of Medicine, Cincinnati, Ohio,
Kingdom U.S.A.
Contributors xiii

Charlotte M. Mistretta, Ph.D. Department of Biological and James E. Schwob, M.D., Ph.D. Department of Anatomy and
Materials Sciences, School of Dentristry, University of Michigan, Cellular Biology, Tufts University School of Medicine, Boston,
Ann Arbor, Michigan, U.S.A. Massachusetts, U.S.A.

Cheil Moon, Ph.D. The Johns Hopkins University School of Thomas R. Scott, Ph.D. College of Sciences, San Diego State
Medicine, Baltimore, Maryland, U.S.A. University, San Diego, California, U.S.A.

Edward E. Morrison, Ph.D. Department of Anatomy, Physi- Nancy L. Segal, Ph.D. Department of Psychology, California
ology, and Pharmacology, Auburn University, Auburn, Alabama, State University, Fullerton, California, U.S.A.
U.S.A.
David V. Smith, Ph.D. Department of Anatomy and Neurobi-
Claire Murphy, Ph.D. Department of Psychology, San Diego ology, University of Tennessee Health Science Center, Memphis,
State University, and Department of Surgery (Otolaryngology), Tennessee, U.S.A.
University of California, San Diego, School of Medicine, San
Diego, California, U.S.A. Gregory S. Smutzer, Ph.D. Smell and Taste Center and De-
partment of Otorhinolaryngology, University of Pennsylvania,
Krishna C. Persaud, Ph.D. Department of Instrumentation Philadelphia, Pennsylvania, U.S.A.
and Analytical Science, University of Manchester Institute of Sci-
ence and Technology, Manchester, United Kingdom Noam Sobel, Ph.D. Wills Neuroscience Institute and Depart-
ment of Psychology, University of California, Berkeley, Califor-
Aras Petrulis, Ph.D. Department of Psychology, Georgia State nia, U.S.A.
University, Atlanta, Georgia, U.S.A.
Alan C. Spector, Ph.D. Department of Psychology, University
Donald W. Pfaff, Ph.D. Laboratory of Neurobiology and Be- of Florida, Gainesville, Florida, U.S.A.
havior, The Rockefeller University, New York, New York, U.S.A.
Regina M. Sullivan, Ph.D. Department of Zoology, University
Richard H. Porter, Ph.D. Laboratoire de Comportement Ani- of Oklahoma, Norman, Oklahoma, U.S.A.
mal, Unité de Physiologie de la Reproduction et des Comporte-
Wyatt C. To, M.D. Department of Otolaryngology and Com-
ments, Institut National de la Recherche Agronomique/Centre
municative Disorders, The Cleveland Clinic Foundation, Cleve-
National de la Recherche Scientifique, Nouzilly, France
land, Ohio, U.S.A.
Evan R. Reiter, M.D. Department of Otolaryngology–Head
Tari D. Topolski, Ph.D. Department of Health Services, Uni-
and Neck Surgery, Virginia Commonwealth University, Rich-
versity of Washington, Seattle, Washington, U.S.A.
mond, Virginia, U.S.A.
John Q. Trojanowski, M.D., Ph.D. Smell and Taste Center,
Klaus Reutter, Ph.D. Anatomical Institute, University of
Center for Neurodegenerative Disease Research, Department of
Tübingen, Tübingen, Germany
Pathology and Laboratory Medicine, University of Pennsylvania,
Philadelphia, Pennsylvania, U.S.A.
Edmund T. Rolls, D. Phil., D.Sc. Department of Experimental
Psychology, University of Oxford, Oxford, United Kingdom Luca Turin, Ph.D. Department of Physiology, University Col-
lege, London, United Kingdom
Gabriele V. Ronnett, M.D., Ph.D. Department of Neuro-
science, The Johns Hopkins University School of Medicine, Bal- Donald A. Wilson, Ph.D. Department of Zoology, University
timore, Maryland, U.S.A. of Oklahoma, Norman, Oklahoma, U.S.A.

Benoist Schaal, Ph.D. Centre Européen des Sciences du Goût, Martin Witt, M.D., Ph.D. Department of Anatomy, University
Dijon, France of Technology Dresden, Dresden, Germany

Hendrik N. J. Schifferstein, Ph.D. Department of Industrial Fumiko Yoshii, Ph.D. Graduate School of Science and Tech-
Design, Delft University of Technology, Delft, The Netherlands nology, Niigata University, Niigata, Japan

Marlene Schwanzel-Fukuda, Ph.D. Laboratory of Neurobiol- David M. Yousem, M.D. Department of Radiology, The Johns
ogy and Behavior, The Rockefeller University, New York, New Hopkins University School of Medicine, Baltimore, Maryland,
York, U.S.A. U.S.A.
Introduction and Historical Perspective

Richard L. Doty
University of Pennsylvania, Philadelphia, Pennsylvania, U.S.A.

I. INTRODUCTION “smoke.” Fire, with its dangerous and magical connota-

tions, must have become associated early on with religious
All environmental nutrients and airborne chemicals re- activities, and pleasant-smelling smoke was likely sent into
quired for life enter our bodies by the nose and mouth. The the heavens in rituals designed to please or appease the
senses of taste and smell monitor the intake of such mate- gods. Importantly, food and drink became linked to numer-
rials, not only warning us of environmental hazards, but de- ous social and religious events, including those that cele-
termining, in large part, the flavor of our foods and bever- brated birth, the attainment of adulthood, graduation to the
ages. These senses are very acute; for example, the human status of hunter or warrior, and the passing of a soul to a
olfactory system can distinguish among thousands of air- better life.
borne chemicals, often at concentrations below the detec- The goal of this introduction is to provide a brief histor-
tion limits of the most sophisticated analytical instruments ical overview of (1) the important role that tastes and odors
(Takagi, 1989). Furthermore, these senses are the most have played in the lives of human beings throughout mil-
ubiquitous in the animal kingdom, being present in one lennia and (2) key observations from the last four centuries
form or another in nearly all air-, water-, and land-dwelling that have helped to form the context of modern chemosen-
creatures. Even bacteria and protozoa have specialized sory research. Recent developments, which are described
mechanisms for detecting environmental chemicals— in more detail elsewhere in this volume, are briefly men-
mechanisms whose understanding may be of considerable tioned to whet the reader’s appetite for what is to follow.
value in explaining their modes of infection and reproduc- Although an attempt has been made to identify, rather
tion (Jennings, 1906; Russo and Koshland, 1983; van specifically, major milestones in chemosensory science
Houten, 2000). since the Renaissance, some important ones have undoubt-
While the scientific study of the chemical senses is of edly been left out, and it is not possible to mention, much
relatively recent vintage, the role of these senses in the ev- less discuss, even a small fraction of the many studies of
eryday life of humans undoubtedly extends far into prehis- this period that have contributed to our current fund of
toric times. For example, some spices and condiments, in- knowledge. Hopefully the material that is presented pro-
cluding salt and pepper, likely date back to the beginnings vides some insight into the basis of the present Zeitgeist.
of rudimentary cooking, and a number of their benefits pre- The interested reader is referred elsewhere for additional
sumably were noted soon after the discovery of fire. The re- perspectives on the history of chemosensory science (e.g.,
lease of odorants from plant products by combustion was Bartoshuk, 1978, 1988; Beidler, 1971a,b; Boring, 1942;
most likely an early observation, the memory of which is Cain, 1978; Cloquet, 1821; Corbin, 1986; Doty, 1976;
preserved in the modern word perfume, which is derived Douek, 1974; Farb and Armelagos, 1980; Farbman, 1992;
from the Latin per meaning “through” and fumus meaning Frank, 2000; Gloor, 1997; Harper et al., 1968; Harrington

xv
xvi Doty

and Rosario, 1992; Johnston et al., 1970; Jones and Jones, abundant in these societies, serving as meeting places for
1953; McBurney and Gent, 1979; McCartney, 1968; persons of all walks of life (Morfit, 1847).
Miller, 1988; Moulton and Beidler, 1967; Mykytowycz, In Grecian mythology, the invention of perfumes was
1986; Ottoson, 1963; Pangborn and Trabue, 1967; Parker, ascribed to the Immortals. Men learned of them from the
1922; Pfaff, 1985; Piesse, 1879; Simon and Nicholelis, indiscretion of Aeone, one of the nymphs of Venus; Helen
2002; Smith et al., 2000; Takagi, 1989; Wright, 1914; von of Troy acquired her beauty from a secret perfume, whose
Skramlik, 1926; Zippel, 1993). formula was revealed by Venus. Homer (eighth century
B.C.) reports that whenever the Olympian gods honored
mortals by visiting them, an ambrosial odor was left, evi-
II. A BRIEF HISTORY OF PERFUME AND dence of their divine nature (Piesse, 1879). Interestingly,
SPICE USE bad odors were a key element of a number of myths, in-
cluding that of Jason and the Argonauts (Burket, 1970). As
The relatively rich history of a number of ancient civi- a result of having been smitten with the wrath of Aphrodite,
lizations, particularly those of Egypt, Greece, Persia, and the women of Lemnos developed a foul odor, which drove
the Roman Empire, provides us with examples of how their husbands to seek refuge in the arms of Thracian slave
perfumes and spices have been intricately woven into the girls. The women were so enraged by their husbands’ ac-
fabric of various societies. Thousands of years before tions that one evening they slew not only their husbands,
Christ, fragrant oils were widely used throughout the but all the men of the island. Thereafter, Lemnos was a
Middle East to provide skin care and protection from the community of women without men, ruled by the virgin
hot and dry environment, and at least as early as 2000 B.C. queen Hypsiple, until the day when Jason and the Argo ar-
spices and fragrances were added to wine, as documented rived, which ended the period of celibacy and returned the
by an inscription on a cuneiform text known as the Enuma island to heterosexual life.
elish (Heidel, 1949). In Egypt, incense and fragrant sub- Perfumes were not universally approved of in ancient
stances played a key role in religious rites and cere- Greece. Socrates, for example, objected to them altogether,
monies, including elaborate burial customs, and whole noting, “There is the same smell in a gentleman and a slave,
sections of towns were inhabited by men whose sole pro- when both are perfumed,” and he believed that the only
fession was to embalm the deceased. As revealed in the odors worth cultivating were those that arose from honor-
general body of religious texts collectively termed the able toil and the “smell of gentility” (Morfit, 1847). Never-
“Book of the Dead”—a number of which predate 3000 theless, the use of perfumes became so prevalent in ancient
B.C. (Budge, 1960)—the Egyptians performed funeral cer-
Greece that laws were passed in Athens in the sixth century
emonies at which prayers and recitations of formulas (in- B.C. to restrain their use. Despite this prohibition, however,
cluding ritualistic repeated burning of various types of in- their use grew unabated, and the Greeks added greatly to
cense) were made, and where the sharing of meat and the stock of fragrant plants from the East that made up the
drink offerings by the attendees occurred. Such acts were core of the perfume industry.
believed to endow the departed with the power to resist Perfume and incense had religious significance to the
corruption from the darkness and from evil spirits that followers of Zoroaster, the Persian religious leader of the
could prevent passage into the next life, as well as to seal sixth century B.C., who offered prayers before altars con-
the mystic union of the friends and loved ones with the taining sacred fires to which wood and perfumes were
dead and with the chosen god of the deceased. The added five times each day (Piesse, 1879). It is noteworthy
prayers of the priests were believed to be carried via in- that, to this day, sandalwood fuels the sacred fires of the
cense into heaven and to the ears of Osiris and other gods Parsees (modem Zoroastrians) in India, and that similar rit-
who presided over the worlds of the dead (Budge, 1960). uals were required of the early Hebrews, as indicated by the
As noted in detail by Piesse (1879), the ancient Greeks following instructions from God to Moses (Exodus 30:1,
and Romans used perfumes extensively, keeping their 7–9, 34–38):
clothes in scented chests and incorporating scent bags to
add fragrance to the air. Indeed, a different scent was often And thou shalt make an altar to burn incense upon: of
applied to each part of the body: mint was preferred for the shittim wood shalt thou make it. And Aaron shall
arms, palm oil for the face and breasts, marjoram extract burn thereon sweet incense every morning: when he
for the hair and eyebrows, and essence of ivy for the knees dresseth the lamps, he shall burn incense upon it [the
and neck. At their feasts, Greek and Roman aristocrats altar]. And when Aaron lighteth the lamps at even, he
adorned themselves with flowers and scented waxes and shall burn incense upon it, a perpetual incense before
added the fragrance of violets, roses, and other flowers to the Lord throughout your generations. Ye shall offer
their wines. As would be expected, perfume shops were no strange incense thereon, nor burnt sacrifice, nor
Introduction and Historical Perspective xvii

meat offering; neither shall ye pour drink offering wash their hands, as it is the habit of other men to do
thereon. in the morning; for he tells them that to do so consti-
And the Lord said unto Moses, take unto thee tutes a sure obstruction to his incantations. This is the
three sweet spices, stacte, and onycha, and gal- case whether it is the witches themselves who wash
banum; these sweet spices with pure frankincense; of their hands, as we learn from the answer freely given
each shall there be a light weight. And thou shalt to her examiners by Alexia Galaea of Betoncourt at
make it a perfume, a confection after the art of the Mirecourt in December 1584, and by countless oth-
apothecary, tempered together, pure and holy: And ers whose names I have not now by me; or whether it
thou shalt beat some of it very small, and put of it be- is the intended victims of their witchcraft who wash
fore the testimony in the tabernacle of the congrega- their hands, as was stated by Claude Fellet (Mersuay,
tion, where I will meet with thee: it shall be unto you February 1587) and Catharina Latomia (Haraucourt,
most holy. And as for the perfume which thou shalt February 1587).
make, ye shall not make to yourselves according to
In contrast to the detection of witches and warlocks by
the composition thereof: it shall be unto thee holy for
stench was the verification of sainthood by a pleasant odor,
the Lord. Whosoever shall make like unto that, to
the so-called “odor of sanctity.” If a saint had been an im-
smell thereto, shall even be cut off from his people.
postor, a nauseating smell, rather than a delectable one, was
Given such instructions from God and the Christian em- present upon exhumation of his body (Rothkrug, 1981).
phasis on cleansing the soul of evil spirits, as well as the This concept bears a striking resemblance to the Greek
fact that Christ himself, after his crucifixion, had been em- myths of the pleasant odors left by the Olympian gods who
balmed in pleasant-smelling myrrh, aloe, and spices (John visited mortals and may well stem from the same tradition.
19:39–40), it is perhaps not surprising that bad smells came It should be noted, however, that cleanliness was not
to signify the unholy at various times in Christian history. always the vogue for Christianity, as described by
Indeed, St. Philip Neri reportedly found the stench emanat- McLaughlin (1971) in a series of interesting accounts
ing from heretics so great that he had to turn his head (Sum- from the Middle Ages. Thus, in their repudiation of Ro-
mers, 1926). One of the more interesting, and tragic, uses man values, early Christians often went unbathed. Every
of bad smells was to identify witches and warlocks in Eu- sensation offensive to humans was believed acceptable to
rope in the late 1500s. Remy, a distinguished appointee of God, and the custom of bathing the limbs and anointing
Charles III to the Provosts of Nancy (a court that judged all them with oil was condemned. Monks shaved their hair,
criminal cases for some 72 villages in the Nancy region of wrapped their heads in cowls to avoid seeing profane ob-
France), wrote the following in his classic 1595 monograph jects, and kept legs naked except in the extreme of winter.
Demonolatry: St. Jerome criticized a number of his followers for being
too clean, and St. Benedict, a key administrator of the
In the Holy Scriptures the Devil is constantly referred early church, pronounced solemnly that “to those that are
to as Behemoth, that is to say, “the impure animal well, and especially to the young, bathing shall seldom be
and the unclean spirit” (see S. Gregory, in Memora- permitted.” St. Agnes reportedly had never washed
bilia, Matthew XII, Mark I and V, Job XI). It is not throughout her life, and a pilgrim to Jerusalem in the
only because the Devil is, as all his actions and pur- fourth century is said to have boasted that her face had
poses show, impure in his nature and character that gone unwashed for 18 years so as not to disturb the holy
we should consider this name to be aptly applied to water used at her baptism.
him; but also because he takes immoderate delight in During the Middle Ages, perfumery and the widespread
external filth and uncleanliness. For often he makes use of spices and flavoring agents was little known in Eu-
his abode in dead bodies; and if he occupies a living rope, being practiced mainly by Arabs in the East. Marco
body, or even if he forms himself a body out of the air Polo, visiting the China of Kublai Khan (1216–1295),
or condensation of vapours, his presence therein noted that pleasantly perfumed silk paper money was used
is always betrayed by some notable foul and noisome for exchange within Khan’s kingdom (Boorstin, 1985).
stench. The gifts of the Demon are also fashioned The dearth of smell in Europe was to change dramatically,
from ordure and dung, and his banquets from however, as a major element of the Renaissance was the re-
the flesh of beasts that have died . . . for the most part lentless search for perfumes and spices, a number of which
[he] has for his servants filthy old hags whose were more valuable than silver or gold. The quest was not
age and poverty serve but to enhance their foulness; only for aesthetic delight; some of these agents made it pos-
and these . . . he instructs in all impurity and un- sible, much like cooking itself, to exploit a wider and more
cleanliness. . . . Above all he cautions them not to diverse range of foodstuffs, including ones that otherwise
xviii Doty

were unsafe or had little gastronomic appeal. In this regard, teeth, false hair, Spanish wool, iron stays, hoops,
it is of interest that at the siege of Rome in A.D. 408, Alaric, high-heeled shoes, bolstered hips, shall incur the
the victorious king of the Goths, demanded 3000 pounds of penalty of the law now in force against witchcraft
pepper as ransom for the city, and when the Genoese cap- and the like misdemeanors, and that the marriage,
tured Caesarea in A.D. 1101, each soldier received two upon conviction, shall stand null and void.
pounds of pepper as his share of the spoils (Verrill, 1940).
Perfume was introduced, at least in a widespread sense, The influences of such attempts to ban perfumes in Eng-
to medieval Europe by the crusaders. After the downfall of land were short-lived, as perfume vendors thrived, al-
the Roman Empire, the perfume industry moved to the though the state taxed them and required them, in 1786, to
Eastern Roman Empire, and Constantinople became the have licenses. By 1800 approximately 40 companies were
perfume center of the world. Reportedly, Avicenna (A.D. making perfumes in London. In the 19th century the revo-
980–1036), the great Arab scientist, philosopher, and lution that occurred in organic chemistry ensured the con-
physician, discovered a way to extract and maintain the fra- tinuance of perfume manufacturing in Britain; the first im-
grances of plants and possibly invented rose water (Takagi, portant successful synthetic odorant, coumarin, was
1989). In part because of its conducive soil and climate, prepared in 1863 by the British chemist Sir William Henry
southern France proved to be a natural place for the culti- Perkin (Vivino, 1960).
vation of flowers for the perfume industry, an industry for
which France gained world supremacy that continues to III. THE CHEMICAL SENSES AND EARLY
this day (Vivino, 1960). In 1190, King Philip II (Philip Au- MEDICINE
gustus, r. 1180–1223) of France granted the first charter to
a perfume maker. In 1370, Queen Elizabeth of Hungary The close association between odors, spices, and medicine
was given a perfume formula based upon rosemary, which was undoubtedly forged long before recorded history and
was the first recorded alcohol-based perfume. This per- was likely fostered not only by stenches associated with
fume, known as “The Queen of Hungary’s Water” or sim- plagues and death, but by the utility of essential oils and
ply “Hungary Water,” was in use for more than five cen- spices in warding off insects and microbes. Indeed, one
turies and may be the precursor to eau de cologne, which is reason why perfumes and spices were major objects of in-
said to have been invented around 1690 in Milan, Italy, by ternational trade in the ancient world was their medicinal
Jean-Paul Feminis, who later resided in Cologne. King properties. According to Morris (1984), such properties
Charles V (Charles the Wise, r. 1364–1380) planted large may have been as important to early civilizations as the de-
fields of flowers in France to obtain perfume materials, and velopment of the x-ray or discovery of penicillin was to
Charles VIII (r. 1483–1498) was reportedly the first French our own, as modern studies confirm that numerous essen-
monarch to appoint a court perfumer. The guild of glove tial oils and spices are very effective in controlling
and perfume-makers was established in Paris in 1656. Per- pathogens, including Staphylococcus and various tubercu-
fume was readily accepted as a substitute for bath in the losis bacilli. Apparently this observation first came to the
court of Louis XIV (r. 1643–1715), as the palace at Ver- attention of European scientists in the latter half of the
sailles totally lacked plumbing. The court of Louis XV (r. 19th century, when the perfumery workers at Grasse,
1715–1774) was well known for the extravagant uses of France, were found to have a much lower rate of cholera
perfumes; indeed, it was named “the perfume court,” re- and tuberculosis than the rest of the European population.
flecting the daily application of scents to skin, fans, cloth- As noted by Morris (1984, p. 15):
ing, furniture.
Perfumes lost their popularity in England for more than Essential oils have shown startling fungitoxic prop-
a century prior to the Victorian era, unlike the case in erties. Oil of clove is toxic to specific growths, and
France, Italy, and Spain (Piesse, 1879). Related to this loss oil of geranium is effective against a broad range of
of popularity was an act, introduced into the English par- fungi. Cymbopogon grasses, an Indian genus of aro-
liament in 1770, that warned women of the use of scents matic grasses, have been found effective against He-
and other materials in the seduction of men (Piesse, 1879, uninthosporium oryzae, a source of food poisoning,
p. 20): Aspergillus niger, a cause of seborrheic dermatitis of
the scalp, Absidia ramosa, a cause of otitis, and Tri-
That all women, of whatever age, rank, profession, or choderma viride, another cause of dermatitis. Man
degree, whether virgins, maids, or widows, that shall, has long guessed that these oils that the plant secreted
from and after such Act, impose upon, seduce or be- to protect itself from insect, fungal, and microbial
tray into matrimony, any of his Majesty’s subjects, dangers could serve him as well. Thus it is that the
by the scents, paints, cosmetic washes, artificial story of perfumery is intimately linked to the story of
Introduction and Historical Perspective xix

pharmacy. Our ancestors could not formulate the penetrating parts, not only affects the olfactory nerve, but is
germ theory of disease, but they assumed that what- spread through the whole brain and can deplete its pituita
ever smelled clean and healthy must be of use in and other overcourse humors, increasing the movement of
hygiene.* animal spirits” (Corbin, 1986, p. 62). During outbreaks of
the plague, defenses included the burning of incense, ju-
The history of hygiene and public health is closely asso- niper, laurel leaves, cypress, pine, balm, rosemary, and
ciated with the view that odors were the source, indeed often lavender, although, if effective, they were only marginally
the cause, of diseases and pestilence. The stenches that de- so. Various plague waters, to be poured on handkerchiefs or
veloped in the cities of Europe during the Middle Ages are into pomanders, were invented, including the original eau
unimaginable to us today. Conditions were so bad that, for de cologne. Unpleasant agents were also believed to keep
example, the monks of White Friars in London’s Fleet Street away the plague, and the members of many households
complained that the smell from the Fleet River overcame all crouched over their privies inhaling the fumes in attempts to
the frankincense burnt at their altars and killed many of their avert the disaster (McLaughlin, 1971). Even in the late
brethren (McLaughlin, 1971). Such problems were the back- 1800s, smells were associated with illnesses, as exemplified
drop of the spread of the plague epidemics that traversed Eu- by the belief that decaying organic matter in swamps pro-
rope and England in the 12th to 17th centuries. duced malaria (mal
bad, aira
air). This theory, appar-
As chronicled by Corbin’s (1986) fascinating account of ently initially proposed by Varro (116–28 B.C.) and Palla-
the history of hygiene and odors in 18th-century France, dius (fourth century A.D.), was brought to the more modern
health administration of that era was based on a catalog of stage by Morton (1697) and Lancisi (1717), but was aban-
noxious odors. Indeed, authorities sought to locate the net- doned after the French physician Alphonse Laveran (1881)
works of miasmas by “mapping the flux of smells that described the responsible parasite and Sir Ronald Ross
made up the olfactory texture of the city” (p. 55). The de- (1923) demonstrated, a few years later, its transmission by
sire to localize odors and to eliminate them in an effort to the female anopheline mosquito.
ward off diseases may well have been one reason why so In the history of medicine, both odors and tastes have
many odor classification schemes arose during the 18th been used at various times in the diagnosis of diseases (see
century, including those of van Haller (1756), Linnaeus Doty, 1981, for review). Even today, diabetes is diagnosed in
(1765), Lorry (1784/85), and Fourcroy (1798). some areas of the world on the basis of the patient’s acetone-
Throughout this period, as well as in earlier times, infec- like breath and sweet-tasting urine, although, in general, the
tion was believed to be stemmed by wearing a perfume or use of odor and taste in diagnosis has become a lost art. In
by burning aromatic pellets in special perfume pans. addition, certain smells and tastes were known to elicit
Lemery’s Pharmacopee universelle (1697) cataloged the symptoms of some diseases, including epilepsy and hysteria.
therapeutic value of aromatics and perfumes and suggested A classic example is reported by the Roman historian Caius
the prescription of “apoplectic balms” because “what is Plinius Caecilius Secundus (Pliny) in his Historia Naturalis
pleasing to the nose, being composed of volatile, subtle, and (circa A.D. 50), where sulfur and burning bitumen (asphalt)
were noted to induce seizures (Bailey, 1932), a phenomenon
*Billing and Sherman (1998) provide empirical support for the that has also been reported in more modern times (West and
hypothesis that the amount of spice in foodstuffs from various Doty, 1994). Alum (alumen), which contained aluminum,
world cuisines is better explained on the basis of their antibacte- was used as a deodorant in the Roman empire, predating the
rial than their sensory properties. These investigators quantified
use of aluminum salts as deodorants in the United States in
the frequency of use of 43 spices in the meat-based cuisines of
the 1880s, as evidenced by the following quotation of Pliny,
136 countries for which traditional cookbooks could be found. A
total of 4578 recipes from 93 cookbooks was examined, along which extols its values (Bailey, 1932, p. 103):
with information on the temperature and precipitation in each
Liquid alumen has astringent, hardening, and corro-
country, the ranges of spice plants, and the antibacterial properties
sive properties. Mixed with honey, it heals sores in
of each spice. As mean annual temperatures (an index of relative
spoilage rates of unrefrigerated foods) increased, the proportion the mouth, pustules, and itchy eruptions. In the latter
of recipes containing spices, number of spices per recipe, total case, the treatment is applied in a bath to which
number of spices used, and use of the most potent antibacterial honey and alumen have been added in the proportion
spices all increased, both within and among countries. The esti- of two to one. Alumen diminishes offensive odours
mated fraction of bacterial species inhibited per recipe in each of the axilla, and reduces sweating in general.
country was positively correlated with annual temperature. Al-
though alternative hypotheses were considered (e.g., that spices To my knowledge, there are no pre-Renaissance trea-
provide macronutrients, disguise the taste and smell of spoiled tises on chemosensory dysfunction per se, although de-
foods, or increase perspiration and thus evaporative cooling), the scriptions of loss of olfactory function are found in the
data did not support any of these alternatives. writings of the ancient Greeks and Romans. Perhaps the
xx Doty

first description of anosmia was that by Theophrastus in the force of a blow, sound, magnetism, and odor (Riti, 1974).
third century B.C. (Stratton, 1917, p. 84): Cardinal Gasparao Contarini (1482–1542), an alchemist,
wrote about the elements and their combinations in five
. . . it is silly to assert that those who have the keen-
brief volumes published posthumously in 1548 by Ioannes
est sense of smell inhale most; for if the organ is not
Gaignaeus. The last of these was dedicated to flavors,
in health or is, for any cause, not unobstructed, more
odors, and colors. Contarini believed that there were eight
breathing is to no avail. It often happens that man has
flavors or tastes and argued that cooking food or preserving
suffered injury [to the organ] and has no sensation
at all. fruit can produce flavors not found in nature. He felt that
the sense of smell was imperfect and noted that the names
Although the early Greeks routinely used surgical inter- of flavors are often employed to explain the variety of
vention for the treatment of polyps and other intranasal ob- odors. Andrea Vesalius devoted one and a half large pages
structive problems (for review, see Wright, 1914), the first to the sense of smell in his classic anatomy treatise De Hu-
description of the use of surgery to specifically correct mani Corporis Fabrica (1543), although he failed to ob-
anosmia was apparently made during the Renaissance by serve the olfactory filaments. In 1566, Gryll published
Forestus (1591; cited in Lederer, 1959): what may be the first work solely devoted to the sense of
If it [anosmia] is from ethmoidal obstruction, or from taste, and in 1581, Fernel listed nine types of basic taste
the humor discharged from catarrh, the latter must qualities, including the seven of Aristotle and Galen
first be cured. If from the flesh growing from within (sweet, bitter, sour, salty, astringent, pungent, harsh) and
the nose . . . it is to be cured by the surgeons by op- “fatty” and “insipid,” the latter apparently reflecting the
erative procedures, either with a cutting instrument, lack of other taste qualities (Bartoshuk, 1978). Casserius
or cautery, or snare. (1609) described the detailed structure of the tongue, and
Malpighi (1664) and Bellini (1665) associated the sense of
Claudius Galenus (Galen; A.D. 130–200), whose writ- taste with lingual papillae. Taste buds were first identified
ings had a major impact on Western medicine in general, on the barbels and skin of fishes by Leydig (1851) and later
attributed anosmia to obstruction of the foramina within the were described in mammals (Loven, 1868; Schwalbe,
cribriform plate (an attribution made by a number of early 1868). In 1587, Iohannes Camerarius presented a thesis to
Greeks, including Plato and Hippocrates). He correctly de- the University of Marburg entitled “Themata Physica de
scribed the role of the nose in warming and filtering the air Odorum Natura et Affectionibus.” In this work, he dis-
and alluded to empirical studies noting the permeability of cussed odor classification, the relationship between taste
the dura matter around the cribriform plate to both water
and smell, a mechanism for explaining the function of ol-
and air (Wright, 1914). He believed that the organ for smell
faction, the ability of smelling in water, and the effect of
was located in the ventricles of the brain and that particles
heat from the sun on odors (Kenneth, 1928).
responsible for olfactory sensations passed through the
In 1673, Robert Boyle wrote an article, “Nature, Prop-
foramina of the cribriform plate during inhalation. As dis-
erties, and Effects of Effluvia,” in which he provides vivid
cussed in more detail later, this compelling idea continued
and accurate observations on such topics as olfaction in
until the 18th century, when light microscopy revealed that
birds, odor tracking in dogs, and the physical nature of the
the nasal secretions came from secretory cells within the
materials released from various odor sources. In his 1675
epithelium. In terms of taste, he posited that the lingual
paper, “Experiments and Observations About the Mechan-
nerve communicated gustatory sensations, in accord with
modern perspectives (see Chapters 32 and 44). ical Production of Odours,” he addresses some simple is-
sues of odorant mixtures and observes that the quality and
intensity of odors can be related. He provides, in his “Ex-
IV. THE RENAISSANCE AND THE BIRTH periments and Considerations About the Profity of Bodies”
OF MODERN STUDIES OF TASTE (1684), perhaps the first description of intravascular olfac-
AND SMELL tion or taste (i.e., the smelling or tasting of substances that
are initially bloodborne):
As is evidenced in this book, major advances have been
made in understanding the senses of taste and smell—ad- One of the notablest instances I ever met with of the
vances that follow on the footsteps of a long tradition of porosity of the internal membranes of the human
scientific observations stemming from treatises written in body, was afforded to me by that British nobleman,
the 16th century. Indeed, the sense of smell did not escape of whom our famous Harvey tells a memorable, not
the attention of Leonardo da Vinci (1452–1519), who, in to say matchiless story. This gentleman, having in his
the Codex Atlanticus, presented nine diagrams next to one youth by an accident, which that doctor relates, had a
another in which he compared the behavior of light, the great and lasting perforation made in his thorax, at
Introduction and Historical Perspective xxi

which the motion of his heart could be directly per- faction than in taste during the post-Renaissance period
ceived, did not only out-live the accident, but grew a stemmed from the compelling, albeit erroneous, conceptual
strong and somewhat corpulent man; and so robust, framework in which olfactory functioning, disease, and
as well as gallant, that he afterwards was a soldier, nasal secretion were viewed. For smelling to occur, odor-
and had the honour to command a body of an army ous bodies had to enter the brain via the foramina of the
for the King. cribriform plate—the same foramina through which body
This earl of Mount-Alexander . . . gave me the op- humors were believed to flow to produce nasal mucus.
portunity of looking into his thorax, and of discern- From this perspective, blockage or alterations in this pas-
ing there the motions of the cone, as they call it, or sageway (e.g., by the changes in the viscosity of the hu-
mucro of the heart. . . . Having then made several in- mors) were closely related to diseases that caused (1) anos-
quiries fit for my purpose, his lordship told me, that, mia, (2) running noses, (3) high fever, and (4) general ill
when he did, as he was wont to due from time to time, feeling.
(though not every day) inject with a syringe some ac- There is no doubt that the major conceptual chemosen-
tually warm medicated liquor into his thorax, to sory advance of this period, indeed perhaps of the entire
cleanse and cherish the parts, he should quickly and modern era, was refutation of this ancient concept. The
plainly find in his mouth the taste and smell of the compelling nature of this theory and its adaptation to a
drugs, wherewith the liquor had been impregnated. more modern era is illustrated by Descartes’ (1644) de-
And I further learned, that, whereas he constantly scription of how olfaction works (Haldane and Ross, 1955,
wore, upon the unclosed part of his chest, a silken p. 292):
quilt fluffed with aromatic and odoriferous powders,
. . . two nerves or appendages to the brain, for they do
to defend the neighboring parts and keep them warm;
not go beyond the skull, are moved by the corporeal
when he came, as he used to do after several weeks,
particles separated and flying in the air—not indeed
to employ a new quilt, the fragrant effluvia of it
by any particles whatsoever, but only by those which,
would mingle with his breath in expiration, and very
when drawn into the nostrils, are subtle and lively
sensibly perfume it, not, as I declared I suspected,
enough to enter the pores of the bones which we call
upon the score of the pleasing exhalations, that might
the spongy, and thus to reach the nerves. And from
get up between his cloathes and his body, but that got
the diverse motions of these particles, the diverse
into the organs of respiration, and came out with his
sensations of smell arise.
breath at his mouth, as was confirmed to me by a
grave and judicious statesman, that happened to be Interestingly, convincing evidence for this notion con-
then present, and knew this general very well. tinued to be amassed during this period, as the following
passage from Thomas Willis (1681, p. 100) indicates:
From the 14th to mid-19th centuries, research on gusta-
tion was much more limited in scope than that on olfaction, The Sieve-like Bone in divers Animals is variously
although notable advances were made in taste research, in- perforated for the manifold necessity and difference
cluding (1) the discovery that dissimilar metals, when of smelling. A Process from the Dura Mater and man-
placed on the tongue, produced an “electric taste” sensation ifold nervous Fibres pass through every one of its
(Sulzer, 1752; Volta, 1792), (2) the observation that taste holes, and besmear the inside of the Nostrils. But as
sensations are localized to papillae (Malpighi, 1664; Bell, the impressions of sensible things, or sensible
1803), (3) the identification of the chorda tympani as the Species, confined as it were by the undulation or wav-
nerve that mediates taste in the anterior tongue (Bellingeri, ing of the animal Spirits, ascend through the passages
1818; see Bartoshuk, 1978), and (4) the demonstration that of these bodies stretched out from the Organ towards
different regions of the oral cavity are differentially sensi- the Sensory; so the humidities watring the same bod-
tive to different taste qualities (Horn, 1825). As noted ies, for as much as some they may be more superflu-
above, the observation that taste buds exist within the ous than usual, may distil into the Nostrils through the
papillae of the mammalian tongue and depend on an intact same ways. For indeed such humors as are perpetu-
nerve supply came in the latter half of the 19th century (see ally to be sent away from the brain, ought so copiously
Chapter 32) (Loven, 1868; Merkel, 1880; Schwalbe, 1867; to be poured upon the Organs of Smelling, as we shall
Vintschgau and Hönigschmied, 1877), as did the painstak- shew hereafter, when we shall speak particularly of
ing mapping of the sensitivity of individual papillae to the smelling Nerves; in the mean time, that there is
stimuli representing the four basic taste qualities (Öhrwall, such a way of Excretion opening into the Nostrils,
1891; Kiesow, 1894). some observations, taken of sick people troubled with
One reason for the comparatively greater interest in ol- Cephalick diseases, do further perswade.
xxii Doty

. . . A Virgin living in this City, was afflicted a rectly noted that the olfactory nerve mediates qualitative
long time with a most cruel Head-ach, and in the odor sensations and the trigeminal nerve somatosensory
midst of her pain much and thin yellow Serum daily sensations (e.g., Good, 1822; Kirkes, 1849).
flowed out from her Nostrils; the last Winter this Ex- Schneider (1655) and Lower (1670) are generally cred-
cretion stopped for some time, and then the sick party ited as being the first to show that nasal secretions arise
growing worse in the Head, fell into cruel Convul- from glands, rather than from secretions secreted through
sions, with stupidity; and within three days dyed the cribriform plate. However, a century earlier Berenger
Apoplectical. Her Head being opened, that kind of del Carpi, who taught surgery at Bologna (1502–1527),
yellow Latex overflowed the deeper turnings and broke the Hippocratic and Galenic tradition and denied that
windings of the Brain and its interior Cavity or Ven- fluids passed through these foramina, suggesting that they
tricles. . . . I could here bring many other reasons, actually passed through the sphenoid sinus (Wright, 1914).
which might seen to perswade, that the Ventricles of The evidence that nasal secretions came from glands,
the Brain, of the Cavity made by the complicature or rather than through the cribriform plate, was clearly an im-
folding up of its border, is a mere sink of the excre- portant observation in the history of medicine.
mentitious Humor; and that the humors there con- Collectively, the aforementioned studies placed the first
gested, are purged out by the Nose and Palate. nails in the coffin of the theory propagated largely by
Galen’s works that the cribriform plate is pervious to odors
The idea of movement of humors from the brain to the and that the sense of smell lies within the ventricles of the
nasal cavity was most likely supported by other types of ev- brain. Other major studies before 1890, a number of which
idence as well. For example, demonstrations that dyes (e.g., are now considered classic, contributed the remaining nails
Indian ink), after injection into the subarachnoid space or to this coffin and include, in chronological order, those by
the cerebral spinal fluid, travel to the nasal mucosa via the Hunter (1786), Todd and Bowman (1847), Schultze (1856,
cribriform plate were made in the 19th century, and there is 1863), Ecker (1856), Eckhard (1858), Clarke (1861), Hoff-
no reason to believe that such information was not available man (1866), Martin (1873), Krause (1876), von Brunn
in earlier times (see Jackson et al., 1979, for review). (1875, 1880, 1892), Sidky (1877), Exner (1878), Ehrlich
It is not clear who deserves the credit for identifying the (1886), and Cajal (1889).
olfactory nerves in the upper nasal cavity, although, ac- The dawn of human chemosensory psychophysics also
cording to Wright (1914), the 7th-century Greek physician occurred in the 19th century, as illustrated by the previ-
Theophilis gave one of the better anatomical accounts of ously mentioned studies in which specific taste qualities
their distribution, despite the potential political ramifica- were painstakingly mapped on the tongue by Öhrwall
tions of going against Galen’s dictates. Graziadei (1971) (1891) and Kiesow (1894). Although Boring (1942) cred-
credits Massa, in 1536, as having first demonstrated the ol- its Fischer and Penzoldt (1886) as having measured the
factory nerves in humans, and Scarpa, in 1789, as having first absolute threshold to an odorant, Valentin, in 1848, de-
shown that the fine fila olfactoria actually end in the regio scribed a procedure that assessed olfactory sensitivity that
olfactoria (note, however, Scarpa’s 1785 article). Wright predated even Fechner’s publication of threshold method-
(1914), on the other hand, notes that the Italian Anatomist ology by a dozen years. Zwaardemaker (1925), who in-
Alessandro Achillini, who died in 1512, had described their vented the important draw-tube olfactometer (see Chapter
intranasal distribution. 10) and who performed sophisticated studies on a wide
Regardless of who is responsible for their first descrip- range of topics, including adaptation and cross-adaptation,
tion, there was considerable disagreement, which spanned credited Passy (1892) as having made an important step in
over a century and a half, among authorities as to whether the development of olfactometry. In essence, Passy dis-
the processes that extended from the olfactory bulbs into solved a given amount of odorant in alcohol in a 1:10 ratio.
the nasal cavity were, in fact, nerves. Indeed, even after This new solution was then again diluted in such a ratio,
they were generally accepted as nerves, debate lasted into and this was repeated over and over to provide a series of
the 1840s as to whether they mediated smell sensations. As dilution steps. For testing, a small amount of solution at
reviewed in Chapter 47, Francois Magendie (1824) was the each concentration was placed in liter bottles, which were
primary proponent of the idea that such sensations were heated slightly to evaporate the alcohol. Such bottles were
mediated via the trigeminal nerve, whereas Sir Charles Bell then sampled from highest concentration to lowest concen-
believed that the olfactory nerves subserved such sensa- tration until no smell was discernible to the subject.
tions (Shaw, 1833). As late as 1860, experiments appeared Zwaardemaker, however, expressed concern that the alco-
in the literature that addressed this point (e.g., Schiff, hol diluent used by Passy might influence the perception of
1860), although the more authoritative general physiology the test odorant. This potential problem was eliminated to
and medical textbooks from the 1820s to the 1850s cor- a large degree in the successive dilution series described by
Introduction and Historical Perspective xxiii

Toulouse and Vaschide (1899) and Proetz (1924), where syphilis, small-pox, and porphyra. The SECOND is
water and mineral oil, respectively, served as the diluents. produced by neglected and long continued coryzas,
As noted above, the more scholarly physicians of the and a persevering indulgence in highly acrid sternu-
18th and 19th centuries were very much aware of the ma- tatories.
jor types of olfactory disorders that we recognize today.
Among the more detailed and vivid descriptions of
Good (1822), for example, classified disorders of olfaction
cases of anosmia in the 19th-century literature are those of
into the following categories: Parosmia acris (acute smell),
Ogle (1870). He describes in detail three cases of anosmia
Parosmia obtusa (obtuse or distorted smell), and Parosmia
due to head injury in which taste function was intact—a
expers (anosmia or lack of smell). Good (1822, pp.
case of anosmia associated with facial palsy, a viral-in-
260–261) notes the following regarding Parosmia obtusa:
duced case of anosmia, and a case of anosmia due to ob-
struction—and three cases of unilateral olfactory loss that
The evil is here so small that a remedy is seldom
were related to aphasia, agraphia, and seizure attributable
sought for in idiopathic cases; and in sympathetic af-
to brain lesions. The olfactory losses due to head injury
fections, as when it proceeds from catarrhs or fevers,
were believed to be caused by the shearing of the olfactory
it usually, though not always, ceases with the cessa-
filaments at the level of the cribriform plate from move-
tion of the primary disease. It is found also as a symp-
ment of the brain produced by the blow. In this explanation
tom of hysteria, syncope, and several species of
of the problem, Ogle notes (p. 266) that “the anterior brain
cephalaea, during which the nostrils are capable of
rests directly upon the bones of the skull, and is not sepa-
inhaling very pungent, aromatic, and volatile er-
rated from them as is the case elsewhere by the interposi-
rhines, with no other effect than that of a pleasing and
tion of cerebro-spinal fluid.”
refreshing excitement.
Where the sense of smell is naturally weak, or
continues so after catarrhs or other acute diseases, V. THE MODERN ERA: MAJOR
many of our cephalic snuffs may be reasonably pre- 20TH-CENTURY ADVANCES IN
scribed, and will often succeed in removing the he- OLFACTION AND GUSTATION
betude. The best are those formed of the natural or-
der verticillatae, as rosemary, lavender, and
The major progress in the field of chemosensory research
marjoram; if a little more stimulus be wanted, these
that took place in the 20th century was due largely to con-
may be intermixed with a proportion of the teucrium
temporaneous advances in other fields of science. In-
Marum; to which, if necessary, a small quantity of
cluded among such advances, which are not mutually ex-
asarum may also be added: but pungent errhines will
clusive, are (1) the development of analytical devices
be sure to increase instead of diminishing the defect.
such as the gas chromatograph and mass spectrometer, (2)
Good’s observations concerning Parosmia expers were the refinement of experimental design and the develop-
as follows: ment of statistical methodology, (3) advances in basic
psychophysical techniques, (4) the invention of sensitive
This species is in many instances a sequel of the pre- methods for recording minute electrical potentials from
ceding [Parosmis obtusa]; for whatever causes oper- the nervous system, including recordings from single cells
ate in producing the former, when carried to an ex- and isolated components of cell membranes, (5) the de-
treme or continued for a long period, may also lay a velopment of not only new histological stains, but radi-
foundation for the latter. But as it often occurs by it- cally novel histological procedures, such as those that uti-
self, and without any such introduction, it is entitled lize autoradiography, immunohistochemistry, and various
to be treated separately. It offers us the two following tracing agents, (6) the development and application of
varieties: biochemical techniques for assessment of endocrine and
Organica. Organic want of smell. From natural neurotransmitter receptor events, (7) the development and
defict, or accidental lesion, injurious to the struc- continued refinement of microimaging systems, including
ture of the organ. the electron microscope, (8) advances in tissue prepara-
Paralytica. Paralytic want of smell. From local tion procedures that optimize such imaging technology,
palsy. such as osmium preparations and freeze fracture tech-
The FIRST VARIETY occurs from a connate desti- niques, (9) the invention of computerized tomography
tution of olfactory nerves, or other structural defect; (CT), magnetic resonance imaging (MRI), and other non-
or from external injuries of various kinds; and is of- invasive tools useful for evaluating the structure of the
ten found as a sequel in ozaenas, fistula lachrymalis, brain in vivo, (10) the development of functional imaging
xxiv Doty

techniques, such as positron emission tomography (PET), and Shibuya, 1987; Firestein and Werblin, 1989), (3) mea-
single photon emission computed tomography (SPECT), surement of ion channel activity in restricted patches of ol-
and functional MRI (fMRI), and (11) numerous other ma- factory or taste cell membranes (Nakamura and Gold, 1987;
jor advances in biology, including the development of the Kinnamon et a1.,1988); (4) topographic analysis of
fields of animal behavior and, importantly, molecular bi- responses within the olfactory epithelium and olfactory
ology and molecular genetics, where recombinant DNA bulb (e.g., Kauer and Moulton, 1974; Kubie and Moulton,
techniques, for example, have been used to identify and 1980; Leveteau and MacLeod, 1966; Mackay-Sim et al.,
confirm the roles of many proteins involved in olfactory 1982; Mozell, 1964, 1966; Moulton, 1976), (5) the applica-
transduction. tion of voltage-sensitive dyes for recording electrical
Obviously, in this introduction it is possible to mention changes in chemosensory neural tissue (Kauer, 1988), (6)
only a few of the many important observations made in this the recording of olfactory- and taste-evoked potentials in
century that have contributed to the present research cli- higher brain regions (e.g., Funakoshi and Kawamura, 1968,
mate. The areas selected for exposition were chosen, in 1971; Kobal and Plattig, 1978; Plattig, 1968/1969), and (7)
part, on the basis of the amount of research they have gen- electrophysiological mapping of local, as well as more
erated and are continuing to generate. A number of these global, olfactory and gustatory brain circuits (e.g., Emmers
events or areas of research are mentioned in more detail et al., 1962; Getchell and Shepherd, 1975; Komisaruk and
elsewhere in this volume. Beyer, 1972; Mori and Takagi, 1978; Motokizawa, 1974;
Nicoll, 1971; Norgren, 1970; Pfaffmann et al., 1961; Rall et
A. Electrophysiological Studies al., 1966; Scott and Pfaffmann, 1972; Shepherd, 1971,
1972; Tanabe et al., 1973, 1975). Electrophysiological data
A major 20th-century milestone that had a significant im- were critical for overturning the long-held notion that the
pact on modern chemosensory science was the develop- entire “limbic lobe,” including the cingulate and parahip-
ment of means for electrophysiologically recording nerve pocampal gyri and hippocampus, were the primary projec-
impulses from the olfactory and gustatory receptors and tion regions of the olfactory system (e.g., Foville, 1844;
pathways. Although crude electrical recordings were ob- Zuckerkandl, 1888). Thus, Hasama (1934) and Allen
tained from the olfactory system in the late 19th century (1943) employed electrophysiological methods to accu-
(e.g., Saveliev, 1892; Garten, 1900), the sophisticated rately localize and delineate the olfactory cortex to a cir-
equipment necessary for reliable recordings, including sen- cumscribed region of the piriform lobe in the rabbit. Kaada
sitive electrodes, was not available until well into the 20th (1951) localized, using similar methods, analogous regions
century [e.g., the oscilloscope was invented in 1922 (Er- in the monkey.
langer and Gasser, 1937)]. The earliest extracellular
recordings of single-cell gustatory primary afferent nerve B. Studies of Receptor Function
activity were made by Zotterman (1935) and Pfaffmann
(1941). Kimura and Beidler (1961) and Tateda and Beidler Considerable progress has been made in the last several
(1964) were first to record from single cells within the decades in elucidating the initial events that occur when a
taste bud. Recordings of olfactory nerve fiber bundles were stimulus molecule activates either a taste or olfactory re-
made by Beidler and Tucker (1955); recordings of single- ceptor cell, as is evidenced by the studies reviewed in detail
cell olfactory receptor activity from extracellular in Chapters 4 and 34. A number of these studies of receptor
electrodes were obtained by Hodgson et al. (1955) and function have been performed solely at the genetic, molec-
Gesteland et al. (1963, 1965). The first evidence for ular genetic, or biochemical level, while others have made
between-species differences in single-cell neural firing was use of modern patch-clamp and dye-physiological measures
presented by Beidler and Tucker. (1955) and Pfaffmann (e.g., Korsching, 2002), sometimes in combination with
(1955), a point that was later exploited by Frank (1973) in biochemical ones, to address conductance changes that oc-
the use of the hamster as a model for species, such as the cur in the cell membrane following receptor activation.
human, that exhibit salient responsiveness to sweet-tasting In both gustation and olfaction, stimulants are initially
stimuli (see Chapters 9 and 35). absorbed by the mucus overlying the receptor cells. In some
Other manifestations of this advance in technology in- cases, soluble proteins within the mucus may assist in the
clude (1) the recording of multicellular summated poten- transport of hydrophobic molecules to receptor regions or
tials at the levels of the vertebrate olfactory mucosa [the possibly aid in stimulus removal and/or deactivation (Lee et
electro-olfactogram (EOG) (Hosoya and Yoshida, 1937; al., 1987; Pevsner et al., 1988a,b; Schmale et al., 1990;
Ottoson, 1956)] and insect antenna [the electroantenno- Pelosi, 2001). Such binding proteins have been found in
gram (Schneider, 1957a,b)], (2) recording of transduction humans (Briand et al., 2002), as well as in insects
currents in isolated olfactory receptor cells (e.g., Kurahashi (Kaissling, 2001), where they have been said to represent “a
Introduction and Historical Perspective xxv

major evolutionary adaptation regarding the terrestrializa- than olfaction. These oligonucleotides were then used as
tion of the olfactory system, converting hydrophobic odor- molecular probes. Eighteen clones were found that coded
ants into hydrophilic ones by increasing their aqueous solu- for proteins with seven transmembrane domains within ol-
bility” (Vogt et al., 1991, p. 74). factory tissue, but not within brain, retina, or various non-
After traversing the mucus, chemosensory stimuli inter- neural tissues. The variability in the amino acid sequences
act in some fashion with the receptor cell membrane (usu- was found to be in regions of the molecule believed to be
ally by binding with membrane-bound receptors, although important in the binding of ligands in other receptor proteins
other mechanisms may also be involved; e.g., Na salts with seven transmembrane domains. Based on this infor-
and some other tastants pass through and/or activate ion mation, Buck and Axel concluded that there is considerable
channels directly; see Chapter 34). In olfaction, Bronshtein diversity in the genes that code for olfactory receptors and
and Minor (1977) provided the first scientific evidence that that ⬃1000 receptors are likely present, although many of
these interactions occur on the olfactory cilia, an observa- these genes are now known to be pseudogenes. Other work-
tion supported by subsequent workers (e.g., Menco, 1977; ers have shown that receptor-encoding complement any
Rhein and Cagan, 1980; Lowe and Gold, 1991; see Chap- DNA can be expressed in nonneuronal cells, which, when
ters 2 and 4). In the taste system, as well as in the stimulated with appropriate odorants, generate second-mes-
vomeronasal organ, interactions occur on microvilli asso- senger responses (implying the receptors recognize odor-
ciated with apical portions of the sensory receptor cells ants and couple to G proteins in the host cells) (Raming et
(Avenet and Lindemann, 1987, 1990; Heck et al., 1984; see al., 1993). Interestingly, the same receptor gene family de-
also Biasi et al., 2001). In taste buds the importance of the scribed by Buck and Axel (1991) may also encode sperm
apical region was first stressed by Renqvist in 1919 (see cell receptors possibly involved in chemotaxis during fer-
Kinnamon and Cummings, 1992). Although occluding tilization (Parmentier et al., 1992). More recently, G-pro-
junctions among cells in the apical region of the taste bud tein–coupled receptors have also been shown to be involved
restrict most stimuli to that region, low molecular weight in vertebrate vomeronasal (Herrada and Dulac, 1997) and
molecules may permeate these junctions (e.g., Holland and taste (Hoon et al., 1999) reception, as well as in invertebrate
Zampighi, 1991). Until the application of electron mi- taste and olfactory reception (Bargmann and Kaplan, 1998;
croscopy to taste buds in the 1950s (de Lorenzo, 1958; En- Clyne et al., 2000; Vosshall et al., 2000).
gström and Rytzner, 1956a,b), taste buds were believed to A number of vertebrate olfactory (Krautwurst et al.,
contain cilia, not microvillae. 1998; Malnic et al., 1999; Touhara et al., 1999; Zhao et al.,
A considerable body of evidence indicates that the in- 1998), vomeronasal (del Punta et al., 2002), and taste (Nel-
teraction of the stimulus molecule with the receptor mem- son et al., 2001) receptors have now been functionally iden-
brane opens or closes, directly or indirectly (i.e., via sec- tified (for invertebrates, see Carlson, 2001; Wetzel et al.,
ond-messenger systems), membrane channels, resulting in 1999). For example, Zhao et al. (1998) used an adenovirus-
a change in the flux of ions and alteration of the cell’s rest- mediated gene transfer procedure to increase the expres-
ing potential. Taste receptor cells possess a number of ion sion of a specific receptor gene in an increased number of
channels identical to some of those found in neurons (i.e., receptor neurons in the rat olfactory epithelium, demon-
voltage-gated Na, Ca2, and K channels, as well as strating ligand-specific increases in EOG amplitude.
Ca-mediated cation channels and amiloride-sensitive Krautwurst et al. (1998) employed a polymerase chain re-
Na channels) and release a neurotransmitter that activates action (PCR) strategy to generate an olfactory receptor li-
the first-order taste neuron (Roper, 1989; Lindemann, brary from which cloned receptors were screened for odor-
2001). In the case of olfaction, the receptor cell is the first- ant-induced responsiveness to a panel of odorants, as
order neuron, so changes in membrane potential, if of suf- measured by an assay sensitive to intracellular Ca2
ficient magnitude, generate the action potential (Frings, changes. Several receptor types with ligand specificity
2001; for vomeronasal organ, see Liman, 2001). were found, including one differentially sensitive to the
Empirical evidence for considerable diversity in olfac- () and () stereoisomers of citronella.
tory receptors was presented by Buck and Axel (1991) (see In 1989, a G-protein with 88% sequence identity to con-
also Young and Trask, 2002). Under the assumption that ol- ventional Gs, designated Gaolf or Golf, was isolated in ol-
factory receptors have elements in common with a large su- factory cilia (Jones and Reed, 1989) and found to the pre-
perfamily of surface receptors that evidence seven trans- dominant signaling G-protein in olfactory receptor cells.
membrane domains and linkage to guanine nucleotide– Interestingly, Golf has also been found elsewhere in the
binding proteins (G proteins) and second-messenger sys- CNS. For example, it has recently been implicated in the
tems, these investigators synthesized oligonucleotides that regulation of dopamine and adenosine action in the stria-
coded for conserved (i.e., nearly invariant) amino acid se- tum (Hervé et al., 2001). Although G-proteins other than
quences found among receptors from sensory systems other Golf (e.g., Gi2 and Go) have been identified in olfactory re-
xxvi Doty

ceptor cells, they appear not to be involved in early trans- (Brunet et al., 1996), or (3) both the cyclic-nucleotide-
duction events, presumably assisting in such processes as gated ion channel and Golf (Belluscio et al., 1998). In the
axonal signal propagation, axon sorting, and target inner- channel knockout mouse, EOG responses to all odorants
vation (Wekesa and Anholt, 1999). However, Gi2 and Go tested were eliminated, including those previously believed
appear to play primary roles vomeronasal sensory trans- to be mediated by the IP3 system (Brunet et al., 1996). To
duction (Biasi et al., 2001). date, IP3-gated channels have not been demonstrated in
A taste tissue–specific G-protein perhaps analogous to mammalian olfactory nerve cells using patch clamp tech-
Golf, called “gustducin,” appears to be important in taste niques (Firestein et al., 1991; Lowe and Gold, 1993).
transduction, although its involvement seems to be primar-
ily for sweet- and sour-tasting stimuli. Mice in whom the C. Studies of the Olfactory Pathways in
gustducin gene has been deleted still are able to discern Transport of Agents from the Nose to
such stimuli, although only at much higher concentrations, the Brain
suggesting that multiple taste pathways may be involved or
that G-proteins can substitute for gustducin in the signaling
process (Ruiz-Avila et al., 2001) (see Chapter 34). A very important empirical observation, made in the first
Early studies indicated that the enzyme adenylyl cy- half of the 20th century, was that the olfactory nerve can
clase, which is usually coupled to a G-protein, is highly serve as a conduit for the movement of viruses and exoge-
active in olfactory (Kurihara and Koyama, 1972; Pace et nous agents from the nasal cavity into the brain (see Chap-
al., 1985) and gustatory (Striem et al., 1991) tissues. ters 3 and 26). This route is direct, since the olfactory neu-
Adenylyl cyclase activity is increased, typically in the pres- rons lack a synapse between the receptive element and the
ence of GTP, in ciliary preparations by a number of olfac- afferent path. The existence this pathway for viral infection
tory ligands (Pace et al., 1985; Shirley et al., 1986; Sklar of the brain has been recognized for some time, as evidenced
al., 1986). A positive correlation was found to exist be- a number of studies from the 1920s and 1930s (see Clark,
tween an odorant’s ability to activate adenylyl cyclase ac- 1929; Hurst, 1936). For example, mice intraperitoneally in-
tivity in a frog ciliary preparation and both its perceived oculated with louping ill virus showed the first signs of cen-
odor intensity to humans (Doty et al., 1990) and the mag- tral nervous system (CNS) localization of the virus in the ol-
nitude of the EOG response it produces in frog epithelia factory bulbs. Mice whose olfactory mucosa was cauterized
(Lowe et al., 1989), suggesting that a functional relation- with zinc sulfate were partly protected against such infection
ship exists between the amount of adenylate cyclase acti- (Barnet and Lush, 1938). Poliomyelitis virus, placed in the
vated and the intensity of odor perception. noses of primates, travels to the olfactory bulbs via the axo-
A number of odorants and tastants increase, in a dose- plasm of the olfactory nerves, rather than along the nerve
related manner, intracellular cyclic adenosine 3,5- bundle sheaths (Bodian and Howe, 1941a,b). In a pioneering
monophosphate (cAMP) in olfactory and taste receptor paper, Armstrong and Harrison (1935) reported that mon-
cells, respectively (Bruch and Teeter, 1989; Pace et al., keys could be protected against intranasal inoculations of po-
1985; Pace and Lancet, 1986; Sklar et al., 1986), thereby liomyelitis virus by previous lavage of the nose with solu-
triggering the opening of cAMP-gated cation channels tions of alum or picric acid (or both). Subsequent studies
(Nakamura and Gold, 1987). In the case of taste, cAMP- (e.g., Schultz and Gebhardt, 1936) found that zinc sulfate
mediated responses may be limited to sweet- and/or bitter- gave a longer-lasting and higher degree of protection from
tasting agents, although other pathways, including the acti- poliomyelitis, leading to the prophylactic spraying of noses
vation of the cyclic guanosine monophosphate (cGMP), of children with this agent during poliomyelitis outbreaks in
may be involved as well. cAMP likely plays a role in the the late 1930s (Peet et al., 1937; Schultz and Gebhardt, 1937;
modulation of the sensitivity of olfactory receptor neurons, Tisdall et al., 1937). Unfortunately, such spraying produced
such as during adaptation (Leinders-Zufall et al., 1996). long-lasting, presumably permanent, anosmia in some indi-
Although it was believed, in the case of olfaction, that a viduals (Tisdall et al., 1938).
second transduction pathway occurs in vertebrates Related to the observation that the olfactory nerves are
(namely, that associated with the activation of the enzyme a major carrier of viruses is the fact that the receptive ele-
phospholipase C to produce the second messenger inositol ments of the olfactory system are exposed, to a large de-
triphosphate or IP3) (Breer and Boekhoff, 1991), recent gree, to the vagaries of the external environment, making
data suggest this may not be the case, at least in mice (Gold, them susceptible to damage from bacteria, viruses, toxins,
1999). The discordant studies have used knockout mice in and other foreign agents. As reviewed in detail in Chapters
which genes have been deleted that are responsible for (1) 3, 26, and 27, there is a wealth of evidence that the olfac-
an olfactory-specific adenylyl cyclase (Wong et al., 2000), tory mucosa is rich in enzymes that presumably minimize
(2) olfactory-specific cyclic-nucleotide-gated ion channels the deleterious influences and uptake of most xenobiotic
Introduction and Historical Perspective xxvii

agents into the olfactory receptor cells, including cy- immature receptor cells that fail to establish synaptic con-
tochromes P-450, flavin-containing monooxygenase, and nections with the olfactory bulb. This hypothesis implies
aldehyde dehydrogenases and carboxylesterases. that environmental agents play an important role in dictat-
ing which elements of the receptor sheet become replaced
D. The Discovery That Taste Buds and and that the rate of regeneration of the olfactory receptor
Olfactory Receptor Cells Regenerate cells is not genetically predetermined, as previously sup-
posed (see Mackay-Sim and Kittle, 1991) (Chapters 5 and
Another important 20th-century development was the dis- 29). The observation that improvement in olfactory function
covery that both the gustatory and olfactory receptor cells after cessation of chronic cigarette smoking occurs over a
can regenerate. Beidler and Smallman (1965) provided the period of years and is dose-related (Frye et al., 1990) sug-
first scientific demonstration that the sensory cells of the gests that either turnover of the olfactory epithelial cell
taste bud are in a dynamic state of flux and are constantly complement takes a much longer time than previously sup-
being renewed, with the more recently formed cells of the posed or growth of olfactory epithelium into damaged areas
periphery migrating centrally to act as receptors for very is relatively slow and dependent on the extent of prior
limited periods of time. The observation that olfactory trauma, or both.
receptor cells, which are derived from ectoderm and which The study of the regeneration of the olfactory neurons
serve as the first-order neurons, can regenerate after they are has been greatly enhanced by the ability to culture the ol-
damaged was first noted in mice by Nagahara (1940) and factory mucosa in vitro (for review, see Mackay-Sim and
later confirmed in primates by Schultz (1960). This obser- Chuah, 2000). This was first demonstrated in the culture of
vation is particularly significant in that it is in conflict with olfactory organs from embryonic mice (Farbman, 1977)
the long-held notion that neurons in the adult animal are ir- and used to show the importance of olfactory bulb in pro-
replaceable (see next section) and suggests that the olfac- moting differentiation of the olfactory sensory neurons
tory system may contain the key to producing neural regen- (Chuah and Farbman, 1983). The next major development
eration in a variety of neural systems (Farbman, 1992). came with the investigations of dissociated cultures from
However, questions remain as to why metaplastic respira- embryonic and newborn rats (Calof and Chikaraishi, 1989;
tory epithelium often invades the region of the damaged ol- Pixley, 1992a). This allowed the growth factors regulating
factory epithelium and why, when such metaplasia occurs, olfactory neurogenesis to be explored in the developing ol-
the epithelium in that region may never convert to olfactory factory epithelium (DeHamer et al., 1994; Mahanthappa
epithelium. Studies in which the olfactory epithelia of ro- and Schwarting, 1993) and in the adult (MacDonald et al.,
dents were exposed to airborne or systemically adminis- 1996; Newman et al., 2000).
tered toxic agents may shed some light on this question. Recently, normal targeting of glomeruli by olfactory re-
Thus, the type of repair seems to correlate with the degree ceptor axons has been demonstrated in mice lacking func-
or extent of the initial epithelial damage (Keenan et al., tional olfactory cycle nucleotide-gated channels (Lin et al.,
1990). For example, when the basilar layer of the mucosa is 2000) and in mice lacking most intrabulbar GABAergic in-
completely damaged, then metaplastic replacement with a terneurons (Bulfone et al., 1998). Thus, establishment of
respiratory-like epithelium occurs. When the damage is not the topographical map from the receptor cells to the
marked or the toxic insult is not sustained, regeneration, glomeruli seems to require neither normal neural activity in
usually with fewer or irregularly arranged cells, occurs. these pathways nor cues provided by the major neural cell
Closely related to the discovery of regeneration within types of the bulb.
the olfactory epithelium is the important observation made Human olfactory neuronal progenitors have now been
by Andres (1966, 1969) that mitotic cells, young sensory grown in vitro (Wolozin et al., 1992). This has been ex-
cells, mature sensory cells, and dying cells coexist within ploited to study biochemical changes in Alzheimer’s dis-
the olfactory epithelium (see Farbman, 1992, for a review). ease (Wolozin et al., 1993). Primary cultures of human ol-
This suggested to Andres the hypothesis that the olfactory factory mucosa (Féron et al., 1998; Murrell et al., 1996)
receptor cells were continually being replaced. The notion have lead to investigations into the etiology of schizophre-
that olfactory receptor cells were in a state of flux received nia (Féron et al., 1999), and the vitro growth of olfactory
subsequent support by others (see Chapters 2–6; Moulton et ensheathing cells (Chuah and Au, 1991; Pixley, 1992b; Ra-
al., 1970; Thornhill, 1970; Graziadei and Metcalf, 1971) mon-Cueto and Nieto-Sampedro, 1992). The latter glial
and led to the idea that they are relatively short-lived. Hinds cells assist sensory neuron regeneration (Doucette, 1984)
et al. (1984), however, found that a number of the olfactory and have, in fact, been employed in cell transplantation
receptor cells of mice reared in a pathogen-free environ- therapy for the damaged nervous system (Li et al., 1998; Lu
ment survived for at least 12 months and hypothesized that et al., 2001; Ramon-Cueto et al., 2000; Ramon-Cueto and
olfactory nerve cell turnover involves recently formed or Nieto-Sampedro, 1994).
xxviii Doty

E. The Discovery That Some Olfactory Bulb ing. It has long been known or suspected that brain circula-
Cells Regenerate tion changes selectively with neuronal activity (e.g., Broca,
1879; Mosso, 1881; Roy and Sherrington, 1890; Fulton,
A long-held dogma regarding the nature of the CNS of ver- 1928), but was not until the late 1950s and early 1960s that
tebrates is now known to be false; namely, that the adult the development of the [131I]trifluoroiodomethane
brain does not exhibit neurogenesis (for review, see Gross, ([131I)CF31]CF31) method provided a potential and novel
2000). Although early studies found mitotic figures within means for quantitatively examining the influences of sen-
the walls of the lateral ventricle (e.g., Allen, 1912; Globus sory, cognitive, and motor processes on local blood flow
and Kuhlenbeck, 1944; Olpalski, 1934; Rydberg, 1932), within regions of the brain (Landau et al., 1955; Freygang
definitive evidence that such cells represented neurogenesis et al., 1958; Kety, 1960; Sokoloff, 1961). This early work
awaited the development of the tritiated thymidine tech- led in the development of the [14C]2-deoxy-D-glucose (2-
nique, the electron microscope, and immunohistochemistry DG) autoradiographic method for determining regional
(Gross, 2000). In the 1960s, Altman and his associates glucose consumption in animals (Reivich et al., 1971;
published a series of classic studies based upon tymidine au- Kennedy et al., 1975; Sokoloff et al., 1977, Sokoloff, 1981,
toradiography demonstrating neurogenesis in several brain 1982), and set the foundation for modern human functional
imaging studies. Reivich et al. (1979) introduced the [18F]
regions of young and adult rats, including the olfactory bulb
flurorodeoxyglucose method for assessing regional glu-
(Altman, 1969), the neocortex (Altman, 1963, 1966a), and
cose metabolism, and Lassen et al. (1963) and Ingvar and
the dentate gyrus of the hippocampus (Altman, 1963; Alt-
Risberg (1965) subsequently developed and applied a pro-
man and Das, 1965). Regarding the olfactory bulb, prolifer-
cedure in which regional blood flow measurements could
ating cells were found within the subventricular zone lining
be established in humans by using scintillation detectors ar-
segments of the lateral ventricles. These cells were found to
rayed over the surface of the scalp. The refinement of such
reach the core of the olfactory bulb via the rostral migratory
approaches led to the practical development of positron
stream. Subsequent studies have confirmed and extended
emission tomography (PET) (Ter-Pogossian et al., 1975;
these observations (e.g., Luskin, 1993; Lois, Garcia-Ver-
Hoffman et al., 1976), which was made possible by the ear-
dugo and Alvarez-Buylla, 1996; O’Rourke, 1996), noting
lier invention of x-ray computed tomography (CT) in 1973
that the precursor cells invade the granule and periglomeru-
(Hounsfield, 1973). The coincidence of these techniques
lar layers of the bulb, where they differentiate into local in-
provided the capability of mapping the regions with in-
terneurons. A major differentiation is into GABAergic
creased blood flow or glucose metabolism to specific re-
granule cells—the most numerous cells of the bulb.
gions of the brain in three-dimensional coordinates.
These stem-cell–related phenomena, which are only now
Magnetic resonance imaging (MRI) technology
beginning to receive widespread attention within the chem-
emerged contemporaneously with the latter developments
ical senses community, are of considerable significance, as
(e.g., Lauterbur, 1973). Based upon a set of earlier princi-
they indicate that the plasticity of the olfactory system goes
ples (Block, 1946; Fox and Raichle, 1986; Lauterbur,
far beyond simply replacing damaged neuroepithelial cells,
1973; Pauling and Coryell, 1936), Ogawa et al. (1990)
and that continual cell replacement may play an integral role
were able to demonstrate that changes in blood oxygena-
in olfactory perception. It is now known, for example, that tion could be detected, in vivo, with MRI, setting the stage
reducing the numbers of interneurons recently generated via for the development of functional MRI (fMRI). This phe-
this process impairs the ability of an animal to discriminate nomenon, known as the blood oxygen level–dependent
among odorants (Gheusi et al., 2000). Moreover, enriching (BOLD) signal, reflects the fact that blood flow changes
the odorous environment of mice enhances such neurogen- more than oxygen consumption does in an activated region,
esis and improves odor memory (Rochefort et al., 2002). reflected by a reciprocal alteration in the amount of local
The degree to which such processes influence, or are influ- deoxyhemoglobin that is present, thereby altering local
enced by, endocrine state and various social processes is not magnetic field properties. Details of fMRI, as well as other
well known, although interestingly glucocorticoids de- imaging procedures, are presented in Chapters 12 and 37.
crease, and estrogens increase, the rate of such neurogene- The first study to employ functioning imaging in the
sis within the hippocampus (Gould and Tanapat, 1999; chemical senses was that of Sharp et al. (1975). These in-
Tanapot et al., 1999). vestigators injected four rats intravenously with 2-DG and
immediately placed them in a sealed glass jar containing
F. Functional Imaging Studies glass wool saturated with pentyl acetate. After 45 minutes
the animals were sacrificed, and sections of the bulbs were
A significant and rapidly evolving modern development in appropriately prepared and autoradiographed. Two regions
the study of the chemical senses is that of functional imag- of heightened optical density were noted bilaterally, which
Introduction and Historical Perspective xxix

tended to be centered in the glomerular layer, with variable aversion paradigms (e.g., Garcia et al., 1955), habituation
spread into the external plexiform and olfactory nerve lay- paradigms (e.g., Krames, 1970), sniff rate analysis
ers. Subsequent studies more clearly defined the regions of paradigms (e.g., Teichner, 1966), and operant conditioning
apparent activation (e.g., Sharp et al., 1977; Stewart et al., paradigms using positive or negative reinforcers (Skinner,
1979), and resulted in the identification of a unique set of 1938). Furthermore, behavioral studies have been critical
glomeruli in weanling rats responsive to odorants in their in the demonstration of the close association between neu-
mothers’ milk (Teicher et al., 1980). roendocrine and chemoreception systems in both verte-
The first published human olfactory PET study was brates and invertebrates and are critical for demonstrating
that of Zatorre et al. (1992). These investigators found the effects of various gene manipulations on smell- or taste-
that odorants increased regional cerebral blood flow mediated behaviors. A number of the chapters in this vol-
(rCBF) bilaterally in the piriform cortex, as well as uni- ume directly relate to this vast literature and, in some cases,
laterally in the right orbitofrontal cortex. The first taste provide means for assessing responses of animals to odor-
study employing PET was that of Small et al. (1997). In- ants (e.g., Chapters 18, 19, 20, 27, and 40). The reader is re-
creased rCBF was noted, in response to citric acid, bilat- ferred to the many general reviews of this topic (Albone,
erally within the caudolateral orbitofrontal cortex, and 1984; Doty, 1974, 1975, 1976, 1980, 1986; Johnston, 2000;
unilaterally within the right anteromedial temporal lobe Johnston et al., 1970; Leon, 1983; Marchlewska-Koj,
and the right caudomedial orbitofrontal cortex. The first 1983; Meredith, 1983; Mykytowycz, 1970; Slotnick, 1990;
published fMRI report on olfaction was that of Yousem et Slotnick and Schellinck, 2002; Smith, 1970; Stevens, 1975;
al. (1997), who demonstrated (1) odor-induced activation Vandenbergh, 1983; Wysocki, 1979).
of the orbitofrontal cortex (Brodmann area 11), with a According to Stürckow (1970), the studies by Barrows
mild right-sided predominance (in accord with the earlier (1907), von Frisch (1919), and Minnich (1921) were semi-
PET study of Zatorre et al., 1992) and (2) unexpected nal for the development of studies of insect chemosensory
cerebellar activation. Sobel et al. (1998a) noted that ol- behavior and physiology, even though earlier, more equiv-
factory stimulation activated lateral and anterior orbito- ocal, studies had been performed (e.g., Hauser, 1880). Bar-
frontal gyri of the frontal lobe, and that sniffing behavior, rows (1907) devised the first insect olfactometer and
regardless of whether an odor is present, induces piriform found, in the pomice fly (Drosophila ampilophila), that dif-
cortex activation. These investigators, following up on the ferent degrees of responding were obtained from different
unexpected observation of cerebellar activation by odor- concentrations of chemical attractants. Von Frisch (1919)
ants, subsequently demonstrated concentration-dependent demonstrated that bees could be trained to fly to a fragrant
odorant activation in the posterior lateral cerebellar hemi- odor using simple reinforcement and later found the loca-
spheres and activation from sniffing alone in the anterior tion of the olfactory sensilla to be on the eight distal seg-
cerebellum, most notably the central lobule (Sobel et al., ments of the antennae (von Frisch, 1921, 1922). Minnich
1998b) (see Chapter 12). (1921, 1926, 1929) explored the responses of various body
parts of butterflies, certain muscid flies, and the bee to taste
G. The Animal Behavior Revolution solutions. For example, he found that they extended their
proboscises when their tarsi or certain mouth parts were
Another large area of research activity that must be men- touched with a sugar solution. These and other studies led
tioned as having had a profound impact on modern to electrophysiological studies of the chemoreceptive sys-
chemosensory research is that of animal behavior and be- tems of insects by Dethier (1941), Boistel (1953), Boistel
havioral endocrinology (see Chapters 17–20, 41, 46). This and Coraboeuf (1953), Kaissling and Renner (1968), and
field, which grew in geometric proportions after World Schneider (1955, 1957a,b).
War II, is still a major contributor to chemosensory studies. A number of important studies published in the 1950s,
In addition to providing detailed explications of the many 1960s, and early 1970s demonstrated a close association
rich and often complicated influences of chemical stimuli between olfaction, social behavior, and reproductive pro-
on wide range of invertebrate and vertebrate behaviors (in- cesses in rodents and other mammalian forms. Pioneering
cluding, in mammals, behaviors related to aggression, reports on this topic include those that showed that
alarm, suckling and feeding, mating, predator–prey rela- volatiles from male and female mice influence the timing
tionships, social status appraisal, territorial marking, and of estrous cycles (Lee and Boot, 1955; Whitten, 1956;
individual and species recognition), this field has provided Whitten et al., 1968), that urine odor from unfamiliar male
important methodology for assessing olfactory, gustatory, mice can block the pregnancy of female mice (Bruce, 1959;
and vomeronasal function in animals, including preference Bruce and Parrott, 1960), and that chemical stimuli can ac-
paradigms (e.g., Richter, 1939; Mainardi et al., 1965), clas- celerate the onset of puberty in mice (Vandenbergh, 1969).
sical conditioning paradigms (Pavlov, 1927), conditioned Other important studies demonstrated that olfactory bul-
xxx Doty

bectomy, anesthetization, or damage to the olfactory re- Animal behavior studies in the 1980s contributed sig-
ceptor region or vomeronasal organ, alone or in combina- nificantly to the understanding of the function of
tion, can dramatically influence mating behavior, depend- vomeronasal organ which was described histologically in
ing on the species involved [e.g., in the male or many species in the 19th century, but whose function was
androgenized female hamster, anesthetization or damage unknown (for review of the early literature, see Wysocki,
of these systems can eliminate male copulatory behavior 1979). In the mouse, removal of the vomeronasal organ
(Doty and Anisko, 1973; Doty et al.,1971; Murphy and eliminates the surge in luteinizing hormone (LH) and sub-
Schneider, 1970; Powers and Winans, 1973, 1975; Winans sequent increase in testosterone that ordinarily follows ex-
and Powers, 1974)]. Such phenomena have been demon- posure of male mice to an anesthetized novel female mouse
strated to one degree or another in a wide variety of mam- or her urine. However, this does not occur following expo-
mals and have important implications for animal ecology, sure to an awake female mouse, suggesting that several
husbandry, and perhaps even human behavior. sensory cues can produce the LH surge (Coquelin et al.,
Other studies of this period that had a considerable im- 1984; Wysocki et al., 1983). In both mice and hamsters,
pact on the field of mammalian social behavior include vomeronasal organ removal impairs male sexual behavior,
those that examined, in a systematic manner, sexual odor particularly in animals that have had no prior adult contact
preferences. Godfrey (1958), for example, found that es- with females (Meredith, 1986; Wysocki et al., 1986). In
trous female bank voles (Clethrionomys) preferred homo- mice whose vomeronasal organs have been removed soon
specific male odors over heterospecific male odors and that after birth, long-lasting influences on male sexual behavior
hybrids were discriminated against. Le Magnen (1952) in adulthood have been noted (Bean and Wysocki, 1985).
demonstrated that adult male rats (Rattus norvegicus) pre- Vomeronasal organ removal also greatly decreases aggres-
fer the odor of receptive females to nonreceptive ones, sion in male house mice, particularly those that have not
whereas prepubertal or castrated males do not (unless they had much fighting experience with other males (Bean,
have been injected with testosterone). Beach and Gilmore 1982; DaVanzo et al., 1983; Wysocki et al., 1986). There is
(1949) noted that sexually active male dogs, but not a sex- now considerable evidence that the adult human has a
ually inactive male dog, preferred estrous to nonestrous vomeronasal lumen and at least a rudimentary vomeronasal
urine. This and other work led to a number of carefully de- gland, although no neural connections have been described
signed studies by Carr and associates in the 1960s, which and the weight of the evidence suggests it is vestigial
sought to determine the influences of sexual behavior and (Doty, 2001; Smith and Bhatnagar, 2000).
gonadal hormones on measures of olfactory function. Carr A significant event for the field of odor communication
and Caul (1962) demonstrated that both castrate and non- was the coining of the term “pheromone” in insects for
castrate male rats can be trained to discriminate between “substances which are secreted to the outside by an indi-
the odors of estrous and nonestrous females in a Y-maze vidual and received by a second individual of the same
test situation, implying that the preference phenomenon species, in which they release a specific reaction, for ex-
observed by Le Magnen (1952) was not due to castration- ample, a definite behavior or a developmental process”
related influences on olfactory discrimination ability, per (Karlson and Lüscher, 1959, p. 55). This term, which un-
se. Carr et al. (1965) subsequently demonstrated the im- fortunately has been applied by some workers to nearly any
portant role of sexual experience in producing strong pref- chemical involved in chemosensory communication in a
erences in male rats for estrous over diestrous odor and in wide variety of species, has permeated most areas of biol-
female rats for noncastrate male odors over castrate male ogy. This term replaced an earlier term (ectohormone) and
odors. These investigators also showed that sexually inex- conjures up the idea that the social organization of animals
perienced females preferred male noncastrate odors if they is akin to the endocrine organization of an organism, with
were administered gonadal hormones that induced estrus.* disparate parts being influenced by chemicals that circulate
These general findings have been observed in a wide range within the social milieu. For many insects, this notion
of species, although some species differences do exist and seems quite appropriate, given the high degree of stereo-
castration has been shown to mitigate the increase in de- typical behavior and evidence for comparatively simple
tection performance of rats that follows repeated testing stimuli that induce behavioral or endocrinological changes.
(Doty and Ferguson-Segall, 1989; for reviews, see Brown However, for many vertebrates, particularly mammals, the
and Macdonald, 1985; Doty, 1974, 1976, 1986). pheromone concept is of questionable value, as the term it-
self has little operational utility and many behavioral and
*In an unpublished M.A. thesis, Keesey (1962) found that sexu- endocrine responses said to be mediated by pheromones
ally experienced, but not sexually inexperienced, male rats pre- are either learned, induced by stress, or not unique to ol-
ferred the odor of female urine collected during proestrus than faction, being mimicked by other types of sensory stimula-
that collected during diestrus. tion (see Chapter 17).
Introduction and Historical Perspective xxxi

Pioneering behavioral studies of mammalian taste func- (1982), who demonstrated that the olfactory system of rats
tion began in the 1930s, heralded by experiments that is functional in utero. In this study, unborn rat pups (gesta-
sought to explain so-called specific hungers, e.g., salt crav- tion day 20) received in utero injections of apple juice and
ing in patients with adrenal gland hypofunction. In seeking lithium chloride. After birth, these individuals showed evi-
to determine whether alterations in taste function are re- dence of having developed a conditioned aversion to the
sponsible for increased NaCl intake of adrenalectomized odor of apple juice.
rats, Richter (1936, 1939) developed the two-bottle taste
test (see also Richter and Campbell, 1940). In this test, dif- H. Clinical Chemosensory Studies
ferential fluid intake from two bottles, one of which con-
tains a tastant (e.g., a NaCl solution) and the other water
alone, is recorded over a period of time. The lowest con- Considerable progress in understanding chemosensory dis-
centration of the tastant that produces a differential intake orders has been made in the last few decades, as reviewed
is taken as the threshold measure. in detail in Chapters 21–30 and 40–45 of this volume. The
Although this pioneering behavioral procedure provided proliferation of clinical studies has been fueled, in large
a means for measuring a preference threshold, postinges- part, by the widespread commercial availability of stan-
tional factors may alter the behavioral response, and such a dardized psychophysical tests of olfactory function (e.g.,
threshold is conceptually different from a sensory threshold. Doty et al., 1984a, 2000, 2001). It is now widely appreci-
Thus, a lack of preference between two solutions need not ated that smell loss is markedly depressed in elderly per-
reflect an inability to discriminate between them [see Chap- sons (Doty et al., 1984b), and that the most common causes
ter 19 and Stevens (1975) for reviews of analogous proce- of permanent smell loss are (1) upper respiratory viral in-
dures for olfaction]. Subsequent workers, including Carr fections, (2) head trauma, and (3) nasal and sinus disease
(1952), Harriman and MacLeod (1953), Morrison (1967), (e.g., Deems et al., 1991). Moreover, it appears that these
and Morrison and Norrison (1966), utilized shock-avoid- disorders largely reflect damage to the olfactory neuroep-
ance paradigms or operant conditioning paradigms that pro- ithelium, as revealed by autopsy and biopsy studies (Douek
vided positive reinforcement to establish NaCl threshold et al., 1975; Hasegawa et al., 1986; Jafek et al., 1989, 1990;
values—values that were much lower than those obtained Moran et al., 1992). Most complaints of taste loss reflect
using Richter’s procedure and which corresponded more the loss of olfactory function, and flavor sensations are
closely to neural thresholds. Numerous modifications of be- largely derived from retronasal stimulation of the olfactory
havioral procedures for assessing taste function in mammals system during active deglutition (Mozell et al., 1969; Bur-
have since been developed which incorporate general prin- dach and Doty, 1987).
ciples that evolved from these pioneering behavioral studies We now know that the olfactory system seems more
(e.g., Brosvic et al., 1985, 1989; Spector et al., 1990) (see susceptible to damage than the taste system, although dam-
Chapter 41). Analogous procedures have been developed in age to regional lingual afferents is particularly striking in
olfaction (e.g., Bowers and Alexander, 1967; Braun et al., old age (Matsuda and Doty, 1995), and taste sensitivity is
1967; Braun and Marcus, 1969; Eayrs and Moulton, 1960; directly related to the number of taste buds or papillae stim-
Goff, 1961; Henton, 1969; Moulton, 1960; Moulton and ulated, regardless of whether stimulation is by chemicals or
Eayrs, 1960; Pfaffmann et al., 1958; Slotnick and Katz, by electrical current (Doty et al., 2001; Miller et al., 2002;
1974; Slotnick and Ptak, 1977) (for reviews, see Chapter 27 Zuniga et al., 1993). Moreover, it has become increasingly
and Slotnick and Schellinck, 2002). apparent that many medicines, including a number of an-
Another noteworthy development in behavioral testing tibiotics, antidepressants, antihypertensives, antilipid
was that of the conditioned aversion paradigm (Garcia et agents, and psychotropic drugs, can produce alterations of
al., 1955, 1974). In one variant of this technique, an animal the taste system (e.g., severe dysgeusia), alone or in com-
is allowed to drink or smell a novel tastant or odorant and bination with alterations in the smell system (Schiffman,
is then injected with an agent that produces nausea (e.g., 1983; Schiffman et al., 1998, 1999a,b, 2000). Importantly,
lithium chloride). The animal quickly learns to avoid the recent studies suggest that damage to one of the major taste
novel stimulus as a result of a single aversive conditioning nerves (e.g., one chorda tympani) may release inhibition on
experience, even if the aversion occurs long after the pre- other taste nerves (e.g., the contralateral glossopharyngeal
sentation of the sensory stimulus. This procedure can be nerve), resulting in hypersensitivity to some tastants and
used to establish whether detection of a given stimulus is the production of phantom dysgeusias (Lehman et al.,
present and is particularly useful for assessing cross-reac- 1995; Yanagisawa et al., 1998).
tivity of stimuli (i.e., the extent to which a stimulus has el- A major advance in the last few years is the discovery
ements in common with other stimuli). One of the more that smell loss is among the first, if not the first, signs of
novel applications of this technique was by Smotherman such common neurodegenerative diseases as Alzheimer’s
xxxii Doty

disease (AD) and idiopathic Parkinson’s disease (PD), and VI. CONCLUSIONS
that disorders sharing similar motor signs to PD, such as
progressive supranuclear palsy (PSP) and MPTP-induced In this introduction, a brief description of the significant
parkinsonism (MPTP-P), are largely unaccompanied by role that tastes and odors have played throughout the
such loss (see Chapter 23). Such observations imply that course of human history has been presented. In addition, a
olfactory testing can be of value not only in the detection of number of key studies, events, and trends have been iden-
some neurodegenerative disorders early in their develop- tified which form the backdrop of much of today’s
ment, but in differential diagnosis. Indeed, odor identifica- chemosensory research enterprise, providing perspective
tion testing accurately differentiates between patients with for the chapters that follow. The chapters of this volume
AD and those with major affective disorder (i.e., depres- provide detailed contemporary information related to most
sion) (Solomon et al., 1999; McCaffrey et al., 2000). Inter- of these trends and address the important role of chemosen-
estingly, longitudinal studies have now appeared indicating sory science in both basic and applied situations.
that olfactory dysfunction can be predictive of AD in indi- Until recently, the chemical senses have engendered,
viduals who are at risk for this disorder, particularly when relative to the other major senses, comparatively little at-
considered in relation to other risk factors (Bacon et al., tention on the part of the scientific and medical communi-
1998; Graves et al., 1999; Devanand et al., 2000). The only ties. This is due to a number of factors, including (1) the
neurodegenerative disorder for which a definitive physio- lack of a simple physical dimension analogous to wave-
logical basis has been found to date, however, is multiple length that correlates with olfactory or taste quality, (2) the
sclerosis, where a –0.94 correlation has been observed be- fact that chemosensory dysfunction rarely produces obvi-
tween odor identification test scores and the number of ous influences on such everyday activities as locomotion
plaques, as measured by MRI, in the subtemporal and sub- and social interaction, and (3) the widespread belief that the
frontal regions of the brain (Doty et al., 1997, 1998, 1999). chemical senses are of little importance to humans. As Bor-
While the olfactory bulbs of patients with schizophrenia ing (1942, p. 437) so aptly put it:
are markedly reduced in size, the functional significance of
If human culture could have been founded on a dog’s
this is yet to be elucidated (Turetsky et al., 2000).
life, smell and not vision would be the great chapter
Diseases or disorders in addition to those noted above
of sensory psychology, and Helmholtz would have
that have been found to be associated with smell loss in-
written three huge volumes of a Handbuch des Phys-
clude severe alcoholism, amyotrophic lateral sclerosis
iologischen Geruchs, as well as a Die Lehre von den
(ALS), chronic obstructive pulmonary disease, cystic fi-
Geschmacksempfindungen als Physiologische
brosis, epilepsy, the Guam ALS/PD complex, head trauma,
Grundlager für die Theorie der Geschmackslehre.
Huntington’s disease, Kallmann’s syndrome, Korsakoff’s
psychosis, pseudohypoparathyroidism, psychopathy, rest- In the not-too-distant future, the amount of knowledge
less leg syndrome, schizophrenia, seasonal affective disor- within the chemical senses may well rival the level of
der, and Sjogren’s syndrome. Neurological disorders in knowledge of the visual sciences that was evidenced in
which olfaction seems to be spared, in addition to PSP and Helmholtz’s classic 19th-century three-volume treatise
MPTP-induced PD, are corticobasal degeneration, depres- (Helmholtz, 1851–1866/1924). When this occurs, the ram-
sion, panic disorder, essential tremor, and multiple chemi- ifications will be far-reaching, as suggested by Lewis
cal hypersensitivity (for review, see Doty, 2001). Thomas (1983, p. 14):
In addition to traditional medical means for treating or I should think that we might fairly gauge the future of
managing diseases responsible for decreased olfactory biological science, centuries ahead, by estimating the
function, surgical intervention at the level of the olfactory time it will take to reach a complete, comprehensive
neuroepithelium (e.g., by selectively ablating or stripping understanding of odor. It may not seem a profound
away the diseased tissue) or the olfactory bulb (e.g., by re- enough problem to dominate all the life sciences, but
moval of one or both olfactory bulbs in an anterior cranial it contains, piece by piece, all the mysteries.
approach) has successfully eliminated or markedly reduced
the symptoms of some forms of chronic dysosmia or phan-
tosmia (see Chapter 22). Recent advances in understanding ACKNOWLEDGMENTS
the deleterious influences of oxygen radicals on neural tis-
sue, as well as changes that occur in olfactory tissue at I thank Drs. Marion Frank, Alan Mackay-Sim, Bert Menco,
menopause, have led to ongoing studies of the prophylac- Igor Kratskin, David Smith, Gabriele Ronnett, Martin Witt,
tic potential of antioxidants, hormones, and other agents in and Klaus Reutter for their constructive comments on this
mitigating toxin-induced damage to the olfactory mucosa introduction. This paper was supported, in part, by Grants
(e.g., Dhong et al., 1999). PO1 DC 00161, RO1 DC 04278, and RO1 DC 02974 from
Introduction and Historical Perspective xxxiii

the National Institute on Deafness and Other Communica- from salt-taste receptor cells. J. Bas. Gin. Physiol. Pharmacol.
tion Disorders, and by Grant R01 AG 08148 from the Na- l: 383–391.
tional Institute on Aging, National Institutes of Health, Bacon, A. W., Bondi, M. W., Salmon, D. P., and Murphy, C.
Bethesda, MD. (1998). Very early changes in olfactory functioning due to
Alzheimer’s disease and the role of apolipoprotein E in olfac-
tion. Ann. NY Acad. Sci. 855:723–731.
REFERENCES Bailey, K. C. (1932). The Elder Pliny’s Chapters on Chemical
Subjects. Vol. II. E. Arnold and Company, London.
Allen, E. (1912). The cessation of mitosis in the central nervous Baisi, E., Silvotti, L., and Tirindelli, R. (2001). Pheromone detec-
system of the albino rat. J. Comp. Neurol. 22:547–569. tion in rodents. NeuroRep. 12:1–4.
Allen, W. F. (1943). Distribution of cortical potentials resulting Bargmann, C. I., and Kaplan, J. M. (1998). Signal transduction in
from insufflation of vapors into the nostrils and from stimula- the Caenorhabditis elegans nervous system. Ann. Rev. Neu-
tion of the olfactory bulbs and the pyriform lobe. Am. J. Phys- rosci. 21:279–308.
iol. 139: 81–92. Barrows, W. M. (1907). The reactions of the pomace fly,
Altman, J. (1963). Autoradiographic investigation of cell prolif- Drosophila ampelophila Loew, to odorous substances. J. Exp.
eration in the brains of rats and cats. Postnatal growth and dif- Zool. 4:515–537.
ferentiation of the mammalian brain, with implications for a Bartoshuk, L. M. (1978). History of taste research. In Handbook
morphological study of memory. Anat. Rec. 145:573–591. of Perception. Vol. VI A. Tasting and Smelling, E. C.
Altman, J. (1966a). Autoradiographic and histological studies of Carterette and M. P. Friedman (Eds.). Academic Press, New
postnatal neurogenesis. II. A longitudinal investigation of the York, pp. 3–17.
kinetics, migration and transformation of cells incorporating Bartoshuk, L. M. (1988). Taste. In Stevens’ Handbook of Experi-
tritiated thymidine in infant rats, with special reference to neu- mental Psychology. Vol. 1. Perception and Motivation, R. C.
rogenesis in some brain regions. J. Comp. Neurol. 128: Atkinson, R. J. Herrnstein, G. Lindzey, and R. D. Luce (Eds.).
Wiley, New York, pp. 461–499.
431–473.
Beach, F.A. and Gilmore, R.W. (1949). Response of male dogs to
Altman, J. (1966b). Proliferation and migration of undifferenti-
urine from females in heat. J. Mammal. 30:391–392.
ated precursor cells in the rat during postnatal gliogenesis.
Bean, N. J. (1982). Modulation of agonistic behavior by the dual
Exp. Neurol. 16:263–278.
olfactory system in male mice. Physiol. Behav. 29:433–437.
Altman, J. (1969). Autoradiographic and histological studies of
Bean, N. J., and Wysocki, C. J. (1985). Behavorial effects of re-
postnatal neurogenesis: IV. Cell proliferation and migration in
moval of the vomeronasal organ in neonatal mice. Chem.
the anterior forebrain, with special reference to persisting neu-
Senses 10:421–422.
rogenesis in the olfactory bulb. J. Comp. Neurol. 137:
Beidler, L. M. (1971a) (Ed.), Handbook of Sensory Physiology.
433–458.
Vol. IV. Chemical Senses. Part 1 Olfaction. Springer-Verlag,
Altman, J., and Das, G.D. (1965). Autoradiographic and histolog-
New York.
ical evidence of postnatal hypocampal neuroregenesis in rats.
Beidler, L. M. (1971b) (Ed.), Handbook of Sensory Physiology.
J. Comp. Neurol. 124:319–335. Vol. IV. Chemical Senses. Part 2. Taste. Springer-Verlag, New
Altman, J., and Das, G.D. (1966a). Autoradiographic and histo- York.
logical studies of postnatal neurogenesis. I. A longitudinal in- Beidler, L. M., and Smallman, R. L. (1965). Renewal of cells
vestigation of the kinetics, migration and transformation of within taste buds. J. Cell Biol. 27:263–272.
cells incorporating tritiated thymidine in neonate rats, with Beidler, L. M., and Tucker, D. (1955). Response of nasal epithe-
special reference to postnatal neurogenesis in some brain re- lium to odor stimulation. Science 122:76.
gions. J. Comp. Neurol. 126:337–390. Bellini, L. (1665). Gustus organum novissime deprehensum prae-
Andres, K. H. (1966). Der Feinbau der Regio olfactoria von missis ad faciliorem intelligentiam quibusdam de saporibus
Makrosmatikem. Z. Zellforsch. 69:140–154. [Taste organs newly observed; with certain premises about the
Andres, K. H. (1969). Der olfaktorische Saum der Katze. Z. Zell- senses given for easier understanding (translation from Latin
forsch. 96:250–274. into German by Jurisch (1922)]. Mangetus Bibliotheca Anat.
Anhof, R. R. H., Mumby, S. M., Stoffers, D. A., Girard, P. R., Kuo, 2, Bologna.
J. F., and Snyder, S. H. (1987). Transduction proteins of olfac- Belluscio, L., Gold, G.H., Nemes, A. and Axel, R. (1998). Mice
tory receptor cells: identification of guanine nucleotide binding deficient in Golf are anosmic. Neuron 20:69–81.
proteins and protein kinase C. Biochemistry 26:78, 795. Bhatnagar, K. P., Kennedy, R.C., Baron, G., and Greenberg, R.A.
Armstrong, C., and Harrison, W. T. (1935). Prevention of in- (1987). Number of mitral cells and the bulb volume in the ag-
tranasally inoculated poliomyelitis of monkeys by instillation ing human olfactory bulb: A quantitative morphological
of alum into nostrils. Public Health Rep. 725–730. study. Anat. Rec. 218:73–87.
Avenet, P., and Lindemann, B. (1987). Patch-clamp study of iso- Bhatnagar, K. P. and Meisami, E. (1998). Vomeronasal organ in
lated taste receptor cells in the frog. J. Membr. Biol. 97: bats and primates: extremes of structural variability and its
223–240. phylogenetic implications. Microsc. Res. Tech. 43:465–475.
Avenet, P., and Lindemann, B. (1990). Fluctuation analysis in the Biasi, E., Silvotti, L., and Tirindelli, R. (2001). Pheromone detec-
amiloride-blockable currents in membrane patches excised tion in rodents. NeuroReort. 12:1–4.
xxxiv Doty

Billing, J., and Sherman, P.W. (1998). Antimicrobial functions of nalling mechanisms in olfaction. In Chemical Senses: Recep-
spices: why some like it hot. Quart. Rev. Biol. 73:3–49 tor Events and Tiansduction in Taste and Olfaction, J. G.
Bodian, D., and Howe, H.A. (1941a). Experimental studies on in- Brand, J. H. Tetter, M. R. Kam, and R. H. Cagan (Eds.). Mar-
traneural spread of poliomyelitis virus. Bull. John Hopkins cel Dekker, New York, pp. 283–298.
Hosp. 68:248–267. Brunet, L. J., Gold, G. H., and Ngai, J. (1996). General anosmia
Bodian, D., and Howe, H. A. (1941b). The rate of progression of caused by a targeted disruption of the mouse cyclic
poliomyelitis virus in nerves. Bull. Johns Hopkins Hosp. nucleotide-gated cation channel. Neuron 17:681–693.
69:79–85. Buck, L., and Axel, R. (1991). A novel multigene family may en-
Boekhoff, L., and Breer, H. (1990). Differential stimulation of code odorant receptors: a molecular basis for odor recognition.
second messenger pathways by distinct classes of odorants. Cell 65:175–187.
Neurochem. Int. 17:553–557. Budge (1960). The Book of the Dead. University Books, New
Boistel, J. (1953). Etude fonctionelle des terminaisons Hyde Park, NY.
sensorielles des antennes d’Hyménoptères. C. R. Soc. Biol. Bulfone, A., Wang, F., Hevner, R., Anderson, S., Cutforth, T.,
147:1683–1688. Chen, S., Meneses, J., Pedersen, R., Axel, R. and Rubenstein,
Boistel, J., and Coraboeuf, E. (1953). L’activité électique dans J.L. (1998). An olfactory sensory map develops in the absence
I’antenne isolée de Léidoptè au cours de l’étude de l’olfaction. of normal projection neurons or GABAergic interneurons.
C. R. Soc. Biol. 147:1172–1175. Neuron 21:1273–1282.
Boorstin, D. J. (1985). The Discoverers. Vintage Books, New Brunet, L. J., Gold, G. H., and Ngai, J. (1996). General anosmia
York. caused by a targeted disruption of the mouse olfactory cyclic
Boring, E. G. (1942). Sensation and Perception in the History of nucleotide-gated cation channel. Neuron 17:681–693.
Experimental Psychology. Appleton-Century-Crofts, New Burdach, K. and Doty, R.L.(1987). Retronasal flavor perception:
York. Influences of mouth movements, swallowing and spitting.
Bowers, J. M., and Alexander, B. K. (1967). Mice: individual Physiol. Behav. 41:353–356.
recognition by olfactory cues. Science 158:1208–1210. Burket, W. (1970). Jason, hypsipyle, and new fire at Lemnos. A
Braun, J. J., and Marcus, J. (1969). Stimulus generalization study in myth and ritual. Class. Q. B:1–16.
among odorants by rats. Physiol. Behav. 4:245. Burnet, F. M., and Lush, D. (1938). Infection of the central ner-
Braun, J. J., Wermuth, B. and Haberly, L. (1967). An olfactory vous system by louping ill virus. Aust. J. Exp. Biol. Med. Sci.
discrimination apparatus: modification of the basic wind-tun- 16:233–240.
nel design. Psychonomic Sci. 9:515. Cain, W. S. (1978). History of research on smell. In Handbook of
Breer, H., and Boekhoff, I. (1991). Odorants of the same odor Perception. Vol. VIA. Tasting and Smelling, E.C. Carterette
class activate different second messenger pathways. Chem. and M. P. Friedman (Eds.). Academic Press, New York, pp.
Senses 16:19–29. 197–229.
Briand, L., Eliot, C., Nespoulous, C., Bézirard, V., Huet, J.-C., Cajal, R. S. (1889). Nuevos aplicaciones del metodo de Golgi:
Hendry, C., Blon, F., Trotier, D., and Pernollet, J.-C. (2002). terminaciones del nervio olfactorio en la mucosa nasal.
Evidence of an odorant-binding protein in the human olfactory Barcelona.
mucus: Location, structural characterization, and odorant- Calof, A. L., and Chikaraishi, D. M. (1989) Analysis of neuroge-
binding properties. Biochemistry 41:7241–7252. nesis in a mammalian neuroepithelium: proliferation and dif-
Broca, P. (1879). Sur les temperatures morbides locales. Bull. ferentiation of an olfactory neuron precursor in vitro. Neuron
Acad. Med. (Paris). 28: 1331–1347. 3:115–127.
Bronshtein, A. A., and Minor A.V. (1977). The regeneration of Carlson, J. R. (2001). Viewing odors in the mushroom body of the
olfactory flagella and restoration of electro-olfactogram after fly. Trends in Neurosciences. 24:497–498.
the treatment of the olfactory mucosa with Triton X-100. Cy- Carr, W. J. (1952). The effect of adrenalectomy upon the
tology 19:33–39. NaCl taste threshold in rat. J. Comp. Physiol. Psychol. 45:
Brosvic, G. M., Slotnick, B. M., and Tandeciarz, S. (1985). A 377–380.
computer-controlled automated system for gustatory psy- Carr, W. J., and Caul, W. F. (1962). The effect of castration in rat
chophysics. Chem. Senses l0:447. upon the discrimination of sex odours. Anim. Behav. 10:
Brosvic, G. M., Risser, J. M., and Doty, R. L. (1989). No influ- 20–27.
ence of adrenalectomy on measures of taste sensitivity in the Carr, W. J., Solberg, B., and Pfaffmann, C. (1962). The olfactory
rat. Physiol. Behav. 46:699–705. threshold for estrous female urine in normal and castrated
Brown, R. E. and Macdonald, D. W. (1985). Social Odours in male rats. J. Comp. Physiol. Psychol. 55:415–417.
Mammals (2 vol.). Clarendon Press, Oxford, 1985. Carr, W. J., Loeb, L. S., and Dissinger, M. E. (1965). Responses
Bruce, H. M. (1959). An exteroceptive block to pregnancy in the of rats to sex odors. J.Comp. Physiol. Psychol. 59:370–377.
mouse. Nature (Lond.) 814: 105. Casserius (1609). Penthaesteseion (cited after Jurisch, 1922).
Bruce, H. M., and Parrott, D. V. M. (1960). Role of olfactory Chuah, M. I., and Au, C. (1991), Olfactory schwann cells are de-
sense in pregnancy block by strange males. Science 131: rived from precursor cells in the olfactory epithelium. J. Neu-
1526. rosci. Res. 29:172–180.
Bruch, R. C., and Teeter, J. H. (1989). Second messenger sig- Chuah, M. I., and Farbman, A. I. (1983). Olfactory bulb increases
Introduction and Historical Perspective xxxv

marker protein in olfactory receptor cells. J. Neurosci. Doty, R. L. (1975). Determination of odour preferences in ro-
3:2197–2205. dents: a methodological review. In Methods in Olfactory Re-
Clark, W. E. L. (1929). Anatomical investigation into the routes by search, D. G. Moulton, A. Turk, and J. W. Johnson, Jr. (Eds.).
which infections may pass from the nasal cavities into the brain. Academic Press, London, pp. 395–406.
Rep. Public Health Med. Subjects No. 54. London, pp. 1–27. Doty, R. L., Ed. (1976). Mammalian Olfaction, Reproductive
Clarke, J.L. (1861). Über den Bau des Bulbus olfactorius and der Processes, and Behavior. Academic Press, New York.
Geruchsschleimhaut. Z. wiss. Zool. 11:31–42. Doty, R. L. (1980). Scent marking in mammals. In Comparative
Cloquet, H. (1821). Osphrèsiologie, ou Traitè,4 des Odeurs, du Psychology: Research in Animal Behavior, M. R. Denny
sens et des organes de l’olfaction. 2nd ed. Mèquignon-Marvis, (Ed.). Wiley, New York, pp. 385–399.
Paris. Doty, R. L. (1981). Olfactory communication in humans. Chem.
Clyne, P., Warr, C. G., and Carlson, J.R. (2000). Candidate taste Senses 6:351–376.
receptors in Drosophila. Science 287:1830–1834. Doty, R. L. (1986). Odor-guided behavior in mammals. Experi-
Contatini, G. (1548). De elementis [et] eorum mixtionibus libri entia 42:257–271.
quinque. Jean de Gaigny, Paris. Doty, R. L. (1998). Cranial Nerve I: Olfaction. In: C. G. Goetz
Coquelin, A., Clancy, A. N., Macrides, F., Nobel, E. P., and and E. J. Pappert (Eds.), Textbook of Clinical Neurology. W.B.
Gorski, R. A. (1984). Pheromonally induced release of Saunders, Philadelphia, pp. 90–101.
luteinizing hormone in male mice: involvement of the Doty, R. L. (2000). The Smell Threshold TestTM Administration
vomeronasal system. J. Neurosci. 4:2230–2236. Manual. Haddon Hts., NJ: Sensonics, 32 pp.
Corbin, A. (1986). The Foul and the Fragrant. Harvard Univer- Doty, R. L. (2001). Olfaction. Annu. Rev. Psychol. 52:423–452.
sity Press, Cambridge, MA. Doty, R. L., and Anisko, J. J. (1973). Procaine hydrochloride ol-
Cowart, B. J., Young, I.M., Feldman, R.S., and Lowry, L.D. factory block eliminates mounting in the male golden hamster.
(1997). Clinical disorders of smell and taste. In: G.K. Physiol. Behav. 10:395–397.
Beauchamp and L.M. Bartoshuk (Eds.), Handbook of Percep- Doty, R. L., Bagla, R., Morgenson, M., and Mirza, N. (2001).
tion and Cognition: Tasting and Smelling. San Diego: Aca- NaCl thresholds: relationship to anterior tongue locus, area
demic Press, pp. 175–198. of stimulation, and number of fungiform papillae. Physiol.
DaVanzo, J. P., Sydow, M., and Garris, D. R. (1983). Influence of Behav. 72:373–378.
isolation and training on fighting in mice with olfactory bulb Doty, R. L., Bromley, S. M., Moberg, P. J., and Hummel, T.
lesions. Physiol. Behav. 31:857–860. (1997). Laterality in human nasal chemoreception. In: S.
Deems, D. A., Doty, R. L., Settle, R. G., Moore-Gillon, V., Christman (Ed.), Cerebral Asymmetries in Sensory and Per-
Shaman, P., Mester, A. F., Kimmelman, C. P., Brightman, V. ceptual Processing (pp. 497–542). Amsterdam: North Hol-
J., and Snow, J. B., Jr. (1991). Smell and taste disorders: a land Publishing.
study of 750 patients from the University of Pennsylvania Doty, R. L, Carter, C. S., and Clemens, L. G. (1971). Olfactory
Smell and Taste Center. Arch. Otolaryngol. Head. Neck Surg. control of sexual behavior in male and early-androgenized fe-
117:519–528. male hamsters. Horm. Behav. 2:325–333.
DeHamer, M. K., Guevara, J. L., Hannon, K., Olwin, B. B., and Doty, R. L., and Ferguson-Segall, M. (1989). Influence of adult
Calof, A. L. (1994). Genesis of olfactory receptor neurons in castration on the olfactory sensitivity of the male rat: a signal
vitro: regulation of progenitor cell divisions by fibroblast detection analysis. Behav. Neurosci. 103:691– 694.
growth factors. Neuron 13:1083–1097. Doty, R. L., Kreiss, D. S., and Frye, R. E. (1990). Human odor in-
de Lorenzo, A. J. (1958). Electron microscopic observations on tensity perception: correlation with frog epithelial adenylate
the taste buds of the rabbit. J. Biophys. Biochem. Cytol. cyclase activity and trasepithelial voltage response. Brain Res.
4:143–150. 527:130–134.
Del Punta, K., Leinders-Zufall, T., Rodriguez, I., Jukam, D., Doty, R. L., Li, C., Mannon, L. J., and Yousem, D. M. (1997). Ol-
Wysocki, C. J., Ogawa, S., Zufall, F., Mombaerts, P. (2002) factory dysfunction in multiple sclerosis. N. Engl. J. Med.
Deficient pheromone responses in mice lacking a cluster of 336:1918–1919.
vomeronasal receptor genes. Nature. 419:70-74. Doty, R. L., Li, C., Mannon, L. J., and Yousem, D. M. (1998). Ol-
Dethier, V. G. (1941). The function of the antennal receptors in factory dysfunction in multiple sclerosis: Relation to plaque
lepidopterous larvae. Biol. Bull. 80:403–414. load in inferior frontal and temporal lobes. Ann. NY Acad. Sci.
Devanand, D. P., Michaels-Marston, K. S., Liu, X., Pelton, G. H., 855:781–786.
Padilla, M., Marder, K., Bell, K., Stern, Y., and Mayeux, R. Doty, R. L., Li, C., Mannon, L. J., and Yousem, D. M. (1999). Ol-
(2000). Olfactory deficits in patients with mild cognitive im- factory dysfunction in multiple sclerosis: relation to longitudi-
pairment predict Alzheimer’s disease at follow-up. Am. J. Psy- nal changes in plaque numbers in central olfactory structures.
chiatry 157:1399–1405. Neurology 53:880–882.
Dhong, H. J., Chung, S. K., and Doty, R. L. (1999). Estrogen pro- Doty, R. L., and Mishra, A. (2001). Olfaction and its alteration by
tects against 3-methylindole-induced olfactory loss. Brain nasal obstruction, rhinitis, and rhinosinusitis. Laryngoscope
Res. 824:312–315. 111:409–423.
Doty, R. L. (1974). A cry for the liberation of the female rodent. Doty, R. L., Shaman, P., and Dann, M. (1984a). Development of
Psychol. Bull. 81:159–172. the University of Pennsylvania Smell Identification Test: A
xxxvi Doty

standardized microencapsulated test of olfactory function. Fourcroy, A. F. (1798). Mèmoire sur 1esprit recteur de Boer-
Physiol. Behav. (Monograph) 32:489–502. rhaave, l’arome des chimistes francais, ou le principe de
Doty, R. L., Shaman, P., Applebaum, S.L., Giberson, R., Siko- l’odeuu des vègètaux. Ann. Chim. 26:232–250.
rsky, L., and Rosenberg, L. (1984b). Smell identification abil- Foville, T. (1844). Traité complet de l’anatomie, de la physiolo-
ity: Changes with age. Science 226:1441–1443. gie et de la pathologie due systèm nerveux cérébrospinal.
Doucette, J. R. (1984) The glial cells in the nerve fibre layer of the Paris: Fortin Nasson Cie.
rat olfactory bulb. Anat. Record 210:385–391. Frank, M. E. (1973). An analysis of hamster afferent taste nerve
Douek, E. (1974). The Sense of Smell and its Abnormalities. Liv- response functions. J. Gen. Physiol. 61:588–623.
ingstone, Edinburgh. Frank, M. E. (2000). Neuron types, receptors, behavior, and taste
Douek, E., Banniester, L.H., and Dodson, H.C. (1975). Recent ad- quality. Physiol. Behav. 69:53–62.
vances in the pathology of olfaction. Proc. Roy. Soc. Med. Freygang, W. H., and Sokoloff, L. (1958). Quantitative measure-
68:467–470. ments of regional circulation in the central nervous system by
Eayrs, J. T., and Moulton, D. G. (1960). Studies in olfactory acu- the use of radioactive inert gas. Adv. Biol. Med. Physics
ity. 1. Measurement of olfactory thresholds in the rat. Q. J. 6:263–279.
Exp. Psychol. 12:90–98. Frings, S. (2001). Chemoelectrical signal transduction in olfac-
Ecker, A. (1856). Über die Geruchschleimhaut des Menschen. tory sensory neurons of air-breathing vertebrates. Cell. Mol.
Über das Epithelium der Riechschleimhaut und die Life Sci. 58:510–519.
wahrscheinliche Endigung des Geruchnerven. Z wiss. Zool. Frye, R. E., Schwartz, B., and Doty, R. L. (1990). Dose-related ef-
8:303–306. fects of cigarette smoking on olfactory function. JAMA
Eckard, C. (1858). Über die Endigungsweise des Geruchsnerven. 263:2133–2136.
Beitr. Anat. Physiol. (Eckhard) Geissen 1:77–84. Fulton, J. F. (1928). Observations upon the vascularity of the hu-
Ehrlich, P. (1886). Uber die Methylenblaureaction der lebenden man occipital lobe during visual activity. Brain 51:310–320.
Nerven substanz. Deutsche Med. Wochenschr. 12:49–52. Funakoshi, M., and Kawamura, Y. (1968). Summated cortical re-
Emmers, R., Benjamin, R. M., and Blomquist, A. J. (1962). Tha- sponses to taste stimulation in man. J. Physiol. Soc. Jpn.
lamic localization of afferents from the tongue in albino rat. J. 30:282–283.
Comp. Neurol. 118:43–48. Funakoshi, M., and Kawamura, Y. (1971). Summated cerebral
Engström, H., and Rytzner. C. (1956a). The structure of taste evoked response to taste stimuli in man. Electroenceph. Clin.
buds. Ann. Oto-laryngol. 46:361–367. Neurophysiol. 30:205–209.
Engström, H., and Rytzner, C. (1956b). The fine structure of taste Garcia, J., Kimeldorf, D., and Koelling, R. A. (1955). A condi-
buds and taste fibres. Ann Oto. Rhinol. Laryngol. 65:361–375. tioned aversion towards saccharin resulting from exposure to
Erlanger, J., and Gasser. H. S. (1937). Electrical Signs of Nervous gamma radiation. Science 122:157–159.
Activity. University of Pennsylvania Press, Philadelphia. Garcia, J., Hankins, W. G., and Rusiniak, K. W. (1974). Behav-
Exner, S. (1878). Fortgesetzte Studien über die Endigungsweise ioral regulation of the milieu interne in man and rat. Science
des Geruchsnerven. Sitzber. Akad. Wiss. Wien Math.-naturw. 185:824–831.
Classe 76–111:171–221. Garten, S. (1900). Physiologie der marklosen Nerven. G. Fischer,
Farb, P., and Armelagos, G. (1980). Consuming Passions: The Jena.
Anthropology of Eating. Houghton Mifflin, Boston. Gesteland, R. C., Leavin, J. Y., Pitts, W. H., and Rojas, A. (1963).
Farbman, A. I. (1977). Differentiation of olfactory receptor cells Odor specificities of the frog’s olfactory receptors. In: Y. Zot-
in organ culture. Anat. Record 189:187–200. terman (Ed.), Olfaction and Taste. Pergamon Press, Oxford,
Farbman, A. I. (1992). Cell Biology of Olfaction. Cambridge Uni- pp. 19–44.
versity Press, Cambridge. Gesteland, R. C., Lettvin, J. Y., and Pitts, W. H. (1965). Chemi-
Féron, F., Perry, C., Hirning, M., McGrath, J. and Mackay-Sim, cal transmission in the nose of the frog. J. Physiol.
A. (1999). Altered adhesion, proliferation and death in neural 181:525–559.
cultures from adults with schizophrenia. Schizophr Res Getchell, T. V., and Shepherd, G. M. (1975). Short-axon cells in
40:211–218. the olfactory bulb: dendrodendritic synaptic interactions. J.
Féron, F., Perry, C., McGrath, J., and Mackay-Sim, A. (1998). New Physiol. 251:523–548.
techniques for biopsy and culture of human olfactory epithelial Getchell, T. V., Doty, R. L., Bartoslark, L. M., and Snow, J. B.,
neurons. Arch. Otolaryngol. Head Neck Surg. 124:861–866. Jr., Eds. (1991). Smell and Taste in Health and Disease. Raven
Firestein, S., and Werblin, F. (1989). Odor-induced membrane Press, New York.
currents in vertebrate olfactory receptor neurons. Science 244: Gheusi, G., Cremer, H., McLean, H., Chazal, G., Vincent, J-D.,
79–82. and Liedo, P-M. (2000). Importance of newly generated neu-
Firestein, S., Darrow, B., and Shepherd, G. M. (1991). Activation rons in the adult olfactory bulb for odor discrimination. Proc.
of the sensory current in salamander olfactory receptor neu- Nat. Acad. Sci. USA 97:1823–1828.
rons depends on a G protein-mediated cAMP second messen- Globus, J.H., and Kuhlenbeck, H. (1944). The subependymal cell
ger system. Neuron 6:825–835. plate (matrix) and its relationship to brain tumors of he
Fischer. E., and Penzoldt, F. (1886). Über die Empfindlichkeit des ependymal type. J. Neuropath. Exp. Neurol. 3:1–35.
Geruchssinnes. Sitzber. physik.-med. Sozietat Erlangen 18: Gloor, P. (1997). The Temporal Lobe and Limbic System. New
7–10. York, Oxford University Press.
Introduction and Historical Perspective xxxvii

Godfrey, J. (1958). The origin of sexual isolation between bank pheromone receptors in mammals with a topographical
voles. Proc. Roy. Phys. Soc. Edinburgh 27:47–55. organized and sexually dimorphic distribution. Cell 90:
Goff, W. R (1961). Measurement of absolute olfactory sensitivity 763–773.
in rats. Am. J. Psychol. 74:384–393. Hervé, D., Le Moine, C., Corvol, J-C., Belluscio, L., Ledent, C.,
Gold, G. H. (1999). Controversial issues in vertebrate olfactory Fienberg, A.A., Jaber, M., Studler, J-M., and Girault, J-A.
transduction. Annu. Rev. Physiol. 61: 857–871. (2001). Golf levels are regulated by receptor usage and con-
Good, J. M. (1822). The Study of Medicine. Vol. III. Baldwin, trol dopamine and adenosine action in the striatum. J. Neu-
Cadock and Joy, London. rosci. 21:4390–4399.
Gould, E. and Tanapat, P. (1999). Stress and hippocampal neuro- Hinds, J. W., Hinds, P. L., and McNelly, N. A. (1984). An au-
genesis. Biol. Psychiatry 46:1472–1479. toradiographic study of the mouse olfactory epithelium: evi-
Graves, A. B., Bowen, J. D., Rajaram, L., McCormick, W. C., dence for long-lived receptors. Anat. Rec. 210:375–383.
McCurry, S. M., Schellenberg, G. D., and Larson, E. B. Hodgson, E. S., Lettvin, J. Y., and Roeder, K. D. (1955). Phys-
(1999). Impaired olfaction as a marker for cognitive decline: iology of a primary chemoreceptor unit. Science 122:
interaction with apolipoprotein E epsilon4 status. Neurology, 417–418.
53:1480–1487. Hoffman, C. K. (1866). Onderzoekingen over den anatomischen
Graziadei, P. P. C. (1971). The olfactory mucosa of vertebrates. Bouw wart de Membranaolfactoria en het peripherische
In: L. M. Beidler (Ed.), Handbook of Sensory Physiology, Vol. Uitende van den Nervus olfactorius. Amsterdam, 58 pp.
IV. Chemical Senses. Sect. 2. Taste. Springer-Verlag, New Hoffman, E. J., Phelps, M. E., Mullani, N.A., Higgins, C.S., and
York, pp. 27–58. Ter-Pogossian, M.M. (1976). Design and performance charac-
Graziadei, P. P. C., and Metcalf, J. E. (1971). Autoradiographic teristics of a whole-body positron transaxial tomography. J.
and ultrastructural observations on the frog’s olfactory mu- Nucl. Med. 17:493–502.
cosa. Z. Zellforsch. 116:305–318. Holland, V. F., and Zampighi, G. A. (1991). Tight junctions in
Gross, C.G. (2000). Neurogenesis in the adult brain: death of a taste buds: possible role in perception of intravascular gusta-
dogma. Nature Rev. Neurosci. 1 :67–73. tory stimuli. Chem. Senses 16:69–80.
Gryllus, L. (1566). De sapore dulci et amaro. Prague: Georgium Hoon, M. A., Adler, E., Lindemeier, J., Battey, J. F., Ryba, N. J. P.,
Melantrichum ab Auentino. and Zuker, C. S. (1999). Putative mammalian taste receptors:
Haldane, E. S., and Ross, G. R. T. (1955). Philosophical Works a class of taste-specific GPCRs with distinct topographic se-
Rendered into English. Dover, New York. lectivity. Cell 96:541–551.
Horn, W. (1825). Uber den Geschmackssinn des Menschen. Karl
Harper, R., Bate Smith, E. C., and Land, D. G. (1968). Odour De-
Groos, Heidelberg.
scription and Odour Classification. American Elsevier, New
Hosoya, Y., and Yoshida, H. (1937). Uber die bioekktrischen Er-
York.
scheinungen an der Riechschleimhaut. Jpn. J. Med. Sci. III
Harriman, A. E., and MacLeod, R. B. (1953). Discriminative
Biophys. 5:22.
thresholds of salt for normal and adrenalectomized rats. Am. J.
Hounsfield, G. N. (1973). Computerized transverse axial scan-
Psychol. 66:465–471.
ning (tomography): Part I. Description of system. Brit. J. Ra-
Harrington, A., and Rosario, V. (1992). Olfaction and the prima-
diol. 46:1016–1022.
tive: nineteeth-century medical thinking on olfaction. In: M. J.
Hunter, J. (1786). Description of the nerves which supply the or-
Serby and K. L. Chobor (Eds.), Science of Olfaction. Springer-
gans of smelling. In Observations on Certain Parts of the An-
Verlag, New York, pp. 3–27. imal Economy, J. Hunter. London, pp. 213–219.
Hasagawa, S., Yamagishi, M., and Nakano, Y. (1986). Micro- Hurst, E. W. (1936). Newer knowledge of virus diseases of ner-
scopic studies of human olfactory epithelia following trau- vous system: review and interpretation. Brain 59:1–34.
matic anosmia. Arch. Otolaryngol. 243:112–116. Ingvar, G. H., and Risberg, J. (1965). Influence of mental activity
Hasama, B. (1934). Über die elektrischen Begleiterscheinungen upon regional cerebral blood flow in man. Acta Neurol. Scand.
an der Riechsphäre bei der Geruchsempfindung. Pflügers Supp. 14:183–186.
Arch. 234:748–755. Jackson, R. T., Tigges, J., and Arnold. W. (1979). Subarachnoid
Hauser, G. (1880). Physiologische und histologische Unter- space of the CNS, nasal mucosa, and lymphatic system. Arch.
suchungen über das Geruchsorgan der Insekten. Z. Wiss. Zool. Otolaryngol. 105: 180–184.
34:367–403. Jafek, B. W., Eller, P.M., Esses, B. A., and Moran, D.T. (1989).
Heck, G. L., Mierson, S., and DeSimone, J. (1984). Salt taste Posttraumatic anosmia. Arch. Neurol. 46:300–304.
tranaduction occurs through an amiloride-sensitive sodium Jafek, B. W., Hartman, D., Eller, P. M., Johnson, E.W., Strahan,
transport pathway. Science 223: 403–405. R.E., and Moran, D.T. (1990). Post-viral olfactory dysfunc-
Heidel, A. (1949). The Gilgamesh Epic and Old Testament Par- tion. Am. J. Rhinol. 4:91–100.
allels, 2nd ed. University of Chicago Press, Chicago. Jennings, H. S. (1906). Behaviour of the Lower Organisms.
Helmholtz H. von. (1856–1866/1924). Physiological Optics. Columbia University Press, New York.
(Translated by J. P. C. Southall from the 3rd German Edition.) Johnston, R. E. (2000). Chemical communication and
Optical Society of America, New York. (3 vol). pheromones: the types of chemical signals and the role of the
Henton, W. W. (1969). Conditioned suppression to odorous stim- vomeronasal system. In: T.E. Finger, W.L. Silver and D. Re-
uli in pigeons. J. Exp. Anal. Behav. 12:175–185. strepo (Eds.), The Neurobiology of Taste and Smell. 2nd edi-
Herrada, G., and Dulac, C. (1997). A novel family of putative tion. New York: Wiley-Liss, pp. 101–127.
xxxviii Doty

Johnston, J. W., Jr., Moulton, D. G., and Turk, A., Eds. (1970). Kobal, G., and Plattig, K. H. (1978). Methodische Anmerkungen
Advances in Chemoreception. Vol 1. Communication by zur Gewinnung olfaktorischer EEGAntworten des wachen
Chemical Signals. Appleton-Century-Crofts, New York. Menschen (objektive Olfaktometrie). Z. EEM-EMG
Jones, D. T., and Reed, R. R. (1989). Golf: an olfactory neuron 9:135–145.
specific-G protein involved in odorant signal transduction. Komisaruk, B. R. and Beyer, C. (1972). Responses of dien-
Science 244:790–795. cephalic neurons to olfactory.bulb stimulation, odor and
Jones, F. N., and Jones, M. H. (1953). Modern theories of olfac- arousal. Brain Res. 36:153–170.
tion: a critical review. J. Psychol. 36:207–241. Korshing, S.I. (2002). Odor maps and odor images. Curr. Opin.
Jurisch, A. (1922). Studien über die Papillae vallatae beim Men- Neurobiol. 12:387–392.
schen. Z. Anat. Entwicklungsgesch. 66:1–149. Krames, L. (1970). Responses of female rats to the individual
Kaada, B. R. (1951). Somato-motor, autonomic and electroacor- body odors of male rats. Psychonomic Sci 20: 274.
ticographic responses to electrical stimulation of “rhinen- Krause, W. (1876). Allgemeine und microscopische Anatomie. In
cephalic” and other structures in primates, cat and dog. Acta Handbuch der menschlichen Anatomie. Bd. 1. C. F. Th.
Physiol. Scand. 24(Suppl 83). Krause, Hrsg., Hannover.
Kaissling, K. E. (2001). Olfactory perireceptor and receptor Krautwurst, D., Yau, K. W., Reed, R. R. (1998). Identification of
events in moths: A kinetic model. Chem. Senses 26:125–150. ligands for olfactory receptors by functional expression of a
Kaissling. K. E., and Renner, M. (1968). Antennale Rezeptoren receptor library. Cell 95:917–26.
für Queen Substance und Sterzelduft bei der Honigbiene. Z. Kubie, J. L., and Moulton, D. G. (1980). Odorant specific patterns
Vergleich. Physiol. 59:357–361. of differential sensitivity inherent in the salamander olfactory
Karlson, P., and Lüscher, M. (1959). “Phermones”: a new term epithelium. Soc. Neurosci. Abstr. 6:243.
for a class of biologically active substances. Nature Kurahashi, T., and Shibuya, T. (1987). The odor responses and
183:55–56. odor-induced current in the solitary olfactory receptor cells
Kauer, J. S. (1988). Real-time imaging of evoked activity in local isolated from newts. Chem. Senses 12:508.
circuits of the salamander olfactory bulb. Nature Kurihara, K. (1986). Transduction mechanisms in chemorxep-
331:166–168. tion. Comp. Biochem. Physiol. 85A:1–22.
Kauer, J. S., and Moulton, D. G. (1974). Response patterns of ol- Kurihara, K. (1990). Molecular mechanisms of reception and
factory bulb neurons using odor stimulation and small nasal transduction in olfaction and taste. Jpn. J. Physiol. 40:
areas in the salamander. J. Physiol. (Lond.) 243:717–737. 305–324.
Keenan, C. M., Kelly, D. P., and Bogdanffy, M. S. (1990). De- Kurihara, K., and Koyama, N. (1972). High activity of adenyl cy-
generation and recovery of rat olfactory epithelium following clase in olfactory and gustatory organs. Biochem. Biophys.
inhalation of dibasic esters. Fund. App. Toxicol. 15:381–393. Res. Commun. 48:30–33.
Keesey, J. C. (1962). Olfactory preference by heterosexually Lancisi, J. M. (1717). De Noxiis Paludun Effluviis, Eorumque
naïve and experienced male rats for estrus and diestrus female Remediis. Rome.
urine. Unpublished M.A. thesis, San Jose State College. [Cited Landau, W. H., Freygang, W.H., Rowland, L.P., Sokoloff, L., and
in Cooper, J. B. (1971). Comparative Psychology. Ronald Kety, S.S. (1955). The local circulation of the living brain: val-
Press, New York, p. 321]. ues in the unanesthetized and anesthetized cat. Trans. Am.
Kennedy, C., DesRosiers, M. H., Jehl, J. W., Reivich, M., Sharp, Neurol. Assoc. 80:125–129.
F., and Sokoloff, R. (1975). Mapping of functional neural Lassen, N. A., Hoedt-Rasmussen, K., Sorensen, S.C., Skinhoj, E.,
pathways by autoradiographic survey of local metabolic rate Cronquist, B., Bodforss, E., and Ingvar, D.H. (1963). Regional
with 14C deoxyglucose. Science 187:850–853. cerebral blood flow in man determined by Krypton-85. Neu-
Kenneth, J. H. (1928). A note on a forgotten sixteenth century dis- rology 13:719–727.
putation on smell. J. Laryngol. Otol. 43:103–104. Lauterbur, P. (1973). Image formation by induced local interac-
Kiesow, F. (1894). Beiträge zur Physiologischen Psychologie des tions: Examples employing nuclear magnetic resonance. Na-
Geschmackssinnes. Philos. Stud. 10:523–561. ture 242:190–191.
Kety, S. S. (1960). Measurement of local blood flow by the ex- Laveran, C. L. A. (1881). Nature parasitaire ties accidents de
change of an inert, diffusible substance. In: H. D. Bruner (Ed.), l’irnpaludisme: description dun Nouvean parasite troue dans
Methods in Medical Research. Year Book, Chicago, le sang ties malades atteints de fievre palustre. Bailliere, Paris.
VIII:228–236. Lederer, F. L. (1959). The problem of nasal polyps. J. Allergy.
Kimura, K., and Beidler, L. M. (1961). Microelectrode study of 30:420–432.
taste receptors of rat and hanister. J. Cell Comp. Physiol. Lee, K., Wells, R. G., and Reed, R. R. (1987). Isolation of an ol-
58:131–140 factory cDNA; similarity to retinolbinding protein suggests a
Kinnamon, S. C., and Cummings, T. A. (1992). Chemosensory role in olfaction. Science 235:1053–1056.
transduction mechanisms in taste. Annu. Rev. Physiol. Lehman, C. D., Bartoshuk, L. M., Catalanotto, F.C., Kveton, J.F.
54:715–731. and Lowlicht, R.A. (1995). Effect of anesthesia of the chorda
Kinnamon, S. C., Dionne, V.E., and Beam, K. G. (1988). Apical tympani nereve on taste perception in humans. Physiol. Behav.
location of K channels in taste cells provides a basis for sour 57:943–951.
taste transduction. Proc. Natl. Acad. Sci. USA 85:7023–7027. Leinders-Zufall, T., Shepherd, G. M., and Zufall, F. (1996). Mod-
Kirkes, W. S. (1849). Manual of Physiology. Lea and Blanchard, ulation by cyclic GMP of the odour sensitivity of vertebrate ol-
Philadelphia, pp. 400–401. factory receptor cells. Proc. Roy. Soc. Lond. B 263:803–811.
Introduction and Historical Perspective xxxix

LeMagnen, J. (1952). Les phènomenes olfacto-sexuells chez le Mackay-Sim, A., and Kittle, P. W. (1991). On the life span of ol-
rat blanc. Arch. Sci. Physiol. 6:295–332. factory receptor neurons. Eur. J. Neurosci. 3:209–215.
Lemery, N. (1697). Pharmacopee universelle. [Cited in Corbin, Mackay-Sim, A., Shaman, P., and Moulton, D. G. (1982). Topo-
A. (1986). The Foul and the Fragrant. Harvard University graphic coding of olfactory quality: odorant specific patterns
Press, Cambridge, MA. of epithelial responsivity in the salamander. J. Neurophysiol.
Leon, M. (1983). Chemical communication in mother-young inter- 48:584–596.
actions. In: J. G. Vandenbergh (Ed.), Pheromones and Repro- Magendie, F. (1824). Le nerf olfactif, est-il l’organe de l’odorat?
duction in Mammals. Academic Press, New York, pp. 39–77. Experiences sur cette question. J. Physiol. Exp. Pathol.
Leydig, F. (1851). Uber die Haut einiger Süwasserfische. Z. 4:169–176.
wiss. Zool. 3:1–12. Mahanthappa, N. K., and Schwarting, G. A. (1993). Peptide
Leveteau, J., and MacLeod, P. (1966). Olfactory discrimination in growth factor control of olfactory neurogenesis and neuron
the rabbit olfactory glomerulus. Science 153:175–176. survival in vitro: roles of EGF and TGF-bs. Neuron
Li, Y., Field, P., and Raisman, G. (1998). Regeneration of adult 10:293–305.
corticospinal axons induced by transplanted olfactory en- Mainardi, D., Marsan, M., and Pasquali, A. (1965). Causation of
sheathing cells. J. Neurosci. 18:10514–10524. sexual preferences of the house mouse. The behaviour of
Liman, E. R. (2001). Sex and the single neuron: pheromones ex- mice reared by parents whose odour was artifically altered.
cite. Trends Neurosci. 24:2–3. Atti Soc. Ital. Sci. Natur. Museo Civ. Stor. Natur. Milano
Lin, D. M., Wang, F., Lowe, G., Gold, G. H., Axel, R., Ngai, J., 104:325–338.
and Brunet, L. (2000). Formation of precise connections in the Malnic, B., Hirono, J., Sato, T., and Buck, L. B. (1999). Combi-
olfactory bulb occurs in the absence of odorant-evoked neu- natorial receptor codes for odors. Cell 96:713–723.
ronal activity. Neuron 26:69–80. Malpighi, M. (1664). Exercitatio epistolica de lingua. (1686; Jo.
Lindemann, B. (2001). Receptors and transduction in taste. Na- Alphonso Borellio). In: Malpighi, M. (ed), Opera omnia.
ture 413:219–225. R.Scott & G. Wells, London, pp. 13–20.
Linnaeus, C. (1765). Odores medicamentorum. Amoenit. Acad.
Marchlewska-Koj, A. (1983). Pregnancy blocking by
3:183–201.
pheromones. In: D. Müller-Schwarze and R. M. Silverstein
Lois, C., Garcia-Verdugo, J.M., and Alvarez-Buylla, A. (1966).
(Eds.), Chemical Signals in Vertebrates 3. Plenum Press, New
Chain migration of neuronal precursors. Science 271:
York, pp. 151–174.
978–981.
Martin, H. N. (1873). Notes on the structure of the olfactory mu-
Lorry, D. (1784/1785). Observations sur les parties volatiles et
cous membrane. J. Anat. Physiol. (Lond.) 8:39–44.
odorantes des médicaments tirés des substances vègètales et
Matsuda, T., and Doty, R.L. (1976). Regional taste sensitivity to
animales. Hist. Mèm. Soc. Roy. Mèd. 7:306–318
NaCl: relationship to subject age, tongue locus and area of
Loven, C. (1868). Beiträge zur Kenntnis vom Bau der
stimulation. Chem. Senses 20:283–290.
Geschmackswärzchen der Zunge. Arch. Mikrosk. Anat. 4:
96–109. McBurney, D. H., and Gent, J. F. (1979). On the nature of taste
Lowe, G., and Gold, G. H. (1991). The spatial distribution of qualities. Psychol. Bull. 86:151–167.
odorant-sensitive and odorant-induced currents in salamander McCaffrey, R. J., Duff, K., and Solomon, G. S. (2000). Olfactory
olfactory receptor cells. J. Physiol. (Lond.) 442:147–168. dysfunction discriminates probable Alzheimer’s dementia
Lowe, G., and Gold, G. H. (1993). Contribution of the ciliary from major depression: a cross-validation and extension. J.
cyclic nucleotide-gated conductance to olfactory transduction Neuropsychiary Clin. Neurosci. 12:29–33.
in the salamander. J. Physiol. 462:175–196. McCartney, W. (1968). Olfaction and Odours. Springer-Verlag,
Lowe, G., Nakamura, T., and Gold, G. H. (1989). Adenylate cy- Berlin.
clase mediates transduction for a wide variety of odorants. McLaughlin, T. (1971). Dirt: A Social History as Seen Through
Proc. Natl. Acad. Sci. USA 86:5641–5645. the Uses and Abuses of Dirt. Dorset Press, New York.
Lower, R. (1670). Dissertatio de origine catarrhi in qua ostenditur McLaughlin, S. K., McKinnon, P. J., and Margolskee, R. F.
non provenire a cerebro. In: Tractato de corde. J. Redmayne, (1992). a Gustducin: a taste cell specific G protein subunit
London, pp. 221–239. closely related to the a transducins. In R. L. Doty and D.
Lu, J., Feron, F., Ho, S., Mackay-Sim, A., and Waite, P. (2001). Müller-Schwarze (Eds.), Chemical Signals in Vertebrates 6.
Transplantation of nasal olfactory tissue promotes partial re- Plenum Press, New York, pp. 9–14.
covery in paraplegic rates. Brain Res. 889:344–357. Meiselman, H. L., and Rivlin, R. S. (Eds.) (1986). Clinical Mea-
Luskin, M.B. (1993). Restricted proliferation and migration of surement of Taste and Smell. Macmillian, New York.
postnatally generated neurons derived from the forebrain sub- Menco, B. P. M. (1991). Ultrastructual localization of the trans-
ventricular zone. Neuron 11:173–189. duction apparatus in the rat’s olfactory epithelium. Chem.
MacDonald, K. P. A., Murrell, W. G., Bartlett, P. F., Bushell, G. Senses 15:555.
R., and Mackay-Sim, A. (1996). FGF2 promotes neuronal dif- Menco, B. P. M. (1977). A qualitataive and quantitative investi-
ferentiation in explant cultures of adult and embryonic mouse gation of olfactory and nasal respiratory mucosal surfaces of
olfactory epithelium. J. Neurosci. Res. 44:27–39. cow and sheep based on various ultrastructural and biochemi-
Mackay-Sim, A., and Chuah, M. (2000). Neurotrophic growth cal techniques. Comm. Agricult. Univ., Wageningen 77–13:
factors in the primary olfactory pathway. Prog. Neurobiol. 1–157.
62:527–559. Meredith, M. (1983). Sensory physiology of pheromone commu-
xl Doty

nication. In: J. G. Vandenbergh (Ed.), Pheromones and mammals—two aspects; proliferation of cells in the olfactory
Reproduction in Mammals. Academic Press, New York, epithelium and sensitivity to odours. In: G. E. W. Wolstenholme
pp. 199–252. and J. Knight (Eds.)., Ciba Foundation Symposium on Taste
Meredith, M. (1986). Vomeronasal organ removal before sexual and Smell in Vertebrates. Churchill, London, pp. 227–250.
experience impairs male hamster mating behavior. Physiol. Moulton, D. G., Turk, A., and Johnson, J. W., Jr. (1975). Methods
Behav. 36: 737–743. in Olfactory Research. Academic Press, London.
Merkel, F. (1880). Über die Endigungen der sensiblen Nerven in Mozell, M. M. (1964). Olfactory discrimination: electrophysio-
der Haut der Wirbelthiere. Rostock, 214 pp. logical spatiotemporal basis. Science 143:1336–1337.
Miller, I. J. Jr., Ed. (1988). The Beidler Symposium on Taste and Mozell, M. M. (1966). The spatiotemporal analysis of odorants at
Smell. Book Service Associates, Winston-Salem, NC. the level of the olfactory receptor sheet. J. Gen. Physiol.
Miller, S. L., Mirza, N., and Doty, R. L. (2002). Electrogustomet- 50:25-i1.
ric thresholds: Relationship to anterior tongue locus, area of Mozell, M. M., Smith, B.P., Smith, P.E., Sullivan, R.L. Jr., and
stimulation, and number of fungiform papillae. Physiol. Swender, P. (1969). Human chemoreception in flavor identifi-
Behav., 75:753–757. cation. Arch. Otolaryngol. 90:131–137.
Minnich, D. E. (1921). An experimental study of the tarsal Murphy, M. R., and Schneider, G. E. (1970). Olfactory bulb re-
chemoreceptors of two nymphalid butterflies. J. Exp. Zool. 33: moval eliminates mating behavior in the male golden hamster.
173–203. Science 167:302–304.
Minnich, D. E. (1926). The chemical sensitity of the tarsi of certain Murrell, W., Bushell, G. R., Livesey, J., McGrath, J., MacDonald,
muscid flies. (Phormia regina Meigen, Phormia terrae-novae K. P. A., Bates, P. R., and Mackay-Sim, A. (1996) Neurogen-
R. D. and Lucilia sericate Meigen). Biol. Bull. 51:166–178. esis in adult human. Neuroreport 7:1189–1194.
Minnich, D. E. (1929). The chemical sensitivity of the legs of the Mykytowycz, R. (1970). The role of skin glands in mammalian
blow-fly, Calliphora vomitoria Linn., to various sugars. Z. communication. In: J. W. Johnston, Jr., D. G. Moulton, and A.
Vergl. Physiol. 11:1–55. Turk (Eds.). Advances in Chemoreception. Vol. 1. Communi-
cation by Chemical Signals. Appleton-Century-Crofts, New
Mombaerts, P. (1999). Molecular biology of odorant receptors in
York, pp. 327–360.
vertebrates. Annu. Rev. Neurosci. 22:487–509.
Mykytowycz, R. (1986). A quarter of a century of studies of
Moran, D. T., Jafek, B.W., Eller, P.M., and Rowley, J.C. III.
chemical communication in vertebrates. In: D. Duvall, D.
(1992). The ultrastructural histopathology of human olfactory
Müller-Schwarze, and R. M. Silverstein (Eds.)., Chemical Sig-
dysfunction. Microsc. Res. Tech. 23:103–110.
nals in Vertebrates 4. Plenum Press, New York, pp. 1–11.
Morfit, C. (1847). Chemistry Applied to the Manufacture of Soap
Nagahara, Y. (1940). Experimentelle Studien über die histologis-
and Candles. Carey and Hart, Philadelphia.
chen Veränderungen des Geruchssorgans nach der Olfacto-
Mori, K., and Takagi, S. F. (1978). An intracellular study of den-
riusdurchschneidung. Jpn. J. Med. Sci. Pt. V. Pathol. 6:
drodendritic inhibitory synapses on mitral cells in the rabbit
165–199.
olfactory bulb. J. Physiol. 719:589–604. Nakamura, T., and Gold, G. H. (1987). A cyclic nucleotide-gated
Morris, E. T. (1984). Fragrance: The Story of Perfume from conductance in olfactory receptor cilia. Nature 325:442–
Cleopatra to Chanel. Scribner, New York. 444.
Morrison, G. R. (1967). Behavioral response patterns to salt stim- Nelson, G., Hoon, M. A., Chandrashekar, J., Zhang, Y., Ryba, N.
uli in the rat. Can. J. Psychol. 21:141–152. J. P. and Zuker, C. S. (2001). Mammalian sweet taste recep-
Morrison, G. R., and Norrison, W. (1966). Taste detection in the tors. Cell 106:381–390.
rat. Can. J. Psychol. 20:208–217. Newman, M., Féron, F., and Mackay-Sim, A. (2000). Growth fac-
Morton, R. (1697). Opera Medica. Sumptibus Anisson and Po- tor control of olfactory neurogenesis. Neuroscience 99:
suel, Lugduni. 343–350.
Mosso, A. (1881). Uber den Kreislauf des Blutes im men- Nicoll, R. A. (1971). Recurrent excitation of secondary olfactory
schlichen Gehirn. Leipzig: Verlag von Veit. neurons: a possible mechanism for signal amplification. Sci-
Mosso, A. (1884). La temperatura del cervèllo. Milan. ence 171:825–825.
Motokizawa, F. (1974). Olfactory input to the thalamus: electro- Norgren, R. (1970). Gustatory responses in the hypothalamus.
physiological evidence. Brain Res. 67:334–337. Brain Res. 21:63–77.
Moulton, D. G. (1960). Studies in olfactory acuity. III. Relative Ogawa, S., Lee, T.M., Kay, A.R., and Tank, D.W. (1990). Brain
detectability of n-alphatic acetates by the rat. Q. J. Exp. Psy- magnetic resonance imaging with contrast dependent on blood
chol. 12:203–213. oxygenation. Proc. Nat. Acad. Sci. USA 87:9868–9872.
Moulton, D. G. (1976). Spatial patterning of response to odors in Ogle, W. (1870). Anosmia (or cases illustrating the physiology
the peripheral olfactory system. Physiol. Rev. 56:578–593. and pathology of the sense of smell). Med. Chir. Trans.
Moulton, D. G., and Beidler, L. M. (1967). Structure and function 35:263–290.
in the peripheral olfactory system. Physiol. Rev. 47:1–52. Öhrwall, H. (1891). Untersuchungen über den Geschmackssinn.
Moulton, D. G., and Eayrs, J. T. (1960). Studies in olfactory acu- Skand. Arch. Physiol. 2:1–69.
ity. II. Relative detectability of n-aliphatic alcohols by the rat. Öpalski, A. (1934). Über locale Unterschiede im Bau der Ven-
Q. J. Exp. Psychol. 12:99–109. trikelwände beim Menschen. Z. Ges. Neurol. Psychiat.
Moulton, D. G., Celebi, G., and Fink, R. P. (1970). Olfaction in 149:221–254.
Introduction and Historical Perspective xli

O’Rourke, N. A. (1996). Neuronal chain gangs: homotypic con- Piesse, G. W. H. (1879). Art of Perfumery. Longmans, London.
tacts support migration into the olfactory bulb. Neuron Pixley, S. K. (1992a). CNS glial cells support in vitro survival, di-
16:1061–1064. vision, and differentiation of dissociated olfactory neuronal
Ottoson, D. (1956). Analysis of the electrical activity of the ol- progenitor cells. Neuron 8:1191–1204.
factory epithelium. Acta Physiol. Scand. 35(Suppl 122):1–83. Pixley, S. K. (1992b) The olfactory nerve contains two popula-
Ottoson, D. (1963). Some aspects of the function of the olfactory tions of glia, identified both in vivo and in vitro. Glia
system. Pharmacol. Rev. 15:1–42. 5:269–284.
Pace, U., and Lancet, D. (1986). Olfactory GTP-binding protein: Plattig, K. H. (1968/1969). Über den elektrischen Geschmack,
signal transducing polypeptide of vertebrate chemosensory Reizstärkeabhängige evozierte Hirnpotentiale nach elek-
neurons. Proc. Natl. Acad. Sci. USA 83:4947–4951. trischer Reizung der Zunge beim Menschen. Habil-Schr. Er-
Pace, U., Hanski, E., Salomon, Y., and Lancet, D. (1985). Odor- langen 1968. Z. Biol. 116:161–211.
ant-sensitive adenylate cyclase may mediate olfactory recep- Powers, J. B., and Winans, S. S. (1973). Sexual behavior in pe-
tion. Nature 316:255–258. riphally anosmic male hamsters. Physiol. Behav. 10:361–368.
Palladius, R. T. A. (4th century A.D.), DeRostica, Lib. 1. Powers, J. B., and Winans, S. S. (1975). Vomeronasal organ: crit-
Pangborn, R. M., and Trabue, I.M. (1967). Bibliography on the ical role in mediating sexual behavior of the male hamster.
sense of taste (1566–1966). In: M.R. Kare and O. Maller Science 187:961–963.
(Eds.), The Chemical Senses and Nutrition. Baltimore, Johns Proetz, A. W. (1924). Exact olfactometry. Ann. Otol. Rhinol.
Hopkins Press, pp. 355–471. Laryngol. 33:275–278.
Parker, G. H. (1922). Smell, Taste, and Allied Senses in the Ver- Rall, W., Shepherd, G. M., Reese. T. S., and Brightman, M. W.
tebrates. Lippincott, Philadelphia. (1966). Dendrodendritic synaptic pathway for inhibition in the
Parmentier, M., Libert, F., Schurmans, S., Schiffmann, S., Lefort, olfactory bulb. Exp. Neurol. 14:44–56.
A., Eggerickx, D., Ledent, C., Mollereau, C., Gerard, C., Per- Ramon-Cueto, A., Cordero, M., Santos-Benito, F. and Avila, J.
ret, J., Grootegoed, A., and Vassart, G. (1992). Expression of (2000). Functional recovery of paraplegic rats and motor axon
members of the putative olfactory receptor gene family in regeneration in their spinal cords by olfactory ensheathing
mammalian germ cells. Nature 355:453–455. glia. Neuron 25:425–435.
Passy, J. (1892). Sur les minimums perceptubles de quelques Ramon-Cueto, A., and Nieto-Sampedro, M. (1992). Glial cells
odeurs. Comp. Rend. Soc. Biol. (Paris) 44:84–88, 239–243. from adult rat olfactory bulb: immunocytochemical properties
Pauling, L., and Coryell, C.D. (1936). The magnetic properties of pure cultures of ensheathing cells. Neuroscience 47:
and structure of hemoglobin, oxyhemoglobin and caron- 213–220.
monoxyhemoglobin. Proc. Nat. Acad. Sci. USA 89:5951– Ramon-Cueto, A., and Nieto-Sampedro, M. (1994). Regenera-
5955. tion into the spinal cord of transected dorsal root axons is
Pavlov, I. P. (1927). Conditioned Reflexes. Oxford University promoted by ensheathing glia transplants. Exp. Neurol.
Press, Oxford. 27:232–244.
Peet, M. M., Echols, D. H., and Richter, H. J. (1937). The chem- Raming, K., Krieger, J., Strotmann, J., Boekoff, I., Kubick, S.,
ical prophylaxis for poliomyelitus: the technique of applying Baumstark, C., and Breer, H. (1993). Cloning and expression
zinc sulfate intranasally. JAMA 108:2184–2187. of odorant receptors. Nature 361:353–356.
Pelosi, P. (2001). The role of perireceptor events in olfaction. Remy, N. (1595). Demonolatry. (Translation to English by E. A.
Cell. Mol. Life Sci. 58:503–509. Ashwin.) John Rodker, London, 1930.
Pevsner, J., Hwang, P. M., Sklar, P. B., Venable, J. C., and Sny- Reivich, M., Sano, N., and Sokoloff, L. (1971). Development of
der, S. H. (1988a). Odorant-binding protein and its mRNA are an autoradiographic method for the determination of regional
localized to lateral nasal gland implying a carrier function. glucose consumption. In: R.W. Ross-Russell (Ed.), Brain and
Proc. Natl. Acad. Sci. USA 85:2382–2387. Blood Flow. London: Pitman, 397–400.
Pevsner, J., Reed, R. R., Feinstein, P. G., and Snyder, S. H. Reivich, M., Kuhl, D., Wolf, A., Greenberg, J., Phelps, M., Ido,
(1988b). Molecular cloning of odorantbinding protein: mem- T., Casella, V., Hoffman, E., Alvai, A., and Sokoloff, L.
ber of a ligand carrier family. Science 241:336–339. (1979). The [18F]fluorodeoxyglucose method for the measure-
Pfaff, D. W. (1985). Taste, Olfaction, and the Central Nervous ment of local cerebral glucose utilization in man. Circ. Res.
System: A Festschrift in Honor of Carl Pfaffmann. Rockefeller 44:127–137.
University Press, New York. Renqvist, Y. (1919). Über den Geschmack. Scand. Arch. Physiol.
Pfaffmann, C. (1941). Gustatory afferent impulses. J. Cell. Comp. 38:7–201.
Physiol. 17:243–258. Restrepo, D., Miyamoto, T., Bryant, B. P., and Teeter, J. H. (1990).
Pfaffman, C. (1959). Gustatory nerve impulses in rat, cat and rab- Odor stimuli trigger influx of calcium into olfactory neurons of
bit. J. Neurophysiol. 23:429–440. the channel catfish. Science 249:1166–1168.
Pfaffmann, C., Goff, W. R., and Bare, J. K. (1958). An olfac- Rhein, L. D., and Cagan, R. H. (1980). Biochemical studies of ol-
tometer for the rat. Science 128: 1007–1008. faction: isolation, characterization, and odorant binding activity
Pfaffmann, C., Erickson, R., Frommer, G., and Halpern, B. of cilia from rainbow trout olfactory rosettes. Proc. Natl. Acad.
(1961). Gustatory discharges in the rat medulla and thalamus. Sci. USA 77:4412–4416.
In W. A. Rosenblith (Ed.), Sensory Communication. MIT Richter, C. P. (1936). Increased salt appetitie in adrenalectomized
Press, Cambridge, MA, pp. 455–473. rats. Amer. J. Physiol. 115:115–161.
xlii Doty

Richter, C. P. (1939). Salt taste thresholds of normal and ad- coryza, haemorrhagia Narium, polypo, steruntatione, a mis-
ranalectomized rats. Endocrinology 24:367–371. sione odoratus; Mebius et Schumacher, Wittebergae
Richter, C. P., and Campbell, K. H. (1940). Taste thresholds and Schneider, D. (1955). Mikro-Elektroden registrieren die elek-
taste preference of rats for five common sugars. J. Nutr. trischen Impulse einzelner Sinneszellen der Schmetterlingsan-
20:31–46. tenne. Ind.-Elektron. 3:3–7.
Riti, L. (1974). The Unknown Leonardo. McGraw-Hill, New York. Schneider, D. (1957a). Electrophysiological investigation on the
Rochefort, C., Gheusi, G., Vincet, J-D., and Liedo, P-M. (2002). antennal receptors of the silk moth during chemical and me-
Enriched odor exposure increases the number of newborn neu- chanical stimulation. Experientia 13:89–91.
rons in the adult olfactory bulb and improves odor memory. Schneider, D. (1957b). Elektrophysiologische Untersuchungen
J. Neurosci. 22:2679–2689. von Chemo- and Menschanorezeptoren der Antenne des
Roper, S. D. (1989). The cell biology of vertebrate taste receptors. Seidenspinners Bombyx mori L. Z. Vergl. Physiol. 40:8–41.
Annu. Rev. Neurosci. 12:329–353. Schultz, E. W. (1960). Repair of olfactory mucosa, with special
Ross, R. (1923). Memoirs. E. P. Dutton, New York. reference to regeneration of olfactory cells (sensory neurones).
Rothkrug, L. (1981). The “odour of sanctity,” and the Hebrew ori- Am. J. Pathol. 37:1–19.
gins of Christian relic veneration. Hist. Reflect. Refex. Hist. Schultz, E. W., and Gebhardt, L. P. (1936). Prevention of in-
8:95–137. tranasally inoculated poliomyelitis in Monkeys by previous
Roy, C. S., and Sherrington, C.S. (1890). On the regulation of the intranasal irrigation with chemical agents. Proc. Soc. Exp.
blood supply of the brain. J. Physiol. (London) 11:85–108. Biol. Med. 34:133–135.
Rydberg, E. (1932). Cerebral injury in newborn children conse- Schultz, E. W., and Gebhardt, L. P. (1937). Zinc sulfate prophy-
quent on birth trauma. Acta Path. Microbiol. Scand, Suppl laxis in poliomyelitis. JAMA 108:2182–2184.
10:1–247. Schultze, M. (1856). Über die Endigungsweise des Geruchsner-
Ruiz-Avila, L., Wong, G.T., Damak, S., and Margolskee, R.F. ven und der Epithelialgebilde der Nasenschleimhaut. Monats-
(2001). Dominant loss of responsiveness of sweet and bitter ber. Deutsche Akad. Wiss. (Berlin) 21:505–515.
compounds caused by a single mutation in a-gustducin. Proc. Schultze, M. (1863). Untersuchungen über den Bau der Nasen-
Nat. Acad. Sci. USA 98:8868–8873. schleimhaut, namentlich die Structur und Endigungsweise der
Russo, A. F., and Koshland, D. E., Jr. (1983). Separation of signal Geruchsnerven bei dem Menschen and den Wirbelthieren.
transduction and adaptation functions of the asparate receptor Abh. Naturforsch. Ges. Halle 7:1–100.
in bacterial sensing. Science 220:1016–1020. Schwalbe (1868). Über die Geschmacksorgane der Säugethiere
Saveliev, N. A. (1892). Fisiologiya nervi olfactorii Istoricheskiya i und des Menschen. Arch. mikr. Anat. 4:154–187.
Eksperimentalniya Izsliedovaniya. Univ. Tipograf, Noskva. Scott, J. W., and Pfaffmann, C. P. (1972). Characteristics of re-
Scarp, A. (1785). Anatomicarum annotationum liber secundus. De sponses of lateral hypothalamic neurons to stimulation of the
organo olfactus praecipuo, deque nervis nasalibus interioribus olfactory bulb. Brain Res. 48:251–264.
e pari quinto nervorum cerebri. Regis, Ticini. Sharp, F. R., Kauer, J. S., and Shepherd, G. M. (1975). Local sites
Scarpa, A. (1789). Anatomicae disquisitiones de auditu et olfacto. of activity-related glucose metabolism in rat olfactory bulb dur-
Regis, Ticini. ing olfactory stimulation. Brain Res. 98:596–600.
Schiff, J. M. (1860). Der erste Hirnnerv ist der Geruchsnerv. Un- Sharp, F. R., Kauer, J. S., and Shepherd, G. M. (1977). Laminar
tersuchungen zur Naturlehre des Menschen und der Thiere 6: analysis of 2-deoxyglucose uptake in olfactory bulb and olfac-
254–267. tory cortex of rabbit and rat. J. Neurophysiol. 40:800–813.
Schiffman, S. S. (1983). Taste and smell in disease. N. Engl. J. Shaw. A. (1833). Narrative of the Discoveries of Sir Charles Bell
Med. 308:1275–1279, 1337–1343. in the Nervous System. Longman, Orme, Brown, Green and
Schiffman, S. S., Zervakis, J., Suggs, M. S., Shaio, E., and Sattely- Longmans, London.
Miller, E.A. (1999a). Effect of medications on taste: example of Shepherd, G. M. (1971). Physiological evidence for dendroden-
amitriptyline HCl. Physiol. Behav. 66:183–191. dritic interactions in the rabbit’s olfactory glomerulus. Brain
Schiffman, S. S., Zervakis, J., Shaio, E., and Heald, A.E. (1999b). Res. 32:212–217.
Effect of the nucleoside analogs zidovudine, didanosine, stavu- Shepherd, G. M. (1972). Synaptic organization of the mammalian
dine, and lamivudine on the sense of taste. Nutrition olfactory bulb. Physiol. Rev. 52: 864–917.
15:854–859. Shirley, S. G., Robinson, C. J., Dickinson, K., Aujla, R., and
Schiffman, S. S., Zervakis, J., Suggs, M. S., Budd, K.C., and Iuga, Dodd, G. H. (1986). Olfactory adenylate cyclase of the rat:
L. (2000). Effect of tricyclic antidepressants on taste responses stimulation by odorants and inhibition by Ca2. Biochem. J.
in humans and gerbils. Pharmacol. Biochem. Behav. 65: 240:605–607.
599–609. Sidky, M. (1877). Recherches anatomo-microscopiques sur la
Schmale, H., Holtgreve-Grez, H., and Christiansen, H. (1990). muqueuse olfactive. Paris, 74 pp.
Possible role for salivary gland protein in taste reception indi- Simon, S.A. and Nicolelis, M. A. L. (2002). Methods in
cated by homology to lipophilic-ligand carrier proteins. Na- Chemosensory Research. Boca Raton: CRC Press.
ture 343:366–369. Skinner, B. F. (1938). The Behavior of Organisms. Appeton-Cen-
Schneider, C. V. (1655). Liber de osse cribriformi, et sensu ac tury, New York.
organo odoratus, et morbis ad utrumque spectantibus, de Sklar, P. B., Anholt, R. R. H., and Snyder, S. H. (1986). The odor-
Introduction and Historical Perspective xliii

ant-sensitive adenylate cyclase of olfactory receptor cells. J. mentia from major depression. J. Neuropsychiat. Clin. Neu-
Biol. Chem. 261:15538–15543. rosci. 10:64–67.
Slotnick. B. M. (1990). Olfactory perception. In: M. A. Berkeley Spector, A. C., Andrews-Labenski, J., and Letterio, F. C. (1990).
and W. C. Stebbins (Eds.), Comparative Perception, Vol. I. A new gustometer for psychophysical taste testing in the rat.
Basic Mechanisms. Wiley, New York, pp. 155–214. Physiol. Behav. 47:795–803.
Slotnick, B. M., and Katz, H. (1974). Olfactory learning-set in Stevens, D. A. (1975). Laboratory methods for obtaining olfac-
rats. Science 185:796–798. tory discrimination in rodents. In: D. G. Moulton. A. Turk, and
Slotnick, B.M., and Ptak. J. E. (1977). Olfactory intensity-differ- J. W. Johnson, Jr. (Eds.), Methods in Olfactory Research. Aca-
ence thresholds in rats and humans. Physiol. Behav. 19: demic Press, London, pp. 375–394.
795–802. Stewart, W. B., Kauer, J. S., and Shepherd, G. M. (1979). Func-
Small, D. M., Jones-Gotman, M., Zatorre, R. J., Petrides, M., and tional organization of rat olfactory bulb analysed by the 2-de-
Evans, A. C. (1997). A role for the right anterior temporal lobe oxyglucose method. J. Comp. Neurol. 185:715–734.
in taste quality recognition. J. Neurosci. 17:5136–5142. Stratton, G. M. (1917). Theophrastus and the Greek Physiologi-
Smith. J. (1970). Conditioned suppression as a animal psychophys- cal Psychology before Aristotle. George Alle & Unwin,
ical technique. In: W. C. Stebbins (Ed.), Animal Psychophysics. London.
Appleton-Century-Crofts, New York, pp. 125–159. Streim, B. J., Naim, M., and Lindemann, B. (1991). Generation of
Smith, D. V. (1990) Taste and smell dysfunction. In: M. M. Pa- cyclic AMP in taste buds of the rat circumvallate papilla in re-
parella, D. A. Shumrick, J. L. Gluckman, std W. L. Meyerhoff sponse to sucrose. Cell Physiol. Biochem. 1:46–54.
(Eds.), Otolaryngology. Vol. III. Head and Neck, 3rd ed. Saun- Striem, B. J., Pace, U., Zehavi, U., Naim, M., and Lancet, D.
ders, Philadelphia, pp. 1911–1934. (1989). Sweet tastants stimulate adenylate cyclase coupled to
Smith, T. D., and Bhatnagar, K. P. (2000) The human GTP-binding protein in rat tongue membranes. Biochem. J.
vomeronasal organ. Part II. prenatal development. J. Anat. 260:121–126.
197: 421–436. Stürckow, B. (1970). Responses of olfactory and gustatory recep-
Smith, D. V., St. John, S. J., and Boughter, J. D. Jr. (2000). Neu- tor cells in insects. In: J. W. Johnston, Jr., D. G. Moulton, and
ronal cell types and taste quality coding. Physiol. Behav. A. Turk (Eds.), Advances in Chemoreception. Vol. I. Commu-
69:77–85. nication by Chemical Signals. Appleton-Century-Crofts, New
Smotherman, W. P. (1982). Odor aversion learning by the rat fe- York, pp. 107–159.
tus. Physiol. Behav. 29:769–777. Sulzer, M. (1752). Resherches sur l’origine des sentiments
Sobel, N., Prabhakaran, V., Desmond, J. E., Glover, G. H., agreables et desagreables. Troisieme partie: Des plaisirs des
Goode, R. L., Sullivan, E. V., and Gabrieli, J. D. (1998a). sens. Historic de l’academie des sciences et belle lettrns
Sniffing and smelling: separate subsystems in the human ol- de Berlin. Cited in Bujas, Z., Electrical Taste. In: L. M. Bei-
factory cortex. Nature, 392:282–286. dler (Ed.), Handbook of Sensory Physiology. Vol. IV. Chemi-
Sobel, N., Prabhakaran, V., Hartley, C. A., Desmond, J. E., Zhao, cal Senses. Sect. 2. Taste. Springer-Verlag, New York,
Z., Glover, G. H., Gabrieli, J. D., Sullivan, E. V. (1998b). pp. 180–199.
Odorant-induced and sniff-induced activation in the cerebel- Summers, M. (1926). The History of Witchcraft and Demonology.
lum of the human. J. Neurosci. 18:8990–9001. Knopf, New York.
Sokoloff, L. (1961). Local cerebral circulation at rest and during Takagi, S. F. (1989). Human Olfaction. University of Tokyo
altered cerebral activity induced by anesthesia or visual stim- Press, Tokyo.
ulation. In: S.S. Kety and J. Elkes (Ed.), The Regional Chem- Tanabe, T., Iino, M., Ooshima, Y., and Takagi, S. F. (1973). The
istry, Physiology, and Pharmacology of the Nervous System. olfactory center in the frontal lobe of the monkey. J. Physiol.
Pergamon Press, Oxford, pp. 107–117. Soc. Jpn. 35:550.
Sokoloff, L. (1981). Localization of functional activity in the cen- Tanabe, T., Yarita, H., Iino, M., Ooshima, Y., and Takagi, S. F.
tral nervous system by measurement of glucose utilization (1975). An olfactory projection area in orbitofrontal cortex of
with radioactive deoxyglucose. J. Cereb. Blood Flow the monkey. J. Neurophysiol. 38:1269–1283.
Metabol. 1:7–36. Tanapat, P., Hastings, N. B., Reeves, A.J., and Gould, E. (1999).
Sokoloff, L. (1982). The radioactive deoxyglucose method. The- Estrogen stimulates a transient increase in the number of new
ory, procedure, and applications for the measurement of local neurons in the dentate gyrus of the adult female rat. J. Neu-
glucose utilization in the central nervous system. In: B. W. rosci. 19:5792–5801.
Agranoff and M. H. Aprison (Eds.), Advances in Neurochem- Tateda, H., and Beidler, L. M. (1964). The receptor potential of
istry. Plenum Press, New York, pp. 1–82. the taste cell of the rat. J. Gen. Physiol. 47:479–486.
Sokoloff, L., Reivich, M., Kennedy, C., Des Rosiers, M. H., Pat- Teichner, W. H. (1966). A method for studying olfaction in the
lak, C. S., et al. (1977). The [14C]deoxyglucose method for the unrestrained rat. J. Psychol. 63:291–297.
meaurement of local cerebral glucose utilization: theory, pro- Teicher, M. H., Stewart, W. B., Kauer, J. S., and Shepherd, G. M.
cedure, and normal values in the conscious and anesthetized (1980). Suckling pheromone stimulation of a modified
albino rat. J. Neurochem. 28:897–916. glomerular region in the developing rat olfactory bulb re-
Solomon, G. S., Petrie, W. M., Hart, J. R., and Brackin, H. B. Jr. vealed by the 2-deoxyglucose method. Brain Res. 194:
(1998). Olfactory dysfunction discriminates Alzheimer’s de- 530–535.
xliv Doty

Ter-Pogossian, M.M., Phelps, M.E., Hoffman, E.J. and Mullani, Verlag, New York, pp. 180–199. [Original: Volta A, 1816 Sull
N.A. (1975). A positron-emission tomography for nuclear elettricitá animale, In: Collezione dell’ Opere. G.Piatti,
imaging (PET). Radiology 114:89–98. Firenze, 268pp. Vol.2, Part1 (see pp. 55–118)].
Thomas, L. (1983). Late Night Thoughts on Mahler’s Ninth Sym- von Brunn, A. (1875). Untersuchungen über das Riechepithel.
phony. Viking Press, New York. Arch. Mikroskop. Anat. 11:468–477.
Thornhill, R. A. (1970). Cell division in the olfactory epithelium von Brunn, A. (1880). Weitere Untersuchungen über das
of the lamprey, Lampetra fluviatilis. Z. Zellforsch. 109: Riechepithel und sein Verhalten zum Nervus olfactorius.
147–157. Arch. Mikroskop. Anat. 17:141–151.
Tisdall, F. F., Brown, A., Defries, R. D., Ross, M. A., and Sellers, von Brunn, A. (1892). Beiträge zur mikroskopischen Anatomie
A. H. (1937). Nasal spraying as preventive of poliomylitis. der menschlichen Nasenhöhle. Arch. Mikroskop. Anat.
Can. Public Health J. 28:431–434. 39:632–651.
Tisdall, F. F., Brown, A., and Defries, R. D. (1938). Persistent von Frisch, K. (1919). Über den Geruchssinn den Biene and seine
anosmia following zinc sulfate nasal spraying. J. Pediatr. blütenbiologische Bedeutung. Zool. Jahrb. Physiol. 37:1–238.
13:60–62. von Frisch, K. (1921). Über den Sitz des Geruchssinnes bei In-
Todd, R. B., and Bowman, W. (1847). The Physiological sekten. Zool. Jahrb. Physiol. 38:449–516.
Anatomy and Physiology of Man, Vol. 11. Parker, London. von Frisch, K. (1922). Morphologische and biologische Unter-
Touhara, K., Sengoku, S., Inaki, K., Tsuboi, A. Hirono, J., Sato, suchungen der Putzapparate der Hymenopteren. Arch.
T., Sakano, H., and Haga, T. (1999). Functional identification Naturgeschichte Abt. A 88:1–63.
and reconstitution of an odorant receptor in single olfactory von Haller, A. (1756). Olfactus, Elementa physiologiae corporis
neurons. Proc. Natl. Acad. Sci. USA 96:4040–4045. humani. Liber XlV Tomus Quintus, Francisci Grasset, Lau-
Toulouse, E., and Vaschide, N. (1899). Mesure de l’odorat chez sanne, pp. 125–185.
1homme et chez la femme. Comp. Rend. Soc. Biol. 51: von Skramlik, E. (1926). Handbuch der Physiologie der niederen
381–383. Sinne. Georg Thieme-Verlag, Leipzig.
Vosshall, L.B., Wong, A. M., and Axel, R. (2000). An olfactory
Turetsky, B. I., Moberg, P. J., Yousem, D. M., Doty, R. L.,
sensory map in the fly brain. Cell 102:147–159.
Arnold, S. E., and Gur, R. E. (2000). Olfactory bulb volume is
Wekesa, K. S., and Anholt, R. R. H. (1999). Differential expres-
reduced in patients with schizophrenia. Am. J. Psychiatry
sion of G proteins in the mouse olfactory system. Brain Res.
157:828–30.
837:117–126.
Valentin, G. (1848). Lehrbuch der Physiologie des Menschen.
West, S. E., and Doty, R. L. (1995). Olfactory function in epilepsy
Braunschweig.
and temporal lobe resection lobectomy: A review. Epilepsia
Vandenbergh, J. G. (1969). Male odor accelerates female sexual
36:531–542.
maturation in mice. Endocrinology 84:658–660.
Wetzel, C. H., Behrendt, H. J., Gisselmann, G., Stortkuhl, K. F.,
Vandenbergh, J. G. (1983). Phermones and Reproduction in
Hovemann, B., Hatt, H. (2001). Functional expression and
Mammals. Academic Press, New York. characterization of a Drosophila odorant receptor in a heterol-
van der Lee, S., and Boot, L. M. (1955). Spontaneous pseudo- ogous cell system. Proc. Natl Acad. Sci. USA 98:9377–9380.
pregnancy in mice. Acta Physiol. Pharmacol. Neer. 4: Whitten, W. K. (1956). Modification of the oestrous cycle of the
442–444. mouse by external stimuli associated with the male. J. En-
Van Houten, J. (2000). Chemoreception in microorganisms. In: T. docrinol. 13:399–404.
E. Finger, W. L. Silver, and D. Restrepo (Eds.), The Neurobi- Whitten, W. K., Bronson, F. H., and Greenstein, J. A. (1968). Es-
ology of Taste and Smell, 2nd edition. Wiley-Liss, New York, trus-inducing phermone of male mice: transport by movement
pp. 11–40. of air. Science 161:584–585.
Varro, M. T. (116–28 B.C.). Rerum Rusticarium, Lib. 1. Winans, S. S., and Powers, J. B. (1974). Neonatal and two-stage
Verrill, A. H. (1940). Perfumes and Spices. L. C. Page, Boston. olfactory bulbectomy: effects on male sexual behavior. Behav.
Vesalius, Andrea. (1543). De Humani Corporis Fabrica. Biol. 10:461–471.
Basileae. Willis, T. (1681). Five Treatises. Dring, Harper, Leight and Mar-
Vintschgau, M. von, Hönigschmied, J. (1877). Nervus glossopha- tin, London.
ryngeus und Schmeckbecher. Arch. Ges. Physiol. 14:443–448. Wolozin, B., Lesch, P., Lebovics, R., and Sunderland, T. (1993).
Vivino, A. E. (1960). Perfumes and perfumery. In: The Encyclo- Olfactory neuroblasts from Alzheimer donors: studies on APP
pedia Americana. Americana, New York, pp. 577–582. processing and cell regulation. Biol. Psychiatry 34:824–838.
Vogt, R. G., Prestwich, G. D., and Lerner, M. R. (1991). Odor- Wolozin, B., Sunderland, T., Zheng, B.-B., Resau, J., Dufy, B.,
ant-binding-protein subfamilies associate with distinct Barker, J., Swerdlow, R., and Coon, H. (1992). Continuous
classes of olfactory receptor neurons in insects. J. Neurobiol. culture of neuronal cells from adult human olfactory epithe-
22:74–84. lium. J. Mol. Neurosci. 3:137–146.
Volta, A. (1792). Briefe über thierische Electricität. In: A. J. Oet- Wong, S.T., Trinh, K., Hacker, R., Chan, G.C.K., Lowe, G., Gag-
tingen (Ed.), Ostwald’s Klassiker der exakten Wissenschaten. gar, A., Xia, Z., Gold, G.H., and Storm, D.R. (2000). Disrup-
Engelmann, Leipzig, 1900. [Cited in Bujas, Z., (1971), Elec- tion of the type III adenylyl cyclase gene leads to peripheral
trical taste. In: L. M. Beidler (Ed.), Handbook of Sensory and behavioral anosmia in transgenic mice. Neuron
Physiology. Vol. IV. Chemical Senses. Part 2. Taste. Springer- 27:487–497.
Introduction and Historical Perspective xlv

Wright, J. (1914). A History of Laryngology and Rhinology. Lea E. T., Brammer, M. J., and Doty, R. L. (1997). Functional
& Febiger, Philadelphia. MR imaging during odor stimulation: preliminary data. Ra-
Wysocki, C. J. (1979). Neurobehavioral evidence for the involve- diology 204:833–838.
ment of the vomeronasal system in mammalian reproduction. Zatorre, R. J., Jones-Gotman, M., Evans, A. C., and Meyer, E.
Neurosci. Biobehav. Rev. 3:301–341. (1992). Functional localization and lateralization of human
Wysocki, C. J., Katz, Y., and Bernard, R. (1983). The male olfactory cortex. Nature 360:339–340.
vomeronasal organ mediates female-induced testosterone Zhao, H., Ivic, L., Otaki, J. M., Hashimoto, M., Mikoshiba, K.,
surges. Biol. Reprod. 28:917–922. and Firestein, S. (1998). Functional expression of a
Wysocki, C. J., Bean, N. J., and Beauchamp, G. K. (1986). The mammalian odorant receptor. Science 279(5348):237–
mammalian vomeronasal system: its role in learning and so- 122.
cial behaviors. In: D. Duvall, D. Müller-Schwarze, and R. M. Zippel, H. P. (1993). Historical aspects of research on the verte-
Silverstein, (Eds.), Chemical Signals in Vertebrates 4. New brate olfactory system. Naturwissenschaften 8D:65–76.
York, Plenum Press, pp. 471–485. Zotterman, Y. (1935). Action potentials in the glossopharyngeal
Yanagisawa, K., Bartoshuk, L. M., Catalanotto, F.A., Karrer, nerve and in the chorda tympani. Scand. Arch. Physiol.
T.A., and Kveton, J.F. (1998). Anesthesia of the chorda 72:73–77.
tympani nerve and taste phantoms. Physiol. Behav. 63: Zuniga, J. R., Davis, S. H., Englehardt, R. A., Miller, I. J., Jr.,
329–335. Schiffman, S. S., and Phillips, C. (1993). Taste performance
Young, J. M., and Trask, B. J. (2002). The sense of smell: ge- on the anterior human tongue varies with fungiform taste bud
nomics of vertebrate olfactory receptors. Hum. Molec. density. Chem. Senses 18:449–460.
Genet. 11:1153–1160. Zuckerkandl, E. (1888). Das Riechbündel des Ammonshornes.
Yousem, D. M., Williams, S. C., Howard, R. O., Andrew, C., Anat. Anz. 3:425–434.
Simmons, A., Allin, M., Geckle, RJ, Suskind, D., Bullmore, Zwaardemaker, H. (1925). L’Odorat. Doin, Paris.
1

Anatomy of the Human Nasal Passages

Dean M. Clerico
Valley ENT, Forty Fort, Pennsylvania, U.S.A.

Wyatt C. To and Donald C. Lanza

The Cleveland Clinic Foundation, Cleveland, Ohio, U.S.A.

I. INTRODUCTION 1993). However, it was not until the work of Max Schultze,
in 1856, that the first reasonably accurate description of
A. Historical Perspective
human olfactory receptor and supporting cells was
presented (Zippel, 1993). In 1892, von Brunn set out to
Scientific advancements leading to our current understanding
determine the precise extent of olfactory mucosa within
of olfaction have evolved considerably since the mid-
the nasal cavity. He studied four men ages 30–45 years,
seventeenth century. Until that time, it was generally held
who succumbed by decapitation. According to von Brunn,
that olfaction occurred via the direct access of odors to the
brain. In 1655, Schneider suggested that the sense of smell the position of the olfactory neuroepithelium is restricted
did not occur as a result of such air passage. Instead, he to the superior turbinate and nasal septum.
reported that the superior aspects of the nasal mucosa were In 1908, Read characterized, in a Ph.D. thesis, the olfac-
extremely sensitive, and he suspected this tissue was tory neuroepithelium in a 1-year-old child and a “30–40
responsible for olfaction. Moreover, he proposed that the year old man”. Read suggested that the distribution of the
secretions that drained through the nose were not produced olfactory mucosa was more extensive than that indicated
by the brain, as previously maintained. He believed that the by von Brunn but less than that depicted by Scarpa and
nasal membranes themselves produced these secretions others. Read also found that the olfactory nerve fibers ori-
(Zippel, 1993). ginating in the neuroepithelium ascended vertically
Subsequent study of olfaction in humans has focused through the cribriform plate without anastomoses or
upon the distribution of olfactory nerves and the nature of formation of a plexus, as had been previously held.
olfactory neuroepithelium. However, further advance- Variations in the distribution of olfactory neuroepithe-
ments in delineating the anatomy of olfactory tissue have lium are now well documented and seem to be related to a
been limited by two key issues: its relative inaccessabilty variety of factors. In 1941, Smith recorded that atrophy of
in living humans and its vulnerability to rapid decomposition the olfactory nerves in adults was very common. He
in the immediate post-mortem period. Thus, without postulated that such atrophy might help prevent the entry
suitable fixatives, histological evaluations prior to the of neurotropic viruses. In 1970, Naessen described a
twentieth century had to be performed quickly and macroscopic staining technique to visualize the extent of
immediately after death. olfactory neuroepithelium. He subsequently (1971)
In 1785, Antonio Scarpa described an extensive plexus explored the effects of aging upon the morphology of the
of olfactory nerve fibers within the human nose (Zippel, olfactory neuroepithelium. In 1984, Nakashima,

1
2 Clerico et al.

Kimmelman, and Snow evaluated 26 specimens from fetal

age through the ninth decade of life. These researchers
found olfactory neuroepithelial degeneration to be charac-
teristic of adult human tissue. The age-related changes
were seen in the cellular arrangement and topographic
distribution of the olfactory mucosa.
Whether or not age alone could be responsible for such
changes remains unclear. Alterations in the olfactory
epithelium seem to be subject to a wide variety of factors,
which include exposures to viral and bacterial infections
(see Chapters 3, 5, and 26), head injury (see Chapter 30),
neurodegenerative disorders (see Chapters 23 and 24), and
chemical exposures (see Chapters 25 and 27). Importantly,
metabolic changes may predispose an individual’s olfactory
system to greater susceptibility to damage from such envi-
ronmental factors (Rehn et al., 1981).
It should be emphasized that the ability to appreciate
smells goes beyond the proper function of the olfactory
neuroepithelium. It is contingent upon the correct func-
tioning of all components in the olfactory system.
Stimulation of this system begins when odorants are delivered Figure 1 Lateral view of external nose: (1) ascending process
through the nasal passages to the olfactory neuropithelium. of maxilla; (2) accessory cartilages; (3 and 4) lateral and medial
Physiological or pathological alterations of the nasal crura of lower lateral cartilages; (5) upper lateral cartilage; (6)
passages can alter the perception of odors. Thus, under- nasal bone.
standing nasal anatomy and the dynamic nature of these
passageways is essential to a complete understanding of
human olfactory function.
protuberances known as turbinates (conchae). A cleft
In this chapter, we review nasal anatomy as it relates to
(meatus) is present lateral to each of these turbinates. Four
human olfaction. When appropriate, structures with
paired groups of paranasal sinuses are ventilated through
synonyms are followed by their alternate names in paren-
these meati. Posteriorly, the nasal passages end at the pos-
theses. The terminology used to describe anatomical
terior choanae. The posterior choanae are bounded by the
relationships of the nose is presented in Figure 1. It should
posterior aspects of the nasal septum and inferior
be noted that the nasal anatomy has not been thoroughly
turbinate; they represent the anterior openings to the
investigated in all ethnic groups. Therefore, some of the
nasopharynx.
anatomical description in this chapter may not generalize
The majority of the bony and cartilaginous structures
to non-Caucasian individuals. The overview that follows is
that support the nasal cavity are covered with mucus-
a brief introduction to the major components that form the
secreting epithelia referred to as mucous membranes
nasal passages.
(mucosa, Schneiderian membranes, tunica mucosa nasi).
Secretions from the nasal mucosa are regulated by the
B. Anatomical Overview
innervation and vascular supply of the nasal cavity. These
regulators are in turn affected by many factors. The nasal
The nasal passages are complex and dynamic conduits
structures that most influence the delivery of odorants are
through which respiration begins. They communicate with
emphasized in the more detailed description of the airway
the external environment through the nose, a pyramid-
that follows.
shaped, bony, cartilaginous, and soft tissue structure which
rests upon an elliptical bony opening into the midface (Fig.
1). This elliptical opening is known as the pyriform aper- II. THE NOSE
ture (anterior choanae) and is the anterior boundary of the
nasal cavity (nasal fossa, cavum). The nose and nasal cav- Externally, the nose can be divided into three separate seg-
ity are separated into two nasal passages by a central par- ments (Lanza et al., 1991). The superior one third is
tition called the nasal septum. The lateral wall of each composed of the paired nasal bones fused with the ascend-
nasal cavity is shaped by three (or occasionally four) bony ing (frontal) processes of the maxilla. The middle third of
Anatomy of the Human Nasal Passages 3

the nose arises at the caudal end of the nasal bones, where
they join the cephalic portions of the upper lateral cartilages
(lateral cartilages). The lower third of the nose originates at
the juncture of the upper and lower lateral cartilages (alar
cartilages). The lower lateral cartilages are comprised of the
medial and lateral crura (Fig. 1). These cartilaginous sup-
port structures help maintain the caudal one third of the
nose. The medial crura of the lower lateral cartilages abut
one another at the midline to form the most caudal partition
known as the columella. The columella and nasal septum
divide the nose into separate nasal passageways. The nasal
septum, which lies directly posterior to the columella, sup-
ports the caudal one third of the nose.
Minor changes in the three-dimensional structure of the
caudal one third of the nose, secondary to either trauma or
surgery, can have a dramatic affect upon nasal airflow, as
well as cosmesis. The nasal musculature regulates nasal
airflow by controlling the aperture of the nares and nasal Figure 2 Schematic diagram, representing a sagittal view of
valve regions (see Chapter 21 and below). nasal septum depicting (a) artery, (aa) arteries, and (n) nerves. A
star marks the perpendicular plate of ethmoid and the (*) indi-
cates the quadrangular cartilage. (From Lanza et al., 1990.)
A. Nasal Septum

The nasal septum (Fig. 2) is composed of three anatomical

regions: the membranous septum, the cartilaginous sep-
tum, and the bony septum. However, a small portion of this vomeronasal organ (Jacobson’s organ, Ruysch tube) may
vertical midline partition is created by contributions from be identified as a blind pouch in the septal mucosa. The
the maxillary and the palatine bones. Both of these bones opening to the vomeronasal organ, first located by Ruysch
send up vertical crests measuring 3–10 mm in height to in 1703 (Bahtnagar, 1996), is located near the base of the
which the cartilaginous and bony septum are attached nasal septum approximately 10–17 mm posterior to the
(Lang, 1989). The membranous septum is the most anterior anterior nasal spine (Smith et al., 1998). The documented
portion of the septum and is comprised of squamous prevalence of this organ varies. Moran et al. (1991) used an
epithelium overlying connective tissue. It extends from the operating microscope to identify bilateral vomeronasal
cephalic border of the columella to the caudal end of the organs in all 200 patients examined. Using anterior
cartilaginous septum. rhinoscopy, Garcia-Velasco and Moudragon (1991) was
The cartilaginous septum (quadrangular or quadrilateral able to identify the vomeronasal organ in 808 of 1000
cartilage) is situated just posterior to the membranous patients. Won et al. (2000) performed rigid nasal
septum and traverses the nose and nasal cavity. Its hyaline endoscopy to identify the organ in 22 of 78 patients. A
cartilage is 2– 4 mm thick and is covered by mucoperi- more detailed discussion of the vomeronasal organ is
chondrium (Lang, 1989). Its basal attachment to the max- found in Chapter 46.
illary crest is termed the footplate. The cartilaginous septum The bony septum lies directly posterior to the cartilagi-
widens at several locations, which include its base, its junc- nous septum and is thus situated within the nasal cavity. It
tion with the upper lateral cartilages, and the anterior septal is formed by the vomer and perpendicular plate of the eth-
body (anterior tubercle, septal intumescence). The anterior moid. Both of these bones are covered by mucoperiosteum
septal body is a thickened area of mucosa that has charac- (Fig. 2). Deviations in the septum are extremely common
teristics resembling erectile tissue. This is situated on the and may result from a number of causes. Most are asymp-
septal cartilage just anterior to the middle turbinates. tomatic but, when severe, can lead to bilateral nasal
According to Lang (1989), Zuckerkandl noted in 1884 obstruction and anosmia. The junction of the cartilaginous
that this body marks the entrance to the olfactory cleft. septum with the bony septum is a common site for septal
Thin cartilaginous strips present at the base of the spurs to occur (a form of septal deviation). Disruption of
quadrilateral cartilage are known as the paraseptal cartilages the septum in childhood by either trauma or surgery has
(Jacobson’s cartilage). They are present in most adults and been reported by some investigators to stunt nasal growth
may ossify to form paraseptal ossicles (Lang, 1989). The (Farrior and Connolly, 1970; Jugo, 1987).
4 Clerico et al.

B. Nasal Vestibule and Nasal Valve positioned as described in the previous paragraph. Ethnic
origin, nasal trauma, and gender are just a few of the import-
The nares open into the anterior nasal chambers, which are ant variables that can affect the components and positioning
known as the vestibules. They are lined by keratinizing of the flow limiting segment to each nasal passage.
squamous epithelium, a hair-bearing epithelium containing Several theories exist concerning the function of the
sweat and sebaceous glands. The nasal hairs are known as nasal valve. Inhalation against upper airway resistance
vibrissae. Caudally, the vestibule is bounded by the free produces increased intrathoracic pressure. This may aid
margin of the ala and nasal sill (Fig. 3). Posteriorly, the alveolar gas exchange by prolonging the expiratory phase
vestibule leads to the pyriform aperture and the nasal cavity. of breathing. With regard to olfaction, the nasal valve
The line demarcating the junction between the skin of the disrupts the laminar airflow entering the nares. The
nasal vestibule and the mucosa of the nasal fossa is called resultant turbulent stream in the nasal cavity can promote
the limen vestibuli. Superiorly, this transition line roughly interaction of odorants with the olfactory neuroepithelium
corresponds with the cephalic border of the lower lateral (Berglund and Lindvall, 1982).
cartilages. Inferiorly, the limen vestibuli approximates the The vestibular counterpart to the nasal valve is the cul-
location of the pyriform aperture. The floor of the vestibule, de-sac (diverticulum, infundibulum). It represents a dilation
at least in Caucasians, usually lies at a slightly lower level of the lateral vestibule between the caudal border of the
than the inferior rim of the pyriform aperture. upper lateral cartilages and the cephalic border of the lower
The narrowest portion of the nasal passage is the func- lateral cartilage. Cottle (1987) suggested that the cul-de-sac
tional segment known as the nasal valve area. It is situated and the nasal valve together represent a series of baffles for
within the nasal vestibule. The superolateral margin of the temperature and humidity control of respired air.
valve area is the caudal border of the upper lateral cartil-
age. The medial boundary is the quadrilateral cartilage. C. Nasal Musculature
Inferiorly and posteriorly, its limits are the pyriform aper-
ture and anterior portion of the inferior turbinate (Bridger The muscles of the nose can be categorized into those that
and Proctor, 1970; Kern, 1978). elevate, depress, dilate, and compress its structure (Tardy
The nasal valve area (Fig. 3) is distinguished from the and Brown, 1990). While all nasal muscles may have an
nasal valve. The nasal valve is a specific slit-like structure impact upon both appearance and function, some appear to
situated between the caudal ends of the upper lateral cartil- play a greater role in affecting mimetic expression, whereas
ages and the septum. In Caucasians it is the major flow- others play a greater role in affecting airflow. Typically,
limiting segment in the entire airway, accounting for about muscle groups function synergistically to achieve either
50% of total resistance to respiratory airflow (Cole, 1993). effect. Muscles of the nose are compartmentalized by two
Whether or not this is true for other ethnic groups is not aponeuroses: the superficial muscular aponeurotic system
well established. The velocity of airflow through the valve (SMAS) and the perichondrial aponeurosis. The muscles
during normal breathing approaches gale-force speed lie deep to the SMAS (Daniel and Letourneau, 1988;
(Cole, 1993). Thus, even small vestibular lesions, such as Tardy and Brown, 1990). The muscle group that elevates
cysts and papillomas, can have a substantial impact upon includes the procerus, levator muscle of the upper lip and
airflow at the entrance of the nose. ala, and the anomalous nasi. The depressor group includes
Although anatomical boundaries can be assigned to the the alar portion of the nasalis muscle and the depressor
nasal valve area and the nasal valve, these structures are best nasi septi labii. The depressor nasi septi labii depresses the
considered functional segments whose anatomical bound- membranous septum and draws the nasal tip downward,
aries may vary from individual to individual. They are thereby narrowing and elongating the vestibule. It also
contributes to expanding the nares during deep inspiration.
The compressors of the nose include the transverse portion
of the nasalis and the compressor narium minor.
According to Tardy (1990), the dilator group includes the
anterior dilator naris. However, Fomon et al. (1950) argue
that the same effect is achieved with the alar portion of the
nasalis (Fig. 4).
All of the nasal muscles are innervated by the lower zygo-
matic and buccal branches of the facial nerve. The procerus
Figure 3 Basal view of the external nose and nasal vestibule receives additional neural input from the temporal branch of
demonstrating the nasal valve area. (From Lanza et al., 1990). the facial nerve. In 1977, Sasaki et al. used electromyography
Anatomy of the Human Nasal Passages 5

The ethmoid bone may be conceptualized as a horizon-

tal bony plate from which a series of parallel vertical plates
emanates (Fig. 5). The center portion of the horizontal
plate is known as the cribriform plate (lamina cribrosa). As
its Latin root implies, this sieve-like structure is perforated
with multiple openings known as foramina. Each side of
the cribriform plate contains between 20 and 71 foramina
(Lang, 1989). Through these foramina pass the fila olfac-
toria, which are the coalescence of unmyelinated axonal
filaments arising from the sensory neuroepithelium in the
olfactory cleft. The olfactory cleft is the space situated
between the medial surface of the turbinates (middle and
superior) and the bony septum (perpendicular plate of the
ethmoid) (Figs. 5 and 6) (Douek et al., 1975; Lanza et al.,
1993; Lovell et al., 1982). Axons from the olfactory cleft
relay their messages centrally through synapses in the
olfactory bulbs.
Figure 4 Nasal musculature: (1) medial fascicle of procerus The cribriform plate is divided at the midline into
muscle; (2) levator of the upper lip; (3) levator of the upper lip approximately equal halves. Superiorly it is divided by the
and ala; (4) anterior dilator naris; (5) compressor narium minor crista galli and inferiorly by the perpendicular plate of the
muscle; (6) orbicularis oris muscle (7) depressor septi nasi labii; ethmoid. The crista galli is occasionally pneumatized and
(8) transverse (a) and alar (b) portions of the nasalis muscle; may even be involved with disease extending from the eth-
(9) anomalous nasi muscle; (10) lateral fascicle of procerus mus- moid sinus. The perpendicular plate of the ethmoid forms
cle. (Adapted from Tardy and Brown, 1992.) the superior bony portion of the nasal septum.

to demonstrate that the nasal dilator muscles functioned in

direct correlation with ventilatory resistance. Cole et al.
showed in 1985 that electrical activity recorded with alar
electromyograms ceased with mouth breathing.

III. THE NASAL CAVITY

Each nasal cavity can be thought of as a modified box that is

open at opposite ends, with a roof, a floor, and two side walls.
The anterior limit of the nasal cavity is the pyriform aperture
and its posterior limit the posterior choanae. These bony
constituents and the soft tissue elements, including the vas-
culature, innervation, and epithelium, are discussed below.

A. Osteology
1. Paranasal Sinuses
a. Ethmoid Complex and Sinuses. Since both the olfac-
tory bulbs and the olfactory neuroepithelium rest upon the
Figure 5 Schematic drawing of the ethmoid bone separated
ethmoid complex, the integrity of this intricate structure is
into anterior and posterior segments. (1) anterior ethmoid sinuses;
essential for normal olfaction. The ethmoid is a bony complex (2) middle turbinate attached to cribriform plate above (ant.
that articulates with 13 bones and forms a central part of the segment) and lateral attachment to the lamina papyracea (post.
nasal roof. Its articulations include the paired frontal, segment); (3) crista galli; (4) perpendicular plate; (5) lateral
sphenoid, nasal, maxillary, lacrimal, palatine, and inferior lamella of cribriform; (6) superior turbinate; (7) olfactory fossa;
turbinate bones and the unpaired vomer (Gray, 1973). (8) lamina papyracea; (9) middle meatus.
6 Clerico et al.

The superolateral edge of the cribriform plate gives rise when the lateral lamellae are very long (8 mm plus) and
to the lateral lamellae of the cribriform. These lamellae the roof of the ethmoid is well above that of the cribriform
form the lateral border of the olfactory fossa intracranially. plate. Under this circumstance the bone of the lateral
A portion of this lateral lamella frequently contains a lamellae is very thin and the olfactory fossa is very deep.
structure referred to as the ethmoidal sulcus of the olfac- This is the anatomical condition under which iatrogenic
tory fossa. This sulcus is a groove through which the anterior CSF leak is most likely to occur during sinus surgery (Fig. 6)
ethmoidal artery courses once it has traversed from lateral (Kainz and Stammberger, 1989; Stammberger, 1991).
to medial along the ethmoid roof (see below). The bone The inferolateral border of the cribriform plate gives
thickness of the sulcus has been recorded at 0.05 mm rise to the paired middle and superior turbinates. Lateral
(Stammberger, 1991). The significance of these measure- and inferior to each of these turbinates are clefts known as
ments lies in the propensity of this region to fracture dur- the middle and superior meati, respectively. The superior
ing ethmoid surgery, resulting in cerebrospinal fluid (CSF) turbinate is often the most posterior of the lamellae within
leaks (Kainz and Stammberger, 1989). the ethmoid complex. Occasionally, a supreme turbinate is
The lateral lamellae vary widely in height and orienta- present. Posterior to the superior concha rests the anterior
tion. Keros described three types that define the variations wall of the sphenoid sinus. Anatomical variations in the
of the olfactory fossa (Kainz and Stammberger, 1989). turbinates, such as their pneumatization (with concha
Type I denotes a flat olfactory fossa where the lateral bullosa or interlamellar cells), are not uncommon (Bolger
lamella of the cribriform are short in height, between 1 and et al., 1991) (Fig. 7). Development of such anatomical
3 mm. Type II olfactory fossa are those where the lateral variations could theoretically compromise airflow to the
lamella are a little taller (4–7 mm) and the roof of the eth-
moid is steeper. A Keros type III olfactory fossa occurs

Figure 6 Coronal CT of paranasal sinuses. Bone is represented

by white, soft tissues are depicted in gray, and black indicates air-
containing spaces. The (*) is situated in right maxillary sinus,
below the right eye. The inferior portion of the left middle
turbinate is marked by the arrowhead and inferior turbinate by the
circle. Note the attachment of the middle turbinate to the cribri-
form above. A short white arrow in the left anterior ethmoid
sinuses points to the anterior ethmoidal neurovascular bundle as
it emerges from the left orbit and courses along the roof of the Figure 7 Coronal CT scan of the paranasal sinuses showing brain
ethmoid. The open arrow located in the anterior cranial fossa is (black check), orbital contents (white check), olfactory fossa (white
directly superior to the bony crista galli. The long thin arrow situ- asterisk), nasal septum (black asterisk), right middle turbinate (black
ated within the olfactory cleft points superiorly to the cribriform star), left middle turbinate with concha bullosa (white star), inferior
plate. The star in the right olfactory fossa is adjacent to the verti- turbinate (large black curved arrow), nasal cavity (small white
cal lamella of the cribriform plate. This most closely resembles a curved arrow), ethmoid bulla (large arrow), uncinate process (small
Keros type II olfactory fossa. arrow), and drainage pathway of the maxillary sinus (arrowhead).
Anatomy of the Human Nasal Passages 7

olfactory cleft. However, to date no studies have docu- allergens) are likely to have a significant impact upon the
mented such an effect. ostiomeatal complex (Wolfsdorf et al., 1969). Resultant eth-
The most lateral vertical plates of the ethmoid bone are moid sinusitis, particularly in the case of polyposis, can
the paired laminae papyracea (orbital plate of ethmoid). produce hyposmia. Appropriate medical and or surgical
The lamina papyracea form the majority of the medial therapy can sometimes reverse these conditions.
orbital wall. Medial to these plates and lateral to the mid-
dle and superior turbinates are the ethmoid sinuses (Figs. 5 b. Maxillary Bone and Sinus. The maxilla is the sec-
and 6). Anatomically, the development and pneumatization ond-largest facial bone (after the mandible) and
of the ethmoid sinuses varies from individual to individual. contributes to the structure of the oral, nasal, and orbital
The anterior and posterior ethmoid sinuses are a maze of cavities. Each maxilla articulates with eight bones: the
individual sinuses (cells), which collectively have earned zygomatic, frontal, palatine, nasal, ethmoid, lacrimal, pala-
the term “ethmoid labyrinth.” The ethmoid sinuses are tine, and inferior turbinate. Its intranasal surfaces form the
divided into anterior and posterior sinus groups by the lat- pyriform aperture, anterior floor of the nasal cavity, infer-
eral attachment of the middle turbinate to the medial ior nasal septum, and the lateral nasal wall below the orbit.
orbital wall. This attachment is known as the basal (grand) The maxillary sinus (antrum, maxillary antrum) develops
lamella of the middle turbinate. The anterior ethmoid within the maxilla and has a mean volume of about 14 mL
sinuses are generally smaller and more numerous than the (Maran and Lund, 1990). The medial maxillary bone is
posterior ethmoid cells. They are ventilated through the open at the maxillary hiatus. This hiatus is partially closed
middle meatus. However, the posterior ethmoid cells drain by bony contributions from the palatine, lacrimal, and infer-
through the superior meatus and occasionally through a ior turbinate bones. Furthermore, a connective tissue sheet,
supreme meatus if a supreme turbinate is present. The roof covered by mucosa, spans the gap that remains.
of the ethmoid sinuses is not completely formed by eth- Zuckerkandl described the portions of this fibrous connect-
moid bone (McMinn and Hutchings, 1977; Stammberger, ive tissue sheet anterior and posterior to the uncinate
1991). Instead, a significant portion of this roof is formed process as “the anterior (inferior) and posterior
by the frontal bone (Stammberger, 1991). fontanelles,” respectively (Fig. 8) (Messerklinger, 1978).
The anterior ethmoid sinuses, as well as the frontal and Mucus generated within the maxillary sinus is propelled
maxillary sinuses, drain through a narrow region located in by cilia off the sinus floor in a star-shaped pattern
the anterior middle meatus. This region is referred to as the (Messerklinger, 1978; Stammberger, 1991) (Fig. 9). It exits
ostiomeatal complex. The ostiomeatal complex is created through the maxillary ostium, an opening along the anterior-
by the ethmoid infundibulum, uncinate process of the superior aspect of the medial antral wall. Typically
ethmoid bone, hiatus semilunaris, ethmoid bulla, and mid- maxillary sinus secretions follow a path through the eth-
dle turbinate (Lanza and Kennedy, 1993) (Fig. 7). The moid infundibulum and cross the hiatus semilunaris into the
hook-shaped, uncinate process is the most anterior struc- middle meatus. From the middle meatus these secretions
ture exposed to inspired air within the middle meatus. A eventually drain into the nasal cavity beneath the eustachian
few millimeters posterior to the uncinate process is the tube orifice in the posterior nasal cavity (Kennedy et al.,
anterior face of ethmoid bulla (bulla ethmoidalis). The 1987). Smaller accessory ostia are commonly found within
ethmoid bulla is generally the largest and most constant the posterior fontanelle. In most cases mucus from the max-
ethmoid air cell (sinus). The two-dimensional space illary sinus bypasses these openings in favor of exiting
between the uncinate process and the ethmoid bulla is the through the natural ostium (Kennedy et al., 1987).
hiatus semilunaris inferioris (Messerklinger, 1978;
Stammberger, 1991). It leads anteriorly to a three-dimen- c. Frontal Bone and Sinus. The paired frontal bones
sional, funnel-shaped space lateral to the uncinate process articulate with the ethmoid, lacrimal, maxillary, nasal,
termed the ethmoid infundibulum. parietal, sphenoid, and zygomatic bones (Gray, 1973).
When inflammatory conditions afflict the nose and Anteriorly the frontal bones meet with one another at the
paranasal sinuses, the ostiomeatal complex is frequently midline. In most cases, the frontal bone contains a nasal
involved. This is believed to be due to the narrowness of this spine, which abuts the perpendicular plate of the ethmoid
region and the fact that only a small amount of mucosal and the undersurface of the nasal bones, helping to stabil-
swelling is required to occlude drainage through this area. ize and support them. Posteriorly, at the midline, the
Stagnation of drainage can promote regional inflammation frontal bone articulates with the cribriform plate. The
and infection (Kennedy et al., 1987). Moreover, since nearly foveolae ethmoidales of the frontal bone joins the lateral
66% of inspired air passes through the anterior middle lamella of the cribriform plate to create the roof of the
meatus, environmental agents (viruses, pollutants, and ethmoid sinuses (Stammberger, 1991). The foveolae
8 Clerico et al.

Figure 8 Schematic sagittal view of lateral nasal wall: (1) frontal bone; (2) ethmoid bone; (2a) bulla ethmoidalis; (2b) uncinate process; (3)
sphenoid bone; (4) perpendicular process of the palatine bone; (5) maxillary bone; (6) lacrimal bone; (7) nasal bone; (8) inferior turbinate bone;
(9a and b) upper and lower lateral cartilage. Note the lacrimal bone (6) is depicted more prominently here than it would be seen in life from
this sagittal view. In actual fact a greater portion of this bone rests more anterolaterally to the uncinate process. This diagram, however, high-
lights its relationships to the ethmoid, inferior turbinate, maxillary, and frontal bones.

ethmoidales region has a mean thickness of 0.5 mm,

largest and most central bone of the skull base. It articulates
whereas the lateral lamellae of the cribriform plate aver-
with the ethmoid, frontal, vomer, occipital, parietal, temporal,
ages 0.2 mm in thickness (Stammberger, 1991).
zygomatic, and palatine bones (Gray, 1973). Its articulation
The frontal sinus is the most variable of the paranasal
with ethmoid appears at the cribriform plate, perpendicular
sinuses, being completely absent in a small percentage of
plate, lamina papyracea, and the posterior aspect of the
the population. It is formed by pneumatization, which
ethmoid labyrinth. The vomer meets the anterior wall of the
originates in the ethmoid. An intersinus septum is usually
sphenoid sinus in the midline at an area termed the sphenoid
present inferiorly at the midline but can deviate markedly
crest. The perpendicular process of the palatine bone
as it courses upwards. Some investigators have documented
continued growth and expansion of the frontal sinus well
into adulthood, though this is thought to be more the
exception than the rule (Lang, 1989).
Mucus generated within the frontal sinus circulates prior
to exiting through the frontal sinus ostium. This opening is
located anteromedially in the floor of the frontal sinus
(Fig. 9). The frontal sinus ostium drains into a channel
within the anterior superior ethmoid complex known as the
frontal recess. The term “nasofrontal duct” had been used
to describe this area of drainage, but is falling out of favor
because only in the minority of cases does a discrete bony
canal exist. Thus, anterior ethmoid sinus disease may pro-
mote obstruction of the frontal recess and block mucocil-
iary clearance from the frontal sinuses. A frontal sinusitis
may develop in association with such an obstruction.

d. Sphenoid Bone and Sinus. The sphenoid bone Figure 9 Schematic diagram of the mucociliary clearance pat-
forms the most posterior extent of the nasal cavity. It is the terns of the frontal (above) and maxillary (below) sinuses.
Anatomy of the Human Nasal Passages 9

articulates with the body of the sphenoid more laterally in the ascending process of the maxilla forms the lacrimal fossa,
area of the sphenopalatine foramen. The medial pterygoid which houses the lacrimal sac in the anterior orbit. The
plate of the sphenoid meets the posterior surface of the per- nasolacrimal duct drains this sac through the medial floor
pendicular process of the palatine bone to form the most pos- of this fossa. The duct traverses inferiorly to drain into the
terolateral recesses of the nasal cavity and nasopharynx. inferior meatus.
The sphenoid sinus has been classified into three
types on the basis of size and degree of pneumatization c. Palatine Bone. The palatine bone articulates with
(Moss-Salentijn, 1991): (1) the rare conchal type, with the inferior concha, maxilla, ethmoid, sphenoid, vomer,
minimal posterior extension, (2) the presellar type, with and opposing palatine bones (Clemente, 1981; Gray,
extension to the anterior wall of the pituitary fossa, and 1973). Its L shape forms the posterior portions of the floor
(3) the common postsellar (sellar, postsphenoid) type, and lateral wall of the nasal cavity. Its perpendicular
with posterior pneumatization beneath and sometimes process has a conchal crest to which the posterior portion
even behind the pituitary fossa. The average volume of of the inferior turbinate bone attaches. This perpendicular
the sphenoid sinus is about 5–7 mL. A mid-sagittal inter- process also forms part of the medial wall of the maxillary
sinus septum usually is present but, as in the case of the sinus and joins the sphenoid bone. The articulation with
frontal sinus septum, may be in an eccentric position. the sphenoid is located just beyond the posterior end of the
The lateral wall of the sinus may be indented by the optic middle turbinate and marks the region of the sphenopala-
nerve and internal carotid artery, forming recesses above tine foramen. This foramen is the opening through which
and below these structures. Of note is that several inves- the sphenopalatine neurovascular bundles emerge (see
tigators have found bony dehiscence overlying these below) (Fig. 8).
structures, making surgical manipulation of the lateral
sphenoid wall extremely dangerous (Kennedy et al.,
1990). Microdehiscences in the lateral wall of the sphe- d. Vomer. This midline unpaired bone forms the
noid sinuses are thought to serve as a route for the inferior portion of the bony nasal septum. It articulates
intracranial spread of infection. with the perpendicular plate of the ethmoid, the sphenoid,
Mucus is actively transported out of the sphenoid sinus palatine, and maxillary bones, as well as with the septal
through the sphenoid ostia. The ostium is located just 2–3 cartilage (Clemente, 1981; Gray, 1973).
mm lateral to the sphenoid crest in the sphenoethmoidal
recess. Inflammatory disease in the sphenoethmoidal B. Soft Tissue Anatomy
recess can cause blockage of the sphenoid os and subse-
quent sphenoid sinusitis. Drainage from the posterior eth- 1. Vasculature of the Nasal Cavity
moidal sinuses eventually joins that of the sphenoid sinus
in the sphenoethmoidal recess. Mucus clearance from The description that follows represents a basis for under-
these sinus groups is usually seen together draining above standing the predominant circulatory patterns of nasal pas-
the eustachian tube orifice. sages. The arterial supply to the nasal passages is formed by
a plexus of vessels derived from several sources. Thus, sev-
eral texts report subtle differences in the origin of vessels.
2. Other Bones of Nasal Cavity
a. Inferior Turbinate Bone. The inferior turbinate a. Arterial Supply. The nasal cavity receives its
bone articulates with the maxilla anteriorly, the perpendicu- blood supply from both the internal and external carotid
lar process of the palatine bone posteriorly, and the lacrimal arteries. Intracranially the internal carotid artery gives rise
bone superiorly (Clemente, 1981; Gray, 1973) (Fig. 8). to the ophthalmic artery. The ophthalmic artery, in turn,
Occasionally it also interfaces superiorly with the uncinate branches to form the anterior and posterior ethmoid arter-
process of the ethmoid bone. The cleft lateral to the inferior ies. The ethmoid arteries cross from the orbit into the
turbinate is known as the inferior meatus. The nasolacrimal ethmoid labyrinth through a foramina located near the
duct opens into the anterior segment of the inferior meatus. frontoethmoidal suture. These arteries usually course within
bony canals situated within the ethmoid labyrinth along
b. Lacrimal Bone. The lacrimal bone is the smallest the foveolae ethmoidales. However, the position of the ves-
bone of the lateral nasal wall and articulates with the sels may vary considerably, from 2 mm below to 4 mm
frontal process of the maxilla, the inferior turbinate, the above the level of the cribriform plate (Lang, 1989;
lamina papyracea, and the frontal bone (Fig. 8) (Zide and Stammberger, 1991) (Fig. 6). Once both ethmoid arteries
Jelks, 1985). The lacrimal bone together with the cross the ethmoid roof, they enter the cranial vault to give
10 Clerico et al.

rise to meningeal and nasal branches. The meningeal

branches supply the dura matter, and the nasal branches
descend through the cribriform plate to supply the nasal
cavity. Specifically, the posterior ethmoidal arteries supply
the superior turbinate and posterior septum. After the
anterior ethmoid artery courses within the ethmoidal
sulcus of the olfactory fossa (Stammberger, 1991), its ter-
minal branches supply the lateral nasal wall, including the
anterior middle turbinate and the septum. A small division
courses to the external nose between the caudal border of
the nasal bones and upper lateral cartilages (Gray, 1973).
The external carotid artery delivers blood to the nasal
cavity via two main branches, the facial and the internal
maxillary arteries. The facial artery has two terminal branches
which supply the nose and anterior nasal cavity: the superior
labial and angular arteries. Significant portions of the nasal
cavity derive their arterial distribution from the internal Figure 10 Schematic diagram of sagittal view of lateral nasal
maxillary artery. Within the pterygopalatine fossa (pterygo- wall depicting (a) artery, (aa) arteries, and (n) nerve. (From Lanza
maxillary space), the internal maxillary artery divides into et al., 1990.)
many branches. Most notable for this discussion are the
sphenopalatine and descending palatine arteries. Variations
can occur in the manner in which these vessels arise, located on the septum in a region known as Little’s area.
anastomose, and supply the nasal cavity. According to Gray This is by far the most common source of epistaxis within
(1973), the descending palatine is also known as the greater the nasal cavity. However, most cases of severe posterior
palatine artery; however, Lang (1989) distinguishes these nasal bleeding involve the sphenopalatine artery.
vessels from one another. Lang asserts that the descending
palatine artery gives rise to the greater palatine vessel. b. Venous Drainage. Nasal veins arise from a rich
Regardless, the descending palatine artery, which arises in network within the nasal mucosa and generally course
the medial aspect of the pterygopalatine fossa, contributes to along the reverse route of the arterial supply. Since a
the blood supply of the septum and lateral nasal wall system of valveless veins constitutes the drainage from the
(Clemente, 1981; Gray, 1973; Lang, 1989; Pansky, 1979). nasal passages, the potential for spread of infection to the
The portion known as the greater palatine artery courses cavernous sinus is real. Venous drainage from the septum
through pterygopalatine canal and exits into the oral cavity generally corresponds with the course of the sphenopala-
through the greater palatine foramen. This vessel courses tine artery. Venous blood coursing in this direction eventu-
along the hard palate, where its terminal branches pass ally reaches the pterygopalatine and infratemporal fossa.
through the incisive canal to supply the nasal septum (Figs. The pterygoid plexus is located in the infratemporal fossa
2, 10). and eventually communicates with the cavernous sinus.
The sphenopalatine artery emerges from the pteryg- The ethmoidal veins exit to the orbit and anterior cranial
opalatine fossa, along with some branches of the descend- fossa. Orbital drainage via the ophthalmic vein is linked to
ing palatine artery via the sphenopalatine foramen. This the cavernous sinus. Ethmoidal drainage can also join the
foramen is located just superior to the posterior attachment venous drainage from the dura mater and exit through the
of the middle turbinate. The sphenopalatine artery divides superior sagittal sinus (Lang, 1989). The area of the nares
into posterior lateral and posterior septal branches. The is drained by a small external nasal plexus and ultimately
posterior lateral branches contribute to the supply of the drains into the facial vein.
turbinates, their respective meati, and to the paranasal
sinuses (Gray, 1973). The posterior septal branch gives rise c. Lymphatic Drainage. Lymph from the nasal cavity
to the nasopalatine artery, which runs in a groove along the drains both anteriorly and posteriorly. The nasal vestibule
vomer to reach the incisive foramen. drains into the facial vein and submandibular lymph nodes
Anteriorly on the nasal septum and just superior to the (Gray, 1973). The lymphatics of the septum run along the
incisive foramen, anastomoses between the septal branches floor of the nose to join drainage from the lateral nasal
of the superior labial, anterior ethmoid, greater palatine, wall. The lymphatic pathways of the lateral nasal wall are
and sphenopalatine arteries form Kiesselbach’s plexus, divided into anteroinferior and posterosuperior trunks
Anatomy of the Human Nasal Passages 11

(Lang, 1989). The anteroinferior trunk drains the inferior 2. Innervation of the Nasal Cavity
turbinate and anterior face of the middle turbinate. The
Besides the special sensory function associated with cra-
posterosuperior trunk filters the olfactory cleft, superior
nial nerve (CN) I, the nasal cavity contains general sensory
turbinate, posterior middle turbinate, and sphenoeth-
and autonomic fibers. The general sensory innervation is
moidal recess (Lang, 1989). These two trunks join poste-
derived from the ophthalmic (V1) and maxillary (V2)
rior to the eustachian tube to drain into the lateral
divisions of the trigeminal nerve. The autonomic input
retropharyngeal nodes. Other lymph node chains, namely
originates from the cervical sympathetic chain and superior
the jugulodigastric and the deep cervical, also receive
salivary nucleus in the midbrain (parasympathetic). There
lymphatic drainage.
are two areas of neural supply within the nose which are
poorly understood in humans: the nervus terminalis and
vomeronasal organ (Jacobson’s organ).
d. Microcirculation and Cavernous Plexuses. Three
different types of capillary vessels supply the nasal cav-
ity. Capillaries that directly supply the epithelial cells a. General Sensory Supply. The trigeminal nerve
are known as the subepithelial capillaries. These exhibit (fifth cranial nerve, on CN V) (see Chapter 47) is the largest
large fenestration in their endothelial lining and pro- of the cranial nerves and relays both sensory and motor
bably play a role in the humidifying the air (Cauna, information (e.g., muscles of mastication). The sensory root
1982). Deeper within the mucosa, capillaries associated of this nerve has its ganglionic cell bodies situated within
with glands are fenestrated to a lesser extent. Capillaries the semilunar (Gasserian, trigeminal) ganglion. The trigem-
not associated with the epithelium or glands are not inal nerve trifurcates as it emerges from the semilunar gan-
fenestrated (Cauna, 1982). glion within the anterior middle cranial fossa.
Discrete regions within the nasal mucosa resemble erec- The ophthalmic division of the trigeminal nerve
tile tissue and are known as cavernous plexuses. These are enters the posterior orbit through the superior orbital
networks of tortuous valveless veins which can rapidly alter fissure and gives rise to the nasociliary nerve. The
the dimensions of the nasal passages. The cavernous plexus nasociliary nerve divides into the anterior and posterior
is best developed over the septum and inferior and middle ethmoid nerves. These join the anterior and posterior
turbinates. Frequently, they may be developed adjacent to ethmoid arteries as they course through their respective
the openings of the paranasal sinuses (Cauna, 1982). foramina at the level of the frontoethmoid suture. The
The cavernous plexuses derive their blood supply from anterior ethmoid nerve divides into an internal and exter-
both arterial and venous sources. They consist of a super- nal branch before it descends upon the anterior septum.
ficial and deep layer. The superficial layer is formed by the The internal branch innervates the anterior lateral nasal
union of veins which drain the subepithelial and glandular wall and the external division supplies a small dorsal
capillaries. The deep layer of the plexus runs along the area of the external nose. The posterior ethmoid nerve
periosteum and perichondrium. supplies the posterior and superior regions of the septum
It is of interest that olfactory ability may improve when and lateral nasal wall (Figs. 2, 10).
the nasal passageways are narrowed somewhat—i.e., when The maxillary division leaves the middle cranial fossa
the mucosa is moderately congested, wet, and red via the foramen rotundum. It crosses the roof of the
(Schneider and Wolf, 1960). Factors such as hypoxia, pterygopalatine fossa and traverses the floor of the orbit
hypercapnia, exercise, and increased sympathetic tone within the infraorbital canal. Within this canal a division
cause constriction, thereby increasing nasal patency, known as the anterior superior alveolar nerve sends fibers
whereas cold air, irritants, and hypocapnia can cause dila- to the upper incisors and canine teeth. This division also
tion (Cole, 1993). Emotional states, posture, and allergic supplies the anterior portions of the inferior meatus and
and inflammatory conditions can also affect cavernous floor of the nasal cavity (Gray, 1973). The infraorbital
plexuses (Cole, 1993). Interestingly, normal noses undergo nerve emerges distally and supplies sensory fibers to the
an alternating pattern of leftright congestion and decon- middle third of the face (including the lower lateral nose).
gestion, a phenomenon termed the nasal cycle (see Chapter Divisions of V2 that supply the nasal cavity diverge
21). In the nonpathological nose, this fluctuating resistance from the maxillary nerve in the pterygopalatine fossa.
to inspiratory airflow is seldomly appreciated. Importantly, Along with autonomic input from the sphenopalatine gan-
greater airflow through the right nostril relative to the left glion and the sphenopalatine vessels, these fibers traverse
is associated with sympathetic predominance, greater left the sphenopalatine foramen. They enter the nasal cavity as
hemispheric integrated EEG activity, and heightened olfac- several branches. The posterolateral nasal branches from
tory sensitivity (Frye and Doty, 1992). V2 supply sensation to the mucosa over the turbinates and
12 Clerico et al.

lateral nasal wall. Medial branches cross the posterior in the nervus intermedius portion of the facial nerve
nasal roof and descend upon the nasal septum as the (cranial nerve VII) to the geniculate ganglion. They exit
nasopalatine (Scarpa’s) nerve (Figs. 2, 10). The nasopala- this ganglion, without synapsing, as the greater super-
tine nerve courses anteriorly and eventually traverses the ficial petrosal nerve. This nerve merges with the deep
incisive foramen. Its terminal branches form anastomoses petrosal nerve (sympathetic) to form the nerve of the
with those from the greater palatine nerve. They may Vidian canal. The nerve of the Vidian canal enters the
innervate the anterior and superior gingiva as far laterally pterygopalatine fossa, where its parasympathetic pregangl-
as the canine teeth (Lang, 1989). ionic fibers synapse in the sphenopalatine (pterygopala-
Of clinical interest is the fact that the nasal mucosa has tine or Meckel’s) ganglion. Postganglionic fibers course
a limited ability to localize tactile and painful stimuli along with the sensory and sympathetic fibers of the
(Cauna, 1982). This may contribute to the phenomenon of nasopalatine and posterior lateral nasal nerves to inner-
referred head and facial pain seen in some sinonasal dis- vate the mucous membranes of the nose and hard palate.
orders (including sinusitis). Though no temperature recep- Other postganglionic fibers from sphenopalatine
tors have been found histologically, clinical and animal ganglion innervate the lacrimal gland.
experiments suggest that the nasal mucosa does indeed
have a thermal sense (Jones et al., 1989). 3. Epithelia of the Nasal Passages
Two other nerves of the nasal septum warrant men-
tion. The nervus terminalis (terminal nerve) is a nerve of Epithelia of different types are topographically distrib-
unknown function lying in the anterior superior aspect of uted within the nasal passages. Anteriorly, the nasal
the septum (see Chapter 48). Olfactory, general sensory, vestibule is lined by stratified squamous epithelium.
and autonomic functions have been attributed to it, but Posterior to the limen vestibuli the epithelium gradually
its actual role in humans remains uncertain (Gray, 1973; changes from squamous to respiratory in nature. Small
Lang, 1989). The cell bodies for this nerve lie in a nerve areas of squamous epithelium which persist over the
plexus in the olfactory region of the septum. anterior ends of the inferior and middle turbinate pro-
Preganglionic fibers course through the cribriform plate bably represent the influence of unmodified inspired air
into the anterior cranial fossa where multiple central contacting these areas (Mygind et al., 1982). Posteriorly,
connections are made. The vomeronasal nerve and organ ciliated cells are found in increasing numbers as the
(Jacobson’s organ) are believed by many to be vestigial nasal mucosa transforms into a true respiratory epithe-
in humans. However, it is highly developed in some lium (pseudostratified, ciliated, columnar epithelium)
animal species. (Mygind et al., 1982). A discrete region of the nasal roof
is covered by a yet different tissue, the olfactory neuro-
b. Autonomic Supply. The vascular bed and glan- epithelium (see below).
dular structures within the nasal mucosa are under
sympathetic and parasympathetic control. The sympa- a. Respiratory Epithelium. The respiratory epitheli-
thetic pathway originates in the thoracolumbar spinal um of the nasal cavity is composed of four basic cell types:
cord as preganglionic fibers. From there, the fibers join basal cells, goblet cells, and ciliated and nonciliated
the vagosympathetic trunk and then terminate in the cer- columnar cells. The basal cells lie on the basement mem-
vical sympathetic ganglion. The postganglionic fibers brane of the mucosa and do not extend to the mucosal
then course along the internal carotid artery and form the surface. They are no longer believed to be progenitor cells;
deep petrosal nerve. This nerve then unites with the they are now thought to support the columnar cell by
greater superficial petrosal nerve to form the Vidian nerve assisting in their adherence to the basement membrane
(nerve of the pterygoid or Vidian canal). The Vidian nerve (Baroody and Naclerio, 1990). Goblet cells are found in
emerges from its canal to arrive within the pterygopala- their greatest concentrations on the inferior turbinate.
tine fossa, where it contributes to the sphenopalatine These concentrations are considerable but diminished on
ganglion. Sympathetic fibers do not synapse there, but middle turbinate and septum. Columnar cells have
proceed to join the nasopalatine and posterior lateral microvilli upon their apical surfaces, which may help to
branches of V2 and are dispersed to the nasal mucosa. prevent dehydration through increases in surface area. The
The ophthalmic division of the trigeminal nerve also con- ciliated columnar cells found throughout the airway are
veys sympathetic fibers from the carotid plexus via the essential to the proper function of the respiratory tree.
ethmoidal nerves. These cells are responsible for mucus transport through a
The parasympathetic pathway to the nose begins in mechanism known as mucociliary clearance (Deitmer,
the superior salivary nucleus of the midbrain. Fibers run 1989; Messerklinger, 1978).
Anatomy of the Human Nasal Passages 13

Cilia, which project from the columnar cells, possess There are three main types of glands associated with the
intrinsic motility. Each ciliary stroke has a biphasic nature, nasal respiratory epithelium: (1) the serous (anterior nasal)
with a rapid active stroke that is followed by a slower glands, located within the vestibular epithelium, (2) the
recovery beat (Deitmer, 1989). During the active stroke, seromucous glands, and the (3) intraepithelial glands.
cilia make contact with the thicker, more superficial gel Most nasal secretions are produced by seromucous glands,
layer of nasal mucus. During the slower recovery beat, the with a lesser contribution from the epithelial goblet cells.
cilia only pass through the thinner sol layer of mucus, Seromucous glands, numbering in the tens of thousands,
which is closer to the cell surface. The overlying mucus are situated submucosally, and their fluid production is pri-
blanket is propelled by this coordinated and synchronized marily responsible for keeping the nasal mucosa moist. The
mucociliary clearance mechanism. Inhaled particulate serous glands outnumber the mucin-producing ones by about
matter trapped by the viscous mucus is thus swept back out 8:1. Intraepithelial glands are thought by some to be
of the nasal cavity and ultimately swallowed. pathological and found by others to be present in normal
The glands associated with the nasal passages produce noses. Regardless, they appear to be few in number, be irreg-
between 1 and 2 L of mucus daily. Mucus is about 96% ularly distributed, and produce only a small amount of
water and 4% glycoprotein (Widdicombe and Wells, mucus.
1982). This essential fluid serves several functions: There is a lower density of glandular elements in the
(1) protection—mucus contains proteins that defend the paranasal sinuses than in the nasal cavity. The glands
underlying epithelium against various harmful particles,
pathogens, and noxious substances inhaled from the
environment; (2) humidification—the nasal passages
lined with mucus warm and humidify inspired air, mak-
ing it more suitable for the lower respiratory tract;
(3) olfaction—mucus affects the ability of odorant mole-
cules to reach and react with the olfactory epithelium.

Figure 11 Comparison of the work of von Brunn in 1892 and Figure 12 Percentage of olfactory tissue in 71 biopsy speci-
Read in 1908 in mapping the extent of the olfactory neuroepithe- mens collected from 23 healthy individuals. The number of
lium. Note both depict the olfactory neuroepithelium ranging samples containing olfactory epithelium was divided by the
onto the superior aspect of the middle turbinate (one half natural number of samples collected in each area (n). (From Feron
size). (From Read, 1908.) et al., 1998).
14 Clerico et al.

Figure 13 Maps of the nasal lining illustrating the location of the biopsy specimens and their histological composition. The plus sign
(+) indicates that the biopsy specimen taken from that area contains olfactory epithelium and/or fasicles of the olfactory nerve (and thus
was olfactory mucosa originally). The minus sign (-) indicates that no olfactory epithelium or nerve was seen. For convenience of illus-
tration, biopsy specimens of the right-side septal mucosa and of the left-side turbinate mucosa were translated onto representations of the
left-side septum and right-side turbinates, respectively. (From Leopold et al., 2000).

within each sinus appear to have a higher concentration this region confirm the presence of olfactory neuroepi-
around the ostial regions. Furthermore, the anterior eth- thelium (Fig. 13). The dimensions and distribution of
moid sinuses contain more glands than the posterior the olfactory neuroepithelium are known to vary with
ethmoid (Tos, 1982). between individuals (Paik et al., 1992). Paik et al., (1992)
performed biopsy of the uppermost portion of the septal
b. Olfactory Neuroepithelium. Traditional teach- mucosa and noted that the ability to obtain a positive
ing places the olfactory neuroepithelum at the cribiform biopsy positive for olfactory neuroepithelium decreases
plate extending a short distance inferiorly and onto the with age. Other factors such as chemical exposure, bac-
superior turbinate (see Sec. I). Several classic diagrams terial or viral infection, and head trauma are also thought
have suggested that the middle turbinate may also have to affect the distribution of olfactory neuroepithelium.
olfactory tissue (Bucher, 1973; Lang, 1989; Read, 1908; Macroscopically, the neuroepithelium has been
von Brunn, 1892) (Fig. 11). Recent studies have demon- described by some authors to have a yellow appearance,
strated a more extensive distribution of olfactory neuro- which distinguishes it from the surrounding respiratory
epithelium, extending as anteriorly as the anterior epithelium (Read, 1908; Zippel, 1993). Further discus-
middle turbinate insertion (Leopold, 2000) and as infer- sion on the microscopic anatomy of the olfactory epithe-
iorly as the body of the middle turbinate itself (Feron lium is presented in Chapters 2, 3, 5, and 6.
et al., 1998). Feron et al., (1998) took 97 biopsies from
six different regions in 33 subjects. Olfactory neuroepithe-
lium was found in 30–76% of specimens, depending on REFERENCES
the region of biopsy (Fig. 12). Leopold et al. (2000) per-
formed electro-olfactographic studies demonstrating Baroody, F., and Naclerio, R. M. (1990). Review of Anatomy and
responses to olfactory stimuli from leads placed at the Physiology of the Nose. American Academy of Otolaryngology-
anterior insertion of the middle turbinate. Biopsies of Head and Neck Surgery Foundation, Inc., Alexandria, VA, p. 13.
Anatomy of the Human Nasal Passages 15

Berglund, B., and Lindvall, T. (1982). Olfaction. In The Nose: olfactory thresholds. In Chemical Signals in Vertebrates 6,
Upper Airway Physiology and the Atmospheric Environment, R. L. Doty and D. Moller-Schwarze (Eds.). Plenum Press,
D. F. Proctor and I. Andersen (Eds.). Elsevier Biomedical New York, pp. 595–598.
Press, Amsterdam pp. 279–285. Garcia-Velasco, J. G., and Mondragon M. (1991). The incidence
Bhatnagar, K. P., and Reid, K. H. (1996). The human of the vomeronasal organ in 1000 human subjects and its pos-
vomeronasal organ, I: Historical perspectives. A study of sible clinical significance. J. Steroid Biochem. Mol. Biol.
Ruysch’s (1703) and Jacobson’s (1811) reports on the 39:561–563.
vomeronasal organ with comparative comments and English Gray, H. (1973). Anatomy of the Human Body. Lea and Febiger,
translations. Biomed Res. 7:219–229. Philadelphia, pp. 164–196, 577–593, 683–690, 740–747,
Bolger, W. E., Butzin, C. A., and Parsons, D. S. (1991). Paranasal 911–930.
sinus bony anatomic variations and mucosal abnormalities: Jafek, B. W., Eller, P. M., Esses, B. A., and Moran, D. T. (1989).
CT analysis for endoscopic sinus surgery. Laryngoscope Post-traumatic anosmia: ultrastructural correlates. Arch.
101:56–64. Neurol. 46:300–304.
Bridger, G. P., and Proctor, D. F. (1970). Maximum nasal inspi- Jones, A. S., Wight, R. G., and Durham, L. H. (1989). The distri-
ratory flow and nasal resistance. Ann. Otol. 79:481–488. bution of thermoreceptors within the nasal cavity. Clin.
Bucher, O. (1973). Cytologie, Histologie und Mikroskopische Otolaryngol. 14:235–239.
Anatomie des Menschen, 8th ed. Huber, Bern, p. 330. Jugo, S. B. (1987). Total septal reconstruction through decortica-
Cauna, N. (1982). Blood and nerve supply of the nasal lining. In tion (external) approach in children. Arch. Otolaryngol. Head
The Nose: Upper Airway Physiology and the Atmospheric Neck Surg. 113:173.
Environment, D. F. Proctor and I. Andersen (Eds.). Elsevier Kainz, J., and Stammberger, H. (1989). The roof of the anterior
Biomedical Press, Amsterdam p. 51. ethmoid: a place of least resistance in the skull base. Am.
Cauna, N., Cauna, D., and Hinderer, K. H. (1972). Innervation of J. Rhinol. 4:191–199.
the human nasal glands. J. Neurocytol. 1:54. Kennedy, D. W., Zinreich, S. J., Shaalan, H., Kuhn, F., and Naclerio,
Clemente, D. D. (1981). Anatomy: A Regional Atlas of the R. (1987). Endoscopic middle meatal antrostomy: theory, tech-
Human Body. Urban and Schwarzenberg, Baltimore-Munich, nique, and patency. Laryngoscope 97(suppl. 43): 81.
pp. 600–615. Kennedy, D. W., Zinreich, S. J., and Hassab, M. H. (1990).
Cole, P. (1993). Respiratory Role of the Upper Airways. Mosby- Internal carotid artery as it relates to endonasal sphenoeth-
Year Book, St. Louis, pp. 8, 93. moidectomy. Am. J. Rhinol. 4:7–12.
Cole, P., Haight, J. S. J., Love, L., and Oprysk, D. (1985). Kern, E. B. (1978). Surgical approaches to abnormalities of the
Dynamic components of nasal resistance. Am. Rev. Respir. nasal valve. Rhinology 16:165–189.
Dis. 132:1229–1232. Lang, J. (1989). Clinical Anatomy of the Nose, Nasal Cavity and
Cottle, M. H. (1987). The anatomy of the nasal septum and Paranasal Sinuses. Thieme Medical Publishers, New York,
external nasal pyramid. In Rhinology, the Collected pp. 16, 30–40, 49–98, 100–121.
Writings of Maurice H. Cottle, MD, P. A. Barelli, W. E. E. Lanza, D. C., and Kennedy, D. W. (1993). Endoscopic sinus
Loch, and E. B. Kern (Eds.). American Rhinologic Society, surgery. In Head and Neck Surgery-Otolaryngology, B. J. Bailey
Chicago, IL, pp. 89–94. (Ed.). J. B. Lippincott Company, Philadelphia, pp. 389–401.
Daniel, R. K., and Letourneau, A. M. (1988). Rhinoplasty: nasal Lanza, D. C., Kennedy, D. W., and Koltai, P. J. (1991). Applied
anatomy. Ann. Plastic Surg. 20:5–13. nasal anatomy and embryology. Ear Nose Throat J.
Deitmer, T. (1989). Physiology and Pathology of the Mucociliary 70(7):416–422.
System: Special Regards to Mucociliary Transport in Lanza, D. C., Moran, D. T., Doty, R. L., Trojanowski, J. Q., Lee,
Malignant Lesions of the Human Larynx. Karger, New York. J. H., Rowley, J. C. III, Crawford, D., and Kennedy, D. W.
Doty, R. L. (1979). A review of olfactory dysfunctions in Man. (1993). Endoscopic human olfactory biopsy technique: a pre-
Am. J. Otolaryngol. 1:1. liminary report. Laryngoscope 103(7):815–819.
Douek, E., Bannister, L. H., and Dodson, H. C. (1975). Recent Leopold, D. (1986). Physiology of olfaction. In Otolaryngology-
advances in the pathology of olfaction. Proc. R. Soc. Med. 68: Head and Neck Surgery, C. W. Cummings (Ed.). C. V. Mosby
467–470. Company, St. Louis, p. 528.
Farrior, R. T., and Connolly, M. E. (1970). Septorhinoplasty in Leopold, D. A., Hummel, T., Schwob J. E., Hong, S. C., Knecht, M.,
children. Otolaryngol. Clin. North Am. 3:345. and Kobal G. (2000). Anterior distribution of human olfactory
Feron, F., Perry, C., McGrath, J. J., and Mackay-Sim, A. (1998). epithelium. Laryngoscope 110(3):417–421.
New techniques for biopsy and culture of human olfactory Lovell, M. A., Jafek, B. W., Moran, D. T., and Rowley, J. C. III
epithelial neurons. Arch. Otolaryngol. Head Neck Surg. (1982). Biopsy of human olfactory mucosa. Arch.
24:861–866. Otolaryngol. 108:247–249.
Fomon, S., Gilbert, J. G., Caron, A. L., and Segal, S. Jr. (1950). Maran, A. G. D., and Lund, V. J. (1990). Clinical Rhinology.
Collapsed ala: pathologic physiology and management. Arch. Thieme Medical Publishers, Inc., New York, pp. 17–23.
Otolaryng. 51:465–484. McMinn, R. M. H., and Hutchings, R. T. (1977). A Colour Atlas
Frye, R. E., and Doty, R. L. (1992). The influences of ultradian of Human Anatomy. Wolfe Medical Publications Ltd.,
anatomic rhythms, as indexed by the nasal cycle, on unilateral London, pp. 25–32.
16 Clerico et al.

Messerklinger, W. (1978). Endoscopy of the Nose. Urban and findings on location, structure, and size. Microsc. Res. Tech.
Schwarzenberg, Baltimore, pp. 11–14. 41:483–491.
Moran, D. T., Jafek, B. W., and Rowley, J. C., III (1991). The Stammberger, H. (1991). Functional Endoscopic Sinus Surgery.
vomeronasal (Jacobson’s) organ in man: ultrastructure and B. C. Decker, Philadelphia, pp. 17–47, 49–87.
frequency of occurrence. J. Steroid Biochem. Mol. Biol. Talamo, B. R., Feng, W-H., Perez-Cruet, M., Adelman, L., Kosik,
39:545–552. K., Lee, V. MY., Cork, L. C., and Kauer, J. S. (1991).
Moss-Salentijn, L. (1991). Anatomy and embryology. In Surgery Pathologic changes in olfactory neurons in Alzheimer’s dis-
of the Paranasal Sinuses, A. Blitzer, W. Lawson, and W. H. ease. Ann. NY Acad. Sci. 640:1–7.
Friedman (Eds.). W. B. Saunders, Philadelphia, p. 18. Tardy, E., and Brown, R. J. (1990). Anatomy of the Nose. Raven
Mygind, N., Pedersen, M., and Nielsen, M. H. (1982). Press, New York.
Morphology of the upper airway epithelium. In The Nose: Tomlinson, A. H., and Esiri, M. M. (1983). Herpes simplex
Upper Airway Physiology and the Atmospheric Environment, encephalitis. Immunohistological demonstration of spread of
D. F. Proctor and I. B. Andersen (Eds.). Elsevier Biomedical virus via olfactory pathways in mice. J. Neurol. Sci.
Press, New York, pp. 70–97. 60:474–484.
Naessen, R. (1970). The identification and topographical local- Tos, M. (1982). Goblet cells and glands in the nose and
ization of the olfactory epithelium in man and other mam- paranasal sinuses. In The Nose: Upper Airway Physiology
mals. Acta Otolaryngol. 70:51–57. and the Atmospheric Environment, D. F. Proctor and I. B.
Naessen, R. (1971). An enquiry on the Morphological character- Andersen (Eds.). Elsevier Biomedical Press, New York,
istics and possible changes with age in the olfactory region of pp. 99–144.
man. Acta Otolaryngol. 71:49–62. Trojanowski, J. Q., Newman, P. D., Hill, W. D., and Lee, V. M-Y.
Nakashima, T. Kimmelman, C. P., and Snow, J. B. (1984). (1991). Human olfactory epithelium in normal aging,
Structure of human fetal and adult olfactory neuroepithelium. Alzheimer’s disease, and other neurodegenerative disorders.
Arch. Otolaryngol. 110:641–646. J. Comp. Neurol. 310:365–376.
Paik, S. I., Lehman, M. N., Seiden, A. M., Duncan, H. J., and von Brunn, A. (1892). Beiträge zur Mikroskopischen Anatomie
Smith, D. V. (1992). Human olfactory biopsy: the influence of der Menschlichen Nasenhöhle. Arch. Mikr. Anat. 39:632–651.
age and receptor distribution. Arch. Otolaryngol. Head Neck Widdicombe, W. G. and Wells, U. M. (1982). Airway secretions.
Surg. 118:731–738. In The Nose: Upper Airway Physiology and the Atmospheric
Pansky, B. (1979). Review of Gross Anatomy, 4th ed MacMillan Environment, D. F. Proctor and I. B. Andersen (Eds.). Elsevier
Publishing Co., Inc. New York, p. 35. Biomedical Press, New York, pp. 215–244.
Read, E. A. (1908). A contribution to the knowledge of the olfac- Wolfsdorf, J., Swift, D. L., and Avery, M. E. (1969). Mist therapy
tory apparatus in dog, cat and man. Am. J. Anatomy 8:17–47. reconsidered: an evaluation of the respiratory deposition of
Rehn, B., Breipohl, W., Schmidt, C., Schmidt, U., and labelled water aerosols produced by jet and ultrasonic nebu-
Effenberger, F. (1981). Chemical blockade of olfactory per- lizers. Pediatrics 43:79–808.
ception by N-methyl-formimino-methylester in albino mice. Won, J., Mair, E. A., Bolger, W. E., and Conran, R. M. (2000).
II. Light microscopical investigations. Chem. Sens. The vomeronasal organ: an objective anatomic analysis of its
7:317–328. prevalence. Ear Nose Throat J. 79:600–605.
Sasaki, C. T., Suzuki, M., Fukuda, H., and Mann, D. G. (1977). Yamagishi, M., and Nakano, Y. (1992). A re-evaluation of the
Dilator naris muscle. Ann. Otol. Rhinol. Laryngol. classification of olfactory epithelia in patients with olfac-
86:362–370. tory disorders. Eur. Arch. Oto-Rhino-Laryngol. 249:
Schneider, R. A. and Wolf, S. (1960). Relation of olfactory activ- 393–399.
ity to nasal membrane function. J. Appl. Physiol. 15:914–920. Yamagishi, M., Hasegawa, S., and Nakano, Y. (1988).
Schultz, E. W., and Gebhardt, L. P. (1934). Olfactory tract and Examination and classification of human olfactory mucosa in
poliomyelitis. Proc. Soc. Exp. Biol. Med. 31:728–730. patients with clinical olfactory disturbances. Arch. Oto-
Smith, C. G. (1941). Incidence of atrophy of the olfactory nerves rhinolaryngol. 245:316–320.
in man. Arch. Otolaryngol. 34:533–539. Zide, B. M., and Jelks, G. W. (1985). Surgical Anatomy of the
Smith, T. D., Siegel, M. I., Burrows, A. M., Mooney, M. P., Burdi, Orbit. Raven Press, New York, p. 5.
A. R., Fabrizio, P. A., and Clemente, F. R. (1998). Searching Zippel, H. P. (1993). Historical aspects of research on the vertebrate
for the vomeronasal organ of adult humans: preliminary olfactory system. Naturwissenschaften Aufsätze 80: 65–76.
2

Morphology of the Mammalian Olfactory Epithelium:

Form, Fine Structure, Function, and Pathology

Bert Ph. M. Menco

Northwestern University, Evanston, Illinois, U.S.A.

Edward E. Morrison
Auburn University, Auburn, Alabama, U.S.A.

I. INTRODUCTION largest gene family in multicellular organisms, including

humans (Glusman et al., 2001).
In most mammals, chemicals, particularly volatile ones, This chapter reviews the morphology and structure of the
are sensed by several intranasal systems, the main olfac- main olfactory epithelium of mammals, including humans,
tory organ (tuned to odors in general), the vomeronasal or in health and disease. The relationship between morphology
Jacobson’s organ (tuned to chemicals employed in social and biochemistry and physiology is addressed when pos-
and sexual activities) (Evans, 2002) (see Chapter 46), the sible, and, when appropriate, contrasts are made with some
septal olfactory organ (a patch of olfactory tissue on the of the other chemosensory systems, such as the vomeronasal
anterior septum of some vertebrates that likely responds to organ. The reader is referred elsewhere for reviews on the
the same agents as the main olfactory system, and perhaps structure of the chemosensory systems of a wide range of
serves an alerting role) (Adams, 1992), and the trigeminal species, including higher-order projection centers (Menco,
intranasal somatosensory system (responsive to pungent 1992a; Smith, 1998; Tolbert, 1993).
and irritative odors) (see Chapter 47). Humans likely do
not possess a septal organ, and their vomeronasal organ is
rudimentary and nonfunctional (Giorgi et al., 2000; Smith II. EARLY OBSERVATIONS
et al., 2001; Trotier et al., 2000).
The chemosensory neurons of main, septal, and Massa (1536) and Scarpa (1789) first described the course
vomeronasal olfactory organs have several characteristics of human olfactory nerves (see Graziadei, 1971; Seifert,
that set them apart from neurons of the central nervous 1969). These early investigators chronicled the presence of
system. First, because of their peripheral location they fine delicate bundles of nerves that originate from the nasal
are exposed to the external environment. Second, their cavity mucosa and extend into the cranial vault, attaching
axons project directly to the forebrain without synapsing in directly to the brain. The early histological investigations
the thalamus (see Chapters 7 and 8). Third, they have a were limited by the difficulty in obtaining fresh olfactory
remarkable capacity for continued postnatal neurogenesis, tissue, unsuitable fixation and staining methods, and the
even into old age (see Chapters 5 and 6). Fourth, reflecting overall poor quality of available microscopes. For example,
a need to distinguish between many compounds, the Ecker (1855) thought that the olfactory region lacked cilia,
subgenome of the main olfactory system comprises the and others reported that only the sustentacular cells have

17
18 Menco and Morrison

Figure 3 Human olfactory dendritic ending (scanning electron

micrograph) with proximal parts of olfactory cilia surrounded by,
Figure 1 Light microscopy of rodent olfactory neuroepithelium in this case short, microvilli of nearby supporting cells. Bar
1
illustrates an apical row of supporting cell (s) nuclei, olfactory m. (After Morrison and Costanzo, 1990.).
receptor cell nuclei (o) that occupy the basal two thirds of the Figure 4 As in Figure 3, but in an 18-day-old rat embryo. Cilia,
epithelium compartment, and globose (g) and horizontal basal radiating from the dendritic ending, have 1–2 m long thick
cells (h). Bar
25 m. proximal parts (about 0.3 m across, fat arrow) that taper to 0.1
m (large arrow). Developing dendritic endings nearby may just
have one (primary) cilium (small arrow) in embryos. Microvilli
(asterisk) of olfactory epithelial supporting cells intermingle with
receptor cell cilia in a course perpendicular to these. Bar
1 m.

flagellum-like processes. Improved optical lenses, new

chemical dyes that yielded superior staining techniques,
and methods for obtaining tissue quickly (thereby allowing
better fixation and tissue preparation) developed in the late
nineteenth century provided opportunities for more accur-
ate and detailed descriptions of the olfactory mucosa.
Early histologists (Ecker, 1855; Krause, 1876; Schultze,
1856) showed that the vertebrate olfactory mucosa was pri-
marily composed of three cellular components: olfactory
receptor, supporting cells, and basal cells (for reviews, see
Seifert, 1969; Zippel, 1993). Schultze (1862) provided what
is thought to be the first accurate description of the vertebrate
olfactory mucosa. He suggested that olfactory cilia were the
endings of olfactory nerves. Further studies on a variety of
animals over the course of the following 140 years support-
ed Schultze’s astute observations. He also emphasized the
uniform structural pattern of the olfactory mucosa in verte-
brates (Fig. 1) (Graziadei, 1973; Seifert, 1969).
Parker (1922) was apparently the first to propose that
olfactory cilia contain the receptive elements for olfactory
Figure 2 High-voltage transmission electron micrograph of a receptor neurons. Such cilia, located on the exposed den-
section (1 m) parallel to the olfactory epithelial surface through dritic tips of olfactory receptor neurons (Figs. 2–15), were
an olfactory mucus layer (calf, septum). Cilia (C) radiate from the hypothesized to increase the sensory surface area available
dendritic endings (De) of olfactory receptor cells. The tissue was for contact with odors. Todd and Bowman (1847) were the
conventionally fixed and embedded. Bar
10 m. first to describe glands in the lamina propria. These glands,
Morphology of the Mammalian Olfactory Epithelium 19

Figure 6 As in Figure 5, but in a rat. Whereas in the adult human

Figure 5 Surface of human olfactory epithelium (scanning
of Figure 5 the thin tapering parts of the cilia are no longer
electron micrograph) where thin parts of olfactory cilia form a
aligned, this is not true in a healthy young (2 months old) rat.
blanket covering the epithelial surface. The opening of a
Thin parts of olfactory cilia of many dendritic endings run paral-
Bowman’s gland duct containing secretory product (asterisk) can
lel to each other. The cilia contain many expansions. Some of
be seen. Bar
10 m.
these may be a fixation artifact, whereas others are genuine. Tips
of supporting cell microvilli “peep” through the parallel array of
cilia (asterisks). Bar
1 m.
later named after Bowman by von Kölliker (1858), pro-
duce mucous secretions that reach the epithelial surface;
olfactory cilia are embedded within this mucus (Fig. 2).
Milne-Edwards (1844) proposed that the specialized  2 microtubule arrangement, typical of almost all cilia and
secretions form a microenvironment that enhances odorant flagella (Bloodgood, 1990; Burton, 1992), that there were
absorption, a hypothesis that is still posed today (see over 1000 cilia per receptor cell, and that these cilia were 1–2
Chapter 3). Thus, odorant-binding proteins within the m in length and 0.1 m in diameter. Later studies showed
nasal mucus are thought to enhance odorant-receptor that the number of cilia was a significant overestimate that
functioning (Pelosi, 1994). Such proteins may be espe- may have been due to respiratory metaplasia within the
cially active in VNO signaling (Pelosi, 2001; Tegoni et al., neuroepithelium. Also, we now know that olfactory cilia are
2000). actually much longer and that only the distal parts of these
After World War II, the development of the electron cilia have the smaller diameter (Figs. 4–15) (Menco, 1977,
microscope ushered in a new era of the study of cell 1983; Seifert, 1970, 1972). Engström and Bloom also
structure. Engström and Bloom (1953) provided the first showed a presence of mitochondria within the receptor cell
electron microscopic observations of the human olfactory dendrites and a system of intracellular membranes, later
epithelium. They determined that olfactory cilia have a 9(2) identified as endoplasmic reticulum.
 Figure 7 Freeze-fracture platinum/
carbon (Pt/C) replica transmission
electron micrograph of an olfactory
receptor cell dendritic ending with
radially oriented cilia (rat, fixed).
Membranes of the cilia have an array
of spiraling particles at their base, the
ciliary necklace (arrow). Several cilia
are seen cleaved just below these
necklaces (arrowhead). The base of
the dendritic ending displays a tight-
junctional belt. The area of this belt
marked with an asterisk is the region
where two other cells joined this cell.
The dendritic ending and its cilia dis-
play many globular membrane parti-
cles, whereas those in the apices of the
surrounding supporting cells are most-
ly dumbbell-shaped (curved arrow)
(Menco, 1980a, 1984, 1988a). As
these particles represent transmem-
brane proteinaceous or lipidic entities
(Menco, 1986), the distinctions give
some indication of molecular differ-
ences between cells and also between
subcellular regions. Bar
1 m.

Figure 8 Freeze-substituted, unfixed rat olfactory epithelial surface (see Menco, 1995a, b, for techniques). Cilia that originate from
the dendritic endings have thick proximal parts with a complete 9(2)  2 axonemal configuration (large arrow, see Fig. 10), that taper
to thin long distal parts that have most commonly two microtubules (small arrows, Fig. 11) and that align parallel to the epithelial sur-
face (see Figs. 6 and 9). Dendrites are packed with microtubules and associated proteins, basal bodies (small asterisks), and mito-
chondria. Supporting cell microvilli (large asterisk) have a course perpendicular to that of the cilia (see Figs. 3, 4, 6, and 27), their
tips reaching the mucus surface. Bar
1 m.
Morphology of the Mammalian Olfactory Epithelium 21

Figure 10 Cross section through the proximal part of a rat

olfactory cilium with a complete 9(2)  2 microtubule axonemal
configuration (see also Burton, 1992). Bar
0.1 m.

turbinates in other vertebrates, such as dog and fox (Morrison

et al., 1983; Negus, 1958), which can expand over many cen-
timeters. The nonsensory respiratory region is covered by a
stratified columnar epithelium consisting of ciliated/microvil-
lous cells interspersed with goblet cells. This epithelium pre-
dominantly lines the inferior, middle, and a portion of the
superior turbinate. The mesenchyme below the basement
membrane contains diffuse lymphoid tissue and blood vessels
and mucous and serous glands. During development, this
Figure 9 Freeze-fracture replica (Pt/C) transmission electron layer plays a major role in the formation of the olfactory path-
micrograph of an olfactory epithelial mucus layer (rat, fixed) way by way of inductive signals (LaMantia et al., 2000).
depicting parallel distal parts of olfactory cilia (small arrow), that Mucous secretions pass through glandular ducts that extend to
emerge from the thicker proximal parts (large arrow; see Figs. 4 the mucosal surface. The nonsensory respiratory portion of
and 8). Ciliary membranes are studded with particles reflecting a the nasal cavity warms, cleans, and humidifies the inspired air.
heterogeneous population of proteins (see also Figs. 7 and
12–14). Bar
1 m.

III. ANATOMY OF THE OLFACTORY MUCOSA

The human nasal cavity typically has three structures extend-

ing from each lateral wall of the ethmoid, termed the inferior,
middle, and superior “turbinates” or “conchae.” Other animals
can have more (Negus, 1958), e.g., the rat has four (Menco
and Jackson, 1997a). The turbinates and septum, the latter
being a cartilaginous structure that separates both halves of
the nose, are covered with an epithelium that, depending on
its location, is either nonsensory (respiratory) or sensory
(olfactory). The human olfactory neuroepithelium is located
high in the superior region of the nasal vault (Chapter 1).
From cadaver measurements, this region appears to be
approximately 1–2 cm2, varying among individuals (Moran Figure 11 Cross section through the distal parts of rat olfacto-
et al., 1982a). This area is modest relative to the nasal ry cilia that here have two to four microtubules. Bar
0.1 m.
22 Menco and Morrison

Figure 12 Parts of distal segments of rat (fixed) olfactory cilia, rotary-replicated with Pt/C from a 45° angle. The cilia have smaller (small
arrows) and larger (large arrows) membrane particles, that may reflect different transmembrane proteins. Bar
0.1 m.
Figure 13 Part of the distal segment of a rat olfactory cilium (unfixed), rotary-replicated with tantalum/tungsten (Ta/W) from a 20°
angle. Some larger membrane particles in this fine grain replica have pores (large arrows), that may reflect ion channels. Smaller parti-
cles lack pores (small arrows). Bar
0.1 m.
Figure 14 Part of the distal segment of a rat olfactory cilium (large arrow), fixed and labeled with the lectin wheat germ agglutinin
(WGA, binding to N-acetylglucosamine residues) conjugated to 5 nm gold grains, rapidly frozen, etched, and rotary-replicated with Pt/C
from a 45° angle (see Fig. 12). WGA bound to several surface particles, reflected as dark dots inside the particles (small arrows) (Menco,
1992b). Surrounding mucus is virtually free of labeling, suggesting that the label bound to molecules specific to the ciliary surface. Bar

0.1 m.

Trapped dust and other particulate matter is transported to the Jackson, 1997b; Moran et al., 1982b; Pixley et al., 1997).
nasopharynx by ciliary movements. The vasculature of the A discussion of the structure and, to some degree, func-
nasal cavity forms an erectable plexiform network beneath tion of most of these cell types is presented below.
the mucous membrane (Proctor and Anderson, 1982).
An adaptation, possibly peculiar to nasal septum and A. Olfactory Receptor Neurons
turbinates of primates, including humans, are small
invaginations termed “olfactory pits” that may serve to Vertebrate olfactory receptor neurons are slender and
better protect some sectors of olfactory epithelium from bipolar, have 5- to 7-m-wide cell bodies that are generally
external damage or, though minimally, enhance the located within the lower two thirds of the neuroepithelium,
receptive surface area in these species (Feng et al., 1997). ciliated dendrites, and occur in densities of 106–107 per cm2
Otherwise, the human olfactory epithelium has a struc- (Güntherschulze, 1979; Menco, 1983). The single dendrites
ture similar to that of other vertebrates. It is pseudostrati- of olfactory receptor cells can take a rather tortuous path,
fied columnar, composed of olfactory receptor neurons, winding around adjacent receptor cell bodies, other den-
nonsensory supporting cells, and two types of basal cells, drites, and supporting cells as they extend toward the
horizontal (HBC) and globose (GBC) (Fig. 1). It also mucosal surface. Each dendrite is slightly thicker near the
contains leukocytes (Suzuki et al., 1995) and other cells soma and contains a Golgi body, smooth and rough
besides supporting cells that have microvilli (Menco and endoplasmic reticulum, mitochondria, microtubules, and
Morphology of the Mammalian Olfactory Epithelium 23

Figure 15 Diagram of mammalian olfactory (top) and respiratory (bottom) cilia, and of lengths of olfactory cilia (inset center) (after
Menco, 1977, 1983). Features in the three diagrams have been drawn to scale. The olfactory cilium is interrupted at two places, indicating
that the cilia are actually much longer (inset). A–E represent basal body cross sections; F–H: cross sections through proximal regions of
olfactory cilia (top) and homologous regions of respiratory cilia (bottom); I–K: cross sections through distal parts of olfactory cilia. The
section of Figure 10 is similar to cross section G. Figure 11 shows the cross sections J and K. Other structures: R: striated rootlet of res-
piratory cilium; 1: fibrogranular microtubule pool (cilium precursor pool); 2: microtubules inside dendritic endings; 3: microvilli of dendritic
endings (sparse) and of ciliated respiratory cells; 4: coated vesicles (Bannister and Dodson, 1992); 5: ciliary necklaces (see Fig. 7; 7 strands
for olfactory cilia, 5 strands for respiratory cilia) (Menco, 1980c); 6: ciliary membranes studded with membrane particles, that reflect
proteins. Olfactory cilia have many more of these than respiratory cilia (see Figs. 7, 9, and 12–14) (Menco, 1977, 1983, 1992a. 1997);
7: nearby glycocalix; 8: bundle of tapers of other, nearby, cilia; 9: vesiculated expansion along distal part of cilium (see Figs. 6 and 9); 10:
ciliary tips; olfactory cilia terminate in a small vesicle. The inset demonstrates that the cilia of one receptor cell dendrite can extend over
about 15 other endings. Olfactory cilia are drawn over about 60 m, that is, 120 m from the tip of one cilium to the tip of an other cilium
across (Seifert, 1970). Bar top and bottom
1 m; center: 10 m.

vesicles. Some of these features can be seen in Figure 8 and measure 1–2 m in diameter (Figs. 2–4, 7, 8). At the
(Andres, 1969; Bannister and Dodson, 1992; Burton, apical surface, just below the dendritic ending, a belt-like
1992; Naguro and Iwashita, 1992). Dendrites have variable tight-junctional complex—a transmembrane barrier,
lengths, mostly extending nearly the total depth of the characteristic of most epithelial tissues — attaches the den-
epithelium, from the surface deep into the epithelium, but drite to adjacent supporting cells, as well as supporting
some are extremely short, only a brief distance from cell cells to other supporting cells (Fig. 7) (Kerjaschki and
bodies lying close to the epithelial surface (Moran et al., Hörandner, 1976; Menco, 1980b, 1988c). Receptor cells
1982a: Morrison and Costanzo, 1990, 1992). are closely associated with olfactory supporting cells at
Dendrites end in a swelling at the epithelial surface, other levels as well, where desmosomes are found between
called the olfactory dendritic ending, knob, or vesicle. them. Olfactory dendritic endings contain basal bodies
Olfactory dendritic endings extend usually, but not always, (Burton, 1992). Many of these give rise to sensory cilia that
above the epithelial surface, are spherical or cylindrical, project perpendicularly from the dendritic ending into the
24 Menco and Morrison

overlying mucus layer. Each cilium consists of a short prox- olfactory cilia have a special morphology, as they are stud-
imal part that tapers to a longer and thinner distal part, ded with numerous (1000–2000/m2), intramembranous
which aligns parallel to the epithelial surface. Thus, olfac- particles (Figs. 7, 9, 12 – 15). The density of these particles
tory cilia are much longer than the nonsensory cilia of the in olfactory cilia is about twice that of such particles of
respiratory epithelium. The distal, aligned, parts of the motile cilia of respiratory ciliated cells (Fig. 15) (Menco,
olfactory cilia form the interface between the external odor- 1977, 1980a, 1983, 1992b, 1997). Particles of olfactory
ous environment and the luminal surface of the olfactory cilia may reflect odorant receptors (Buck, 1996, 2000;
epithelium (Figs. 2 – 9, 15, 27). In most mammals, includ- Buck and Axel, 1991; Mombaerts, 1999), but also the
ing humans, lengths of olfactory cilia are around 50 m transmembrane signaling proteins Type III adenylyl
(Figs. 5, 6) (Seifert, 1970). In nonmammalian vertebrates, cyclase (AC) (Bakalyar and Reed, 1990; Krupinski et al.,
such as the frog, they can be as long as 200 m (Reese, 1989) and olfactory cyclic-nucleotide gated (CNG) chan-
1965). Individual olfactory receptor cells possess 1–50 sen- nels (Dhallan et al., 1990). Ultrastructural research
sory cilia (Figs. 2 – 4) (Chuah and Zheng, 1992; Menco, (reviewed in Menco, 1997) supports physiological and bio-
1983; Menco and Farbman, 1985b; Morrison and chemical evidence (Ache and Restrepo, 2000; Paysan and
Costanzo, 1990, 1992; Ohno et al., 1981). Thus, sensory Breer, 2001; McClintock, 2000; Nakamura, 2000) (see
cilia number and special morphology result in an increased Chapters 4 and 11) that the cilia contain the biochemical
surface area, as much as 40 times (inset Fig. 15) (Menco, mechanisms of olfactory signal-transduction.
1983, 1992b), for interaction with odors. In somewhat more detail, olfactory signal transduction
As noted earlier, the proximal parts of the olfactory cilia begins when odorants interact with members of the GTP-
have a 9(2)  2 axonemal configuration (Figs. 10, 15). At binding protein (or G-protein)–linked odorant-receptor
their very base they have a ciliary necklace (Figs. 7, 15) superfamily that characteristically traverse the membrane
that consists of spiraling arrays of membrane particles, pre- seven times (Buck, 1996; Buck and Axel, 1991;
sumably special proteins. Both features are typical of Mombaerts, 1999; Sullivan and Dryer, 1996). This stimu-
almost all forms of cilia. However, olfactory cilia have lus receptor interaction leads to activation of a G-protein,
more of such spiraling strands than nonsensory respiratory probably Golf , but perhaps Gs (especially in embryos) as
cilia (Menco, 1980c, 1988b; Naguro and Iwashita, 1992). well (Belluscio et al., 1998; Menco et al., 1994). The G-pro-
The exact function of the necklaces is still unclear, but it tein  subunits, Golf and Gs, activate calcium(Ca2)/
has been suggested that they may serve as anchors, mole- calmodulin-sensitive Type III AC, making cyclic AMP
cular barriers, and calcium-binding entities (see references (cAMP). The cAMP opens CNG channels. This results in
in Menco, 1980c, 1988b; Plattner and Klauke, 2001). The an electrical signal (Belluscio et al., 1998; Brunet et al.,
thin membrane-lined distal parts of mammalian olfactory 1996; Gold and Nakamura, 1987; Jones and Reed, 1989;
cilia have only one to four, but most commonly two, Kleene, 1994; Wong et al., 2000; reviewed by Nakamura,
microtubules inside (Figs. 11, 15). Mammalian olfactory 2000; Schild and Restrepo, 1998) (see Chapter 4). Fine-
cilia, including those of humans, are not intrinsically structural studies have shown that all proteins involved in
motile, unlike the case in some other vertebrates (Lidow the onset of the AC/cAMP cascade are highly concentrated
and Menco, 1984). The nine doublets in the proximal parts in the olfactory cilia, particularly the distal parts. This
lack dynein arms (Fig. 10). Dynein is a Mg2-ATPase pro- includes odorant receptors (Figs. 16, 17) (Menco et al.,
tein necessary to generate the force for cilium motility 1997), Gs and Golf , Type III AC (Asanuma and Nomura,
(Stephens, 1974). The overall fine structure of the receptor 1991; Mania-Farnell and Farbman, 1990; Menco et al.,
cell cilia of the septal olfactory organ resembles that of 1992, 1994), and CNG channels (Fig. 18) (Matsuzaki et al.,
those of the main olfactory organ (Adams, 1992; Miragall 1999a) (Fig. 27 and Table 1).
et al., 1984). Regulators of G-protein–signaling (RGS) proteins are
The blanket of sensory cilia covering the olfactory a group of GTPase-activating proteins (GAPs) (Kehrl,
region varies with location. Some regions of the septum and 1998). These RGS proteins have recently also been
superior turbinates can have a dense, matted sensory ciliary identified in rodent (Norlin and Berghard, 2001) and
surface (e.g., Figs. 5, 6), whereas adjacent regions can have canine olfactory (unpublished) and vomeronasal systems,
only a few scattered olfactory receptor cells. This topo- with some showing spatial restrictions correlating with
graphic distinction is discussed in Sec. IV of this chapter. other olfactory signaling molecules (Ressler et al., 1993;
Since the aforementioned early investigations by Vassar et al., 1993). RGS proteins were first discovered
Schultze (1856, 1862), cilia have been suspected of in the yeast Saccharomyces cerevisiae and the nematode
harboring the odorant receptors (Menco, 1977; Parker, Caenorhabditis elegans. To date, 19 mammalian genes
1922; Rhein and Cagan, 1981). The plasma membranes of are known to encode RGS-cognate sequences. In the
Morphology of the Mammalian Olfactory Epithelium 25

Figures 16 and 17 Two nearby sections through the same mouse olfactory epithelial surface labeled with polyclonal antibodies to puta-
tive odorant receptor M4 (dilution: 1:100, arrowhead). Proximal (large arrow) and distal segments (thin arrow) of cilia of one receptor
cell dendritic show binding, while those of nearby receptor cells do not (asterisk). Freeze-substituted tissue was fixed with paraformalde-
hyde (Menco et al., 1997). Gold particles, conjugated to secondary goat-anti-rabbit antibodies, are 10 nm across. Bar
1 m.
26 Menco and Morrison

Figure 18 Distal parts of mouse olfactory

cilia (thin arrow) are immunopositive for poly-
clonal antibodies to -subunits of CNG chan-
nels (dilution: 1:25), unlike dendritic endings,
proximal cilium parts (large arrow), and sup-
porting cell microvilli (asterisk). Unfixed tissue
was rapidly frozen and freeze-substituted
(Matsuzaki et al., 1999a). Gold particles, conju-
gated to goat-anti-rabbit antibodies, are 10 nm
across. Bar
1 m.

main olfactory system RGS2 probably contributes to the 1980c, 1988b) contains regulatory components as well as tar-
ability of olfactory neurons to discern odors by con- get structures involved in Ca2 signaling (Plattner and
trolling AC activity (Sinnarajah et al., 2001). Klauke, 2001). Other fine structural evidence suggests that
Alternative routes, particularly in invertebrates (Hatt olfactory cilia possess at least some signal-terminating and
and Ache, 1994), may work through activation of a phos- -modulating proteins, phosphodiesterases (PDEs; Asanuma
pholipase C (PLC)/trisphosoinositide (IP3) cascade. and Nomura, 1993) and Na, K-ATPase (Table 1 and Fig.
G-proteins are thought to be the catalysts for these routes 27) (Kern et al., 1991; Menco et al., 1998). The latter may
as well (Nakamura, 2000; Schild and Restrepo, 1998). play a role in the restoration of the receptor potential. CO and
However, in vertebrates evidence for a role of the PLC/IP3 NO conceivably also help to regulate and modulate olfactory
cascade in olfactory signaling, though present (Cadiou signaling. However, heme-oxygenase-2 immunoreactivity is
et al., 2000; Vogl et al., 2000), is ambiguous: knockout found in perinuclear regions of receptor cells rather than in
mice studies favor the cAMP/AC cascade (Belluscio et al., their cilia (Wenisch et al., 2001), and cytochemical activities
1998; Brunet et al., 1996; Wong et al., 2000), and two pro- of proteins involved in CO and NO metabolism are not
teins thought to be involved in the PLC/IP3 cascade, Gq restricted to the receptor cells, but also involve supporting
(DellaCorte et al., 1996) and IP3 receptors (Cunningham cells (Wenisch et al., 2000). In vertebrates, fine structural
et al., 1993; Kalinoski et al., 1994), have been localized to localization is lacking for odorant-binding proteins
supporting cell microvilli besides receptor cell cilia. (Bastianelli et al., 1995), but in insects binding proteins for
For some other proteins conceivably involved in olfactory pheromonal compounds occur in different antennal
signal onset and most proteins putatively involved in signal hemolymph compartments than those for more general
termination and signal modulation, the exact subcellular odorants (Steinbrecht, 1999).
location is less clear (Table 1). The former includes several Antibodies to putative odorant receptors label in par-
proteins implemented in the multiple roles potentially ticular cilia and only those of a few receptor cells (Figs.
played by Ca2 in olfactory signaling (Lindemann, 2001; 16, 17) (Menco et al., 1997), the latter as would be pre-
Nakamura, 2000; Schild and Restrepo, 1998; Zufall and dicted from in situ hybridization studies (Buck, 1996;
Leinders-Zufall, 2000), such as Ca2-exchanger (Noë Buck and Axel, 1991). Interestingly, several other proteins
et al., 1997) and -binding proteins (Kishimoto et al., 1993; are also present in only a select group of olfactory recep-
Yamagishi et al., 1993). However, fine structural energy- tor cells. One of these is the heat shock protein 70
dispersive x-ray microanalysis suggests that Ca2-gated Cl (HSP70) (Carr et al., 1994), which is confined to a subset
channels in olfactory cilia conduct inward currents carried by that is much smaller than that of cells expressing specific
Cl efflux into the mucus (Reuter et al., 1998). In motile cilia odorant receptors (Buck and Axel, 1991).
there is evidence that the ciliary necklace (Fig. 7) (Menco, Immunoreactivity for HSP70 is present throughout the
Morphology of the Mammalian Olfactory Epithelium 27

Table 1 Olfactory Signal-Onset Molecules and the Signal-Modulating and -Terminating Molecules Targeting Them
Signal RGS2 PKC γ, δ, and
modulation λ (?)

Signal Odorant Golf ␣ Type III AC (minor cell CNG channel

onset receptors population: GC)

Signal GRK3, PKA, PDE (PDE1C2, PDE4A?), CaMKII, CaM

termination PKC (?) -arrestin-2, (PDE2 for GC)
Horizontal arrows give the sequentially activated onset molecules, from odorant receptor to current generating channel. Exact locations
of these molecules have been established (bold italics; see also Fig. 27). For all molecules marked in plain italics, especially those of sig-
nal modulation and signal termination, exact such knowledge is still absent. The vertical arrows near signal-terminating and -modulat-
ing molecules point to the part of the signal-onset cascade targeted by these molecules. Signal-onset molecules: Golf
olfactory
GTP-binding protein; AC
adenylyl cyclase; CNG channels
cyclic nucleotide-gated channels. Signal-termination molecules:
Protein kinases A and possibly C (PKA and PKC) may act in concert with GRK3 (formerly called -adrenergic receptor kinase-2 or -
ARK2) (Borisy et al., 1982; Dawson et al., 1993; Peppel et al., 1997; Schleicher et al., 1993) or other GRKs at the level of odorant recep-
tors (Boekoff and Breer, 1992; Breer, 1994). CaM interacts with CNG channels (Chen et al., 1994; Kurahashi and Menini, 1997), while
phosphodiesterases (PDEs) (Firestein and Shepherd, 1991), possibly CaM-activated PDE (PDE1C2, formerly called CaM-PDE) (Borisy
et al., 1982; Juilfs et al., 1997; Yan et al., 1995), -arrestin-2 (Dawson et al., 1993), and CaM kinase II (CaMKII) (Wei et al., 1998) act
at the level of AC. Some PDEs other than PDE1C2, PDE4A (formerly called PDE2) (Cherry and Davis, 1995) and PDE2 (not to be con-
fused with PDE4A, formerly called PDE2), may also be involved in signal termination. PDE2 is expressed in a minor population of
receptor cells that use guanylyl cyclase (GC) instead of AC (Gibson and Garbers, 2000; Juilfs et al., 1997) and, likely, cGMP-selective
CNG channels instead of cAMP-selective CNG channels (Meyer et al., 2000). Ultrastructurally GC has also been localized to olfactory
dendritic knobs and cilia, but to supporting cell apices and microvilli as well (Spreca and Rambotti, 1994). Signal-modulation mole-
cules: A regulator of G-protein signaling proteins [RGS2; RSGs act as GTPase-activating proteins (GAPs)] attenuates odorant-elicited
cAMP production (Sinnarajah et al., 2001). Besides a possible role in signal termination, PKCs (, , and
) may modulate signals by
increasing CNG channel sensitivity. For reasons of comprehension, not all molecules that may be involved in olfactory signaling, such
as several Ca2-binding proteins, are included in this table, but see text (Müller et al., 1998; also Nakamura, 2000; Schild and Restrepo,
1998) (see also Chapter 4).

cytoplasm but seems to exclude the cilia (Fig. 19) where- Table 1). The two cellular subsets considered here are likely
as odorant receptors are primarily located in membranes not the same, as carbonic anhydrase positive cells also occur
of receptor cell cilia (Figs. 16, 17) (Menco et al., 1997). in the nasal respiratory epithelium. In the VNO microvilli of
The implication of the fact that morphologically similar receptor cells are thought to be the subcellular sites that
receptor cells can have different, membranous as well as interact with incoming VNO-targeted odors (see Chapter
cytoplasmic, proteins is unclear. 46). Indeed, analogous to the cilia of main olfactory epithe-
There are at least two other, larger, subsets of olfactory lium receptor cells, these microvilli are selectively enriched
epithelial receptor cells. Both of these may have specific in proteins putatively involved in VNO signal transduction
roles. One of these is a subset of receptor cells that displays (Matsuoka et al., 2001; Menco et al., 2001).
carbonic anhydrase activity (Brown et al., 1984; Coates Olfactory marker protein (OMP) is a low molecular
2001; Okamura et al., 1996). Carbonic anhydrase is a zinc- weight soluble protein. It may, either directly or indirectly,
dependent metalloenzyme that catalyzes the reversible modulate part of the olfactory signaling cascade (Buiakova
hydration of CO2 to produce HCO3 and H. This enzyme et al., 1996) and olfactory neurogenesis (Carr et al., 1998).
is thought to play a role in CO2 chemoreception (Coates, Labeling for OMP is rather evenly distributed throughout
2001). The other subset is one that uses guanylyl cyclase and receptor cells that have sprouted cilia. The labeling includes
cGMP-gated CNG channels instead of ACIII and cAMP- the cilia (Johnson et al., 1993; Margolis, 1988; Menco,
gated CNG channels. These cells of this subset terminate in 1989). Thus, the labeling pattern for OMP differs from
a special region of the olfactory bulb, the so-called necklace that of the antibodies to the signaling proteins that
region.* These cells may, like vomeronasal receptor cells, be label most receptor cells, such as those to CNG channels
involved in certain aspects of conspecific recognition (Fig. 18). The latter label the cilia much more prominently
(Gibson and Garbers, 2000; Meyer et al., 2000) (see also than other cellular compartments (Menco et al., 1992,
1994). However, antibodies to OMP and signaling proteins
*The bulbar necklace region, a glomerular cellular assembly, have in common the fact that they label many receptor cells.
should not be confused with the ciliary necklaces mentioned ear- This contrasts with the labeling patterns of antibodies to
lier. The latter are special subcellular structures at the base of cilia. odorant receptors (Figs. 16, 17), HSP70 (Fig. 19), and
28 Menco and Morrison

Figure 19 An olfactory receptor cell dendrite

(large asterisk) shows binding for monoclonal
antibodies to HSP70 (undiluted) (Carr et al.,
1994) throughout its cytoplasm apart from that
inside the cilia (large arrow). This receptor cell
is the only one out of thousands that displays
labeling; an unlabeled dendrite can be seen
nearby (small asterisk). Surrounding support-
ing cells, including their microvilli, are also
devoid of label. The tissue was fixed with
paraformaldehyde and glutaraldehyde before
cryofixation and freeze-substitution (Griffith
1993; Menco, 1995b). Gold particles, conju-
gated to secondary goat-anti-rabbit antibodies,
are 15 nm across. Bar
1 m.

cGMP cascade proteins and of carbonic anhydrase axon fibers but extend tongues of cytoplasm that wrap
cytochemistry. bundles of 50 – 200 olfactory axons (see, e.g., Figs. 2D
Olfactory axons arise from the basal region of the recep- and 8A in Doucette, 1992). This unique packaging of
tor cell bodies and transmit information about odor intensity axons, in direct contact with one another, provides the
and quality to the brain. Olfactory axons are always unmyeli- opportunity for interaction of fibers in terms of metabo-
nated and unbranched and are among the smallest fibers lism, ionic flux, and electrical currents during transduc-
(0.1–0.7 m) in the nervous system (Fig. 20). Olfactory tion (Eng and Kocsis, 1987; Gesteland, 1986; Zhang et
axons form small intraepithelial bundles, pass through the al., 2000). Olfactory ensheathing cells lack a surrounding
basal lamina, and then combine in larger fascicles, the fila basement membrane and contain GFAP and S-100 pro-
olfactoria. The latter are surrounded by ensheathing or tein, two biochemical markers characteristic of central
Schwann cells (Doucette, 1992). It is noteworthy that odor- nervous system (CNS) astrocytes (Barber and Lindsay,
ant receptors are also expressed in axons and axon terminals 1982; Takahashi et al., 1984). However, not all olfactory
(Harrington et al., 1997; Ressler et al., 1994; Vassar et al., ensheathing cells are immunopositive for both proteins
1994; reviewed by Buck, 1996; Mombaerts, 1999) in addi- (Pixley, 1993). Some resemble astrocytes, while others
tion to olfactory cilia (Menco et al., 1997). This intriguing resemble Schwann cells (Franklin and Barnett, 2000).
finding suggests that these receptor cells use the same odor- They are derived from the olfactory placode, accompany
ant receptors for odor recognition and for axonal targeting to the axons they surround, and cross the peripheral nervous
appropriate secondary mitral and tufted cells in the olfactory system (PNS)-CNS boundary (Doucette, 1992). Thus,
bulb (Mombaerts, 1999; Mori et al., 2000). along with their unique morphological characteristics,
Olfactory ensheathing cells have several unique char- olfactory ensheathing Schwann cells are more similar to
acteristics. For example, they do not surround individual central glial cells than to peripheral Schwann cells.
Morphology of the Mammalian Olfactory Epithelium 29

some present deep within the bulb relative to the olfactory

nerve layer.

B. Supporting Cells

Supporting cells can be distinguished from receptor cells,

which occasionally are found in the upper epithelium, by
their width and their oval and elongated nuclei. Supporting
cells are columnar; they span the neuroepithelium
throughout and taper basally where they attach by foot-like
processes to the basal lamina (Fig. 20). Like receptor cells,
supporting cells exhibit a cellular polarity, also cytochem-
ically (Figs. 24, 25) (Menco et al., 1998 and unpublished).
Olfactory receptor cell bodies, dendrites, and axons are
often surrounded by supporting cell sleeve-like extensions.
Scanning microscopy has shown many fine cellular exten-

Figure 20 Olfactory axons (human, transmission electron micro-

graph) form small intraepithelial fascicles that exit through the
basal lamina (arrow) into the underlying lamina propria. O, olfac-
tory receptor cells; S, supporting cells; B, horizontal basal cells.
The tissue was conventionally fixed and embedded. Bar
5 m.

Because of their axon growth–promoting properties, they

may be an important therapeutic asset in nerve reconstitu-
tion following nerve injury (Franklin and Barnett, 2000;
Imaizumi et al., 2000; Ramón-Cueto and Avila, 1998).
After projecting, unbranched, centrally through small
foramina in the cribriform plate of the ethmoid bone
(Fig. 21), the olfactory axons terminate in the olfactory bulb
in characteristic spherical neuropils called glomeruli
(Fig. 23) (see Chapter 7). Within these glomeruli, olfactory
receptor axons form asymmetrical synapses with second-
order mitral and tufted neuronal cells (Fig. 22). Axons of
these second-order neurons project to subcortical and corti-
Figure 21 Intracranial view of the human anterior cranial fossa,
cal regions where higher-level processing of olfactory infor- cribriform plate ethmoid bone region. Olfactory axon fascicles
mation and discrimination occurs (Chapters 8, 9). The (arrows) project from the nasal epithelium through the foramen of
human glomerular layer appears not to exhibit the continuity the cribriform plate to reach the olfactory bulb (OB). It is within
observed in other species (Smith et al., 1991). Glomeruli this region that the delicate olfactory axon fascicles are suscep-
tend to be somewhat smaller (25–100 m) and fewer and are tible to injury, i.e., head trauma. Asterisk: crista galli, d: dura mater.
more widely dispersed than that seen in other mammals, Bar
1 mm.
30 Menco and Morrison

material from the mucus. Mammalian olfactory support-

ing cells, however, do not contain glycoconjugates
characteristic of mucus-producing cells (Foster et al.,
1991). This role is mainly played by Bowman’s gland
cells (see Sec. III. E). Also, supporting cell apical glyco-
proteins differ distinctly from those of surrounding ciliat-
ed receptor and other microvillous cells (Ferrari et al.,
1999; Foster et al., 1992; Menco, 1992c).
The membrane appearance of supporting cell apices
and microvilli is quite different from that of receptor cell
dendritic endings and cilia. Densities of membrane-asso-
ciated particles are considerably higher in the supporting
cell apical structures. Also, membranes of supporting cell
apices contain a special type of rod- or dumbbell-shaped
particle (Fig. 7) (Menco, 1980a, 1988a). These have often
been associated with transport processes within epithelia
(Menco et al., 1998). Indeed, several lines of evidence
suggest that one of the putative functions of the support-
ing cells is to maintain a water and salt balance by way of

Figure 22 Plastic section, 1 m thick, toluidine blue stained, of

human olfactory bulb. Olfactory axons (arrows) from the outer
fiber layer, enter the bulb, and terminate in characteristic neu-
ropile structures, called glomeruli (G). Bar
100 m.

sions, forming multiple contacts with olfactory receptor

cells throughout the epithelium (Breipohl et al., 1974;
Morrison and Costanzo, 1990). The apical part of the sup-
porting cell is covered with long microvilli. These
microvilli extend into the mucus and terminate at the
mucous surface, where they intermingle with the thin parts
of the olfactory cilia (Figs. 6, 8, 27) (Andres, 1969;
Bannister and Dodson, 1992; Naessen, 1971b; Okano
et al., 1967; Seifert, 1970, 1972).
Ultrastructural observations of supporting cells have
shown differences between them and the columnar mucus
secretory goblet cells of the respiratory epithelium.
Unlike that of the latter cells, supporting cell apical cyto-
plasm contains a rich supply of organelles that become
scarce basally (Carr et al., 2001; Moran et al., 1982a;
Naguro and Iwashita, 1992). Cytoplasmic vesicles have Figure 23 Transmission electron micrograph of an olfactory bul-
been observed fusing with apical supporting cell surface bar glomerulus showing receptor cell axon terminals with synaptic
membranes (Bannister and Dodson, 1992), suggesting vesicles (arrows) and dendrites (D) of second-order neurons. The
that the supporting cells release materials in and/or absorb tissue was conventionally fixed and embedded. Bar
1 m.
Morphology of the Mammalian Olfactory Epithelium 31

Figure 24 Olfactory supporting cell microvilli (large asterisk) bind polyclonal antibodies to amiloride-sensitive Na-channels (dilution:
1:5); apical regions of the supporting cells from which the microvilli sprout (arrow), and olfactory receptor cell dendritic knobs (small
asterisk) and cilia (curved arrow) do much less so, if at all. Unfixed tissue was rapidly frozen and freeze-substituted. Protein G, conju-
gated to 5 nm colloidal gold, was used as secondary probe (Menco et al., 1998). Bar
1 m.

transporting channels. For example, supporting cell Vomeronasal epithelial supporting cells lack this array.
microvilli have amiloride-sensitive sodium channels (Fig. Ubiquitination serves to modify proteasomes, multiprotein
24) (Menco et al., 1998), and at least one water channel, complexes involved in the regulated breakdown of pro-
aquaporin Type 3, is present in the lateral membranes of teins. Chains of added ubiquitin enable these proteasomes
these cells (Fig. 25) (Verkman and Mitra, 2000). The lat- to participate in protein degradation (Bonifacino and
ter is in line with findings that aquaporin 3 is present in Weissman, 1998). Supporting cells may also be involved in
basolateral membranes of some ciliated cells (Matsuzaki removing debris of dying cells and act as phagocytes
et al., 1999b). Aquaporin 3 does not appear to be present (Suzuki et al., 1996). Their apical surfaces undergo
elsewhere in the olfactory epithelium. While aquaporins remarkable morphological transformations paralleling
1 and 2 have been found to be immunopositive in VNO endocrine activity during the ovarian cycle (Da Pos and
tissues, this was not the case in main olfactory epithelial Arimondi, 1983; Saini and Breipohl, 1976).
tissues. This includes receptor as well as supporting cells The close association between supporting cells and
(unpublished). receptor neurons (Breipohl et al., 1974) has led to the
Besides playing a role in ion and water regulation (Kern belief that supporting cells have glial characteristics. They
and Pitovski, 1997; Menco et al., 1998), supporting cells are thought to electrically isolate adjacent olfactory recep-
are, together with those of Bowman’s glands, involved in tor neurons and to regulate the potassium concentration in
metabolism of xenobiotic compounds. This includes odor- the extracellular fluid compartment (Graziadei, 1971;
ant metabolism (Chapters 3, 27). The specific expression Morrison and Costanzo, 1989; Rafols and Getchell, 1983).
of an ubiquitin-positive membrane array in supporting cell However, neither supporting cells nor any of the other
supranuclear regions, following excessive odor exposure, olfactory epithelial cells express glial fibrillary acidic
is one sign of supporting cell involvement in metabolism protein (GFAP). This suggests that these cells do not
of xenobiotic compounds (Fig. 26) (Carr et al., 2001). resemble glial ensheathing cells (Ophir and Lancet, 1988;
32 Menco and Morrison

Figure 26 Supranuclear region of olfactory epithelial support-

Figure 25 Lateral membranes of olfactory supporting cells are ing cells of a rat exposed to 1.0 mL lavender essential oil extract
immunopositive for polyclonal antibodies to aquaporin 3 (small for 6 hours prior to sacrifice. -Ubiquitin (dilution: 1:100)
arrows; dilution: 1:100). No other structure is seen labeled, immunoreactivity outlines a conical, somewhat electron-dense
including supporting cell microvilli (large asterisk), olfactory array, in this region (asterisk) (Carr et al., 2001). The array con-
receptor cell dendritic knobs (small asterisk), and cilia (large sists of a heterogeneous assembly of fragmented membranes of
arrow). Freeze-substituted tissue was fixed with paraformalde- organelles normally present in the supranuclear regions, such as
hyde. Gold particles, conjugated to secondary goat-anti-rabbit those of endoplasmic reticulum, Golgi body, and mitochondria.
antibodies, were 15 nm across. Bar
1 m. Tissue treatment was as in Figure 26. n: nucleus of labeled sup-
porting cell. Bar
1 m.

Volrath et al., 1985). Nevertheless, in some mammalian exclusively of supporting cells and horizontal basal cells
species, supporting cells (their apical structures), as well as (Suzuki et al., 2000). Such variations may be due, in part,
receptor cells and Schwann cells, may contain S-100 (S- to altered physiological conditions (Saini and Breipohl,
100) proteins, a biochemical marker for glial cells. In 1976). Other supporting cell heterogeneity may be a
the receptor cells this protein may be involved in micro- specific topographic protein expression (Miyawaki et al.,
tubule assembly. The presence of S-100 in supporting cells 1996), paralleling the topographic expression of odorant
suggests that these cells may share at least some properties receptors in olfactory receptor cells (Mori et al., 2000;
with glial (Schwann) cells (Rambotti et al., 1989). Ressler et al., 1993; Strotmann et al., 1994; Vassar et al.,
A heterogeneous population of supporting cells has 1993) and topographic physiological odor responsivity
been observed in humans and other vertebrates (Costanzo (Scott and Brierley, 1999). Indeed, scanning electron
and Morrison, 1989; Rafols and Getchell, 1983; Yamada, microscopic observations suggest that, at least in part, the
1983). Some areas of the nasal cavity may even consist expression of the odorant-receptor zones is determined by
Morphology of the Mammalian Olfactory Epithelium 33

a distinct morphological appearance of supporting cell

apices as well as of receptor cell apices in each zone
(Menco and Jackson, 1997a). Thus, supporting cell hetero-
geneity may play a role in the formation of the odorant
receptor-specific epithelial zones (Fig. 27).

C. Basal Cells

There are two types of basal cells—horizontal (HBC) and

globose (GBC) (Figs. 1, 20). Both are roughly 4–7 m in
diameter and have a round, centrally located nucleus.
HBCs are found near the basal lamina and contain
keratins, intermediate filaments, or tonofilaments charac-
teristic of proliferating epithelial cells (Holbrook et al.,
1995; Suzuki and Takeda, 1991a, b). They also contain
ecto-5-nucleotidase, a marker for neural development
(Braun and Zimmerman, 1998). Several of the histochem-
ical characteristics of the HBCs are shared with basal cells
of the nasal respiratory epithelium (Holbrook et al., 1995).
GBCs are possibly a heterogeneous population of cells in
themselves (Goldstein and Schwob, 1996). They are
located above the HBCs. Their cytoplasm is more electron- Figure 27 Summarizing diagram of the fine structural localization
lucent and contains basal bodies. GBCs are not immunore- of important olfactory epithelial signal-transduction proteins (see
active for keratin. Animal studies show basal cells to be also Table 1 and Menco, 1997) and of several proteins that may play
stem cells capable of postnatal neurogenesis; mitotic fig- supportive roles in this transduction process. Antibodies to all signal-
ures are evident in the lower epithelial region. In vitro and transduction proteins mainly label cilia, in most cases especially
in vivo evidence suggests that at least some GBCs give rise their distal parts (Cunningham et al., 1994; DellaCorte et al., 1996;
to new olfactory neurons. In rodents, HBCs are more Kern et al., 1991; Matsuzaki et al., 1991a; Menco et al., 1992, 1994,
slowly dividing and replenish the GBCs (Figs. 1, 20) 1997, 1998). Immunoreactivity for OMP includes cytoplasmic com-
(Caggiano et al., 1994; Goldstein and Schwob, 1996; partments of knobs and dendrites (Menco, 1989), whereas, when
Huard et al., 1998; Ohta and Ichimura, 2001; Suzuki and present, HSP70-immunoreactivity (Carr et al., 1994) is localized in
all cytoplasmic compartments of the cells apart from cilia (unpub-
Takeda, 1993; Suzuki et al., 1998). Supporting cell prog-
lished preliminary observations). Supporting cell apices seem to be
enitors may be multipotent basal cells (Caggiano et al., involved in ion (Menco et al., 1998) and water transport (Matsuzaki
1994; Goldstein and Schwob, 1996; Schwob et al., 1994) et al., 1999b), and in detoxification (Carr et al., 2001) (see Chapters
and/or reside in Bowman’s gland ducts (Huard et al., 1998; 3 and 27). The thin parts of the cilia align near the interface
Weiler and Farbman, 1998) (see Chapters 5, 6). mucus/external odorous environment where they intermingle with
the tips of supporting cell microvilli (see Chapter 4). Bar
1 m.
D. Microvillous Cells

Besides the major populations of olfactory epithelial Jourdan, 1975; Menco, 1977). A second type of infrequent
cells—the ciliated receptor cells, microvillous supporting, microvillous cell has its microvilli aligned in parallel, and
and basal cells—there are at least five other much less these microvilli have a more uniform diameter and length
abundant cell types that line the nasal cavity with micro- than those of supporting cells. Depending on the fixation
villi. The term microvillous is used here generically for all method used, the cytoplasm of this cell is either more elec-
cell types that have microvilli to prevent confusion with tron-opaque (Agasandyan, 1990; Carr et al., 1991; Erhardt
the term “microvillar,” which has been used to describe and Meinel, 1979; Johnson et al., 1993; Jourdan, 1975;
specific microvillous cell types in the nose (Moran et al., Pyatkina and Agasandyan, 1991; Rowley et al., 1989) or
1982a,b; Rowley et al., 1989) (see below under cell Types more electron-lucent than that of surrounding supporting
2 and 4). First, there are brush cells, which occur in cells (Menco, 1992c, 1994; Pixley et al., 1997). A third
olfactory and respiratory epithelia and which have type of infrequent microvillous cell is more electron-lucent
microvilli with a more rigid appearance than those of sup- in conventionally fixed tissues than surrounding support-
porting cells. Collectively, the microvilli of these cells ing cells, while its microvilli are more compacted than
resemble a brush (Andres, 1969; Jeffery and Reid, 1975; those of cell Type 2 above (Miller et al., 1995). A fourth
34 Menco and Morrison

microvillous cell, also electron-lucent in conventionally (Yamagishi et al., 1993). Membranes of its microvilli have
fixed tissues and found in humans (Moran et al., 1982a, b), a specific lectin-labeling pattern distinct from that of cilia
is flask-shaped and has short microvilli (Fig. 28) and a of surrounding receptor cells and microvilli of supporting
subnuclear pole-like process (Morrison and Costanzo, cells (Fig. 29) (Menco, 1992c). Membranes of microvilli
1990, 1992). These cells are present at a level of approxi- of Type 2 cells immunolabel in their apical membranes
mately 10% of the neuronal population. A fifth cell resem- with an antibody named 1A6 (Carr et al., 1991) and
bles in its apex hair cells of the ear and has, so far, only display ecto-5-nucleotidase activity (Braun and
been shown to be present during development. This cell is Zimmerman, 1998). Also, despite the different appear-
very sparse and is zonally distributed (Menco and Jackson, ances, Type 2 and Type 4 cells may be a same polymor-
1997b). There is no evidence that any of these cells resem- phous cell. The shape of the apex of this cell is conceivably
ble microvillous receptor cells of the vomeronasal organ. affected by fixation; in fixed tissues we saw more Type 2
For example, the latter have ample basal bodies in their cells and in unfixed tissues we saw more Type 4 cells. The
apices (e.g., Vaccarezza et al., 1981), unlike the microvil- functional implications of these labeling patterns are still
lous cells that may have, at most, two (Bannister and unclear, and the exact role of any of the microvillous cells
Dodson, 1992; Menco and Jackson, 1997b). Also, none of is unknown. The brush cell, Type 1, may help to regulate
them resemble fish olfactory epithelial microvillous recep- concentrations of electrolytes, probably NaHCO3 (Ogata,
tor cells (Moran et al., 1992d; Rhein et al., 1981; Zielinski 2001). Speculatively, because of their resemblance to inner
and Hara, 1992). The latter are more similar to ear hair cells, Type 5 cells with stiff microvilli could be
vomeronasal receptor cells of higher vertebrates mechanoreceptors (Menco and Jackson, 1997b). Type 2
(Anderson et al., 1999; Eisthen, 1992). and/or 4 were thought to be bipolar neurons (Rowley et al.,
Microvillous cell Type 4 above is immunopositive for 1989), but the evidence for this is controversial (Carr et al.,
the calcium-binding proteins Spot-35 and calbindin 1991).

Figure 29 Microvilli (small arrows) of microvillous cell Type 4

bind the lectin peanut agglutinin (specific for -galactose
Figure 28 Transmission electron micrograph of a flask-shaped residues; lectin conjugated to colloidal gold, 5 nm) in a neu-
microvillous cell (M; human), probably Type 4, surrounded by apices of raminidase-treated section (removes sialic acid). Nearby support-
supporting cells (S) and by olfactory epithelial sensory receptor ing cell microvilli (large asterisk) and receptor cell cilia (small
cell dendrites (D). The tissue was conventionally fixed and asterisk) are devoid of label (Menco, 1992c). Unfixed tissue was
embedded. Bar
1 m. rapidly frozen and freeze-substituted. Bar
1 m.
Morphology of the Mammalian Olfactory Epithelium 35

E. Lamina Propria, Bowman’s Glands, and Mucus receptor cells by week 11. During development and
throughout the receptor cell’s life, centrioles migrate
The olfactory mucosa resides on a lamina propria that through the dendrites to the receptor cell dendritic knobs.
contains axon fascicles, blood vessels, connective tissue, This process is thought to be important in cilium forma-
and Bowman’s glands. Axons of olfactory receptor cells tion, cilium replacement, and possibly cell renewal (Heist
fasciculate, form small intraepithelial bundles, and enter and Mulvaney, 1968; Mulvaney and Heist, 1971). In this
the lamina propria, where they form larger bundles context it is noteworthy that vomeronasal receptor cell
(20–100 m) that project centrally to the olfactory bulb. dendrites and dendritic endings are stacked with centri-
Bowman’s glands are present in the olfactory region of all oles that do not give rise to cilia, but to microvilli instead
vertebrates except for fish. Human Bowman’s glands are (Vaccarezza et al., 1981). That abundance of centrioles is
spherical (20–40 m diameter) and are composed of still enigmatic.
serous and stem cells (Breipohl, 1972; Getchell and In general, olfactory receptor cell dendritic endings
Getchell, 1992; Huard et al., 1998; Seifert, 1971, 1972). sprout primary cilia before the full complement of olfac-
The serous cells are pyramidal, with a spherical nucleus tory cilia arises (Fig. 3) (Menco, 1988a; Menco and
and short stubby microvilli, and surround a central lumen. Farbman, 1985a,b).* There is a distinct topographical
Myoepithelial cells surround the acini and contain actin transition period, probably within hours, when the olfac-
filaments. They squeeze secretory cells and aid in moving tory epithelial surface becomes characteristically olfac-
secretory products toward a simple duct, which extends tory. This includes the appearance of tight junctions
through the epithelium to deliver the products to the (Menco, 1988c). In rats this occurs by day 14 following
mucous surface. Thus, Bowman’s gland cells, together conception (total time of pregnancy 22 days). As part of
with supporting cells (especially in lower vertebrates), olfactory ciliogenesis, densities of membrane particles
produce the microenvironment in which sensory transduc- (Menco, 1988a) and strands of ciliary necklaces (Menco,
tion occurs (Getchell and Getchell, 1990, 1992; Getchell 1988b) increase, and most of the signaling proteins begin
and Mellert, 1991; Pelosi, 2001; Seifert, 1971; 1972). to become apparent (Matsuzaki et al., 1999; Menco et al.,
There are at least two types of Bowman’s gland serous 1994; reviewed in Menco, 1997, especially Fig. 15, and
cells, one with electron-lucent droplets and one with Tarozzo et al., 1995, Table 2). Immunohistochemical
opaque droplets, suggesting that these glands secrete mul- (Menco et al., 1994) and knockout studies (Belluscio et
tiple mucous products (Seifert, 1971). Such heterogeneity al., 1998) suggest that a Gs signaling cascade precedes
is reflected in the mucus, which can consist of several dis- the one involving Golf. Initial receptor cell formation
tinct domains (Foster et al., 1992; Menco and Farbman, does not need odorant receptors, but such receptors are
1992). The exact implications of this heterogeneity, needed for their proper projections to secondary cells
although still unclear, may relate to odorant binding, (Lin and Ngai, 1999). Neurotrophic factors influence the
clearance, and/or maintenance of a mucus consistency, all formation of the neuroepithelium (Mackay-Sim and
allowing the process of olfaction to properly take place Chuah, 2000). Special pioneer cells may precede this
(Getchell and Getchell, 1990; Pelosi, 1994). process (Whitlock and Westerfield, 1998). All of this paral-
lels onset of physiological responsivity (Gesteland et al.,
1982).†
IV. TRANSIENT ASPECTS OF
THE OLFACTORY MUCOSA *A case can be made for parallel evolution in some
form. Whereas in many invertebrates single modified primary
A. Structural Aspects of Embryonic Development cilia seem to form the odorant-receptive sensory cellular
and of Neuronal Plasticity structures, in most vertebrates these structures involve modi-
fied secondary cilia. In again other instances such structures
Since other chapters in this book deal with the development involve modified microvilli, such as in vomeronasal chemo-
reception (Eisthen, 1992; Menco, 1992b; Steinbrecht, 1999;
of the olfactory mucosa and its special plasticity (see
Vaccarezza et al., 1981).
Chapters 5, 6, and 29), only a few aspects of these process- † Interestingly, in the invertebrate nematode C. elegans, there is
es, notably those touching on fine structure, are discussed distinct evidence that special transcription factors are involved in
here. sensory neuron cilium formation (Swoboda er al., 2000). So far
The developmental aspects of the olfactory epithelium none of these factors has been directly implicated in vertebrate
have been staged in mice (Cuschieri and Bannister, olfactory cilium formation, but receptor cells that do not yet
1975a,b), as well as in humans (Bossy, 1980; Pyatkina, have cilia lack the immunocytochemical expression of OMP
1982). The latter authors noted differentiated olfactory (Menco, 1989).
36 Menco and Morrison

The olfactory epithelium has a zonal topography during Another developmental process important for correct
fetal development (Menco and Jackson, 1997a) that roughly tissue formation is that of programmed cell death or apop-
parallels zones in which odorant receptors (Ressler tosis. In the olfactory mucosa apoptosis involves
et al., 1993; Strotmann et al., 1994; Vassar et al., 1993; programmed developmental elimination of neurons and of
review Mori et al., 2000) and perhaps also RGS proteins mesenchymal cells (Pellier et al., 1996).
(Norlin and Berghard, 2001), as well as zones of odor-
induced functional responsivity (Scott and Brierley, 1999) B. Aging
are expressed. Supporting cells, too, show the zonal pat-
terning (Menco and Jackson, 1997a; Miyawaki et al., It is difficult to determine the life span of human olfac-
1996). Also, like receptor cells, they undergo developmen- tory neurons. However, in nonhuman vertebrates, olfac-
tal transformations; their microvilli become longer and the
tory receptor neurons appear to have a life span of at
number of supranuclear organelles increases (Cuschieri
least one year, depending on factors such as the environ-
and Bannister, 1975b; Menco and Farbman, 1985a,b;
ment, health of the animal, and ability to form synapses
Mendoza and Kühnel 1991).
(Hinds et al., 1984; Mackay-Sim and Kittel, 1990;
Olfactory neurons have a number of unique characteris-
Weiler and Farbman, 1997). In human adult epithelium
tics that set them apart from most other neurons within the
it is not uncommon to find olfactory receptor neurons
nervous system. First, their peripheral location, which
in all regions, even close to the epithelial surface. Based
exposes them to the external environment, renders them
on their morphology receptor neurons observed near
especially vulnerable; most other sensory receptor neurons
the epithelial surface may be “long-lived” or “old.”
are located internally, protecting them from the environ-
Typically, they have short, thick (2–3 m) dendrites and
ment. Second, olfactory neurons are among the few
often a more irregular “bumpy” cell surface than neurons
neurons that can replace themselves (postnatal neurogenesis),
found in lower parts of the epithelium (Morrison and
normally and when injured. This remarkable capacity for
regeneration, which involves trophic (Mackay-Sim and Costanzo, 1990). However, Strotmann et al. (1996) offer
Chuah, 2000; Newman et al., 2000; Plendl et al., 1999) as a distinct alternative. They showed that, based on odor-
well as adhesive factors (Plendl and Sinowatz, 1998) (see ant-receptor expression, a definite vertical zonal organ-
Chapter 5), allows them to keep functioning in an often ization of olfactory neurons exists besides the horizontal
hostile environment. In vitro studies suggest that the zonal organization (Ressler et al., 1993; Strotmann et al.,
human olfactory epithelium also retains the capacity for 1994; Vassar et al., 1993). Each laminar zone contains
neurogenesis (Murrell et al., 1996). In primates, olfactory receptor cells that express a distinct group of odorant
axotomy results in immediate retrograde degeneration of receptors. Conceivably, then, the distribution of cells
the olfactory receptor neurons. Third, the receptor cells lit- could reflect intrinsic differences in cell populations,
erally form a conduit from the environment to the central age-related processes, or both.
nervous system, providing a pathway for movement of The number of receptor cells decreases with age
exogenous agents into the brain (see Chapter 26). (Breckenridge et al., 1997; Naessen, 1971a; Rosli et al.,
At a structural level, experimental animal studies 1999). In epithelial surfaces, the numbers of cilia and sup-
have shown that injury to olfactory axons (axotomy) porting cell microvilli are reduced (Hirai et al., 1996). Areas
results in profound changes in the neuroepithelium. containing sparse numbers of receptor cells may have been
Olfactory receptor neurons degenerate after their axons subject to local insults, e.g., from airborne toxic agents, bac-
are severed, leaving an epithelium mostly populated by teria, or viruses, and/or may reflect gradual epithelial
supporting cells and basal cells, besides degenerating changes inherent in the aging process (Lenz, 1977; Morrison
neurons. Following degeneration, GBCs become mito- and Costanzo, 1990; Naessen, 1971b; Seifert, 1969) (see
tically active. These then give rise to new neurons that next section). Age-related accumulation of several types of
mature, grow axons to the olfactory bulb, where they electron-dense granules in supporting cell apices (Naessen
reestablish anatomical and functional connections, even- 1971a) and basal feet (Naguro and Iwashita, 1992) may
tually reconstituting the neuroepithelium (Costanzo, reflect processes that ultimately compromise function. The
1991; Farbman, 1992; Graziadei and Monti Graziadei, 1979; basis for these accumulations, including their chemical
Monti Graziadei et al., 1980; Morrison and Costanzo, nature, is not clear, although the basal feet granules are
1989, 1995). Olfactory receptor cells display a remark- thought to contain lipofuscin. The various transformations
able degree of target independence with regard to this and the decrease in amount of sensory epithelium, overall or
regeneration, and this independence involves odorant in cell numbers and cilia, likely contribute to the decreased
receptors (Conzelmann et al., 1998; Lin et al., 2000; olfactory ability experienced by many elderly (Spielman,
Wang et al., 1998). 1998) (see Chapters 23 and 24).
Morphology of the Mammalian Olfactory Epithelium 37

C. Ultrastructural Correlates of Olfactory cal evaluation of human olfactory tissues (Getchell et al.,
Pathologies and Biopsies 1991). An irregular and patchy distribution of olfactory
epithelium mixed with respiratory epithelium (Figs.
A thorough understanding of the distribution of olfac- 30 – 33) (Morrison and Costanzo, 1990; Naessen, 1970,
tory epithelia is relevant to studies of olfactory biopsy 1971a; Nakashima et al., 1991; Rossli et al., 1999;
material in clinical cases. Obtaining a reliable tissue Schultze, 1862; Talamo et al., 1994; von Brunn, 1892;
sample from the nearly inaccessible olfactory region Yamada et al., 1980) must be taken into account when
has presented a challenge to the clinician. The proce- sampling for biopsies or for studies on the physiological
dure is also dangerous, since part of the olfactory responsivity of the human olfactory epithelium
mucosa is located on a thin, bony shelf (cribriform (Leopold et al., 2000; Rawson, 2000). Several attempts
plate of the ethmoid bone) that separates the nasal and may be needed to obtain samples that contain olfactory
anterior cranial cavities. In 1982 Lovell and colleagues neuroepithelium.
developed an instrument and technique to obtain small Olfactory dysfunction may have a genetic component
biopsies. Together with direct endoscopic observation (Belluscio et al., 1998; Brunet et al., 1996; Wong et al.,
(Lanza et al., 1993), this and similar instruments have 2000). Here we restrict ourselves to some structural
been used for safe removal of human olfactory epi- aspects of olfactory dysfunction. However, emerging
thelium (e.g., Lehman et al., 2000; Leopold et al., 2000; methods of functional imaging that take the whole olfacto-
Paik et al., 1992; Yamagishi et al., 1988) (see Chapter ry system into account (see Chapters 12 and 28) may be
1). This has led to an increased use of biopsy samples very helpful in the diagnosis of olfactory deficits (e.g.,
in ultrastructural, immunocytochemical, and pathologi- Yousem et al., 1996). (Ultra)structural abnormalities of the

Figure 31 A transition region between the two epithelia

Figure 30 Scanning electron micrograph of transition region (human) at higher magnification. The bottom half displays ol-
between sensory (darker areas, labeled O) and respiratory epithe- factory epithelium, the top half respiratory epithelium. Arrows
lial regions (lighter areas, R) in human nasal septum. The border identify olfactory receptor cell dendritic endings with cilia.
between the two epithelia is irregular. Bar
1 mm. Bar
5 m.
38 Menco and Morrison

1997; Schwob et al., 1995; Sunderman, 2001) (see

Chapters 3, 25–28). Dysfunction can be caused by dam-
aged receptor, supporting, basal, and Bowman’s gland
cells (Mancuso et al., 1997; Nakashima et al., 1991;
Schwob et al., 1995) or combinations of the above, but
special protective mechanisms (Carr et al., 2001) may
make supporting and gland cells more resistant to damage
than receptor cells (Nakashima et al., 1991).
Significant olfactory deficits often occur as a result of
head injury (Chapter 30). This may lead to transection of
olfactory receptor axons, either by fracture of the cribri-
form plate or through rapid displacement of the brain. The
limited data available suggest that a number of so-called
“traumatic anosmics” show marked changes in the
ultrastructure of their olfactory epithelia (Hasegawa et al.,
1986; Jafek et al., 1989). The olfactory epithelium looks
disorganized, lacking the normal “layered” appearance
seen in normal individuals. Many pyknotic and metaboli-
cally active neurons cause the epithelium to resemble a
regenerative one. Numerous olfactory axon fascicles are
displaced within the epithelium and lamina propria,
indicating axon proliferation. The number of olfactory
receptor cells is greatly reduced. Few dendrites reach the
surface; those that do usually are devoid of cilia. New
receptor cells that develop following trauma-induced
axotomy may try to send their axons centrally, but most
are unable to penetrate the fibrotic healing of the cribri-
form plate. Though, in rare instances, slight recovery of
olfactory function occurs, suggesting some potential for
regeneration and reconnection (Doucette et al., 1983).
Figures 32 and 33 Schematic sagittal sections through In Kallmann’s syndrome the symptoms are somewhat
the human nose corresponding to Figure 30. The dashed lines similar, but possibly even more extreme. As the olfactory
outline the olfactory epithelial areas (O; dark), patches of respi- bulb is hypoplastic or aplastic (Truwitt et al., 1993),
ratory epithelium within them (R; light). OB: olfactory bulb. The developing olfactory neurons cannot reach their targets.
somewhat erratic distributions of both epithelia has to be taken Consequently the olfactory epithelium is severely degener-
into account for biopsies (see also Fig. 4 in Naessen, 1970).
ated with vastly reduced numbers of receptor and
supporting cells. The few axons present also reflect this
olfactory mucosa accompany pathological states that lead degeneration (Schwob et al., 1993).
to total or partial loss of olfactory function (Rawson, 2000; Olfactory function is also commonly lost following
Spielman, 1998). These states include traumatic anosmia intranasal viral infections; the condition is known as
(Hasegawa et al., 1986; Jafek et al., 1989; Moran et al., postviral olfactory dysfunction (see Chapter 26). Patients
1985, 1992a, c; Yamagishi et al., 1988) (see Chapters 29 who have experienced this problem often also show ultra-
and 30), postviral olfactory dysfunction (Douek et al., structural changes in their olfactory epithelia that are
1975; Jafek et al., 1990; Moran et al., 1992a; Yamagishi et quite similar to those in traumatic anosmics, that is, the
al., 1988), Alzheimer’s and Parkinson’s diseases number of ciliated olfactory receptor cells is reduced, and
(Brouillard et al., 1994; Moran et al., 1992b) (see those that are present have few olfactory cilia. Postviral
Chapters 23–26), Kallmann’s syndrome (Schwob et al., hyposmics, who have partial loss of their sense of smell,
1993; Truwitt et al., 1993), olfactory epithelial tumors have more ciliated olfactory receptor cells than do post-
(Reznik-Schüller, 1983; Takahashi et al., 1986; Taxy et al., viral anosmics, emphasizing that there is a correlation
1986), rhinosinusitis (Kern, 2000), and exposure to xeno- between function and the number of olfactory neurons.
biotic toxic compounds (Hurtt et al., 1988; Mancuso et al., (Jafek et al., 1990).
Morphology of the Mammalian Olfactory Epithelium 39

Various degrees of loss of olfactory function as a con-

sequence Alzheimer’s and Parkinson’s disease (Chapters
23 and 24) are accompanied by olfactory epithelial ultra-
structural alterations (Brouillard et al., 1994; Moran et
al., 1992b). Compared to other brain areas, these degen-
erative changes preferentially extend to the olfactory
cortex (Reyes et al., 1993). Olfactory epithelia of
patients having Alzheimer’s and Parkinson’s disease
exhibit a greatly reduced number of ciliated receptor
cells. When present, these cells have reduced numbers of
cilia (Figs. 34 – 37). In Alzheimer’s disease some of the
bipolar neurons have thickened dendrites; many dying
cells are evident. Near the basement membrane, the
olfactory epithelium of Alzheimer’s patients contains
increased numbers of axons, many of them swollen (Fig.
36). Supporting cells also show signs of degeneration
(Brouillard et al., 1994). In patients with Parkinson’s
disease, the “layering” of nuclei normally seen in
healthy olfactory epithelia is disrupted in places.
Receptor cell supranuclear regions are often swollen.
Numbers of axon profiles near the basement membrane

Figure 35 Transmission electron micrograph of the same biop-

sy material as used for Figure 34. Olfactory dendritic knobs lack
cilia (arrows) and axon bundles have invaded the base of the
epithelium (arrowheads). O, olfactory receptor neurons. Bar
5
m.

are greatly increased, and these axons are often enlarged

and of variable diameter. Large axon bundles “invade”
the epithelium (Figs. 34, 35).

V. SUMMARY AND CONCLUSIONS

In this chapter we have reviewed the overall and func-

tional morphology of the mammalian olfactory system,
including that of the human, as well as structural aspects
of normal and regenerative development, aging, and
some important pathologies. The evidence is compelling
that the olfactory receptor cell cilia (and also VNO
receptor cell microvilli) possess all the properties neces-
sary to transform odorant-receptor interactions into an
electrical signal. Thus, these cilia and microvilli are
highly specialized organelles that resemble in many
Figure 34 Light micrograph (1 m thick section) of an olfac- respects the modified cilia that form vertebrate retinal
tory epithelial biopsy of a patient with Parkinson’s disease. The photoreceptor cell outer segments (Müller and Kaupp,
epithelium has a disorganized appearance (arrows). Olfactory 1998). The olfactory epithelial supporting cells appear to
axons (Ax) and Bowman’s glands are present in the underlying play a large number of roles. These include insulation of
lamina propria. Bar
20 m. receptor cells, transport and regulation of ions and other
40 Menco and Morrison

Figure 36 Transmission electron micrograph of olfactory tissue Figure 37 As in Figure 36, but showing an absence of normal
from an Alzheimer’s patient. The epithelium is disorganized with ciliated olfactory receptor cells: swollen dendrites (arrows),
several degenerating neurons (arrows). Olfactory knobs generally Supporting cells (S). Bar
5 m.
lack cilia and there is an increased number of axon fibers invading
epithelium near the basal lamina (arrowheads). Bar
5 m.

Gene Minner (BPhMM), Debbie Allgood and Karen Wolfe

substances in surrounding receptor cells and extracel- (EEM) for their help, and Dr. D.T. Moran, whose collabo-
lular fluid, metabolism of xenobiotic compounds, ration on the earlier version of this chapter made our work
protection against aging, phagocytosis, response to here so much easier. The work was supported by NIH-
hormonal variations, structural support, maintaining of a NIDCD (DC02491, BPhMM and DC01532, EEM), NSF
transmembrane permeability boundary, and guiding of (IBN-0094709, BPhMM), ONDCP, and FAA (EEM).
developing receptor neurons. There are two types of
basal cells, horizontal and globose, the latter being the
REFERENCES
precursors for the receptor cells. Bowman’s glands make
a heterogeneous mucus and may contain the stem cells
Ache, B. W., and Restrepo, D. (2000). Olfactory transduction. In
for the supporting cells. Though some insights as to the The Neurobiology of Taste and Smell, 2nd ed. T. E. Finger,
nature of changes in the olfactory epithelia associated W. L. Silver, and D. Restrepo (Eds.). Wiley-Liss, Inc., New
with aging and a number of diseases have been observed, York, pp. 167–177.
the data are still sparse. New biopsy techniques offer the Adams, D. R. (1992). Fine structure of the vomeronasal and sep-
opportunity for a more systematic study of the tal olfactory epithelia and of glandular glands. Microsc. Res.
histopathology of the olfactory epithelium in a variety of Techn. 23:86–97.
disease states. Agasandyan, Kh. V. (1990). Microvillar cells in swine olfactory
epithelium. J. Evol. Biochem. Physiol. 26:194–198.
Anderson, K. T., Hansen, A., and Finger, T. E. (1999).
ACKNOWLEDGMENTS Localization of olfactory-type (Ors) and vomeronasal type
(V2Rs) receptors in different olfactory receptor neurons of
The authors thank the editorial staff of Marcel Dekker Inc., goldfish. Chem. Senses 24:593, abstr. 265.
the editor of this book, Dr. Richard Doty, Dr. Virginia Carr Andres, K.-H. (1969). Der olfaktorische Saum der Katze. Z.
(the HSP70 and ubiquitin project), Maya Yankova and Zellforsch. mikrosk. Anat. 96:250–274.
Morphology of the Mammalian Olfactory Epithelium 41

Asanuma, N., and Nomura, H. (1991). Cytochemical localization Brunet, L. J., Gold, G. H., and Ngai, J. (1996). General anosmia
of adenylate cyclase activity in rat olfactory receptor cell. caused by a targeted disruption of the mouse cyclic nucleotide
Histochem. J. 23:83–90. gated cation channel. Neuron 17:681–693.
Asanuma, N., and Nomura, H. (1993). Cytochemical localization Buck, L. B. (1996). Information coding in the vertebrate olfactory
of cyclic 3, 5-nucleotide phosphodiesterase activity in the rat system. Ann. Rev. Neurosci. 19:517–544.
olfactory receptor mucosa. Histochem. J. 25:348–356. Buck, L. B. (2000). The molecular architecture of odor and
Bakalyar, H. A., and Reed, R. R. (1990). Identification of a special- pheromone sensing in mammals. Cell 100:611–618.
ized adenylyl cyclase that may mediate odorant detection. Buck, L. B., and Axel, R. (1991). A novel multigene family may
Science 250:1403–1406. encode odorant receptors: a molecular basis for odor recogni-
Bannister, L. H., and Dodson, H. C. (1992). Endocytotic path- tion. Cell 65:157–167.
ways in the olfactory and vomeronasal epithelia of the mouse: Buiakova, O. I., Baker, H., Scott, J. W., Farbman, A., Kream, R.,
ultrastructure and uptake of tracers. Microsc. Res. Techn. 23: Grillo, M., Franzen, L., Richman, M., Davis, L. M.,
128–141. Abbondanzo, S., Stewart, C. L., and Margolis, F. L. (1996).
Barber, P. C., and Lindsay, R. M. (1982). Schwann cells of the Olfactory marker protein (OMP) gene deletion causes altered
olfactory nerves contain glial fibrillary acidic protein and physiological activity of olfactory sensory neurons. Proc.
resemble astrocytes. Neuroscience 7:3077–3090. Natl. Acad. Sci. USA 93:9858–9863.
Bastianelli, E., Polans, A. S., Hidaka, H., and Pochet, R. (1995). Burton, P. R. (1992). Ultrastructural studies of microtubules and
Differential distribution of six calcium-binding proteins in the microtubule organizing centers of the vertebrate olfactory
rat olfactory epithelium during postnatal development and neuron. Microsc. Res. Techn. 23:142–156.
adulthood. J. Comp. Neurol. 354:395–409. Cadiou, H., Sienaert, I., Vanlingen, S., Parys, J. B., Molle, G., and
Belluscio, L., Gold, G. H., Nemes, A., and Axel, R. (1998). Mice Duclohier, H. (2000). Basic properties of an inositol 1,4,5-
deficient in Golf are anosmic. Neuron 20:69–81. trisphosphate-gated channel in carp olfactory cilia. Eur. J.
Bloodgood, R. A. (Ed.) (1990). Ciliary and Flagellar Membranes. Neurosci. 12:2805–2811.
Plenum Press, New York. Caggiano, M., Kauer, J. S., and Hunter, D. G. (1994). Globose
Boekhoff, I., and Breer, H. (1992). Termination of second mes- basal cells are neuronal progenitors in the olfactory epithelium:
senger signaling in olfaction. Proc. Natl. Acad. Sci. USA a lineage analysis using a replication-incompetent retrovirus.
89:471–474. Neuron 13:339–352.
Bonifacino, J. S., and Weismann, A. M. (1998). Ubiquitin and the Carr, V. McM., Farbman, A. I., Coletti, L. M., and Morgan, J. I.
control of protein fate in the secretory and endocytotic path- (1991). Identification of a new nonneuronal cell type in rat
ways. Annu. Rev. Cell Dev. Biol. 14:19–57. olfactory epithelium. Neuroscience 45:433–449.
Borisy, F. F., Ronnett, G. V., Cunningham, A. M., Juilfs, D., Carr, V. McM., Murphy, S. P., Morimoto, R. I., and Farbman, A. I.
Beavo, J., and Snyder, S. H. (1992). Calcium/calmodulin- (1994). Small subclass of rat olfactory neurons with specific
activated phosphodiesterase expressed in olfactory receptor bulbar projections is reactive with monoclonal antibodies to
neurons. J. Neurosci. 12:915–923. the HSP70 heat shock protein. J. Comp. Neurol. 348:150–160.
Bossy, J. (1980). Development of olfactory and related structures Carr, V. McM., Walters, E., Margolis, F. L., and Farbman, A. I.
in staged human embryos. Anat. Embryol. 161:225–236. (1998). An enhanced olfactory marker protein immunore-
Braun, N., and Zimmerman, H. (1998). Association of ecto-5- activity in individual olfactory receptor neurons following
nucleotidase with specific cells types in the adult and devel- olfactory bulbectomy may be related to increased neurogene-
oping rat olfactory organ. J. Comp. Neurol. 393:528–537. sis. J. Neurobiol. 34:377–390.
Breckenridge, L. J., Cameron, J., Reid, O., McGadey, J., and Carr, V. McM., Menco, B. Ph. M., Yankova, M. P., Morimoto, R.
Smith, R. A. (1997). Localised degeneration occurs in aged I., and Farbman, A. I. (2001). Odorants as cell-type specific
mouse olfactory epithelium. J. Anat. 191:151–154. activators of a heat shock response in the rat olfactory
Breer, H. (ed.) (1994). The Molecular and Cell Biology of Smell. mucosa. J. Comp. Neurol. 432:425–439.
Semin. Cell Biol. 5(1). Chen, T.-Y., and Yau, K.-W. (1994). Direct modulation by Ca2-
Breipohl, W. (1972). Licht- und electronenmikoskopische Befunde calmodulin of cyclic nucleotide-activated channel of rat olfac-
zur Struktur de Bowmanschen Drüsen im Riechepithel der tory receptor neurons. Nature 368:545–548.
weißen Maus. Z. Zellforsch. mikrosk. Anat. 131:329–346. Cherry, J. A., and Davis, R. L. (1995). A mouse homolog of
Breipohl, W., Laugwitz, H. J., and Bornfeld, N. (1974). dunce, a gene important for learning and memory in
Topological relations between the dendrites of olfactory sens- Drosophila, is preferentially expressed in olfactory receptor
ory cells and sustentacular cells in different vertebrates. An neurons. J. Neurobiol. 28:102–113.
ultrastructural study. J. Anat. 117:89–94. Chuah, M. I., and Zheng, D. R. (1992). The human primary
Brouillard, M., Laccourreye, L., Jabbour, W., Emile, P., and olfactory pathway: fine structural and cytochemical
Pouplard-Barthelaix, A. (1994). Étude ultrastructurale et aspects during development in adults. Microsc. Res. Techn.
immunohistochimique de la muqueuse olfactive dans la mal- 23:76–85.
adie d’Alzheimer. Bull. l’Assoc. Anat. 78:25–28. Coates, E. L. (2001). Olfactory CO2 chemoreceptors. Resp.
Brown, D., Garcia-Segura, L.-M., and Orci, L. (1984). Carbonic Physiol. 129:219–229.
anhydrase is present in olfactory receptor cells. Conzelmann, S., Saucier, D., Strotmann, J., Breer, H., and Astic,
Histochemistry 80:307–309. L. (1998). Decline and recovery of olfactory receptor expres-
42 Menco and Morrison

sion following unilateral bulbectomy. Cell Tissue Res. Engström, H. and Bloom, G. (1953). The structure of the olfac-
294:421–430. tory region in man. Acta Otolaryngol. 43:11–21.
Costanzo, R. M. (1991). Regeneration of olfactory receptor Erhardt, H., and Meinel, W. (1979). Electron microscope observa-
cells. In Regeneration of Vertebrate Sensory Receptor Cells, tions on the olfactory epithelium of the hedgehog, Erinaceus
Ciba Foundation Symposium 160, G. R. Bock and J. Whelan europaeus Linnaeus, 1758 (Insectivora, Erinaceidae). Zool. Jb.
(Eds.). John Wiley & Sons, Chichester, UK, pp. 233–248. Anat. 101:113–121.
Costanzo, R. M., and Morrison, E. E. (1989). Three-dimensional Evans, C. (2002). Vomeronasal Chemoreception in Vertebrates. A
scanning electron microscopic study of the normal hamster Study of the Second Nose. Imperial College Press, London, UK.
olfactory epithelium. J. Neurocytol. 18:381–391. Farbman, A. I. (1992). Cell Biology of Olfaction. Cambridge
Cunningham, A. M., Ryugo, D. K., Sharp, A. H., Reed, R. R., University Press, Cambridge, UK.
Snyder, S. H., and Ronnett, G. V. (1994). Neuronal inositol Farbman, A. I. (2000). Cell biology of olfactory epithelium. In The
1,4,5-trisphosphate receptor localized to the plasma mem- Neurobiology of Taste and Smell, 2nd ed., T. E. Finger, W. L. Silver
brane of olfactory cilia. Neuroscience 57:339–352. and D. Restrepo (Eds.). Wiley-Liss, Inc., New York, pp. 131–158.
Cuschieri, A., and Bannister, L. H. (1975a). The development of Feng, W.-H., Kauer, J. S., Adelman, L., and Talamo, B. R. (1997).
the olfactory mucosa in the mouse: light microscopy. J. Anat. New structure, the “olfactory pit”, in human olfactory mucosa. J.
119:277–286. Comp. Neurol. 378:443–453.
Cuschieri, A., and Bannister, L. H. (1975b). The development of Ferrari, C. C., Carmanchahi, P. D., Marcos, H. J. A., Mugnaini,
the olfactory mucosa in the mouse: electron microscopy. M. T., Affani, J. M., and Paz, D. A. (1999). Identification and
J. Anat. 119:471–498. localisation of glycoconjugates in the olfactory mucosa of the
Da Pos, O., and Arimondi, C. (1983). Rat olfactory epithelial varia- armadillo Chaetophractus villosus. J. Anat. 194: 395–405.
tions related to age and sex. Bull. Zool. 50:87–92. Firestein, S., and Shepherd, G. M. (1991). A kinetic model of the
Dawson, T. M., Arriza, J. L., Jaworsky, D. E., Borisy, F. F., odor response in single olfactory receptor neurons. J. Steroid
Attramadal, H., Lefkowitz, R. J., and Ronnett, G. V. (1993). Biochem. Molec. Biol. 39:615–620.
 -Adrenergic receptor kinase-2 and -arrestin-2 as mediators Foster, J. D., Getchell, M. L., and Getchell, T. V. (1991).
of odorant-induced desensitization. Science 259:825–829. Identification of sugar residues in secretory glycoconjugates
DellaCorte, C., Restrepo, D., Menco, B. Ph. M., Andreini, I., of olfactory mucosae by lectin histochemistry. Anat. Rec.
and Kalinoski, D. L. (1996). G q /G 11 : immunolo- 229:525–544.
calization in the neuroepithelium of the rat (Rattus rattus) Foster, J. D., Getchell, M. L., and Getchell, T. V. (1992).
and the channel catfish (Ictalurus punctatus). Neuroscience Ultrastructural localization of sialylated glycoconjugates in
74:261–273. cells of the salamander olfactory mucosa using lectin cyto-
Dhallan, R. S., Yau, K.-W., Schrader, K. A., and Reed, R. R. chemistry. Cell Tissue Res. 267:113–124.
(1990). Primary structure and functional expression of a cyclic Franklin, R. J. M., and Barnett, S. C. (2000). Olfactory ensheath-
nucleotide-activated channel from olfactory neurons. Nature ing cells and CNS regeneration: the sweet smell of success?
347:184–187. Neuron 25:115–118.
Doucette, R. (1992). Glial cells in the nerve fiber layer of the Gesteland, R. C. (1986). Speculations on receptor cells as ana-
main olfactory bulb of embryonic and adult mammals. lyzers and filters. Experientia 42:287–291.
Microsc. Res. Techn. 24:113–130. Gesteland, R. C., Yancey, R. A., and Farbman, A. I. (1982).
Doucette, J. R., Kiernan, J. A., and Flumerfelt, B. A. (1983). Two Development of olfactory receptor neuron selectivity in the
different patterns of retrograde degeneration in the olfactory rat fetus. Neuroscience 7:3127–3136.
epithelium following transection of primary axons. J. Anat. Getchell, M. L., and Getchell, T. V. (1992). Fine structural
136:673–689. aspects of secretion and extrinsic innervation in the olfactory
Douek, E., Bannister, L. H., and Dodson, H. C. (1975). Recent mucosa. Microsc. Res. Techn. 23:111–127.
advances in the pathology of olfaction. Proc. Roy. Soc. Med. Getchell, M. L., and Mellert, T. K. (1991). Olfactory mucus secretion.
68:467–470. In Smell and Taste in Health and Disease, T. V. Getchell, L. M.
Ecker, A. (1855). Über das Epithelium der Riechschleimhaut und Bartoshuk, R. L. Doty and J. B. Snow, Jr. (Eds.). Raven Press, New
die wahrscheinliche Endigung des Geruchsnerven beim York, pp. 83–95.
Menschen und Säugetieren. Ber. Verh. Beförd. Naturwiss. Getchell, T. V., Bartoshuk, L. M., Doty, R. L., and Snow, Jr., J. B.
Freiburg 12:199–206. (Eds.). (1991). Smell and Taste in Health and Disease. Raven
Eckhard, C. (1855). Über die Endigungsweise des Geruchsnerven. Press, New York.
Beitr. Anat. Physiol. 1:77–84. Getchell, T. V., and Getchell, M. L. (1990). Regulatory factors in
Eisthen, H. (1992). Phylogeny of the vomeronasal system and of the vertebrate olfactory mucosa. Chem. Senses 15:223–231.
receptor cell types in the olfactory and vomeronasal epithe- Gibson, A. D., and Garbers, D. L. (2000). Guanylyl cyclases as
lial of vertebrates. Microsc. Res. Techn. 23:1–21. a family of putative odorant receptors. Ann. Rev. Neurosci. 23:
Eng, D. L., and Kocsis, J. D. (1987). Activity dependent changes 417–439.
in extracellular potassium and excitability in the turtle olfac- Giorgi, D., Friedman, C., Trask, B. J., and Rouquier, S. (2000).
tory nerve. J. Neurophysiol. 57:740–754. Characterization of non-functional V1R-like pheromone
Morphology of the Mammalian Olfactory Epithelium 43

receptor sequences in human. Genome Res. 10: Hurtt, M. E., Thomas, D. A., Working, P. K., Monticello, T.
1979–1985. M., and Morgan, K. T. (1988). Degeneration and regenera-
Glusman G., Yanai, I., Rubin, I., and Lancet, D. (2001). The com- tion of the olfactory epithelium following inhalation
plete human olfactory subgenome. Genome Res. 11:685–702. exposure to methyl bromide: pathology, cell kinetics, and
Gold, G. H., and Nakamura, T. (1987). Cyclic nucleotide-gated olfactory function. Toxicol. Appl. Pharmacol. 94: 311–328.
conductances: a new class of ion channels mediates visual and Imaizumi, T., Lankford, K. L., and Kocsis, J. D. (2000).
olfactory transduction. TIPS 8:312–316. Transplantation of olfactory ensheathing cells or Schwann cells
Goldstein, B. J., and Schwob, J. E. (1996). Analysis of the glob- restores rapid and secure conduction across the transected spinal
ose basal cell compartment in rat olfactory epithelium using cord. Brain Res. 854:70–78.
GBC-1, a new monoclonal antibody against globose basal Jafek, B. W., Eller, P. M., Esses, B. A., and Moran, D. T. (1989).
cells. J. Neurosci. 16:4005–4016. Posttraumatic anosmia. Arch. Neurol. 46:300–304.
Graziadei, P. P. C. (1971). The olfactory mucosa of vertebrates. Jafek, B. W., Hartman, D., Eller, P. M., Johnson, E. W., Strahan,
In Handbook of Sensory Physiol, IV, Chemical Senses, I. R. E., and Moran, D. T. (1990). Post-viral olfactory dysfunc-
Olfaction, L. M. Beidler (Ed.). Springer-Verlag, New York, tion. Am. J. Rhinol. 4:91–100.
pp. 27–58. Jeffery, P. K., and Reid, L. (1975). New observations of rat air-
Graziadei, P. P. C. (1973). The ultrastructure of vertebrates olfac- way epithelium: a quantitative and electron microscopic
tory mucosa. In The Ultrastructure of Sensory Organs, I. study. J. Anat. 120:295–320.
Friedman (Ed.). Elsevier, Oxford, pp. 267–305. Johnson, E. W., Eller, P. M., and Jafek, B. W. (1993). An immuno-
Graziadei, P. P. C., and Monti-Graziadei, G. A. (1979). electron microscopic comparison of olfactory marker protein
Neurogenesis and neuron regeneration in the olfactory system of localization in the supranuclear region in the rat olfactory epithe-
mammals. I. Morphological aspects of differentiation and struc- lium and vomeronasal organ neuroepithelium. Acta Otolaryngol.
tural organization of the olfactory sensory neurons. (Stockh.) 113:766–771.
J. Neurocytol. 8:1–18. Jones, D. T., and Reed, R. R. (1989). Golf: an olfactory neuron
Griffith, G. (1993). Fine Structure Immunocytochemistry. specific-G protein involved in odorant signal transduction.
Springer-Verlag, Berlin. Science 244:790–795.
Güntherschulze, J. (1979). Studien zur Kenntnis der Regio Jourdan, F. (1975). Ultrastructure de l épithelium olfactif du
olfactoria von Wild- und Hausschwein (Sus scrofa scrofa L. rat: polymorphisme des récepteurs. C. R. Hebd. Séance
1758 und Sus scrofa f. domestica). Zool. Anz. (Jena) 202: Acad. Sc. Paris 280:443–446.
256–279. Juilfs, D. M., Fülle, H.-J., Zhao, A. Z., Houslay, M. D., Garbers,
Harrington, C., Buckland, M., Levy, N., Reed, R., and D. L., and Beavo, J. A. (1997). A subset of olfactory neurons
Cunningham, A. (1997). Odorant receptor proteins: expres- that selectively express cGMP-stimulated phosphodiesterase
sion in olfactory axons and olfactory bulb glomeruli supports (PDE2) and guanylyl cyclase-D define a unique olfactory
a role in axonal guidance and/or target recognition. Chem. signal transduction pathway. Proc. Natl. Acad. Sci. USA
Senses 22:181–182, abstr. 2. 94:3388–3395.
Hasegawa, S., Yamagishi, M., and Nakano, Y. (1986). Kalinoski, D. L., DellaCorte, C., Menco, B. Ph. M., and
Microscopic studies of human olfactory epithelia following Restrepo, D. (1994). Localization of inositol 1,4,5-trisphos-
traumatic anosmia. Arch. Otorhinolaryngol. 243:112–116. phate receptors in the olfactory neuroepithelium of the rat and
Hatt, H., and Ache, B. W. (1994). Cyclic nucleotide- and inositol channel catfish. Chem. Senses 19:493, abstr. 137.
phosphate-gated ion channels in lobster olfactory receptor Kehrl, J. H. (1998). Heterotrimeric G protein signaling: roles in
neurons. Proc. Natl. Acad. Sci. USA 91:6264–6268. immune function and fine tuning by RGS proteins. Immunity
Heist, H. E., and Mulvaney, B. D. (1968). Centriole migration. 8:1–10.
J. Ultrastruct. Res. 24:68–101. Kerjaschki, D., and Hörandner, H. (1976). The development of
Hinds, J. W., Hinds, P. L., and McNelly, N. A. (1984). An auto- mouse olfactory vesicles and their contacts: a freeze-etching
radiographic study of the mouse olfactory epithelium: evidence study. J. Ultrastruct. Res. 54:420–444.
for long-lived receptors. Anat. Rec. 210:375–383. Kern, R. C. (2000). Chronic sinusitis and anosmia: pathologic
Hirai, T., Kojima, S., Shimada, A., Umemura, T., Sakai, M., and changes in the olfactory mucosa. Laryngoscope 110: 1071–1077.
Itakura, C. (1996). Age-related changes in the olfactory Kern, R. C., and Pitovski, D. Z. (1997). Localization of 11-
system of dogs. Neuropathol. Appl. Neurobiol. 22:531–539. hydroxysteroid dehydrogenase: specific protector of the min-
Holbrook, E. H., Mieleszko Szumowski, K. E., and Schwob, J. E. eralocorticoid receptor in mammalian olfactory mucosa. Acta
(1995). An immunochemical, ultrastructural, and developmen- Otolaryngol. (Stockh.) 117:738–743.
tal characterization of the horizontal basal cells of rat olfac- Kern, R. C., Kerr, T. P., and Getchell, T. V. (1991). Ultrastructural
tory epithelium. J. Comp. Neurol. 363:129–146. localization of Na/K-ATPase in rodent olfactory epithe-
Huard, J. M. T., Youngentob, S. L., Goldstein, B. J., Luskin, M. lium. Brain Res. 546:8–17.
B., and Schwob, J. E. (1998). Adult olfactory epithelium con- Kleene, S. J., Gesteland, R. C., and Bryant, S. H. (1994). An elec-
tains multipotent progenitors that give rise to neurons and trophysiological survey of frog olfactory cilia. J. Exp. Biol.
non-neural cells. J. Comp. Neurol. 400:469–486. 195:307–328.
44 Menco and Morrison

Kishimoto, J., Keverne, E. B., Emson, P. C. (1993). Calretinin, Margolis, F. L. (1988). Molecular cloning of olfactory-specific gene
calbindin-D28k and parvalbumin-like immunoreactivity in products. In Molecular Neurobiology of the Olfactory System.
mouse chemoreceptor neurons. Brain Res. 610:325–329. Molecular, Membranous, and Cytological Studies,
Krause, W. (1876). Allgemeine und mikroskopische Anatomie. F. L. Margolis and T. V. Getchell (Eds.). Plenum Press, New
In Handbuch der menschlichen Anatomie, Vol. 1, C. F. Th. York, pp. 237–265.
Krause (Ed). Hahnsche Hof-Buchhandlung, Hannover, Massa, N. (1536). Introductorius Anatomiae sive Dissectionis.
Germany. Cap. XXXIX. Venezia, pp. 88–90.
Krupinski, J., Coussen, F., Bakalyar, H. A., Tang, W.-J., Feinstein, Matsuoka, M., Yoshida-Matsuoka, J., Iwasaki, N., Norita, M.,
P. G., Orth, K., Slaughter, C., Reed, R. R., and Gilman, A. G. Costanzo, R. M., and Ichikawa, M. (2001). Immuno-
(1989). Adenylyl cyclase amino acid sequence: possible cytochemical study of Gi2 and Go on the epithelium surface
channel- or transporter-like structure. Science 244:1558–1564. of the rat vomeronasal organ. Chem Senses 26:161–166.
Kurahashi, T., and Menini, A. (1997). Mechanism of odorant Matsuzaki, O., Bakin, R. E., Menco, B. Ph. M., Cai, X., and
adaptation in the olfactory receptor cell. Nature 385: Ronnett, G. V. (1999a). Immunolocalization of the olfactory
725–729. cyclic nucleotide-gated channel subunit 1 (OCNC1) in normal
LaMantia, A.-S., Bhasin, N., Rhodes, K., and Heemskerk, and regenerating olfactory neuroepithelium. Neuroscience
J. (2000). Mesenchymal/epithelial induction mediates olfactory 94:131–140.
pathway formation. Neuron 28:411–425. Matsuzaki, T., Suzuki, T., Koyama, H., Tanaka, S., and Takata, K.
Lancet, D. (1986). Vertebrate olfactory reception. Annu. Rev. (1999b). Water channel protein AQP3 is present in epithelia
Neurosci. 9:329–355. exposed to the environment of possible water loss.
Lanza, D. C., Moran, D. T., Doty, R. L., Trojanowski, J. Q., Lee, J. H., J. Histochem. Cytochem. 47:1275–1286.
Rowley, J. C., Crawford, D., and Kennedy, D. W. (1993). McClintock, T. S. (2000). Molecular biology of olfaction. In The
Endoscopic human biopsy technique: a preliminary report. Neurobiology of Taste and Smell, 2nd ed., T. E. Finger,
Laryngoscope 103:815–819. W. L. Silver and D. Restrepo (Eds.). Wiley-Liss, Inc., New York,
Leopold, D. A., Hummel, T., Schwob, J. E., Hing, S. C., Knecht, pp. 179–199.
M., and Kobal, G. (2000). Anterior distribution of human Menco, B. Ph. M. (1977). A qualitative and quantitative investiga-
olfactory epithelium. Laryngoscope 110:417–421. tion of olfactory and nasal respiratory mucosal surfaces of cow
Lenz, H. (1977). Surface of the olfactory region on man observed and sheep based on various ultrastructural and biochemical
by scanning electron microscope. Acta Otolaryngol. (Stockh.) techniques. Commun. Agricult. Univ., Wageningen
84:145–154. 77–13:1–157.
Lidow, M. S., and Menco, B. Ph. M. (1984). Observations on Menco, B. Ph. M. (1980a). Qualitative and quantitative freeze-
axonemes and membranes of olfactory and respiratory cilia in fracture studies on olfactory and nasal respiratory epithelial
frogs and rats using tannic acid-supplemented fixation and surfaces of frog, ox, rat, and dog. II. Cell apices, cilia, and
photographic rotation. J. Ultrastruct. Res. 86:18–30. microvilli. Cell Tissue Res. 211:5–30.
Lin, D. M., and Ngai, J. (1999). Development of the vertebrate Menco, B. Ph. M. (1980b). Qualitative and quantitative freeze-
main olfactory system. Curr. Opin. Neurobiol. 9:24–78. fracture studies on olfactory and nasal respiratory epithelial
Lin, D. M., Wang, F., Lowe, G., Gold, G. H., Axel, R., Ngai, J., and surfaces of frog, ox, rat, and dog. III. Tight-junctions. Cell
Brunet, L. (2000). Formation of precise connections in the Tissue Res. 211:361–373.
olfactory bulb occurs in the absence of odorant-evoked neu- Menco, B. Ph. M. (1980c). Qualitative and quantitative freeze-
ronal activity. Neuron 26:69–80. fracture studies on olfactory and nasal respiratory epithelial
Lindemann, B. (2001). Predicted profiles of ion concentrations in surfaces of frog, ox, rat, and dog. IV. Ciliogenesis and ciliary
olfactory cilia in the steady state. Biophys. J. 80:1712–1721. necklaces (including high-voltage observations). Cell Tissue
Lovell, M. A., Jafek, B. W., Moran, D. T., and Rowley, J. C. III. Res. 212:1–16.
(1982). Biopsy of human olfactory mucosa: an instrument and Menco, B. Ph. M. (1983). The ultrastructure of olfactory and
a technique. Arch. Otolaryngol. 108:247–249. nasal respiratory epithelium surfaces. In Nasal Tumors in
Mackay-Sim, A., and Chuah, M. I. (2000). Neurotrophic factors Animals and Man, Vol. 1, Anatomy, Physiology and
in the primary olfactory pathway. Progr. Neurobiol. Epidemiology, G. Reznik and S. F. Stinson (Eds.). CRC Press,
62:527–559. Inc., Boca Raton, FL, pp. 45–102.
Mackay-Sim, A., and Kittel, P. W. (1990). On the life span of Menco, B. Ph. M. (1984). Ciliated and microvillous structures of
olfactory receptor neurons. Eur. J. Neurosci. 3:209–215. rat olfactory and nasal respiratory epithelia. A study using
Mancuso, M., Giovanetti, A., and Brittebo, E. B. (1997). Effects ultra-rapid cryo-fixation followed by freeze-substitution or
of dichlobenil on ultrastructural morphology and cell replica- freeze-etching. Cell Tissue Res. 235:225–241.
tion in the mouse olfactory mucosa. Toxicol. Pathol. Menco, B. Ph. M. (1986). A survey of ultra-rapid cryofixation
25:186–194. methods with particular emphasis on applications to freeze-
Mania-Farnell, B., and Farbman, A. I. (1990). fracturing, freeze-etching, and freeze-substitution.
Immunohistochemical localization of guanine nucleotide- J. Electron. Microsc. Techn. 4:177–240.
binding proteins in rat olfactory epithelium during development. Menco, B. Ph. M. (1988a). Pre-natal development of nasal
Develop. Brain Res. 51:103–112. epithelia. IV. Freeze-fracturing on apices, microvilli and pri-
Morphology of the Mammalian Olfactory Epithelium 45

mary and secondary cilia of olfactory and respiratory epithe- cyclase in peripheral olfactory systems. Light microscopic
lial cells, and on olfactory axons. Anat. Embryol. 178: and freeze-substitution electron microscopic immunocyto-
309–326. chemistry. J. Neurocytol. 23:708–727.
Menco, B. Ph. M. (1988b). Pre-natal development of rat nasal Menco, B. Ph. M., Cunningham, A. M., Qasba, P., Levy, N.,
epithelia. V. Freeze-fracturing on necklaces of primary and and Reed, R. R. (1997). Putative odour receptors localize in
secondary cilia of olfactory and respiratory epithelial cells. cilia of olfactory receptor cells in rat and mouse; a freeze-
Anat. Embryol. 178:382–388. substitution ultrastructural study. J. Neurocytol. 26:297–312,
Menco, B. Ph. M. (1988c). Tight junctional strands first appear in 691–706 (erratum).
regions where three cells meet in differentiating olfactory Menco, B. Ph. M., Birrell, G. B., Fuller, C. M., Ezeh, P. I., Keeton,
epithelium: a freeze fracture study. J. Cell Sci. 89:495–505. D. A., and Benos, D. J. (1998). Ultrastructural localization of
Menco, B. Ph. M. (1989). Electron microscopic demonstration amiloride-sensitive sodium channels and Na, K-ATPase in
of olfactory-marker protein with protein G-gold in freeze- the rat’s olfactory epithelial surface. Chem. Senses 23:137–149.
substituted, Lowicryl K11M-embedded rat olfactory receptor Menco, B. Ph. M., Carr, V. McM., Ezeh, P. I., Liman E. R.,
cells. Cell Tissue Res. 256:275–281. and Yankova, M. P. (2001). Ultrastructural localization of
Menco, B. Ph. M. (Ed.) (1992a). Electron Microscopy of G-proteins and the channel protein TRP2 to microvilli of
Olfactory Tissues: Peripheral Aspects. Microsc. Res. Techn. rat vomeronasal receptor cells. J. Comp. Neurol. 438:
22, 23: Parts I–IV. 468–489.
Menco, B. Ph. M. (1992b). Ultrastructural studies on membrane, Mendoza, A. S. and Kühnel, W. (1991). Postnatal changes in the
cytoskeletal, mucous, and protective compartments in olfac- ultrastructure of the rat olfactory epithelium: the supranuclear
tion. Microsc. Res. Techn. 22:215–224. region in supporting cells. Cell Tissue Res. 261:193–196.
Menco, B. Ph. M. (1992c). Lectins bind differentially to cilia and Meyer, M. R., Angele, A., Kremmer, E., Kaupp, U. B., and
microvilli of major and minor cell populations in olfactory and Müller, F. (2000). A cGMP-signaling pathway in a subset of
nasal respiratory epithelia. Microsc. Res. Techn. 23:181–199. olfactory sensory neurons. Proc. Natl. Acad. Sci. USA 97:
Menco, B. Ph. M. (1994). Ultrastructural aspects of olfactory trans- 10595–10600.
duction and perireceptor events. Semin. Cell Biol. 5: 11–24. Miller, M. L., Andringa, A., Evans, J. E., and Hastings, L. (1995).
Menco, B. Ph. M. (1995a). Scanning- and transmission electron Microvillar cells of the olfactory epithelium: morphology and
microscopy of olfaction. In Experimental Cell Biology of regeneration following exposure to toxic compounds. Brain
Taste and Olfaction. Current Techniques and Protocols, A. I. Res. 669:1–9.
Spielman and J. G. Brand (Eds.). CRC Press, Boca Raton, FL, Milne-Edwards, H. (1844). Histoire Naturelle, 1ère Partie:
pp. 115–125. Anatomie et Physiologie. Masson, Paris.
Menco, B. Ph. M. (1995b). Freeze-fracture, deep-etch, and Miragall, F., Breipohl, W., Naguro, T., and Voss-Wermbter, G.
freeze-substitution studies of olfactory epithelia, with special (1984). Freeze-fracture study of the plasma membranes of the
emphasis on immunocytochemical variables. Microsc. Res. septal olfactory organ of Masera. J. Neurocytol. 13: 111–125.
Techn. 32:337–356. Miyawaki, A., Homma, H., Tamura, H-O., Matsui, M., and
Menco, B. Ph. M., and Farbman, A. I. (1985a). Genesis of cilia Mikoshiba, K. (1996). Zonal distribution of sulfotransferase
and microvilli of rat nasal epithelia during pre-natal develop- in olfactory sustentacular cells. EMBO J. 15:2050–2055.
ment. I. Olfactory epithelium, qualitative studies. J. Cell Sci. Mombaerts, P. (1999). Molecular biology of odorant receptors in
78:283–310. vertebrates. Ann. Rev. Neurosci. 22:487–509.
Menco, B. Ph. M., and Farbman, A. I. (1985b). Genesis of cilia Monti Graziadei, G. A., Karlan, M. S., Bernstein, J. J., and
and microvilli of rat nasal epithelia during pre-natal develop- Graziadei, P. P. C. (1980). Reinnervation of the olfactory bulb
ment. 1I. Olfactory epithelium, a morphometric analysis. after the section of the olfactory nerves in monkey (Samiri
J. Cell Sci. 78:311–336. sciureus). Brain Res. 189:343–354.
Menco, B. Ph. M., and Farbman, A. I. (1992). Ultrastructural evi- Moran, D. T., Rowley, J. C. III, Jafek, B. W., and Lovell, M. A.
dence for multiple mucous domains in frog olfactory epithe- (1982a). The fine structure of the olfactory mucosa in man.
lium. Cell Tissue Res. 270:47–56. J. Neurocytol. 11:721–746.
Menco, B. Ph. M., and Jackson, J.E. (1997a). A banded topog- Moran, D. T., Rowley, J. C., III, and Jafek, B. W. (1982b).
raphy in the developing rat’s olfactory epithelial surface. Electron microscopy of human olfactory epithelium reveals a
J. Comp. Neurol. 388:293–306. new cell type: the microvillar cell. Brain Res. 253:39–46.
Menco, B. Ph. M., and Jackson, J. E. (1997b). Cells resembling Moran, D. T., Jafek, B. W., Rowley, J. C., III, and Eller, P. M.
hair cells in developing rat olfactory and nasal respiratory (1985). Electron microscopy of olfactory epithelia in two
epithelia. Tissue Cell 29:707–713. patients with anosmia. Arch. Otolaryngol. 111:122–126.
Menco, B. Ph. M., Bruch, R. C., Dau, B., and Danho, W. (1992). Moran, D. T., Jafek, B. W., Eller, P. M., and Rowley, J. C., III
Ultrastructural localization of olfactory transduction compon- (1992a). The ultrastructural histopathology of human olfac-
ents: the G protein subunit Golf and type III adenylyl cyclase. tory dysfunction. Microsc. Res. Techn. 23:103–110.
Neuron 8:441–453. Moran, D. T., Rowley, J. C., III, Lanza, D. C., Kratskin, I., Kennedy,
Menco, B. Ph. M., Tekula, F. D., Farbman, A. I., and Danho, W. D. W., and Doty, R. L. (1992b). Intermediate voltage electron
(1994). Developmental expression of G-proteins and adenylyl microscopy (IVEM) of olfactory epithelia in patients with
46 Menco and Morrison

Parkinson’s disease and Alzheimer’s disease. Chem. Senses Nakamura, T., and Gold, G. H. (1987). A cyclic nucleotide-gated
17:673, abstr. 199. conductance in olfactory receptor cilia. Nature 325:
Moran, D. T., Jafek, B. W., and Rowley, J. C., III (1992c). The 442–444.
ultrastructure of the human olfactory mucosa. In The Human Nakashima, T., Tanaka, M., Inamitsu, M., and Uemura, T. (1991).
Sense of Smell, D. Laing, R. L. Doty, and W. Breipohl (Eds.). Immunohistopathology of variations of human olfactory
Springer-Verlag, Berlin, pp. 1–25. mucosa. Eur. Arch. Otorhinolaryngol. 248:370–375.
Moran, D. T., Rowley, J. C., III, Aiken, G. R., and Jafek, B. W. Negus, V. (1958). The Comparative Anatomy and Physiology of the
(1992d). Ultrastructural neurobiology of the olfactory mucosa of Nose and Paranasal Sinuses. E. & S. Livingstone, Edinburgh.
the brown trout, Salmo trutta. Microsc. Res. Techn. 23:28–48. Newman, M. P., Féron, F., and Mackay-Sim, A. (2000). Growth fac-
Mori, K., Von Campenhausen, H., and Yoshihara, Y. (2000). tor regulation of neurogenesis in adult olfactory epithelium.
Zonal organization of the mammalian main and accessory Neuroscience 99:343–350.
olfactory systems. Phil. Trans. Roy. Soc. Lond., Ser. B., Noë, J., Tareilus, E., Boekhoff, I., and Breer, H. (1997).
Biol. Sci. 355:1801–1812. Sodium/calcium exchanger in rat olfactory neurons.
Morrison, E. E., and Costanzo, R. M. (1989). Scanning electron Neurochem. Int. 30:523–531.
microscopic study of degeneration and regeneration in the Norlin, E. M., and Berghard, A. (2001). Spatially restricted
olfactory epithelium after axotomy. J. Neurocytol expression of regulators of G-protein signaling in primary
18:393–405. olfactory neurons. Mol. Cell. Neurosci. 17:872–882.
Morrison, E. E., and Costanzo, R. M. (1990). Morphology of the Ogata, T. (2001). Mammalian tuft (brush) cells and chloride cells
human olfactory epithelium. J. Comp. Neurol. 297:1–13. of other vertebrates share a similar structure and cytochemical
Morrison, E. E., and Costanzo, R. M. (1992). Morphology of reactivities. Acta Histochem. Cytochem. 33: 439–449.
olfactory epithelium in humans and other vertebrates. Ohno, I., Ohyama, M., Hanamura, Y., and Ogawa, K. (1981).
Microsc. Res. Techn 23:49–61. Comparative anatomy of olfactory epithelium. Biomed Res.
Morrison, E. E., and Costanzo, R. M. (1995). Regeneration 2(Suppl):455–458.
of olfactory sensory neurons and reconnection in the Ohta, Y., and Ichimura, K. (2001). Globose basal cells are identi-
aging hamster central nervous system. Neurosci. Lett. fied as proliferating cells in mouse olfactory epithelium. Ann.
198:213–217. Otol. Rhino. Laryngol. 110:53–55.
Morrison, E. E., Leipold, H., and Kruckenberg, S. (1983). The Okamura, H., Sugai, N., and Ohtani, I. (1996). Identification of
microscopic anatomy of the respiratory epithelium in the nasal epithelial cells with carbonic anhydrase activity. Brain
coyote Canis latrans. Anat. Histol. Embryol. Zentralblatt, Res. 728:263–266.
Series C 12:325–340. Okano, M., Weber, A. F., and Frommes, S. P. (1967). Electron
Müller, F., and Kaupp, U. B. (1998). Signaltransduktion in microscopic studies of the distal border of the canine olfac-
Sehzellen. Naturwissenschaften 85:49–61. tory epithelium. J. Ultrastruct. Res. 17:487–502.
Müller, F., Bönigk, W., Sesti, F., and Frings, S. (1998). Ophir, D., and Lancet, D. (1988). Expression of intermediate fila-
Phosphorylation of mammalian olfactory cyclic nucleotide- ments and desmoplakin in vertebrate olfactory mucosa.
gated channels increases ligand sensitivity. J. Neurosci. Anat. Rec. 221:754–760.
18:164–173. Paik, S. I., Lehman, M. N., Seiden, A. M., Duncan, H. J., and
Mulvaney, B. D., and Heist, H. E. (1971). Centriole migration Smith, D. V. (1992). Human olfactory biopsy. The influence
during regeneration and normal development of olfactory of age and receptor distribution. Arch. Oto.-Head Neck Surg.
epithelium. J. Ultrastruct. Res. 35:274–281. 118:731–738.
Murrell, W., Bushell, G. R., Livesey, J., McGrath, J., MacDonald Parker, G. H. (1922). Smell, Taste and Allied Senses in the
K. P. A., Bates, P. R., and Mackay-Sim, A. (1996). Vertebrates. Monographs of Experimental Biology, Vol. VI.
Neurogenesis in adult human. NeuroRep. 7:1189–1194. Lippincott, Philadelphia.
Naessen, R. (1970). The identification and topographical localisation Paysan, J., and Breer, H. (2001). Molecular physiology of odor
of the olfactory epithelium in man and other mammals. Acta detection: current views. Pflügers Arch.-Eur. J Physiol.
Otolaryngol. (Stockh.) 70:51–57. 441:579–586.
Naessen, R. (1971a). An enquiry on the morphological character- Pellier, V., Saucier, D., Oestreicher, A. B., and Astic, L. (1996).
istics and possible changes with age in the olfactory region of Ultrastructural and cytochemical identification of apoptotic
man. Acta Otolaryngol (Stockh.) 71:49–62. cell death accompanying development of the fetal rat olfac-
Naessen, R. (1971b). The receptor surface of the olfactory organ tory nerve layer. Anat. Embryol. 194: 99–109.
(epithelium) of man and guinea pig. Acta Otolaryngol. Pelosi, P. (1994). Odorant-binding proteins. Crit. Rev. Biochem.
(Stockh.) 71:335–348. Mol. Biol. 29:199–228.
Naguro, T., and Iwashita, K. (1992). Olfactory epithelium in young Pelosi, P. (2001). The role of perireceptor events in vertebrate
adult and aging rats as seen with high-resolution scanning elec- olfaction. Cell. Mol. Life Sci. 58:503–509.
tron microscopy. Microsc. Res. Techn. 23:62–75. Peppel, K., Boekhoff, I., McDonald, P., Breer, H., Caron, M. G.,
Nakamura, T. (2000). Cellular and molecular constituents of and Lefkowitz, R. J. (1997). G protein-coupled receptor
olfactory sensation in vertebrates. Comp. Biochem. Physiol. A kinase 3 (GRK3) gene disruption leads to loss of odorant
126:17–32. receptor desensitization. J. Biol. Chem. 41:25425–25428.
Morphology of the Mammalian Olfactory Epithelium 47

Pixley, S. K. (1992). The olfactory nerve contains two popula- Cagan and M. R. Kare (Eds.). Academic Press, New York,
tions of glia, identified both in vivo and in vitro. Glia pp. 47–68.
5:269–284. Rhein, L. D., Cagan, R. H., Orkand, P. M., and Dolack, M. K.
Pixley, S. K., Farbman, A. I., and Menco, B. Ph. M. (1997). A (1981). Surface specializations of the olfactory epithelium of
monoclonal antibody marker for olfactory sustentacular cell rainbow trout, Salmo gairdneri. Tissue Cell 13:577–587.
microvilli. Anat. Rec. 248:307–321. Rosli, Y., Breckenridge, L. J., and Smith, R. A. (1999). An ultra-
Plattner, H., and Klauke, N. (2001). Calcium in ciliated protozoa: structural study of age-related changes in mouse olfactory
sources, regulation, and calcium-regulated cell functions. Int. epithelium. J. Electron Microsc. 48:77–84.
Rev. Cytol. 201:115–208. Rowley, J. C., Moran, D. T., and Jafek, B. W. (1989).
Plendl, J., and Sinowatz, F. (1998). Glycobiology of the olfac- Peroxidase backfills suggest the mammalian olfactory
tory system. Acta Anat. 161:234–253. epithelium contains a second morphologically distinct class
Plendl, J., Stiersstirfer, B., and Sinowatz, F. (1999). Growth fac- of bipolar sensory neuron: the microvillar cell. Brain Res.
tors and their receptors in the olfactory system. Anat. Histol. 502:387–400.
Embryol. 28:73–79. Saini, K. D., and Breipohl, W. (1976). Surface morphology in the
Proctor, D. F., and Anderson, I. B. (Eds.) (1982). The Nose: olfactory epithelium of normal and female Rhesus monkeys.
Upper Airway Physiology and the Atmospheric Environment. Am. J. Anat. 147:433–445.
Elsevier, Amsterdam. Scarpa, A. (1789). Anatomicae disquisitiones de audito et olfacto.
Pyatkina, G. A. (1982). Development of the olfactory epithelium Secto tertia. Ticini, pp. 64–85.
in man. Z. mikrosk.-anat. Forsch., Leipzig 96:361–372. Schild, D., and Restrepo, D. (1998). Transduction mechanisms in
Pyatkina, G., and Agasandyan, Kh. (1991). Microvillar cells—a vertebrate olfactory receptor cells. Physiol. Rev. 78:429–466.
new type of receptor cells in pig olfactory mucosa. Schleicher, S., Boekhoff, I., Arriza, J., Lefkowitz, R. J., and
Pheromones 1:129–137. Breer, H. (1993). A -adrenergic receptor kinase-like enzyme
Rafols, J. A., and Getchell, T. V. (1983). Morphological relations is involved in olfactory signal termination. Proc. Natl. Acad.
between the receptor neurons, sustentacular cells and Schwann Sci. USA 90:1420–1424.
cells in the olfactory mucosa of the salamander. Anat. Rec. Schultze, M. (1856). Über die Endigungsweise des Geruchs-
206:87–101. nerven und der Epitelialgebilde des Nasenschleimhaut. M.
Rambotti, M. G., Saccardi, C., Spreca, A., Aisa, M. C., Ber. Preuss. Acad. Wiss., Berlin 21: 504–514.
Giambanco, I., and Donato, R. (1989). Immunocytochemical Schultze, M. (1862). Untersuchungen über den Bau der Nasen-
localization of S-100 protein in olfactory and supporting schleimhaut, namentlich die Struktur and Endigungsweise der
cells of lamb olfactory epithelium. J. Histochem. Cytochem. Geruchsnerven bei dem Menschen und den Wirbeltiere.
37:1825–1833. Abhandl. naturforsch. Ges., Halle 7:1–100.
Ramón-Cueto, A., and Avila, J. (1998). Olfactory ensheathing Schwob, J. E., Leopold, D. A., Mieleszko Szumowski, K. E., and
cells: properties and function. Brain Res. Bull. 46:175–187. Emko, P. (1993). Histopathology of olfactory mucosa in
Rawson, N. E. (2000). Human olfaction. In The Neurobiology of Kallmann’s syndrome. Ann. Otol. Rhinol. Laryngol.
Taste and Smell, 2nd ed., T. E. Finger, W. L. Silver and D. 102:117–122.
Restrepo (Eds.). Wiley-Liss, Inc., New York, pp. 257–284. Schwob, J. E., Huard, J. M. T., Luskin, M. B., and Youngentob, S. L.
Reese, T. S. (1965). Olfactory cilia in the frog. J. Cell Biol. (1994). Retroviral lineage studies of the rat olfactory epithelium.
25:209–230. Chem. Senses 19:671–682.
Ressler, K. J., Sullivan, S. L., and Buck, L. B. (1993). A zonal Schwob, J. E., Youngentob, S. L., and Mezza, R. C. (1995).
organization of odorant receptor gene expression in the olfac- Reconstitution of the rat olfactory epithelium after methyl
tory epithelium. Cell 73:597–609. bromide-induced lesion. J. Comp. Neurol. 359:15–37.
Ressler, K. J., Sullivan, S. L., and Buck, L. B. (1994). Scott, J. W., and Brierley, T. (1999). A functional map in rat olfac-
Information coding in the olfactory system: evidence for a tory epithelium. Chem. Senses 24:679–690.
stereotyped and highly organized epitope map in the olfac- Seifert, K. (1969). Geschichte und Bibliographie der Erforschung
tory bulb. Cell 79:1245–1255. des peripheren Geruchsorgans. Clio Medica 4:305–337.
Reuter, D., Zierold, K., Schröder, W. H., and Frings, S. (1998). Seifert, K. (1970). Die Ultrastruktur des Riechepithels beim
A depolarizing chloride current contributes to chemoelec- Makrosmatiker. Eine elektronenmikroskopische Untersuchung.
trical transduction in olfactory sensory neurons in situ. In Normale und pathologische Anatomie, Heft 21, W.
J. Neurosci. 18:6623–6630. Bargmann and E. Doerr (Eds.), Georg Thieme Verlag, Stuttgart.
Reyes, P. F., Deems, D. A., and Suarez, M. G. (1993). Olfactory- Seifert, K. (1971). Licht- und elektronenmikroskopische
related changes in Alzheimer’s disease: a quantitative neuro- Untersuchungen der Bowman-Drüsen in der Riechschleimhaut
pathologic study. Brain Res. Bull. 32:1–5. makrosmatischer Säuger. Arch. klin. exp. Ohr.-, Nas.- u Kehlk.
Reznik-Schüller, H. M. (1983). Pathogenesis of tumors induced Heilk. 200:252–274.
with N-nitrosomethylpiperazine in the olfactory region of the Seifert, K. (1972). Neue Ergebnisse licht- und elektronen-
rat nasal cavity. J. Natl. Canc. Inst. 71:165–171. mikroskopischer Untersuchungen am peripheren Geruchsorgan
Rhein, L. D., and Cagan, R. H. (1981). Role of cilia in olfactory einschließlich der Bowman-Drüsen und des Organen vomero-
recognition. In Biochemistry of Taste and Olfaction, R. H. nasal. Acta Oto-rhino-laryngol. Belg. 26: 463–492.
48 Menco and Morrison

Sinnarajah, S., Dessauer, C. W., Srikumar, D., Chen, J., Yuen, J., Swoboda, P., Adler, H. T., and Thomas, J. H. (2000). The RFX-
Yilma, S., Dennis, J. C., Morrison, E. E., Vodyanoy, V., and type transcription factor DAF-19 regulates sensory neuron cil-
Kehrl, J. H. (2001). RSG2 regulates signal transduction in ium formation in C. elegans. Mol. Cell 5:411–421.
olfactory neurons by attenuating activation of adenylyl Takahashi, S., Iwanega, T., Takahashi, Y., Nakano, Y., and Fujita, T.
cyclase III. Nature 409:1051–1055. (1984). Neuron specific enolase, neurofilament protein and
Smith, R. L., Baker, H., Kolstad, K., Spencer, D. D., and Green, S-100 protein in the olfactory mucosa of human fetuses. An
C. A. (1991). Localization of tyrosine hydroxylase and olfac- immunohistochemical study. Cell Tissue Res. 238: 231–234.
tory marker protein immunoreactivities in the human and Takahashi, H. Ohara, S., Yamada, M., Ikuta, F., Tanimura, K., and
macaque olfactory bulb. Brain Res. 548:140–148. Honda, Y. (1987). Estesioneuroepithelioma: a tumor of true
Smith, T. D. (Ed.) (1998). Histology and Histochemistry of the olfactory epithelium origin. An ultrastructural and immuno-
Vomeronasal Organ. Microsc. Res. Techn. 41, 43: Parts I and II. histochemical study. Acta Neuropathol. (Berl.) 75:147–155.
Smith, T. D., Siegel, M. I., and Bhatnagar, K. P. (2001). Talamo, B. R., Feng, W.-H., and Stockmayer, M. (1994). Human
Reappraisal of the vomeronasal system of catarrhine pri- olfactory epithelium: Normal patterns and types of lesions found
mates: ontogeny, morphology, functionality, and persisting in the general population. Inhal. Toxicol. 6(Suppl): 249–275.
questions. Anat. Rec. 265:179–192. Tarozzo, C., Peretto, P., and Fasolo, A. (1995). Cell migration
Spielman, A. I. (1998). Chemosensory function and dysfunction. from the olfactory placode and the ontogeny of the neuro-
Crit. Rev. Oral. Biol. Med. 9:267–291. endocrine compartments. Zool. Sci. 12:367–383.
Spreca, A., and Rambotti, M. G. (1994). Ultracytochemical local- Taxy, J. B., Bharani, N. K., Mills, S. E., Frierson, H. F., and
ization of particulate guanylate cyclase after stimulation with Gould, V. E. (1986). The spectrum of olfactory neural tumors.
natriuretic peptides in lamb olfactory mucosa. Histochem. J. A light-microscopic immunohistochemical and ultrastructural
26:778–786. analysis. Am. J. Surg. Pathol. 10:687–695.
Steinbrecht, R. A. (1999). Olfactory receptors. In Atlas of Arthropod Tegoni, M., Pelosi, P., Vincent, F., Spinelli, S., Campanacci, V., Grolli,
Sensory Receptors, E. Eguchi and Y. Tominaga (Eds.). Springer- S., Ramoni, R., and Cambillau, C. (2000). Mammalian odorant
Verlag, Berlin, pp. 155–176. binding proteins. Biochem. Biophys. Acta 1482:229–240.
Stephens, R. E. (1974). Enzymatic and structural proteins of the Todd, R. B., and Bowman, W. (1847). The Physiological
axoneme. In Cilia and Flagella, M. A. Sleigh (Ed.). Academic Anatomy and Physiology of Man, Vol. II. Parker, London, p. 1.
Press, New York, pp. 39–76. Tolbert, L. P. (Ed.) (1993). Olfactory Centers in the Brain.
Strotmann, J., Wanner, I., Helfrich, T., Beck, A., and H. Breer, Microsc. Res. Techn. 24: Parts I and II.
H. (1994). Rostro-caudal patterning of receptor-expressing Trotier, D., Eliot, C., Wassef, M., Talmain, G., Bensimon, J. L.,
olfactory neurones in the rat nasal cavity. Cell Tissue Res. Døving, K. B., and Ferrand, J. (2000). The vomeronasal
278:11–20. cavity in adult humans. Chem. Senses 25:369–380.
Strotmann, J., Conzelmann, S., and Breer, H. (1996). Laminar Truwitt, C. L., Barkovich, A. J., Grumbach, M. M., and Martini,
segregation of odorant receptor expression in the olfactory J. J. (1993). MR imaging of Kallmann syndrome, a genetic
epithelium. Cell Tissue Res. 284:347–354. disorder of neuronal migration affecting the olfactory and
Sullivan, S. L., and Dryer, L. (1996). Information processing in genital systems. Am. J. Neurorad. 14:827–838.
mammalian olfactory system. J. Neurobiol. 30:20–36. Vaccarezza, O. L., Septa, L. N., and Tramezzani, J. H. (1981). The
Sunderman, F. W. (2001). Review: nasal toxicity, carcinogenicity vomeronasal organ of the rat. J. Anat. 132:167–185.
and olfactory uptake of metals. Ann. Clin. Lab. Sci. 31: 3–24. Vassar, R., Ngai, J., and Axel, R. (1993). Spatial segregation of
Suzuki, Y., and Takeda, M. (1991a). Keratins in the developing odorant receptor expression in the mammalian olfactory
olfactory epithelia. Dev. Brain Res. 59:171–178. epithelium. Cell 74:309–318.
Suzuki, Y., and Takeda, M. (1991b). Basal cells in the mouse Vassar, R., Chao, S. K., Sitcheran, R., Nunez, J. M., Vosshall, L.
olfactory epithelium after axotomy: immunohistochemical B., and Axel, R. (1994). Topographic organization of sensory
and electron microscopic studies. Cell Tissue Res. projections to the olfactory bulb. Cell 79:981–991.
266:239–245. Verkman, A. S., and Mitra, A. K. (2000). Structure and function
Suzuki, Y., and Takeda, M. (1993). Basal cells in the mouse olfac- of aquaporin water channels. Am. J. Physiol. 278:F13–F28.
tory epithelium during development: immunohistochemical Vogl, A., Noé, J., Breer, H., and Boekhoff, I. (2000). Cross-talk
and electron microscopic studies. Dev. Brain Res. 73: 107–113. between olfactory second messenger pathways. Eur. J.
Suzuki, Y., Schafer, J., and Farbman, A. I. (1995). Phagocytotic Biochem. 267:4529–4535.
cells in the rat olfactory epithelium after bulbectomy. Exp. Vollrath, M., Altmannsberger, M., Weber, K., and Osborn, M.
Neurol. 136:225–233. (1985). An ultrastructural and immunohistological study of
Suzuki, Y., Takeda, M., and Farbman, A. I. (1996). Supporting cells the rat olfactory epithelium: unique properties of olfactory
as phagocytes in the olfactory epithelium after bulbectomy. sensory cells. Differentiation 29:243–253.
J. Comp. Neurol. 376:509–517. von Brunn, A. (1892). Beiträge zur mikroskopischen Anatomie der
Suzuki, Y., Takeda, M., Obara, N., Suzuki, N., and Takeichi, N. menschlichen Nasenhöhle. Arch. mikrosk. Anat. 39: 632–651.
(2000). Olfactory epithelium consisting of supporting cells and von Kö1liker, R. A. (1858). Ausbreitung der Nerven in der
horizontal basal cells in the posterior nasal cavity of mice. Cell Geruchsschleimhaut um Plagiostomen. Verh. phys.-med. Ges.,
Tissue Res. 299:313–325. Würzburg 8:31–36.
Morphology of the Mammalian Olfactory Epithelium 49

Wang, F., Nemes, A., Mendelsohn, M., and Axel, R. (1998). Yamada, S., Miyoshi, Y., Skakura, Y., Yamagiwa, M., Taniguchi,
Odorant receptors govern the formation of a precise topo- T., and Fukukita, K. (1980). Scanning electron microscopic
graphic map. Cell 93:47–60. studies of human olfactory epithelia. J. Clin. Electron Microsc.
Wei, J., Zhao, A. Z., Chan, G. C. K., Baker, L. P., Impey, S., 13:63–68.
Beavo, J. A., and Storm, D. R. (1998). Phosphorylation and Yamagishi, M., Hasegawa, S., and Nakano, Y. (1988). Examination
inhibition of olfactory adenylyl cyclase by CaM kinase II in and classification of human olfactory mucosa in patients
neurons: a mechanism for attenuation of olfactory signals. with clinical olfactory disturbances. Arch. Otorhinolaryngol.
Neuron 21:495–504. 245:316–320.
Weiler, E., and Farbman, A. I. (1997). Proliferation in the rat Yamagishi, M., Ishizuka, Y., Fujiwara, M., Nakamura, H.,
olfactory epithelium: age-dependent changes. J. Neurosci. Igarashi, S., Nakano, Y., Kuwano, R. (1993). Distribution
17:3610–3622. of calcium binding proteins in sensory organs of the ear,
Weiler, E., and Farbman, A. I. (1998). Supporting cell prolifera- nose and throat. Acta Otolaryngol. (Stockh.) 506(Suppl.):
tion in the rat olfactory epithelium decreases postnatally. Glia 85–89.
22:315–328. Yan, C., Zhao, A. Z., Bentley, J. K., Loughney, K., Ferguson, K.,
Wenisch, S., Andressen, C., Derouiche, A., Arnhold, S., Addicks, and Beavo, J. A. (1995). Molecular cloning and characteriza-
K., and Leiser, R. (2000). Heme oxygenase-2 and nitric oxide tion of a calmodulin-dependent phosphodiesterase enriched in
synthase immunoreactivity of bovine olfactory receptor olfactory sensory neurons. Proc. Natl. Acad. Sci. USA
neurons and a comparison with the distribution of 92:9677–9681.
NADPH-diaphorase staining. Histochem. J. 32:381–388. Yousem, D. M., Geckle, R. J., Bilker, W. B., McKeown, D., and
Wenisch, S., Andressen, C., Addicks, K., Arnhold, S., and Leiser, Doty, R. L. (1996). MR evaluation of patients with congenital
R. (2001). Heme oxygenase-2 immunoreactivity in develop- hyposmia or anosmia. Am. J. Radiol. 166:439–443.
ing and mature bovine olfactory epithelium. Dev. Brain Res. Zhang, C., Finger, T. E., and Restrepo, D. (2000). Mature olfac-
126:101–108. tory receptor neurons express connexin 43. J. Comp. Neurol.
Whitlock, K. E., and Westerfield, M. (1998). A transient popula- 426:1–12.
tion of neurons pioneers the olfactory pathway in the zebrafish. Zielinski, B. S., and Hara, T. J. (1992). Ciliated and microvillar
J. Neurosci. 18:8919–8927. receptor cells degenerate and then differentiate in the olfac-
Wong, S. T., Trinh, K., Hacker, B., Chan, G. C. K., Lowe, G., tory epithelium of rainbow trout following olfactory nerve
Gaggar, A., Xia., Z., Gold, G. H., and Storm, D. R. (2000). section. Microsc. Res. Techn. 23:22–27.
Disruption of the type III adenylyl cyclase gene leads to Zippel, H. P. (1993). Historical aspects of research on the verte-
peripheral and behavioral anosmia in transgenic mice. Neuron brate olfactory system. Naturwissenschaften 80:65–76.
27:487–497. Zufall, F., and Leinders-Zufall, T. (2000). The cellular
Yamada, S. (1983). Scanning electron microscopic study of and molecular basis of odor adaptation. Chem. Senses
olfactory epithelia. J. Clin. Electron Microsc. 16:95–108. 25:473–481.
3

Olfactory Mucosa: Composition, Enzymatic Localization,

and Metabolism

Xinxin Ding
New York State Department of Health and State University of New York at Albany, Albany, New York, U.S.A.
Alan R. Dahl
Battelle Memorial Institute, Columbus, Ohio, U.S.A.

I. INTRODUCTION their metabolic capacities are compared across several

species. Finally, the potential for these enzymes to modu-
The olfactory mucosa, as well as the nasal respiratory late the toxicity of inhalants and to influence odor signal
mucosa, has a very high metabolic capacity for endogen- detection is discussed.
ous and exogenous, or xenobiotic, substrates. Olfactory This chapter is not a review of all specific isozymes
tissue also has a high degree of inflammatory and immune detected in nasal tissues, their localization, or the spe-
responsiveness stimulated by contact with foreign cific toxic effects they are thought to mediate. For such a
substances, exfoliates in response to toxic insult, and detailed review of nasal enzymes, the reader is referred to
regenerates to varying degrees following this exfoliation. Dahl and Hadley (1991), Ding and Coon (1993), and
The olfactory epithelium is unique in containing the only Thornton-Manning and Dahl (1997). This chapter
recognized mammalian neurons that regenerate from pre- describes the complexity of the composition and regula-
cursor basal cells. In addition, these neurons are unique in tion of the major known biotransformation enzymes.
contacting the external environment with their dendritic Additionally, it discusses various physiological and
processes while the axonal processes of the same cells pathological processes in which nasal metabolic activity
synapse within the central nervous system (CNS) in the is thought to play a role, providing the reader with a
olfactory bulbs. The olfactory mucosa, therefore, framework in which to incorporate enzyme-specific
represents a tissue where interactions are continually information and relevant sources for a more detailed
occurring between secretory processes, immune responses, examination of specific questions. In many cases, data on
neural signaling, and cell death and development. This nasal metabolism have been obtained from whole tissue
chapter examines the basic structure and cell types of the homogenates, making it impossible to determine the rela-
olfactory mucosa and then focuses primarily on the tive contributions of epithelial or subepithelial enzymes,
enzymatic capacity of this tissue. The anatomical charac- or olfactory versus respiratory mucosa. Because the
teristics generally common to all species are outlined and olfactory mucosa occupies the most caudal region of the
followed by a brief discussion of interspecies variability nose, metabolism in other nasal regions can affect olfac-
in the magnitude, localization, or occurrence of these tory function as well. Therefore, data from nasal
characteristics. The localization of nasal enzymes and homogenates are also discussed and noted as such.

51
52 Ding and Dahl

II. ANATOMY OF THE NASAL CAVITY displays metaplastic changes in response to toxicants. For
example, chronic exposure to ozone results in a metaplastic
To understand the contribution of metabolism in the olfac- change of the transitional epithelium to secretory, respira-
tory mucosa, it is necessary to understand its relationship tory epithelium, and cigarette smoke exposure produces
to other nasal tissues illustrated in Figure 1. Before reaching squamous metaplasia in this region (Harkema, 1990).
the olfactory mucosa, inhaled air comes in contact with Continuing in a caudal direction, the nasal cavity is
three other epithelial types in the nasal cavity: squamous, lined by a respiratory mucosa consisting of an epithelium
transitional, and respiratory. These epithelial regions differ made up of ciliated cells and mucus-secreting goblet cells.
in their metabolic capacities, but metabolism in these The ciliated cells are responsible for movement of the
tissues can alter the chemical composition of inhaled toxi- mucous layer through the nasal cavity. Underlying this
cants before they reach the olfactory mucosa. epithelium are subepithelial glands that produce the major-
The anterior vestibule of the nasal cavity is lined with a ity of serous secretions in the nose. The subepithelial
stratified squamous epithelium. Although enzymes primar- glands also secrete mucus to the mucous layer. The respi-
ily involved in metabolism of endogenous substrates such ratory mucosa plays the major role in both production of
as alkaline phosphatase and gamma-glutamyl peptidase nasal secretions and clearance of inhaled materials. High
have been localized to squamous epithelium (Randall et al., metabolic capacity is found in the respiratory mucosa as
1987), this epithelium has not been reported to have well.
xenobiotic-metabolizing capacity. The squamous epithelium The most dorsal and caudal region of the nasal cavity is
has neither secretory capacity nor cilia. The squamous epi- lined by the olfactory mucosa. The surface area of this
thelium gives way, in some species, to a narrow region of mucosa is greatly enhanced by a convoluted turbinate
transitional epithelium, which is a cuboidal, nonciliated structure, which varies greatly across species. The olfactory
epithelium that has a high metabolic capacity for substrates mucosa is composed of the olfactory epithelium lining
of specific cytochrome P450 (P450 or CYP) enzymes the nasal cavity and separated from the underlying lamina
(Bond et al., 1988). This transitional epithelium often propria by the basal lamina (Fig. 2).

III. COMPOSITION OF THE OLFACTORY

MUCOSA

The following is a brief, general description of the olfactory

mucosa discussing primarily those elements common
to most species. For more detail, the reader is referred
to the following reviews: for general nasal and olfactory
tissue anatomy, Sorokin (1988) and Uraih and Maronpot
(1990); for comparative anatomy, Reznik (1990); and
for human olfactory anatomy, Chapters 1 and 2 in this
volume.

A. Epithelial Cell Types

The olfactory epithelium is made up of four primary cell

types: the olfactory receptor cells, the sustentacular or sup-
porting cells, the basal cells, and the duct cells of Bowman’s
glands (Fig. 2). Cilia containing the olfactory receptors
project from the receptor cells into the mucous layer
lining the nasal cavity. These cells are unique in two
respects: (1) they project directly into the brain before their
Figure 1 Epithelia of the human nasal cavity. Olfactory
epithelium (OE); Bowman’s gland (BG); olfactory nerve (ON);
first synapse, which makes them the only cells directly con-
olfactory receptor cell (R); sustentacular cell (S); respiratory tacting both the CNS and the external environment; and (2)
epithelium (RE); squamous epithelium (SE); transitional epithe- in contrast to almost all other neuronal cells, olfactory
lium (TE); nasopharynx (NP); hard palate (HP); inferior turbinate receptor cells regenerate from basal cells after damage
(IT); middle turbinate (MT); superior turbinate (ST). (Graziadei and Monti-Graziadei, 1983; Huard et al., 1998).
Olfactory Mucosa 53

Figure 2 Cellular anatomy of the olfactory mucosa. The left panel shows a transverse section in the region of the ethmoturbinates of an
adult rat. Tissues on the lumenal side of the basal lamina compose the olfactory epithelium (OE), and tissue inferior to the basal lamina
forms the lamina propria (LP). The two layers are included in the olfactory mucosa (OM). Structures identified include sustentacular cell
nuclei (sn), olfactory neuronal cells (n), basal cells (b), olfactory nerve bundles (on), Bowman’s glands (bg), Bowman’s gland ducts (d),
blood vessels (bv), and nasal airway (NA). 5 m paraffin-embedded sections stained with hematoxylin and eosin; approximately x250
(Modified from Gu et al., 1997). The right panel shows a transmission electron micrograph of the olfactory epithelial surface of an adult
rat (approximately x3500). The structures identified are olfactory receptor cells (RC); olfactory dendritic knobs (OK); cilia of receptor
neurons (C); sustentacular (or supporting) cells (SC), and microvilli (MV) at the lumenal surface of supporting cells.

The cilia on the receptor cells are nonmotile. These receptor B. Subepithelial Structure
cells generally have very little xenobiotic-metabolizing
capacity. The lamina propria consists of the acinar cells of
The majority of xenobiotic-metabolizing enzymes in Bowman’s glands, olfactory nerves and their associated
the olfactory epithelium have been localized to the susten- Schwann cells, blood vessels, and connective tissue.
tacular cells, the duct cells of Bowman’s glands, and the Because this tissue is so often exposed to inhaled foreign
progenitor basal cells. Sustentacular cells have secretory substances, cells associated with inflammation and immun-
functions in some species (Getchell et al., 1988; Zielinski ity—including neutrophils, plasma cells, monocytes, and
et al., 1988), but generally are not the primary source of the macrophages—often are present within the submucosa and
seromucous secretions covering the olfactory epithelium. epithelium.
54 Ding and Dahl

1. Bowman’s Glands a number of antioxidants (Cross et al., 1994), such as

reduced glutathione (GSH), mucin, and an abundant,
These subepithelial glands are the primary source of
thiol-specific antioxidant protein belonging to the mono-
mucous and serous secretions in the olfactory mucosa. The
cysteine subfamily of peroxiredoxins (Novoselov et al.,
acinar and duct cells of Bowman’s glands contain many
1999). As mentioned previously, preliminary evidence
xenobiotic-metabolizing enzymes, although this localiza-
indicates that some xenobiotic-metabolizing enzymes may
tion is species dependent. In some cases there is evidence
be secreted to nasal mucus as well. Nasal mucus also con-
that these enzymes are secreted, again depending on the
tains regulatory proteins and peptides, including secretory
species (Bogdanffy et al., 1987; Chen et al., 1992; Lewis
leukoprotease inhibitor (Lee et al., 1993), substance P,
et al., 1992a). However, whether the secretion is from
vasoactive intestinal peptide (Chaen et al., 1993), and
Bowman’s glands or from sustentacular cells has not been
insulin-like growth factor I and its binding proteins
determined.
(Federico et al., 1999), as well as transport proteins such as
odorant-binding proteins (Pelosi, 1996) and vomeromod-
2. Blood Vessels
ulin (Krishna et al., 1995b). The proteins in the mucus are
The highly vascularized lamina propria in the olfactory thought to result from either serum transudation (e.g.,
mucosa is supplied by the ethmoidal artery, a source dis- albumin, transferrin, and carboxypeptidase) or local
tinct from the sphenopalatine supply of the respiratory synthesis and secretion (e.g., the metal-binding protein
mucosa. In mice, the blood flow through the total nasal lactoferrin, lysozyme, neutral endopeptidase, and antipro-
mucosa has been estimated at 0.87% of cardiac output teases) (Ohkubo et al., 1994, 1995).
(Stott et al., 1983), and in rats, 0.32 mL/min (Morris and Nasal secretion may be controlled by nerve stimulation
Cavanagh, 1986), or 0.53% of cardiac output (Stott et al., (Revington et al., 1997) and by corticosteroids (Fong et al.,
1983). The high perfusion rates in nasal tissues allow for 1999). The latter, via mineralocorticoid receptors in sup-
rapid absorption and systemic distribution of substances porting cells and Bowman’s glands (Robinson et al.,
that penetrate the olfactory epithelium. Conversely, this 1999), may modulate olfactory Na,K-ATPase and
high perfusion can also allow toxicants in the bloodstream active ion transport, which results in hyperosmolarity of
to come in contact with xenobiotic-metabolizing enzymes mucus with respect to serum as well as secretion of water.
in the olfactory mucosa. Therefore, because of the high The viscoelastic properties and fluidity of mucus are deter-
metabolic capacity for xenobiotics, the olfactory mucosa mined by interactions between mucous components, ion
can show significant tissue damage following even content, and pH.
systemic administration of toxicants that require metabolic The composition of nasal secretions can change
activation. Examples of this are discussed in Sec. V.E. dramatically with inflammation, disease, or toxicant expos-
ure. For example, inflammation, with the transient influx
C. Secretions of the Olfactory Mucosa of neutrophils into the mucosa, results in large increases in
the secretion of stored mucosubstances from the respira-
Acinar cells of Bowman’s glands and, in some species, tory mucosa (Harkema et al., 1988). Chronic bronchitis and
sustentacular cells secrete acidic, sulfated, or neutral asthma also increase the quantity of nasal mucus; cigarette
mucopolysaccharides, the percentage of each depending smoke and formaldehyde alter the surface viscoelasticity
on the species and specific physiological, neuronal, and of the mucous layer, and cigarette smoke, antigens, and
environmental conditions. The distribution of different diethyl ether cause leakage of macromolecules and small
carbohydrate residues in the mucociliary complex is not ions into the mucous layer (Morgan et al., 1986). The lev-
homogeneous (Getchell et al., 1993b). Human nasal secre- els of growth factors and neuropeptides in the mucus also
tions contain immune factors including IgA, IgM, and IgG change in pathological conditions. For instance, the levels
(Kaliner, 1991). Secretory component and J chain have of substance P and vasoactive intestinal peptide in nasal
been localized to acinar and duct cells of Bowman’s secretions of patients with nasal allergy are significantly
glands, as well as the mucociliary apparatus in the human higher than in normal subjects (Chaen et al., 1993), and the
olfactory mucosa (Mellert et al., 1992). Other components levels of insulin-like growth factor I and its binding pro-
of mucus thought to play a defensive role include the teins in the mucus of olfactory epithelium are decreased in
antimicrobial proteins lysozyme and lactoferrin (Mellert et patients with certain neurodegenerative diseases (Federico
al., 1992; Mullol et al., 1992), enzymatic constituents, et al., 1999).
including aminopeptidases, endopeptidases, carboxypepti- The cilia on respiratory epithelial cells move mucus
dases, angiotensin-converting enzyme, peroxidase, and over the surface of the epithelium to produce mucociliary
kallikrein (Kaliner, 1991; Ohkubo et al., 1998), and clearance of environmental airway contaminants.
Olfactory Mucosa 55

However, the olfactory epithelium does not contain beating

cilia and therefore must rely on the movement and flow
created by the cilia in the respiratory epithelium for clear-
ance. As is the case with mucous composition, the efficacy
of ciliary beating can be altered by toxicant exposure or
disease. For example, beat frequency is reduced by cadmium
salts and acetaldehyde, and the amplitude of the beat is
reduced by dimethylamine (Morgan et al., 1986). The beat
frequency and proportion of epithelial area with normal
ciliary beat frequency are also decreased by oxygen rad-
icals (Min et al., 1999). Cigarette smoke can cause loss of
cilia, uncoordinated beating, and even reversal in the direc-
tion of beat (Iravani and Melville, 1974). The significance
of these alterations in clearance for given individuals can
vary widely. In humans, clearance rates have been Figure 3 Comparative anatomy of the nasal cavity. Shading
described as characteristic of a given individual, which represents the portion of the nasal lumen lined with olfactory
may vary from 1 to 20 mm/min (Proctor et al., 1978), and epithelium. Note the decrease in proportional surface area of
nasal ciliary beat frequency appears to be age independent olfactory tissue progressing phylogenetically from rat to monkey
(Jorissen et al., 1998). to human. Also note the parallel proportional decrease in relative
size of the olfactory bulbs. Olfactory bulb (OB); superior
The presence in nasal secretions of macromolecules
turbinate (ST); middle turbinate (MT); inferior turbinate (IT);
suggests that the protective function of secretions is not
hard palate (HP); nasopharynx (NP); nares (N); ethmoturbinate
simply related to clearance [which decreases with (ET); maxilloturbinate (MX); nasal turbinate (NT).
increased secretion (Proctor et al., 1978)], but includes
reactions such as bacterial destruction by lysozyme, proin-
flammatory peptide degradation by peptidases, viral inacti- than in the human, as can be seen in Figure 3 (Harkema,
vation by IgA interaction, and possibly metabolism of 1991). Increased infolding in the turbinates also results in
toxicants prior to tissue contact or absorption into systemic alteration in airflow patterns, and therefore in intranasal
circulation. The precise activity and in vivo function of deposition patterns. However, studies on airflow indicate
these macromolecules in mucous secretions have not been that the percentage of inspired air reaching the olfactory
well studied to date. However, transport, binding, and mucosa is roughly 15% in rat, monkey, and human (Hahn
clearance of xenobiotics occurring in the mucociliary et al., 1993; Jaillardon et al., 1992; Kimbell et al., 1993).
apparatus will influence deposition of these substrates in Because of the differences in relative proportion of olfac-
nasal tissues. Therefore, alterations in mucous constituents tory tissue, however, the percentage of inhaled dose
and ciliary function will alter metabolism as well. deposited in olfactory tissue may still be quite different
across these species. These differences are therefore
D. Comparative Aspects of Mucosal Composition important considerations in extrapolating data derived
from laboratory animal research to the human population.
The primary interspecies anatomical differences in Although the cytoarchitecture of the olfactory epithe-
olfactory mucosa result from differences in turbinate struc- lium is remarkably similar across mammalian species, the
ture and related proportion of the nasal cavity lined with capacity and localization of xenobiotic metabolism can be
olfactory mucosa. In general, the surface area to nasal markedly different. Thus, the relative activity of specific
cavity volume ratio reflects the reliance on olfaction of a P450 enzymes varies between rat and human with some
given species (Fig. 3). For example, the rat has a surface isoforms that show high activity in one species being
area to lumenal volume ratio of 3350 mm2/cm3; macaque apparently absent in the other (Dahl and Lewis, 1993;
monkey, 775 mm2/cm3; and human, 820 mm2/cm3 with Gervasi et al., 1989; Hadley and Dahl, 1983; Morris, 1997;
comparative lumenal volumes of 0.4 cm3 for rat and 25 Sheng et al., 2000). Epoxide hydrolase and glutathione S-
cm3 for human (Harkema, 1991). Increased surface area transferase (GST) both show greater activity in human
results from an increase in the complexity of turbinates in respiratory tissue than in rat tissue; however, NADPH-
the nasal cavity; generally the greatest difference occurs in cytochrome c reductase activity in human respiratory tis-
the number of olfactory turbinates. For example, in the rat, sue is only 25% that observed in rat tissue (Gervasi et al.,
the percentage of the nasal surface area covered by olfac- 1989). Other interspecies differences in metabolic activity
tory epithelium is nearly 50%, a much greater percentage will be discussed in more detail in Sec. IV. Notably,
56 Ding and Dahl

although activities of specific enzymes may show large The distribution of specific enzymes within the nasal
differences between species, no consistent differences cavity can also be different. For example, zonal distrib-
across specific enzyme families or even isozymes within a ution has been observed for the expression of microsomal
given family allow for reliable generalizations regarding epoxide hydrolase in rat olfactory mucosa. The enzyme is
the relative overall enzymatic activity across species. absent from most of the dorsal medial meatus where
Because biopsy samples of human olfactory mucosa are immunoreactivity of a GST has been found to be abundant
difficult to obtain, most human nasal enzyme activity to (Genter et al., 1995a). In contrast, expression of a sulfo-
date has been studied in respiratory mucosa. Generally, the transferase in mouse olfactory mucosa is localized to the
olfactory mucosa has a higher or equal level of activity for most dorsal and medial zone (Miyawaki et al., 1996).
xenobiotic substrates than does the respiratory mucosa. An
exception to this rule is aldehyde dehydrogenase. In rats,
nasal respiratory mucosa shows higher aldehyde dehydro- IV. IDENTITY, TISSUE-AND CELL-SELECTIVE
genase activity than does olfactory tissue (Bogdanffy et al., EXPRESSION, AND DEVELOPMENTAL
1998; Casanova-Schmitz et al., 1984), and olfactory tis- REGULATION OF NASAL
sue has very low immunoreactivity for this enzyme BIOTRANSFORMATION ENZYMES
(Bogdanffy et al., 1986).
The cells containing xenobiotic metabolic activity in the The dramatic capacity of mammalian nasal mucosa to
olfactory mucosa are relatively consistent across species. metabolize inhaled substances has only been recognized in
The primary localization for xenobiotic metabolizing the last two decades. Reports of alkaline phosphatase
enzymes is within the sustentacular, basal, and duct cells of localization in olfactory tissue date back to 1948 (Bourne,
the epithelium and within the acinar and duct cells of 1948), and the possibility that esterases present in the
Bowman’s glands in the lamina propria. Much like squa- olfactory apparatus of moths might play a role in metabo-
mous epithelial cells, olfactory receptor cells contain lizing olfactory signals was suggested in 1981 (Vogt and
enzymes having primarily a homeostatic function, such as Riddiford, 1981). Since the first reports that P450 activity
alkaline phosphatase (Bourne, 1948) and carbonic anhy- in rat nasal mucosa sometimes exceeded activity in liver
drase (Brown et al, 1984); however, xenobiotic-metabolizing when normalized to tissue protein content (Hadley and
enzymes have generally not been localized to these cells. Dahl, 1982), numerous laboratories have reported activity
Although the cell types identified in the previous para- in the nasal mucosa for families of xenobiotic-metabolizing
graph are consistent sites of enzyme localization across enzymes, including flavin-containing monooxygenases,
species, the specific distribution of a given enzyme within aldehyde dehydrogenases, alcohol dehydrogenase, car-
these cell types can vary across species. For example, the boxylesterases, epoxide hydrolases, UDP glucuronosyl-
cyanide-metabolizing enzyme rhodanese is found in the transferase, GST, and rhodanese (Dahl and Hadley, 1991).
acinar cells and duct cells of Bowman’s glands in bovine In addition, xenobiotic-metabolizing capacity has been
olfactory mucosa (Lewis et al., 1991). However, in the rat, demonstrated in olfactory and other nasal tissues from
rhodanese is localized to the sustentacular and basal cells a broad range of species, including Drosophila
rather than to Bowman’s glands. As will be discussed later, melanogaster (Wang et al., 1999), lobsters (Trapido-
this localization can be an important determinant of toxi- Rosenthal et al., 1990), rainbow trout (Starcevic and
cant-induced damage in different species and must be kept Zielinski, 1995), rabbits (Ding and Coon, 1988, 1990a;
in mind when generalizing from one species to another. Shehin-Johnson et al., 1995), rodents (Genter et al., 1995b;
Conversely, some enzymes show remarkable similarity Hadley and Dahl, 1982), dogs (Dahl et al., 1982), pigs
in distribution across species. For example, car- (Marini et al., 1998), sheep (Larsson et al., 1994), cows
boxylesterase localization by immunostaining is highly (Longo et al., 1997), and humans (Gervasi et al., 1989;
similar in the rat, dog, and human. However, the presence Getchell et al., 1993a; Gu et al., 2000; Lewis et al, 1991,
of inflammation in human respiratory tissue correlates 1994b).
with a marked reduction in immunoreactivity for the Rapid progress has been made in the identification and
enzyme, whereas metaplastic lesions in the tissue are asso- characterization of nasal biotransformation enzymes.
ciated with a total loss of staining (Lewis et al., 1994b). Many new enzymes have been identified since this subject
Such findings indicate that caution must be used in was recently reviewed (Dahl and Hadley, 1991; Ding and
extrapolating to the human population from clean labora- Coon, 1993). The majority of work was focused on mem-
tory studies because this extremely plastic tissue is vulner- bers of the P450 gene superfamily, but significant progress
able to toxicant and irritant-induced damage that can has also been made in the molecular identification of other
dramatically alter its enzymatic capacity. biotransformation enzymes. The biological model systems
Olfactory Mucosa 57

ranged from insects to fish and to humans. Although most major forms (Ding and Coon, 1990a; Genter et al., 1998;
of the xenobiotic-metabolizing enzymes localized in the Gu et al., 1998). Multiple genes are present in the CYP2A
nose are also found in other tissues, several enzymes have subfamily, which were named sequentially according to
been found to be uniquely or preferentially expressed in the time of discovery. The CYP2A genes expressed in the
the olfactory mucosa in a number of species. olfactory mucosa include CYP2A3 in rats, CYP2A5 in
mice, CYP2A6 and CYP2A13 in humans, and CYP2A10
A. Cytochrome P450 and CYP2A11 in rabbits (Koskela et al., 1999; Peng et al.,
1993; Su et al., 1996, 2000). There appears to be only a
The P450 gene superfamily encodes over 500 structurally single CYP2G gene in all mammalian animals studied;
similar monooxygenases (Nelson et al., 1996). All P450s thus they are all called CYP2G1 (Ding et al., 1991; Hua et
contain a heme prosthetic group ligated to a highly con- al., 1997; Nef et al., 1990). Originally, the rabbit CYP2As
served cysteine residue in the carboxyl terminal portion of were called P450 NMa, which included both CYP2A10
the proteins. In a single species, e.g., the humans, the total and 2A11 when purified from nasal microsomes (Ding and
number of P450 genes can be more than 50, and individual Coon, 1988; Peng et al., 1993); similarly, the rabbit
genes are expressed more or less in tissue- and cell-selective CYP2G1 was called P450 NMb (Ding and Coon, 1988),
fashions. Within a cell, the majority of P450s are located in and the rat CYP2G1 was called P450 olf1 (Nef et al.,
the endoplasmic reticulum (microsomal fraction), while 1989). In humans there may be two copies of the CYP2G
some are specifically located in the mitochondria. gene, but both contain loss-of-function mutations in the
The substrates for microsomal P450s include physio- majority of individuals, and a functional cDNA has not
logically important substances such as steroid hormones, been identified to date (Sheng et al., 2000).
eicosanoids, and retinoids, and xenobiotics such as drugs, In addition to the P450 forms in gene families 1–4,
procarcinogens, antibiotics, organic solvents, anesthetics, which are often referred to as the xenobiotic-metabolizing
pesticides, and odorants. P450-catalyzed biotransforma- P450s, there are also several microsomal P450 gene fam-
tions lead to the formation of more polar compounds that ilies specifically involved in steroid biosynthetic pathways
are more readily excreted directly or after conjugation with in the endocrine and reproductive organs or bile acid
water-soluble agents such as glucuronic acid and GSH metabolism in liver, such as CYP7, 17, 19, 21, and 51
(Porter and Coon, 1991). (Nelson et al., 1996). The expression of these genes in the
P450 and NADPH-cytochrome P450 reductase (CPR), olfactory mucosa has not been examined.
a flavoprotein required for microsomal P450-catalyzed Several olfactory mucosal P450s are specifically or
monooxygenase reactions, have been found in relatively preferentially expressed in this tissue. For example,
high concentration in the olfactory mucosa of rodents, rab- CYP2G1 is only expressed in the olfactory mucosa (Ding
bits, cows, dogs, pigs, monkeys (Dahl and Hadley, 1991; and Coon, 1990a; Hua et al., 1997; Nef et al., 1989) and,
Ding and Coon, 1993; Hua et al., 1997; Longo et al., 1997; at much lower levels, in the vomeronasal organ (Gu et al.,
Marini et al., 1998), and humans (Getchell et al., 1993a; Su 1999). Several CYP2As are expressed in olfactory mucosa
et al., 1996). Both P450 and CPR have also been identified at much higher levels than in other tissues (Ding and Coon,
in the olfactory organ of D. melanogaster (Hovemann et 1990a; Su et al., 1996). Preferential expression of a P450
al., 1997; Wang et al., 1999). On a per mg microsomal pro- in the olfactory organ was also found in Drosophila (Wang
tein basis, the level of total microsomal P450 in olfactory et al., 1999). Although the specific roles of these tissue-
mucosa is second only to liver among all tissues examined selective P450s have not been identified, their unique or
in rodents and rabbits; the level of CPR in olfactory preferential presence in the olfactory mucosa strongly sug-
mucosa microsomes is even higher than in liver (Ding et gests functional importance in the chemosensory organ.
al., 1986; Reed et al., 1986). The evolutionarily conserved Immunohistochemical studies of several olfactory
presence of the P450 enzymes supports their functional mucosa microsomal P450s, including CYP1A, 2A, 2B,
importance in olfaction. 2G, and 4B, indicated that they are expressed in nonneu-
More than 10 different P450s have been identified in ronal cells, particularly in the sustentacular cells in the
mammalian olfactory mucosa, including members of the epithelium and in the Bowman’s glands in the submucosa
CYP1A, 2A, 2B, 2C, 2E, 2G, 2J, 3A, 4A, and 4B subfam- (Adam et al., 1991; Chen et al., 1992; Getchell et al.,
ilies (Dahl and Hadley, 1991; Deshpande et al., 1999; Ding 1993a; Thornton-Manning et al., 1997; Voigt et al., 1985,
and Coon, 1993; Gu et al., 1998; Zhang et al., 1997). 1993; Zupko et al., 1991). Distribution of CPR was found
Additional forms are expected to be found since several to resemble that of the P450s (Adam et al., 1991; Baron
subfamilies have not been examined, such as CYP2D, 2F, et al., 1986; Voigt et al., 1985); however, CPR expression in
and 4F. Of these, CYP1A2, CYP2A, and CYP2G1 are the olfactory receptor neurons (ORNs) has also been reported
58 Ding and Dahl

(Verma et al., 1993; Voigt et al., 1985). The lack of known 1995a; Rogers et al., 1999; Starcevic and Zielinski, 1995).
microsomal P450 expression in the neuronal cells is also In rats, these include rGSTA3, rGSTA4, rGSTM1,
supported by toxicological studies implicating the rGSTM2, rGSTM6, and rGSTP1 (Banger et al., 1993,
Bowman’s glands and the supporting cells as the initial tar- 1996; Ben-Arie et al., 1993). In humans, GSTA and GSTP,
gets following chemical treatment (Brittebo, 1997). The but not GSTM, were detected in the olfactory mucosa by
localization of P450s to the mucus-producing cells in the immunohistochemistry (Krishna et al., 1995a). In rats and
Bowman’s glands and the detection of P450 immuno- cows, high GST activity was found in olfactory mucosa
reactivity in the mucociliary complex at the epithelial surface toward model substrates and odorants (Aceto et al., 1993;
led to suggestions that they may be secreted to the mucous Ben-Arie et al., 1993). An olfactory tissue-specific GST has
layer where they may directly act on inhaled chemicals not been found in mammals, although one has been found
(Adam et al., 1991; Chen et al., 1992). in the sphinx moth Manduca sexta (Rogers et al., 1999). In
Little is known about the molecular mechanisms that rats, GSTA and GSTM immunoreactivity was detected in
regulate the tissue- and cell-selective expression of P450s sustentacular cells and Bowman’s glands in the olfactory
and other biotransformation enzymes in the olfactory mucosa. In humans, GSTA immunoreactivity was detected
mucosa. Two recent in vitro studies identified nuclear fac- mainly in the acinar cells of the Bowman’s glands, as well
tor I–like cis-acting elements in the proximal promoter as in the supranuclear region of supporting cells, but GSTP
region of both CYP2A3 and CYP1A2 genes (Zhang and immunoreactivity was detected only in the supporting cells
Ding, 1998; Zhang et al., 2000). These highly conserved in the olfactory mucosa (Krishna et al., 1994, 1995a).
DNA sequences, which are critical for transcriptional Olfactory mucosal GSTA and GSTM immunoreactivity
activity of the cognate P450 promoters in vitro, appear to was detectable at E16 in rats and increased postnatally,
interact with olfactory mucosa–restricted nuclear proteins. with peak expression around P11 (Krishna et al., 1994).
Identification of these potentially novel tissue-selective The postnatal increases in the levels of GSTA and GSTM
transcription factors will be important for understanding isoforms were confirmed by immunoblot analysis of olfac-
the regulation of these and other genes preferentially tory S9 fractions (Banger et al., 1996). However, cytosolic
expressed in the olfactory mucosa. GST activity measured with 1-chloro-2,4-dinitrobenzene
The developmental expression of P450s and CPR in as a substrate was constant in rat olfactory mucosa
olfactory mucosa has also been examined. In rabbits, between P3 and P84, while microsomal GST activity
CYP2G1 was detected at 2 days before birth (Ding et al., remained low until P21 and then increased to reach adult
1992). In rats, CYP2G1 expression was detected at E20, levels at about P60 (Banger et al., 1996). In humans, both
which was suggested to coincide with the appearance of P450 2A and GST immunoreactivities were decreased in
Bowman’s glands (Margalit and Lancet, 1993). Prenatal older adults (Getchell et al., 1993a; Krishna et al., 1995a).
expression of several P450s and CPR has also been found UDP glucuronosyltransferases (UDPGTs), also named
in humans (Gu et al., 2000). The earlier onset of P450 UDP glycosyltransferases (UGTs), catalyze the conjuga-
expression in olfactory mucosa than in other tissues may tion of UDP glucuronic acid with a variety of substrates
indicate a functional significance in the perinatal period (Mackenzie et al., 1997). In mammals, the UDPGTs are
when olfactory ability is important for the survival of the found in microsomal fractions and belong to two different
newborn. gene families, each having multiple genes (Mackenzie et
al., 1997). An olfactory mucosa-specific UDPGT has been
B. Other Enzymes identified in rats, cows, and humans, named UGT2A1,
which is active toward numerous compounds, including
GSTs catalyze the conjugation of GSH with numerous many odorants (Jedlitschky et al., 1999; Lazard et al.,
electrophilic substrates, including reactive intermediates 1990, 1991; Mackenzie et al., 1997). A tissue-specific
formed in P450-catalyzed reactions, which decrease their UGT (DmeUgt35a) has also been identified in the olfac-
reactivity with proteins and other cellular macromolecules tory organ of D. melanogaster (Wang et al., 1999). Multiple
(Armstrong, 1997; Eaton and Bammler, 1999), as well as UDPGTs are believed to be expressed in mammalian
unaltered odorants (Ben-Arie et al., 1993). Most GSTs are olfactory mucosa (Marini et al., 1998), but the specific
located in the cytosol, although some have also been found enzymes have not been characterized, except for UGT2A1.
in microsomes and mitochondria (Eaton and Bammler, Sulfotransferases (ST), which include phenol ST (PST),
1999). At least five cytosolic GST gene families are known hydroxysteroid ST (HSST), and, in plants, flavonol
in humans. Multiple GSTs have been detected in the olfac- ST (FST) gene families, catalyze the transfer of a
tory mucosa in a number of species (Aceto et al., 1993; sulfonate group from 3-phosphoadenosine-5-phospho-
Banger et al., 1993; Ben-Arie et al., 1993; Krishna et al., sulfate to both endogenous and xenobiotic compounds
Olfactory Mucosa 59

(Weinshilboum et al., 1997). Mouse nasal cytosol had high Table 1 Instances in Which Nasal Metabolism Probably
activity for a number of phenolic aromatic odorants Results in Detoxication
(Miyawaki et al., 1996; Tamura et al., 1997). A PST cDNA
Substrates Enzyme activities
has been isolated from a mouse olfactory cDNA library
(Matsui et al., 1998). PST proteins were also detected in Nitropyrenes Oxidases and hydroxylases
the cytosol of rat and mouse nasal tissues using an anti- 2,6-dichlorobenzonitrile Hydroxylase
body to rat liver PSTg (Miyawaki et al., 1996). Mouse Coumarin 7-hydroxylase
PSTG immunoreactivity, which is detectable prenatally, is Cocaine Demethylation
localized mainly in the sustentacular cells (Miyawaki et Alkoxycoumarins Dealkylation
Lactones Carboxylesterases
al., 1996).
Styrene oxide Epoxide hydrolases
Naphthol Transferases
V. FUNCTIONS OF NASAL Chlorodinitrobenzene Transferases
Cyanide S-transferases (rhodanese)
BIOTRANSFORMATION ENZYMES
Nicotine Demethylases
Formaldehyde Aldehyde dehydrogenases
Nasal xenobiotic metabolism likely serves multiple func-
tions. Four possibilities, discussed in more detail below, Source: Modified from Dahl and Hadley, 1991.
are (1) detoxication of inhaled and systemically derived
xenobiotics, (2) protection of other tissues, such as the olfactory mucosa (Dahl and Hadley, 1983). Although
lung and CNS, from inhaled toxicants, (3) modification of inhalation of 15 ppm formaldehyde produces squamous
inhaled odorants, including the special case of steroids as metaplasia and squamous cell carcinomas of the respira-
reproductive stimuli, and (4) modulation of endogenous tory mucosa of the nasal cavity (Morgan and Monticello,
signaling molecules. In addition, the roles of nasal bio- 1990), the presence of formaldehyde dehydrogenase can
transformation enzymes in the metabolic activation and increase clearance and therefore protect tissues from meta-
toxicity of inhaled or systemically derived xenobiotics are bolically produced formaldehyde. The activity of
also considered. formaldehyde dehydrogenase in the olfactory mucosa is
approximately double that of the respiratory mucosa
A. Detoxication of Inhaled Toxicants (Bogdanffy, 1990). One would therefore predict that the
olfactory mucosa will be less sensitive to formaldehyde
The nose is the portal of entry for inhaled chemicals and, toxicity. Indeed, olfactory mucosal lesions resulting from
as such, is continually exposed to toxic insults. Therefore, formaldehyde exposure are less common than respiratory
one function of xenobiotic metabolism could be detoxica- mucosal damage. Nevertheless, the regional distribution of
tion of inhaled toxicants. Because of the small mass of damage will be influenced by the relationships among air-
nasal mucosa, even though the activity of nasal enzymes is flow, deposition, and chemical solubility and reactivity, as
high, the total capacity to metabolize inhaled substrates is well as enzyme localization.
probably not high enough in most cases to provide The enzyme rhodanese metabolizes cyanide to the less
systemic protection from inhaled toxicants. (Exceptions toxic metabolite thiocyanate. The activity of rhodanese in
are discussed in the next section.) The protective function human nasal respiratory mucosa is high and probably serves
of nasal metabolism, therefore, is probably more a form of to protect against toxic effects of inhaled cyanide (Lewis et
local tissue protection than protection of downstream tis- al., 1991). There are many diverse environmental sources
sues such as lung. It can be said for the vast majority of for inhaled cyanide such as combustion products from syn-
lipophilic compounds that would normally build up in the thetic materials and cooking of cyanogenic fruits such as
nasal tissue that combined P450 and phase II metabolism, apricots and cherries. However, it is possible that rhodanese
or metabolism by other routes, decreases toxicity either by serves an additional protective function in secondarily
increasing solubility and subsequent clearance or by other metabolizing cyanide produced from the P450 metabolism
chemical modification to less toxic forms. Examples of of inhaled organonitrile compounds such as benzylnitrile
substrates for which this is the case are given in Table 1. and acetonitrile (Dahl and Waruszewski, 1990).
In some instances, nasal metabolic systems may work With respect to the secondary detoxication of toxic
in tandem to provide local protection. For example, a wide metabolites, the cellular localization of specific enzymes
range of inhaled substrates, including methamphetamine, may be important when generalizing across species. The
cocaine, nicotine, diesel soot extracts, and pyrilamine, are organonitrile ,-iminodipropionitrile (IDPN) is toxic to
metabolized to formaldehyde by P450 isozymes in the the acinar cells of Bowman’s glands following systemic
60 Ding and Dahl

administration in rats (Genter et al., 1992). P450 metab- enzyme families. The capacity for nasal metabolism of some
olism of IDPN would yield cyanide. The cells of Bowman’s P450 substrates is considerably lower: 0.1–5 ppm for
glands in the rat contain several isoforms of P450 (Dahl p-nitroanisole and 0.1–3 ppm for aniline. At these levels of
and Hadley, 1991); however, they do not contain rhodanese activity, significant systemic protection from inhaled toxic
(Lewis et al., 1992b). It is therefore likely that the toxicity substrates would probably not result from nasal metabolism.
of IDPN in these cells is caused by a buildup of the
metabolite cyanide. Because the cellular distribution of 2. Central Nervous System
rhodanese differs across species, the target cells for IDPN
Xenobiotic metabolism in the olfactory epithelium as well
toxicity may also differ.
as in the olfactory bulbs may be a component of a “nose-
It should be remembered that nasal xenobiotic enzymes
brain barrier.” The olfactory epithelium has a unique anatomy
act not only on inhaled substrates, but on substrates in the
wherein a single receptor cell contacts the external
systemic circulation as well, as evidenced by metabolite-
environment in the nasal lumen and projects directly to its
induced toxicity to the olfactory mucosa seen following
synapse within the CNS in the glomeruli of the olfactory
intravenous administration of toxicants such as 4-(methyl-
bulbs. These cells, then, provide direct access for inhalants
nitrosamino)-1-(3-pyridyl)-1-butanone (NNK) (Belinsky
to the CNS. Several studies have demonstrated that a vari-
et al., 1990), 3-methylindole (Turk et al., 1986), and aceta-
ety of materials instilled or surgically implanted into the
minophen (Jeffery and Haschek, 1988). Teleologically, this
nasal cavity can be transported to the olfactory bulbs
metabolism of systemic compounds could serve to reduce
(Henriksson and Tjälve, 2000; Larsson and Tjälve, 2000;
stimulation of olfactory receptors by circulating odorants,
McLean et al., 1989; Schultz and Gebhardt, 1934; Shipley,
thereby eliminating possible interference with or masking
1985; Tomlinson and Esiri, 1983). Generally, these studies
of inhaled odorants. In addition, such metabolism could
have used concentrations of material far in excess of those
serve to protect this important sensory tissue from damage
encountered environmentally. Therefore, the importance of
induced by circulating toxicants. While, in the cases noted
this phenomenon for inhalation of environmentally rele-
above, toxic metabolites are formed and the metabolic
vant concentrations of toxicants is not yet clear. A more in-
activity of the nasal tissue results in damage rather than
depth discussion of this topic can be found in Chapter 26
protection, this is almost certainly not the general case, as
and in the reviews by Lewis et al. (1994a) and Tjälve and
discussed below.
Henriksson (1999). The factors involved in a nose-brain
barrier that might protect the brain from toxicant exposure
have not yet been elucidated. However, nasal xenobiotic
B. Protection of Other Tissues from Inhaled
metabolism is likely to be involved. In addition, mucocil-
Toxicants
iary clearance, immune responses in the olfactory or other
1. Lung nasal mucosa, tight junctions between epithelial cells, and
the rapid death of epithelial cells following toxicant expos-
Based on activity, most nasal enzymes probably have little
ure are also likely to play a role (Lewis et al., 1994a).
effect on reducing concentrations of toxicants entering sys-
temic circulation unless inhaled concentrations are very
low. Two exceptions to this might be toxic substrates of C. Modification of Olfactory Stimuli
carboxylesterases and cyanogenic compounds detoxified
1. Odorants
by the cyanide-metabolizing enzyme rhodanese (Dahl,
1988; Lewis et al., 1991). Across several species, the A third possible function for nasal metabolism is either
capacity of nasal carboxylesterases is sufficient to detoxify activation of inhaled nonodorants to odorants or, conversely,
inhaled concentrations of esters such as ethyleneglycol clearance of odorants from the olfactory receptor cells
monomethyl ether acetate in the 1000–3000 ppm range to allow reactivation of receptors (Dahl, 1988; Getchell
(Dahl, 1988), a concentration in excess of occupational and Getchell, 1977). Although most biotransformation
exposure limits. However, in many cases, such as with enzymes are located in the nonneuronal cells, the
dibasic esters, the metabolites of esters are themselves lipophilic substrates can quickly diffuse to all cells in the
toxic to olfactory tissues (Bogdanffy, 1990). Likewise, olfactory mucosa. This phenomenon of receptor reactiva-
nasal rhodanese activity is sufficient to detoxify inhaled tion has been demonstrated in lobsters (Trapido-Rosenthal
concentrations of hydrogen cyanide as high as 2800 ppm et al., 1990) and in silk moths (Vogt and Riddiford, 1981),
in the rat (Lewis et al., 1991). but as yet not in mammalian species. However, this
This capacity to significantly alter systemic toxicant possibility has been suggested for mammalian cells where
exposure through nasal metabolism does not hold for all nasal-specific UDPGTs have been shown to have a greater
Olfactory Mucosa 61

substrate specificity for odorant molecules than do the olites when compared to those produced by other known
UDPGTs isolated from liver (Lazard et al., 1991). P450s (Ding and Coon, 1990a, 1994; Hua et al., 1997).
Odorant metabolites may contribute to potency and The vomeronasal organ is also capable of metabolizing sex
odor quality. Many odorant metabolites are more water- steroids (Gu et al., 1999). Accumulation of sex steroids
soluble than the parent odorants, and they may reach very and other endogenous or exogenous compounds that are
high concentrations in the mucus bathing ORNs (Dahl, normally removed by P450 metabolism could affect signal
1988; Price, 1984). If olfactory stimulation includes sum- transduction by competing for receptors. To that end,
mation of signals from both parent odorant and its metab- Rosenblum et al. (1991) reported that receptor binding of
olites (Kashiwayanagi et al., 1987; Price, 1984), such 17,20-dihydroxy-4-pregnen-3-one to goldfish olfactory
metabolites could be important to the sensitivity, intensity, mucosa is competitively inhibited by progesterone and
and quality perception of an odor. Thus, it has been other sex steroids. Androstenone is metabolized by the
hypothesized that odor quality and intensity may reflect nasal mucosa in pigs (Gennings et al., 1974) and is a com-
effects of the odorant and its metabolites on ORNs (Dahl, petitive inhibitor of steroid metabolism by CYP2G1 (Ding
1988; Price, 1984, 1986). and Coon, 1994), which suggests that it may be metab-
Lipophilic odorants partition favorably into the mem- olized by this or other P450 enzymes. The role of nasal
branous structure of the neuroepithelium. Their accumula- metabolism in reproductive function has received very lit-
tion may adversely affect many aspects of cellular function. tle attention to date, but the discovery of olfactory
They may saturate the odorant clearance mechanism, mucosa–specific P450s that metabolize sex steroids is likely
disturb mitochondrial energy production, and suppress or to make this an important area of research in the future.
sensitize local immune systems. They may also change the
biophysical properties of the plasma membrane and thus D. Modulation of Endogenous Signaling Molecules
the functional capacity of ion channels and other signal
transduction components. Biotransformation reactions that The P450 isoforms identified in the olfactory mucosa are
convert these lipophilic compounds into more water-solu- all active in the metabolism of endogenous compounds,
ble metabolites may thus be indispensable for maintaining although they are also involved in metabolizing foreign
the homeostasis of the chemosensory tissue. chemicals. For example, CYP1B, 2A, 2B, 2G, and 3A are
active in the hydroxylation of sex steroids (Ding and Coon,
1994; Hayes et al., 1996; Liu et al., 1996; Waxman et al.,
2. Steroids
1991), CYP1A, 2B, 2C, 2E, 2J, and 4A are active in the
Steroids are likely to represent a special case of metabolic hydroxylation or epoxygenation of arachidonic acid
modulation of olfactory stimuli. Inhaled steroids can serve (Laethem et al., 1992; Luo et al., 1998; Scarborough et al.,
as primary olfactory cues in the regulation of reproductive 1999), CYP1A2 and CYP2J4 are active in converting
function in a number of species, as well as modulators of retinals to retinoic acids (RAs), and CYP1A and 2B are
olfactory function. Androstenone is a steroid found in the active in the hydroxylation of RA (Roberts et al., 1992;
urine and saliva of pigs and humans as well as in human Zhang et al., 1998). The consequences of microsomal P450-
sweat. In pigs, androstenone excreted by boars has been catalyzed metabolism of endogenous compounds are usually
shown to initiate mating behavior in estrus sows with intact inactivation of the bioactive substance, as with hydroxyla-
olfactory function (Beauchamp et al., 1976). Sensitivity to tion of testosterone and RAs. However, the epoxygenated
the odor of androstenone varies widely in humans (Dorries or hydroxylated products of arachidonic acid have been
et al., 1989). Interaction of exogenous steroids with the implicated in many biological processes, such as regula-
olfactory system has also been demonstrated through modi- tion of vascular tone, ion transport, calcium release from
fication of serum testosterone, testicular size, and spermato- endoplasmic reticulum, and modification of biophysical
genesis in rhesus monkeys by intranasal administration of properties of plasma membrane (Capdevila et al., 1992;
estradiol and progesterone (Anand-Kumar et al., 1980). Makita et al., 1996). It is believed (Nebert, 1990, 1991)
It is likely that the ability of the olfactory mucosa to that the xenobiotic-metabolizing P450s regulate steady-
metabolize steroids will influence responses to exogenous state levels of endogenous compounds important for
steroids. Mammalian olfactory mucosa has very high growth, homeostasis, differentiation, and neuroendocrine
activities in the metabolism of all three major sex steroids functions.
(Brittebo and Rafter, 1984; Brittebo, 1985; Ding and P450-catalyzed formation of arachidonic acid epoxide
Coon, 1990a, 1994; Hua et al., 1997, Longo et al., 1997; and hydroxides can regulate vascular tone and thus rate of
Marini et al., 1998). The olfactory mucosa–specific P450 blood flow (Capdevila et al., 1992; Makita et al., 1996).
2G1 metabolizes sex steroids to unique patterns of metab- Decreased ability to produce these regulatory molecules
62 Ding and Dahl

may lead to congestion and restrictions in air flow in the Table 2 Instances in Which Nasal Metabolism Probably
nasal cavity, which could affect threshold sensitivity in Results in Activation
odor detection. On the other hand, accumulation of arachi- Substrates Enzyme activities
donic acid may lead to increased production of
leukotrienes and other mediators through the lipoxygenase 2,6-dichlorobenzonitrile Epoxygenase
pathway and potentially induce airway hypersensitivity Coumarin Epoxygenase
(Pinto et al., 1997). Ferrocene Oxidases
ORNs are one of the few vertebrate neuronal popula- Benzo(a)pyrene Oxidases
Hexamethylphosphoramide Demethylases
tions that undergo turnover and replacement throughout
Diethylnitrosamine Deethylases
the life of an animal and following injury (Goldstein et al.,
Organonitriles Oxidases
1998; Graziadei and Monti-Graziadei, 1983) (see Chapter Phenacetin Oxidases
5). Such remarkable regenerative capacity may be at least Esters Carboxylesterases
partly related to the presence of the highly active biotrans- Acetaminophen Oxidases
formation enzymes, which control the availability and Trifluoromethylpyridine N-oxidases
level of various endogenous bioactive substances capable
Source: Modified from Dahl and Hadley, 1991.
of regulating growth and differentiation in the target tissue.
The relatively high efficiency and broad substrate speci-
ficity of the P450 enzymes toward steroid hormones (Ding from 2,6-dichlorobenzonitrile or coumarin (Ding et al.,
and Coon, 1994) and retinoids (Roberts et al., 1992) 1996; Zhuo et al., 1999), can usually be efficiently removed
suggest that these compounds may accumulate in olfactory by phase II enzymes such as GST. However, when the dose
mucosa when the P450s are inhibited or downregulated. is high or when the phase II enzymes are compromised due
RAs have powerful differentiation-promoting effects to chemical inhibition or genetic deficiency, the reactive
(Chambon, 1996) and induce apoptosis (e.g., Josefsen et metabolites would accumulate and cause cytotoxicity in the
al., 1999). RA receptors have been detected in mouse olfac- olfactory mucosa. In cases where a reactive intermediate
tory mucosa (Zhang, 1999). A role of RA in ORN differen- with relatively long half-life is generated, such as the
tiation has been reported (Whitesides et al., 1998), and RA benzo(a)pyrene epoxides (Dahl et al., 1985), it may be
has also been shown to regulate neurogenesis in adult-derived transported to nearby organs, such as the pharynx, the
neural stem cell cultures (Takahashi et al., 1999). RAs are esophagus, the anterior nasal cavity, and the olfactory bulb,
degraded in target tissues by microsomal P450s (Duester, where local biotransformation activities are much lower
1996). RA inactivation catalyzed by an embryonic P450 compared to the olfactory mucosa, and potentially cause
isoform (P450RA) has been found to result in RA hyposen- toxicity (Dahl et al., 1985; Ghantous et al., 1990).
sitivity in cultured cells (Fujii et al., 1997). Numerous compounds, such as ferrocene (Sun et al.,
Olfactory mucosa is also a known target tissue for 1991), 3-trifluoromethylpyridine (Gaskell, 1990), aceta-
steroid hormone action (Balboni, 1967; Balboni and minophen (Genter et al., 1998; Jeffery and Haschek, 1988),
Vannelli, 1982; Fong et al., 1999; Vannelli and Balboni, NNK (Belinsky et al., 1990), 2,6-dicholorobenzonitrile
1982). In male rats, olfactory mucosa morphology is (Brittebo, 1997), and coumarin (Gu et al., 1997) are metabo-
altered by castration, and testosterone replacement coun- lized to toxicants that produce necrosis of the olfactory epithe-
teracts these alterations (Balboni, 1967). In addition, cortico- lium. This toxicity can occur following not only inhalation,
steroids may regulate olfactory secretion by modulating but systemic administration as well. A more detailed discus-
Na,K-ATPase (Fong et al., 1999). Thus, prolonged sion of this subject can be found in Chapter 26.
accumulation of these endogenous compounds may lead to Often, the relative toxicity of a compound in different
changes in olfactory mucosa structure, cell biology, and species or in tissues within a given species is affected by
functional capacity. levels of activating or detoxicating enzymes. While this
probably holds for nasal toxicants as well, the relationships
E. Metabolic Activation and Xenobiotic Toxicity can be complex. For example, although activity levels in
in the Nasal Mucosa hamster nasal tissues for many enzymes known to activate
toxicants are higher than those in rats, hamsters nonethe-
The powerful biotransformation enzymes, particularly the less are less susceptible than rats to the toxic effects
P450 enzymes, generate reactive intermediates from induced by metabolites of 3-methylfuran and N-nitrosodi-
inhaled or systemically derived xenobiotic substrates, ethylamine (Dahl and Hadley, 1991). On the other hand, a
which could lead to toxicity (Table 2). The activated dose of 100 mg/kg of dichlobenil is needed to cause olfac-
metabolites, such as the proposed epoxide intermediates tory toxicity in the rat, whereas toxicities are observed in
Olfactory Mucosa 63

the mouse at a dose of 12 mg/kg (Brandt et al., 1990; least some instances, such an alteration has been demon-
Genter et al., 1996). There may be several reasons for such strated to subsequently alter toxicity as well. Induction of
apparent discrepancies. When compounds are admin- P450 activity by administration of -naphthoflavone
istered systemically, the capacity for other organs to clear decreased the severity of 3-methylindole–induced olfac-
the compound must be considered, thereby reducing the tory lesions (Turk et al., 1986), possibly as a result of lower
concentration reaching the nose. An example is the finding blood levels due to enhanced liver metabolism.
that although rat and mouse olfactory P450s are equally Conversely, treatment with dexamethasone potentiates the
active in metabolic activation of coumarin, rats are much 3-methylindole olfactory toxicity, which could be partly
more sensitive to the nasal toxicity of coumarin than mice due to the inducing action of dexamethasone on the P450
because of a lower hepatic clearance of the parent com- responsible for metabolic bioactivation of 3-methylindole
pound (Zhuo et al., 1999). When a toxicant is inhaled, dif- in the olfactory mucosa (Kratskin et al., 1999). In addition,
ferences in nasal airflow patterns, mucociliary clearance, changes in endogenous steroid hormones may modify the
or epithelial status may also affect the toxicity. biotransformation capacity of the olfactory mucosa, as
Nasal cancers are relatively uncommon in humans, demonstrated by the effects of castration on nasal metabo-
although in certain populations, notably Chinese males, the lism of testosterone. Castrated male rats have a reduced
rate of occurrence is quite high (Tricker and Preussman, ability to metabolize testosterone, while testosterone
1991). The occurrence of nasal tumors in laboratory ani- replacement restores metabolic capacity for the steroid
mals exposed by inhalation to toxic materials, on the other (Lupo et al., 1986). However, little else is known about this
hand, is a common finding. Often, the toxic materials are potentially very interesting subject.
procarcinogens requiring metabolic activation, suggesting As the following section will detail, many common
that differences in nasal xenobiotic metabolism between environmental exposure scenarios can result in alterations
humans and laboratory animals may underlie the observed of nasal enzymatic activity, thereby enhancing individual
differences in nasal tumor formation. Such differences have variations in responses to toxicant exposures. These alter-
been predictive among laboratory animal species. Thus, ations can result from either direct inhalation or systemic
inhalation of the procarcinogen benzo(a)pyrene results in exposure to chemicals that induce or inhibit nasal enzymes
nasal tumors in Syrian hamsters, but not in other species or from toxicant insults that alter the histology of the tis-
such as rats (Thyssen et al., 1981). The capacity for nasal sue. For example, many toxicants cause exfoliation of the
metabolism of benzo(a)pyrene in hamster is ~400 olfactory epithelium and a concomitant loss of metabolic
pmol/mg/min (Dahl et al., 1985), whereas in rats, the capacity from those lost cells. Enzymatic expression also
capacity to metabolize benzo(a)pyrene is only ~20 appears sensitive to hyperplastic or metaplastic alterations
pmol/mg/min (Bond, 1983). Although such relationships in the epithelium that can result from either toxicant
are compelling as explanations of toxicity, as in the case of exposures, infections, or inflammatory processes.
noncarcinogenic responses, relative metabolic capacity is Consideration of a patient’s exposure history may there-
not always sufficient to explain differences in toxicity. The fore be helpful in diagnosing what appears to be an atyp-
nasal carcinogen NNK produces DNA adducts in nasal tis- ical response to a subsequent toxicant exposure.
sue via the reactive -hydroxylated N-nitrosamine metab-
olites. Although metabolic capacity of the tissues might A. Enzyme Induction
lead to the prediction that the olfactory mucosa would pro-
duce comparatively more adducts than the respiratory Early reports, primarily from studies on rat nasal tissue,
mucosa, more adducts were actually found in the respirato- indicated that nasal P450s were relatively refractory to
ry mucosa (Belinsky et al., 1990). Again, other factors, per- induction by a wide range of inducers effective for hepatic
haps in this case route of exposure and DNA repair rates, enzymes. Either no induction or mild induction of rat nasal
must be taken into account to explain or predict toxicity. P450 activity was observed following treatment with the
classic inducers: phenobarbital, benzo(a)pyrene, 2,3,7,8-
tetrachlorodibenzo-p-dioxin, or 3-methylcholanthrene (3-
VI. MODIFICATION OF OLFACTORY MC) (Baron et al., 1988; Bond, 1983; Hadley and Dahl,
XENOBIOTIC METABOLISM 1982; Longo et al., 1988). Although one study reported the
induction of mouse nasal P450 activity by phenobarbital
As is the case with hepatic xenobiotic-metabolizing (Brittebo, 1982), the apparently increased activity may
enzymes, nasal enzymes are susceptible to modification in have resulted from induction of phase II enzymes
their levels of activity. Specific chemical inhibitors, and in occurring downstream from the P450 breakdown step, as
some cases inducers, can alter nasal metabolic capacity. In at only increased 14CO2 production was reported (Dahl and
64 Ding and Dahl

Hadley, 1991). Induction of phase II enzymes would be et al., 1999). The tissue-differential inducibility of some,
consistent with reports of induction of these enzymes in rat but not all, P450 enzymes may be related to the unique (yet
nasal tissue by phenobarbital (Guengerich et al., 1982; still unknown) function of each P450 enzyme and the need
Longo et al., 1988). The phase II enzyme UDPGT is also for the olfactory mucosa to maintain certain enzymes at a
induced by both Arochlor 1254 and 3-MC (Bond, 1983; relatively constant level. Alternatively, it is possible that
Longo et al., 1988). Nevertheless, not all phase II enzymes nasal biotransformation enzymes may respond preferen-
are readily inducible in the olfactory mucosa. Olfactory tially to inhaled odorants. For example, carboxylesterase is
GSTs were not induced in rats by trans-stilbene oxide, induced in the olfactory mucosa following inhalation
which caused a 2-fold induction in the liver (Banger et al., exposure to the common solvent pyridine (Nikula et al.,
1996); only a marginal induction (1.3-fold) by PB was 1995), which is not a substrate for this enzyme.
achieved in the olfactory mucosa, while a 2.8-fold induc-
tion was found in liver (Banger et al., 1996). B. Inhibition of Nasal Xenobiotic Metabolism
Rabbit nasal CYP2E1 (involved in the metabolism of
ethanol and other alcohols, acetone, acetaminophen, Unlike the case for nasal enzyme induction, inhibition of
nitrosamines, and diethyl ether) can be increased twofold nasal xenobiotic metabolism occurs in a wide range of
by treatment with ethanol and sixfold following acetone enzyme families. Several P450 isoforms have been inhibited
treatment (Ding and Coon, 1990b). These data represent the in homogenates of nasal mucosa by hepatic P450 inhibi-
first evidence of an increase in nasal xenobiotic-metaboliz- tors such as metyrapone, -naphthoflavone, piperonyl
ing capacity of a magnitude that can be considered import- butoxide, and a number of odorants, including 5--
ant physiologically. Induction of CYP2E1 in the olfactory androstenone. Because these inhibitors are common ingredi-
mucosa has been confirmed in other species (Genter et al., ents in many products in everyday usage such as perfumes,
1994; Gu et al., 1998; Longo et al., 1993). Nasal CYP2E1 cosmetics, and household insecticides, exposure to these
and CYP1A2 can also be induced in rats by fasting (Longo compounds may alter nasal metabolic capacity from that
et al., 2000). CYP1A1, which is active in the metabolic observed in controlled laboratory situations (Dahl, 1982;
activation of polycyclic aromatic hydrocarbons, was not Dahl and Brezinski, 1985; Ding and Coon, 1994; Laethem
induced in the olfactory mucosa by 3-MC, but was signifi- et al., 1992).
cantly induced in Bowman’s glands and in the olfactory and Cigarette smoke is another common environmental pol-
respiratory epithelia following a single intraperitoneal lutant that is known to modify olfaction (Frye et al., 1989)
injection of Arochlor 1254 in rats (Voigt et al., 1993); the and to alter the capacity for nasal xenobiotic metabolism
increase in CYP1A1 protein was accompanied by dramat- (Wardlaw et al., 1998). Alterations in nasal metabolism
ically enhanced benzo(a)pyrene hydroxylase activity in the may be the direct result of exposure to the myriad of com-
same sites. More recent studies indicated an induction of ponents of cigarette smoke known to be metabolized in the
CYP1A1 protein in olfactory mucosa of mainstream cig- nasal epithelium including benzo(a)pyrene, N-nitroso-
arette smoke–exposed rats, but a corresponding increase in nornicotine, and cyanide. Rhodanese, the primary enzyme of
CYP1A1 activity was not observed (Wardlaw et al., 1998). cyanide metabolism, shows nearly a 50% reduction in
Induction of nasal P450 enzymes by tobacco smoke has activity in respiratory mucosa from human smokers com-
been proposed as a possible mechanism for developing pared to nonsmokers (Lewis et al., 1991).
resistance to the environmental toxins implicated in parkin- Inhibitors of biotransformation enzymes have been
sonism and other neurological diseases (Gresham et al., used in vivo to demonstrate the role of local metabolism in
1993) (see Chapters 23 and 24). xenobiotic toxicity. For example, treatment with
CYP2As represent major P450 isoforms in the olfac- metyrapone reduced or abolished cytotoxicity caused by a
tory mucosa of a number of species. A study by Beréziat number of toxic chemicals, such as methimazole
et al. (1995) suggested that a CYP2A-like P450 may be (Bergman and Brittebo, 1999), 2,6-dichlorothiobenzamide
induced in rats by treatment with coumarin in drinking (Eriksson and Brittebo, 1995), 2,6-dichlorobenzonitrile
water. However, the same results were not obtained in (Walter et al., 1993), and IDPN (Genter et al., 1994).
another study with a different strain of rats (Gu et al., Metyrapone has also been used to demonstrate that
1997), and no induction of CYP2A was found following inspired styrene is metabolized in nasal tissues in the rat
treatment of mice with several chemicals known to induce and mouse (Morris, 2000). Other P450 inhibitors used for
the same enzyme in the liver (Su et al., 1998). in vivo studies include diethyldithiocarbamate (Deamer
Interestingly, one of the known CYP2E1 and CYP2A5 and Genter, 1995; Eriksson and Brittebo, 1995), carbon
inducers, pyrazole, was found to induce CYP2J4 in rat tetrachloride (Genter et al., 1994), disulfiram (Deamer and
olfactory mucosa as well as in other tissues (Zhang Genter, 1995), 3-aminobenzamide (Eriksson et al., 1996),
Olfactory Mucosa 65

and cobalt protoporphyrin IX (Chamberlain et al., 1998); parallel alterations in histopathology and metabolic activity
the latter depletes P450 by interfering with heme syn- in nasal tissue, interpretation of data from these correlative
thesis. Some inhibitors cause inactivation of a subset of studies can be complicated by the multifaceted nature of
P450s, such as xylene (Blanchard and Morris, 1994) and the toxic response. For example, tissue damage may lead
chlormethiazole (Longo et al., 2000). Inhibition of nasal to increased influx of immune cells, which may contribute
GST-dependent conjugation activity has been achieved by to local metabolic activity.
depleting GSH with phorone (Larsson and Tjälve, 1995) or Another complexity in interpretation of data indicating
phorone plus L-buthionine sulfoximine (Chamberlain et al., altered metabolism following exposure to specific toxi-
1998). In addition, the role of aldehyde dehydrogenase on cants lies in the parallel pathological alterations to the
nasal uptake of inspired acetaldehyde has been examined nasal epithelium. Biochemical data are often normalized
using cyanamide as an inhibitor. While these inhibitors per mg protein, per mg tissue, per mg mitochondrial or
have been useful for the initial identification of biotrans- microsomal protein, and so forth. If cellularity has
formation enzymes or pathways involved in the metabolism decreased in the tissue (as is often the case in the olfactory
and toxicity of a compound, they are generally not spe- epithelium), the normalized data may show no alteration
cific for any single enzyme or even a single family in metabolism, but the total capacity of the tissue to metab-
of enzymes (e.g., Eriksson et al., 1996). Furthermore, these olize inhalants may be severely reduced. Conversely,
inhibitors are most likely also toxicants. Therefore, caution hyperplastic responses may greatly increase the metabolic
should be exercised in interpreting the results obtained capacity without altering the normalized biochemical data.
using chemical inhibitors as tools. Alternatively, mouse Although this problem exists in other tissues as well, the
models with targeted gene deletion of specific P450 and structure of the nasal epithelium and the close association
other biotransformation enzymes are becoming available with cartilage and bone make it difficult to control the
and have been used in limited cases to examine the role of problem by normalizing to total tissue weight, as can read-
biotransformation enzymes in nasal toxicity of xenobiotics ily be done in most other organs. In addition, toxicant
(Genter et al., 1998). exposure and age are both known to produce metaplastic
alteration in the olfactory epithelium. Because these alter-
C. Effects of Mucosal Damage on Nasal Metabolism ations in cell type can also affect metabolic capacity, close
evaluation of both biochemical and histopathological alter-
Expression of olfactory mucosa P450s and CPR is sup- ations in the interpretation of data from nasal epithelium is
pressed when ORNs undergo degeneration as a conse- necessary for valid extrapolations. Finally, because
quence of either chemical toxicity, unilateral naris closure, inhalants can contact and be metabolized in the respiratory
or olfactory bulbectomy (Gu et al., 1997; Schwob et al., or transitional mucosa before reaching the olfactory
1995; Walters et al., 1992, 1993). P450 expression returns to mucosa, metabolic processes occurring in these nasal
normal following successful regeneration of ORNs, but not mucosa will also affect olfactory processes.
when degenerated olfactory mucosa was replaced by respir-
atory type of epithelium (Schwob et al., 1995). The sup-
pressed expression of P450 following olfactory bulbectomy VII. CONCLUSIONS
is particularly intriguing since the P450-expressing cells
were apparently intact after the operation (Walters et al., Both general and research interests in the olfactory system
1992). This result contrasts with the report that the expres- have increased over the last two decades owing to several
sion of PSTg protein is not affected in the olfactory mucosa unique aspects of this system. Its continuous exposure to
following olfactory bulbectomy (Miyawaki et al., 1996). inhaled environmental toxicants, its vulnerability to cell
Decreases in GSH and GST levels have also been found in loss resulting from toxicant insult, and its capacity to
the peripheral olfactory organ of rainbow trout during retro- regenerate neuronal cells following this loss make it a
grade olfactory nerve degeneration, which are followed by unique neural tissue. In addition, its histological structure,
widespread recovery as the ORNs begin to repopulate the with neuronal cells contacting the external environment at
olfactory mucosa (Starcevic and Zielinski, 1997). the nasal lumen and projecting directly to the olfactory
Tissue damage resulting from chemically induced nasal bulb, makes it a viable portal of entry for inhaled environ-
toxicity may underlie some of the in vivo inhibitory effects mental toxicants, as well as a potential route of entry for
of enzyme inhibitors described in the previous section, as therapeutic drugs, into the CNS. Although olfaction has
well as the apparent resistance of nasal biotransformation traditionally been thought of as a sensory system of minor
enzymes to xenobiotic induction (Su et al., 1996). importance in humans, evidence is accumulating that
Furthermore, although it may appear logical to assume olfaction plays an important role in learning and memory,
66 Ding and Dahl

hedonic responses, and reproductive function in humans as REFERENCES

well as in other species.
Only a small fraction of the biochemical repertoire of Aceto, A., Sacchetta, P., Dragani, B., Bucciarelli, T., Angelucci, S.,
the olfactory mucosa has been characterized to date. This Longo, V., Gervasi, G. P., Martini, F., and Di Ilio, C. (1993).
will change rapidly as the new genomics and proteomics Glutathione transferase isoenzymes in olfactory and respiratory
techniques are applied to the olfactory system. What has epithelium of cattle. Biochem. Pharmacol. 46:2127–2133.
been learned so far already indicates that the nasal Adams, D. R., Jones, A. M., Plopper, C. G., Serabjit-Singh, C. J.,
and Philpot, R. M. (1991). Distribution of cytochrome P-450
biotransformation enzymes are very likely to play
monoxygenase enzymes in the nasal mucosa of hamster and
important roles in many cellular processes in the olfactory rat. Am. J. Anatomy 190:291–298.
mucosa, as supported by their high metabolic capacity, Anand-Kumar, T. C., Sehgal, A., David, G. F. X., Bajaj, J. S., and
their diverse substrates of both endogenous and exogenous Prasad, M. R. N. (1980). Effect of intranasal administration of
origins, and their tissue and cell-type specific expression. hormonal steroids on serum testosterone and spermatogenesis in
The precise nature of these roles is not yet fully under- rhesus monkey (Macaca mulatta). Biol. Reprod. 22:935–940.
stood; however, rapid progress is anticipated, as an Armstrong, R. N. (1997). Structure, catalytic mechanism, and evo-
increasing number of knockout mouse and other genetic- lution of the glutathione transferases. Chem. Res. Toxicol.
ally modified animal models become available. 10:2–18.
As reviewed in this chapter, nasal biotransformation Balboni, G. C. (1967). L’ultrastuttura dell’epitelio olfattiva nel
enzymes can alter inhaled toxicants either by converting ratto e sue modificazioni in seguito a castrazione e alla som-
ministrazione, a ratti castrati, di testosterone. Arch. Ital. Anat.
them to less toxic metabolites or by activating them to
Embriol. 52:203–223.
reactive chemicals that result in local damage, and in some Balboni, G. C., et al. (1982). Morphological features of the olfac-
cases damage to other tissues as well. As such, the role of tory epithelium in prepubertal and postpubertal rats. In
nasal biotransformation enzymes has historically received Olfaction and Endocrine Regulation, Breiphol (Ed.). IRL Press,
the most attention in the field of toxicology. The biotrans- London, pp. 285–295.
formation capacity in the olfactory mucosa is vulnerable to Banger, K. K., Lock, E. A., and Reed, C. J. (1993). The character-
modification by a variety of toxicant exposures, histological ization of glutathione S-transferases from rat olfactory epithe-
changes, and disease or inflammatory processes. Further, lium. Biochem. J. 290:199–204.
the nasal xenobiotic-metabolism activities will be influ- Banger, K. K., Lock, E. A., and Reed, C. J. (1996). Regulation of
enced by genetic polymorphisms of the participating bio- rat olfactory glutathione S-transferase expression. Investigation
transformation enzymes. Thus, the capacity to detoxify or of sex differences, induction, and ontogenesis. Biochem.
Pharmacol. 52:801–808.
activate inhaled toxicants is likely to be a fluid system best
Baron, J., Voigt, J. M., Whitter, T. B., Kawabata, T. T., Knapp, S.
understood with respect to an individual case history and A., Guengerich, F. P., and Jakoby, W. B. (1986). Identification
genetic makeup. Future studies on the interactions of vari- of intratissue sites for xenobiotic activation and detoxication.
ous components of the olfactory system, the activities of Adv. Exp. Med. Biol. 197:119–144.
individual biotransformation enzymes and their genetic Baron, J., Burke, J. P., Guengerich, F. P., Jakoby, W. B., and Voigt,
polymorphisms, and the impact of nasal xenobiotic-metab- J. M. (1988). Sites for xenobiotic activation and detoxication
olism on other systems such as the CNS will make it pos- within the respiratory tract: implications for chemically
sible to identify situations, periods, or individuals of induced toxicity. Toxicol. Appl. Pharmacol. 93:493–505.
increased vulnerability to inhaled xenobiotics and to Beauchamp, G. K., Doty, R. L., Moulton, D. G., and Mugford, R.
reduce the risk of toxicity through targeted prevention. A. (1976). The pheromone concept in mammalian chemical
communication: a critique. In Mammalian Olfaction,
Reproductive Processes, and Behavior, R. L. Doty (Ed.).
ACKNOWLEDGMENTS Academic Press, New York, pp. 147–153.
Belinsky, S. A., Foley, J. F., White, C. M., Anderson, M. W., and
The authors acknowledge Jack Harkema for Figures 1 and Maronpot, R. R. (1990). Dose-response relationship between
3 and Dr. Bert Menco and Virginia Carr for the transmis- O6-methylguanine formation in Clara cells and induction of
pulmonary neoplasia in the rat by 4-(methylnitrosamino)-1-(3-
sion electron micrograph in Figure 2 in this chapter. We
pyridyl)-1-butanone. Cancer Res. 50:3772–3780.
also thank Drs. Mary Beth Genter and Katherine Ben-Arie, N., Khen, M., and Lancet, D. (1993). Glutathione S-
Henrikson for reading the manuscript. This work was transferases in rat olfactory epithelium: purification, molecular
supported in part by NIH Grants ES07462 from the properties and odorant biotransformation. Biochem. J.
National Institute of Environmental Health Sciences and 292:379–384.
DC02640 from the National Institute on Deafness and Bergman, U. and Brittebo, E. B. (1999). Methimazole toxicity in
Other Communication Disorders. rodents: covalent binding in the olfactory mucosa and detection
Olfactory Mucosa 67

of glial fibrillary acidic protein in the olfactory bulb. Toxicol. Chaen, T., Watanabe, N., Mogi, G., Mori, K., and Takeyama, M.
Appl. Pharmacol. 155:190–200. (1993). Substance P and vasoactive intestinal peptide in nasal
Beréziat, J. C., Raffalli, F., Schmezer, P., Frei, E., Geneste, O., and secretions and plasma from patients with nasal allergy. Ann.
Lang, M. A. (1995). Cytochrome P450 2A of nasal epithelium: Otol. Rhinol. Laryngol. 102:16–21.
regulation and role in carcinogen metabolism. Mol. Chamberlain, M. P., Lock, E. A., Gaskell, B. A., and Reed, C. J.
Carcinogen. 14:130–139. (1998). The role of glutathione S-transferase- and cytochrome
Blanchard, K. T. and Morris, J. B. (1994). Effects of m-xylene on P450-dependent metabolism in the olfactory toxicity of methyl
rat nasal cytochrome P450 mixed function oxidase activities. iodide in the rat. Arch. Toxicol. 72:420–428.
Toxicol. Lett. 70:253–259. Chambon, P. (1996). A decade of molecular biology of retinoic
Bogdanffy, M. S. (1990). Biotransformation enzymes in the rodent acid receptors. FASEB J. 10:940–954.
nasal mucosa: the value of a histochemical approach. Environ. Chen, Y., Getchell, M. L., Ding, X., and Getchell, T. V. (1992).
Health Perspect. 85:177–186. Immunolocalization of two cytochrome P450 isozymes in rat
Bogdanffy, M. S., Randall, H. W., and Morgan, K. T. (1986). nasal chemosensory tissue. Neuroreport 3:749–752.
Histochemical localization of aldehyde dehydrogenase in the Cross, C. E., Van der Vliet, V, O’Neill, C. A., Louie, S., and
respiratory tract of the Fischer-344 rat. Toxicol. Appl. Halliwell, B. (1994). Oxidants, antioxidants, and respiratory tract
Pharmacol. 82:560–567. lining fluids. Environ. Health Perspect. 102 (Suppl 10):185–191.
Bogdanffy, M. S., Randall, H. W., and Morgan, K. T. (1987). Dahl, A. R. (1982). The inhibition of rat nasal cytochrome P-450-
Biochemical quantitation and histochemical localization of car- dependent monooxygenase by the essence heliotropin (piper-
boxylesterase in the nasal passages of the Fischer-344 rat and onal). Drug Metab. Dispos. 10:553–554.
B6C3F1 mouse. Toxicol. Appl. Pharmacol. 88:183–194. Dahl, A. R. (1988). The effect of cytochrome P-450-dependent
Bogdanffy, M. S., Sarangapani, R., Kimbell, J. S., Frame, S. R., metabolism and other enzyme activities on olfaction. In
and Plowchalk, D. R. (1998). Analysis of vinyl acetate metab- Molecular Neurobiology of the Olfactory System: Molecular,
olism in rat and human nasal tissues by an in vitro gas uptake Membranous, and Cytological Studies, F. L. Margolis and T. V.
technique. Toxicol. Sci. 46:235–246. Getchell (Eds.). Plenum Press, New York, pp. 51–70.
Bond, J. A. (1983). Some biotransformation enzymes responsible Dahl, A. R., and Brezinski, D. A. (1985). The inhibition of rabbit
for polycyclic aromatic hydrocarbon metabolism in rat nasal nasal and hepatic cytochrome P-450-dependent hexam-
turbinates: effects on enzyme activities of in vitro modifiers and ethylphosphoramide (HMPA) N-demethylase by methylene-
intraperitoneal and inhalation exposures of rats to inducing dioxphenyl compounds. Biochem. Pharmacol. 34:631–636.
agents. Cancer Res. 43:4804–4811. Dahl, A. R., and Hadley, W. M. (1983). Formaldehyde production pro-
Bond, J. A., Harkema, J. R., and Russell, V. I. (1988). Regional dis- moted by rat nasal cytochrome P450-dependent monooxygenases
tribution of xenobiotic metabolizing enzymes in respiratory air- with nasal decongestants, essences, solvents, air pollutants, nicotine,
ways of dogs. Drug Metab. Dispos. 16:116–124. and cocaine as substrates. Toxicol. Appl. Pharmacol. 67:200–205.
Bourne, G. H. (1948). Alkaline phosphatase in taste buds and nasal Dahl, A. R., and Hadley, W. M. (1991). Nasal cavity enzymes
mucosa. Nature 161:445–446. involved in xenobiotic metabolism: effects on the toxicity of
Brandt, I., Brittebo, E. B., Feil, V. J., and Bakke, J. E. (1990). inhalants. Crit. Rev. Toxicol. 21:345–372.
Irreversible binding and toxicity of the herbicide dichlobenil Dahl, A. R., and Lewis, J. L. (1993). Respiratory tract uptake of
(2,6-dichlorobenzonitrile) in the olfactory mucosa of mice. inhalants and metabolism of xenobiotics. Annu. Rev.
Toxicol. Appl. Pharmacol. 103:491–501. Pharmacol. Toxicol. 32:383–407.
Brittebo, E. B. (1982). Demethylation of aminopyrine by the nasal Dahl, A. R., and Waruszewski, B. A. (1990). Metabolism of
mucosa in mice and rats. Acta Pharmacol. Toxicol. 51:227–232. organonitriles to cyanide by rat nasal tissue enzymes.
Brittebo, E. B. (1985). Localization of oestradiol in the rat nasal Xenobiotica 19:1201–1205.
mucosa. Acta Pharmacol. Toxicol. 57:285–290. Dahl, A. R., Hadley, W. M., Hahn, F. F., Benson, J. M., and
Brittebo, E. B. (1997). Metabolism-dependent activation and toxi- McClellan, R. O. (1982). Cytochrome P-450-dependent
city of chemicals in nasal glands. Mutat. Res. 380:61–75. monooxygenases in olfactory epithelium in dogs: possible role
Brittebo, E. B. and Rafter, J. J. (1984). Steroid metabolism by rat in tumorigenicity. Science 216:57–59.
nasal mucosa: studies on progesterone and testosterone. J. Dahl, A. R., Coslett, D. S., Bond, J. A., and Hesseltine, G. R.
Steroid Biochem. 20:1147–1151. (1985). Metabolism of benzo(a)pyrene on the nasal mucosa of
Brown, D., Garcia-Segura, L. M., and Orci, L. (1984). Carbonic Syrian hamsters: comparison to metabolism by other extrahep-
anhydrase is present in olfactory receptor cells. Histochemistry atic tissues and possible role of nasally produced metabolites in
80:307–309. carcinogenesis. J. Natl. Cancer Inst. 75:135–139.
Capdevila, J. H., Falck, J. R., and Estabrook, R. W. (1992). Cytochrome Deamer, N. J., and Genter, M. B. (1995). Olfactory toxicity of
P450 and the arachidonate cascade. FASEB J. 6:731–736. diethyldithiocarbamate (DDTC) and disulfiram and the protec-
Casanova-Schmitz, M., David, R. M., and Heck, H. D. (1984). tive effect of DDTC against the olfactory toxicity of dichlobe-
Oxidation of formaldehyde and acetaldehyde by NAD-depen- nil. Chem. Biol. Interact. 95:215–226.
dent dehydrogenases in rat nasal mucosal homogenates. Deshpande, V. S., Genter, M. B., Jung, C., and Desai, P. B. (1999).
Biochem. Pharmacol. 33:1137–1142. Characterization of lidocaine metabolism by rat nasal micro-
68 Ding and Dahl

somes: implications for nasal drug delivery. Eur. J. Drug system which is altered in some neurodegenerative diseases.
Metab. Pharmacokinetics 24:177–182. Brain Res. 835:306–314.
Ding, X., and Coon, M. J. (1988). Purification and characterization Fong, K. J., Kern, R. C., Foster, J. D., Zhao, J. C., and Pitovski, D.
of two unique forms of cytochrome P-450 from rabbit nasal Z. (1999). Olfactory secretion and sodium, potassium-adeno-
microsomes. Biochemistry 27:8330–8337. sine triphosphatase: regulation by corticosteroids.
Ding, X., and Coon, M. J. (1990a). Immunochemical characteriza- Laryngoscope 109:383–388.
tion of multiple forms of cytochrome P-450 in rabbit nasal Frye, R. E., Doty, R. L., and Schwartz, B. (1989). Influence of cig-
microsomes and evidence for tissue-specific expression of arette smoking on olfaction: evidence for a dose-response rela-
P450s NMa and NMb. Mol. Pharmacol. 37:489–496. tionship. JAMA 263:1233–1236.
Ding, X., and Coon, M. J. (1990b). Induction of cytochrome P-450 Fujii, H., Sato, T., Kaneko, S., Gotoh, O., Fujii-Kuriyama, Y.,
isozyme 3a (P-450IIE1) in rabbit olfactory mucosa by ethanol Osawa, K., Kato, S., and Hamada, H. (1997). Metabolic inacti-
and acetone. Drug Metab. Dispos. 18:742–745. vation of retinoic acid by a novel P450 differentially expressed
Ding, X., and Coon, M. J. (1994). Steroid metabolism by rabbit in developing mouse embryos. EMBO J. 16:4163–4173.
olfactory-specific P450 2G1. Arch. Biochem. Biophys. Gaskell, B. A. (1990). Nonneoplastic changes in the olfactory
315:454–459. epithelium—experimental studies. Environ. Health Perspect.
Ding, X., Koop, D. R., Crump, B. L., and Coon, M. J. (1986). 85:275–289.
Immunochemical identification of cytochrome P-450 isozyme Gennings, J. N., Gower, D. B., and Bannister, L. H. (1974). Studies
3a (P-450alc) in rabbit nasal and kidney microsomes and evi- on the metabolism of the odoriferous ketones, 5-androst-16-
dence for differential induction by alcohol. Mol. Pharmacol. en-3-one and 4, 16-androstadien-3-one by the nasal epithelium
30:370–378. of the mature and immature sow. Biochim. Biophys. Acta
Ding, X., Porter, T. D., Peng, H. M., and Coon, M. J. (1991). 369:294–303.
cDNA and derived amino acid sequence of rabbit nasal Genter, M. B., Llorens, J., O’Callaghan, J. P., Peele, D. B., Morgan,
cytochrome P450NMb (P450IIG1), a unique isozyme possibly K. T., and Crofton, K. M. (1992). Olfactory toxicity of ,-
involved in olfaction. Arch. Biochem. Biophys. 285:120–125. iminodipropionitrile in the rat. J. Pharmacol. Exp. Ther.
Ding, X., Peng, H. M., and Coon, M. J. (1992). P450 cytochromes 263:1432–1439.
NMa, NMb (2G1), and LM4 (1A2) are differentially expressed Genter, M. B., Deamer, N. J., Cao, Y., and Levi, P. E. (1994).
during development in rabbit olfactory mucosa and liver. Mol. Effects of P450 inhibition and induction on the olfactory toxic-
Pharmacol. 42:1027–1032. ity of beta,beta-iminodipropionitrile (IDPN) in the rat. J.
Ding, X., and Coon, M. J. (1993). Olfactory cytochrome P450. In Biochem. Toxicol. 9:31–39.
Cytochrome P450, Handbook of Experimental Pharmacology, Genter, M. B., Owens, D. M., and Deamer, N. J. (1995a).
Vol. 105, Schenkman J. B., and Greim, H. (Eds.). Springer- Distribution of microsomal epoxide hydrolase and glutathione
Verlag, New York, pp. 351–361. S-transferase in the rat olfactory mucosa: relevance to distribu-
Ding, X., Spink, D. C., Bhama, J. K., Sheng, J. J., Vaz, A. D., and tion of lesions caused by systemically-administered olfactory
Coon, M. J. (1996). Metabolic activation of 2,6-dichloroben- toxicants. Chem. Senses 20:385–392.
zonitrile, an olfactory-specific toxicant, by rat, rabbit, and Genter, M. B., Deamer, N. J., Blake, B. L., Wesley, D. S., and Levi,
human cytochromes P450. Mol. Pharmacol. 49:1113–1121. P. E. (1995b). Olfactory toxicity of methimazole: dose-response
Dorries, K. M., Schmidt, H. J., Beauchamp, G. K., and Wysocki, and structure-activity studies and characterization of flavin-
C. J. (1989). Changes in sensitivity to the odor of androstenone containing monooxygenase activity in the Long-Evans rat
during adolescence. Dev. Psychobiol. 22:423–435. olfactory mucosa. Toxicol. Pathol. 23:477–486.
Duester, G. (1996). Involvement of alcohol dehydrogenase, short- Genter, M. B., Owens, D. M., Carlone, H. B., and Crofton, K. M.
chain dehydrogenase/reductase, aldehyde dehydrogenase, and (1996). Characterization of olfactory deficits in the rat follow-
cytochrome P450 in the control of retinoid signaling by activa- ing administration of 2,6-dichlorobenzonitrile (dichlobenil),
tion of retinoic acid synthesis. Biochemistry 35:12221–12227. 3,3-iminodipropionitrile, or methimazole. Fund. Appl. Toxicol.
Eaton, D. L., and Bammler, T. K. (1999). Concise review of the 29:71–77.
glutathione S-transferases and their significance to toxicology. Genter, M. B., Liang, H. C., Gu, J., Ding, X., Negishi, M.,
Toxicol. Sci. 49:156–164. McKinnon, R. A., and Nebert, D. W. (1998). Role of CYP2A5
Eriksson, C. and Brittebo, E. B. (1995). Effects of the herbicide and 2G1 in acetaminophen metabolism and toxicity in the
chlorthiamid on the olfactory mucosa. Toxicol. Lett. olfactory mucosa of the Cyp1a2(/) mouse. Biochem.
76:203–208. Pharmacol. 55:1819–1826.
Eriksson, C., Busk, L., and Brittebo, E. B. (1996). 3- Gervasi, P. G., Longo, V., Ursino, F., and Panattoni, G. (1989).
Aminobenzamide: effects on cytochrome P450-dependent Drug metabolizing enzymes in respiratory mucosa of humans.
metabolism of chemicals and on the toxicity of dichlobenil in Comparison with rats. In Cytochrome P-450: Biochemistry and
the olfactory mucosa. Toxicol. Appl. Pharmacol. 136:324–331. Biophysics, I. Schuster (Ed.). Taylor & Francis, New York,
Federico, G., Maremmani, C., Cinquanta, L., Baroncelli, G. I., pp. 97–100.
Fattori, B., and Saggese, G. (1999). Mucus of the human Getchell, T. V., and Getchell, M. L. (1977) Early events in verte-
olfactory epithelium contains the insulin-like growth factor-I brate olfaction. Chem. Senses 2:313–326.
Olfactory Mucosa 69

Getchell, M. L., Zielinski, B., and Getchell, T. V. (1988). Odorant Hadley, W. M., and Dahl, A. R. (1983). Cytochrome P-450-depen-
and autonomic regulation of secretion in the olfactory mucosa. dent monooxygenase activity in nasal membranes of six
In Molecular Neurobiology of the Olfactory System: species. Drug Metab. Dispos. 11:275–276.
Molecular, Membranous, and Cytological Studies, F. L. Hahn, I., Scherer, P. W., and Mozell, M. M. (1993). Velocity pro-
Margolis and T. V. Getchell (Eds.). Plenum Press, New York, files measured for airflow through a large scale model of the
pp. 71–98. human nasal cavity. J. Appl. Physiol. 75:2273–2287.
Getchell, M. L., Chen, Y., Ding, X., Sparks, D. L., and Getchell, T. Harkema, J. R. (1990). Comparative pathology of the nasal mucosa
V. (1993a). Immunohistochemical localization of a cytochrome in laboratory animals exposed to inhaled irritants. Environ.
P-450 isozyme in human nasal mucosa: age-related trends. Health Perspect. 85:231–238.
Ann. Otol. Rhinol. Laryngol. 102:368–374. Harkema, J. R. (1991). Comparative aspects of nasal airway anato-
Getchell, T. V., Su, Z., and Getchell, M. L. (1993b). Mucous my: relevance to inhalation toxicology. Toxicol. Pathol.
domains: microchemical heterogeneity in the mucociliary com- 19:321–336.
plex of the olfactory epithelium. Ciba Foundation Symp. Harkema, J. R., Hotchkiss, J. A., Harlnsen, A. G., and Henderson,
179:27–40. R. F. (1988). In vivo effects of transient neutrophil influx on
Ghantous, H., Dencker, L., Gabrielsson, J., Danielsson, B. R. G., nasal respiratory epithelial mucosubstances: quantitative histo-
and Bergrnan, K. (1990). Accumulation and turnover of chemistry. Am. J. Pathol. 130:605–615.
metabolites of toluene and xylene in nasal mucosa and olfac- Hayes, C. L., Spink, D. C., Spink, B. C., Cao, J. Q., Walker, N. J.,
tory bulb in the mouse. Pharmacol. Toxicol. 66:87–92. and Sutter, T. R. (1996). 17 beta-estradiol hydroxylation cat-
Goldstein, B. J., Fang, H. S., Youngentob, S. L., and Schwob, J. E. alyzed by human cytochrome P450 1B1. Proc. Natl. Acad. Sci.
(1998). Transplantation of multipotent progenitors from the USA 93:9776–9781.
adult olfactory epithelium. Neuroreport 9:1611–1617. Henriksson, J. and Tjälve, H. (2000). Manganese taken up into the
Graziadei, P. P., and Monti-Graziadei, A. G. (1983). Regeneration CNS via the olfactory pathway in rats affects astrocytes.
in the olfactory system of vertebrates. Am. J. Otolaryngol. Toxicol. Sci. 55:392–398.
4:228–233. Hovemann, B. T., Sehlmeyer, F., and Malz, J. (1997). Drosophila
Gresham, L. S., Molgaard, C. A., and Smith, R. A. (1993). melanogaster NADPH-cytochrome P450 oxidoreductase: pro-
Induction of cytochrome P-450 enzymes via tobacco smoke: a nounced expression in antennae may be related to odorant
potential mechanism for developing resistance to environmen- clearance. Gene 189:213–219.
tal toxins as related to parkinsonism and other neurologic dis- Hua, Z., Zhang, Q.-Y., Su, T., Lipinskas, T. W., and Ding, X.
eases. Neuroepidemiology 12:114–116. (1997). cDNA cloning, heterologous expression, and character-
Gu, J., Walker, V. E., Lipinskas, T. W., Walker, D. M., and Ding, X. ization of mouse CYP2G1, an olfactory-specific steroid
(1997). Intraperitoneal administration of coumarin causes tis- hydroxylase. Arch. Biochem. Biophy. 340:208–214.
sue-selective depletion of cytochromes P450 and cytotoxicity Huard, J. M. T., Youngentob, S. L., Goldstein, B. J., Luskin, M. B.,
in the olfactory mucosa. Toxicol. Appl. Pharmacol. and Schwob, J. E. (1998). Adult olfactory epithelium contains
146:134–143. multipotent progenitors that give rise to neurons and non-neur-
Gu, J., Zhang, Q.-Y., Genter, M. B., Lipinskas, T. W., Negishi, M., al cells. J. Comp. Neurol. 400:469–486.
Nebert, D. W., and Ding, X. (1998). Purification and character- Iravani, J., and Melville, G. N. (1974). Long-term effect of ciga-
ization of heterologously expressed mouse CYP2A5 and rette smoke on mucociliary function in animals. Respiration
CYP2G1: role in metabolic activation of acetaminophen and 31:358–366.
2,6-dichlorobenzonitrile in mouse olfactory mucosal micro- Jaillardon, E., Astic, L., Cattarelli, M., and Gay, B. (1992). A
somes. J. Pharmacol. Exp. Therapeutics 285:1287–1295. numerical model of the flow in a rat nasal cavity. Chem. Senses
Gu, J., Dudley, C., Su, T., Spink, D. C., Zhang, Q.-Y., Moss, R. L., 17(6):844.
and Ding, X. (1999). Cytochrome P450 and steroid hydroxy- Jedlitschky, G., Cassidy, A. J., Sales, M., Pratt, N., and Burchell, B.
lase activity in mouse olfactory and vomeronasal mucosa. (1999). Cloning and characterization of a novel human olfac-
Biochem. Biophys. Res. Commun. 266:262–267. tory UDP-glucuronosyltransferase. Biochem. J. 340:837–843.
Gu, J., Su, T., Chen, Y., Zhang, Q.-Y., and Ding, X. (2000). Jeffery, E. H., and Haschek, W. M. (1988). Protection by
Expression of biotransformation enzymes in human fetal olfac- dimethyl sulfoxide against acetaminophen-induced hepatic,
tory mucosa: potential roles in developmental toxicity. Toxicol. but not respiratory toxicity in the mouse. Toxicol. Appl.
Appl. Pharmacol. 165:158–162. Pharmacol. 93:452–461.
Guengerich, F. P., Dannan, G. A., Wright, S. T., Martin, M. V., and Jorissen, M., Willems, T., and Van der Schueren, S. B. (1998).
Kaminsky, L. S. (1982). Purification and characterization of Nasal ciliary beat frequency is age independent.
liver microsomal cytochromes P-450: electrophoretic, spectral, Laryngoscope 108:1042–1047.
catalytic, and immunochemical properties and inducibility of Josefsen, D., Blomhoff, H. K., Lomo, J., Blystad, A. K., and
eight isozymes isolated from rats treated with phenobarbital or Smeland, E. B. (1999). Retinoic acid induces apoptosis of
-naphthoflavone. Biochemistry 21:6019–6030. human CD34() hematopoietic progenitor cells: Involvement
Hadley, W. M., and Dahl, A. R. (1982). Cytochrome P-450 depen- of retinoic acid receptors and retinoid X receptors depends on
dent monooxygenase activity in rat nasal epithelial membranes. lineage commitment of the hematopoietic progenitor cells. Exp.
Toxicol. Lett. 10:417–422. Hematol. 27:642–653.
70 Ding and Dahl

Kaliner, M. A. (1991). Human nasal respiratory secretions and host inhibitor in the human nasal airway. Am. Rev. Respir. Dis.
defense. Am. Rev. Respir. Vis. 144:552–556. 147:710–716.
Kashiwayanagi, M., Sai, K., and Kurihara, K. (1987) Cell suspen- Lewis, J. L., Rhoades, C. E., Gervasi, P. G., Griffith, W. C., and
sions from porcine olfactory mucosa. Changes in membrance Dahl, A. R. (1991). The cyanide-metabolizing enzyme rho-
potential and membrane fluidity in response to various odor- danese in human nasal respiratory mucosa. Toxicol. Appl.
ants. J. Gen. Physiol. 89:443–457. Pharmacol. 108:114–120.
Kinlbell, J. S., Gross, E. A., Joyner, D. R., Godo, M. N., and Lewis, J. L., Nikula, K. J., and Dahl, A. R. (1992a). Comparative
Morgan, K. T. (1993). Application of computational fluid analysis of activity and distribution of nasal carboxylesterases
dynamics to regional dosimetry of inhaled chemicals in the (CE). Toxicologist 12:398.
upper respiratory tract of the rat. Toxicol. Appl. Pharmacol. Lewis, J. L., Rhoades, C. E., Bice, D. E., Harkema, J. R.,
121:253–263. Hotchkiss, J. A., Sylvester, D. M., and Dahl, A. R. (1992b).
Koskela, S., Hakkola, J., Hukkanen, J., Pelkonen, O., Sorri, M., Interspecies comparison of cellular localization of the cyanide
Saranen, A., Anttila, S., Fernandez-Salguero, P., Gonzalez, F., metabolizing enzyme rhodanese within olfactory mucosa. Anat.
and Raunio, H. (1999). Expression of CYP2A genes in human Rec. 232:620–627.
liver and extrahepatic tissues. Biochem. Pharmacol. Lewis, J. L., Hahn, F. F., and Dahl, A. R. (1994a). Transport of
57:1407–1413. inhaled toxicants to the central nervous system: characteristics
Kratskin, I. L., Kimura, Y., Hastings, L., and Doty, R. L. (1999). of a nose-brain barrier. In The Vulnerable Brain and
Chronic dexamethasone treatment potentiates insult to olfac- Environmental Risks, Vol. 3, Toxins in Air and Water, R. L.
tory receptor cells produced by 3-methylindole. Brain Res. Isaacson and K. F. Jensen (Eds.). Plenum Press, New York,
847:240–246. pp. 77–103.
Krishna, N. S., Getchell, T. V., and Getchell, M. L. (1994). Lewis, J. L., Nikula, K. J., Novak, R., and Dahl, A. R. (1994b).
Differential expression of alpha, mu, and pi classes of glu- Comparative localization of carboxylesterase in F344 rat, bea-
tathione S-transferases in chemosensory mucosae of rats during gle dog, and human nasal tissue. Anat. Rec. 239:55–64.
development. Cell Tissue Res. 275:435–450. Liu, C., Zhuo, X., Gonzalez, F. J., and Ding, X. (1996).
Krishna, N. S., Getchell, T. V., Dhooper, N., Awasthi, Y. C., and Baculovirus-mediated expression and characterization of rat
Getchell, M. L. (1995a). Age- and gender-related trends in the CYP2A3 and human CYP2A6: role in metabolic activation of
expression of glutathione S-transferases in human nasal nasal toxicants. Mol. Pharmacol. 50:781–788.
mucosa. Ann. Otol. Rhinol. Laryngol. 104:812–822. Longo, V. and Ingelman-Sundberg, M. (1993). Acetone-dependent
Krishna, N. S., Getchell, M. L., Margolis, F. L., and Getchell, T. V. regulation of cytochromes P4502E1 and P4502B1 in rat nasal
(1995b). Differential expression of vomeromodulin and odor- mucosa. Biochem. Pharmacol. 46:1945–1951.
ant-binding protein, putative pheromone and odorant trans- Longo, V., Citti, L., and Gervasi, P. G. (1988). Biotransformation
porters, in the developing rat nasal chemosensory mucosae. enzymes in nasal mucosa and liver of Sprague-Dawley rats.
J. Neurosci. Res. 40:54–71. Toxicol. Lett. 44:289–297.
Laethem, R. M., Laethem, C. L., Ding, X., and Koop, D. R. (1992). Longo, V., Amato, G., Santucci, A., and Gervasi, P. G. (1997).
P-450-dependent arachidonic acid metabolism in rabbit olfac- Purification and characterization of three constitutive
tory microsomes. J. Pharmacol. Exp. Therapeutics cytochrome P-450 isoforms from bovine olfactory epithelium.
262:433–438. Biochem. J. 323:65–70.
Larsson, P., and Tjälve, H. (1995). Extrahepatic bioactivation of Longo, V., Ingelman-Sundberg, M., Amato, G., Salvetti, A., and
aflatoxin B1 in fetal, infant and adult rats. Chem. Biol. Interact. Gervasi, P. G. (2000). Effect of starvation and chlormethiazole
94:1–19. on cytochrome P450s of rat nasal mucosa. Biochem.
Larsson, P., and Tjälve, H. (2000). Intranasal instillation of afla- Pharmacol. 59:1425–1432.
toxin B-1 in rats: Bioactivation in the nasal mucosa and neu- Luo, G., Zeldin, D. C., Blaisdell, J. A., Hodgson, E., and Goldstein,
ronal transport to the olfactory bulb. Toxicol. Sci. 55:383–391. J. A. (1998). Cloning and expression of murine CYP2CS and
Larsson, P., Busk, L., and Tjälve, H. (1994). Hepatic and extrahep- their ability to metabolize arachidonic acid. Arch. Biochem.
atic bioactivation and GSH conjugation of aflatoxin B1 in Biophys. 357:45–57.
sheep. Carcinogenesis 15:947–955. Lupo, D., Lodi, L., Canonaco, M., Valenti, A., and Dessi-
Lazard, D., Tal, N., Rubinstein, M., Khen, M., Lancet, D., and Fulgheri, F. (1986). Testosterone metabolism in the olfactory
Zupko, K. (1990). Identification and biochemical analysis of epithelium of intact and castrated male rats. Neurosci. Lett.
novel olfactory-specific cytochrome P-450IIA and UDP-glu- 69:259–262.
curonyl transferase. Biochemistry 29:7433–7440. Mackenzie, P. I., Owens, I. S., Burchell, B., Bock, K. W., Bairoch,
Lazard, D., Zupko, K., Poria, Y., Nef, P., Lazarovits, J., Horn, S., A., Belanger, A., Fournel-Gigleux, S., Green, M., Hum, D. W.,
Khen, M., and Lancet, D. (1991). Odorant signal termination Iyanagi, T., Lancet, D., Louisot, P., Magdalou, J., Chowdhury,
by olfactory UDP glucuronosyl transferase. Nature J. R., Ritter, J. K., Schachter, H., Tephly, T. R., Tipton, K. F.,
349:790–793. and Nebert, D. W. (1997). The UDP glycosyltransferase gene
Lee, C. H., Igarashi, Y., Hohman, R. J., Kaulbach, H., White, M. V., superfamily: recommended nomenclature update based on
and Kaliner, M. A. (1993). Distribution of secretory leukoprotease evolutionary divergence. Pharmacogenetics 7:255–269.
Olfactory Mucosa 71

Makita, K., Falck, J. R., and Capdevila, J. H. (1996). Cytochrome enzyme possibly involved in chemoreception. J. Biol. Chem.
P450, the arachidonic acid cascade, and hypertension: new vis- 264:6780–6785.
tas for an old enzyme system. FASEB J. 10:1456–1463. Nef, P., Larabee, T. M., Kagimoto, K., and Meyer, U. A. (1990).
Margalit, T., and Lancet, D. (1993). Expression of olfactory recep- Olfactory-specific cytochrome P-450 (P-45001fl; IIG1): gene
tor and transduction genes during rat development. Dev. Brain structure and developmental regulation. J. Biol. Chem.
Res. 73:7–16. 265:2903–2907.
Marini, S., Longo, V., Mazzaccaro, A., and Gervasi, P. G. (1998). Nelson, D. R., Koymans, L., Kamataki, T., Stegeman, J. J.,
Xenobiotic-metabolizing enzymes in pig nasal and hepatic tis- Feyereisen, R., Waxman, D. J., Waterman, M. R., Gotoh, O.,
sues. Xenobiotica 28:923–935. Coon, M. J., Estabrook, R. W., Gunsalus, I. C., and Nebert, D.
Matsui, M., Tamura, H., Nagai, F., Homma, H., Miyawaki, A., and W. (1996). P450 superfamily: update on new sequences, gene
Mikoshiba, K. (1998). On the nature of rat hepatic and mouse mapping, accession numbers and nomenclature.
olfactory sulfotransferases. Chem. Biol. Interact. 109:69–80. Pharmacogenetics 6:1–42.
McLean, J. H., Shipley, M. T., and Bernstein, D. I. (1989). Golgi- Nikula, K. J., Novak, R. F., Chang, I. Y., Dahl, A. R., Kracko, D. A.,
like, transneuronal retrograde labeling with CNS injections of Zangar, R. C., Kim, S. G., and Lewis, J. L. (1995). Induction of
herpes simplex virus type 1. Brain Res. Bull. 22:867–881. nasal carboxylesterase in F344 rats following inhalation expo-
Mellert, T. K., Getchell, M. L., Sparks, L., and Getchell, T. V. sure to pyridine. Drug Metab. Disposition 23:529–535.
(1992). Characterization of the immune barrier in human olfac- Novoselov, S. V., Peshenko, I. V., Popov, V. I., Novoselov, V. I.,
tory mucosa. Otolaryngol. Head Neck Surg. 106:181–188. Bystrova, M. F., Evdokimov, V. J., Kamzalov, S. S., Merkulova,
Min, Y. G., Ohyama, M., Lee, K. S., Rhee, C. S., Oh, S. H., Sung, M. I., Shuvaeva, T. M., Lipkin, V. M., and Fesenko, E. E.
M. W., Yun, J. B., and Jung, I. H. (1999). Effects of free radi- (1999). Localization of 28-kDa peroxiredoxin in rat epithelial
cals on ciliary movement in the human nasal epithelial cells. tissues and its antioxidant properties. Cell Tissue Res.
Auris Nasus Larynx 26:159–163. 298:471–480.
Miyawaki, A., Homma, H., Tamura, H., Matsui, M., and Ohkubo, K., Okuda, M., and Kaliner, M. A. (1994).
Mikoshiba, K. (1996). Zonal distribution of sulfotransferase Immunological localization of neuropeptide-degrading
for phenol in olfactory sustentacular cells. EMBO J. enzymes in the nasal mucosa. Rhinology 32:130–133.
15:2050–2055. Ohkubo, K., Baraniuk, J. N., Merida, M., Hausfeld, J. N., Okada,
Morgan, K. T., and Monticello, T. M. (1990). Airflow, gas deposi- H., and Kaliner, M. A. (1995). Human nasal mucosal car-
tion, and lesion distribution in the nasal passages. Environ. boxypeptidase: activity, location, and release. J. Allergy Clin.
Health Perspect. 88:209–218. Immunol. 96:924–931.
Morgan, K. T., Patterson, D. L., and Gross, E. A. (1986). Ohkubo, K., Baraniuk, J. N., Hohman, R., Merida, M., Hersh, L.
Responses of the nasal mucociliary apparatus to airborne irri- B., and Kaliner, M. A. (1998). Aminopeptidase activity in
tants. In Toxicology of the Nasal Passages, C. S. Barrow (Ed.). human nasal mucosa. J. Allergy Clin. Immunol. 102:741–750.
Hemisphere Publishing Corp., New York, pp. 123–134. Pelosi, P. (1996). Perireceptor events in olfaction. J. Neurobiol.
Morris, J. B. (1997). Uptake of acetaldehyde vapor and aldehyde 30:3–19.
dehydrogenase levels in the upper respiratory tracts of the Peng, H. M., Ding, X., and Coon, M. J. (1993). Isolation and het-
mouse, rat, hamster, and guinea pig. Fundam. Appl. Toxicol. erologous expression of cloned cDNAs for two rabbit nasal
35:91–100. microsomal proteins, CYP2A10 and CYP2A11, that are relat-
Morris, J. B. (2000). Uptake of styrene in the upper respiratory ed to nasal microsomal cytochrome P450 form a. J. Biol. Chem.
tract of the CD mouse and Sprague-Dawley rat. Toxicol. Sci. 268:17253–17260.
54:222–228. Pinto, S., Gallo, O., Polli, G., Boccuzzi, S., Paniccia, R., Brunelli,
Morris, J. B., and Cavanagh, D. G. (1986). Deposition of ethanol T., and Abbate, R. (1997). Cyclooxygenase and lipoxygenase
and acetone vapors in the upper respiratory tract of the rat. metabolite generation in nasal polyps. Prostaglandins Leukot.
Fundam. Appl. Toxicol. 6:78–88. Essent. Fatty Acids 57:533–537.
Mullol, J., Raphael, G. D., Lundgren, J. D., Baraniuk, J. N., Porter, T. D. and Coon, M. J. (1991). Cytochrome P-450.
Merida, M., Shelhamer, J. H., and Kaliner, M. A. (1992). Multiplicity of isoforms, substrates, and catalytic and regula-
Comparison of human nasal mucosal secretion in vivo and in tory mechanisms. J. Biol. Chem. 266:13469–13472.
vitro. J. Allergy Clin. Immunol. 89:584–592. Price, S. (1984). Mechanisms of stimulation of olfactory neurons:
Nebert, D. W. (1990). Drug metabolism. Growth signal pathways. an essay. Chem. Senses 8:341–354.
Nature 347:709–710. Price, S. (1986). Effects of odorant mixtures on olfactory receptor
Nebert, D. W. (1991). Proposed role of drug-metabolizing cells. Ann. NY Acad. Sci. 512:55–60.
enzymes: regulation of steady state levels of the ligands that Proctor, D. F., Adams, G. K., Andersen, I., and Man, S. F. (1978).
effect growth, homeostasis, differentiation, and neuroendocrine Nasal mucociliary clearance in man. Ciba Found. Symp.
functions. Mol. Endocrinol. 5:1203–1214. 54:219–234.
Nef, P., Heldmann, J., Lazard, D., Margalit, T., Jaye, M., Randall, H. W., Bogdanffy, M. S., and Morgan, K. T. (1987).
Hanukoglu, I., and Lancet, D. (1989). Olfactory-specific Enzyme histochemistry of the rat nasal mucosa embedded in
cytochrome P-450: cDNA cloning of a novel neuroepithelial cold glycol methacrylate. Am. J. Anal. 179:10–17.
72 Ding and Dahl

Reed, C. J., Lock, E. A., and De Matteis, (1986). NADPH: during retrograde degeneration and regeneration of the olfac-
cytochrome P-450 reductase in olfactory epithelium. tory nerve. Exp. Neurol. 146:331–340.
Relevance to cytochrome P-450-dependent reactions. Stott, W. T., Dryzga, M. D., and Ramsey, J. C. (1983). Blood-flow
Biochem. 240:585–592. distribution in the mouse. J. Appl. Toxicol. 3:310–312.
Revington, M., Lacroix, J. S., and Potter, E. K. (1997). Su, T., Sheng, J. J., Lipinskas, T. W., and Ding, X. (1996).
Sympathetic and parasympathetic interaction in vascular and Expression of CYP2A genes in rodent and human nasal
secretory control of the nasal mucosa in anaesthetized dogs. J. mucosa. Drug Metab. Disposition 24:884–890.
Physiol. 505:823–831. Su, T., He, W., Gu, J., Lipinskas, T. W., and Ding, X. (1998).
Reznik, G. (1990). Comparative anatomy, physiology, and function Differential xenobiotic induction of CYP2A5 in mouse liver,
of the upper respiratory tract. Environ. Health. Perspect. kidney, lung, and olfactory mucosa. Drug Metab. Disposition
85:171–184. 26:822–824.
Roberts, E. S., Vaz, A. D., and Coon, M. J. (1992). Role of Su, T., Bao, Z., Zhang, Q.-Y, Smith, T. J., Hong, J.-Y., and Ding,
isozymes of rabbit microsomal cytochrome P-450 in the X. (2000). Human CYP2A13: predominant expression in the
metabolism of retinoic acid, retinol, and retinal. Mol. respiratory tract and its high efficiency metabolic activation of
Pharmacol. 41:427–433. a tobacco-specific carcinogen, 4-(methylnitrosamino)-1-
Robinson, A. M., Kern, R. C., Foster, J. D., Krozowski, Z. S., and (3-pyridyl)-1-butanone. Cancer Res. 60:5074–5079.
Pitovski, D. Z. (1999). Mineralocorticoid receptors in the mam- Sun, J. D., Dahl, A. R., Gillett, N. A., Barr, E. B., Crews, M. L.,
malian olfactory mucosa. Ann. Otol. Rhinol. Laryngol. Eidson, A. F., Burt, D. G., Dieter, M. P., and Hobbs, C. H.
108:974–981. (1991). Two-week repeated inhalation exposure of F344/N rats
Rogers, M. E., Jani, M. K., and Vogt, R. G. (1999). An olfactory- and B6C3F1 mice to ferrocene. Fundam. Appl. Toxicol.
specific glutathione-S-transferase in the sphinx moth Manduca 17:150–158.
sexta. J. Exp. Biol. 202:1625–1637. Takahashi, J., Palmer, T. D., and Gage, F. H. (1999). Retinoic acid
Rosenblum, P. M., Sorensen, P. W., Stacey, N. E., and Peter, R. E. and neurotrophins collaborate to regulate neurogenesis in adult-
(1991). Binding of the steroidal pheromone 17a,20b-dihy- derived neural stem cell cultures. J. Neurobiol. 38:65–81.
droxy-4-pregnen-3-one to goldfish (Carassius auratus) olfac- Tamura, H., Miyawaki, A., Inoh, N., Harada, Y., Mikoshiba, K.,
tory epithelium membrane preparations. Chem. Senses and Matsui, M. (1997). High sulfotransferase activity for phe-
16:143–154. nolic aromatic odorants present in the mouse olfactory organ.
Scarborough, P. E., Ma, J. X., Qu, W., and Zeldin, D. C. (1999). Chem. Biol. Interact. 104:1–9.
P450 subfamily CYP2J and their role in the bioactivation of Thornton-Manning, J. R. and Dahl, A. R. (1997). Metabolic capac-
arachidonic acid in extrahepatic tissues. Drug Metab. Rev. ity of nasal tissue interspecies comparisons of xenobiotic-
31:205–234. metabolizing enzymes. Mutat. Res. 380:43–59.
Schultz, E. W., and Gebhardt, L. P. (1934). Olfactory tract and Thornton-Manning, J. R., Nikula, K. J., Hotchkiss, J. A., Avila, K.
poliomyelitis. Proc. Soc. Exp. Biol. Med. 31:728–730. J., Rohrbacher, K. D., Ding, X., and Dahl, A. R. (1997). Nasal
Schwob, J. E., Youngentob, S. L., and Mezza, R. C. (1995). cytochrome P450 2A: identification, regional localization, and
Reconstitution of the rat olfactory epithelium after methyl bro- metabolic activity toward hexamethylphosphoramide, a known
mide-induced lesion. J. Comp. Neurol. 359:15–37. nasal carcinogen. Toxicol. Appl. Pharmacol. 142:22–30.
Shehin-Johnson, S. E., Williams, D. E., Larsen-Su, S., Stresser, D. Thyssen, J., Althoff, L., Kimmerle, G., and Mohr, U. (1981).
M., and Hines, R. N. (1995). Tissue-specific expression of Inhalation studies with benzo(a) pyrene in Syrian golden ham-
flavin-containing monooxygenase (FMO) forms 1 and 2 in the sters. J. Natl. Cancer Inst. 6:575–577.
rabbit. J. Pharmacol. Exp. Therapeutics 272:1293–1299. Tjälve, H., and Henriksson, J. (1999). Uptake of metals in the brain
Sheng, J., Guo, J., Hua, Z., Caggana, M., and Ding, X. (2000). via olfactory pathways. Neurotoxicology 20:181–195.
Characterization of human CYP2G genes: widespread loss- Tomlinson, A. H., and Esiri, M. M. (1983). Herpes simplex
of-function mutations and genetic polymorphism. encephalitis. Immunohistological demonstration of spread of
Pharmacogenetics 10:667–678. virus via olfactory pathways in mice. J. Neurol. Sci.
Shipley, M. T. (1985). Transport of molecules from nose to brain: 60:473–484.
transneuronal anterograde and retrograde labeling in the rat Trapido-Rosenthal, H. G., Carr, W. E. S., and Gleeson, R. A.
olfactory system by wheat germ agglutinin-horseradish peroxi- (1990). Ectonucleotidase activities associated with the olfac-
dase applied to the nasal epithelium. Brain Res. Bull. tory organ of the spiny lobster. J. Neurochem. 55:88–96.
15:129–142. Tricker, A. R., and Preussman, R. (1991). Carcinogenic N-
Sorokin, S. P. (1988). The respiratory system. In Cell and Tissue nitrosamines in the diet: occurrence, formation, mechanisms
Biology: A Textbook of Histology, L. Weiss (Ed.) Uban & and carcinogenic potential. Mutat. Res. 259:277–289.
Schwarzenberg, Baltimore, pp. 753–814. Turk, M. A. M., Flory, W., and Henk, W. G. (1986). Chemical mod-
Starcevic, S. L., and Zielinski, B. S. (1995). Immunohistochemical ulation of 3-methylindole toxicosis in mice: effect of bronchio-
localization of glutathione S-transferase pi in rainbow trout lar and olfactory mucosal injury. Vet. Pathol. 23:563–570.
olfactory receptor neurons. Neurosci. Lett. 183:175–178. Uraih, L. C., and Maronpot, R. R. (1990). Normal histology of the
Starcevic, S. L., and Zielinski, B. S. (1997). Glutathione and glu- nasal cavity and application of special techniques. Environ.
tathione S-transferase in the rainbow trout olfactory mucosa Health Perspect. 85:187–208.
Olfactory Mucosa 73

Vannelli, G. B., and Balboni, G. C. (1982). On the presence of Whitesides J, Hall M, Anchan R, Lamantia AS (1998) Retinoid
estrogen receptors in the olfactory epithelium of the rat. In signaling distinguishes a subpopulation of olfactory receptor
Olfaction and Endocrine Regulation, Breiphol (Ed.). IRL Press, neurons in the developing and adult mouse. J. Comp. Neurol.
London, pp. 279–282. 394:445–461.
Verma, A., Hirsch, D. J., Glatt, C. E., Ronnett, G. V., and Snyder, Zhang, J. H., and Ding, X. (1998). Identification and characteriza-
S. H. (1993). Carbon monoxide: a putative neural messenger. tion of a novel tissue-specific transcriptional activating element
Science 259:381–384. in the 5-flanking region of the CYP2A3 gene predominantly
Vogt, R. G., and Riddiford, L. M. (1981). Pheromone binding and expressed in rat olfactory mucosa. J. Biol. Chem.
inactivation by moth antennae. Nature 293:161–163. 273:23454–23462.
Voigt, J. M., Guengerich, F. P,, Baron, J. (1985). Localization of a Zhang, J. H., Zhang, Q.-Y., Guo, J. C., Zhou, Y. L., and Ding, X.
cytochrome P-450 isozyme (cytochrome P-450 PB-B) and (2000). Identification and functional characterization of a con-
NADPH-cytochrome P-450 reductase in rat nasal mucosa. served, nuclear factor 1-like element in the proximal promoter
Cancer Lett. 27:241–247. region of CYP1A2 gene specifically expressed in the liver and
Voigt, J. M., Guengerich, F. P., and Baron, J. (1993). Localization olfactory mucosa. J. Biol. Chem. 275:8895–8902.
and induction of cytochrome P450 1A1 and aryl hydrocarbon Zhang, Q.-Y. (1999). Retinoic acid biosynthetic activity and
hydroxylase activity in rat nasal mucosa. J. Histochem. retinoid receptors in the olfactory mucosa of adult mice.
Cytochem. 41:877–885. Biochem. Biophys. Res. Commun. 256:346–351.
Walters, E., Buchheit, K., and Maruniak, J. A. (1992). Receptor Zhang, Q.-Y., Ding, X., and Kaminsky, L. S. (1997). cDNA
neuron losses result in decreased cytochrome P450 immunore- cloning, heterologous expression, and characterization of rat
activity in associated non-neuronal cells of mouse olfactory intestinal CYP2J4. Arch. Biochem. Biophys. 340:270–278.
mucosa. J. Neurosci. Res. 33:103–111. Zhang, Q.-Y., Raner, G., Ding, X., Dunbar, D., Coon, M. J., and
Walters, E., Buchheit, K., and Maruniak, J. A. (1993). Olfactory Kaminsky, L. S. (1998). Characterization of the cytochrome
cytochrome P-450 immunoreactivity in mice is altered by P450 CYP2J4 – expression in rat small intestine and role in
dichlobenil but preserved by metyrapone. Toxicology retinoic acid biotransformation from retinal. Arch. Biochem.
81:113–122. Biophys. 353:257–264.
Wang, Q., Hasan, G., and Pikielny, C. W. (1999). Preferential Zhang, Q.-Y., Ding, X., Dunbar, D., Cao, L., and Kaminsky, L. S.
expression of biotransformation enzymes in the olfactory (1999). Induction of rat small intestinal cytochrome P-450 2J4.
organs of Drosophila melanogaster, the antennae. J. Biol. Drug Metab. Disposition 27:1123–1127.
Chem. 274:10309–10315. Zhuo, X., Gu, J., Zhang, Q.-Y., Spink, D. C., Kaminsky, L. S., and
Wardlaw, S. A., Nikula, K. J., Kracko, D. A., Finch, G. L., Ding, X. (1999). Biotransformation of coumarin by rodent and
Thornton-Manning, J. R., and Dahl, A. R. (1998). Effect of cig- human cytochromes P-450: metabolic basis of tissue-selective
arette smoke on CYP1A1, CYP1A2 and CYP2B1/2 of nasal toxicity in olfactory mucosa of rats and mice. J. Pharmacol.
mucosae in F344 rats. Carcinogenesis 19:655–662. Exp. Therapeutics 288:463–471.
Waxman, D. J., Lapenson, D. P., Aoyama, T., Gelboin, H. V., Zielinski, B. S., Getchell, M. L., and Getchell, T. V. (1988).
Gonzalez, F. J., and Korzekwa, K. (1991). Steroid hormone Ultrastructural characteristics of sustentacular cells in control
hydroxylase specificities of eleven cDNA-expressed human and odorant-treated olfactory mucosa of the salamander. Anal.
cytochrome P450s. Arch. Biochem. Biophys. 290:160–166. Rec. 221:769–779.
Weinshilboum, R. M., Otterness, D. M., Aksoy, I. A., Wood, T. C., Zupko, K., Poria, Y., Lancet, D. (1991) Immunolocalization of
Her, C., and Raftogianis, R. B. (1997). Sulfation and sulfo- cytochrome P-450olf1 and P-450olf2 in rat olfactory mucosa.
transferases 1: Sulfotransferase molecular biology: cDNAs and Eur. J. Biochem. 196:51–58.
genes. FASEB J. 11:3–14.
4

Molecular Neurobiology of Olfactory Transduction

Cheil Moon and Gabriele V. Ronnett

The Johns Hopkins University School of Medicine, Baltimore, Maryland, U.S.A.

I. INTRODUCTION (see Chapters 3 and 4) (Graziadei and Monti-Graziadei,

1979; Moulton and Beidler, 1967).
The correct interpretation of sensory information is vital to ORNs are bipolar, extending apical dendrites to the
an organism’s survival. Among sensory modalities, the surface of the neuroepithelium and sending unmyeli-
olfactory system has daunted many investigators seeking nated axons through the basal lamina and cribiform plate
to understand the molecular aspects of its signal trans- of the ethmoid bone to terminate in glomeruli on mitral
duction and coding mechanisms (Buck, 1996; Getchell, and tufted neurons in the olfactory bulb of the brain. The
1986; Getchell et al., 1985). The ability of the olfactory apical dendrites form dendritic knobs from which arise
system to discriminate among thousands of odors specialized, nonmotile cilia, where the initial events of
comprised of chemically divergent structures (odorants) olfactory transduction occur (Getchell, 1986; Labarca
posed unique challenges that have been answered only by and Bacigalupo, 1988; Lowe and Gold, 1993a).
a combination of molecular, electrophysiological, and cell Electrophysiological studies indicate that odorant sensi-
biological approaches. What has emerged is that olfactory tivity and the odorant-induced current are uniformly
transduction combines unique receptive molecules with distributed along the cilia, suggesting that all the compon-
classical transduction cascades to detect olfactory stimuli. ents of the immediate responses to odorants are localized
What is provocative is that many cascades are activated in to the cilia. Immunoelectron microscopic studies have
response to odorant detection whose roles we are only confirmed the cilial localization of many of these compo-
beginning to be appreciated. nents (Menco, 1997; Menco et al., 1992a). ORNs comprise
75–80% of the cells in the epithelium (Farbman, 1992).
They are functionally homogeneous: they all detect odor-
II. CELLULAR ANATOMY OF ants. As they mature, ORNs move apically in the epithelium,
THE OLFACTORY EPITHELIUM permitting determination of neuronal age by position
(Roskams et al., 1998). Mature ORNs express olfactory
The peripheral olfactory system is well adapted structurally marker protein (OMP) (Farbman and Margolis, 1980;
to perform its function. The olfactory primary sensory Margolis, 1980). ORNs senesce throughout life at a
neurons are located in a portion of the olfactory regular rate and are replenished by the differentiation of
epithelium, thus facilitating their direct contact with globose basal cells (Caggiano et al., 1994; Graziadei,
inhaled odorants. There are several principal cell types in 1973; Graziadei and Metcalf, 1971). This neurogenesis
the olfactory epithelium, including olfactory receptor can be hyperinduced by ablation of the olfactory bulb
neurons (ORNs), supporting sustentacular cells, microvil- (termed bulbectomy) (Carr and Farbman, 1993; Costanzo
lar cells, Bowman gland cells, and two types of basal cells and Graziadei, 1983; Hirsch and Margolis, 1980). Thus,

75
76 Moon and Ronnett

understanding of the functions of signaling components and Cagan, 1980). Receptors subsequently couple to a G-
in signal transduction is facilitated by the spatial organi- protein to activate adenylyl cyclase (Pace et al., 1985;
zation of ORNs. Ronnett et al., 1993; Sklar et al., 1986).
Sustentacular cells share features in common with glia. Electrophysiological and biochemical studies confirm that
They stretch from the epithelial surface to the basal lami- cAMP is the key messenger in the initial phase of odorant
na, where they maintain foot processes (Getchell, 1986; detection (Breer et al., 1990; Brunet et al., 1996; Jaworsky
Getchell et al., 1985). Sustentacular cells electrically isol- et al., 1995; Pace et al., 1985; Ronnett et al., 1991; Ronnet
ate ORNs, secrete components into the mucus, and contain and Snyder, 1992; Sklar et al., 1986; Wong et al., 2000).
detoxifying enzymes (Okano, 1974). The sustentacular cAMP levels increase and open a cyclic nucleotide-gated
cells contain high concentrations of cytochrome P450–like channel, resulting in an influx of Na and calcium
enzymes (Lazard et al., 1991). These enzymes may modify (Firestein and Werblin, 1989; Nakamura and Gold, 1987).
odorants to make them less membrane permeable or The immediate response is the generation of a graded
inactivate them. Recent studies indicate that sustentacular receptor potential (Getchell and Shepherd, 1978; Ottoson,
cells may produce growth factors important to ORN devel- 1956). Several other second messenger cascades are acti-
opment (Hansel et al., 2001). Neuropeptide Y (NPY) is an vated upon odorant detection and may regulate secondary
amidated neuropeptide that performs many functions in events or odorant responsivity. The increase in calcium may
mammalian physiology (Baraban et al., 1997; Danger et regulate downstream events (Frings et al., 1995; Kaupp,
al., 1990). NPY mRNA is upregulated following periph- 1991). Odorants also increase phosphoinositide hydrolysis
eral axotomy and in pheochromocytoma and ganglioneu- and the production of inositol-1,4,5-trisphosphate (IP3)
roblastoma tissue (Adrian et al., 1983). Whereas NPY is (Breer and Boekhoff, 1991; Miyamoto et al., 1992; Ronnett
expressed in developing ORNs during embryogenesis, it is et al., 1993; Schandar et al., 1998). Cyclic GMP production
expressed in sustentacular cells in the adult olfactory is also increased with odorant exposure (Ingi et al., 1996;
epithelium. NPY functions as a neuroproliferative factor Verma et al., 1993). Interestingly, the odorant-induced
for olfactory neuronal precursors in vivo and in vitro cGMP response is much slower than the cAMP or IP3
(Hansel et al., 2001). This is the first of potentially many responses, which normally peak within 500 msec. Thus, the
growth factors that sustentacular cells contribute to ORN cGMP response does not appear to function in the immedi-
homeostasis. ate detection phase of olfaction, such as modulating cyclic
The basal cells underlie the ORNs and serve as precur- nucleotide gated cation channels or IP3 receptors, but rather
sors for the generation of new ORNs throughout adulthood in desensitization or the modulation of the cellular response
(Caggiano et al., 1994; Graziadei and Monti Graziadei, during longer exposures to odorants (Breer et al., 1992;
1979; Moulton and Beidler, 1967). Basal cells have been Moon et al., 1998, 1999; Zufall and Leinders-Zufall, 1997).
divided into two general classes. Horizontal cells are flat These messengers are discussed in subsequent sections.
and express cytokeratin (Calof and Chikaraishi, 1989;
Graziadei and Monti-Graziadei, 1979). Globose basal cells
are rounded in shape and express several markers, includ- IV. ODORANT-BINDING PROTEINS
ing GBC-1, GBC-3, and GBC-5 (Goldstein and Schwob,
1996; Huard et al., 1998). Compared to other neurons, The existence of carrier proteins for odorants in the nasal
many ORNs have a relatively short survival time, in the mucus was predicted based upon the fact that hydrophobic
range of several months. This may be due to the fact that odorants must travel through the aqueous mucus barrier
ORNs are exposed to a variety of toxic or infectious towards the cilia of ORNs. In fact, odorant-binding pro-
agents. Thus, the function of globose basal cells in provid- teins (OBPs) were discovered by several laboratories in
ing new ORNs is crucial to the maintenance of the sense of early attempts to identify odorant receptors using radio-
smell. These issues are discussed in greater detail in active odorants such as 3-isobutyl-2-methyloxypyrazine
Chapter 4. (Pelosi et al., 1982; Pevsner et al., 1986; Pevsner et al.,
1985). Purified OBP is a homodimer comprised of two
identical 19 kDa subunits and binds to odorants with
III. GENERAL MECHANISMS OF affinities in the micromolar range (Pevsner et al., 1990).
ODORANT TRANSDUCTION The molecular cloning of OBP helped to clarified its
function. OBP is a member of the lipophilic molecule
Olfactory signal transduction (Fig. 1) is initiated when carrier protein family. A well-characterized member of
odorants interact with specific receptors on cilia on ORNs this family is a retinol-binding protein. This protein con-
(Buck, 1996; Dwyer et al., 1998; Malnic et al., 1999; Rhein veys retinol from retinal pigment epithelium to rods and
Molecular Neurobiology of Olfactory Transduction 77

Figure 1 Model of odorant signal transduction (see text for details). Signaling cascades mediate the initial phase of odorant detection
and potential long-term responses to odorant detection. Abbreviations: AC, adenylyl cyclase; bARK, beta-adrenergic receptor kinase;
bARR, beta-adrenergic receptor arrestin; CaM, calmodulin, CO, carbon monoxide; CREB, camp-responsive element binding protein;
GCAP, guanylyl cyclase activating protein; Golf, olfactory G protein; Gq, G protein q; HO, heme oxygenase; InsP3, inositol-1,4,5 phos-
phates; InsP3R, InsP3 receptor; MAPK, a mitogen-activated protein kinase, MEK, MAP or ERK kinase; OBP, odorant binding protein;
oCNC, olfactory cyclic neucleotide-gated channel; OR, odorant receptor; PDE, phosphodiesterase; pGC, particulate guanylyl cyclase;
PKA, cAMP-dependent protein kinase; PKG, cGMP-dependent protein kinase; PLC, phospholipase C; Raf, MEK kinase; RSK, 90 kDa
ribosomal S6 kinase; sGC, soluble guanylyl cyclase.

cones where it is incorporated into rhodopsin (Heller, Further studies have revealed that multiple forms of
1975). In situ hybridization studies of OBP mRNA in rats OBP may be expressed in the nasal epithelium. Rabbitts
revealed its selective concentration in the lateral nasal and colleagues (Dear et al., 1991) identified a second form
gland, the largest of 20 discrete nasal glands (Pevsner of OBP, OBPII. OBPII encodes a secretory protein with
et al., 1988). OBP thus appears to be secreted from this significant homology to OBPI, and it is also expressed in
gland down a long duct to the tip of the nose, where the lateral nasal gland, which is the site of OBP expression.
watery secretions are atomized in order to humidify Interestingly, the OBPII sequence also shows significant
inspired air. OBP thus localized might trap odorants and homology to the VEG protein, which is thought to be
carry them with inspiration to ORNs. Alternatively, OBP involved in taste transduction (Burova et al., 2000). Breer
may function to remove odorants from sensory and colleagues demonstrated that rat OBPI and OBPII
epithelium and cilia. contain distinct ligand specificities (Lobel et al., 1998).
78 Moon and Ronnett

Recombinant OBP proteins appear to share many struc- odorant receptor proteins. Polyclonal antibodies have been
tural features, but each has been shown to interact with dis- raised against some odorant receptors, permitting visualiza-
tinct sets of odorants. OBPI binds specifically to a pyrazine tion of odorant receptor proteins. In rats, an odorant receptor
derivative, 2-isobutyl-3-methoxypyrazine, whereas OBPII is expressed as early as E14 in a zonally restricted pattern
binds to the chromophore, 1-anilinonaphthalene 8-sulfonic (Koshimoto et al., 1994). The expression of odorant recep-
acid (1,8-ANS), specifically. In other vertebrates, multiple tors is restricted to the cilia and dendritic knobs of ORNs.
forms of OBP have been identified. There are four OBPs The cilia-specific expression of odorant receptors support-
in mice (Pes and Pelosi, 1995), three OBPs in rabbit ed a role for odorant receptors in olfactory transduction
(Garibotti et al., 1997), and two OBPs in cow (Bianchet (Menco et al., 1997a, b, c). A concern with studies utilizing
et al., 1996; Dal Monte et al., 1991). OBP has also been antibodies to identify discrete members of the odorant
cloned from insects (Vogt et al., 1990, 1991). receptor family is the specificity of the antibodies, given the
large numbers of receptors. Despite the general utility of
antisera for immunohistochemical and biochemical studies,
V. ODORANT RECEPTORS the enormous size of the odorant receptor repertoire limits
the feasibility of proving the specificity of an antibody to a
Mammals perceive a huge variety of environmental odors. specific receptor.
The initial step in odor perception requires the interaction Significant difficulties with heterologous expression of
of odorous ligands with specific receptors on the surface of odorant receptors severely limited studies designed to pro-
olfactory receptor neurons (Buck, 1996; Dwyer et al., vide functional confirmation of the role of such receptors.
1998; Malnic et al., 1999; Rhein and Cagan, 1980). Based The most convincing data concerning function were pro-
upon the assumption derived from biochemical evidence vided initially by genetic studies in Caenorhabditis ele-
that odorant signal transduction involved G proteins, and gans (Senhupta et al., 1996), which demonstrated that the
thus G protein–coupled receptors, a very large gene family odor 10 mutant lacked a seven transmembrane receptor
of closely related olfactory-specific seven transmembrane and was deficient in its ability to detect acetyl (Senhupta et
spanning domain receptors was identified by polymerase al., 1996). Krautwurst et al., (1998) first achieved func-
chain reaction (PCR) (Buck and Axel, 1991; Buck, 1992, tional heterologous expression of odorant receptors using
1996). In vertebrates, the family of odorant receptors HEK-293 cells. This group (Krautwurst et al., 1998) gen-
(ORs) is known to encode as many as 1000 genes, sug- erated an expression library of mouse odorant receptors
gesting that the first steps of odorant recognition are and identified three receptors responding to carvone,
accomplished within the primary sensory neurons them- ()citronellal, and limonene using micromolar concentra-
selves. To date, odorant receptor genes have been isolated tions of these odorants. Firestein and colleagues also
from 12 vertebrate species: rat, mouse, human, catfish, demonstrated functional expression of a cloned odorant
zebrafish, dog, frog, chicken, pig, opossum, mudpuppy, receptor in rat nasal epithelium by using a recombinant
and lamprey (Mombaerts, 1999a). In humans, estimates of adenovirus containing a putative odorant receptor to infect
the size of the receptor family range from 500 to 1000 rat nasal epithelium in vivo (Zhao et al., 1998). They
genes. Compared to the other species, human odorant demonstrated that this specific odorant receptor was over-
receptor clones display a high frequency of pseudogenes expressed in the rat olfactory epithelium and the expressed
(Mombaerts, 1999b). odorant receptor transduced a response to a small subset of
The expression pattern of odorant receptors in ORNs of odorants by EOG. Malnic et al., (1999) performed single
the olfactory epithelium has an unusual spatial distribution cell PCR on ORNs whose odorant responses had been
(Ressier et al., 1993; Vassar, et al., 1993). In situ hybridiza- determined to demonstrate that a combinatorial code exists
tion studies have shown that odorant receptor mRNAs are for odorant perception. These approaches to develop func-
expressed within one of several broad, nonoverlapping tional expression systems for odorant receptors can be
zones. Within a zone, odorant receptors are expressed in a extremely useful to screen odorant receptors on a large
random manner. Each zone occupies about a quarter of the scale as well as to understand the molecular mechanism of
olfactory epithelium (Ressier et al., 1993) and is repre- odorant recognition.
sented on the turbinates and on the septum (Mombaerts, Besides functioning in the detection of odorants, odor-
1999a). However, the physiological meaning of zonal ant receptors are hypothesized to be involved in determin-
expression remains unclear. ing or guiding ORN axonal projections to the olfactory
While a number of studies have been done on the bulb and possibly to specific glomeruli (Mombaerts et al.,
expression and distribution of odorant receptors at the mes- 1996; Ressler et al., 1994). In rodents, the axons of ORNs
sage level, relatively little is known about the expression of expressing the same odorant receptors converge onto
Molecular Neurobiology of Olfactory Transduction 79

defined glomeruli in the olfactory bulb, suggesting that the (Firestein et al., 1990; Firestein and Werblin, 1989)
rodent olfactory bulb is topographically organized and, in suggested that the latency of the odorant response (several
turn, that ORN expressing a specific odorant receptor pro- hundred milliseconds) indeed supported a role for a second
jects to and forms a synapse with the representing messenger such as cAMP. The first direct biochemical stud-
glomeruli in the olfactory bulb. This represents an interest- ies reported an odorant-induced cAMP response in olfacto-
ing hypothesis that an environmental odor is encoded by ry sensory cilia isolated from both frog and rat (Pace et al.,
activation of specific glomeruli that perceive a signal from 1985; Sklar et al., 1986). The olfactory sensory cilia were
ORNs expressing a specific odorant receptor out of the prepared by subcellular fractionation after calcium-shock
odorant receptor repertoire. of the olfactory epithelium (Rhein and Cagan, 1983). The
odorant-stimulated production of cAMP was tissue-specif-
ic and occurred only in the presence of GTP, suggesting the
VI. G-PROTEINS
involvement of receptors coupled to G-proteins. Further
characterization using isolated rat olfactory sensory cilia
The first evidence for the involvement of G-proteins in
showed that cAMP was best produced by fruity, floral, and
odorant transduction was provided by the observation that
herbaceous odors (Nakamura and Gold, 1987; Sklar et al.,
the odorant-induced activation of olfactory sensory cilia
1986). Screening many odorants at a single concentration
depended upon the presence of GTP (Rhein and Cagan,
revealed only minimal cAMP production by some, generat-
1983). Subsequently, a G-protein was cloned from an
ing the hypothesis that those odorants with small or absent
olfactory cDNA library that was highly and almost exclu-
cAMP responses employed another cascade, perhaps inos-
sively expressed in ORNs; this G-protein was named Golf
itol phosphates (Nakamura and Gold, 1987; Sklar et al.,
(Jones and Reed, 1987). Golf was shown to stimulate
1986). These initial measurements were made 15 minutes
adenylyl cyclase in heterologous systems. Aside from
after the exposure of isolated cilia to odorants.
its expression in ORNs of the olfactory epithelium, Golf is
To demonstrate that the production of cAMP occurs on
expressed in basal ganglia (Drinnan et al., 1991). As men-
a relevant time scale, subsecond kinetics of odorant-
tioned, odorants also increase IP3 production, causing
induced changes were analyzed by using a rapid quench-
many to postulate that cilia might contain olfactory-specif-
flow device (Boekhoff et al., 1990; Breer et al., 1990). In
ic Gq-proteins. To date, these have not been reported.
this device, cilia membranes and odorant solutions were
Mice with targeted disruption of the gene for Golf dis-
subjected to computer-controlled mixing, with subsequent
played a striking reduction in the electrophysiological
quenching of samples at intervals from 8 to 500 msec.
response of ORNs to a wide variety of odors, supporting
cAMP was produced rapidly and transiently in response to
the hypothesis that Golf, and thus this G-protein–mediated
odorants, with increases evident as early as 25 msec.
cascade, is required for odorant signal transduction
Certain odorants such as fruity odors were able to stimu-
(Belluscio et al., 1998). Despite this intense attenuation in
late cAMP production at concentrations as low as 10 nM,
response to odors, the topographic map of ORN pro-
whereas others such as putrid odors had no effect, even at
jections to the olfactory bulb was unaltered in Golf-deficient
higher concentrations. Those odorants that did not stimu-
mice. Thus, odorant stimulation may not be an essential
late cAMP production were hypothesized to act through
process in determining the targets of ORN axonal pro-
the phosphoinositide (PI) cycle. High (millimolar) levels
jections to the olfactory bulb. However, these studies may
of calcium inhibited the response, but intermediate con-
need to be done at higher resolution.
centration ranges were not tested.
The odorant-induced cAMP response was investigated
further using isolated rat olfactory cilia to determine the
VII. SECOND MESSENGERS
generality of the odorant-induced cAMP response and the
A. cAMP calcium dependence of this response (Jaworsky et al.,
1995). Odorants indeed cause rapid and transient eleva-
Electrophysiological studies provided some of the first evi- tions of cAMP, as well as the more sustained signal, as
dence for the central role of cAMP in odorant detection. seen by Lancet (Pace et al., 1985) and Sklar (Sklar et al.,
Patch clamp experiments on cilia demonstrated a cAMP- 1986). Different from the observation from Breer’s group
gated conductance (Nakamura and Gold, 1987). (Boekhoff et al., 1990; Breer et al., 1990), all odorants
Investigators proposed that an odorant would increase stimulated cAMP production. Interestingly, responses
cyclic nucleotide levels to gate a cationic conductance, were non-linear. Basal and odorant-induced cAMP levels
initiating a depolarizing response. Kinetic studies of odor- in cilia demonstrated biphasic calcium dependence, with
ant-induced currents using whole patch-clamp techniques peak cAMP stimulation in the range of 1–10 M free
80 Moon and Ronnett

calcium. Dose-response curves done at two calcium levels Cooper, 1995). Although AC3 is highly enriched in ORNs,
showed that the influence of calcium on odor responses other adenylyl cyclases, such as AC2 or AC4, have also
was complex, suggesting the possible involvement of cal- been associated with olfactory neuroepithelium, raising the
cium both in signal generation and termination. issue that other adenylyl cyclases may be important in dif-
To evaluate olfactory signal transduction in intact cells, ferent aspects of olfactory signal transduction. These other
primary cultures of olfactory epithelium enriched in ORNs adenylyl cyclases may function in other aspects of ORN
were developed (Ronnett et al., 1991a, b, 1993). Using this homeostasis or signaling.
primary culture system, cAMP responses to odorant stim- The various adenylyl cyclases are regulated by different
ulation were monitored in intact ORNs. Odorants were mechanisms. Studies by Storm and colleagues (Choi et al.
quite potent at producing cAMP, with as little as 0.1 nM 1992, 1993; Wayman, 1995) indicated that the mech-
isobutylmethoxypyrazine (IBMP) generating a response anisms of regulation of adenylyl cyclases may not only be
(Ronnett et al., 1991b, 1993). Responses were multiphasic; dependent upon the specific kind of adenylyl cyclase
cAMP production increased with increasing odorant con- expressed in a tissue, but by local influences and the
centration, decreased at even higher odorant concentra- expression of regulatory molecules in that specific cell.
tions, and sometimes reappeared at still higher (1–10 mM) Thus, while ectopically expressed AC3 may be stimulated
concentrations. Signals were calcium dependent, with by calcium, in vivo studies in certain tissues argue for the
maximal activity at 10 M free calcium and inhibition at inhibition of AC3 by calcium. Equally diverse are the
higher calcium concentrations. Odorant induction of effects of protein kinases on adenylyl cyclases. Phorbol
cAMP production was rapid, with peak effects observed at esters are used to mimic the effects of protein kinase C
10–15 sec, but signals continued well above baseline for (PKC) activation and elicit a stimulatory effect on AC2,
minutes, confirming results from Sklar et al. (1986) and while barely stimulating AC1 or AC8. These latter adeny-
Pace and Lancet (1986). lyl cyclases are stimulated up to 8 times by calcium
Cyclic AMP is produced by adenylyl cyclase. There are (Cooper et al., 1995). Frings (1993) has reported that
at least nine identified isoforms of adenylyl cyclases activation of PKC by phorbol esters increased cAMP in
(Hanoune and Defer, 2001). A novel adenylyl cyclase, frog olfactory tissue. The stimulation by calcium of AC1,
referred to as type III AC (AC3), was cloned by Bakalyar AC3, and AC8 is mediated by calmodulin (Tang and
and Reed (1990). Northern blot analysis indicated that Gilman, 1992); it is unclear how the calcium sensitivity of
AC3 mRNA was enriched in the olfactory epithelium and the calcium inhibition of AC5 and AC6 are achieved. There
that AC3 mRNA disappeared after bulbectomy. When is also evidence that PKA may affect adenylyl cyclase
expressed in HEK293 cells, AC3 had almost no basal activity.
activity. In contrast, AC1 and AC2 have high basal acti- The elevation of cAMP results in the gating of the
vities. Golf and AC3 have been ultrastructurally localized to olfactory cyclic nucleotide-gated channel, OCNC, to
olfactory cilia, indicating that Golf may mediate the activa- depolarize ORNs, and thus OCNC, is an integral compo-
tion of AC3 (Menco et al., 1992b). nent in olfactory transduction (Zufall et al., 1994). The
To evaluate the role of AC3 in the olfactory trans- OCNC is a nonspecific cation channel, activated by both
duction, the AC3 gene has been disrupted in mice (Wong cAMP and cGMP (Zagotta, 1996). Three OCNC subunits
et al., 2000). Odorant-induced responses measured by are expressed in the olfactory epithelium: the OCNC1 and
electro-olfactogram (EOG) were completely eliminated in OCNC2 subunits and an olfactory-enriched splice variant
AC3-null mice. Moreover, odor-dependent learning was of the rod photoreceptor  subunit (OCNC) (Bonigk
impaired in these mice. Interestingly, both fruity odors et al., 1999; Bradley et al., 1994; Dhallan et al., 1990;
(transduced by cAMP) and putrid odors (formally thought Liman and Buck, 1994; Sautter et al., 1998). OCNC1
to act through IP3) failed to evoke any response in these forms functional homodimers in vitro (Dhallan 1990) but
animals. This observation was mimicked by a pharma- in vivo is thought to be associated with either the OCNC
cological study that showed that adenylyl cyclase antagonists and or OCNC2. The association with these subunits con-
reversibly inhibit EOG responses, even to putrid odors fers greater sensitivity to cyclic nucleotides and changes in
(Chen et al., 2000). Taken together, these results confirmed single channel kinetics (Bonigk et al., 1999; Bradley et al.,
earlier biochemical studies that implicated adenylyl 1994; Liman and Buck, 1994; Sautter et al., 1998). Ngai
cyclase and cAMP as essential for the initial phases of and colleagues generated OCNC1-deficient mice and
odorant transduction. IP3 was therefore postulated to play reported that these mice were anosmic and died within a
more of a modulatory role in the odorant transduction. few days after birth (Brunet et al., 1996). Later, Parent
Certain enzymes are rather broadly expressed, while et al. (1998) developed a method to promote the survival of
others are restricted in their distribution (Mons and OCNC1-null mice to permit further analysis. Similar to
Molecular Neurobiology of Olfactory Transduction 81

AC3-null mice, the OCNC1-deficient mice showed no primary second messenger required for the initial events of
EOG responses to both fruity and putrid odorants, even to odor detection and cellular depolarization, while IP3 is
complex odorants such as urine. This again suggests that involved in secondary responses, such as adaptation or
cAMP is the essential messenger for odorant transduction. activity-driven cellular responses, not EOG generation.
Using immunohistochemistry, IP3 receptors have been
B. Inositol-1,4,5-trisphosphate localized to the ciliary surface membrane (Cunningham et al.,
1993), positioning IP3 to trigger the influx of extracellular
In the brain and peripheral tissues, receptor-mediated calcium. There is also evidence for plasma membrane
stimulation of phospholipase C (PLC) generates IP3, IP3-sensitive channels in lobster ORNs (Fadool and Ache,
which releases calcium from endoplasmic reticulum (ER) 1992; Munger et al., 2000). Kalinoski and colleagues
stores by binding to specific IP3 receptors (Berridge and (1992) have also demonstrated an IP3-like receptor in
Irvine, 1984, 1989). Plasma membrane IP3 receptors have isolated catfish cilia, although its micromolar Kd for IP3
been identified in lymphocytes (Kuno and Gardner, 1987) suggests a different form of IP3 receptor (Kalinoski et al.,
and neurons (Bush et al., 1994; Fijimoto et al., 1992) to 1992). Several PLC isoforms are demonstrated in olfactory
permit calcium entry from extracellular sources. Thus, cal- epithelium (Abogadie et al., 1995; Bruch et al., 1995;
cium becomes available to modulate enzyme activities. Munger et al., 2000).
There are now five families of IP3 receptors (Joseph, 1996; Reconciliation of the data thus far obtained for IP3 will
Taylor and Richardson, 1991; Taylor and Traynor, 1995). require further work. For over 10 years, debate existed as
Studies in number of species implicate IP3 in olfaction. to whether cGMP or calcium was the visual second mes-
However, electrophysiological experiments have in many senger (Zuker, 1996). We now know that while cGMP is
cases failed to demonstrate a role for IP3 in ORN depolar- central, calcium is the major modulator of cGMP levels
ization. Huque and Brunch (1986) showed PLC activity in (Coccia and Cote, 1994; Mitchell et al., 1995; Somlyo and
isolated catfish olfactory cilia. Restrepo et al., (1990) Walz, 1995; Udovichenko et al., 1994). Additionally, there
showed that amino acids enhanced calcium flux in isolated are striking interspecies differences: while IP3 is important
catfish ORNs. Utilizing the rapid mixing technique, Breer in amphibian phototransduction, no role has thus far been
and colleagues (1990) demonstrated increases in IP3 levels found in mammals. Olfaction may have similar
in response to some odorants. complexities.
Studies in primary cultures of ORNs confirmed that
odorants stimulate the production of IP3. Exposure of cells C. cGMP
to low nanomolar concentrations of odorants resulted in
IP3 formation (Ronnett et al., 1993; Wood et al., 1990). All Cyclic GMP is the primary second messenger in visual
odorants were found to stimulate cAMP and IP3 produc- signal transduction. A number of studies indicate that
tion in primary culture, although with different potencies, cGMP may play an important role in the olfactory trans-
suggesting interactions with different receptors. These duction. Odorants augment cGMP levels in olfactory tis-
responses were very sensitive to ambient calcium and odor sues (Breer et al., 1992) and ORNs (Verma et al., 1993).
concentrations. The enhancement by single odors of both Compared to the odorant-induced increase in cAMP and
cAMP and IP3 production affords a mechanism for IP3 levels, the rise in cGMP levels occurred with a slower,
increased specificity of odor detection. However, these sustained time course. These kinetics suggested that cGMP
studies were only performed at longer (1 sec and beyond) may not be involved in initial signaling events, but rather
times after odor encounter. Ache and coworkers confirmed in long-term cellular events such as desensitization
that odors differentially stimulate dual pathways in isol- (Leinders-Zufall et al., 1996), or in the activation of neur-
ated lobster antennules (Boekhoff et al., 1994). Odors onal activity–dependent transcription (Moon et al., 1999).
elevated cAMP and IP3 in the outer dendritic membranes cGMP levels are regulated by two distinct classes of
of lobster in vitro. IP3 carried the stimulatory current, guanylyl cyclases: soluble guanylyl cyclase and particulate
while cAMP was inhibitory, providing a mechanism for guanylyl cyclase. Soluble guanylyl cyclase is activated by
fine-tuning of responses. gaseous messengers such as NO or CO, whereas particu-
The relevance of IP3 to mammalian olfaction has been late guanylyl cyclase is activated by specific extracellular
questioned by several groups, whose knock-outs affecting ligands or calcium. Both guanylyl cyclases are expressed
the cAMP signaling cascade failed to generate an EOG for in ORNs, implying a complex regulation of cGMP levels
any odor, suggesting that cAMP is the sole odorant-gener- in olfaction (Moon et al., 1998; Verma et al., 1993).
ated second messenger (Brunet et al., 1996). These dis- Diffusible gaseous messenger molecules such as NO or
crepancies may be reconciled if cAMP is indeed the CO can stimulate soluble guanylyl cyclase by binding to
82 Moon and Ronnett

the heme group in soluble guanylyl cyclases (Snyder, GC-G (Simpson, Moon, and Ronnett, unpublished data).
1994). NO and CO are produced by NO synthase (NOS) GC-B is highly expressed throughout the epithelium.
and heme oxygenase (HO), respectively. In ORNs, NOS is These guanylyl cyclases are stimulated by specific nautri-
expressed at embryonic stages and is markedly reduced at uretic peptides. At present, the role and the regulation of
early postnatal stage, whereas HO is highly expressed after these guanylyl cyclases in the olfactory system are unclear.
birth (Ingi and Ronnett, 1995; Roskams et al., 1994). Recent studies have identified odorant-responsive
These data suggest that NO plays an important role during particulate guanylyl cyclases in rat olfactory cilia (Moon
development, whereas HO functions in mature ORNs. Two et al., 1998). At least two particulate guanylyl cyclases
forms of HO have been identified: HO-1 and HO-2. HO-1 exist in cilia, a low Km and a high Km isoform (Moon et al.,
is a heat shock protein (hsp-32) induced by heme, heavy 1998). Odorants were shown to elevate cGMP levels in
metals, stress, or hormones (Bauer et al., 1998; Beschorner cultured ORNs (Ingi and Ronnett, 1995) and in isolated
et al., 2000; Ewing et al., 1994; Hirata et al., 2000; olfactory cilia (Moon et al., 1998) in a calcium-dependent
Koistinaho et al., 1996; Kutty and Maines, 1989) and is manner. A number of experiments suggested that calcium
highly expressed in the spleen and liver, where it is respon- plays a role in odorant transduction and can fluctuate upon
sible for the destruction of heme from red blood cells. HO-1 odorant exposure (Dhallan et al., 1990; Hatt and Ache,
is present in rodent olfactory epithelium (J. Chen, C. 1994; Yau, 1994). Hence, it was hypothesized that an
Moon, and G.V. Ronnett, unpublished data), but its role olfactory particulate guanylyl cyclase could be regulated
and function in olfactory transduction are unclear. by a calcium-binding protein, such as guanylyl
HO-2 is not inducible and is distributed throughout the cyclase–activating protein (GCAP), similar to that found in
body. HO-2 is highly expressed in the brain, especially in the visual transduction pathway. In fact, immunohisto-
neurons of the olfactory epithelium and in the neuronal and chemical studies using anti-GCAP1 antibodies revealed
granule cell layer of the olfactory bulb. In situ hybridiza- that GCAP1 was highly localized to the olfactory cilia
tion analysis showed that guanylyl cyclase and HO-2 are (Moon et al., 1998). Moreover, GCAP1 regulated the odor-
found in ORNs (Verma et al., 1993). Incubation of ORNs ant-induced cGMP response in isolated rat olfactory cilia
with the HO inhibitor, zinc protoporphyrin-9 (Zn PP-9), in a calcium-dependent manner (Moon et al., 1998). Thus,
lowered cGMP levels in ORNs (Ingi and Ronnett, 1995). ORNs contain multiple cGMP pathways that mediate
In addition, odorants augment cGMP levels in ORNs (Ingi delayed and sustained cGMP responses to odorants.
and Ronnett, 1995; Verma et al., 1993). This odorant-
induced cGMP increase could be inhibited by Zn PP-9, but D. Olfactory Phosphodiesterases
not by a NOS inhibitor. Interestingly, the inhibition of HO
could not entirely deplete cGMP levels in ORNs, suggest- The ambient level of cAMP in a cell is dependent upon
ing that particulate guanylyl cyclases may also contribute both the synthesis and degradation of cAMP. Although
to cGMP production in ORNs (Ingi and Ronnett, 1995). odorants clearly activate adenylyl cyclase, is there any
Exposure of isolated cilia derived from olfactory receptor effect of odorants on phosphodiesterases (PDEs)? There
neurons to various odorants increased cGMP levels (Moon are at least seven different gene families of PDEs whose
et al., 1998). Thus, there was a strong suspicion that both activities are regulated by calcium, cyclic nucleotides, and
soluble and particulate guanylyl cyclases have significant phosphorylation (Beavo, 1995; Beavo et al., 1994,
roles in olfactory signal transduction. Beltman et al., 1993; Burns et al., 1996). Thus, odorants
The observation that the inhibition of HO in ORNs could have an indirect effect on the degradation of cAMP,
could not totally block the cGMP response suggested the thus potentially providing a second site of regulation for
involvement of particulate guanylyl cyclases in olfactory the odorant-induced cAMP response. Several forms of
transduction. The fact that an NO donor and soluble cAMP-PDE are expressed in rat olfactory cilia (Borisy
guanylyl cyclase activator, sodium nitroprusside, could not et al., 1991, 1993). A novel calcium/calmodulin PDE
alter the cGMP levels in isolated cilia supported the idea (CaM-PDE) is selectively found in ORNs, with prominent
that the particulate guanylyl cyclases might play a role in cilial expression. This novel CaM-PDE has a high affinity
olfactory cilia. An olfactory specific particulate guanylyl (Km 1.4 M) for cAMP and could be activated by
cyclase, guanylyl cyclase-D (GC-D), has been identified in odorants in response to intracilial calcium increases.
olfactory epithelium (Fulle et al., 1995). GC-D has been Cloning of the high-affinity PDE revealed it to have a
suggested to function as the receptor of sensory neurons to higher affinity for cAMP than any known brain isoform
specific odors. Other members of the particulate guanylyl (Yan et al., 1995). This PDE, designated PDE1C2, is well
cyclase family that are expressed in the olfactory epithe- suited for restoring the submicromolar levels of cAMP
lium have been identified by RT-PCR: GC-A, GC-B, and after odorant stimulation. In an ectopic expression system,
Molecular Neurobiology of Olfactory Transduction 83

maximum activation by calcium was reached at 10M and Shibuya, 1990). Calcium/calmodulin can also affect
calcium concentration. CNG channel activity (Kurahashi and Yau, 1993).
A subset of olfactory neurons expresses cGMP-stimul- Neurocalcin, a calcium-binding protein with three EF
ated phosphodiesterase (PDE2) (Juilfs et al., 1997). In these hand motifs, is also expressed in the rat olfactory epithelium
specific ORNs, GC-D is also expressed, suggesting that (lino et al., 1995). Neurocalcin is localized to ORNs and dis-
GC-D may play an important role in odorant transduction tributed in the cytoplasm, where it is associated with outer
for a specific subset of responses. PDE2 and GC-D are both mitochondrial membrane, endoplasmic reticulum, and axon
expressed in olfactory cilia of these neurons; however, only fibers. The intracellular distribution of neurocalcin in ORNs
PDE2 is expressed in axons (Juilfs et al., 1997). In contrast suggests that this protein may participate in cytoskeletal
to most other ORNs, these neurons appear to project to a arrangement in ORNs. The expression of neurocalcin in
distinct group of glomeruli in the olfactory bulb similar to postnatal development was also studied (Bastianelli et al.,
the subset that have been termed necklace glomeruli. 1995). Neurocalcin showed a gradient of expression pattern
Furthermore, this subset of neurons are unique in that they descending from the central to the lateral areas in the nasal
do not contain several of the previously identified compon- cavity during childhood, and this expression pattern became
ents of olfactory signal transduction cascades involving identical to the adult profile after 20 days.
cAMP and calcium, including a calcium/calmodulin- Additional calcium-binding proteins have been
dependent PDE (PDE1C2), AC3, and cAMP-specific PDE described. A 26 kDa calcium-binding protein named p26olf
(PDE4A) (Juilfs et al., 1997; Meyer et al., 2000). was identified from the frog olfactory epithelium (Miwa et
Interestingly, these latter three proteins are expressed in the al., 1998). p26olf consists of two S-100–like regions and is
same neurons; however, their subcellular distributions are localized to the cilia layer of the olfactory epithelium, sug-
distinct. PDE1C2 and AC3 are expressed almost gesting that p26olf is a dimeric form of S-100 proteins and
exclusively in the olfactory cilia, whereas PDE4A is pre- may be involved in the olfactory transduction or adaptation.
sent only in the cell bodies and axons. Taken together, these Visinin-like protein (VILIP), a member of the neuronal
data strongly suggest that selective compartmentalization subfamily of EF-hand calcium-sensor proteins, was found
of different PDEs and cyclases is an important feature for to be expressed in ORNs of the rat olfactory epithelium
the regulation of signal transduction in ORNs. (Boekhoff et al., 1997). VILIP is localized prominently to
cilia and dendritic knobs. In vitro recombinant VILIP
E. Calcium attenuates odorant-induced cAMP formation in a calcium-
dependent manner. The observation that VILIP does not
Calcium regulates diverse cellular functions, and in general interfere with odorant-induced receptor desensitization and
these functions are mediated by specific calcium-binding pro- that VILIP inhibits the forskolin-induced cAMP production
teins (Baimbridge et al., 1992). Odorant stimulation of ORNs suggests that VILIP may directly affect adenylyl cyclases
results in a calcium influx, which in turn can modulate a and in turn may play a role in adaptation of ORNs.
number of transduction pathways. Calmodulin and other cal- A GCAP1-like calcium-binding protein is present in rat
cium-binding proteins may participate in the processing of olfactory cilia (Moon et al., 1998). GCAP1 was initially
olfactory information. Therefore, study of the calcium-bind- purified and later cloned from bovine retina by Palczewski
ing proteins may provide important background about the and colleagues (1994). GCAP1 is a 21 kDa cytosolic EF-
complex signal transduction pathway involved in olfaction. hand family protein and is proposed to function as a
Olfactory tissues contains various calcium-binding proteins: photoreceptor-specific calcium-binding protein to activate
calmodulin, calretinin, calbindin-D28k, neurocalcin, recov- particulate guanylyl cyclase, thus restoring cGMP level in
erin (Bastianelli et al., 1995). Another calcium-binding pro- light-activated photoreceptor cells. Immunohistochemical
tein, S-100, is restricted to glial cells, primarily around the studies using anti-GCAP1 antibodies revealed the presence
cribiform plate. of GCAP1 in rat olfactory cilia (Moon et al., 1998).
Calmodulin is expressed in olfactory cilia at a concen- Interestingly, purified GCAP1 potentiated cGMP produc-
tration of about 1 µM (Anholt and Rivers, 1990). The odor- tion at high calcium concentrations in isolated rat olfactory
ant-induced intracellular elevation of calcium is thought to cilia (Moon et al., 1998). In photoreceptor cells, GCAP1
promote adaptation because calcium/calmodulin can reduce activates particulate guanylyl cyclase when intracellular cal-
the affinity of the CNG channel for cAMP by 20-fold (Chen cium level is low. The size of the olfactory GCAP (19 kDa)
and Yau, 1994; Hsu and Molday, 1993). Extracellular was not identical to the retinal GCAP1. Thus, the olfactory
calcium is absolutely required for the decay phase of the GCAP is referred to as a GCAP1-like protein. The cloning
odorant-induced whole cell current, which in the absence of of the olfactory GCAP will answer the precise function and
extracellular calcium remains at a steady state (Kurahashi mechanism of the olfactory guanylyl GCAP in olfaction.
84 Moon and Ronnett

A novel calcium-binding protein was recognized in the antibodies to ARK-2 and ARR-2 increased the absolute
olfactory tissues by using R2D5, a mouse monoclonal anti- levels of odorant-induced cAMP as much as four fold and
body that labels rabbit olfactory receptor neurons (Nemoto completely blocked desensitization. Later mice targeted dis-
et al., 1993). Immunoblot analysis showed that R2D5 anti- rupted of ARK-2 have been available, and cilia prepara-
body recognizes a 22 kDa protein that is abundant in the tions derived from the ARK-2–deficient mice showed lack
olfactory epithelium and in the olfactory bulb. This protein of the agonist-induced desensitization (Peppel et al., 1997).
contains three calcium-binding EF hands and potent phos- Taken together, the expression of ARK-2 and ARR-2
phorylation sites for calcium/calmodulin-dependent pro- within the olfactory cilia, the inhibition of desensitization
tein kinase II (CaMPK II) and cAMP-dependent protein with ARK-2–and ARR-2–neutralizing antibodies, and
kinase (PKA). Different from ubiquitously expressed the lack of the agonist-induced desensitization in the
calmodulin, this calcium-binding protein is expressed ARK-2–deficient mice suggest that ARK-2 and ARR-2
specifically in ORNs, indicating that this protein may par- mediate the odorant-dependent desensitization in olfaction.
ticipate in olfactory signal transduction. In addition, it has been suggested that PKA or PKC may
Calcium itself mediates Cl conductance in ORNs play a role in olfactory desensitization (Boekhoff and Breer,
(Kleene and Gesteland, 1991a; Kleene, 1993; Lowe and 1992). PKA has been implicated in olfactory desensitization
Gold, 1993b). The odor-induced currents show little recti- following increase in cAMP by odorant stimulation, and
fication. It appears that the depolarizing current has two PKC may mediate desensitization following Pl cycle acti-
components, an initial inward cationic conductance fol- vation by odorant stimulation. However, these results need
lowed by an inward anionic Cl conductance (Kleene, to be reexamined, given more recent data using knockout
1993; Kurahashi and Yau, 1993; Lowe and Gold, 1993b). animals that indicate that cAMP mediates odorant detection.
Calcium, which enters the cilia through the cyclic Cyclic GMP may also be involved in desensitization.
nucleotidegated channel, triggers a calcium-activated Cl Zufall and Leinders-Zufall (1997) showed that cGMP
channel in olfactory cilia membrane (Kleene and mediated a long-lasting form of odor response adaptation
Gesteland, 1991b). This conductance may serve as a “fail- in tiger salamander. The long-lasting adaptation lasted for
safe” so that cells can depolarize, irrespective of changes several minutes and was attributable to cyclic nucleotide-
in extracellular milieu. gated channel modulation by cGMP. They also showed
that this form of long-lasting adaptation was abolished
selectively by HO inhibitors (thus preventing CO release
VIII. DESENSITIZATION and cGMP formation), whereas odor excitation was
unaffected. The results suggest that endogenous
Desensitization occurs through a variety of processes, CO/cGMP signals contribute to olfactory desensitization.
including phosphorylation, internalization, and receptor-
effector uncoupling (Hausdorff et al., 1990; Huganir and
Greengard, 1990; Sibley et al., 1987). The homologous IX. LONG-TERM RESPONSES TO ODORANT
desensitization of G-protein–coupled receptors is well DETECTION
established in 2-adrenergic receptor (AR-2) as a model
(Benovic et al., 1988, 1989). Phosphorylation of receptors The theory that extracellular signals, such as hormones,
by a specific receptor kinase termed -adrenergic receptor growth factors, and neuronal activity, modulate transcrip-
kinase (ARK) mediates homologous desensitization. tional events to produce long-term changes in cellular
Complete quenching of signal transduction requires the activity is well established (Hill and Treisman, 1995).
binding of a protein called -arrestin (ARR) to phospho- However, the long-lasting effects of odorant stimulation in
rylated receptor (Lohse et al., 1990). Specific isoforms of ORNs are unknown.
ARK and ARR, ARK-2 and ARR-2 were localized to A delayed cAMP response upon odorant stimulation
olfactory neurons, specifically to olfactory cilia and was characterized and was mediated by cGMP via acti-
dendritic knobs (Dawson et al., 1993). Other isoforms of vation of a cGMP-dependent protein kinase (PKG) (Moon
ARK or ARR were not present in these regions. et al., 1998). Based on the kinetics of the delayed cAMP
Functional studies of ARK-2 and ARR-2 in the olfactory response previously discussed, it was postulated that cGMP
cilia were performed (Dawson et al., 1993; Schleicher et al., might mediate a delayed cAMP response to regulate long-
1993). The odorant-induced cAMP production was moni- term cellular responses to odorant detection, including gene
tored in the presence or absence of neutralizing antibodies expression. Recent work supports this idea. Odorant stimu-
against specific isoforms of ARK and ARR. lation can result in transcriptional changes via CREB
Preincubation of isolated olfactory cilia with neutralizing activation (Moon et al., 1999). While incubation with either
Molecular Neurobiology of Olfactory Transduction 85

8-Br-cGMP or a soluble guanylyl cyclase activator (sodium inositide-specific phospholipase C from catfish olfactory
nitroprusside) increased CREB activation, PKG could not rosettes. Brain Res. 31:10–16.
directly phosphorylate CREB in ORNs. Thus, cGMP pro- Adrian, T. E., Allen, J. M., Bloom, S. R., Ghatei, M. A., Rossor,
duced upon odorant stimulation may generate a sustained M. N., Roberts, G. W., Crow, T. J., Tatemoto, K., and Polak,
J. M. (1983). Neuropeptide Y distribution in human brain.
cAMP signal capable of activating CREB.
Nature 306:584–586.
Involvement of the Ras-MAPK (mitogen-activated pro-
Anholt, R. R. H., and Rivers, A. M. (1990). Olfactory transduc-
tein kinase) signal transduction pathway in olfaction was tion: cross-talk between second-messenger systems.
recently reported in C. elegans (Hirotsu et al., 2000). The Biochemistry 29:4049–4054.
Ras-MAPK pathway plays important roles in cellular pro- Baimbridge, K. G., Celio, M. R., and Rogers, J. H. (1992).
liferation and differentiation in response to extracellular Calcium-binding proteins in the nervous system. Trends
signals. Mutational inactivation and hyperactivation of this Neurosci. 15:303–308.
pathway impaired efficiency of chemotaxis to a set of odor- Bakalyar, H. A., and Reed, R. R. (1990). Identification of a spe-
ants. The activation of MAPK upon odorant stimulation cialized adenylyl cyclase that may mediate odorant detection.
was dependent on calcium via the nucleotide-gated channel Science 250:1403–1406.
and the voltage-activated calcium channel. More recently, Baraban, S. C., Hollopeter, G., Erickson, J. C., Schwartzkroin, P.
A., and Palmiter, R. D. (1997). Knock-out mice reveal a crit-
Storm and colleagues demonstrated that odorants activate
ical antiepileptic role for neuropeptide Y. J. Neurosci.
MAPK in rodent ORNs (Watt and Storm, 2001). The
17:8927–8936.
odorant-activation of MAPK pathway led to the activation Bastianelli, E., Polans, A. S., Hidaka, H., and Pochet, R. (1995).
of cAMP response element (CRE)–mediated transcription. Differential distribution of six calcium-binding proteins in the
The odorant stimulation of MAPK activation was ablated rat olfactory epithelium during postnatal development and
by inhibition of CaM-dependent protein kinase II adulthood. J. Comp. Neurol. 354:395–409.
(CaMKII), suggesting that odorant activation of MAPK is Bauer, I., Wanner, G. A., Rensing, H., Alte, C., Miescher, E. A.,
mediated through CaMKII. Moreover, discrete populations Wolf, B., Pannen, B. H., Clemens, M. G., and Bauer, M.
of ORNs display CRE-mediated gene transcription when (1998). Expression pattern of heme oxygenase isoenzymes
stimulated by odorants in mice. Taken together, these data 1 and 2 in normal and stress-exposed rat liver. Hepatology
suggest that ORNs may undergo long-term adaptive 27:829–838.
Beavo, J. A. (1995). Cyclic nucleotide phosphodiesterases: func-
changes mediated through CRE-mediated transcription.
tional implications of multiple isoforms. Physiol. Rev.
75:725–748.
X. CONCLUSIONS Beavo, J. A., Conti, M., and Heaslip, R. J. (1994). Multiple cyclic
nucleotide phosphodiesterases. Mol. Pharmacol. 46:399–405.
Belluscio, L., Gold, G. H., Nemes, A., and Axel, R. (1998).
Olfaction is an essential sensory modality that influences Mice deficient in G(olf) are anosmic. Neuron 20:69–81.
the quality and in many cases the survival of an organism. Beltman, J., Sonnenburg, W. K., and Beavo, J. A. (1993). The role
Tremendous progress has been made in the last decade of protein phosphorylation in the regulation of cyclic
regarding our understanding of odorant transduction. nucleotide phosphodiesterases. Mol. Cell Biochem. 127/128:
Challenges remain. Understanding the olfactory code will 239–253.
allow us to manipulate olfactory perception in both health Benovic, J. L., Bouvier, M., Caron, M. G., and Lefkowitz, R. J.
and disease. Our appreciation of the ability of odor per- (1988). Regulation of adenylyl cyclase-coupled b-adrenergic
ception to influence long-term neuronal responses, and receptors. Ann. Rev. Cell Biol. 4:405–428.
potentially neuronal survival, may provide clues to under- Benovic, J. L., DeBlasi, A., Stone, W. C., Caron, M. G., and
standing this process in other neuronal systems. Given the Lefkowitz, R. J. (1989). -Adrenergic receptor kinase: pri-
mary structure delineates a multigene family. Science 246:
tools that we have available, it is clear that the olfactory
235–240.
system is an excellent model for signal transduction and Berridge, M. J., and Irvine, R. F. (1984). Inositol trisphosphate, a
neuronal homeostasis. The cytoarchitecture, life cycle, novel second messenger in cellular signal transduction.
availability of in vitro models, and straightforward axonal Nature 312:315–321.
projections of ORNs make them amenable to future stud- Berridge, M. J., and Irvine, R. F. (1989). Inositol phosphates and
ies that seek to investigate stimulus processing. cell signalling. Nature 341:197–204.
Beschorner, R., Adjodah, D., Schwab, J. M., Mittelbronn, M.,
Pedal, I., Mattern, R., Schluesener, H. J., and Meyermann, R.
REFERENCES (2000). Long-term expression of heme oxygenase-1 (HO-1,
HSP-32) following focal cerebral infarctions and traumatic
Abogadie, F. C., Bruch, R. C., Wurzburger, R., Margolis, F. L., brain injury in humans. Acta Neuropathol. (Berl) 100:
and Farbman, A. I. (1995). Molecular cloning of a phospho- 377–384.
86 Moon and Ronnett

Bianchet, M. A., Bains, G., Pelosi, P., Pevsner, J., Snyder, S. H., Burns, F., Zhao, A. Z., and Beavo, J. A. (1996). Cyclic nucleotide
Monaco, H. L., and Amzel, L. M. (1996). The three-dimen- phosphodiesterases: gene complexity, regulation by phospho-
sional structure of bovine odorant binding protein and its rylation, and physiological implications. Adv. Pharmacol.
mechanism of odor recognition [see comments]. Nat. Struct. 36:29–48.
Biol. 3:934–939. Burova, T. V., Rabesona, H., Choiset, Y., Jankowski, C. K.,
Boekhoff, I., and Breer, H. (1992). Termination of second mes- Sawyer, L., and Haertle, T. (2000). Why has porcine VEG pro-
senger signaling in olfaction. Proc. Natl. Acad. Sci. USA tein unusually high stability and suppressed binding ability?
89:471–474. Biochim. Biophys. Acta 1478:267–279.
Boekhoff, I., Tareilus, E., Strotmann, J., and Breer, H. (1990). Bush, K. T., Stuart, R. O., Li, S. H., Moura, L. A., Sharp, A. H.,
Rapid activation of alternative second messenger pathways in Ross, C. A., and Nigam, S. K. (1994). Epithelial inositol
olfactory cilia from rats by different odorants. EMBO J. 9: 1,4,5-trisphosphate receptors. Multiplicity of localization,
2453–2458. solubility, and isoforms. J. Biol. Chem. 269:23694–23699.
Boekhoff, I., Michel, W. C., Breer, H., and Ache, B. W. (1994). Caggiano, M., Kauer, J. S., and Hunter, D. D. (1994). Globose
Single odors differentially stimulate dual second messenger basal cells are neuronal progenitors in the olfactory epithe-
pathways in lobster olfactory receptor cells. J. Neurosci. 14: lium: a lineage analysis using a replication-incompetent retro-
3304–3309. virus. Neuron 13:339–352.
Boekhoff, I., Braunewell, K. H., Andreini, I., Breer, H., and Calof, A. L., and Chikaraishi, D. M. (1989). Analysis of neuro-
Gundelfinger, E. (1997). The calcium-binding protein VILIP genesis in a mammalian neuroepithelium: proliferation and
in olfactory neurons: regulation of second messenger signal- differentiation of an olfactory neuron precursor in vitro.
ing. Eur. J. Cell Biol. 72:151–158. Neuron 3:115–127.
Bonigk, W., Bradley, J., Muller, F., Sesti, F., Boekhoff, I., Carr, V. M., and Farbman, A. I. (1993). The dynamics of
Ronnett, G. V., Kaupp, U. B., and Frings, S. (1999). The cell death in the olfactory epithelium. Exp. Neurol.
native rat olfactory cyclic nucleotide-gated channel is com- 124:308–314.
posed of three distinct subunits. J. Neurosci. 19:5332–5347. Chen, S., Lane, A. P., Bock, R., Leinders-Zufall, T., and Zufall, F.
Borisy, F. F., Ronnett, G. V., Cunningham, A. M., Juilfs, D., (2000). Blocking adenylyl cyclase inhibits olfactory generator
Beavo, J., and Snyder, S. H. (1991). Calcium/calmodulin currents induced by “IP(3)-odors”. J.Neurophysiol. 84:
activated phosphodiesterase selectively expressed in olfac- 575–580.
tory receptor neurons. J. Neurosci. 12:915–923. Chen, T.-Y., and Yau, K.-W. (1994). Direct modulation by Ca2-
Borisy, F. F., Hwang, P. M., Ronnett, G. V., and Snyder, S. H. calmodulin of cyclic nucleotide-activated channel of rat olfac-
(1993). High affinity cyclic AMP phosphodiesterase and adeno- tory receptor neurons. Nature 368:545–548.
sine localized in sensory organs. Brain Res. 610:199–207. Choi, E. J., Xia, Z., and Storm, D. R. (1992). Stimulation of the
Bradley, J., Li, J., Davidson, N., Lester, H. S., and Zinn, K. type III olfactory adenylyl cyclase by calcium and calmod-
(1994). Heteromeric olfactory cyclic nucleotide-gated chan- ulin. Biochemistry 31:6492–6498.
nels: a subunit that confers increased sensitivity to cAMP. Choi, E.-J., Wong, S. T., Dittman, A. H., and Storm, D. R. (1993).
Proc. Natl. Acad. Sci. USA 91:8890–8894. Phorbol ester stimulation of the type I and type III adenylyl
Breer, H., and Boekhoff, I. (1991). Odorants of the same odor cyclases in whole cells. Biochem. J. 32(8):1891–1894.
class activate different second messenger pathways. Chem. Coccia, V. J., and Cote, R. H. (1994). Regulation of intracellu-
Senses 16:19–29. lar cyclic GMP concentration by light and calcium in
Breer, H., Boekhoff, I., and Tareilus, E. (1990). Rapid kinetics of electropermeabilized rod photoreceptors. J. Gen. Physiol.
second messenger formation in olfactory transduction. Nature 103:67–86.
345:65–68. Constanzo, R. M., and Graziadei, P. P. C. (1983). A quantitative
Breer, H., Klemm, T., and Boekhoff, I. (1992). Nitric oxide medi- analysis of changes in the olfactory epithelium following
ated formation of cyclic GMP in the olfactory system. bulbectomy in hamster. J. Comp. Neurol. 215: 370–381.
NeuroReport 3(11):1030–1031. Cooper, D. M. F., Mons, N., and Karpen, J. W. (1995). Adenylyl
Bruch, R. C., Abogadie, F. C., and Farbman, A. I. (1995). cyclases and the interaction between calcium and cAMP sig-
Identification of three phospholipase C isotypes expressed in naling. Nature 374:421–424.
rat olfactory epithelium. NeuroReport 6:233–237. Cunningham, A. M., Ryugo, D. K., Sharp, A. H., Reed, R. R.,
Brunet, L. J., Gold, G. H., and Ngai, J. (1996). General anosmia Snyder, S. H., and Ronnett, G. V. (1993). Neuronal inositol
caused by a targeted disruption of the mouse olfactory cyclic 1,4,5-trisphosphate receptor localized to the plasma mem-
nucleotide-gated cation channel. Neuron 17:681–693. brane of olfactory cilia. Neuroscience 57:339–352.
Buck, L. B. (1992). The olfactory multigene family. Curr. Biol. Dal Monte, M., Andreini, I., Revoltella, R., and Pelosi, P. (1991).
2:467–473. Purification and characterization of two odorant-binding pro-
Buck, L. B. (1996). Information coding in the vertebrate olfac- teins from nasal tissue of rabbit and pig. Comp. Biochem.
tory system. Ann. Rev. Neurosci. 19:517–544. Physiol. [B] 99:445–451.
Buck, L., and Axel, R. (1991). A novel multigene family may Danger, J. M., Tonon, M. C., Jenks, B. G., Saint-Pierre, S.,
encode odorant receptors: a molecular basis for odor recogni- Martel, J. C., Fasolo, A., Breton, B., Quirion, R., Pelletier, G.,
tion. Cell 65:175–187. and Vaudry, H. (1990). Neuropeptide Y: localization in the
Molecular Neurobiology of Olfactory Transduction 87

central nervous system and neuroendocrine functions. Garibotti, M., Navarrini, A., Pisanelli, A. M., and Pelosi, P.
Fundam. Clin. Pharmacol. 4:307–340. (1997). Three odorant-binding proteins from rabbit nasal
Dawson, T. M., Arriza, J. L., Lefkowitz, R. J., Jaworsky, D. E., mucosa. Chem. Senses 22:383–390.
and Ronnett, G. V. (1993). Beta-adrenergic receptor kinase-2 Getchell, T. V. (1986). Functional properties of vertebrate olfac-
and beta-arrestin-2: mediators of odorant-induced desensitiza- tory receptor neurons. Physiol. Rev. 66:772–818.
tion. Science 259:825–829. Getchell, T. V., and Shepherd, G. M. (1978). Adaptive properties
Dear, T. N., Campbell, K., and Rabbitts, T. H. (1991). Molecular of olfactory receptor analysed with odour pulses of varying
cloning of putative odorant-binding and odorant-metabolizing durations. J. Physiol. 282:541–560.
proteins. Biochemistry 30:10376–10382. Getchell, T. V., Margolis, F. L., and Getchell, M. L. (1985).
Dhallan, R. S., Yau, K. W., Schrader, K. A., and Reed, R. A. Perireceptor and receptor events in vertebrate olfaction. Prog.
(1990). Primary structure and functional expression of a Neurobiol. 23:317–345.
cyclic nucleotide-activated channel from olfactory neurons. Goldstein, B. J., and Schwob, J. E. (1996). Analysis of the glo-
Nature 347:184–187. bose basal cell compartment in rat olfactory epithelium using
Drinnan, S. L., Hope, B. T., Snutch, T. P., and Vincent, S. R. GBC-1, a new monoclonal antibody against globose basal
(1991). Golf in the basal ganglia. Mol. Cell. Neurosci. cells. J. Neurosci. 16:4005–4016.
2:66–70. Graziadei, P. P. C. (1973). Cell dynamics in the olfactory mucosa.
Dwyer, N. D., Troemel, E. R., Sengupta, P., and Bargmann, C. I. Tissue Cell 5:113–131.
(1998). Odorant receptor localization to olfactory cilia is Graziadei, P. P. C., and Metcalf, J. F. (1971). Autoradiographic
mediated by ODR-4, a novel membrane-associated protein. and ultrastructural observation on the frog’s olfactory
Cell 93:455–466. mucosa. Zellforschung 116:305–318.
Ewing, J. F., Raju, V. S., and Maines, M. D. (1994). Induction Graziadei, P. P. C., and Monti-Graziadei, G. A. (1979).
of heart heme oxygenase-1 (HSP32) by hyperthermia: Neurogenesis and neuron regeneration in the olfactory system
possible role in stress-mediated elevation of cyclic 3:5- of mammals. J. Neurocytol. 8:1–18.
guanosine monophosphate. J. Pharmacol. Exp. Ther. Hanoune, J., and Defer, N. (2001). Regulation and role of adenylyl
271:408–414. cyclase isoforms. Annu. Rev. Pharmacol. Toxicol. 41:145–174.
Fadool, D. A., and Ache, B. W. (1992). Plasma membrane inositol Hansel, D. E., Eipper, B. A., and Ronnett, G. V. (2001).
1,4,5-trisphosphate-activated channels mediate signal trans- Neuropeptide Y functions as a neuroproliferative factor.
duction in lobster olfactory receptor neurons. Neuron Nature 410:940–944.
9:907–918. Hatt, H., and Ache, B., W. (1994). Cyclic nucletide-and inositol
Farbman, A. I. (1992). Development and plasticity. In Cell phosphate-gated ion channels in lobster olfactory receptor
Biology of Olfaction, Barlow, P. W., Bray, D., Green, P. B. and neurons. Proc. Natl. Acad. Sci. USA 91:6264–6268.
Slack, J. M. W. (Eds.). Cambridge University Press, Hausdorff, W. P., Caron, M. G., and Lefkowitz, R. J. (1990).
Cambridge, pp. 167–206. Turning off the signal: desensitization of b-adrenergic recep-
Farbman, A. I., and Margolis, F. L. (1980). Olfactory marker pro- tor function. FASEB J. 4:2881–2889.
tein during ontogeny: immunohistochemical localization. Heller, J. (1975). Interactions of plasma retinol-binding protein
Dev. Biol. 74:205–215. with its receptor. J. Biol. Chem. 250:3613–3619.
Fijimoto, T., Nakade, S., Miyawaki, A., Mikoshiba, K., and Hill, C. S., and Treisman, R. (1995). Transcriptional regulation
Ogawa, K. (1992). Localization of inositol 1,4,5-trisphos- by extracellular signals: mechanisms and specificity. Cell
phate receptor-like protein in plasmalemmal caveolae. J. Cell 80:199–211.
Biol. 119:1507–1513. Hirata, K., He, J. W., Kuraoka, A., Omata, Y., Hirata, M., Islam,
Firestein, S., and Werblin, F. S. (1989). Odor-induced membrane A. T., Noguchi, M., and Kawabuchi, M. (2000). Heme oxyge-
currents in vertebrate olfactory receptor neurons. Science nase 1 (HSP-32) is induced in myelin-phagocytosing
244:79–82. Schwann cells of injured sciatic nerves in the rat. Eur. J.
Firestein, S., Shepherd, G. M., and Werblin, F. S. (1990). Time Neurosci. 12:4147–4152.
course of the membrane current underlying sensory transduc- Hirotsu, T., Saeki, S., Yamamoto, M., and Iino, Y. (2000). The
tion in salamander olfactory receptor neurones. J. Physiol. Ras-MAPK pathway is important for olfaction in
430:135–158. Caenorhabditis elegans. Nature 404:289–293.
Frings, S. (1993). Protein kinase C sensitizes olfactory adenylate Hirsch, J. D., and Margolis, F. L. (1980). Influence of unilateral
cyclase. J. Gen. Physiol. 101:183–205. bulbectomy on opiate and other binding sites in the contralat-
Frings, S., Seifert, R., Godde, M., and Kaupp, U. B. (1995). eral bulb. Brain. Res. 199:39–47.
Profoundly different calcium permeation and blockage deter- Hsu, Y.-T., and Molday, R. S. (1993). Modulation of the cGMP-
mine the specific function of distinct cyclic nucleotide-gated gated channel of rod photoreceptor cells by calmodulin.
channels. Neuron 15:169–179. Nature 361:76–79.
Fulle, H.-J., Vassar, R., Foster, D. C., Yang, R.-B., Axel, R., and Huard, J. M., Youngentob, S. L., Goldstein, B. J., Luskin, M. B.,
Garbers, D. L. (1995). A receptor guanylyl cyclase expressed and Schwob, J. E. (1998). Adult olfactory epithelium contains
specifically in olfactory sensory neurons. Proc. Natl. Acad. multipotent progenitors that give rise to neurons and non-
Sci. USA 92:3571–3575. neural cells. J. Comp. Neurol. 400:469–486.
88 Moon and Ronnett

Huganir, R. L., and Greengard, P. (1990). Regulation of neuro- Krautwurst, D., Yau, K. W., and Reed, R. R. (1998). Identification
transmitter receptor desensitization by protein phosphoryla- of ligands for olfactory receptors by functional expression of
tion. Neuron 5:555–567. a receptor library. Cell 95:917–926.
Huque, T., and Brunch, R. C. (1986). Odorant-and guanine Kuno, M., and Gardner, P. (1987). Ion channels activated by inos-
nucleotide-stimulated phosphoinositide turnover in olfactory itol 1,4,5-trisphosphate in plasma membrane of human T-lym-
cilia. Biochem. Biophys. Res. Commun. 137:36–42. phocytes. Nature 326:301–304.
Iino, S., Kobayashi, S., Okazaki, K., and Hidaka, H. (1995). Kurahashi, T., and Shibuya, T. (1990). Ca2-dependent adaptive
Neurocalcin-immunoreactive receptor cells in the rat olfactory properties in the solitary olfactory receptor cell of the newt.
epithelium and vomeronasal organ. Neurosci. Lett. 191: Brain Res. 515:261–268.
91–94. Kurahashi, T., and Yau, K.-W. (1993). Co-existence of cationic
Ingi, T., Cheng, J., and Ronnett, G. V. (1996). Carbon monoxide: and chloride components in odorant-induced current of verte-
an endogenous modulator of the nitric oxide-cyclic GMP sig- brate olfactory receptor cells. Nature 363:71–74.
naling system. Neuron 16:835–842. Kutty, R. K., and Maines, M. D. (1989). Selective induction of
Ingi, T., and Ronnett, G. V. (1995). Direct demonstration of a heme oxygease-1 isozyme in rat testis by human chorionic
physiological role for carbon monoxide in olfactory receptor gonadotropin. Arch. Biochem. Biophys. 268:100.
neurons. J. Neurosci. 15:8214–8222. Labarca, P., and Bacigalupo, J. (1988). Ion channels from
Jaworsky, D. E., Matsuzaki, O., Borisy, F. F., and Ronnett, G. V. chemosensory olfactory neurons. J. Bioenerg. Biomem.
(1995). Calcium modulates the rapid kinetics of the odorant- 20:551–569.
induced cyclic AMP signal in rat olfactory cilia. J. Neurosci. Lazard, D., Zupko, K., Poria, Y., Nef, P., Lazarovits, J., Horn, S.,
15:310–318. Khen, M., and Lancet, D. (1991). Odorant signal termination
Jones, D. T., and Reed, R. R. (1987). Molecular cloning of five by olfactory UDP-glucuronosyl transferase. Nature
GTP-binding protein cDNA species from rat olfactory neuro- 349:790–793.
epithelium. J. Biol. Chem. 262:14241–14249. Leinders-Zufall, T., Shepherd, G. M., and Zufall, Z. (1996).
Joseph, S. K. (1996). The inositol triphosphate receptor family. Modulation by cyclic GMP of the odour sensitivity of verte-
Cell. Signal. 8:1–7. brate olfactory receptor cells. Proc. R. Soc. Bio. 263:803–811.
Juilfs, D. M., Fulle, H. J., Zhao, A. Z., Houslay, M. D., Garbers, Liman, E. R., and Buck, L. B. (1994). A second subunit of the
D. L., and Beavo, J. A. (1997). A subset of olfactory neurons olfactory cyclic nucleotide-gated channel confers high sensi-
that selectively express cGMP-stimulated phosphodiesterase tivity to cAMP. Neuron 13:611–621.
(PDE2) and guanylyl cyclase-D define a unique olfactory sig- Lobel, D., Marchese, S., Krieger, J., Pelosi, P., and Breer, H.
nal transduction pathway. Proc. Natl. Acad. Sci. USA (1998). Subtypes of odorant-binding proteins—heterologous
94:3388–3395. expression and ligand binding. Eur. J. Biochem. 254:318–324.
Kalinoski, D. L., Aldinger, S. B., Boyle, A. G., Huque, T., Lohse, M. J., Benovic, J. L., Codina, J., Caron, M. G., and
Maracek, J. F., Prestwich, G. D., and Restrepo, D. (1992). Lefkowitz, R. J. (1990). b-arrestin: a protein that regulates b-
Characterization of a novel inositol 1,4,5-trisphosphate recep- adrenergic receptor function. Science 248:1547–1550.
tor in isolated olfactory cilia. Biochem. J. 281:449–456. Lowe, G., and Gold, G. H. (1993a). Contribution of the ciliary
Kaupp, U. B. (1991). The cyclic nucleotide-gated channels of cyclic nucleotide-gated conductance to olfactory transduction
vertebrate photoreceptors and olfactory epithelium. Trends in the salamander. J. Physiol. 462:175–196.
Neurosci. 14:150–157. Lowe, G., and Gold, G. H. (1993b). Nonlinear amplification by
Kleene, S. J. (1993). Origin of the chloride current in olfactory calcium-dependent chloride channels in olfactory receptor
transduction. Neuron 11:123–132. cells. Nature 366:283–286.
Kleene, S. J., and Gesteland, R. C. (1991a). Calcium-activated Malnic, B., Hirono, J., Sato, T., and Buck, L. (1999).
chloride conductance in frog olfactory cilia. J. Neurosci. Combinatorial receptor codes for odors. Cell 96:713–723.
11(11):3624–3629. Margolis, F. L. (1980). A marker protein for the olfactory
Kleene, S. J., and Gesteland, R. C. (1991b). Calcium-activated chemoreceptor neuron. In Proteins of the Nervous System,
chloride conductance in frog olfactory cilia. J. Neurosci. Bradshaw, R. A., and Schneider, D. M. (Eds.). Raven, New
11:3624–3629. York, pp. 59–84.
Koistinaho, J., Miettinen, S., Keinanen, R., Vartiainen, N., Menco, B. P. (1997). Ultrastructural aspects of olfactory signal-
Roivainen, R., and Laitinen, J. T. (1996). Long-term induction ing. Chem. Senses 22:295–311.
of haem oxygenase-1 (HSP-32) in astrocytes and microglia Menco, B. P., Bruch, R. C., Dau, B., and Danho, W. (1992a).
following transient focal brain ischaemia in the rat. Eur. J. Ultrastructural localization of olfactory transduction compon-
Neurosci. 8:2265–2272. ents: the G protein subunit Golf alpha and type III adenylyl
Koshimoto, H., Katoh, K., Yoshihara, Y., Nemoto, Y., and Mori, cyclase. Neuron 8:441–453.
K. (1994). Immunohistochemical demonstration of Menco, B. P. M., Bruch, R. C., Dau, B., and Danho, W. (1992b).
embryonic expression of an odor receptor protein and its Ultrastructural localization of olfactory transduction compon-
zonal distribution in the rat olfactory epithelium. Neurosci. ents: the G protein subunit Golfa and type III adenylyl cyclase.
Lett. 169:73–76. Neuron 8:441–453.
Molecular Neurobiology of Olfactory Transduction 89

Menco, B. P., Cunningham, A. M., Qasba, P., Levy, N., and Reed, Nakamura, T., and Gold, G. H. (1987). A cyclic-nucleotide
R. R. (1997a). Putative odour receptors localize in cilia of gated conductance in olfactory receptor cilia. Nature
olfactory receptor cells in rat and mouse: a freeze-substitution 325:442–444.
ultrastructural study. J. Neurocytol. 26:691–706. Nemoto, Y., Ikeda, J., Katoh, K., Koshimoto, H., Yoshihara, Y.,
Menco, B. P., Cunningham, A. M., Qasba, P., Levy, N., and Reed, and Mori, K. (1993). R2D5 antigen: a calcium-binding phos-
R. R. (1997b). Putative odour receptors localize in cilia of phoprotein predominantly expressed in olfactory receptor
olfactory receptor cells in rat and mouse: a freeze-substitution neurons. J. Cell Biol. 123:963–976.
ultrastructural study. J. Neurocytol. 26:297–312. Okano, T. M. (1974). Secreation and electrogenesis of the sup-
Menco, B. P., Cunningham, A. M., Qasba, P., Levy, N., and Reed, porting cell in the olfactory epithelium. J. Physiol. (London)
R. R. (1997c). Putative odour receptors localize in cilia of 242:353–370.
olfactory receptor cells in rat and mouse: a freeze-substitution Ottoson, D. (1956). Analysis of the electrical activity of the olfac-
ultrastructural study [corrected and republished article ori- tory epithelium. Acta Physiol. Scan. 122:1–83.
ginally printed in J. Neurocytol. 1997 May;26(5):297–312]. Pace, U., and Lancet, D. (1986). Olfactory GTP-binding protein:
J. Neurocytol. 26:691–706. signal transducing polypeptide of vertebrate chemosensory
Meyer, M. R., Angele, A., Kremmer, E., Kaupp, U. B., and neurons. Proc. Natl. Acad. Sci. USA 83:4947–4951.
Muller, F. (2000). A cGMP-signaling pathway in a subset of Pace, U., Hanski, E., Salomon, Y., and Lancet, D. (1985).
olfactory sensory neurons. Proc. Natl. Acad. Sci. USA Odorant-sensitive adenylate cyclase may mediate olfactory
97:10595–10600. reception. Nature 316:255–258.
Mitchell, J., Gutierrez, J., and Northup, J. K. (1995). Purification, Palczewski, K., Subbaraya, I., Gorczyca, W. A., Helekar, B. S.,
characterization, and partial amino acid sequence of a G pro- Ruiz, C. C., Ohguro, H., Huang, J., Zhao, X., Crabb, J. W.,
tein-activated phospholipase C from squid photoreceptors. Johnson, R. S., Walsh, K. A., Gray-Keller, M. P., Detwiler,
J. Biol. Chem. 270:854–859. P. B., and Baehr, W. (1994). Molecular cloning and character-
Miwa, N., Kobayashi, M., Takamatsu, K., and Kawamura, S. ization of retinal photoreceptor guanylyl cyclase-activating
(1998). Purification and molecular cloning of a novel calcium- protein. Neuron 13:395–404.
binding protein, p26olf, in the frog olfactory epithelium. Parent, A., Schrader, K., Munger, S. D., Reed, R. R., Linden, D.
Biochem. Biophys. Res. Commun. 251:860–867. J., and Ronnett, G. V. (1998). Synaptic transmission and hip-
Miyamoto, T., Restrepo, D., Cragoe, E. J., and Teeter, J. H. pocampal long-term potentiation in olfactory cyclic
(1992). IP3 and cAMP-induced responses in isolated olfactory nucleotide-gated channel type 1 null mouse. J. Neurophysiol.
receptor neurons from the channel catfish. J. Membrane Biol. 79:3295–3301.
127:173–183. Pelosi, P., Baldaccini, N. E., and Pisanelli, A. M. (1982).
Mombaerts, P. (1999a). Molecular biology of odorant receptors Identification of a specific olfactory receptor for 2-isobutyl-3-
in vertebrates. Annu. Rev. Neurosci. 22:487–509. methoxypyrazine. Biochem. J. 201:245–248.
Mombaerts, P. (1999b). Odorant receptor genes in humans. Curr. Peppel, K., Boekhoff, I., McDonald, P., Breer, H., Caron, M. G.,
Opin. Genet. Dev. 9:315–320. and Lefkowitz, R. J. (1997). G protein-coupled receptor
Mombaerts, P., Wang, F., Dulac, C., Chao, S. K., Nemes, A., kinase 3 (GRK3) gene disruption leads to loss of odorant
Mendelsohn, M., Edmondson, J., and Axel, R. (1996). receptor desensitization. J. Biol. Chem. 272:25425–25428.
Visualizing an olfactory sensory map. Cell 87:675–686. Pes, D., and Pelosi, P. (1995). Odorant-binding proteins of the
Mons, N., and Cooper, D. M. F. (1995). Adenylate cyclases: crit- mouse. Comp. Biochem. Physiol. B Biochem. Mol. Biol.
ical foci in neuronal signaling. Trends Neurosci. 18(12): 112:471–479.
536–542. Pevsner, J., Trifiletti, R. R., Strittmatter, S. S., and Snyder, S. H.
Moon, C., Jaberi, P., Otto-Bruc, A., Baehr, W., Palczewski, K., (1985). Isolation and characterization of an olfactory receptor
and Ronnett, G. V. (1998). Calcium-sensitive particulate protein for odorant pyrazines. Proc. Natl. Acad. Sci. USA
guanylyl cyclase as a modulator of cAMP in olfactory recep- 82:3050–3054.
tor neurons. J. Neurosci. 18:3195–3205. Pevsner, J., Sklar, P. B., and Snyder, S. H. (1986). Odorant-bind-
Moon, C., Sung, Y., Reddy, R., and Ronnett, G. V. (1999). ing protein: localization to nasal glands and secretions. Proc.
Odorants induce the phosphorylation of the cAMP response Natl. Acad. Sci. USA 83:4942–4946.
element binding protein in olfactory receptor neurons. PNAS Pevsner, J., Hou, V., Snowman, A. M., and Snyder, S. H. (1990).
96:14605–14610. Odorant-binding protein. Characterization of ligand binding.
Moulton, D. G., and Beidler, L. M. (1967). Structure and function J. Biol. Chem. 265:6118–6125.
in the peripheral olfactory system. Physiol. Rev. 47:1–52. Pevsner, J., Hwang, P. M., Sklar, P. B., Venable, J. C., and Snyder,
Munger, S. D., Gleeson, R. A., Aldrich, H. C., Rust, N. C., Ache, S. H. (1988). Odorant-binding protein and its mRNA are
B. W., and Greenberg, R. M. (2000). Characterization of a localized to lateral nasal gland implying a carrier function.
phosphoinositide-mediated odor transduction pathway reveals Proc. Natl. Acad. Sci. USA 85:2383–2387.
plasma membrane localization of an inositol 1,4,5-trisphos- Ressier, K. J., Sullivan, S. L., and Buck, L. B. (1993). A zonal
phate receptor in lobster olfactory receptor neurons. J. Biol. organization of odorant receptor gene expression in the olfac-
Chem. 275:20450–20457. tory epithelium. Cell 73:597–609.
90 Moon and Ronnett

Ressler, K. J., Sullivan, S. L., and Buck, L. B. (1994). Snyder, S. H. (1995). Nitric oxide and carbon monoxide:
Information coding in the olfactory system: evidence for a unprecedented signalling molecules in the brain. In:
stereotyped and highly organized epitope map in the olfactory Encyclopedia Britannica, Chicago, pp. 84–101.
bulb. Cell 79:1245–1255. Somlyo, A. V., and Walz, B. (1995). Ca2 in visual transduction
Restrepo, D., Miyamoto, T., and Bryant, B. P. (1990). Odor and adaptation in vertebrate and invertebrates. Cell Cal.
stimuli trigger influx of Ca2 into olfactory neurons of the 18:253–255
channel catfish. Science 249:1166–1168. Tang, W.-J., and Gilman, A. G. (1992). Adenylyl cyclases. Cell
Rhein, L. D., and Cagan, R. H. (1980). Biochemical studies of 70:869–872.
olfaction: isolation, characterization and odorant binding Taylor, C. W., and Richardson, A. (1991). Structure and function
activity of cilia from rainbow trout olfactory rosettes. Proc. of inositol trisphosphate receptors. Pharmac. Ther. 51:97–137.
Natl. Acad. Sci. USA 77:4412–4416. Taylor, C. W., and Traynor, D. (1995). Calcium and inositol
Rhein, L. D., and Cagan, R. H. (1983). Biochemical studies of trisphosphate receptor. J. Membrane Biol. 145:109–118.
olfaction: binding specificity of odorants to cilia preparation Udovichenko, I. P., Cunnick, J., Gonzalez, K., and Takemoto, D.
from rainbow trout olfactory rosettes. J. Neurochem. J. (1994). The visual transduction and the phosphoinositide
41:569–577. system. Cell. Signal. 6:601–605.
Ronnett, G. V., and Snyder, S. H. (1992). Molecular messengers Vassar, R., Ngai, J., and Axel, R. (1993). Spatial segregation of
of olfaction. Trends Neurosci. 15:508–512. odorant receptor expression in the mammalian olfactory
Ronnett, G. V., Hester, L. D., and Snyder, S. H. (1991a). Primary epithelium. Cell 74:309–318.
culture of neonatal rat olfactory neurons. J. Neurosci. Verma, A., Hirsch, D. J., Glatt, C. E., Ronnett, G. V., and Snyder,
11:1243–1255. S. H. (1993). Carbon monoxide: a putative neural messenger.
Ronnett, G. V., Parfitt, D. J., Hester, L. D., and Snyder, S. H. Science 259:381–384.
(1991b). Odorant-sensitive adenylate cyclase: rapid potent Vogt, R. G., Prestwich, G. D., and Lerner, M. R. (1990).
activation and desensitization in primary olfactory neuronal Odorant-binding-protein subfamilies associate with distinct
cultures. Proc. Natl. Acad. Sci. USA 88:2366–2369. classes of olfactory receptor neurons in insects. J. Neurobiol
Ronnett, G. V., Cho, H., Hester, L. D., Wood, S. F., and Snyder, 22:74–84.
S. H. (1993). Odorants differentially enhance phosphoinosi- Vogt, R. G., Rybczynski, R., and Lerner, M. R. (1991). Molecular
tide turnover and adenylyl cyclase in olfactory receptor neur- cloning and sequencing of general odorant-binding proteins
onal cultures. J. Neurosci. 13:1751–1758. GOBP1 and GOBP2 from the tobacco hawk moth manduca
Roskams, A. J. I., Cai, X., and Ronnett, G. V. (1998). Expression sexta: comparisons with other insect OBPs and their signal
of neuron-specific beta-III tubulin during olfactory neurogen- peptides. J. Neurosci. 11:2972–2984.
esis in the embryonic and adult rat. Neuroscience 83:191–200. Watt, W. C., and Storm, D. R. (2001). Odorants stimulate the
Roskams, J. A., Bredt, D. S., and Ronnett, G. V. (1994). Nitric Erk/MAP kinase pathway and activate CRE-mediated tran-
oxide expression during olfactory neuron development and scription in olfactory sensory neurons. J. Biol. Chem. 276:
regeneration. AChemS 16:308. 2047–2052.
Sautter, A., Zong, X., Hofmann, F., and Biel, M. (1998). An iso- Wayman, G. A., Impey, S., and Storm, D. R. (1995). Ca2 inhi-
form of the rod photoreceptor cyclic nucleotide-gated channel bition of type III adenylyl cyclase in vivo. J. Biol. Chem.
beta subunit expressed in olfactory neurons. Proc. Natl. Acad. 270(37):21480–21486.
Sci. USA 95:4696–4701. Wong, S. T., Trinh, K., Hacker, B., Chan, G. C., Lowe, G.,
Schandar, M., Laugwitz, K. L., Boekhoff, I., Kroner, C., Gaggar, A., Xia, Z., Gold, G. H., and Storm, D. R. (2000).
Gudermann, T., Schultz, G., and Breer, H. (1998). Odorants Disruption of the type III adenylyl cyclase gene leads to
selectively activate distinct G protein subtypes in olfactory peripheral and behavioral anosmia in transgenic mice. Neuron
cilia. J. Biol. Chem. 273:16669–16677. 27:487–497.
Schleicher, S., Boekoff, I., Arriza, J., Lefkowitz, R. J., and Breer, Wood, S. F., Ronnett, G. V., and Snyder, S. H. (1990). Activation
H. (1993). A -adrenergic receptor kinase-like enzyme is of inositol-phosphate metabolism in primary olfaction cell
involved in olfactory signal termination. Proc. Natl. Acad. Sci. cultures. Chem. Senses 15:252.
USA 90:1420–1424. Yan, C., Zhao, A. Z., Bentley, J. K., Loughney, K., Ferguson, K.,
Senhupta, P., Chou, J. H., and Bargmann, C. I. (1996). odr-10 and Beavo, J. A. (1995). Molecular cloning and characteriza-
encodes a seven transmembrane domain olfactory receptor tion of a calmodulin-dependent phophodiesterase enriched in
required for responses to the odorant diacetyl. Cell olfactory sensory neurons. Proc. Natl. Acad. Sci. USA
84:899–909. 92:9677–9681.
Sibley, D. R., Benovic, J. L., Caron, M. G., and Lefkowitz, R. J. Yau, K.-W. (1994). Cyclic nucleotide-gated channels: an expand-
(1987). Regulation of transmembrane signaling by receptor ing new family of ion channels. Proc. Natl. Acad. Sci. USA
phosphorylation. Cell 48:913–922. 91:3481–3483.
Sklar, P. B., Anholt, R. R. H., and Snyder, S. H. (1986). The odor- Zagotta, W. N., and Siegelbaum, S. A. (1996). Structure and func-
ant-sensitive adenylate cyclase of olfactory receptor neurons. tion of cyclic nucleotide-gated channels. Ann. Rev. Neurosci.
J. Biol. Chem. 261:15538–15543. 19:235–263.
Molecular Neurobiology of Olfactory Transduction 91

Zhao, H., Ivic, L., Otaki, J. M., Hashimoto, M., Mikoshiba, K., Zufall, F., Firestein, S., and Shepherd, G. M. (1994). Cyclic
and Firestein, S. (1998). Functinal expression of a mammalian nucleotide-gated ion channels and sensory transduction in
odorant receptor. Science 279:237–241. olfactory receptor neurons. Annu. Rev. Biophys. Biomol.
Zufall, F., and Leinders-Zufall, T. (1997). Identification of a long- Struct. 23:577–607.
lasting form of odor adaptation that depends on the carbon Zuker, C. S. (1996). The biology of vision of Drosophila. Proc.
Monoxide/cGMP second-messenger system. J. Neurosci. Natl. Acad. Sci. USA 93:571–576.
17:2703–2712.
5

Neurogenesis in the Adult Olfactory Neuroepithelium

Alan Mackay-Sim
Griffith University, Brisbane, Queensland, Australia

I. INTRODUCTION II. OVERVIEW

Neurogenesis has long been recognized as a property of After the early reports of basal cell mitosis in mouse olfac-
the adult olfactory epithelium. Over 50 years ago mitot- tory epithelium (Nagahara, 1940) and regeneration of
ic activity was first observed in the olfactory epithelium olfactory sensory neurons after zinc sulfate lesion in mon-
of adult mice (Nagahara, 1940). Olfactory sensory neu- key (Schultz, 1941), there followed numerous reports con-
rons regenerate in monkey (Graziadei et al., 1980; firming these observations in a variety of vertebrates: frog
Schultz, 1941) and human (Murrell et al., 1996; Wolozin (Smith, 1951), fish (Westerman and Baumgarten, 1964),
et al., 1992). Human olfactory neurogenesis continues cat and dog (Andres, 1966), lamprey (Thornhill, 1970),
into old age, making the olfactory system one of the and mouse (Smart, 1971). These early histological obser-
most continually variable regions of the nervous system. vations were supported by analyses using tritiated thymi-
It is now recognized that neurogenesis occurs in a num- dine to label cells during S-phase (DNA replication) of the
ber of sites within the adult brain. A recent study has cell cycle (Graziadei and Metcalf, 1971; Moulton et al.,
even identified newly formed neurons in the brain of 1970; Thornhill, 1970). The field of olfactory neurogene-
aged humans (Eriksson et al., 1998). Sites of neurogen- sis was greatly expanded in the 1970s and 1980s by inten-
esis in the brain include the dentate gyrus and the sub- sive efforts to document and understand the morphological
ventricular zone of the forebrain (recently reviewed in features of neurogenesis and especially the stimulus to
Scheffler et al., 1999). Neurogenesis in the subventricu- neurogenesis brought about by destruction of the sensory
lar zone gives rise to neurons which migrate forward to neurons.
populate the olfactory bulb, providing interneurons in The quantitative 3[H]-thymidine analyses and electron
the periglomerular and granule cell layers (Luskin, microscopic investigations led to the oft-repeated view that
1993). neurogenesis in the adult olfactory epithelium is unique in the
This chapter presents a review of investigations of neu- adult nervous system, now known to be untrue. Another oft-
rogenesis in the adult olfactory epithelium. This process is repeated view is that the olfactory sensory neurons live for
shown to be regulated by endocrine, autocrine, and only 30 days and are more-or-less automatically replaced by
paracrine factors and modulated by environment factors division and differentiation of the basal cells (Graziadei and
presented in the inspired air. Current hypotheses for the Monti Graziadei, 1979; Moulton, 1975). This model of short-
lineage and regulation of neurogenesis are discussed and lived sensory neurons and “automatic” replacement became
explored to provide a cellular and molecular model of this the “orthodoxy” and is cited in primary papers, reviews, and
unusual and interesting “embryonic” feature of adult textbooks. This model was challenged by evidence that some
olfactory epithelium. sensory neurons may live for at least one year (Hinds et al.,

93
94 Mackay-Sim

1984). It was also challenged by evidence that the rate of basal basal region into the midzone containing sensory neuron
cell mitosis may be inversely proportional to epithelial thick- nuclei, the loss of labeled cells by 30 days was interpreted
ness, indicating regulatory mechanisms at work within the to mean that the neurons “remain in the epithelium as
epithelium (Mackay-Sim and Patel, 1984). These and other mature functional elements for approximately 25 days”
data led to an alternative model, which proposed that olfacto- (Graziadei and Monti Graziadei, 1979). This view was
ry sensory neurons do not die “automatically” after 30 days— reinforced by a quantitative analysis, which indicated that
that their lifespan is regulated by extrinsic factors, such as the the “turnover time of the entire population” of cells in the
odorous environment in the nose, rather than immutable, cell- olfactory epithelium is 28.6 days and that the “turnover
intrinsic factors and that olfactory neurogenesis is a process time of the receptor cells should approximate that of the
regulated by endocrine, autocrine, and paracrine factors simi- entire population” (Moulton, 1975). A quantitative study
lar to those operating during embryonic organogenesis of regeneration of the hamster olfactory epithelium also
(Breipohl et al., 1986). A model of regulated neurogenesis is led to an estimate of the life span of receptor neurons of
more in keeping with current views of the development of cells 25–35 days (Samanen and Forbes, 1984). These conclu-
and tissues during embryogenesis. Research in the last 10 sions—that mature olfactory sensory neurons live for
years has centered on the factors that regulate olfactory neuro- about 30 days—were based on the assumption that cells
genesis, and many growth factors have now been implicated. entered the population via division of basal cells and left it
Although it is recognized that basal cells give rise to as mature neurons (Fig. 1). This assumption was later
neurons, less obvious are the identities of the cells in the shown to be false. Nevertheless, there was no doubt that
lineage hierarchy from uncommitted stem cell to precursor new neurons arise in the olfactory epithelium from
cells to neurons in the adult olfactory epithelium. These division of the basal cells. Along with the concept of a
investigations are reviewed in this chapter, and the impli- short-lived sensory neuron, there came to be an assump-
cations for cell therapy based on olfactory epithelium are tion that turnover of sensory neurons from basal cell
discussed. mitosis was a “predetermined . . . genetic characteristic”
(Graziadei and Monti Graziadei, 1978). The prevailing
model thus came to be one of disposable neurons in the
III. OLFACTORY SENSORY NEURONS: olfactory epithelium, similar to cells in other epithelia such
DISPOSABLE OR REPLACEABLE? as the epidermis and the intestinal epithelium in which the
neurons seemed to be inherently obsolescent. It was
Injection of 3[H]-thymidine into the adult mouse labels believed that the olfactory epithelium was unique in (1)
many dividing cells in the olfactory epithelium. At early having short-lived neurons and (2) having the ability to
survival periods after injection the dividing cells are replace them (Graziadei and Monti Graziadei, 1978).
located in two places—most of them are among the basal The notion that mature olfactory sensory neurons live for
cells, close to the basement membrane, with a few located only about 30 days was first challenged by evidence of labeled
apically, among the supporting cells (Graziadei and Monti neurons present one year after injection (Hinds et al., 1984).
Graziadei, 1979; Moulton et al., 1970). A similar distribu- This evidence questioned the prevailing view of a short-lived,
tion of labeled cells is observed in amphibian (Graziadei disposable neuron, but it was only one of several lines of evi-
and Metcalf, 1971; Graziadei, 1973; Mackay-Sim and dence that olfactory neurogenesis may be actually a highly
Patel, 1984). With increasing periods after injection of regulated process, rather than being driven by an genetically
3[H]-thymidine, the labeled basal cells migrate away from predetermined, “clock-like” process (Breipohl et al., 1986).
the basement membrane until their nuclei lie in the mid- This “regulated” model places adult olfactory neurogenesis as
zone of the epithelium in the region of the sensory neuron an extension of the same processes occurring during embry-
nuclei (Graziadei and Monti Graziadei, 1979; Moulton et al., onic development of the nervous system in general (Mackay-
1970). These observations are consistent with basal cells Sim and Kittel, 1991a) with the difference being that the
giving rise to neurons, and under the electron microscope mature olfactory neurons are directly exposed to the external
there appear to be transitional cell types whose morpho- environment and at risk of damage by it (Breipohl et al., 1986;
logy suggests that they are immature neurons (Graziadei, Hinds et al., 1984; Mackay-Sim and Kittel, 1991b). Significant
1973; Graziadei and Monti Graziadei, 1979). By 30 days predictions of this “regulated” model were (1) that the majori-
after injection of 3[H]-thymidine, the labeled cells have ty of dying cells in the olfactory epithelium would be develop-
either disappeared from the epithelium (Graziadei and ing, immature neurons, rather than mature sensory neurons
Monti Graziadei, 1979) or are reduced in number and (2) that mature sensory neurons would remain alive and
(Mackay-Sim and Kittel, 1991a; Moulton et al., 1970). In connected to the olfactory bulb unless damaged by the
combination with the migration of labeled cells from the environment (Breipohl et al., 1986).
Neurogenesis in the Adult Olfactory Neuroepithelium 95

Figure 1 Disposable neuron model of the genesis of the olfactory sensory neuron. The sensory neuron arises from division and differ-
entiation of the basal cell, lives for about 1 month, and is automatically lost from the epithelium.

Many investigations bear out these predictions. mature sensory neurons (Mahalik, 1996). Cell death can
Retrograde labeling by injection of colloidal gold occur within one day of birth (Carr and Farbman, 1993),
provided direct evidence that olfactory neurons remain indicating that apoptosis in the olfactory epithelium is an
connected to the olfactory bulb for at least 3 months integral and early part of neurogenesis in the adult olfac-
(Mackay-Sim and Kittel, 1991b), supporting 3[H]-thymi- tory epithelium.
dine evidence for long-lived neurons (Hinds et al., 1984; In summary, there are two complementary hypotheses
Mackay-Sim and Kittel, 1991a; Moulton et al., 1970). supported by the published data. The first of these is that
Quantitative analyses after 3[H]-thymidine injection olfactory neurogenesis in the adult reflects ontogeny in
show that 70–80% of the labeled cells are lost from the other parts of the nervous system except that adult olfac-
epithelium between 14 and 21 days, after migrating into tory neurogenesis is an ongoing process. In the embryo,
the neuronal layer (Mackay-Sim and Kittel, 1991a; neuronal precursors are born and developing neurons reach
Moulton et al., 1970). The surviving labeled cells survive stages of differentiation within a limited time period so
for at least 3 months (Mackay-Sim and Kittel, 1991a; that the developmental events in the population are reflect-
Moulton et al., 1970) or longer (Hinds et al., 1984). The ed in the molecular events guiding the differentiation of
most parsimonious explanation for this is that the cells each cell. In contrast, all stages of development are seen
lost early are immature neurons which were not success- simultaneously in the adult olfactory epithelium. The
ful in making connections in the olfactory bulb, whereas second hypothesis supported by the published data is that
the surviving cells are those neurons that found synaptic adult olfactory neurogenesis is a regulated process with
space at the bulb and dendritic space at the epithelial sur- fine controls over cell proliferation, cell differentiation,
face (Breipohl et al., 1986; Mackay-Sim and Kittel, and cell death. According to these models of olfactory
1991a). In other parts of the nervous system during devel- neurogenesis, the emphasis shifts from considering the
opment, immature neurons pass through a “critical period” olfactory sensory neuron being unusual for its alleged
during which they must make the correct connections. short lifespan to investigating the cellular mechanisms of
This is a period of intense competition for synaptic space. the regulation of neurogenesis (Breipohl et al., 1986;
For example, 80% of retinal ganglion cells die during Mackay-Sim and Kittel, 1991a, 1991b).
development (Williams and Herrup, 1988). It is probable
that competition for space in the bulb and at the epithelial
surface is a major determinant of whether immature neu- IV. REGULATION OF OLFACTORY
rons survive beyond 2–3 weeks. Cell death is an integral NEUROGENESIS IN VIVO
part of neurogenesis during embryonic development, and
analyses of cell death in the adult olfactory epithelium Olfactory neurogenesis has now been studied for about 60
indicate that all cell types undergo apoptosis, not just years, and it is appropriate to bring together all the available
96 Mackay-Sim

unaffected (Costanzo and Graziadei, 1983). The cell death

observed after olfactory nerve section is apoptotic
(Deckner, 1997; Holcomb et al., 1995; Michel et al., 1994)
and reaches a peak at about 1.5–2 days (Costanzo and
Graziadei, 1983; Deckner, 1997; Michel et al., 1994),
declining to control levels at 4 days after nerve section
(Deckner, 1997). The loss of neurons is accompanied by
an increase in basal cell mitosis (Graziadei, 1973), which
reaches its peak 4 days after nerve section (Camara and
Harding, 1984). The sensory neuron population recovers in
number following olfactory nerve section, but the cell
numbers and epithelial thickness reach only 60% of con-
trol levels (Costanzo and Graziadei, 1983; Samanen and
Forbes, 1984), although functional recovery is observed
(Costanzo, 1985). When the sensory neurons are destroyed
by application of zinc sulfate to the nose, the recovery in
Figure 2 Replaceable neuron model of the genesis of the olfac-
epithelial thickness is more complete, reaching control lev-
tory sensory neuron. Like other neurons, the olfactory sensory
els after about 1 month (Matulionis, 1975) followed by
neuron lives until damaged by its environment. It is replaced by a
dynamic process involving many paracrine and autocrine signals, restoration of olfactory function (Harding et al., 1978).
which eventually select a few neurons to undergo final maturation.
B. Synaptic Contact with the Olfactory Bulb
and Sensory Neuron Survival

information to develop hypotheses that can help direct The olfactory nerve is most commonly sectioned by
future research. A working hypothesis is that neurogenesis in complete removal of the olfactory bulb, their synaptic
the adult olfactory system is similar to embryonic develop- target. When this occurs, the subsequent development of
ment, following similar developmental rules that govern the the olfactory axons is seriously disrupted, and their aber-
development of other sensory systems, such as the retina rant growth can lead to neuromas within and below the
(Breipohl et al., 1986). Most of the data described above can olfactory epithelium (Schwob et al., 1994b). This can
be interpreted to support this “developmental” hypothesis: also occur when the olfactory nerve is simply transected
(1) cell death occurs at all stages of neuronal development, without removal of the olfactory bulb (Schwob et al.,
(2) developing neurons are overproduced, (3) developing 1994b). Thus, although there can be some functional
neurons pass through a “critical period” during which they recovery after olfactory nerve transection (Costanzo,
must find synaptic space at their target, (4) successful neu- 1985), the epithelium still fails to recover to control lev-
rons are dependent on their target for survival, and (5) mature els (Costanzo, 1984).
neurons are not programmed to die but may die from exter- When the olfactory bulb is removed, the many newly
nal influence. Figure 2 summarizes the cycle of neurogene- formed neurons fail to fully differentiate and die within 2
sis and some of the regulating factors described below. weeks (Carr and Farbman, 1992; Schwob et al., 1992).
Death of newly formed cells is maximal death at 6–8 days
A. Stimulation of Neurogenesis by Death after cell birth even when the animal is killed 12 days or
of the Sensory Neurons 7 weeks after olfactory bulbectomy (Carr and Farbman,
1993). Taken together these observations suggest that the
Olfactory sensory neurons are lost and then regenerate developing sensory neurons require contact with cells in
after the olfactory nerve is cut (Graziadei, 1973; Nagahara, the olfactory bulb for their survival, perhaps because bulb
1940) or the epithelium is washed with zinc sulfate cells release some trophic factor (Schwob et al., 1992).
(Margolis et al., 1974; Schultz, 1941; Smith, 1951). There This was tested directly when the mitral cells, the main
is a decrease in epithelial thickness and a decrease in the target for olfactory sensory axons, were reduced in num-
number of nuclei in the epithelium and sensory dendrites ber by sectioning of their axons in the lateral olfactory
at the epithelial surface (Costanzo and Graziadei, 1983; tract (Weiler and Farbman, 1999). This reduction in mitral
Samanen and Forbes, 1984). The loss of cells after nerve cells in the olfactory bulb stimulated basal cell prolifera-
section is confined to the basal cell and sensory neuron tion in the olfactory epithelium at all time points (up to 14
layers of the epithelium, with the supporting cell numbers months) (Weiler and Farbman, 1999).
Neurogenesis in the Adult Olfactory Neuroepithelium 97

The importance of the olfactory bulb for sensory neu- neurons (Maruniak et al., 1989, 1990). This effect was
ron differentiation was shown in organ cultures of greater rostrally than caudally, leading to the conclusion
embryonic olfactory epithelium, cultured with or that all of the inspired air passing through the one side of
without an olfactory bulb (Chuah and Farbman, 1983). the nose leads to accelerated sensory neuron death and a
Sensory neuron differentiation was assessed by measur- lack of the regenerative ability to maintain their numbers
ing the amount of olfactory marker protein (OMP) in the (Maruniak et al., 1989, 1990). The mechanism for the sen-
cultured olfactory epithelium. Under these conditions sory neuron loss is unknown, but speculations are that it is
the olfactory bulb increased the amount of OMP only if due to overuse or to toxins or to infections, supported by
it was co-cultured in contact with the olfactory epithe- evidence for a large number of polymorphonuclear leuko-
lium, indicating that physical contact between the tissues cytes on the open side after 7 and 8 months of closure
was necessary for the effect (Chuah and Farbman, 1983). (Maruniak et al., 1990).
Similarly, contact co-culture increased the numbers of
OMP-positive cells (Chuah and Farbman, 1983) and D. Regulation of Neurogenesis by the Density
the numbers of ciliated dendritic knobs at the surface of of Immature Neurons
the differentiating epithelium (Chuah et al., 1985). The
induction of sensory neuron maturation was tissue- When basal cell proliferation is observed using 3[H]-
specific: culture of the olfactory epithelium with brain, thymidine, it is evident that the density of proliferating
spinal cord, or heart did not increase OMP levels cells is not constant across the epithelial sheet. There are
(Chuah and Farbman, 1983). Taken together with the in obvious regions where proliferation is more active
vivo observations, there seems no doubt that the sur- (Graziadei and Monti Graziadei, 1978; Moulton et al.,
vival of olfactory sensory neurons depends on synaptic 1970; Weiler and Farbman, 1997). This suggests that in
contact with mitral cells in the olfactory bulb, probably the normal epithelium neurogenesis is under local control
due to a nondiffusible trophic factor provided by contact mechanisms. Another example of this is the observation
with mitral or other cells in the olfactory bulb. that in the salamander the rate of basal cell proliferation is
inversely proportional to the thickness of the epithelium
C. Neurogenesis is Regulated by Usage (Mackay-Sim and Patel, 1984). In this species the olfac-
tory epithelium varies in thickness from anterior to poste-
If one naris is closed during early development, there are rior, being thicker anteriorly. Quantitative analysis of the
marked differences in the olfactory epithelia from the con- cell types in this epithelium demonstrated that the only
trol and occluded sides (Farbman et al., 1988) (see Chapter cell type whose numbers increased with epithelial thick-
29). The thickness of the epithelium is reduced on the ness were the immature neurons (Mackay-Sim et al.,
occluded side, accompanied by a reduction in cell number 1988). The numbers of mature sensory neurons, support-
and a reduction in the number of proliferating basal cells ing cells, and basal cells were all constant and indepen-
(Farbman et al., 1988). Despite these differences there was dent of epithelial thickness. Therefore, the rate of basal
no effect on the number of sensory neurons, indicated by cell proliferation was effectively inversely proportional to
the numbers of olfactory dendrites at the epithelial surface the number of immature sensory neurons, leading to the
(Farbman et al., 1988). These observations suggest that the conclusion that the developing neurons exert an inhibitory
rate of basal cell proliferation was reduced because of a influence basal cell proliferation (Mackay-Sim et al.,
protective effect of naris occlusion with a reduction in the 1988). This conclusion is supported by a report that the
access of infectious or toxic agents (Farbman et al., 1988). rate of proliferation in vitro was reduced when precursor
A lack of toxic environmental effects was also proposed cells were co-cultured with sensory neurons (Mumm
for the observation of long-lived sensory neurons living in et al., 1996).
a clean air environment (Hinds et al., 1984).
When the naris is closed in adult mice, it is the open side E. Regulation of Neurogenesis by Thyroxine
that is reduced in thickness (Maruniak et al., 1989). The
reduction in thickness is associated with loss of sensory Adult mice made hypothyroid exhibit olfactory dysfunc-
neurons (Maruniak et al., 1989) in marked contrast to the tion, from which they recover if normal thyroxine levels are
effects of naris occlusion during development in which the restored (Beard and Mackay-Sim, 1987). After 7 weeks of
mature neurons are unaffected (Farbman et al., 1988). The hypothyroidism there is a reduction in epithelial thickness
loss of sensory neurons was not accompanied by loss of without loss of sensory neurons (Mackay-Sim and Beard,
other cell types in the epithelium, leading to the conclusion 1987). The reduction in epithelial thickness is due to loss of
that naris occlusion had an effect specifically on the sensory immature neurons (Mackay-Sim and Beard, 1987).
98 Mackay-Sim

F. Cell Death as an Integral Part of Neurogenesis apoptosis in vitro (Farbman et al., 1999). Several enzymes
of the caspase family, enzymes known to be involved in
As indicated above, a single intraperitoneal injection of TNF-–induced cell death in other cell types, are present
3[H]-thymidine labels a large number of basal cells in the in the adult olfactory epithelium, and inhibition of these
olfactory epithelium (Graziadei and Monti Graziadei, enzymes blocks apoptosis in a dose-dependent manner in
1978; Hinds et al., 1984; Mackay-Sim and Kittel, 1991a; olfactory epithelial cultures (Suzuki and Farbman, 2000).
Moulton et al., 1970) indicating a high rate of prolifer- In summary, there is strong evidence that the cell
ation in the normal, undisturbed epithelium. Despite this death seen in the undisturbed olfactory epithelium and
continual proliferation of neuronal precursors, there is no after olfactory nerve section is apoptosis, or programmed
continual increase in epithelial thickness in the adult rat cell death, and there is evidence for autocrine or
from 60–330 days of age (Hinds and McNelly, 1981; paracrine signaling pathways involved in apoptosis in the
Weiler and Farbman, 1997). In rat there is an increase in olfactory epithelium. These observations further confirm
surface area of the epithelium in adulthood (Hinds and that olfactory neurogenesis is a regulated process with an
McNelly, 1981; Paternostro and Meisami, 1993; Weiler intricate balance between production of new neurons and
and Farbman, 1997), and it is probable that basal cell pro- death of all cell types to maintain equilibrium within the
liferation at the edges of the epithelium could contribute epithelium.
to its expansion. There is no evidence that dividing basal
cells of their progeny migrate very far laterally to popu- G. Neurogenesis and Aging
late new regions. Given the lack of increase in epithelial
thickness, it follows that the constant proliferation must Olfactory neurogenesis continues throughout adult life,
be balanced by a concomitant constant cell death. This is observed in aged rat (Loo et al., 1996; Weiler and
supported by evidence that from 1 to 17 weeks of age, the Farbman, 1997) and human (Murrell et al., 1996; Wolozin
rates of basal cell proliferation and of cell death in the et al., 1992). As rats and humans age, there are histopatho-
epithelium show a similar age-related decline (Fung logical changes that suggest that neurogenesis is not able
et al., 1997). to fully maintain the epithelium. In rat the anterodorsal
It is now evident that cell death occurs at all stages of region of the epithelium shows a greater average number
development after basal cell division. This cell death in the of proliferating basal cells but also a greater level of intra-
normal undisturbed epithelium is apoptotic (Magrassi and and interanimal variability of basal cell proliferation (Loo
Graziadei, 1995), similar to death induced by olfactory et al., 1996; Weiler and Farbman, 1997). Histologically,
nerve section. By labeling dividing cells with 3[H]-thymi- this region also appears disordered with a reduction of
dine and looking for thymidine-labeled pyknotic nuclei, it lamination, a loss of neurons, and increased proliferation
was shown that cell death can occur as early as one day of supporting cells (Loo et al., 1996). These changes are
after birth (Carr and Farbman, 1993). When cells were consistent with damage to this area, with a concomitant
identified with cell-type specific markers and double- attempt to reconstitute the sensory neuron population (Loo
labeled to identify dying cells, it was clear that the et al., 1996). Similar changes are observed in olfactory
apoptotic cells could be horizontal basal cells, globose epithelium sampled from adult humans. In aged humans
basal cells, immature neurons, or mature neurons the olfactory epithelium may show a reduction in thick-
(Holcomb et al., 1995; Mahalik, 1996). ness, a reduction in sensory neuron number, and patchy
Apoptotic cell death is a highly regulated process (Vaux distribution of olfactory epithelium within the respiratory
and Strasser, 1996). Evidence for the involvement of epithelium (Naessen, 1971; Nakashima et al., 1984, 1985).
regulatory genes in olfactory neurogenesis are that overex- It is suggested that the pathology seen in the aged olfac-
pression of the bcl-2 gene protects the adult animal from tory epithelium resembles the changes induced in the open
apoptotic death after olfactory nerve transection (Jourdan side of adult animals subject to unilateral naris occlusion
et al., 1998). Further evidence for apoptotic regulation of (Loo et al., 1996). When one naris is closed, sensory neu-
neurogenesis is given by experiments describing the rons are lost on the open side, resulting, in places, in an
presence of apoptotic regulatory molecules in the adult epithelium composed of supporting cells only (Maruniak
olfactory epithelium. The apoptotic cascade can be et al., 1989, 1990; Walters et al., 1992). These observa-
induced by activation of the cell surface receptors Fas and tions suggest that a similar mechanism may act during
tumor necrosis factor (TNF) receptor-1 by their ligands aging, which is accelerated by naris occlusion, that is, the
FasL and TNF-. Both the receptors and their ligands were most exposed regions of the olfactory epithelium (e.g., the
observed in olfactory epithelium in vivo (Farbman et al., anterodorsal region in the rat) may be subject to overusage
1999), and addition of either FasL and TNF- induced or airborne factors that continually stimulate neurogenesis.
Neurogenesis in the Adult Olfactory Neuroepithelium 99

With age the level of neurogenesis may not be able to be hypothyroidism suggests that thyroxine is involved in pro-
maintained, thus leading to a reduced capacity for repair moting survival of these cells.
and replacement of olfactory epithelium for respiratory Another form of “regulation” of neurogenesis occurs via
epithelium. the influence of the environment on the sensory neurons.
Already discussed is the use of nasal lavage of zinc sulfate
H. Summary to experimentally destroy the sensory neurons. Inhaled
toxic gases can also destroy the olfactory epithelium and
As the discussion above indicates, there is increasing the neurons within it. Inhalation of N-methyl-formimino-
evidence for various regulatory controls on olfactory neu- methylester and methyl bromide led to a temporary loss of
rogenesis in the adult. In the normal epithelium these smell and a reversible loss of the sensory neurons (Hurtt et
controls act to maintain the epithelial thickness and the al., 1988; Rehn et al., 1981; Schmidt et al., 1984). Even
number of sensory neurons and to balance the rate of cell nasal lavage with a large protein conjugate, wheat germ
birth with the rate of cell death. From the manipulations of agglutinin-horseradish peroxidase, led to loss of sensory
olfactory neurogenesis in vivo, the action of various neurons from the epithelium and stimulated basal cell pro-
unknown but potential regulatory factors can be implied. liferation (Moon and Baker, 1998). Taken together these
Table 1 summarizes these. The regulatory pathways sug- observations indicate that the sensory neurons are vulnera-
gested in Table 1 are only speculative, but they can explain ble to damage from inhaled molecules leading to sensory
the observations. Clearly basal cell proliferation can be neuron death and stimulation of neurogenesis.
regulated up and down by the density of sensory neurons It is informative to speculate that the high rate of pro-
and immature neurons. Such regulation could be achieved liferation of neuronal precursors, the high mortality of the
by the release of a stimulatory factor when neurons die and immature neurons, and the low rate of replacement of sen-
an inhibitory factor while they live. Regulation could be sory neurons may be related to the requirement to balance
achieved with a single stimulatory or inhibitory factor, but the birth and death of sensory neurons expressing individ-
two would provide finer control and would allow a greater ual receptor genes.
variation in local control mechanisms. Similarly, multiple Each olfactory neuron appears to expresses a single
stimulatory or inhibitory factors would provide even odorant receptor gene (Buck and Axel, 1991), which is
greater variation and some redundancy. In addition to fac- involved in targeting the axon to restricted glomeruli in the
tors regulating basal cell proliferation, the evidence sug- olfactory bulb (Vassar et al., 1994). The receptor gene is
gests that there are also factors promoting survival of the expressed early in differentiation before the sensory neu-
sensory neurons, as shown by the continuing death of neu- ron establishes connections with the olfactory bulb
rons unable to make contact with their target, the olfactory (Leibovici et al., 1996), and its expression in the nasal cav-
bulb. Similarly, the loss of immature neurons during ity is independent of the presence of the olfactory bulb in

Table 1 Regulation of Basal Cell Proliferation In Vivo

Experiment Stimulus Effect on basal cell proliferation Possible regulatory pathway
Olfactory nerve cut Death of mature neurons Increased Proliferating factor released from
dying sensory neurons
Chemical destruction Death of neurons Increased Proliferating factor released from
of neurons dying sensory neurons
Destruction of Death of neurons Increased Lack of trophic factor in olfactory
mitral cells bulb leads to death of sensory neu-
rons and release of proliferating
factor
Naris occlusion Increase in usage/ Increased Proliferating factor released from
open side loss of neurons dying sensory neurons
Naris occlusion Reduction in usage/ Decreased Lack of proliferating factor
closed side no loss of sensory neurons
Epithelial thickness Increased density
of immature neurons Decreased Lack of proliferating factor or
of anti-proliferative factor
100 Mackay-Sim

the adult (Konzelmann et al., 1998; Margalit and Lancet, regulatory pathway. Another pathway could act via the
1993; Strotmann et al., 1995; Sullivan et al., 1995). Cells supporting cells. The supporting cells surround the den-
expressing each receptor gene are expressed stochastically drites of the sensory neurons (Breipohl et al., 1974;
within restricted regions of the olfactory epithelium (Buck Graziadei and Monti Graziadei, 1979) with which they
and Axel, 1991; Ressler et al., 1993; Strotmann et al., make tight junctions close to the surface (Menco, 1980).
1996), so the question arises as to how the numbers of Therefore, the supporting cells are in a position to monitor
neurons expressing each gene are maintained. Perhaps the the local density of sensory neurons and release factors to
most critical property of a developing neuron in the adult regulate proliferation of basal cells or differentiation of
is whether its receptor gene matches the receptor gene of neuronal precursors and immature neurons. Other cells
the dying neuron it replaces. With approximately 1000 that could be important for the survival and differentiation
genes distributed in four epithelial zones, dying sensory of the developing neurons are the horizontal basal cell and
neurons could express one of 250 receptor genes. Is it pos- the olfactory nerve ensheathing cell. The horizontal basal
sible that 250 developing neurons are necessary for each cell wraps around the axons before they leave the epitheli-
dying neuron to be replaced by a cell expressing the cor- um (Holbrook et al., 1995), and the ensheathing cells do so
rect receptor gene? The selection of the successful devel- when they enter the lamina propria and guide them to the
oping neuron may be regulated by its finding synaptic olfactory bulb (Doucette, 1984; Gong et al., 1994). Either
space in target glomeruli in the olfactory bulb, although all of these cell types could regulate neuronal survival and dif-
successful neurons would also require dendritic space at ferentiation.
the epithelial surface. According to this argument, there Broadly speaking, there are two types of signals that
could be 249 unsuccessful neuronal precursors for every could regulate olfactory neurogenesis at the local or cellu-
cell that accomplishes differentiation into a functioning lar level: diffusible and fixed. Growth factors are diffusible
sensory neuron. signals which can have paracrine or autocrine actions.
Fixed signals include physical interactions via direct cell
surface contacts and indirect contacts through the extracel-
V. MOLECULAR REGULATION OF lular matrix. Such signals act via cell surface integrin
OLFACTORY NEUROGENESIS receptors, and cells can be switched from growth to apop-
tosis simply by changing their shape (Chen et al., 1997).
Presumably the principles by which the olfactory epithe- Extracellular signals have not been extensively investi-
lium is maintained in the adult animal are similar to the gated in olfactory neurogenesis; much more is known
principles by which it develops in the embryo, namely, the about growth factors.
cells are subject to autocrine and paracrine signals as well
as cell-cell contact signals, which maintain or induce dif- A. Growth Factors and Receptors Present in
ferent cells types. In the olfactory epithelium it is possible Olfactory Epithelium
that signals arise from any or all of the cell types (horizon-
tal and globose basal cells, immature and mature neurons, Growth factors are proteins or peptides found in tissues
supporting cells), but signaling molecules will not be con- which exert highly specific effects at very low concentra-
fined to the epithelium. In addition to the putative signals tions. Each growth factor acts through a specific cell-
from the olfactory bulb, there may be signals from the surface receptor or set of receptors, which convey signals
olfactory nerve ensheathing cells. Furthermore, because via kinases and other second-messenger systems. In the
the surface density of sensory neuron dendrites is con- nervous system the first growth factor to be isolated was
trolled and stable (Mackay-Sim and Kittel, 1991b), it is nerve growth factor (NGF), and its initially defined effect
possible that signals may be present in the mucus, acting was the promotion of neuron survival (Levi-Montalcini,
as target-derived factors for the dendrites. In that case 1987). “Growth factor” is now a term for increasing
signaling molecules may arise from the Bowman’s glands numbers of molecules that regulate cell proliferation, cell
and other cells that produce the olfactory mucus. differentiation, and cell death. Growth factors may have
In considering the signaling molecules that regulate the multiple actions on multiple cell types. For example,
different aspects of olfactory neurogenesis (proliferation, platelet-derived growth factor (PDGF), named for the cell
differentiation, survival, death), it is important to be open type it was originally identified in, can act on fibroblasts,
to the possibilities of multiple factors operating in multiple smooth muscle, and neuroglia. In other parts of the nervous
pathways. For example, when sensory neurons die they system it is evident that the function of growth factors and
may release a factor that stimulates proliferation of the their influence on individual cells can vary with stages of
basal precursor cells, but that is not the only possible development and the actions of single growth factors can
Neurogenesis in the Adult Olfactory Neuroepithelium 101

be different in the presence of others. The overall picture A large number of growth factors and their receptors
of the functions of growth factors is increasingly complex: have been identified in the olfactory epithelium (Table 2).
neurons can require different growth factors at specific Although dopamine is not a peptide or protein, nor is it
stages of development and can require several growth usually classified as a growth factor, it is included here
factors simultaneously. A recent review presents a fuller because of its growth factor–like effects in vitro (see
discussion of growth factors and their roles in the olfac- below). For many of growth factors the cell types that
tory system (Mackay-Sim and Chuah, 2000). express them are not identified. The exceptions are ciliary

Table 2 Growth Factors and Receptors in the Olfactory Mucosa

Growth factor family Ligands Ref. Receptors Ref.
Cytokines CNTF Buckland and Cunningham,
1999
Dopamine DA Lucero and Squires, 1998 D2 Coronas et al., 1997b;
Féron et al., 1999c;
Koster et al., 1999
EGF family TGF Farbman and Buchholz, 1996 EGFR Farbman et al., 1994;
Holbrook et al., 1995;
Rama Krishna et al.,
1996; Salehi-Ashtiani
and Farbman, 1996
NDF Salehi-Ashtiani and ErbB2 Salehi-Ashtiani and
Farbman, 1996 Farbman, 1996
ErbB3 Perroteau et al., 1998
ErbB4 Perroteau et al., 1998
FGF family FGF2 Chuah and Teague, 1999; FGFR1 DeHamer et al., 1994
Goldstein et al., 1997;
Hsu et al., 2001
FGFR1b,c Hsu et al., 2001
FGFR2 DeHamer et al., 1994
FGFR2b,c Hsu et al., 2001
FGFR3b,c Hsu et al., 2001
GDNF family GDNF Buckland and Cunningham, Ret Nosrat et al., 1997
1999; Nosrat et al., 1996;
Woodhall et al., 2001
GFR1 Nosrat et al., 1997;
Woodhall et al., 2001
GFR2 Woodhall et al., 2001
IGF family IGF-I Ayer-LeLievre et al., 1991 IGFR-I Pixley et al., 1998
IGF-II Ayer-LeLievre et al., 1991 IGFBP-2 Bondy and Lee, 1993;
Federico et al., 1999
IGFBP-3 Federico et al., 1999
IGFBP-4 Federico et al., 1999
IGFBP-5 Bondy and Lee, 1993
Neurotrophins NGF Ayer-LeLievre et al., 1983; TrkA Miwa et al.,1998;
Williams and Rush, 1988; Roskams et al., 1996
Woodhall et al., 2001
BDNF Buckland and Cunningham, TrkB Roskams et al., 1996;
1999; Woodhall et al., 2001 Woodhall et al., 2001
TrkC Roskams et al., 1996
PDGF family PDGFA Orr-Urtreger and Lonai, 1992 PDGFR Lee et al., 1990; Orr-
Urtreger and Lonai, 1992
TGF family BMP2,4,7 Shou et al., 2000 BMPR-Ib Zhang et al., 1998
ActR-Ib Verscheuren et al., 1995
102 Mackay-Sim

neurotrophic factor (CNTF) (basal cells and neurons) growth factor must be shown to have a specific action on
(Buckland and Cunningham, 1999), dopamine (mucus) the target cell. In defining the actions of growth factors, in
(Lucero and Squires, 1998), dopamine D2 receptor (basal vitro techniques are used because the simplification of cell
cells and neurons) (Féron et al., 1999c; Koster et al., 1999), and tissue culture allows more variables to be controlled.
epidermal growth factor (EGF) receptor and transforming The causative pathway becomes less obvious as the system
growth factor alpha (TGF) (horizontal basal cells and increases in complexity. For example, does the growth fac-
supporting cells) (Farbman and Buchholz, 1996; Farbman tor act directly or via a neighboring cell? Even in vitro,
et al., 1994; Holbrook et al., 1995; Rama Krishna et al., under relatively simple conditions, it can be difficult to dis-
1996), fibroblast growth factor 2(FGF2) (supporting cells, tinguish between the possible actions of a growth factor.
neurons, basal cells) (Chuah and Teague, 1999; Gall et al., For example, an increase in the number of neurons in a cul-
1994; Goldstein et al., 1997; Hsu et al., 2001; Matsuyama ture (or in the tissue) induced by an added growth factor
et al., 1992), glial cell line–derived growth factor (GDNF) may be caused by increased proliferation, increased differ-
(neurons) (Buckland and Cunningham, 1999), insulin-like entiation, increased survival, or a combination of these.
growth factor I (IGF-I) and binding proteins 2–4 (mucus) Table 3 summarizes the existing data on growth factor
(Federico et al., 1999), tyrosine kinase A (TrkA) (horizon- function in olfactory neurogenesis. These data derive from
tal basal cells) (Miwa et al., 1998; Roskams et al., 1996), culture systems of different complexity with variations in
TrkB and TrkC (neurons) (Roskams et al., 1996), nerve the cell types present and in the culture media.
growth factor (NGF) (neurons) (Aiba et al., 1993; Consequently the conclusions drawn from these studies
Roskams et al., 1996; Williams and Rush, 1988), and should be considered as working hypotheses. Nonetheless,
brain-derived growth factor (BDNF) (horizontal basal converging data suggest distinct and definable roles for
cells) (Buckland and Cunningham, 1999). The variety of TGF, FGF2, and TGF2 in olfactory neurogenesis.
growth factors and receptors present in the olfactory Proliferation of the horizontal basal cells is stimulated by
epithelium and the variation in expression by the different EGF and the related TGF (Farbman and Buchholz, 1996;
olfactory cell types suggests a rich complexity in the regu- Féron et al., 1999a; Satoh and Takeuchi, 1995). Note that
lation of olfactory neurogenesis. TGF, but not EGF, is expressed in the olfactory epithe-
lium, and their cognate receptor (EGFR) is also present
B. Growth Factor Function in Olfactory (see above).
Epithelium FGF2 stimulates proliferation of a “stem cell”
(DeHamer et al., 1994) and the globose basal cell
Defining the actions of growth factors can be very difficult. (Newman et al., 2000). In a globose basal cell–like cell line
In order to be certain that a specific growth factor functions FGF2 stimulated proliferation and inhibited differentiation
in olfactory neurogenesis, the growth factor must be shown (Goldstein et al., 1997). In contrast, in a human olfactory
to be available to the putative target cells, the target cell cell line, which also produced FGF2, this growth factor
must be shown to express the appropriate receptors, and the stimulated proliferation and induced differentiation

Table 3 Growth Factors Active in Olfactory Epithelium

Growth factor Cell type Action Ref.
BMP2/4/7 Neuronal precursors Inhibit proliferation Shou et al., 1999, 2000
BMP4 Immature neurons Promotes survival Shou et al., 2000
Dopamine Immature neurons Stimulates differentiation Féron et al., 1999c
EGF/TGF Horizontal basal cells Stimulates proliferation Farbman and Buchholz, 1996;
Farbman et al., 1994
EGF/TGF Supporting cells Stimulates proliferation Farbman and Buchholz, 1996
FGF2 Globose basal cells/ Stimulates proliferation DeHamer et al., 1994; Newman
neuronal precursors et al., 2000
PDGF Mature neurons Promotes survival Newman et al., 2000
TGF2 Globose basal cells/ Stimulates differentiation Mahanthappa and Schwarting, 1993;
neuronal precursors Newman et al., 2000
Neurogenesis in the Adult Olfactory Neuroepithelium 103

(Ensoli et al., 1998). FGF2 also induced differentiation in 1998), and it modulates an inwardly rectifying current in
explant cultures of mouse and human olfactory epithelium sensory neurons (Vargas and Lucero, 1999) via adenylyl
(MacDonald et al., 1996; Murrell et al., 1996), although cyclase (Coronas et al., 1999; Mania-Farnell et al.,
we now believe this to have been an indirect effect via 1993). These observations suggest that dopamine present
stimulation of globose basal cell proliferation (Newman in the mucus could act as signal to the developing neu-
et al., 2000). All studies are in agreement that FGF2 stim- ron that its dendrite has reached the epithelial surface,
ulates proliferation of a neuronal precursor, both in pri- thereby triggering cessation of dendritic extension and
mary culture and as a cell line; it remains to be proven initiation of cilial growth. Dopamine is also present in
whether FGF2 also has a differentiating effect. In vivo, the glomerulus, the site of axon termination (Davis and
appropriate FGF receptor subtypes (FGFR1 and FGFR2) Macrides, 1983; Halasz et al., 1977), and it could act
are present and FGF2 immunoreactivity is also present in there as a signal to the developing neuron that its axon
a number of cell types (Hsu et al., 2001). has reached it target. There is evidently a positive feed-
TGF2 induces differentiation of neuronal precursors back when the axon makes connection with the bulb
(Mahanthappa and Schwarting, 1993), a keratin-positive because dopamine and its synthetic enzyme, tyrosine
basal cell line (Satoh and Takeuchi, 1995) and the globose hydroxylase, are selectively downregulated by chemical
basal cell (Newman et al., 2000). We have identified mRNA destruction of the sensory neurons and upregulated when
for TGF-receptor subtypes I, II, and III in the olfactory sensory innervation returns (Baker et al., 1983; Nadi
epithelium, although the cellular distribution is currently et al., 1981). Even occlusion of the naris can reduce tyro-
unknown (P. Hsu and A. Mackay-Sim, unpublished). sine hydroxylase and dopamine expression in the olfac-
In vivo (Mackay-Sim and Patel, 1984) and in vitro tory bulb (Baker et al., 1993; Philpot et al., 1998). In
experiments (Mumm et al., 1996) indicated that neurons or vitro experiments indicate that the upregulation of
immature neurons exert an inhibitory effect on basal cell tyrosine hydroxylase by sensory neurons acts via odor-
proliferation. It is possible that this inhibition is mediated ant-stimulated glutamate release by the sensory neuron
via BMPs and their receptors. Recent experiments indicate terminals (Puche and Shipley, 1999). A model emerging
that the bone morphogenic proteins (BMPs) 2, 4, and 7 from all these data is that dopamine may signal that the
can inhibit proliferation of neuronal precursors in vitro dendrite and axon have reached their targets and are
(Shou et al., 1999). BMP receptor subtype Ib is present in active. This is then reinforced by odorant-stimulated
embryonic olfactory epithelium (Zhang et al., 1998), and activity in the dendrite and subsequent synaptic activity
we have identified mRNA for BMP receptor subtypes Ia, in the bulb, leading to dopamine synthesis in the
Ib, and II in adult olfactory epithelium (P. Hsu and A. periglomerular cells. In the epithelium the level of
Mackay-Sim, unpublished). In the embryo, BMPs 2, 4, and dopamine is regulated by activity in the trigeminal nerve
7 are expressed by cells in the lamina propria beneath the (Lucero and Squires, 1998), whose activity is stimulated
olfactory epithelium and noggin, a BMP antagonist, inhib- by odorant stimulation (Cain, 1974; Doty, 1975; Silver
ited olfactory neurogenesis in embryonic cultures (Shou et and Moulton, 1982). It is possible, therefore, that
al., 2000). In these cultures low concentrations of BMP4 dopamine may act continually as a trophic factor at both
but not BMP7 promoted survival of newly generated olfac- ends of the active sensory neuron.
tory receptor neurons (Shou et al., 2000). These results Other growth factors have been implicated in olfac-
suggest both antiproliferative and neuronal survival roles tory neurogenesis, although their functions are less well
for BMPs, at least during embryogenesis. Their roles in defined. IGF-I is present in human olfactory mucus
adult olfactory epithelium remain to be defined. (Federico et al., 1999), and infusion of IGF-I into the
Dopamine, although it is not a traditional growth fac- external naris increased the thickness of the olfactory
tor, was shown to induce apoptosis and differentiation of epithelium and increased the number of proliferating
an olfactory cell line (Coronas et al., 1997a) and to pro- cells (Pixley et al., 1998). Of the neurotrophins, BDNF
mote differentiation in explant culture of olfactory and neurotrophin 3 (NT-3) but not NGF, increased the
epithelium of adult mouse (Féron et al., 1999c). In numbers of immature neurons in primary cultures of
human explant cultures dopamine inhibited mitosis and olfactory neurons (Holcomb et al., 1995; Liu et al.,
induced apoptosis (Féron et al., 1996b). These effects 1998; Roskams et al., 1996). Given the distribution of
were mediated via the dopamine D2 receptor (Féron the neurotrophin receptors (see above), it is not surpris-
et al., 1999c), which has been identified in the neuronal ing that sensory neurons were not affected by the pres-
layer of the olfactory epithelium (Féron et al., 1999b; ence of NGF. The increased cell numbers may have
Koster et al., 1999). Dopamine is present in the mucus resulted from the survival-promoting effects of BDNF
above the olfactory epithelium (Lucero and Squires, and NT-3. It is interesting to note that in co-cultures of
104 Mackay-Sim

Figure 3 Growth factor regulation of cell dynamics in the olfactory mucosa. Most of these functions have been demonstrated in vitro.
The functions of the neurotrophins (NGF, BDNF, and NT3) and IGF-I are inferred from the presence of their receptors on the cell types
indicated. A transition from horizontal basal cell to globose basal cell is disputed.

neurons and ensheathing cells, withdrawal of NGF stimulates ensheathing cell differentiation (Key et al.,
resulted in a dramatic decrease in neuron number 1996). Figure 3 summarizes the actions of various growth
(Bakardjiev, 1997). This may have been an indirect factors in the olfactory epithelium and lamina propria.
effect via loss of the ensheathing cells, which have the
low-affinity NGF receptor (see below). Finally, the C. Regulation of Neurogenesis by Ensheathing Cells
cytokine leukemia inhibitory factor (LIF) stimulated
proliferation of a neuronal precursor population (Satoh The remarkable capacity of the olfactory epithelium to
and Yoshida, 1997). regenerate lies, in part, in the properties of the olfactory
Because of the importance of olfactory ensheathing nerve ensheathing glia that accompany the sensory axons
cells in the promotion of sensory neuron differentiation from the epithelium to the bulb. Olfactory ensheathing glia
(see below), it is interesting to consider the evidence for derive from the olfactory placode (Chuah and Au, 1991;
growth factor regulation of ensheathing cell growth and Doucette, 1989; Farbman and Squinto, 1985) and are pre-
development. Of the growth factors investigated, the activ- sent in the olfactory nerve and the outer region of the olfac-
ity of the neuregulins is the most well defined. The neu tory bulb in the adult (Doucette, 1984; Franceschini and
differentiation factors (NDF -1, -2, and -3) stimulate prolif- Barnett, 1996; Marin-Padilla and Amieva, 1989; Valverde
eration of olfactory ensheathing cells and the ensheathing and Lopez-Mascaraque, 1991). In the adult the morphology
cells express ErbB2 receptors (Pollock et al., 1999). FGF2 of the ensheathing cells appears homogeneous (Doucette,
also stimulates proliferation of ensheathing cells (Chuah 1991), whereas during development two morphotypes are
and Teague, 1999). Another neuregulin, glial growth factor evident (Cuschieri and Bannister, 1975; Doucette, 1989;
2 (GGF2), is weakly proliferative (Chuah et al., 2000). Valverde et al., 1993). Two types of ensheathing cells are
GGF2 induces differentiation of ensheathing cells and is seen in cultures from embryonic (Kafitz and Greer, 1999)
expressed by them (Chuah et al., 2000), and FGF1 also and newborn animals (Pixley, 1992).
Neurogenesis in the Adult Olfactory Neuroepithelium 105

Olfactory ensheathing glia have properties that are 1994) and promote their extension via soluble factors
similar both to peripheral Schwann cells and to astroglia (Kafitz and Greer, 1999) and extracellular matrix (Tisay
of the central nervous system. Although they do not and Key, 1999). Ensheathing glia promote axonal exten-
myelinate the olfactory nerve, like Schwann cells, they sion of retinal ganglion cells (Goodman et al., 1993) and
evidently allow and promote axon growth and can myeli- can myelinate neurites from the dorsal root ganglion
nate dorsal root neurites in vitro (Devon and Doucette, (Devon and Doucette, 1992). These observation indicate
1992). Unlike Schwann cells but like astroglia, they exist that the supportive role of olfactory ensheathing glia is
in the central nervous system. The olfactory ensheathing attributable not simply to specific interactions with olfac-
glia can be recognized, and distinguished from astrocytes tory sensory axons, but to interactions with growing or
and Schwann cells, by the expression of a combination of regenerating axons in general. This ability is in evidence in
proteins. Like Schwann cells they express the calcium- their ability to assist the regeneration of dorsal root axons
binding protein S-100 (Pixley, 1992) and the p75 low- to reenter the dorsal horns (Ramon-Cueto and Nieto-
affinity neurotrophin receptor (p75NTR Gong et al., 1994; Sampedro, 1994), to assist remyelination in the descending
Pixley, 1992; Roskams et al., 1996; Turner and Perez- motor axons after nerve crush (Imaizumi et al., 1998) and
Polo, 1992; Vickland et al., 1991). Like astroglia and non- electrolytic lesions (Li et al., 1997; Li et al., 1998), and,
myelinating Schwann cells, they express the glial acidic remarkably, to promote spinal regrowth and behavioral
fibrillary protein (GFAP) (Pixley, 1992). There appear to recovery after complete spinal transection (Ramon-Cueto
be two types of olfactory ensheathing glia, one of which et al., 1998, 2000).
expresses both GFAP and S-100, the other only GEAP, The regenerative properties of olfactory ensheathing
detected at high antibody dilution (Pixley, 1992). In vitro glia probably arise from the variety of growth factors
these two glial types have distinguishable morphologies, and extracellular matrix molecules that they secrete
the former being spindly and bipolar, and the latter, flatter (Liesi, 1985). In addition to laminin, ensheathing cells
(Pixley, 1992). Others have defined two types of olfac- express cell-surface antigens and extracellular
tory ensheathing glia based in their expression of p75NTR matrix–associated molecules such as L1, laminin, colla-
and polysialated neural cell adhesion molecule (E- gen IV, NCAM, heparan-sulfate proteoglycans, and glia-
NCAM) (Franceschini and Barnett, 1996). During devel- derived nexin (Chuah and Au, 1992; Doucette, 1990;
opment S-100 immunoreactivity emerges before GFAP Liesi, 1985; Miragall and Dermietzel, 1992; Miragall
immunoreactivity (Astic et al., 1998), but in the adult et al., 1988, 1989, 1992; Reinhard et al., 1988; Scotti et
olfactory nerve and bulb, all four antigens (S-100, GFAP, al., 1994; Treloar et al., 1996; Whitesides and LaMantia,
E-NCAM, and p75NTR) are present (Franceschini and 1996). Several of these extracellular molecules are
Barnett, 1996). Curiously, in vitro p75NTR-immunoreac- important for survival and differentiation of sensory neu-
tive cells were reported to be GFAP-immunoreactive but rons in vitro. Olfactory ensheathing cells are also a rich
not S-100–immunoreactive, even though GFAP and S-100 source of growth factors: NGF, (Woodhall et al., 2001)
are co-expressed in many cells (Kafitz and Greer, 1999). BDNF (Woodhall et al., 2001), FGF1 (Key et al., 1996),
This is an unusual finding considering that in some cul- FGF2 (Chuah and Teague, 1999; Gall et al., 1994;
tures virtually all cells appear to express all three antigens Matsuyama et al., 1992), GDNF (Woodhall et al.,
(Franceschini and Barnett, 1996; Tisay and Key, 1999). 2001), and CNTF (Guthrie et al., 1997). This list is not
These discrepancies may be explained by the ages of the exhaustive and no doubt has more members because sev-
rats from which the ensheathing cells arise: embryonic eral growth factors have been identified in the lamina
day 15 (Kafitz and Greer, 1999), a stage at which S-100 propria of the olfactory mucosa without identifying the
but not GFAP is expressed in vivo (Astic et al., 1998), and expressing cells (for a recent review, see Mackay-Sim
postnatal day 7 (Tisay and Key, 1999) and adult and Chuah, 2000).
(Franceschini and Barnett, 1996), by which time both
antigens are expressed in the olfactory nerve (Astic et al., D. Summary
1998). Culture conditions can also affect the differentia-
tion of ensheathing cells (Franceschini and Barnett, Recent research is beginning to flesh out the molecular
1996). signals that regulate proliferation, differentiation, survival,
Olfactory sensory neurons, when given the choice in and death in olfactory neurogenesis. There are now several
vitro, preferentially grow on ensheathing glia (Chuah and growth factors whose roles have been defined in vitro and
Au, 1994; Tisay and Key, 1999). The ensheathing glia whose presence and the presence of whose receptors are
extend processes around neurites in vitro (Chuah and Au, confirmed in the epithelium. The paracrine and autocrine
106 Mackay-Sim

pathways by which these growth factors act are still being Takeuchi, 1995). In contrast, experiments using in vivo
established but will include cell-cell communication both retroviral labeling of proliferating cells in undisturbed
within the epithelium and between sensory neurons and epithelium suggest that neurons arise only from glo-
the ensheathing cells within the lamina propria, as well as bose basal cells (Caggiano et al., 1994; Schwob et al.,
sensory neurons and their synaptic targets in the olfactory 1994a). There is no direct evidence linking immuno-
bulb. As in other parts of the nervous system, extracellular logical phenotype with a role as progenitor or stem cell.
factors and physical cell surface interactions are also Mackay-Sim and Kittel (Mackay-Sim and Kittel,
expected to play important roles as well. 1991a) identified the stem cell as an asymmetrically
dividing cell, located on the basement membrane, but
did not identify these cells immunologically or mor-
VI. CELL LINEAGE IN OLFACTORY phologically as horizontal basal cells. Both
NEUROGENESIS Mahanthappa and Schwarting (1993), and Satoh and
Takeuchi (1995) used only one basal cell antibody (ker-
The olfactory epithelium is a pseudo-stratified, columnar atin) and one neuronal antibody (NCAM) to character-
epithelium containing four cell types: the sensory neuron; ize the cells in vitro. They did not account for either
the supporting cell, a glial-like cell; the globose basal cell; globose basal cells or supporting cells in their cultures.
and the horizontal basal cell. These cells can be identified In other words, the observed neurons may have arisen
immunologically. The supporting cell is identified with from unidentified cells but ascribed to the keratin-pos-
the antibody SUS1 (Hempstead and Morgan, 1983). itive horizontal basal cells. In the retroviral labeling
Mature olfactory sensory neurons are distinguished from experiments of undisturbed epithelium (Caggiano
immature neurons by their expression of olfactory marker et al., 1994; Schwob et al., 1994a), the animals were
protein (OMP) (Margolis, 1985). Immature and mature killed too early to observe the division of a horizontal
neurons express neuron-specific -tubulin and some iso- basal cell with a cell cycle period of about 50 days
forms of NCAM, while immature neurons express other (Mackay-Sim and Kittel, 1991a). The data from the
isoforms of NCAM, GAP43, as well as -tubulin retroviral lineage study of methyl bromide–treated
(Goldstein and Schwob, 1996). All globose basal cells are epithelium also do not allow one to distinguish the
identified with the antibodies GBC1–3 (Goldstein and direction of lineage relations between the globose and
Schwob, 1996). The horizontal basal cell expresses ker- horizontal basal cells, although the authors favor the
atin and a surface glycoprotein, which binds to the lectin hypothesis that horizontal basal cells arose from glo-
BS-I (Holbrook et al., 1995). bose basal cells (Huard et al., 1998).
The olfactory epithelium contains a stem cell that gives There can be no doubt now that the immediate neuronal
rise to sensory neurons, but is it multipotent? Can it give rise precursor is a globose basal cell and that the globose basal
to other cell types? There is evidence that it can. After cells can proliferate (Calof and Chikaraishi, 1989;
destruction of the olfactory epithelium with methyl bro- Goldstein and Schwob, 1996; Graziadei and Monti
mide, the epithelium regenerates all cell types including Graziadei, 1979; Newman et al., 1999). These properties
neurons, supporting cells, basal cells, and duct cells of define the globose basal cell at least as a proliferating pre-
Bowman’s glands (Schwob et al., 1995). Retroviral lineage cursor. They may be a committed precursor population
analysis of the regenerating epithelium indicated that there because they reduced in numbers with time as the neuronal
may be two multipotent progenitors, one of which gave rise population was reconstituted after methyl bromide damage
to nonneuronal cells only (supporting cells, Bowman’s (Huard et al., 1998). Still in doubt is whether the globose
gland cells and duct cells) and another that gave rise to basal basal cell population contains an undifferentiated, uncom-
cells, neurons, and supporting cells (Huard et al., 1998). mitted stem cell (Huard et al., 1998) or whether such a
It is generally accepted that new neurons arise in the stem cell resides among the horizontal basal cells
olfactory epithelium from proliferation and differen- (Mackay-Sim and Kittel, 1991a).
tiation of basal cells, but the roles of globose versus
horizontal basal cells are in dispute. Early in vivo quan-
titative analysis suggested that the stem cell resides on
the basement membrane, in the location of the horizon- VII. NEUROGENESIS IN THE ADULT
tal basal cell (Mackay-Sim and Kittel, 1991a). In agree- OLFACTORY EPITHELIUM
ment with these conclusions, in vitro experiments also
suggested that neurons can arise from horizontal basal This review has drawn together observations from in vivo
cells (Mahanthappa and Schwarting, 1993; Satoh and and in vitro studies of neurogenesis in the adult. The case is
Neurogenesis in the Adult Olfactory Neuroepithelium 107

made here that olfactory neurogenesis is a continuing, 1996), thus raising the possibility that olfactory tissues
regulated process, which is similar in many respects to could be used for transplantation repair of the brain.
neurogenesis in the embryonic nervous system. Obviously
neurogenesis in the adult olfactory epithelium cannot
mimic all aspects of neurogenesis in other parts of the ner- ACKNOWLEDGMENTS
vous system. There are peculiarities of this tissue that
define it; for example, olfactory sensory neurons are Alan Mackay-Sim is supported by the Garnett Passe and
exposed to the atmosphere, hence they might be more prone Rodney Williams Memorial Foundation.
to die by external influences than neurons in other parts of
the nervous system. Perhaps the evolutionary selection
pressure to preserve olfactory neurogenesis into adulthood
REFERENCES
can be due to a number of interrelated factors. On the one
hand, olfaction evolved as essential for survival of the indi-
Aiba T, Mori J, Nakai Y. Nerve growth factor (NGF) and its
vidual through food finding and selection and essential to
receptor in rat olfactory epithelium. Acta Otolaryngol
survival of the species through social, sexual, and predator (Stockholm) 1993; (suppl 506):37–40.
selection. On the other hand, the transduction mechanism Andres K. Der Feinbau der Regio olfactoria von Makrosmatikern.
required that sensory neurons be exposed to the odorous air, Z Zellforsch Mikrosk Anat 1966; 69: 140–154.
which could also contain toxic substances, viruses, and bac- Astic L, Pellier-Monnin V, Godinot F. Spatio-temporal patterns of
teria, whose actions could destroy the sensory neurons or ensheathing cell differentiation in the rat olfactory system
pass via the sensory neurons to the brain. Under these con- during development. Neuroscience 1998; 84:295–307.
ditions there might be selection pressure to evolve labile Ayer-LeLievre C, Ebendal T, Olson L, Seiger A. Localization of
sensory neurons, which can be replaced through a contin- nerve growth factor-like immunoreactivity in rat nervous tis-
ued neurogenesis occurring through the reproductive life of sue. Med. Biol. 1983; 61:296–304.
the organism. It is noteworthy that although neurogenesis Ayer-LeLievre C, Stahlbom P-A, Sara V. Expression of IGF-I and
-II mRNA in the brain and craniofacial region of the rat fetus.
continues throughout adult life, the rate of basal cell prolif-
Development 1991; 111:105–115.
eration declines in old age in rodents (Weiler and Farbman, Bakardjiev A. Biosynthesis of carnosine in primary cultures of rat
1997) and there is a significant decrease in olfactory func- olfactory bulb. Neurosci. Lett. 1997; 227:115–118.
tion in aging humans (Doty et al., 1984). Baker H, Morel K, Stone DM, Maruniak JA. Adult naris closure
The regulation of olfactory neurogenesis is still not well profoundly reduces tyrosine hydroxylase expression in mouse
understood. There are a large number of candidate factors olfactory bulb. Brain Res. 1993; 614:109–116.
whose function is unknown but whose presence is inferred Baker HT, Kawano T, Margolis FL, Joh TH. Transneuronal regu-
by the expression of appropriate receptors in the lation of tyrosine hydroxylase expression in olfactory of
epithelium (Mackay-Sim and Chuah, 2000). The olfactory mouse and rat. J. Neurosci. 1983; 3:9–78.
epithelium provides a useful model tissue to investigate Beard M, Mackay-Sim A. Loss of sense of smell in adult
neurogenesis and has the advantages of relatively large hypothyroid mice. Dev. Brain Res. 1987; 36:181–189.
Bondy C, Lee W-H. Correlation between insulin-like growth fac-
volume and accessibility compared to the embryo.
tor (IGF)-binding protein 5 and IGF-I gene expression during
Apart from its intrinsic interest in scientific terms, brain development. J. Neurosci. 1993; 13: 5092–5104.
understanding olfactory neurogenesis is becoming impor- Breipohl W, Laugwitz HJ, Bornfeld N. Topological relations
tant now for its potential clinical applications in diseases between the dendrites of olfactory sensory cells and susten-
of brain development and in nervous system repair. tacular cells in different vertebrates. An ultrastructural study.
Differences in olfactory neurogenesis were observed in J. Anat. 1974; 117:89–94.
persons with schizophrenia (Féron et al., 1999b). These Breipohl W, Mackay-Sim A, Grandt D, Rehn B, Darrelmann C.
differences may lead to a better understanding of the neu- Neurogenesis in the vertebrate main olfactory epithelium. In:
rodevelopmental origins of the disease. Additionally, it is Breipohl W, ed. Ontogeny of Olfaction. Berlin: Springer-
evident that the olfactory nerve–ensheathing glia are quite Verlag, 1986:21–33.
different from any others in the nervous system (Doucette, Buck L, Axel R. A novel multigene family may encode odorant
receptors: a molecular basis for odor recognition. Cell 1991;
1990; Ramon-Cueto and Nieto-Sampedro, 1992), and
65:175–187.
they show great promise in promoting repair and recovery Buckland ME, Cunningham AM. Alterations in expression of the
after spinal damage (Lu et al., 2001; Ramon-Cueto et al., neurotrophic factors, glial cell line-derived neurotrophic fac-
2000). Additionally, when cells from the olfactory mucosa tor, ciliary neurotrophic factor and brain-derived neurotrophic
were transplanted into the embryonic brain, some devel- factor, in the target-deprived olfactory neuroepithelium.
oped into nonolfactory neurons (Magrassi and Graziadei, Neuroscience 1999; 90:333–347.
108 Mackay-Sim

Caggiano M, Kauer JS, Hunter DD. Globose basal cells are neur- Costanzo RM. Neural regeneration and functional reconnection
onal progenitors in the olfactory epithelium: a lineage analy- following olfactory nerve transection in hamster. Brain Res.
sis using a replication-incompetent retrovirus. J. Neurosci. 1985; 361:258–266.
1994; 13:339–352. Costanzo RM, Graziadei PPC. A quantitative analysis of changes
Cain WS. Contribution of the trigeminal nerve to perceived odor in the olfactory epithelium following bulbectomy in hamster.
magnitude. Ann. NY Acad. Sci. 1974; 237:28–34. J. Comp. Neurol. 1983; 215:370–381.
Calof AL, Chikaraishi DM. Analysis of neurogenesis in a mam- Cuschieri A, Bannister LH. The development of the olfactory
malian neuroepithelium: proliferation and differentiation of an mucosa in the mouse: electron microscopy. J. Anat. 1975;
olfactory neuron precursor in vitro. Neuron 1989; 3:115–127. 119:471–498.
Camara CG, Harding JW. Thymidine incorporation in the olfac- Davis BJ, Macrides F. Tyrosine hydroxylase immunoreactive
tory epithelium of mice: early exponential response induced neurons and fibers in the olfactory system of the hamster.
by olfactory neurectomy. Brain Res. 1984; 308:63–68. J. Comp. Neurol. 1983; 214:427–440.
Carr VM, Farbman AI. Ablation of the olfactory bulb up-regu- Deckner M-L, Risling, M, Frisen, J. Apoptotic death of olfactory
lates the rate of neurogenesis and induces precocious cell sensory neurons in the adult rat. Exp. Neurol. 1997;
death in olfactory epithelium. Exp. Neurol. 1992; 115: 143:132–140.
55–59. DeHamer MK, Guevara JL, Hannon K, Olwin BB, Calof AL.
Carr VM, Farbman AI. The dynamics of cell death in the olfac- Genesis of olfactory receptor neurons in vitro: regulation of
tory epithelium. Exp. Neurol. 1993; 124:308–314. progenitor cell divisions by fibroblast growth factors. Neuron
Chen C, Mrksich M, Huang S, Whitesides G, Ingber D. 1994; 13:1083–1097.
Geometric control of life and death. Science 1997; Devon R, Doucette R. Olfactory ensheathing cells myelinate dor-
276:1425–1428. sal root ganglion neurites. Brain Res. 1992; 589:175–179.
Chuah MI, Au C. Olfactory Schwann cells are derived from pre- Doty RL. Intranasal trigeminal detection of chemical vapors by
cursor cells in the olfactory epithelium. J. Neurosci. Res. humans. Physiol. Behav. 1975; 14:855–859.
1991; 29:172–180. Doty RL, Shaman P, Applebaum SL, Giberson R, Siksorski L,
Chuah MI, Au C. Neural cell adhesion molecules are present in Rosenberg L. Smell identification ability: changes with age.
the fetal human primary olfactory pathway. Dev. Neurosci. Science 1984; 26:1441–1443.
1992; 14:357–361. Doucette R. The glial cells in the nerve fibre layer of the rat olfac-
Chuah MI, Au C. Olfactory cell cultures on ensheathing cell tory bulb. Anat. Rec. 1984; 210:385–391.
monolayers. Chem. Senses 1994; 19:25–34. Doucette R. Development of the nerve fiber layer in the olfac-
Chuah MI, Farbman AI. Olfactory bulb increases marker protein tory bulb of mouse embryos. J. Comp. Neurol. 1989;
in olfactory receptor cells. J. Neurosci. 1983; 3:2197–2205. 285:514–527.
Chuah M, Teague R. Basic fibroblast growth factor in the Doucette R. Glial influences on axonal growth in the primary
primary olfactory pathway: mitogenic effect on ensheathing olfactory system. Glia. 1990; 3:433–449.
cells. Neuroscience 1999; 88:1043–1050. Doucette R. PNS-CNS transition zone of the first cranial nerve.
Chuah MI, Farbman AI, Menco BPM. Influence of olfactory bulb J. Comp. Neurol. 1991; 312:451–466.
on dendritic density of rat olfactory receptor neurons in vitro. Ensoli F, Fiorelli V, Vannelli B, Barni T, De Cristofaro M, Ensoli
Brain Res. 1985; 338:259–266. B, et al. Basic fibroblast growth factor supports human olfac-
Chuah M, Cossins J-M, Woodhall E, Tennent R, Nash G, West tory neurogenesis by autocrine/paracrine mechanisms.
A. Glial growth factor 2 induces proliferation and structural Neuroscience 1998; 86:881–893.
changes in ensheathing cells. Brain Res. 2000; 857:265–274. Eriksson P, Perfilieva E, Bjork-Eriksson T, Alborn A, Nordborg
Coronas V, Feron F, Hen R, Sicard G, Jourdan F, Moyse E. In C, Peterson D, et al. Neurogenesis in the adult human hip-
vitro induction of apoptosis or differentiation by dopamine in pocampus. Nat. Med. 1998; 11:1313–1317.
an immortalized olfactory neuronal cell line. J. Neurochem. Farbman A, Buchholz J. Transforming growth factor- and other
1997a; 69:1870–1881. growth factors stimulate cell division in olfactory epithelium
Coronas V, Srivastava L, Liang J-J, Jourdan F, Moyse E. in vitro. J. Neurobiol. 1996; 30:267–280.
Identification and localization of dopamine receptor subtypes Farbman A, Buchholz J, Bruch R. Growth factor regulation of
in rat olfactory mucosa and bulb: a combined in situ olfactory cell proliferation. In: Kurihara K, Suzuki N, Ogawa
hybridization and ligand binding radioautographic approach. H, eds. Olfaction and Taste XI. Tokyo: Springer-Verlag,
J. Neurochem. 1997b; 12:243–257. 1994:45–48.
Coronas V, Krantic S, Jourdan F, Moyse E. Dopamine receptor Farbman A, Buchholz J, Suzuki Y, Coines A, Speert D. A molecu-
coupling to adenylyl cyclase in rat olfactory pathway: a lar basis of cell death in olfactory epithelium. J. Comp.
combined pharmacological-radioautographic approach. Neurol. 1999; 414:306–314.
Neuroscience 1999; 90:69–78. Farbman AI, Squinto LM. Early development of olfactory recep-
Costanzo RM. Comparison of neurogenesis and cell replacement tor cell axons. Dev. Brain Res. 1985; 19:205–213.
in the hamster olfactory system with and without a target Farbman AI, Brunjes PC, Rentfro L, Michas J, Ritz S. The
(olfactory bulb). Brain Res. 1984; 307:295–301. effect of unilateral naris occlusion on cell dynamics in the
Neurogenesis in the Adult Olfactory Neuroepithelium 109

developing rat olfactory epithelium. J. Neurosci. 1988; Graziadei, P. P. C., Karlan, M. S., Monti Graziadei, G. A., and
8:3290–3295. Bernstein, J. J. Neurogenesis of sensory neurons in the pri-
Federico G, Maremmani C, Cinquanta L, Baroncelli G, Fattori B, mate olfactory system after section of the filia olfactoria.
Saggese G. Mucus of the human olfactory epithelium contains Brain Res. 1980; 186:289–300.
the insulin-like growth factor-I system which is altered in Guthrie, K., Woods, A., Nguyen, T., and Gall, C. Astroglial
some neurodegenerative diseases. Brain Res. 1999; neurotrophic factor mRNA expression is increased in fields of
835:306–314. axonal sprouting in deafferented hippocampus. J. Comp.
Féron F, Mackay-Sim A, Andrieu J, Matthaei K, Holley A, Sicard Neurol. 1997; 386:137–148.
G. Stress induces neurogenesis in non-neuronal cell cultures Halasz, N., Ljungdahl, A., Hokfelt, T., Johansson, O., Goldstein,
of adult olfactory epithelium. Neuroscience 1999a; M., Park, D., et al. Transmitter histochemistry of the rat olfac-
88:571–583. tory bulb. I. Immunohistochemical localization of monoamine
Féron F, Perry C, Hirning M, McGrath J, Mackay-Sim A. Altered synthesizing enzymes. Support for intrabulbar, periglomerular
adhesion, proliferation and death in neural cultures from adults neurons. Brain Res. 1977; 126:253–271.
with schizophrenia. Schizophrenia Res. 1999b; 40:211–218. Harding, J. W., Getchell, T. V., and Margolis, F. L. Denervation in
Féron F, Vincent A, Mackay-Sim A. Dopamine promotes differ- the primary pathway of mice. V. long term effect of intranasal
entiation of olfactory neuron in vitro. Brain Res. 1999c; ZnSO4 irrigation on behavior, biochemistry and morphology.
845:252–259. Brain Res. 1978; 140:271–285.
Franceschini, I., and Barnett, S. Low-affinity NGF-receptor and Hempstead, J. L., and Morgan, J. I. Monoclonal antibodies to the
E-N-CAM expression define two types of olfactory nerve rat olfactory sustentacular cell. Brain Res. 1983; 288:
ensheathing cells that share a common lineage. Dev. Biol. 289–295.
1996; 173:327–343. Hinds, J. W., and McNelly, N. A. Aging in the rat olfactory
Fung, K.-M., Peringa, J., Venkatachalam, S., Lee, V.-Y., and system: correlation of changes in the olfactory epithe-
Trojanowski, J. Coordinate reduction in cell proliferation and lium and olfactory bulb. J. Comp. Neurol. 1981; 203:
cell death in mouse olfactory epithelium from birth to matur- 441–453.
ity. Brain Res. 1997; 761:347–351. Hinds, J. W., Hinds, P. L., and McNelly, N. A. An autoradio-
Gall, C., Berschauer, R., and Isackson, P. Seizures increase basic graphic study of the mouse olfactory epithelium: evidence for
fibroblast growth factor mRNA in adult rat forebrain neurons long-lived receptors. Anat. Rec. 1984; 210: 375–383.
and glia. Mol. Brain Res. 1994; 21:190–205. Holbrook, E. H., Szumowski, K. E., and Schwob, J. E. An
Goldstein, B. J., and Schwob, J. E. Analysis of the globose basal immunochemical, ultrastructural, and developmental charac-
cell compartment in rat olfactory epithelium using GBC-1, a terization of the horizontal basal cells of rat olfactory epithe-
new monoclonal antibody against globose basal cells. lium. J. Comp. Neurol. 1995; 363:129–146.
J. Neurosci. 1996; 16:4005–4016. Holcomb, J., Mumm, J., and Calof, A. Apoptosis in the neuronal
Goldstein, B. J., Wolozin, B. L., and Schwob, J. E. FGF2 sup- lineage of the mouse olfactory epithelium: regulation in vivo
presses neurogenesis of a cell line derived from rat olfactory and in vitro. Dev. Biol. 1995; 172:307–323.
epithelium. J. Neurobiol. 1997; 33:411–428. Hsu, P., Yu, F., Féron, F., Pickles, J., and Sneesby, K., Mackay-
Gong, Q., Bailey, M. S., Pixley, S., Ennis, M., Liu, W., and Sim, A. Basic fibroblast factor and fibroblast growth factor
Shipley, M. T. Localization and regulation of low affinity receptors in adult olfactory epithelium. Brain Res. 2001;
nerve growth factor receptor expression in the rat olfactory 896:188–197.
system during development and regeneration. J. Comp. Huard, J., Youngentob, S., Goldstein, B., Luskin, M., and
Neurol. 1994; 344:336–348. Schwob, J. Adult olfactory epithelium contains multipotent
Goodman, M., Silver, J., and Jacobberger, J. Establishment and progenitors that give rise to neurons and non-neural cells.
neurite outgrowth properties of neonatal and adult rat olfac- J. Comp. Neurol. 1998; 400:469–486.
tory bulb glial cell lines. Brain Res. 1993; 619:199–213. Hurtt, M. E., Thomas, D. A., Working, P. K., Monticello, T. M.,
Graziadei, P., and Metcalf, J. Autoradiographic and ultrastruc- and Morgan, K. T. Degeneration and regeneration of the olfac-
tural observations on the frog’s olfactory mucosa. Z. tory epithelium following inhalation exposure to methyl bro-
Zellforsch. Microsk. Anat. 1971; 116:305–318. mide: pathology, cell kinetics, and olfactory function. Toxicol.
Graziadei, P. P. C. Cell dynamics in the olfactory mucosa. Tissue Appl. Pharmacol. 1988; 94:311–328.
Cell 1973; 5:113–131. Imaizumi, T., Lankford, K., Waxman, S., Greer, C., and Kocsis,
Graziadei, P. P. C., and Monti Graziadei, G. A. Continuous nerve J. Transplanted olfactory ensheathing cells remyelinate and
cell renewal in the olfactory system. In: Jacobson M, ed. enhance axonal conduction in the demyelinated dorsal
Handbook of Sensory Physiology. IX. Development of columns of the rat spinal cord. J. Neurosci. 1998;
Sensory Systems. New York: Springer-Verlag, 1978:55–83. 18:6176–6185.
Graziadei, P. P. C., and Monti Graziadei, G. A. Neurogenesis and Jourdan, F., Moyse, E., De Bilbao, F., and Dubois-Dauphin, M.
neuron regeneration in the olfactory system of mammals. I. mor- Olfactory neurons are protected from apoptosis in adult trans-
phological aspects of differentiation and structural organization genic mice over-expressing the bcl-2 gene. Neuroreport 1998;
of the olfactory sensory neurons. J. Neurocytol. 1979; 8:1–18. 9: 921–926.
110 Mackay-Sim

Kaakkola, S., Palo, J., Malmberg, H., Sulkava, R., and Virtanen, Mackay-Sim, A., and Beard, M. Hypothyroidism disrupts neural
I. Neurofilament profile in olfactory mucosa of patients with development in the olfactory epithelium of adult mice. Dev.
a clinical diagnosis of Alzheimer’s disease. Virchows Arch. Brain Res. 1987; 36:190–198.
1994; 424:315–319. Mackay-Sim, A., and Chuah, M. Neurotrophic growth factors in the
Kafitz, K., and Greer, C. Olfactory ensheathing cells promote primary olfactory pathway. Prog. Neurobiol. 2000; 62:527–559.
neurite extension from embryonic olfactory receptor cells in Mackay-Sim, A., and Kittel, P. Cell dynamics in the adult mouse
vitro. Glia 1999; 25:99–110. olfactory epithelium: a quantitative autoradiographic study.
Key, B., Treloar, H., Wangerek, L., Ford, M., and Nurcombe, V. J. Neurosci. 1991a; 11:979–984.
Expression and localization of FGF-1 in the developing rat Mackay-Sim, A., and Kittel, P. On the life span of olfactory
olfactory system. J. Comp. Neurol. 1996; 366:197–206. receptor neurones. Eur. J. Neurosci. 1991b; 3:209–215.
Konzelmann, S., Saucier, D., Strotmann, J., Breer, H., and Astic, Mackay-Sim, A., and Patel, U. Regional differences in cell gene-
L. Decline and recovery of olfactory receptor expression fol- sis in the olfactory epithelium of the salamander, Ambystoma
lowing unilateral bulbectomy. Cell Tissue Res. 1998; tigrinum. Exp. Brain Res. 1984; 57:99–106.
294:421–430. Mackay-Sim, A., Breipohl, W., and Kremer, M. Cell dynamics in
Koster, N., Norman, A., Richard, N., Nickell, W., Puche, A., the olfactory epithelium of the salamander: a morphometric
Pixley, S., et al. Olfactory receptor neurons express dopamine analysis. Exp. Brain Res. 1988; 71:189–198.
D2 receptors. J. Comp. Neurol. 1999; 411:666–673. Magrassi, L., and Graziadei, P. Cell death in the olfactory epithe-
Lee, K. H., Bowenpope, D. F., and Reed, R. R. Isolation and lium. Anat. Embryol. (Berlin) 1995; 192:77–87.
characterization of the alpha-platelet-derived growth factor Magrassi, L., and Graziadei, P. Lineage specification of olfactory
receptor from rat olfactory epithelium. Mol. Cell Biol. 1990; neural precursor cells depends on continuous cell interactions.
10:2237–2246. Dev. Brain Res. 1996; 96:11–27.
Leibovici, M., Lapointe, F., Aletta, P., and Ayer-LeLievre, C. Mahalik, T. Apparent apoptotic cell death in the olfactory epithe-
Avian olfactory receptors: differentiation of olfactory neurons lium of adult rodents: death occurs at different developmental
under normal and experimental conditions. Dev. Biol. 1996; stages. J. Comp. Neurol. 1996; 372:457–464.
175:118–131. Mahanthappa, N. K., and Schwarting, G. A. Peptide growth
Levi-Montalcini, R. The nerve growth factor 35 years later. factor control of olfactory neurogenesis and neuron survival
Science 1987; 237:1154–1162. in vitro: roles of EGF and TGF-s. Neuron 1993; 10:
Li, Y., Field, P., and Raisman, G. Repair of adult rat corticospinal 293–305.
tract by transplants of olfactory ensheathing cells. Science Mania-Farnell, B. L., Farbman, A. I., and Bruch, R. C.
1997; 277:2000–2002. Bromocriptine, a dopamine D2 receptor agonist, inhibits
Li, Y., Field, P., and Raisman, G. Regeneration of adult cortico- adenylyl cyclase activity in rat olfactory epithelium.
spinal axons induced by transplanted olfactory ensheathing Neuroscience 1993; 57:173–180.
cells. J. Neurosci. 1998; 18:10514–10524. Margalit, T., and Lancet, D. Expression of olfactory receptor and
Liesi, P. Laminin-immunoreactive glia distinguish regenerative transduction genes during rat development. Dev. Brain Res.
adult CNS systems from non-regenerative ones. EMBO J. 1993; 73:7–16.
1985; 4:2505–2511. Margolis, F. L. Olfactory marker protein: from PAGE band to
Liu, N., Shields, C. B., and Roisen, F. J. Primary culture of adult cDNA clone. Trends Neurosci. 1985; 8:542–546.
mouse olfactory receptor neurons. Exp. Neurol. 1998; Margolis, F. L., Roberts, N., Ferriero, D., and Feldman, J.
151:173–183. Denervation in the primary olfactory pathway of mice: bio-
Loo, A. T., Youngentob, S. L., Kent, P. F., and Schwob, J. E. The chemical and morphological effects. Brain Res. 1974;
aging olfactory epithelium: neurogenesis, response to dam- 81:469–483.
age, and odorant-induced activity. Int. J. Dev. Neurosci. 1996; Marin-Padilla, M., and Amieva, M. R. Early neurogenesis of the
14:881–900. mouse olfactory nerve: Golgi and electron microscopic stud-
Lu, J., Féron, F., Ho, S., Mackay-Sim, A., and Waite, P. ies. J. Comp. Neurology 1989; 288:339–352.
Transplantation of nasal olfactory tissue promotes partial Maruniak, J. A., Lin, P. J., and Henegar, J. R. Effects of unilateral
recovery in paraplegic rates. Brain Res. 2001; 889: naris closure on the olfactory epithelia of adult mice. Brain
344–357. Res. 1989; 490:212–218.
Lucero, M., and Squires, A. Catecholamine concentrations in rat Maruniak, J. A., Henegar, J. R., and Sweeney, T. P. Effects of
nasal mucus are modulated by trigeminal stimulation of the long-term unilateral naris closure on the olfactory epithelia of
nasal cavity. Brain Research. 1998; 807:234–236. adult mice. Brain Res. 1990; 526:65–72.
Luskin, M. Restricted proliferation and migration of postnatally Matsuyama, A., Iwata, H., Okumura, N., Yoshida, S., Imaizumi,
generated neurons derived from the forebrain subventricular K., Lee, Y., et al. Localization of basic fibroblast growth factor-
zone. Neuron 1993; 11:173–189. like immunoreactivity in the rat brain. Brain Res. 1992;
MacDonald, K. P. A., Murrell, W. G., Bartlett, P. F., Bushell, G. 587:49–65.
R., and Mackay-Sim, A. FGF2 promotes neuronal differen- Matulionis, D. H. Ultrastructural study of mouse epithelium fol-
tiation in explant cultures of adult and embryonic mouse lowing destruction by ZnSO4 and its subsequent regeneration.
olfactory epithelium. J. Neurosci. Res. 1996; 44:27–39. Am. J. Anat. 1975; 142:67–90.
Neurogenesis in the Adult Olfactory Neuroepithelium 111

Menco, B. P. M. Qualitative and quantitative freeze-fracture stud- Nakashima, T., Kimmelmann, C., and Snow, J. Structure of
ies on olfactory and nasal respiratory epithelial surfaces of human fetal and adult olfactory neuroepithelium. Acta
frog, ox, rat and dog. III. Tight-junctions. Cell Tissue Res. Otolaryngol. 1984; 110:641–646.
1980; 211:361–373. Nakashima, T., Kimmelmann, C., and Snow, J. Immuno
Michel, D., Moyse, E., Brun, G., and Jourdan, F. Induction of histopathology of human olfactory epithelium, nerve and bulb.
apoptosis in rat olfactory neuroepithelium by synaptic target Laryngoscope 1985; 995:391–396.
ablation. Neuroreport 1994; 5:1329–1332. Newman, M., Féron, F., and Mackay-Sim, A. Growth factor reg-
Miragall, F., and Dermietzel, R. Immunocytochemical localiza- ulation of neurogenesis in adult olfactory epithelium.
tion of cell adhesion molecules in the developing and mature Neuroscience 2000; 99:343–350.
olfactory system. Microsc. Res. Tech. 1992; 23:157–172. Nosrat, C. A., Tomac, A., Lindqvist, E., Lindskog, S., Humpel,
Miragall, F., Kadmon, G., Husmann, M., and Schachner, M. C., Stromberg, I., Ebendal, T., Hoffer, B., and Olson, L.
Expression of cell adhesion molecules in the olfactory system Cellular expression of GDNF mRNA suggests multiple func-
of the adult mouse: presence of the embryonic form of N- tions inside and outside the nervous system. Cell Tiss Res.
CAM. Dev. Biol. 1988; 129:516–531. 1996; 286:191–207.
Miragall, F., Kadmon, G., and Schachner, M. Expression of L1 Nosrat, C. A., Tomac, A., Hoffer, B. J., and Olson, L. Cellular and
and N-CAM cell adhesion molecules during development developmental patterns of expression of Ret and glial cell
of the mouse olfactory system. Dev. Biol. 1989; 135: line-derived neurotrophic factor receptor alpha mRNAs. Exp.
272–286. Brain Res. 1997; 115:410–422.
Miragall, F., Hwang, T.-K., Traub, O., Hertzberg, E. L., and Orr-Urtreger, A., and Lonai, P. Platelet-derived growth factor-A
Dermietzel, R. Expression of connexins in the developing olfac- and its receptors are expressed in separate, but adjacent cell
tory system of the mouse. J. Comp. Neurol. 1992; 325: 359–378. layers of the mouse embryo. Development 1992; 115:
Miwa, T., Horikawa, I., Uramoto, N., Ishimaru, T., Yamamoto, 1045–1058.
K., Furukawa, M., et al. TrkA expression in mouse olfactory Paternostro, M. A., and Meisami, E. Developmental plasticity of
tract following axotomy of olfactory nerves. Acta the rat olfactory receptor sheet as shown by complete
Otolaryngol. 1998; 539:79–82. recovery of surface area and cell number from extensive early
Moon, Y., and Baker, H. Induction of cell division in olfactory hypothyroid growth retardation. Dev. Brain Res. 1993;
basal epithelium following intranasal irrigation with wheat 76:151–161.
germ agglutinin-horseradish peroxidase. J. Comp. Neurol. Perroteau, I., Oberto, M., Ieraci, A., Bovolin, P., and Fasolo, A.
1998; 393:472–481. ErbB-3 and ErbB-4 expression in the mouse olfactory system.
Moulton, D. G. Cell renewal in the olfactory epithelium of the Ann. NY. Acad. Sci. 1998; 855:255–259.
mouse. In: Denton, D. A., Coughlan, J. P., eds. Olfaction and Philpot, B., Men, D., McCarty, R., and Brunjes, P. Activity-
Taste V. New York: Academic Press, 1975:111–114. dependent regulation of dopamine content in the olfactory
Moulton, D. G., Çelebi, G., and Fink, R. P. Olfaction in bulbs of naris-occluded rats. Neuroscience 1998; 85:969–977.
mammals—two aspects: proliferation of cells in the Pixley, S., Dangoria, N., Odoms, K., and Hastings, L. Effects of
olfactory epithelium and sensitivity to odours. In: insulin-like growth factor 1 on olfactory neurogenesis in vivo
Wolstenholme G. E. W., Knight J., eds. Ciba Foundation on and in vitro. Ann. NY Acad. Sci. 1998; 855:244–247.
Taste and Smell in Vertebrates. London: J. & A. Churchill, Pixley, S. K. The olfactory nerve contains two populations of
1970:227–250. glia, identified both in vivo and in vitro. Glia 1992;
Mumm, J. S., Shou, J., and Calof, A. L. Colony-forming progeni- 5:269–284.
tors from mouse olfactory epithelium: evidence for feedback Pollock, G., Franceschini, I., Graham, G., Marchionni, M., and
regulation of neuron production. Proc. Natl. Acad. Sci. USA Barnett, S. Neuregulin is a mitogen and survival factor for
1996; 93:11167–11172. olfactory bulb ensheathing cells and an isoform is produced
Murrell, W., Bushell, G. R., Livesey, J., McGrath, J., MacDonald, by astrocytes. Eur. J. Neurosci. 1999; 11:769–780.
K. P. A., Bates, P. R., and Mackay-Sim, A. Neurogenesis in Puche, A., and Shipley, M. Odor-induced, activity-dependent
adult human. Neuroreport 1996; 7:1189–1194. transneuronal gene induction in vitro: mediation by NMDA
Nadi, N. S., Head, R., Grillo, M., Hempstead, J., Grannot-Reisfeld, receptors. J. Neurosci. 1999; 19:1359–1370.
N., and Margolis, F. L. Chemical deafferentation of the olfac- Rama Krishna, N. S., Little, S. S., and Getchell, T. V. Epidermal
tory bulb: plasticity of the levels of tyrosine hydroxylase, growth factor receptor mRNA and protein are expressed in
dopamine, norepinephrine. Brain Res. 1981; 213:365–377. progenitor cells of the olfactory epithelium. J. Comp. Neurol.
Naessen, R. An enquiry on the morphological characteristics and 1996; 373:297–307.
possible changes with age in the olfactory region of man. Acta Ramon-Cueto, A., and Nieto-Sampedro, M. Glial cells from
Otolaryngol. 1971; 71:49–62. adult rat olfactory bulb: immunocytochemical properties of
Nagahara, Y. Experimentelle Studien über die histologischen pure cultures of ensheathing cells. Neuroscience 1992; 47:
Veränderungen des Geruchsorgans nach der Olfactorius 213–220.
durchschneidung. Beiträge zur Kenntnis des feineren Baus Ramon-Cueto, A., and Nieto-Sampedro, M. Regeneration into the
des Geruchsorgans. Jpn. J. Med. Sci. V Pathol. 1940; spinal cord of transected dorsal root axons is promoted by
5:46–63. ensheathing glia transplants. Exp. Neurol. 1994; 127: 232–244.
112 Mackay-Sim

Ramon-Cueto, A., Plant, G., Avila, J., and Bunge, M. Long-dis- Scotti, A., Hoffmann, M.-C., and Nitsch, C. The neurite growth
tance axonal regeneration in the transected adult rat spinal promoting protease nexin 1 in glial cells of the olfactory bulb
cord is promoted by ensheathing glia transplants. J. Neurosci. of the gerbil: an ultrstructural study. Cell Tissue Res. 1994;
1998; 18:3803–3815. 278:409–413.
Ramon-Cueto, A., Cordero, M., Santos-Benito, F., and Avila, J. Shou, J., Rim, P., and Calof, A. BMPs inhibit neurogenesis by a
Functional recovery of paraplegic rats and motor axon regen- mechanism involving degradation of a transcription factor.
eration in their spinal cords by olfactory ensheathing glia. Nat. Neurosci. 1999; 2:339–345.
Neuron 2000; 25:425–435. Shou, J., Murray, R., Rim, P., and Calof, A. Opposing effects of
Rehn, B., Breipohl, W., Schmidt, C., Schmidt, U., and bone morphogenic proteins on neuron production and survival
Effenberger, F. Chemical blockade of olfactory perception by in the olfactory receptor neuron lineage. Development 2000;
N-methyl-formimino-methylester in albino mice. II. Light 127:5403–5413.
microscopical investigations. Chem. Senses 1981; 6:317–328. Silver, W. L., and Moulton, D. G. Chemosensitivity of rat nasal
Reinhard, E., Meier, R., Halfter, W., Rovelli, G., and Monard, D. trigeminal receptors. Physiol. Behav. 1982; 28:927–931.
Detection of glia-derived nexin in the olfactory system of the Smart, I. H. M., Location and orientation of mitotic figures in the
rat. Neuron 1988; 1:387–394. developing mouse olfactory epithelium. J. Anat. 1971;
Ressler, K. J., Sullivan, S. L., and Buck, L. B. A zonal organiza- 109:243–251.
tion of odorant receptor gene expression in the olfactory Smith, C. G. Regeneration of sensory epithelium and nerves in
epithelium. Cell 1993; 73:597–609. adult frogs. Anat. Rec. 1951; 109:661–671.
Roskams, A. J. I., Bethel, M. A., Hurt, K. J., and Ronnett, G. V. Strotmann, J., Konzelmann, S., Breer, H. Laminar segregation of
Sequential expression of Trks A, B, and C in the regen- odorant receptor expression in the olfactory epithelium. Cell
erating olfactory neuroepithelium. J. Neurosci. 1996; Tissue Res. 1996; 284:347–354.
16(4):1294–1307. Strotmann, J., Wanner, I., Helfrich, T., and Breer, H. Receptor
Salehi-Ashtiani, K., and Farbman, A. I. Expression of neu and expression in olfactory neurons during rat development: in
Neu differentiation factor in the olfactory mucosa of rat. Int. situ hybridization studies. Eur. J. Neurosci. 1995; 7:492–500.
J. Dev. Neurosci. 1996; 14:801–811. Sullivan, S. L., Bohm, S., Ressler, K. J., Horowitz, L. F., and
Samanen, D. W., and Forbes, W. B. Replication and differentia- Buck, L. B. Target-independent pattern specification in the
tion of olfactory receptor neurons following axotomy in the olfactory epithelium. Neuron 1995; 15:779–789.
adult hamster: a morphological analysis of postnatal neuro- Suzuki, Y., and Farbman, A. Tumor necrosis factor-α-induced
genesis. J. Comp. Neurol. 1984; 225:201–211. apoptosis in olfactory epithelium in vitro: possible roles of
Satoh, M., and Takeuchi, M. Induction of NCAM expression in caspase 1 (ICE), caspase 2 (ICH-1), and caspase 3 (CPP32).
mouse olfactory keratin-positive basal cells in vitro. Dev. Exp. Neurol. 2000; 165:35–45.
Brain Res. 1995; 87:111–119. Thornhill, R. Cell division in the olfactory epithelium of the lam-
Satoh, M., and Yoshida, T. Promotion of neurogenesis in mouse prey, Lampetra fluviatilis. Z. Zellforsch. Mikrosk. Anat. 1970;
olfactory neuronal progenitor cells by leukemia inhibitory 109:147–157.
factor in vitro. Neurosci. Lett. 1997; 225:165–168. Tisay, K., Key, B. The extracellular matrix modulates olfactory
Scheffler, B., Horn, M., Blumcke, I., Laywell, E., Coome, D., neurite outgrowth on ensheathing cells. J. Neurosci. 1999;
Kukekov, V., et al. Marrow-mindedness: a perspective on 19:9890–9899.
neuropoiesis. Trends Neurosci. 1999; 22:348–357. Treloar, H., Nurcombe, V., and Key, B. Expression of extracellu-
Schmidt, U., Schmidt, C., Breipohl, W., and Effenberger, F. The lar matrix molecules in the embryonic rat olfactory pathway.
effects of N-methyl-formimino-methylester on the neural J. Neurobiol. 1996; 31:41–55.
olfactory threshold in albino mice. Arch Otorhinolaryngol Turner, C. P., and Perez-Polo, J. R. Regulation of the low affinity
1984; 239:25–29. receptor for nerve growth factor, p75NGFR, in the olfactory sys-
Schultz, E. Regeneration of olfactory cells. Proc. Soc. Exp. Biol. tem of neonatal and adult rat. Int. J. Dev. Neurosci. 1992;
Med. 1941; 46:41–43. 10:343–359.
Schwob, J. E., Szumowski, K. E. M., and Stasky, A. A. Olfactory Valverde, F., Lopez-Mascaraque, L. Neuroglial arrangements in
sensory neurons are trophically dependent on the olfactory bulb the olfactory glomeruli of the hedgehog. J. Comp. Neurol.
for their prolonged survival. J. Neurosci. 1992; 12: 3896–3919. 1991; 307:658–674.
Schwob, J. E., Huard, J. M. T., Luskin, M. B., and Youngentob, Valverde, F., Heredia, M., Santacana, M. Characterization of
S. L. Retroviral lineage studies of the rat olfactory epithelium. neuronal cell varieties migrating from the olfactory
Chem. Senses 1994a; 19:671–682. epithelium during prenatal development in the rat.
Schwob, J. E., Youngentob, S. L., and Meiri, K. F. On the forma- Immunocytochemical study using antibodies against olfactory
tion of neuromata in the primary olfactory projection. marker protein (OMP) and luteinizing hormone-releasing hor-
J. Comp. Neurol. 1994b; 840:361–380. mone (LH-RH). Dev. Brain Res. 1993; 71:209–220.
Schwob, J. E., Youngentob, L., and Mezza, R. C. Reconstitution Vargas, G., and Lucero, M. Dopamine modulates inwardly rectify-
of the rat olfactory epithelium after methyl bromide-induced ing hyperpolarization-activated current (Ih) in cultured rat olfac-
lesion. J. Comp. Neurol. 1995; 359:15–37. tory receptor neurons. J. Neurophysiol. 1999; 81: 149–158.
Neurogenesis in the Adult Olfactory Neuroepithelium 113

Vassar, R., Chao, S. K., Sitcheran, R., Nuez, J. M., Vosshall, L. Westerman, R., and von Baumgarten, R. Regeneration of olfacto-
B., and Axel, R. Topographic organization of sensory projec- ry paths in carp (Cyprinus carpio L.). Experientia 1964;
tions to the olfactory bulb. Cell 1994; 79:981–991. 20:519–520.
Vaux, D., and Strasser, A. The molecular biology of apoptosis. Whitesides, J., and LaMantia, A. Differential adhesion and the
Proc. Natl. Acad. Sci. USA 1996; 93:2239–2244. initial assembly of the mammalian olfactory nerve. J. Comp.
Verscheuren, K., Dewulf, N., Goumans, M-J, Lonnoy, O., Feijen, Neurol. 1996; 373:240–254.
A., Grimsby, S., et al. Expression of type I and type 1B recep- Williams, R., and Herrup, K. The control of neuron number. Ann.
tors for activin in midgestation mouse embryos suggests dis- Rev. Neurosci. 1988; 11:423–453.
tinct functions in organogenesis. Mech. Dev. 1995; Williams, R., and Rush, R. A. Electron microscopic immuno-
52:109–123. cytochemical localization of nerve growth factor in devel-
Vickland, H., Westrum, L. E., Kott, J. N., Patterson, S. L., and oping mouse olfactory neurons. Brain Res. 1988;
Bothwell, M. A., Nerve growth factor receptor expression in 463:21–27.
the young and adult rat olfactory system. Brain Res. 1991; Wolozin, B., Sunderland, T., Zheng, B.-B., Resau, J., Dufy, B.,
565:269–279. Barker, J., et al. Continuous culture of neuronal cells from
Walters, E., Buchheit, K., and Maruniak, J. A. Receptor neuron adult human olfactory epithelium. J. Mol. Neurosci. 1992;
losses result in decreased cytochrome P-450 immunoreactiv- 3:137–146.
ity in associated non-neuronal cells of mouse olfactory Woodhall, E., West, A., and Chuah, M. Cultured olfactory
mucosa. J. Neurosci. Res. 1992; 33:103–111. ensheathing cells express nerve growth factor, brain-
Weiler, E., and Farbman, A. I. Proliferation in the rat olfactory derived neurotrophic factor, glia cell-line derived neuro-
epithelium: age-dependent changes. J. Neurosci. 1997; trophic factor and their receptors. Mol. Brain Res. 2001;
17:3610–3622. 88:203–213.
Weiler, E., and Farbman, A. Mitral cell loss following Zhang, D., Mehler, M. F., Song, Q., and Kessler, J. A.
lateral olfactory tract transection increases proliferation densi- Development of bone morphogenic protein receptors in the
ty in rat olfactory epithelium. Eur. J. Neurosci. 1999; nervous system and possible roles in regulating trkC expres-
11:3265–3275. sion. J. Neurosci. 1998; 18:3314–3326.
6

Developmental Anatomy of the Olfactory System

Meng Inn Chuah

University of Tasmania, Hobart, Australia

James E. Schwob
Tufts University School of Medicine, Boston, Massachusetts, U.S.A.

Albert I. Farbman
Northwestern University, Evanston, Illinois, U.S.A.

The descriptive anatomy of olfactory system develop- placode had been removed, showed considerable differ-
ment had been done by nineteenth-century anatomists, ences in the size of the two hemispheres, the operated side
who had shown that in vertebrates the paired nasal being the smaller.” Thus, Burr (1916) was the first to
(olfactory) placodes on the anterolateral region of the demonstrate that development of the olfactory bulb was, in
embryonic head were the precursors of the nasal cavity, some way, dependent on the nerves growing out of the
which contained the olfactory sensory epithelium. The first olfactory placode and reaching the forebrain. In this chap-
successful experimental study on development of the ter we review current knowledge about the major steps in
olfactory system was done in 1916 by Burr, who surgically development of the sensory epithelium and the bulb and
removed one or both placodes from the larval form of describe some of the experimental studies used to examine
Amblystoma, a salamander, and showed that the olfactory the developing olfactory system.
bulb(s) on the operated side(s) did not develop. A previous
effort to do the same experiment in the frog, Hyla esculenta,
had failed because the placode regenerated and I. DEVELOPMENT OF THE NASAL CAVITY
development of the bulb was not compromised (Bell, AND CONCHAE
1907). Burr’s success in his experiments was dependent on
the fact that in Amblystoma the olfactory placode was The structures of the human nose develop from the
sharply outlined and easily distinguishable from the sur- nasal/olfactory placodes, which emerge bilaterally as
rounding epidermis, whereas in the frog, the experimental thickenings of the surface ectoderm on the anterolateral
animal used by Bell, the outline of the placode was dif- sides of the head at stage 11 (24 days gestation and the
ficult to discern. Indeed, Burr showed that partial removal of formation of 21 somites), i.e., just after closure of the ante-
the placode in the frog permitted regeneration of the plac- rior neuropore (Bossy, 1980; Verwoerd and van Oostrum,
ode, as Bell (1907) had reported, but complete removal 1979). Fate mapping has determined that the paired
resulted in failure of the bulb to develop. Burr concluded: placodes derive from cells located in the anterior neural
“The removal of the nasal epithelium deprives the devel- ridge just lateral to the midline at the neural-fold stage of
oping forebrain of a stimulus necessary for its complete embryonic development in chicks and mammals (Couly
development. This is evidenced by the fact that the and Le Douarin, 1985; Verwoerd and van Oostrum, 1979),
forebrain of the six months old larva from which one although in fish there may be a different origin (Whitlock
115
116 Chuah et al.

and Westerfield, 1998). Experimental embryological identify with any confidence the molecular pathways that
analyses of transplantation experiments indicate that the direct that part of the anterior neural ridge to form the
development of the olfactory placode is induced and/or olfactory placode.
maintained via the action of two distinct organizing cen- The nature and role of the prechordal plate is somewhat
ters, identified as the prechordal plate and the posteriorly better understood. At stages subsequent to neural fold for-
adjacent neural plate (Jacobson, 1963). mation the prechordal plate probably does not participate
Several molecular candidates have been proposed for in induction of placodal tissue and eventually olfactory
inducers/organizers that act to specify the anterior end of epithelium, per se. However, it is absolutely critical to the
the neural plate and anterior neural ridge. The process of separation of anterior neurectoderm into two placodes
anterior neural specification begins with the invagination (Jacobson, 1963; Macdonald et al., 1995). It is likely that
of the prospective endoemesoderm through the primitive the molecular factor that is responsible for splitting the
node during gastrulation and the secretion of the neural anterior end of the embryo into symmetrical sides is sonic
inducers Noggin, Chordin, Follistatin, and Cerberus. hedgehog (sHH), whose action is opposed by members of
These inducers impart an anterior character to the neuro- the bone morphogenetic protein (BMP) family (Roelink,
ectoderm if unopposed by posteriorizing factors like 1996). Thus, in the absence of sHH or the overexpression
FGFs and Wnts that derive from the regressing node (Sasai of BMP, the eye and olfactory placode fields, marked by
and De Robertis, 1997). A number of homeobox-containing the expression of the paired box transcription factor Pax6,
transcription factors are expressed in the anterior neural remain continuous across the anterior midline resulting in
ridge at a stage subsequent to gastrulation, including cylcopia, holoprosencephaly, and proboscis formation,
members of the Anf class, the empty spiracles homolog with an area within the proboscis that is recognizable as
Emx2, Dlx5, BF-1, Vaxl, Pax6, Six3, etc. (Zaraisky et al., olfactory epithelium although reduced in size (Chiang et
1992; Simeone et al., 1992; Yang et al., 1998; Tao and Lai, al., 1996; Golden et al., 1999).
1992; Hallonet et al., 1998; Puschel et al., 1992; Oliver et The appearance of the nasal/olfactory placode is fol-
al., 1995, respectively). Likewise, secreted factors, such as lowed by the rapid growth of the mesenchyme around the
FGF8, and members of signaling cascades, such as the placode, forming a horseshoe-shaped ridge, the sides of
Wnt receptor frizzled7, are also expressed in the anterior which are called the medial and lateral nasal prominences
neural fold region of the developing nervous system in (Fig. 1). The shallow depression between the lateral and
advance of placodal emergence (Shanmugalingam et al., medial prominences is known as the nasal pit. The medial
2000; Stark et al., 2000). It is important to note that muta- nasal prominences are separated from one another by the
tion of several genes will disrupt the peripheral olfactory frontonasal prominence, which contributes to the forma-
system along with forebrain or other anterior structures, tion of the nasal septum. Mesenchymal growth appears to
e.g., empty spiracles in Drosphila and Emx2 in mice be tightly regulated and may also impact and/or reflect
(Cecchi et al., 1999; Hirth et al., 1995), likewise, Pax-6 neural development in the periphery. Mutation of zinc fin-
(small-eye mouse mutant) (Hill et al., 1991; Hogan et al., ger transcription factors expressed at later stages of nasal
1986), Otx2 (Acampora et al., 1995), and BF-1 (Hatini et development, such as the cubitus interruptus homolog
al., 1999). However, it is also fair to say that we cannot yet Gli3 (which is truncated in the mouse mutant known as

Figure 1 Frontal and side views of a human embryo head, approximately 33 days old. A horseshoe-shaped rim of mesenchyme sur-
rounds the nasal pit. The lateral part of the rim forms the lateral nasal prominence; the medial part forms the medial nasal prominence.
(Adapted from Moore, 1988.)
Developmental Anatomy of the Olfactory System 117

Figure 2 Sketches of coronal sections of the human embryonic head from weeks 6 through 12, illustrating development of the palate.
(A) The nasal septum becomes established and grows ventrally. (B) The conchae (turbinates) develop as elevations on the lateral walls
of the nasal cavity, while the lateral palatine processes extend medially to meet with each other and the nasal septum. (C) Fusion of the lat-
eral palatine processes with one another and with the nasal septum completes the formation of the palate. As a result, the oral cavity is
separated from the nasal cavities. (Adapted from Moore, 1988.)

extra-toes), and members of the Dlx family (vertebrate The later stages of nasal and craniofacial development
homologs of the fly gene distal-less, a homeobox-containing can be disrupted in human populations, resulting, for
gene downstream of the homeotic selector genes), affect example, in cleft lip and palate. However, much less is
mesenchymal structures (Hui and Joyner, 1993; known about the molecular controls on these morpho-
Schimmang et al., 1992). Mutation of these genes also dis- genetic events. During the later stages, the nasal pits deepen
rupts the formation of axonal connections between the into nasal sacs, which grow dorsocaudally ventral to the
epithelium and telencephalon (Johnson, 1967; Qiu et al., developing forebrain. Initially, these sacs are separated
1995). Thus, in extra-toes mutant mice, the epithelium from the oral cavity by the oronasal membrane, but this
apparently differentiates and grows axons as in normal ani- membrane soon ruptures, thus establishing continuity
mals (Sullivan et al., 1995) and even contacts the telen- between the nasal and oral cavities. The opening between
cephalon long enough to permit migration of LHRH() these two regions is called the posterior choana, and the
cells from the placode to the basal forebrain (S. Wray, per- midline piece of tissue anterior to it is the median palatine
sonal communication), but the bulb does not form because process. As the lateral part of the embryonic head grows
the contact between the olfactory axons and the telen- rapidly, the anterior nasal openings are shifted relatively
cephalon is not maintained. Whether the disruption is pri- closer to the midline. A palate forms, which separates the
marily a consequence of excessive growth of branchial oral and nasal passages (Fig. 2). Lateral palatine processes
arch–derived mesenchymal tissues or an abnormality in extend from the maxillary processes and fuse with each
the telencephalic target area remains to be determined. other in the midline. Anteriorly they fuse with the median
118 Chuah et al.

palatine process, and superiorly with the nasal septum,

resulting in the separation of right and left nasal cavities.
Palate formation enables the neonate to feed and breathe at
the same time (Moore, 1988).
While these events are occurring, the superior, middle,
and inferior turbinates (conchae) develop as elevations on
the lateral wall of each nasal cavity in the human embryo.
They are supported by a cartilaginous framework, which is
gradually replaced by bone. In macrosmatic mammals, i.e.,
those that have a relatively powerful sense of smell as
compared with the microsmatic human, several elaborately
scrolled turbinates form and become more complex as
the snout grows in postnatal life, thus dramatically expand-
ing the surface area of the olfactory epithelium. For
example, in the rat there is an eightfold increase in the
olfactory area during the first postnatal month (Meisami, Figure 3 Histological section through olfactory mucosa of a
newborn rat. The supporting cell nuclei stain slightly darker than
1989). In contrast, olfactory epithelium of the human is
the others and are arranged in a single layer nearest to the surface
restricted to the roof of the nasal cavity, the adjacent super-
of the epithelium (arrows). The basal cell nuclei are immediately
ior part of the nasal septum, and the superior concha. above the basal lamina and appear indistinct in this specimen. The
sensory cell nuclei are located between the supporting cell nuclei
and the basal cell nuclei. Nerve bundles (NB) consisting of olfac-
II. DIFFERENTIATION OF OLFACTORY tory axons can be found in the lamina propria. Bar
25 m.
EPITHELIAL CELLS

In the adult, the olfactory epithelium is a pseudostratified

columnar epithelium overlying a lamina propria. It is A. Cell Division
composed of five basic cell types, distinguishable on
morphological and biochemical grounds, which are arrayed The tissue lining the nasal pit is also a pseudostratified
in stereotyped layers in the epithelium (Fig. 3). From the columnar epithelium. In human and other mammalian
apical surface deep, they are the sustentacular cells (of embryos, a dramatic increase of mitotic figures is observed
which the microvillar cells are a variant), olfactory sensory in the olfactory epithelium immediately before the first
neurons (mature ones positioned superficial to immature emergence of olfactory axons at stage 13 (28 days) (Bossy,
ones), globose basal cells, and horizontal basal cells 1980; Cuschieri and Bannister, 1975a; Smart, 1971). At
(which are tightly apposed to the basal lamina) (Carr et this early stage, mitosis in the olfactory epithelium is sim-
al., 1991; Graziadei and Monti Graziadei, 1978, 1979; ilar to the process of cell division in the developing neural
Holbrook et al., 1995; Schwartz Levey et al., 1991). The tube. Thus, epithelial cells undergo interkinetic nuclear
supporting cell nuclei are arranged in a single layer near migration (Sauer, 1937): G1- and S-phases take place in
the surface, the several layers of olfactory neuron nuclei deeper parts of the epithelium and the cells migrate super-
are deep to it, and the basal cell nuclei are closely related ficially to complete mitosis at the apical epithelial surface
to the basal lamina (Fig. 3). The olfactory neurons make (homologous to the ventricular surface of the neural tube)
up 80–85% of the epithelial cells, the supporting cells (Cuschieri and Bannister, 1975a; Smart, 1971). The mitotic
12–15%, and the basal cells about 5%, although those pattern changes in older fetuses when the basal cell layer
proportions vary depending on location (neurons being is established; there is a progressive shift of mitotic activity
densest in the posterior dorsal part of the olfactory to the base of the epithelium (Smart, 1971). The precise
epithelium) (Farbman et al., 1988; Youngentob et al., relationship between the two spatially distinct precursor
1997). The fifth element is the Bowman’s gland/duct populations has not been clarified, but it may be amenable
complex that extends from the glands in the lamina pro- to better understanding given the recent isolation of mark-
pria to the ducts within the epithelium, which carry the ers that label basal cells in the postnatal epithelium
secretions to the apical epithelial surface. All of these ele- (Goldstein and Schwob, 1996; Goldstein et al., 1997).
ments derive from the nasal/olfactory placode, as does Nonetheless, neurons are being generated even at the times
the respiratory epithelium lining more rostral and ventral when mitoses are concentrated apically. The dividing basal
portions of the nasal cavity. cells in the older embryo are predominantly of the globose
Developmental Anatomy of the Olfactory System 119

variety, as in the adult. Indeed, the horizontal basal cell type (Goldstein et al., 1996), although other investigators
population emerges somewhat later in development after have suggested an alternative role for FGF-2 (Mackay-Sim
the translocation of the proliferating population basalward and Chuah, 2000). In muscle development, FGF-2 has the
(Holbrook et al., 1995). effect of blocking myocyte differentiation, which is more
Genesis of olfactory neurons occurs continuously analogous to the effect on basal cell differentiation noted
throughout the life of the animal (Graziadei and Monti by Schwob’s lab (Goldstein et al., 1996).
Graziadei, 1978), promoting anatomical and functional
recovery after injury either to the epithelium or the olfac- B. Cellular Differentiation
tory nerve (Costanzo, 1991). We have a better understand-
ing of the process of neurogenesis during postnatal life The most detailed studies on the ontogeny of the various
than during embryonic development (Chapter 5). A variety cell types of the olfactory epithelium have been done in rat
of experimental approaches (including pulse-chase studies and mouse fetuses. The first cells to differentiate are neurons,
with 3H-thymidine in vivo and in vitro and application of ca. E10.25 in mice and E12 in rats, which is shortly
retroviral or other markers for tracing lineage in vivo) have after the placode begins to invaginate (Cau et al., 2000;
indicated that the immediate neuronal precursor cell, i.e., Cuschieri and Bannister, 1975a,b). (The terminology used
the dividing cell whose daughter(s) differentiate into neu- here indicates embryonic development in terms of gesta-
rons, resides among the population of globose basal cells tional age; in rodents conception is assumed to occur at
(Caggiano et al., 1994; Calof and Chikaraishi, 1989; midnight preceding the morning when the dam is found to
Graziadei and Monti Grazeidei, 1979; Schwartz Levey et be sperm-positive.) Slightly later there is a major transition
al., 1991; Schwob et al., 1994). The distribution and densi- in the appearance of the epithelium, which is coincident
ty (number per unit length of epithelium) of proliferating with the shift of mitotic profiles from the apical epithelium
globose basal cells declines dramatically with increasing into the basal compartment. At this time the characteristic
age of the animal (Weiler and Farbman, 1997). In younger distribution of the three major cell types––the lamination
animals (up to about 40 days postnatal) proliferating neu- of supporting cells, olfactory neurons, and basal cells in
ronal precursors are evenly distributed in the basal epithe- the olfactory epithelium into relatively distinct, progres-
lium (Weiler and Farbman, 1997), whereas in older ani- sively deeper epithelial zones––emerges, and the apical
mals they are unevenly distributed and form roughly circu- aspects of the neurons and supporting cells begin to elabor-
lar patches that are distributed across the tangential extent ate the adult-like complement of dendrites and microvilli
of the adult epithelium (Graziadei and Monti Graziadei, (Farbman, 1991; Menco and Farbman, 1985a).
1979; Loo et al., 1996; Weiler and Farbman, 1997). The
surrounding areas are populated with fewer globose basal
1. Olfactory Neurons
cells and are described as quiescent, containing mainly
mature olfactory neurons. Although it has been suggested In olfactory neuron development, genesis of the axon pre-
that horizontal basal cells serve as stem cells that support cedes differentiation of the dendritic process (Cuschieri
ongoing neurogenesis, direct evidence for that notion is and Bannister, 1975a,b; Farbman and Squinto, 1985). Each
lacking (Graziadei and Monti Graziadei, 1979; Holbrook sensory cell body gives rise to a small-diameter axon,
et al., 1995; Mackay-Sim and Kittel, 1991). which fasciculates with other axons, and the bundles pro-
The molecular regulation of basal cell division ject into the lamina propria, at E12 in rats (Farbman and
remains largely unexplored. Analysis of explants of Squinto, 1985) and at E10.5 in mice (Cuschieri and
embryonic/neonatal olfactory epithelium or epithelium- Bannister, 1975a,b). These bundles are the fila olfactoria of
derived cell lines provides a few clues. FGF2 has modest the olfactory nerve (Fig. 4), and they are surrounded by
effects on the proliferation of neuronal precursors in vitro ensheathing cells (see Sec. II. C). The fibers from each
(Calof and Chikaraishi, 1989; DeHamer et al., 1994) and nasal cavity can be roughly divided into two projections: a
of olfactory-derived cell lines (Goldstein et al., 1997; medial projection composed of fibers from the nasal sep-
Vawter et al., 1996). TGF- also stimulates basal cell pro- tum and a lateral one from the turbinates. The olfactory
liferation in explants and cell lines; in particular, it is a nerve fibers reach the presumptive olfactory bulb roughly
potent stimulus for proliferation of the horizontal basal one day later, and morphologically distinct synapses are
cells, which express the EGF receptor (Farbman and demonstrable 2–3 days after that (Farbman, 1986; Gong
Buchholz, 1996; Getchell et al., 2000; Mahanthappa and and Shipley, 1995; Hinds, 1972a,b). In the human embryo,
Schwarting, 1993). Some of the analysis on cell lines axonogenesis also takes place relatively early in gestation.
suggests that FGF-2 suspends neuronal differentiation, At stage 15 (33 days gestation), olfactory axon bundles are
holding the cells in a more globose basal cell-like pheno- seen in the lamina propria (Pyatkina, 1982); they reach the
120 Chuah et al.

third postnatal week, when they reach the adult value of an

average of 11 per knob in the rat (Menco and Farbman,
1985b). About this time, freeze-fracture electron
microscopy reveals also that increasing numbers of
intramembranous particles are inserted into the ciliary
membrane (Menco, 1988). In human embryos, ciliogenesis
begins at the ninth week of gestation, and a dramatic
increase in ciliary number occurs in the subsequent
2 weeks (Pyatkina, 1982). When maturation is complete, it
is estimated that the number of cilia on each dendritic knob
is between 10 and 50 (Chuah and Zheng, 1992; Ohno et al.,
1981). During ciliogenesis the perikaryon of the human
olfactory neuron also undergoes differentiation. The rough
endoplasmic reticulum in the perinuclear region becomes
Figure 4 Specimen from 15-day embryonic rat in which the more elaborate, and free ribosomes are organized increas-
nasal pits containing olfactory placode cells (OP) were impreg- ingly into polyribosomes (Pyatkina, 1982). In vitro studies
nated with a fluorescent dye that traces the olfactory nerves show that although ciliogenesis can occur to a limited
(arrows) projecting medially and laterally towards the olfactory extent in the absence of the bulb (Chuah et al., 1985;
bulb. Bar
50 m.
Farbman, 1977), it is enhanced in the presence of the
presumptive olfactory bulb. This suggests that the final
maturation of olfactory neurons may be regulated by the
forebrain in the region where the olfactory bulb forms establishment of contact with its target tissue (Chuah et al.,
within a week (Bossy, 1980). 1985). However, no cause-and-effect relationship between
The formation of dendrites marks another stage of these two events has been established definitively, as dis-
olfactory neuron development. At the time that neurons cussed more thoroughly below. Indeed, the evidence in the
first appear, the primitive olfactory dendritic processes ter- adult rat suggests that the relative lack of mature neurons in
minate as cytoplasmic expansions at the surface of the the absence of the bulb can be explained by the premature
epithelium and contain several centrioles, which have death of the neurons that occurs in the absence of the troph-
migrated from the perikaryon (Menco and Farbman, ic influence exerted by the bulb on sensory neuron survival
1985a; Mulvaney and Heist, 1971). The cells go through a (Schwob et al., 1992). In humans, ciliogenesis lags genesis
stage when they elaborate a primary cilium. Eventually, the of axons significantly; olfactory axons in humans reach
centrioles become basal bodies and give rise to multiple their target tissue 3 weeks before the first olfactory neurons
cilia (Fig. 5), which are first seen at E15 to E16. Cilia begin to sprout cilia (Bossy, 1980; Pyatkina, 1982).
increase in number and length until about the second or Roughly coincident with the elaboration of cilia is the
expression of the elements that constitute the signal trans-
duction cascade for olfactory stimuli. Members of the
odorant receptor (OR) gene family are expressed by rare
neurons ca. E11.5 in mice (Sullivan et al., 1995), lagging
slightly the morphological emergence of neurons at the
nasal pit stage (E10.5). Probe-positive cells become more
abundant at the time of the aforementioned morphological
transition in the lamination of proliferating cells, the dif-
ferentiation of the surface of epithelium, and ciliogenesis
(E12.5) (Royal and Key, 1999; Sullivan et al., 1995).
Similar timing is observed in rats (Saito et al., 1998;
Strotmann et al., 1995). Initial contact of olfactory axons
with the olfactory bulb occurs about the time that odorant
receptors are first expressed (Hinds, 1972a,b; Gong and
Shipley, 1995). As a consequence, it is unlikely that a
Figure 5 An electron micrograph of rat olfactory epithelium retrograde signal from the bulb to the epithelium is respon-
showing dendritic knobs (D) of olfactory sensory neurons. sible for eliciting the pattern of receptor expression.
Arrows indicate cross sections of cilia. Bar
0.5 m. Indeed, receptor genes are expressed with a spatial distrib-
Developmental Anatomy of the Olfactory System 121

ution that is indistinguishable from normal in the extra- The timing of their expression allows them to be assigned
toes mouse mutant in which the olfactory bulb does not to a particular stage in the process of neuronal differentia-
develop normally due to truncation of Gli3, a zinc finger tion. Thus, neurogenin1/Math4C is expressed by precur-
protein that participates in the sHH signal transduction sors that are downstream of the Mash1-expressing transit
cascade (also described above). G-proteins, which are amplifying cells, and NeuroD is seen in the differentiating
involved in transduction of the olfactory stimulus, are neurons of the olfactory epithelium (Cau et al., 1997). Still
expressed in cilia shortly after they are formed (Mania- other members of the bHLH superfamily, Hes1 and Hes5,
Farnell and Farbman, 1990; Sullivan et al., 1995). Finally, vertebrate homologs of the Hairy-enhancer of Split family
a newly cloned member of the NCAM family of Ig cell in Drosophila, seem to operate to hold in check the elabo-
adhesion molecules termed OCAM or mamFas II (for the ration of Mash1-expressing precursors (Hes1) and, later
mammalian homolog of Fasciclin II) is expressed around on, the differentiation of immature neurons (both Hes1 and
the same time as the first appearance of odorant receptors Hes5 acting together) (Cau et al., 2000). The regulation of
(Yoshihara et al., 1997). OCAM/mamFas II is differential- early neuronal differentiation is little understood beyond
ly expressed according to zone of origin in the olfactory these few items. Cell lines and neurons in explants can be
epithelium, i.e., there are high levels on axons of neurons pushed to differentiate by application with members of the
derived from ventrolateral epithelium (Zone 2 and higher) TGF- superfamily, either alone or in combination with
and very low levels on axons from dorsomedial epithelium other growth factors and cytokines (Mahanthappa and
(Zone 1) in developing and adult olfactory system (Mori et Schwarting, 1989; Vawter et al., 1996).
al., 1985; Schwob, 1992; Schwob and Gottlieb, 1986,
1988; Yoshihara et al., 1997). Like ORs, differential
OCAM expression is maintained independent of the bulb
2. Supporting Cells
in both settings. Additional components of the signal trans-
duction cascade and the multitude of adhesion and matrix Supporting cells are easily distinguished ultrastructurally
molecules that accompany genesis and growth of axons are from olfactory neurons in the early human embryo, but
detailed further below. the literature on their differentiation is scarce. In the 7-
At a molecular level, we know little about the events week-old human embryo they are cylindrical in shape,
that accompany the differentiation of neurons from the with numerous microvilli projecting from the dome-
daughters of immediate neuronal precursors. As for other shaped apical surface (Pyatkina, 1982). A layer of
neuronal precursors and their neuronal descendants, glycocalyx is deposited on the microvilli. The perikaryon
members of the basic helix-loop-helix (bHLH) family of characteristically contains short profiles of rough endo-
transcription factors are expressed in the olfactory epithe- plasmic reticulum, abundant microfilaments, and a large
lium and seem to participate in the process of olfactory number of glycogen granules. By the ninth week of gesta-
neuronal differentiation. Mash1 and its cognate protein tion in humans, a morphologically different supporting
MASH1, a mammalian homolog of the Drosophila cell appears. This cell is considerably narrower, lacks
proneural gene, Achaete-Scute, is expressed by basal cells glycogen granules, and possesses bundles of filaments
that seem to function as transit amplifying cells, i.e., cells along its longitudinal axis. However, it resembles the
that proliferate to expand the population of precursors and predominant type of supporting cell by the presence of
do not give rise to neuronal daughters directly, yet are vesicular inclusions in the perinuclear cytoplasm. It is
committed to neuronal differentiation (Gordon et al., 1995; likely that this second type of supporting cell is actually a
Guillemot et al., 1993). In keeping with that role, elimina- morphologically differentiated state of the first because
tion of Mash1 by homologous recombination produces an ultrastructural and immunohistochemical studies show
epithelium that is largely depleted of neurons by birth that the supporting cells in the adult human contain large
(Guillemot et al., 1993). Interestingly, some differentiating bundles of filaments, particularly in the lower two thirds
neurons are observed at the nasal placode/pit stages in of the cell (Graziadei and Monti Graziadei, 1979;
Mash1 mutant mice before the onset of massive cell death Holbrook et al., 1995; Moran et al., 1982a,b). Maturation
eliminates the neuronal population (Cau et al., 1997), sug- of supporting cells in the perinatal period is marked by a
gesting that not all neurons are equivalent in the early pla- decrease in the electron density of the cytoplasm and a
code. Indeed, a distinct population of pioneer neurons is conspicuous increase in the amounts of smooth and rough
responsible for establishing the olfactory nerve in endoplasmic reticulum.
zebrafish, which then disappear (Whitlock and A nonneuronal microvillar cell has been identified in rat
Westerfield, 1998). Several other bHLH neuronal differen- olfactory epithelium on the basis of its immunoreactivity
tiation genes are expressed in the embryonic epithelium. with a specific monoclonal antibody, 1A-6 (Carr et al.,
122 Chuah et al.

1991). This microvillar cell type is believed to be non-

neural because (1) it has no identifiable axonal process, (2)
it is not reactive with an antibody against olfactory marker
protein, and (3) it survives ablation of the olfactory bulb.
These cells fail to react with a supporting cell-specific
monoclonal antibody (Hempstead and Morgan, 1983) and
consequently are thought to be different from ordinary
supporting cells. It is not clear whether this cell type is the
same as that thought to be a microvillar sensory neuron
(Moran et al., 1982b). However the latter investigators
showed no convincing experimental evidence that the
microvillar cell they described had an axon and was
connected to the olfactory bulb.
Figure 6 A photomicrograph showing a group of cells (arrows)
migrating out of the fetal rat olfactory epithelium. These cells
3. Horizontal Basal Cells
accompany the growing olfactory axons as they project toward
As early as the ninth week of gestation in humans, two the olfactory bulb. Bar
20 m.
types of basal cells are recognizable in the human olfactory
epithelium: one with electron-dense cytoplasm,
corresponding to the horizontal basal cells, the other with ble of giving rise to typical olfactory neurons also
lighter cytoplasm, corresponding to the globose basal cells migrate out of the epithelium along the fascicles of the
(Pyatkina, 1982). A similar delay in the emergence of phe- olfactory nerve during embryonic development as well
notypically distinct horizontal basal cells relative to the as after injury in the adult (Monti Graziadei, 1992;
differentiation of neurons or sustentacular cells is observed Schwob et al, 1995). Hence, the cells of the original
in rodent epithelium as well (Holbrook et al., 1995). olfactory placode give rise not only to intrinsic epithelial
cells, but also to those that eventually function at distant
sites.
C. Extraepithelial Cell Migration

During ontogeny, groups of cells migrate out of the

epithelium before the outgrowth of the first axons from
the olfactory epithelium (Fig. 6) (e.g., Bossy, 1980;
Farbman and Squinto, 1985; Mendoza et al., 1982;
Schwanzel-Fukuda and Pfaff, 1989; Wray et al., 1989).
Recent evidence indicates that these cells are functional-
ly heterogeneous. A large mass of cells derived from the
epithelium occupies the region between the epithelium
and the telencephalon in advance of contact with the
telencephalon. The cell mass may be serving as an inter-
mediate target for the olfactory nerve, given the sharp
bend that the fibers take on contacting the mass (Drapkin
and Silverman, 1999; Gong and Shipley, 1995). Some of
the cells migrating from the medial side of the olfactory
epithelium contain luteinizing hormone-releasing hor-
mone (LHRH); some LHRH-positive cells eventually
reside in the hypothalamus (Schwanzel-Fukuda and
Pfaff, 1989), whereas others become ganglion cells of
the terminal nerve (Schwanzel-Fukuda and Silverman, Figure 7 A photomicrograph of ensheathing cells that had been
1980). Still other migrating cells become the ensheath- isolated from the olfactory nerve layer of the newborn rat olfac-
ing cells of the olfactory nerve (Fig. 7) (Chuah and Au, tory bulb and grown in culture. Most of the ensheathing cells are
1991a) and of the nerve bundles in the outermost layer of spindle-like and bipolar in shape, but some have a few processes.
the bulb (Doucette, 1989). Finally, precursor cells capa- Bar
20 m.
Developmental Anatomy of the Olfactory System 123

III. BIOCHEMICAL AND FUNCTIONAL olfactory neurons (Matsuzaki et al., 1999; Saito et al.,
ASPECTS OF DIFFERENTIATION 1998). Two days later, at E16, the first action potential in
single olfactory neurons can be recorded. At E18 to E19,
Immunochemical and biochemical methods have also been when synaptogenesis between olfactory axons and mitral
used to assess maturation of olfactory neurons. In the E12 dendrites is first observed, an increasing number of olfac-
rat embryo, adenosine deaminase immunoreactivity can be tory neurons become responsive only to specific types of
demonstrated in the olfactory epithelium cells (Senba stimuli, suggesting that they can discriminate among
et al., 1987). With the first axon growth at E13, certain odorants (Gesteland et al., 1982). Interestingly, it is also at
membrane-related antigens are expressed on the olfactory E18 that distinct immunoreactivity for OcNCI at the level
neurons. A monoclonal antibody that binds N-CAM–like of the cilia can be demonstrated ultrastructurally
moieties, Neu-5, is reactive with rat olfactory axons at E13 (Matsuzaki et al., 1999). Consistent with the notion of a
and with the perikarya the next day (Carr et al., 1989). functioning fetal olfactory system, behavioral studies have
Members of the N-CAM family are known to mediate shown that fetal rats at 20 days of gestation are sensitive to
neuron-neuron adhesion in vivo and neurite growth and odor molecules dissolved in amniotic fluid and are able to
fasciculation in vitro (for review, see Rutishauser and undergo odor aversion conditioning (Stickrod et al., 1982).
Jessell, 1988). In the mouse, N-CAM expression in The time of onset of the electro-olfactogram (EOG) in
the olfactory placode is detected at E9, while the rat at E14 coincides with the first expression of the
OCAM/mamFasII, a recently isolated member of the olfactory marker protein (OMP) in the olfactory neurons
N-CAM family, is associated with subsets of olfactory (Allen and Akeson, 1985a). Olfactory marker protein is a
axons at E13 (Yoshihara et al., 1997). Similarly early onset cytosolic, acidic protein, molecular weight about 18 KDa
was noted in rats in vivo (Schwob, 1992). Another cell (Margolis, 1972; Margolis, 1982), found throughout the
adhesion molecule, L1, appears at E11 (Miragall et al., olfactory neuron, from the dendritic knob to the axon ter-
1989). In the mouse, L1 is thought to be involved in inter- minal (e.g., Farbman and Margolis, 1980; Monti
actions between neurons and extracellular matrix, neu- Graziadei et al., 1980). OMP is widely accepted as a
roglia, and other neurons (see Persohn and Schachner, marker for mature olfactory neurons (Farbman and
1987; Seilheimer and Schachner, 1988). More recently, it Margolis, 1980), although its function has not been clearly
has been shown that growing olfactory axons express defined. Studies with OMP-null mice suggest that OMP
retinoic acid–binding protein (CRABP I) as early as the may play a modulatory role in odor detection or olfacto-
12th day of gestation in the mouse (Gustafson et al., 1999). ry signal transduction (Buiakova et al., 1996) and /or in
It is uncertain when N-CAM is first synthesized in the neurogenesis (Carr et al., 1998). The appearance of OMP
human olfactory system, although it is present as early as the in humans is first apparent at about 24 weeks postconcep-
17th week of gestation. At this time, N-cadherin, another cell tion, i.e., about 1 month after synaptogenesis has occurred
adhesion molecule, is also demonstrable in the olfactory (Chuah and Zheng, 1987; Johnson et al., 1995).
epithelium (Chuah and Au, 1991b). The expression of these In addition to OMP, blood group antigens have been
membrane-related molecules at a period of extensive olfac- detected by immunostaining in rat olfactory neurons (Astic
tory nerve growth suggests that these molecules are involved et al., 1989). H antigen immunoreactivity appears on E14
in modulating axonal growth. Interestingly, ensheathing cells in the cell body, dendrite, and axon terminal, and the B
that accompany the growing axons also express N-CAM and antigen is detectable on E16 in some of the cells express-
L1 (Gong and Shipley, 1996; Miragall et al., 1989). ing the H antigen. Blood group antigens are normally
Several lines of evidence from morphological, behav- found in low amounts on a very small number of CNS or
ioral, and electrophysiological studies indicate that the PNS neurons; the significance of the presence in olfactory
mammalian olfactory system may be functional before neurons is not known.
birth (Cuschieri and Bannister, 1975a,b; Hinds and Hinds, Another cell surface glycoprotein expressed on olfac-
1976; Gesteland et al., 1982; Stickrod et al., 1982). Given tory neurons is recognized by a monoclonal antibody
the early onset of receptor expression that has been referred to as 2B-8 (Allen and Akeson, 1985b). In adult
documented in the rat and mouse (as described above), it rats, about 25% of OMP-positive olfactory neurons are
is not surprising that at E14 rat olfactory neurons begin to immunoreactive with 2B-8. The relationship between this
exhibit odorant-induced electrical activity (Gesteland et antigen and early differentiation is yet to be elucidated.
al., 1982). At this stage, molecules involved in olfactory Carnosine synthetase activity has been detected as early
signal transduction, such as olfactory cyclic nucleotide- as E16 in the embryonic rat (Margolis et al., 1985). This
gated channel subunit 1 (OcNC1) and Gb, can be demon- enzyme is involved in the synthesis of the dipeptide carno-
strated immunohistochemically in select populations of sine (β-alanylhistidine), a major constituent of mature
124 Chuah et al.

olfactory cells and their terminals in the olfactory bulb (Talamo et al., 1989). The reasons for the shift in loca-
(Margolis, 1980). The biological function of carnosine tion of this neurofilament subunit are unclear. The 73
remains unknown, although a neurotransmitter role has and 200 kDa neurofilament subunits are not detectable in
been hypothesized (Margolis, 1980). Biochemical and human olfactory tissue. However, the 200 kDa subunit
neurochemical data are consistent with this idea (Hirsch has been convincingly demonstrated in cell lines derived
and Margolis, 1979; Margolis, 1974, 1980; Margolis et al., from the human olfactory epithelium (Vawter et al.,
1979), but electrophysiological studies have produced 1996).
conflicting results (Frosch and Dichter, 1984; Gonzales-
Estrada and Freeman, 1980; Macleod and Straughan,
1979; Tonosaki and Shibuya, 1979). During ontogeny,
IV. EXTRINSIC INFLUENCE ON OLFACTORY
carnosine can be demonstrated immunohistochemically in
NEURON MATURATION
rat olfactory neurons at E17 (Biffo et al., 1992). Carnosine-
like immunoreactivity is also present in human olfactory A. Influence from the Olfactory Bulb—Maturation
neurons (Sakai et al., 1990). and Survival
One marker that olfactory neurons have in common
with mature neurons in other systems is neuron-specific In many parts of the nervous system, the target organ of
enolase (NSE). Following axotomy of olfactory nerves in the growing axon has a major effect on the survival or
the guinea pig, mature neurons die and within 7 days are maturation of the neurons. For example, embryonic
replaced by newly differentiating neurons which show motoneurons are unable to survive beyond a certain period
immunoreactivity for NSE (Yamagishi et al., 1989). In the without connections to muscle cells (e.g., Oppenheim
human embryo, NSE is already present at the end of the et al., 1978). Similarly, neurons of the ciliary ganglion
first trimester (Takahashi et al., 1984). are dependent upon their target tissue for survival (Pilar
More complicated is the question of which type IV and Landmesser, 1976). Superior cervical ganglia grown
intermediate filament proteins are expressed by olfactory with target salivary glands show greater elaboration and
neurons. The predominant intermediate filament proteins directionality of nerve fiber outgrowth than control
expressed in the olfactory axons in the rodent are explants (Coughlin et al., 1978). In 1975 Cuschieri and
vimentin and peripherin (Escurat et al., 1990; Gorham Bannister hypothesized that the olfactory bulb had an
et al., 1991; Schwob et al., 1986). Vimentin is also seen in influence over the maturation of olfactory neurons
the olfactory nerve fibers of other species (Ophir and because ciliogenesis was complete only after the axons
Lancet, 1988). Different research groups have produced had reached their target.
conflicting reports of whether neurofilaments are Support for this hypothesis comes from results of organ
expressed in olfactory neurons (Bruch and Carr, 1991; culture experiments and degeneration/reconstitution
Ophir and Lancet, 1988; Schwob et al., 1986; Takahashi experiments. When olfactory mucosa is explanted alone,
et al., 1984; Talamo et al., 1989; Vollrath et al., 1985; the olfactory neurons differentiate to some extent but fail
Yamagishi et al., 1989). The evaluations have all been to reach full maturation. The differentiating olfactory
immunohistochemical and have employed a number of neurons grow axons and express some cilia on the dendritic
different polyclonal and monoclonal antibodies. Vollrath knobs; EOGs can be recorded from these cells, and some
and coworkers (1985) failed to demonstrate any neuro- of them synthesize OMP (Chuah and Farbman, 1983;
filament staining in rat olfactory epithelium, whereas Chuah et al., 1985; Farbman, 1977; Farbman and
Schwob and colleagues (1986), using several affinity- Gesteland, 1975). When the olfactory mucosa is cocul-
purified polyclonal antisera, demonstrated that expression tured with the bulb, twice as many olfactory neurons con-
of neurofilament proteins is limited to a small subpopu- tain OMP, and the number of cells with ciliated dendritic
lation of neurons in the lateral olfactory epithelium and knobs also increases twofold (Chuah and Farbman, 1983;
their axons projecting via the lateral side of the olfactory Chuah et al., 1985). This enhancing influence appears to
nerve layer of the bulb. In contrast, Bruch and Carr be mediated by interaction between the bulb and olfactory
(1991) found that prominent immunostaining for the 200 mucosa; however, the target tissue does not appear to con-
kDa neurofilament was present in the cell bodies of trol directly the trajectory of the olfactory axons (Gonzales
rodent olfactory neurons. et al., 1985).
In humans, the 145 kDa neurofilament is demonstra- After unilateral ablation of the olfactory bulb in mam-
ble histologically in the olfactory neuron cell bodies at mals, the olfactory neurons undergo degeneration and new
the 16th week of gestation (Takahashi et al., 1984). In neurons are generated from globose basal cells to repopu-
the adult this molecule is located only in the axons late the olfactory epithelium. However, reconstitution of
Developmental Anatomy of the Olfactory System 125

the epithelium is incomplete as it does not usually reach its In the early stages of axon growth during ontogeny or
preoperative thickness (Costanzo and Graziadei, 1983). reconstitution of the olfactory epithelium, axons are
The number of mature olfactory neurons, as evidenced by enveloped by cytoplasmic processes of ensheathing
the presence of OMP, is greatly reduced. On the other cells. Doucette (1990) has suggested that the ensheath-
hand, the number of immature elements is greatly ing cells guide the olfactory axons to their target and that
increased compared to the control side (Monti Graziadei, the guidance is probably modulated by cell adhesion
1983; Monti Graziadei and Graziadei, 1992; Schwob et al., molecules, extracellular matrix molecules, and
1992; Verhaagen et al., 1990). Interestingly, a recent study chemotropic substances. This notion is supported by the
has shown that the remaining OMP-positive neurons pre- positive immunohistochemical staining for the neuronal
sent after bulbectomy demonstrate a significant elevation cell adhesion molecules, N-CAMs and L1, on olfactory
in their OMP level (Carr et al., 1998). neurons, ensheathing cells, and in the mesenchyme sur-
A possible explanation for the failure of the epithelium rounding the developing olfactory pathway in rodents
to be fully reconstituted is that the sensory neurons are (Gong and Shipley, 1996; Miragall et al., 1988, 1989;
trophically dependent on the bulb for their prolonged sur- Whitesides and LaMantia, 1996). The use of antibodies
vival (Schwob et al., 1992; Weiler and Farbman, 1999). In to adhesion molecules on cultures of olfactory neurons
the absence of the bulb, neurons are produced at a twofold growing on astrocyte monolayers has provided some
greater rate than in a control situation (Carr and Farbman, insight into the functional role of these molecules in
1992; Schwob et al., 1992), but nearly 90% of the neurons olfactory axonal growth. Neurite outgrowth from cul-
on the operated side die before they reach the age of tured olfactory neurons is inhibited by antibodies to
2 weeks (Schwob et al., 1992). That is shorter by far than NCAM, N-cadherin, and L1 (Chuah et al., 1991). The
any determinations of life span in the normal epithelium fact that OCAM, a member of the N-CAM family, is
(Graziadei and Monti Graziadei, 1978; Hinds et al., 1984; expressed by olfactory neurons in the early stage of pre-
Mackay-Sim and Kittel, 1991). In other words, the data natal development and is also present in restricted zones
suggest that if an olfactory sensory neuron does not receive of the olfactory epithelial sheet suggests that OCAM
a required trophic factor from the bulb at a critical stage in may be involved in broad segregation of nerve fascicles
development, it is very likely to die before reaching full (Yoshihara et al., 1997).
maturity. Consequently, most of the cells in the reconsti- In humans, N-CAM immunoreactivity in the olfactory
tuted epithelium are relatively immature. These data have nerve bundles can be demonstrated as early as the 18th
led to the suggestion that olfactory neurons are trophically week of gestation. Western blot analysis shows that the
dependent on the bulb not only for maturation as discussed N-CAMs of the human olfactory nerve consist of all three
above (Chuah and Farbman, 1983; Chuah et al., 1985), but molecular isoforms: N-CAM 180, N-CAM 140, and N-CAM
also for their survival (Schwob et al., 1992). Indeed, it is 120 (Chuah and Au, 1992).
difficult to separate a direct effect on maturation per se In addition to adhesion molecules, extracellular matrix
from a failure to mature due to abbreviated neuronal sur- has also been implicated in stimulating and sorting olfac-
vival. A recent study has in fact shown that mitral cells, the tory axons as they course towards the olfactory bulb. In the
major postsynaptic target, are crucial in maintaining the rat, laminin and heparan sulfate proteoglycans, which are
survival of olfactory neurons (Weiler and Farbman, 1999). expressed by ensheathing cells, are associated with the
It was found that depletion of mitral cells resulting from developing nerve pathway as early as E13 (Liesi, 1985;
transection of their axons in the lateral olfactory tract led Treloar et al., 1996). It is thought that these two molecules
to increased numbers of proliferating neurons in the olfac- provide a conducive substrate on which olfactory axons
tory epithelium, presumably related to and resulting from elongate. In contrast, chondroitin sulfate proteoglycans,
an increased level of cell death. which are selectively present in the mesenchyme and mar-
ginal zone of the telencephalon, may contribute to restrict-
B. Regulation of Olfactory Axonal Growth ing axonal growth along a particular trajectory (Treloar
et al., 1996). Concerning the effect of laminin on olfactory
In the process of attaining full functional maturity, olfac- nerve outgrowth, cell cultures of dissociated olfactory
tory neurons need to extend axons from the epithelium to neurons on laminin have produced conflicting results.
the olfactory bulb and to make accurate contact with the Although some studies showed that olfactory neurons were
appropriate dendrites in specific glomeruli. This type of able to adhere to laminin and subsequently differentiate
precise target recognition is probably regulated by a series into bipolar cells (Pevsner et al., 1988; Pixley and Pun,
of distinct guidance cues, arranged in a hierarchical man- 1990), Chuah and colleagues found that the percentage of
ner (Lin and Ngai, 1999). olfactory neurons growing neurites when placed on
126 Chuah et al.

laminin was negligible (Chuah et al., 1991). Recent in Recently, the isolation of several members of the OR37
vitro studies show that exogenous laminin enhances the subfamily of receptors has elucidated the extent of the
spreading and migration of ensheathing cells, and this topographic specificity. OR37A-E are each expressed in
probably facilitates their function as a conduit of axonal a patch in the posterodorsal epithelium and are not
elongation (Tisay and Key, 1999). found throughout the full anteroposterior extent of the
Other classes of molecules have been implicated in the epithelium, in contrast to the expression patterns of
sorting out and fasciculation of axons during development. more typical receptors (Kubick et al., 1997). Each set of
A candidate is the carbohydrate-binding protein galectin-1, neurons of the OR37 subfamily project to a single, dif-
which is first apparent in the mesenchyme surrounding the ferent glomerulus at the ventral margin of the bulb
nasal cavity at E15 and which can be localized distinctly to (Strotmann et al., 2000). In contrast to the absolute
ensheathing cells at E17 when the first nerve fibers have positional specificity reported for zebrafish glomeruli
reached the olfactory bulb (St. John and Key, 1999). (Dynes and Ngai, 1998; Friedrich and Korsching,
Expression of galectin-1 is maintained throughout devel- 1997), the position of the OR37 subgroups of glomeruli
opment, and in the postnatal rat it can be observed in the are not fixed but vary relative to each other (Strotmann
ensheathing cells surrounding the axon bundles in the lam- et al., 2000). The means by which receptor choice
ina propria as well as those residing in the olfactory nerve directs targeting and glomerular convergence remains a
layer of the bulb (St. John and Key, 1999). subject of intense investigation. Recent evidence
In summary, the data are consistent with the notion that suggests that activity, and more importantly coordinate
ensheathing cells express specific surface molecules that activity by a set of olfactory neurons, is crucial for the
promote axonal elongation. Ultrastructural observations acquisition and maintenance of glomerular territory.
reveal that these cells are intimately related to the early for- The experiments take advantage of the fact that elimi-
mation of olfactory axons—their processes are always pre- nation of the OcNC1 subunit of the cyclic nucleotide
sent ahead of growing axonal terminals (Tennent and receptor abolishes stimulus-evoked olfactory neuronal
Chuah, 1996). In addition to the expression of extracellular activity (Brunet et al., 1996). Some OR-defined classes
matrix molecules, ensheathing cells may also be a rich of neurons innervate multiple glomeruli in the region
source of growth factors. It has been shown that these cells where they would normally terminate, rather than the
produce brain-derived neurotrophic factor, glial cell single glomerulus that is typical (Zheng et al., 2000).
line–derived neurotrophic factor, and neuregulins such as For other neuron types, targeting appears normal (Lin
glial growth factor 2 (Chuah et al., 2000; Salehi-Ashtiani et al., 2000; Zheng et al., 2000). Intriguingly, Zhao and
and Farbman, 1996; Woodhall et al., 2001). Whether Reed (2001) have taken advantage of the fact that rough-
the secreted growth factors play a role in promoting ly half of the neurons in a female mouse heterozygous
axon growth is yet to be determined. The olfactory bulb for the channel mutation are silenced via allelic inacti-
represents another candidate for directional guidance of vation of genes, such as OcNC1, on the X chromosome.
olfactory axons. In vitro studies show that the olfactory bulb In this case, their lab has shown that silenced neurons
secretes soluble molecules that act as chemoattractants to are gradually excluded from glomeruli and eventually
ensheathing cells (Liu et al., 1995). The identity of these from the epithelium as a whole.
chemoattractants has not been elucidated, and they could
well be any number of the growth factors that are known to
be present in the olfactory bulb (Mackay-Sim and Chuah,
V. DEVELOPMENT OF THE OLFACTORY BULB
2000).
Once the olfactory axons enter the olfactory bulb, A. Anatomy
additional cues may be required to fine-tune target
recognition and ensure that axons terminate in the The general anatomical features of bulb development in all
appropriate glomeruli. Experiments with genetically mammals are similar to one another. Most of the descrip-
engineered mice show that expression of an odorant tive and experimental studies have been done on rodent
receptor is required for convergence to specific embryos, and most of what follows is based on the results
glomeruli (Mombaerts et al., 1996; Wang et al., 1998). of these studies. Where information is available, we have
Although the exact mechanisms are not known, recep- included the stages of human olfactory bulb development.
tor-swap experiments have shown that odorant receptors The bulb is derived from the rostral region of the cere-
play an instructive role in glomerular targeting and bral (telencephalic) vesicle of the early mammalian
establishing a topographic map of the projection onto embryo. Before the bulb becomes apparent as an entity,
the bulb (Mombaerts et al., 1996; Wang et al., 1998). the cerebral vesicle is a fluid-filled cavity lined with an
Developmental Anatomy of the Olfactory System 127

epithelium divisible into two regions: a highly cellular Maintained contact between sensory epithelium and
ventricular region bordering the ventricle and an acellular the telencepalon is required for induction and proper for-
marginal region. In the rat, between the 11th and 14th mation of the olfactory bulb. We have already noted that
embryonic days, an increasing number of axons from the ablation of the placode prevents contact, telencephalic
olfactory epithelium reach the cerebral vesicle and align evagination, and bulbar differentiation (Burr, 1916;
themselves parallel to the vesicle surface. A subpopulation Stout and Graziadei, 1980). A variety of malformations
of these axons penetrate the epithelium lining the most ros- and teratogens, including the extra-toes mouse mutation
tral and inferior part of the cerebral vesicle and extend past mentioned above (Johnson, 1967) and the small eye
the marginal layer to end deep in the ventricular layer. The mutation (Dellovade et al., 1998), may also cause induc-
entry of the first or “pioneer” olfactory axons into the bulb tion to fail. In the majority of these cases, the olfactory
precursor is correlated with an average increase in the epithelium has formed and has elaborated axons. One of
length of the cell cycle in the ventricular region of the bulb the more instructive examples is the human disorder
precursor, compared to that in the ventricular region of the termed Kallmann syndrome. Patients with Kallmann
adjacent cerebral vesicle (Gong and Shipley, 1995). The syndrome are identified on the basis of anosmia and
significance of this increase in the average cell cycle length hypogonadotrophic hypogonadism (Kallmann et al.,
is that, in the bulb precursor, more postmitotic cells exit 1944). The olfactory bulb is absent (Yousem et al.,
from the cell cycle and begin their differentiation into pre- 1993), but the olfactory epithelium forms and generates
sumptive mitral/tufted cells than is the case in the adjacent olfactory neurons (Schwanzel-Fukuda et al., 1989;
cerebral vesicle epithelium. Thus, in the forebrain the Schwob et al., 1993). The mutated gene, KAL1, has been
mitral/tufted cell precursors are among the first to begin identified as mutated in the X-linked variant of the
their differentiation. disease and encodes anosmin-1, a cell adhesion protein
At E15-E16 in rats (at E13-14 in mice) the presumptive expressed by the presumptive telencephalic anlage of the
bulb becomes more obvious as an evagination of the ros- bulb around the time that it is contacted by axons from
tral end of the telencephalic vesicle. In the human embryo, the epithelium (Franco et al., 1991; Legouis et al., 1991,
this stage begins at 37–41 days after conception. At the 1993; Rugarli, 1999). It has been suggested that the con-
same time, the sensory epithelium is expanding. Along tact between the olfactory nerve and bulbar anlage
with this expansion, more olfactory axons project from the breaks down due to the lack of anosmin-1, leading to the
sensory epithelium to the presumptive bulb and add to the failure of bulb induction (Legouis et al., 1993; Rugarli,
numbers of parallel axons along the bulb surface, outside 1999). As a likely consequence of the absence of the
of the marginal layer. These axons form what will later bulb, the olfactory epithelium is highly abnormal in
become the outer nerve layer of the bulb. For a few days biopsy specimens from a Kallmann patient and from
axons do not penetrate into the presumptive bulb (see another individual with congenital anosmia without
Bailey et al., 1999). This raises the question of how accompanying endocrinologic abnormalities. Only
molecular interactions between these axons and the bulb immature olfactory neurons, which lack cilia on the
precursor might differ from those of the pioneer axons dendritic knobs, are seen (Leopold et al., 1992; Schwob
which, at an earlier stage, do have the ability to penetrate et al., 1993). Some of the axon bundles in the lamina
into the lining of the cerebral vesicle. propria are devoid of ensheathing cells and form dense
The bulb increases in size and takes on a more defini- neuromas. Ultrastructural observations show that the
tive shape after a constriction forms in the ventricular abnormal axons are either swollen or their membranes
cavity. The constriction occurs between what will are fragmented, suggesting that the axons are degenerat-
become the olfactory ventricle, a transient central cavity, ing (Leopold et al., 1992; Schwob et al, 1993).
and the remainder of the lateral ventricle that persists in
the adult cerebrum. As the bulb increases in size, it B. Proliferation of Neuron Precursors and
grows rostrally and expands in diameter. The growth pat- Formation of Laminae
tern results in the relative caudal displacement of the
ventricle so that it is no longer seen in histological Before the evagination of the bulb from the cerebral vesicle,
sections of the bulb proper. In humans this occurs by the many of the first cells to exit from the cell cycle are the fore-
19th week (Humphrey, 1940). In rats and mice, the runners of mitral cells. In human embryos (Fig. 8) mitral
ventricle recedes caudally during the first 2–3 weeks fol- cells are first noticeable in the presumptive bulb at stage 21
lowing birth. Associated with the expansion in size is the (~52 days), and they become significantly larger at stage 22
loss of the embryonic marginal layer and the appearance (~60 days) (Humphrey, 1940; Humphrey and Crosby, 1938).
of the adult laminae. After bulb evagination (10 weeks in human embryos), mitral
128 Chuah et al.

Figure 8 Embryonic development of the human olfactory bulb. (A) At about 41 days gestation, olfactory nerve fibers form a rudimen-
tary olfactory nerve layer (ONL) along the ventral side of the presumptive olfactory bulb. Lamination is absent at this stage. (B) At about
60 days gestation the mitral cell layer (MCL), which consists of the large cell bodies of mitral cells, appears most distinct. The external
plexiform layer (EPL) can be distinguished as a thin lamina immediately deep to the external granular layer (EGL), a cellular region sand-
wiched between the developing EPL and ONL. (C) At about 70 days of gestation, all the laminae of the olfactory bulb are formed, with
the glomerular layer (GmL) developing last. The MCL appears less distinct as the distance between adjacent mitral cells increases. The
thickness of the EPL has increased dramatically. GL, granular layer.

cells become larger and have migrated away from the ven- Rosselli-Austin and Altmann, 1979). The ventricular zone as
tricular zone, where they were generated, to a more periph- a site of neurogenesis disappears at E18 in mice, one day
eral location (Chuah and Zheng, 1992), where they begin to before birth. The interneurons generated after this time are
form a mitral layer. However, in the human bulb the mitral generated in a subventricular (subependymal) zone, first in
layer is never as clearly defined as that in rodents, so that it the olfactory ventricle, but later, after the ventricle recedes
is difficult to distinguish a clear boundary between it and the from the olfactory bulb, they are generated in the subven-
adjacent external plexiform layer. tricular zone of the anterior region of the lateral ventricle.
The time of origin of the various neuron types has been Although most––perhaps as many as 90%—of these new
well studied in the mouse embryo (Hinds, 1968a,b). At neurons die soon after they are produced (Morshead and
E12 (the 12th embryonic day; the mouse has a gestation Van der Kooy, 1992), some do survive and migrate via the
period of about 19 days) precursors to mitral cells undergo rostral migratory stream into the bulb, where they become
their final cell divisions in the ventricular layer and migrate granule or periglomerular neurons and are incorporated
peripherally. They reach their definitive locations about into the neural network within the bulb (Altman, 1969; Lois
3 days later where they form a discrete layer, the mitral cell et al., 1996; Luskin, 1993). Neuronal precursors migrate as
layer, much more obvious and clearly defined than in the chains of cells passing through “tunnels” of glial cells in the
human bulb. Generally the smaller and more superficially rostral migratory stream (Lois et al., 1996). These neuronal
located tufted cells arise in the ventricular layer later than precursors continue to divide as they migrate. Migration
the mitral cells and they migrate past the mitral cells to the appears to be dependent on the presence of polysialic
presumptive external plexiform layer where they are found acid–rich N-CAM (neural cell adhesion molecule) on the
at all levels, some close to the mitral cell layer, some surfaces of the neuronal precursors (Bonfanti and
intermediate and others more superficial, near the Theodosis 1994; Hu et al. 1996).
glomerular layer. Tufted cells appear to be formed in an In development of the neuronal populations in the bulb,
inside-out gradient, i.e., those nearest the mitral layer are then, the mitral cells develop first and establish a one-cell-
formed first, and those that take up residence higher in the thick layer. Tufted cells arise later and come to rest at
external plexiform layer are formed later. varying levels between the mitral cell and the nerve layer
In the mouse, and in other mammals, the last neurons to of the bulb. Because most of the granule and periglomeru-
form are the interneurons, the small periglomerular and lar cells are not present at birth, the granule cell layer and
granule neurons. In rodents, the earliest interneurons are the external plexiform layers are rather thin, and glomeruli
generated in the subventricular (subependymal) layer a few are not outlined by a ring of periglomerular cells as they
days before birth, but most are generated within the first 2 are in the mature bulb. Most of the growth of the granule
or 3 weeks after birth (see Bayer, 1983; Hinds, 1968a; cell layer and the external plexiform layer occurs in the
Developmental Anatomy of the Olfactory System 129

layer between E17 and postnatal day 10 in the mouse

(Hinds 1968a,b). These probably include both astrocytes
and ensheathing cells originating from the olfactory pla-
code. The stages are similar in the rat, but they begin at
E13-E16 (Bayer, 1983).
The number of the various neuronal types in the devel-
oping olfactory bulb can be influenced by physical and
chemical teratogens, such as x-irradiation (Bayer and
Altman, 1975), phenobarbital (Rosselli-Austin and Yanai,
1989), and alcohol (Bonthius and West, 1991). In experi-
mental animals postnatal administration of alcohol appears
to induce cell death or interfere with neurogenesis or
migration of granule cells. Although some degree of recov-
ery in the granule cell population occurs following cessa-
tion of alcohol treatment, the effect on mitral cells is per-
manent (Bonthius and West, 1991). This is probably
because all of the mitral and tufted cells are produced
during a brief window of development, whereas granule
and periglomerular neurons are produced continually
throughout life.

C. Formation of Glomeruli

Within the presumptive bulb (rat, E11-12) a population of

glial cells is organized radially from the ventricular to the
marginal layer. By E15, the proximal and distal ends of
these radial glia form two plexuses: one in the marginal
zone of the presumptive bulb, the other in the subventricu-
lar zone (Bailey and Shipley, 1993; Bailey et al., 1999).
The plexus in the marginal (sub-pial) layer may act as a
temporary barrier, of sorts, against invasion of olfactory
axons in the nerve layer (Bailey et al., 1999; Treloar et al.,
1999; Valverde et al., 1992).
The details of glomerulus formation have been studied
Figure 9 Photomicrographs demonstrating changes in the lam-
most intensively in the developing rat brain and in an
inae of the rat olfactory bulb from the time of birth (A) until invertebrate, the moth. The principles of construc-
10 days postnatal (B). Most granule cells are generated postnatally; tion appear to be similar. After the radial glia and the
consequently the granular layer (GL) undergoes extensive growth presumptive nerve layer of the bulb become established at
after birth. The external plexiform layer (EPL) also increases in E16 in the rat, axon bundles from the nerve layer begin to
thickness because of the more complex dendritic arborization of penetrate into the presumptive bulb (in the moth, the anten-
mitral cells and the arrival of centripetal inputs. The glomerular nal lobe), forming knots of neuropil or “protoglomeruli”
layer (GmL) is better defined as the number and size of the indi- just beneath the bulbar surface (Graziadei and Samanen,
vidual glomeruli increase. MCL, mitral cell layer; ONL, olfac- 1980; Graziadei et al., 1979; Malun and Brunjes, 1996) as
tory nerve layer. Bar
30 m. they intermingle with the radial glia (Bailey et al., 1999;
Treloar et al., 1999; Valverde et al., 1992) [see also devel-
opment in the moth (Oland and Tolbert, 1996; Oland et al.,
first weeks after birth in rodents (Fig. 9) (reviewed by 1990)]. These protoglomeruli are the glomerular precur-
Brunjes and Frazier, 1986). sors and are demarcated by glial processes. Some sorting
Neuroglia arise from scattered proliferating glioblasts process must occur before olfactory axons form the
originally derived from the subventricular germinal layer. protoglomeruli because all of the axons from cells express-
Many proliferating cells are found in the olfactory nerve ing a particular olfactory receptor go to only two glomeruli
130 Chuah et al.

in the bulb—one located laterally and one medially edges that restrict olfactory axons inside protoglomeruli
(Mombaerts et al., 1996; Ressler et al., 1994; Vassar et al., (González et al., 1996; Valverde et al., 1992). It has been
1994). Experiments with genetically manipulated mice shown that experimentally induced reduction of the num-
have shown that deletion of a particular receptor gene ber of glial cells causes disruption or the complete disap-
results in spreading of axons across a wider expanse of pearance of glomeruli in the olfactory lobe of the moth
bulb, whereas insertion of an “incorrect” receptor gene (Oland et al., 1988). Efforts to determine whether the actual
into an olfactory neuron results in the axons terminating in presence of the postsynaptic partners of olfactory axons,
a different glomerulus than in the unperturbed animal i.e., periglomerular or mitral/tufted cells, are necessary for
(Mombaerts et al., 1996). glomerular formation have indicated that they are not. As
Soon after the axons reach the target glomerular region, shown in the moth (Oland and Tolbert, 1998) and in
they form extensive branches and begin to form synaptic mutant mice (Buffone et al., 1998), olfactory glomeruli
relationships with dendrites of mitral and tufted cells develop in the absence of target cells.
(Halaász and Greer, 1993; Kasowski et al., 1999, Klenoff Olfactory axons clearly play a key role in development
and Greer, 1998; Treloar et al., 1999). On a molecular of glomeruli. Indeed, as indicated above, if afferent input
level, the expression of the gene, sonic hedgehog, by mitral is experimentally removed by extirpating the olfactory
and tufted cells is associated with branching of axons with- placode, the vertebrate olfactory bulb does not form at all
in glomeruli and glomerular formation (Gong and (e.g., Burr, 1916; Stout and Graziadei, 1980; Venneman et al.,
Farbman, 1999). Axons form their synapses on the 1982). However, glomeruli do form in mutant mice in
dendrites of mitral cells and on periglomerular cell which the axons are rendered unable to conduct an
dendrites. It has been shown that as the olfactory epithe- impulse, by deleting either a gene coding for a channel
lium expands and more axons grow to the bulb, the newly subunit (Brunet et al., 1996) or a gene coding for a G-protein
arrived axons, which express GAP-43, a marker for young necessary for signal transduction (Belluscio et al., 1998).
olfactory neurons, occupy the central region of the These experiments show that neural activity of olfactory
glomerulus, whereas those that had arrived previously, axons is not required for glomerulus formation but the
more mature axons that express OMP, are pushed to the presence of the axons is. Undoubtedly axons either
lateral regions of the glomerulus (Treloar et al., 1999). The produce a factor or carry a factor on their surface that is
synapse-specific molecule synaptophysin is expressed in important in glomerular formation.
GAP-43-positive axons, suggesting that synapses are The ability of olfactory axons to induce formation of
formed by these axons before becoming fully mature, i.e., glomeruli is not restricted to the glomerular layer of the
before expressing OMP (Kim and Greer, 2000; Treloar et bulb, but under experimental conditions can also induce
al., 1999). However, beginning at around the 12th post- glomeruli at ectopic sites in the bulb, in other parts of the
natal day in rats, immature axons entering glomeruli were telencephalon, in the diencephalon, midbrain, and hind-
distributed in the periphery and moved toward the center as brain (Graziadei and Kaplan, 1980; Graziadei and
they matured (Kim and Greer, 2000). This change in Samanen, 1980; Magrassi and Graziadei, 1985; Monti
trajectories of axons into glomeruli suggested that different Graziadei and Graziadei, 1992). In some of these experi-
rules may be followed in establishing glomeruli and main- ments fragments of cerebral or cerebellar cortex were trans-
taining them. planted into the space following partial or complete bul-
Early in glomerular development a mitral cell may pro- bectomy. The ingrowing axons were able to reorganize
ject apical dendrites to more than one glomerulus, but dur- components of these “inappropriate” target tissues to form
ing maturation all but one apical dendrite are withdrawn, so glomeruli. However, when fragments of olfactory epithe-
that each mitral cell projects to only one glomerulus in the lium are excised and transplanted to other parts of the brain,
adult. Thus, given that (1) sensory neurons each express a the axons invade the host tissue but fail to form glomeruli
single receptor and (2) all neurons expressing a given odor- (Monti Graziadei and Graziadei, 1983; Morrison and
ant receptor project to only two or three glomeruli in the Graziadei, 1983). Graziadei and Monti Graziadei (1986)
bulb (Ressler et al., 1994; Vassar et al., 1994), the fact that have suggested that the glomerulus can form in any part of
each mitral cell is postsynaptic to a single glomerulus indi- the brain only if it receives axonal contributions from a
cates that it is responsive to a very narrow range of odorant broad region of the olfactory epithelial sheet. This is con-
stimuli (reviewed in Mori et al., 1999). sistent with the fact that innervation of individual glomeruli
The organization of glomeruli into a distinct lamina is derived from the convergence of several thousand indi-
seems to depend largely on an interaction between olfac- vidual olfactory axons derived from cells broadly distrib-
tory axons and radial glia (Bailey et al., 1999; Oland and uted within four zones of the olfactory sheet (Astic and
Tolbert, 1990). Glial cells form boundaries delimiting the Saucier, 1986; Astic et al., 1987; Ressler et al., 1993).
Developmental Anatomy of the Olfactory System 131

Olfactory sensory neurons may also play a role in the Baker, H., and Farbman, A. I. (1993). Olfactory afferent regula-
initiation and maintenance of neurotransmitter expres- tion of the dopamine phenotype in the fetal rat olfactory sys-
sion in some bulbar neurons. Embryonic olfactory bulbs tem. Neuroscience 52:115–134.
co-cultured with olfactory epithelium possess more tyro- Baker, H., Kawano, T., Margolis, F. L., and Joh, T. H. (1983).
Transneuronal regulation of tyrosine hydroxylase expression
sine hydroxylase–containing neurons than if they are cul-
in olfactory bulb of mouse and rat. J. Neurosci. 3:69–78.
tured alone (Baker and Farbman, 1993). In deafferented
Bayer, S. A. (1983). 3H-thymidine-radiographic studies of neuro-
olfactory bulbs, dopamine in periglomerular, external, genesis in the rat olfactory bulb. Exp. Brain Res. 50:329–340.
and middle tufted cells is severely reduced (Kawano and Bayer, S. A., and Altman, J. (1975). The effects of x-irradiation
Margolis, 1982; Kream et al., 1984); so is the number of on the postnatally-forming granule cell populations in the
periglomerular cells expressing tyrosine hydroxylase and olfactory bulb, hippocampus, and cerebellum of the rat. Exp.
its messenger RNA (Baker et al., 1983). Following rein- Neurol. 48:167–174.
nervation of the bulb, dopamine levels and the number of Bell, E. T. (1907). Some experiments on the development and
tyrosine hydroxylase immunoreactive cells return to con- regeneration of the eye and nasal organ in frog embryos. Arch.
trol values, indicating that olfactory neurons are crucial Entwicklungsmech. Org. 23:457–478.
in the maintenance of neurotransmitter synthesis in bul- Belluscio, L., Gold, G. H., Nemes, A., Axel, R. (1998). Mice defi-
cient in Golf are anosmic. Neuron 20:69–81.
bar cells (Baker et al., 1983; Kawano and Margolis,
Biffo, L., Marti, E., and Fasolo, A. (1992). Carnosine, nerve
1982).
growth factor receptor and tyrosine hydroxylase expression
during the ontogeny of the rat olfactory system. J. Chem.
Neuroanat. 5:51–62.
Bonfanti, L., and Theodosis, D. T. (1994). Expression of polysia-
REFERENCES lylated neural cell adhesion molecule by proliferaing cells in
the subependymal layer of the adult rat, in its rostral extension
Acampora, D., Mazan, S., Lallemand, Y., Avantaggiato, V., and in the olfactory bulb. Neuroscience 62:291–305.
Maury, M., Simeone, A., and Brulet, P. (1995). Forebrain and Bonthius, D. J., and West, J. R. (1991). Acute and long-term neuronal
midbrain regions are deleted in Otx2 -/- mutants due to defect- deficits in the rat olfactory bulb following alcohol exposure dur-
ive anterior neuroectoderm specification during gastrulation. ing the brain growth spurt. Neurotoxicol. Teratol. 13: 611–619.
Development 121:3279–3290. Bossy, J. (1980). Development of olfactory and related structures.
Allen, W. N., and Akeson, R. (1985a). Identification of an olfac- Anat. Embryol. 161:225–236.
tory receptor neuron subclass: cellular and molecular analysis Bruch, R. D., and Carr, V. M. (1991). Rat olfactory neurons
during development. Dev. Biol. 109:393–401. express a 200 kDa neurofilament. Brain Res. 550:133–136.
Allen, W. K., and Akeson, R. (1985b). Identification of a cell sur- Brunet, L. J., Gold, G. H., and Ngai, J. (1996). General anosmia
face glycoprotein family of olfactory receptor neurons with a caused by a targeted disruption of the mouse olfactory cyclic
monoclonal antibody. J. Neurosci. 5:284–296. nucleotide-gated cation channel. Neuron 17:681–693.
Altman, J. (1969). Autoradiographic and histological studies of Brunjes, P. C., and Frazier, L. L. (1986). Maturation and plasticity
postnatal neurogenesis. IV. Cell proliferation and migration in in the olfactory system of vertebrates. Brain Res. Rev. 11:1–45.
the anterior forebrain, with special reference to persisting Buffone, A., Wang, F., Hevner, R., Anderson, S., Cutforth, T.,
neurogenesis in the olfactory bulb. J. Comp. Neurol. Chen, S., Meneses, J., Pedersen, R., Axel, R., and Rubenstein,
137:433–458. J. L. R. (1998). An olfactory sensory map develops in the
Astic, L., and Saucier, D. (1986). Anatomical mapping of the absence of normal projjection neurons or GABAergic
neuroepithelial projection to the olfactory bulb in the rat. interneurons. Neuron 21:1273–1282).
Brain Res. Bull. 16:445–454. Buiakova, O. I., Baker, H., Scott, J. W., Farbman, A., Kream, R.,
Astic, L., Saucier, D., and Holley, A. (1987). Topographical rela- Grillo, M., Franzen, L., Richman, M., Davis, L. M.,
tionships between olfactory receptor cells and glomerular foci Abbondanzo, S., Stewart, C. L., and Margolis, F. L. (1996).
in the rat olfactory bulb. Brain Res. 424:144–152. Olfactory marker protein (OMP) gene deletion causes altered
Astic, L., LePendu, J., Mollicone, R., Saucier, D., and Oriol, R. physiological activity of olfactory sensory neurons. Proc.
(1989). Cellular expression of H and B antigens in the rat Natl. Acad. Sci. USA 93:9858–9863.
olfactory system during development. J. Comp. Neurol. Burr H. S. (1916). The effects of the removal of the nasal pits in
289:386–394. Amblystoma embryos. J. Exp. Zool. 20:27–57.
Bailey, M. S., and Shipley, M. T. (1993). Astrocyte subtypes in Caggiano, M., Kauer, J. S., and Hunter, D. D. (1994). Globose
the rat olfactory bulb: morphological heterogeneity and dif- basal cells are neuronal progenitors in the olfactory epithe-
ferential laminar distribution. J. Comp. Neurol. 328:501–526. lium: a lineage analysis using a replication-incompetent retro-
Bailey, M. S., Puche, A. C., and Shipley, M. T. (1999). virus. Neuron 13:339–352.
Development of the olfactory bulb: evidence for glia-neuron Calof, A. L., and Chikaraishi, D. M. (1989). Analysis of neurogen-
interactions in glomerular formation. J. Comp. Neurol. esis in a mammalian neuroepithelium: proliferation and differen-
415:423–448. tiation of an olfactory neuron precursor in vitro. Neuron 3:115–127.
132 Chuah et al.

Carr, V. M., and Farbman, A. I. (1992). Ablation of the olfactory eration and structural changes in ensheathing cells. Brain Res.
bulb up-regulates the rate of neurogenesis and induces preco- 857:265–274.
cious cell death in olfactory epithelium. Exp. Neurol. Costanzo R. M. (1991). Regeneration of olfactory receptor cells.
115:55–59. Ciba Found. Symp. 160:233–242.
Carr, V. M., Farbman, A. I., Lidow, M. S., Colletti, L. M., and Costanzo, R. M., and Graziadei, P. P. C. (1983). A quantitative
Morgan, J. I. (1989). Developmental expression of reactivity analysis of changes in the olfactory epithelium following bul-
to monoclonal antibodies generated against olfactory epithe- bectomy in hamster. J. Comp. Neurol. 215:370–381.
lia. J. Neurosci. 9:1179–1198. Coughlin, M. D., Dibner, M. D., Boyer, D. M., and Black, I. B.
Carr, V. M., Farbman, A. I., Colletti, L. M., and Morgan, J. 1. (1978). Factors regulating development of an embryonic
(1991). Identification of a new non-neuronal cell type in rat mouse sympathetic ganglion. Dev. Biol. 66:513–528.
olfactory epithelium. Neuroscience 45:433–449. Couly, G. F., and Le Douarin, N. M. (1985). Mapping of the early
Carr, V. M., Walters, E., Margolis, F. L. and Farbman, A. I. neural primordium in quail-chick chimeras. I. Developmental
(1998). An enhanced olfactory marker protein immunoreac- relationships between placodes, facial ectoderm, and prosen-
tivity in individual olfactory receptor neurons following olfac- cephalon. Dev Biol. 110:422–439.
tory bulbectomy may be related to increased neurogenesis. Cuschieri, A., and Bannister, L. H. (1975a). The development of
J. Neurobiol. 34:377–390. the olfactory mucosa in the mouse: light microscopy. J. Anat.
Cau, E., Gradwohl, G., Fode, C., and Guillemot, F. (1997). 119:277–286.
Mash1 activates a cascade of bHLH regulators in olfactory Cuschieri, A., and Bannister, L. H. (1975b). The development of
neuron progenitors. Development 124:1611–1621. the olfactory mucosa in the mouse: electron microscopy.
Cau, E., Gradwohl, G., Casarosa, S., Kageyama, R., and J. Anat. 119:471–498.
Guillemot, F. (2000). Hes genes regulate sequential stages of DeHamer, M. K., Guevara, J. L., Hannon, K., Olwin, B. B., and
neurogenesis in the olfactory epithelium. Development Calof, A. L. (1994). Genesis of olfactory receptor neurons in
127:2323–2332. vitro: regulation of progenitor cell divisions by fibroblast
Cecchi, C., Mallamaci, A., and Boncinelli, E. (1999). Mouse growth factors. Neuron 13:1083–1097.
forebrain development. The role of Emx2 homeobox gene. Dellovade, T. L., Pfaff, D. W., Schwanzel-Fukuda, M. (1998)
C. R. Acad. Sci. III 322:837–842. Olfactory bulb development is altered in small-eye (Sey)
Chiang, C., Litingtung, Y., Lee, E., Young, K. E., Corden, J. L., mice. J. Comp. Neurol. 402:402–418.
Westphal, H., Beachy, P. A. (1996). Cyclopia and defective Doucette, R. (1989). Development of the nerve fiber layer in the
axial patterning in mice lacking Sonic hedgehog gene func- olfactory bulb of mouse embryos. J. Comp. Neurol.
tion. Nature 383:407–413. 285:514–527.
Chuah, M. I., and Au, C. (1991a). Olfactory Schwann cells are Doucette, R. (1990). Glial influences on axonal growth in the pri-
derived from precursor cells in the olfactory epithelium. mary olfactory system. Glia 3:433–449.
J. Neurosci. Res. 29:172–180. Drapkin, P. T., and Silverman, A. J. (1999). Development of the
Chuah, M. I., and Au, C. (1991b). Localization of N-CAM and chick olfactory nerve. Dev Dyn. 214:349–360.
N-cadherin in the fetal human primary olfactory pathway. Dynes, J. L., and Ngai, J. (1998). Pathfinding of olfactory neuron
Soc. Neurosci. Abstr. 17:634. axons to stereotyped glomerular targets revealed by dynamic
Chuah, M. I., and Au, C. (1992). Neural cell adhesion molecules imaging in living zebrafish embryos. Neuron 20:1081–1091.
are present in the fetal human primary olfactory pathway. Dev. Escurat, M., Djabali, K., Gumpel, M., Gros, F., and Portier, M. M.
Neurosci. 14:357–361. (1990). Differential expression of two neuronal intermediate-
Chuah, M. I., and Farbman, A. I. (1983). Olfactory bulb increases filament proteins, peripherin and the low-molecular-mass
marker protein in olfactory receptor cells. J. Neurosci. neurofilament protein (NF-L), during the development of the
3:2197–2205. rat. J. Neurosci. 10:764–784.
Chuah, M. I., and Zheng, D. R. (1987). Olfactory marker protein Farbman, A. I. (1977). Differentiation of olfactory receptor cells
is present in olfactory receptor cells of human fetuses. in organ culture. Anat. Rec. 189:187–200.
Neuroscience 23:363–370. Farbman, A. I. (1986). Prenatal development of mammalian
Chuah, M. L., and Zheng, D. R. (1992). The human primary olfactory receptor cells. Chem. Senses 11:3–18.
olfactory pathway: fine structural and cytochemical aspects Farbman, A. I. (1991). Developmental neurobiology of the olfac-
during development and in adults. Microsc. Res. Tech. tory system. In Smell and Taste in Health and Disease,
23:76–85. T. V. Getchell, R. L. Doty, L. M. Bartoshuk, and J. B. Snow
Chuah, M. I., Farbman, A. I., and Menco, B. P. M. (1985). (Eds.). Raven Press, New York, pp. 19–33.
Influence of olfactory bulb on dendritic knob density of rat Farbman, A. I., and Buchholz, J. A. (1996). Transforming
olfactory receptor neurons in vitro. Brain Res. 338:259–266. growth factor-alpha and other growth factors stimulate cell
Chuah, M. I., David, S., and Blaschuk, O. (1991). Differentiation division in olfactory epithelium in vitro. J. Neurobiol.
and survival of rat olfactory epithelial neurons in dissociated 30:267–280.
cell culture. Dev. Brain Res. 60:123–132. Farbman, A. I., and Gesteland, R. C. (1975). Development and
Chuah, M. I., Cossins, J.-M., Woodhall, E., Tennent, R., Nash, G., electrophysiological studies of olfactory mucosa in organ cul-
and West, A. K. (2000). Glial growth factor 2 induces prolif- ture. In International Symposium on Olfaction and Taste,
Developmental Anatomy of the Olfactory System 133

D. A. Denton and J. P. Coghlan (Eds.). Academic Press, New Gonzales-Estrada, M. T., and Freeman, W. J. (1980). Effects of
York, pp. 107–110. carnosine on olfactory bulb EEG, evoked potentials and D. C.
Farbman, A. I., and Margolis, F. L. (1980). Olfactory marker pro- potentials. Brain Res. 202:373–386.
tein during ontogeny: immunohistochemical localization. Gordon, M. K., Mumm, J. S., Davis, R. A., Holcomb, J. D., and
Dev. Biol. 74:205–215. Calof, A. L. (1995). Dynamics of MASH1 expression in vitro
Farbman, A. I., and Squinto, L. M. (1985). Early development of and in vivo suggest a non-stem cell site of MASH1 action in
olfactory receptor cell axons. Dev. Brain Res. 19:205–213. the olfactory receptor neuron lineage. Mol. Cell Neurosci.
Farbman, A. I., Brunjes, P. C., Rentfro, L., Michas, J., and Ritz, S. 6:363–379.
(1988). The effect of unilateral naris closure on cell dynamics in Gorham, D., Ziff, E. B., and Baker, H. (1991). Differential spa-
developing rat olfactory epithelium. J. Neurosci. 8:3290–3295. tial and temporal expression of two type III intermediate fila-
Franco, B., Guioli, S., Pragliola, A., Incerti, B., Bardoni, B., ment proteins in olfactory receptor neurons. Neuron
Tonlorenzi, R., Carrozzo, R., Maestrini, E., Pieretti, M., 7:485–497.
Taillon-Miller, P., et al. (1991). A gene deleted in Kallmann’s Graziadei, P. P. C., and Kaplan, M. (1980). Regrowth of olfac-
syndrome shares homology with neural cell adhesion and tory sensory axons into transplanted neural tissue. 1.
axonal path-finding molecules. Nature 353:529–536. Development of connections with the occipital cortex. Brain
Friedrich, R. W., and Korsching, S. I. (1997). Combinatorial and Res. 201:39–44.
chemotopic odorant coding in the zebrafish olfactory bulb Graziadei, P. P. C., and Monti Graziadei, G. A. (1978).
visualized by optical imaging. Neuron 18:737–752. Continuous nerve cell renewal in the olfactory system. In
Frosch, M. P., and Dichter, M. A. (1984). Physiology and phar- Handbook of Sensory Physiology, Vol. IX, M. Jacobson (Ed.).
macology of olfactory bulb neurons in dissociated cell cul- Springer-Verlag, Berlin, pp. 55–82.
ture. Brain Res. 290:321–332. Graziadei, P. P. C., and Monti Graziadei, G. M. (1979).
Gesteland, R. C., Yancey, R. A., and Farbman, A. I. (1982). Neurogenesis and neuron regeneration in the olfactory system
Development of olfactory receptor neuron selectivity in the of mammals. I. Morphological aspects of differentiation and
rat fetus. Neuroscience 7:3127–3136. structural organization of the olfactory sensory neurons.
Getchell, T. V., Narla, R. K., Little, S., Hyde, J. F., and J. Neurocytol. 8:1–18.
Getchell, M. L. (2000). Horizontal basal cell proliferation in Graziadei, P. P. C., and Monti Graziadei, G. M. (1986) Principles
the olfactory epithelium of transforming growth factor-alpha of organization of the vertebrate olfactory glomerulus: an
transgenic mice. Cell Tissue Res. 299:185–192. hypothesis. Neuroscience 19:1025–1035.
Golden, J. A., Bracilovic, A., McFadden, K. A., Beesley, J. S., Graziadei, P. P. C., and Samanen, D. W. (1980). Ectopic glomeru-
Rubenstein, J. L., and Grinspan, J. B. (1999). Ectopic bone lar structures in the olfactory bulb of neonatal and adult mice.
morphogenetic proteins 5 and 4 in the chicken forebrain lead Brain Res. 187:467–472.
to cyclopia and holoprosencephaly. Proc. Natl. Acad. Sci. Graziadei, P. P. C., Levine, R. R., and Monti Graziadei, G. A.
USA 96:2439–2444. (1979). Plasticity of connections of the olfactory sensory
Goldstein, B. J., and Schwob, J. E. (1996). Analysis of the glo- neuron: regeneration into the forebrain following
bose basal cell compartment in rat olfactory epithelium using bulbectomy in the neonatal mouse. Neuroscience 4: 713–727.
GBC-1, a new monoclonal antibody against globose basal Guillemot, F., Lo, L. C., Johnson, J. E., Auerbach, A.,
cells. J. Neurosci. 16:4005–4016. Anderson, D. J., and Joyner, A. L. (1993). Mammalian
Goldstein, B. J., Wolozin, B. L., and Schwob, J. E. (1997). FGF2 achaete-scute homolog 1 is required for the early development
suppresses neuronogenesis by a cell line derived from rat of olfactory and autonomic neurons. Cell 75:463–476.
olfactory epithelium. J. Neurobiol. 33:411–428. Gustafson, A.-L., Eriksson, U. and Dencker, L. (1999). CRBP I
Gong, Q., and Farbman, A. I. (1999) Sonic hedgehog expression and CRABP I localisation during olfactory nerve develop-
in the glomeruli during rat olfactory system development. ment. Dev. Brain Res. 114:121–126.
Chem. Senses 214:574–575. Halász, N. and Greer, C. A. (1993). Terminal arborizations of
Gong, Q., and Shipley, M. T. (1995). Evidence that pioneer olfactory nerve fibers in the glomeruli of the olfactory bulb.
olfactory axons regulate telencephalon cell cycle kinetics to J. Comp. Neurol. 337:307–316.
induce the formation of the olfactory bulb. Neuron Hallonet, M., Hollemann, T., Wehr, R., Jenkins, N. A., Copeland,
14:91–101. N. G., Pieler, T., and Gruss, P. (1998). Vax1 is a novel homeobox-
Gong, Q., and Shipley, M. (1996). Expression of extracellular containing gene expressed in the developing anterior ventral
matrix molecules and cell surface molecules in the olfactory forebrain. Development 125:2599–2610.
nerve pathway during early development. J. Comp. Neurol. Hatini, V., Ye, X., Balas, G., and Lai, E. (1999). Dynamics of pla-
366:1–14. codal lineage development revealed by targeted transgene
Gonzales, F., Farbman, A. I., and Gesteland, R. C. (1985). Cell expression. Dev. Dyn. 215:332–343.
and explant culture of olfactory chemoreceptor cells. Hempstead, J. L., and Morgan, J. I. (1983). Monoclonal anti-
J. Neurosci. Methods 14:77–90. bodies to the rat olfactory sustentacular cell. Brain Res.
González, M. de L., Malemud, C. J., and Silver, J. (1993). Role 288:289–295.
of astroglial extracellular matrix in the formation of rat olfac- Hill, R. E., Favor, J., Hogan, B. L., Ton, C. C., Saunders, G. F.,
tory bulb glomeruli. Exp. Neurol. 123:91–105. Hanson, I. M., Prosser, J., Jordan, T., Hastie, N. D., and van
134 Chuah et al.

Heyningen, V. (1991). Mouse small eye results from muta- Johnson, E. W., Eller, P. M., and Jafek, B. W. (1995). Distribution
tions in a paired-like homeobox-containing gene. Nature of OMP-, PGP 9.5- and CaBP-like immunoreactive chemo-
354:522–525. receptor neurons in the developing human olfactory epithelium.
Hinds, J. W. (1968a). Autoradiographic study of histogenesis in Anat. Embryol. 191:311–317.
the mouse olfactory bulb. I. Time of origin of neurons and Kallmann, F. J., Schoenfeld, W. A., and Barrera, S. E. (1944). The
neuroglia. J. Comp. Neurol. 134:287–304. genetic aspects of primary eunuchoidism. Am. J. Ment. Defic.
Hinds, J. W. (1968b). Autoradiographic study of histogenesis in 48:203–236.
the mouse olfactory bulb. II. Cell proliferation and migration. Kasowski, H. J., Kim, H., and Greer, C. A. (1999).
J. Comp. Neurol. 134:305–322. Compartmental organization of the olfactory bulb glomerulus.
Hinds, J. W. (1972a). Early neuron differentiation in the mouse of J. Comp. Neurol. 407:261–274.
olfactory bulb. I. Light microscopy. J. Comp. Neurol. Kawano, T., and Margolis, F. L. (1982). Transsynaptic regulation
146:233–252. of olfactory bulb catecholamine in mice and rats.
Hinds, J. W. (1972b). Early neuron differentiation in the mouse J. Neurochem. 39:342–348.
olfactory bulb. II. Electron microscopy. J. Comp. Neurol. Kim, H., and Greer, C. A. (2000). The emergence of compart-
146:253–276. mental organization in olfactory bulb glomerulil during post-
Hinds, J. W., and Hinds, P. L. (1976). Synapse formation in the natal development. J. Comp. Neurol. 422:297–311.
mouse olfactory bulb. I. Quantitative studies. J. Comp. Klenoff, J. R., and Greer, C. A. (1998). Postnatal development of
Neurol. 169:15–40. olfactory receptor cell axonal arbors. J. Neurosci.
Hinds, J. W., Hinds, P. L. and McNelly, N. A. (1984). An auto- 390:256–267.
radiographic study of the mouse olfactory epithelium: evidence Kream, R. M., Davis, B. J., Kawano, T., Margolis, F. L., and
for long-lived receptors. Anat. Rec. 210:375–383. Macrides, F. (1984). Substance P and catecholaminergic
Hirsch, J. D., and Margolis, F. L. (1979). L-[3H]Carnosine bind- expression in neurons of the hamster main olfactory bulb.
ing in the olfactory bulb. II. Biochemical and biological stud- J. Comp. Neurol. 222:140–154.
ies. Brain Res. 174:81–94. Kubick, S., Strotmann, J., Andreini, I., and Breer, H. (1997).
Hirth, F., Therianos, S., Loop, T., Gehring, W. J., Reichert, H., Subfamily of olfactory receptors characterized by unique
and Furukubo-Tokunaga, K. (1995). Developmental defects structural features and expression patterns. J. Neurochem.
in brain segmentation caused by mutations of the homeobox- 69:465–475.
genes orthodenticle and empty spiracles in Drosophila. Legouis, R., Hardelin, J. P., Levilliers, J., Claverie, J. M.,
Neuron 15:769–778. Compain, S., Wunderle, V., Millasseau, P., Le Paslier, D.,
Hogan, B. L., Horsburgh, G., Cohen, J., Hetherington, C. M., Cohen, D., Caterina, D., et al. (1991). The candidate gene for
Fisher, G., and Lyon, M. F. (1986). Small eyes (Sey): a the X-linked Kallmann syndrome encodes a protein related to
homozygous lethal mutation on chromosome 2 which affects adhesion molecules. Cell 67:423–435.
the differentiation of both lens and nasal placodes in the Legouis, R., Lievre, C. A., Leibovici, M., Lapointe, F., and Petit, C.
mouse. J. embryol. Exp. Morphol. 97:95–110. (1993). Expression of the KAL gene in multiple neuronal sites
Holbrook, E. H., Mieleszko Szumowski, K. E., and Schwob, J. E. during chicken development. Proc. Natl. Acad. Sci. USA
(1995). Characterization of the horizontal basal cells of the 90:2461–2465.
olfactory epithelium. J. Comp. Neurol. 363:129–146. Leopold, D. A., Hornung, D. E., and Schwob, J. E. (1992).
Hu, H. Y., Tomasiewicz, H., Magnuson, T., and Rutishauser, U. Congenital lack of olfactory ability. Ann. Otol. Rhinol.
(1996). The role of polysialic acid in migration of olfactory Laryngol. 101:229–236.
bulb interneuron precursors in the subventricular zone. Liesi, P. (1985). Laminin-immunoreactive glia distinguish regen-
Neuron 16:735–743. erative adult CNS systems from non-regenerative ones.
Hui, C. C., and Joyner, A. L. (1993). A mouse model of greig EMBO J. 4:2505–2511.
cephalopolysyndactyly syndrome: the extra-toes mutation Lin, D. M., and Ngai, J. (1999). Development of the vertebrate
contains an intragenic deletion of the Gli3 gene. Nat. Genet. main olfactory system. Curr. Opin. Neurobiol. 9:74–78.
3:241–246. Lin, D. M., Wang, F., Lowe, G., Gold, G. H., Axel, R., Ngai, J.,
Humphrey, T. (1940). The development of the olfactory and the and Brunet, L. (2000). Formation of precise connections in
accessory olfactory formations in human embryos and fetuses. the olfactory bulb occurs in the absence of odorant-evoked
J. Comp. Neurol. 73:431–468. neuronal activity. Neuron 26:69–80.
Humphrey, T., and Crosby, E. C. (1938). The human olfactory Liu, K. L., Chuah, M. I., and Lee, K. K. H. (1995). Soluble fac-
bulb. Univ. (of Mich.) Hosp. Bull. 4:61–62. tors from the olfactory bulb attract olfactory schwann cells.
Jacobson, A. G. (1963). The determination and positioning of the J. Neurosci. 15:990–1000.
nose, lens and ear. I. Interaction within the ectoderm, and Lois, C., Garcia-Verdugo, J. M., Alvarez-Buylla, A. (1996). Chain
between the ectoderm and underlying tissues. J. Exp. Zool. migration of neuronal precursors. Science 271: 978–981.
154:273–284. Loo, A. T., Youngentob, S. L., Kent, P. F., and Schwob, J. E.
Johnson, D. R. (1967). Extra-toes: a new mutant gene causing (1996). The aging olfactory epithelium: neurogenesis,
multiple abnormalities in the mouse. J. Embryol. Exp. response to damage and odorant-induced activity. Int. J.
Morphol. 17:543–581. Develop. Neurosci. 14:881–900.
Developmental Anatomy of the Olfactory System 135

Luskin, M. B. (1993). Restricted proliferation and migration of mary and secondary cilia of olfactory and respiratory epithe-
postnatally generated neurons derived from the forebrain sub- lial cells, and on olfactory axons. Anat. Embryol.
ventricular zone. Neuron 11:173–189. 178:309–326.
Macdonald, R., Barth, K. A., Xu, Q., Holder, N., Mikkola, I., and Menco, B. P. M., and Farbman, A. 1. (1985a). Genesis of cilia
Wilson, S. W. (1995). Midline signaling is required for Pax and microvilli of rat nasal epithelia during pre-natal develop-
gene regulation and patterning of the eyes. Development ment. 1. Olfactory epithelium, qualitative studies. J. Cell Sci.
121:3267–3278. 78:283–310.
Mackay-Sim, A., and Chuah, M. I. (2000). Neurotrophic factors in Menco, B. P. M., and Farbman, A. 1. (1985b). Genesis of cilia
the primary olfactory pathway. Prog. Neurobiol. 62:527–559. and microvilli of rat nasal epithelia during pre-natal develop-
Mackay-Sim, A., and Kittel, P. (1991). Cell dynamics in the adult ment. II. Olfactory epithelium, a morphometric analysis.
mouse olfactory epithelium: a quantitative autoradiograph J. Cell Sci. 78:311–336.
study. J. Neurosci. 11:979–984. Mendoza, A. A., Breipohl, W., and Miragall, F. (1982). Cell
MacLeod, N. K., and Straughan, D. W. (1979). Responses of migration from the chick olfactory placode: a light and elec-
olfactory bulb neurons to the dipeptide carnosine. Exp. Brain tron microscopic study. J. Embryol. Expr. Morphol. 69:47–59.
Res. 34:183–189. Miragall, F., Kadmon, G., Husmann, M., and Schachner, M.
Magrassi, L., and Graziadei, P. P. G. (1985). Interaction of the (1988). Expression of cell adhesion molecules in the olfac-
transplanted olfactory placode with the optic stalk and the tory system of the adult mouse: presence of the embryonic
diencephalon in Xenopus laevis embryos. Neuroscience form of N-CAM. Dev. Biol. 129:516–531.
15:903–921. Miragall, F., Kadmon, G., and Schachner, M. (1989).
Mahanthappa, N. K., and Schwarting, G. A. (1993). Peptide Expression of L1 and N-CAM cell adhesion molecules dur-
growth factor control of olfactory neurogenesis and neuron ing development of the mouse olfactory system. Dev. Biol.
survival in vitro: roles of EGF and TGF-betas. Neuron 135:272–286.
10:293–305. Mombaerts, P., Wang, F., Dulac, C., Chao, S., Nemes, A.,
Malun, D., and Brunjes, P. C. (1996). Development of olfactory Mendelsohn, M., Edmondson, J., and Axel, R. (1996).
glomeruli: temporal and spatial interactions between olfactory Visualizing an olfactory sensory map. Cell 87:675–686.
receptor axons and mitral cells in opossums and rats. Monti Graziadei, G. A. (1983). Experimental studies on the
J. Comp. Neurol. 368:1–16. olfactory marker protein. III. The olfactory marker protein in
Mania-Farnell, B., and Farbman, A. I. (1990). Immuno- the olfactory neuroepithelium lacking connections with the
histochemical localization of guanine nucleotide-binding forebrain. Brain Res. 262:303–308.
proteins in rat olfactory epithelium during development. Dev. Monti-Graziadei, A. G. (1992). Cell migration from the olfactory
Brain Res. 51:103–112. neuroepithelium of neonatal and adult rodents. Dev. Brain
Margolis, F. L. (1972). A brain protein unique to the olfactory Res. 70:65–74.
bulb. Proc. Natl. Acad. Sci. USA 69:1221–1224. Monti Graziadei, G. A., and Graziadei, P. P. C. (1983) The olfactory
Margolis, F. L. (1974). Carnosine in the primary olfactory path- organ: neural transplantation. In Neural Transplants, J. R. Sladek
way. Science 184:909–911. and D. M. Gash (Eds.). Plenum Press, New York, pp. 167–186.
Margolis, F. L. (1980). An olfactory neuropeptide. In Role of Monti Graziadei, G. A., and Graziadei, P. P. C. (1992). Sensory
Peptides in Neuronal Function, J. L. Baker and T. Smith reinnervation after partial removal of the olfactory bulb.
(Eds.). Marcel Dekker, New York, pp. 545–572. J. Comp. Neurol. 316:32–44.
Margolis, F. L. (1982). Olfactory marker protein (OMP). Scand. Monti Graziadei, G. A., Stanley, R. S., and Graziadei, P. P. C.
J. Immunol. 15(Suppl. 9):181–199. (1980). The olfactory marker protein in the olfactory system
Margolis, F. L., Grillo, M., Brown, C. E., Williams, T. H., Pitcher, of the mouse during development. Neuroscience
R. G., and Elgar, G. J. (1979). Enzymatic and immunological 5:1239–1252.
evidence for 2 forms of carnosinase in the mouse. Biochim. Moore, K. L. (1988). The Developing Human. W. B. Saunders,
Biophys. Acta 570:311–323. Philadelphia, pp. 190–196.
Margolis, F. L., Grillo, M., Kawano, T., and Farbman, A. I. Moran, D. T., Rowley, J. C., and Jafek, B. W. (1982a). The fine
(1985). Carnosine synthesis in olfactory tissue during onto- structure of olfactory mucosa in man. J. Neurocytol. 11:
geny: effect of exogenous b-alanine. J. Neurochem. 721–746.
44:1459–1464. Moran, D. T., Rowley, J. C., and Jafek, B. W. (1982b). Electron
Matsuzaki, O., Bakin, R. E., Cai, X., Menco, B. P. M., and microscopy of human olfactory epithelium reveals a new cell
Ronnett, G. V. (1999). Localization of the olfactory cyclic type: the microvillar cell. Brain Res. 253:39–46.
nucleotide-gated channel subunit 1 in normal, embryonic and Mori, K., Fujita, S. C., Imamura, K., and Obata, K. (1985).
regenerating olfactory epithelium. Neuroscience 94:131–140. Immunohistochemical study of subclasses of olfactory nerve
Meisami, E. (1989). A proposed relationship between increases in fibers and their projections to the olfactory bulb in the rabbit.
the number of olfactory receptor neurons, convergence ratio J. Comp. Neurol. 242:214–229.
and sensitivity in the developing rat. Dev. Brain Res. 46: 9–19. Mori, K., Nagao, H., and Yoshihara, Y. (1999). The olfactory
Menco, B. P. M. (1988). Pre-natal development of rat nasal bulb: coding and processing of odor molecule information.
epithelia. IV. Freeze-fracturing on apices, microvilli and pri- Science 286:711–715.
136 Chuah et al.

Morrison, E. E., and Graziadei, P. P. C. (1983). Transplants Pyatkina, G. A. (1982). Development of the olfactory epithelium
of olfactory mucosa in the rat brain. I. A light microscop- in man. Z. Mikrosk. Anat. Forsch. 96:361–372.
ic study of transplant organization. Brain Res. 279: Qiu, M., Bulfone, A., Martinez, S., Meneses, J. J., Shimamura,
241–245. K., Pedersen, R. A., and Rubenstein, J. L. (1995). Null muta-
Morshead, C. M., and Van der Kooy, D. (1992). Postmitotic death tion of Dlx-2 results in abnormal morphogenesis of proximal
is the fate of constitutively proliferating cells in the first and second branchial arch derivatives and abnormal dif-
subependymal layer of the adult mouse brain. J. Neurosci. ferentiation in the forebrain. Genes Dev. 9:2523–2538.
12:249–256. Ressler, K. J., Sullivan, S. L., and Buck, L. B. (1993). A zonal
Mulvaney, B. D., and Heist, H. E. (1971). Centriole migration organization of odorant receptor gene expression in the olfac-
during regeneration and normal development of olfactory tory epithelium. Cell 73:597–609.
epithelium. J. Ultrastruct. Res. 35:274–281. Ressler, K. J., Sullivan, S. L., and Buck, L. B. (1994).
Ohno, I., Ohyama, M., Hanamure, Y., and Ogawa, K. (1981). Information coding in the olfactory system: evidence for a
Comparative anatomy of olfactory epithelium. Biomed. Res., stereotyped and highly organized epitope map in the olfac-
(Suppl.) 2:455–458. tory bulb. Cell 79:1245–1255.
Oland, L. A., and Tolbert, L. P. (1996). Multiple factors shape Roelink, H. (1996). Tripartite signaling of pattern: interactions
development of olfactory glomeruli: insights from an insect between Hedgehogs, BMPs and Wnts in the control of verte-
model system. J. Neurobiol. 30:92–109. brate development. Curr. Opin. Neurobiol. 6:33–40.
Oland, L. A., and Tolbert, L. P. (1998). Glomerulus development Rosselli-Austin, L., and Altmann, J. (1979). The postnatal devel-
in the absence of a set of mitral-like neurons in the insect opment of the main olfactory bulb of the rat. J. Dev. Physiol.
olfactory lobe. J. Neurobiol. 36:41–52. 1:295–313.
Oland, L. A., Tolbert, L. P., and Mossman, K. L. (1988). Rosselli-Austin, L., and Yanai, J. (1989). Neuromorphological
Radiation-induced reduction of the glial population during changes in mouse olfactory bulb after neonatal exposure to
development disrupts the formation of olfactory glomeruli in phenobarbital. Neurotoxicol. Teratol. 11:227–230.
an insect. J. Neurosci. 8:353–367. Royal, S. J., and Key, B. (1999). Development of P2 olfactory
Oland, L. A., Orr, G., and Tolbert, L. P. (1990). Construction of a glomeruli in P2-internal ribosome entry site-tau-LacZ trans-
protoglomerular template by olfactory axons initiates the for- genic mice. J. Neurosci. 19:9856–9864.
mation of olfactory glomeruli in the insect brain. J. Neurosci. Rugarli, E. I. (1999). Kallmann syndrome and the link between
10:2096–2112. olfactory and reproductive development. Am. J. Human
Oliver, G., Mailhos, A., Wehr, R., Copeland, N. G., Jenkins, N. Genet. 65:943–948.
A., and Gruss, P. (1995). Six3, a murine homologue of the Rutishauser, U., and Jessell, M. (1988). Cell adhesion molecules
sine oculis gene, demarcates the most anterior border of the in vertebrate neural development. Physiol. Rev. 68:819–857.
developing neural plate and is expressed during eye develop- Saito, H., Mimmack, M., Kishimoto, J., Keverne, E. B., and
ment. Development 121:4045–4055. Emson, P. C. (1998). Expression of olfactory receptors, G-
Ophir, D., and Lancet, D. (1988). Expression of intermediate fila- proteins and AxCAMs during the development and matur-
ments and desmoplakin in vertebrate olfactory mucosa. ation of olfactory sensory neurons in the mouse. Dev. Brain
Anat. Rec. 221:754–760. Res. 110:69–81.
Oppenheim, R. W., Chuwang, I.-W., and Maderdrut, J. S. (1978). Sakai, M., Ashihara, M., Nishimura, T., and Nagatsu, I. (1990)
Cell death of motoneurons in chick embryo spinal cord. III. Caronosine-like immunoreactivity in human olfactory
The differentiation of motoneurons prior to their induced mucosa. Acta Otolaryngol. (Stockholm) 109:450–453.
degeneration following limb bud removal. J. Comp. Neurol. Salehi-Ashtiani, K., and Farbman, A. I. (1996). Expression of neu
177:87–112. and Neu differentiation factor in the olfactory mucosa of rat.
Persohn, E., and Schachner, M. (1987). Immunoelectron-micro- Int. J. Dev. Neurosci. 14:801–811.
scopic localization of the neural crest cell adhesion molecules Sasai, Y., and De Robertis, E. M. (1997). Ectodermal patterning
L1 and N-CAM during postnatal development of the mouse in vertebrate embryos. Dev. Biol. 182:5–20.
cerebellum. J. Cell Biol. 105:569–576. Sauer, F. W. (1937). Some factors in the morphogenesis of verte-
Pevsner, J., Ronnett, G., Hester, L. D., and Snyder, S. H. (1988). brate embryonic epithelia. J. Morph. 61:563–579.
Development of a method for primary rat olfactory neuron Schimmang, T., Lemaistre, M., Vortkamp, A., and Ruther, U.
culture. Soc. Neurosci. Abstr. 14:1062. (1992). Expression of the zinc finger gene Gli3 is affected in
Pilar, G., and Landmesser, L. (1976). Ultrastructural differences the morphogenetic mouse mutant extra-toes (Xt).
during embryonic cell death in normal and peripherally Development 116:799–804.
deprived ciliary ganglia. J. Cell Biol. 68:339–356. Schwanzel-Fukuda, M., and Pfaff, D. W. (1989). Origin of
Pixley, S. K., and Pun, R. Y. K. (1990). Cultured rat olfactory luteinizing hormone releasing hormone neurons. Nature
neurons are excitable and respond to odors. Dev. Brain Res. 338:161–164.
53:125–130. Schwanzel-Fukuda, M., and Silverman, A. J. (1980). The nervus
Puschel, A. W., Gruss, P., and Westerfield, M. (1992). Sequence terminalis of the guinea-pig; a new luteinizing hormone-
and expression pattern of pax-6 are highly conserved between releasing hormone (LHRH) neuronal system. J. Comp.
zebrafish and mice. Development 114:643–651. Neurol. 191:213–225.
Developmental Anatomy of the Olfactory System 137

Schwanzel-Fukuda, M., Bick, D., and Pfaff, D. W. (1989). Stickrod, G., Kimble, D. P., and Smotherman, W. P. (1982). In
Luteinizing hormone-releasing hormone (LHRH)- utero taste/odor aversion conditioning in the rat. Physiol.
expressing cells do not migrate normally in an inherited Behav. 28:5–7.
hypogonadal (Kallmann) syndrome. Mol. Brain Res. St. John, J. A., and Key, B. (1999). Expression of galectin-1 in
6:311–326. the olfactory nerve pathway of rat. Dev. Brain Res.
Schwartz Levey, M. A., Chikaraishi, D. M., and Kauer, J. S. 117:171–178.
(1991). Characterization of potential precursor populations in Stout, R. P., and Graziadei, P. P. C. (1980). Influence of the olfac-
the mouse olfactory epithelium using immunocytology and tory placode on the development of the brain in Xenopus laevis
autoradiography. J. Neurosci. 11:3556–3564. (Daudin). I. Axonal growth and connections of the transplanted
Schwob, J. E. (1992). The biochemistry of olfactory neurons: placode. Neuroscience 5:2175–2186.
Stages of differentiation and neuronal subsets. In Science of Strotmann, J., Wanner, I., Helfrich, T., and Breer, H. (1995).
Olfaction, M. L. Serby and K. L. Chodor (Eds.). Springer- Receptor expression in olfactory neurons during rat develop-
Verlag, New York, pp. 80–125. ment: in situ hybridization studies. Eur. J. Neurosci.
Schwob, J. E., and Gottlieb, D. I. (1986). The primary olfactory 7:492–500.
projection has two chemically distinct zones. J. Neurosci. Strotmann, J., Conzelmann, S., Beck, A., Feinstein, P., Breer, H.,
6:3393–3404. and Mombaerts, P. (2000). Local permutations in the
Schwob, J. E., and Gottlieb, D. I. (1988). Purification and char- glomerular array of the mouse olfactory bulb. J. Neurosci.
acterization of an antigen that is spatially segregated in the 20:6927–6938.
primary olfactory projection. J. Neurosci. 8:3470–3480. Sullivan, S. L., Bohm, S., Ressler, K. J., Horowitz, L. F., and
Schwob, J. E., Farber, N. B., and Gottlieb, D. I. (1986). Neurons Buck, L. B. (1995). Target-dependent pattern specification in
of the olfactory epithelium in adult rats contain vimentin. the olfactory epithelium. Neuron 15:779–789.
J. Neurosci. 6:208–217. Takahashi, S., Iwanaga, I., Takahashi, Y., Nakano, Y., and Fujita,
Schwob, J. E., Szumowski, K. E. M., and Stasky, A. A. (1992). T. (1984). Neuron-specific enolase, neurofilament protein
Olfactory sensory neurons are trophically dependent on the and S-100 protein in the olfactory mucosa of human fetuses:
olfactory bulb for their survival. J. Neurosci. 12:3896–3919. an immunohistochemical study. Cell Tissue Res.
Schwob, J. E., Mieleszko Szumowski, K. E., Leopold, D. A., and 238:231–234.
Emko, P (1993). Histopathology of olfactory mucosa in Talamo, B. R., Rudel, R., Kosik, K. S., Lee, V. M. Y., Neff, S.,
Kallmann’s syndrome. Ann. Otol. Rhin. Laryngol. 102:117–122. Adelman, L., and Kauer, J. S. (1989). Pathological changes in
Schwob, J. E., Youngentob, S. L., and Meiri, K. F. (1994). On the olfactory neurons in patients with Alzheimer’s disease. Nature
formation of neuromata in the primary olfactory projection. 337:736–739.
J. Comp. Neurol. 340:361–380. Tao, W., and Lai, E. (1992). Telencephalon-restricted expression
Schwob, J. E., Youngentob, S. L., and Mezza, R. C. (1995). The of bf-1, a new member of the HNF-3/fork head gene family,
reconstitution of the rat olfactory epithelium after methyl in the developing rat brain. Neuron 8:957–966.
bromide-induced lesion. J. Comp. Neurol. 359:15–37. Tennent, R., and Chuah, M. I. (1996). Ultrastructural study of
Seilheimer, B., and Schachner, M. (1988). Studies of adhesion ensheathing cells in early development of olfactory axons.
molecules mediating interactions between cells of peripheral Dev. Brain Res. 95:135–139.
nervous system indicate a major role for L1 in mediating Tisay, K. T., and Key, B. (1999). The extracellular matrix modu-
sensory neuron growth on Schwann cells. J. Cell Biol. lates olfactory neurite outgrowth on ensheathing cells. J.
107:341–351. Neurosci. 19:9890–9899.
Senba, E., Daddona, P. E., and Nagy, J. I. (1987). Adenosine Tonosaki, K., and Shibuya, T. (1979). Action of some drugs on
deaminase-containing neurons in the olfactory system of the gecko olfactory bulb mitral cell, responses to odor stimula-
rat during development. Brain Res. Bull. 18:635–648. tion. Brain Res. 16:180–184.
Shanmugalingam, S., Houart, C., Picker, A., Reifers, F., Treloar, H. B., Nurcombe, V., and Key, B. (1996). Expression of
Macdonald, R., Barth, A., Griffin, K., Brand, M., and Wilson, extracellular matrix molecules in the embryonic rat olfactory
S. W. (2000). Ace/Fgf8 is required for forebrain commissure pathway. J. Neurobiol. 31:41–55.
formation and patterning of the telencephalon. Development Treloar, H. B., Purcell, A. L., and Greer, C. A. (1999). Glomerular
127:2549–2561. formation in the developing rat olfactory bulb. J. Comp.
Simeone, A., Acampora, D., Gulisano, M., Stornaiuolo, A., and Neurol. 413:289–304.
Boncinelli, E. (1992). Nested expression domains of four Valverde, F., Santacana, M., and Heredia, M. (1992). Formation
homeobox genes in developing rostral brain. Nature of an olfactory glomerulus: morphological aspects of devel-
358:687–690. opment and organization. Neuroscieuce 49:255–275.
Smart, I. H. M. (1971). Location and orientation of mitotic fig- Vassar, R., Chaol, S. K., Sitcheran, R., Nunez, J. M., Vosshall, L.
ures in the developing mouse olfactory epithelium. J. Anat. B., and Axel, R. (1994) Topographic organization of sensory
109:243–251. projections to the olfactory bulb. Cell 79:981–991.
Stark, M. R., Biggs, J. J., and Schoenwolf, G. C., and Rao, M. S. Vawter, M. P., Basaric-Keys, J., Li, Y., Lester, D. S., Lebovics, R.
(2000). Characterization of avian frizzled genes in cranial pla- S., Lesch, K. P., Kulaga, H., Freed, W. J., Sunderland, T., and
code development. Mech. Dev. 93:195–200. Wolozin, B. (1996). Human olfactory neuroepithelial cells:
138 Chuah et al.

tyrosine phosphorylation and process extension are increased Wray, S., Grant, P., and Gainer, H. (1989). Evidence that
by the combination of IL-1beta, IL-6, NGF, and bFGF. Exp. cells expressing luteinizing hormone-releasing hormone
Neurol. 142:179–194. mRNA in the mouse are derived from progenitor cells in
Venneman, W., Van Nie, C. J., and Tibboel, D. (1982). the olfactory placode. Proc. Natl. Acad. Sci. USA
Developmental abnormalities of the olfactory bulb: a compar- 86:8132–8136.
ative study of the pig and chick embryo. Teratology 26:65–70. Yamagishi, M., Nakamura, H., Takahashi, S., Nakano, Y., and
Verhaagen, J., Oestreicher, A. B., Grillo, M., Khew-Goodall, Y.- Iwanaga, T. (1989). Olfactory receptor cells:
S., Gispen, W. H., and Margolis, F. L. (1990). Neuroplasticity Immunocytochemistry for nervous system-specific proteins
in the olfactory system: differential effects of central and and re-evaluation of their precursor cells. Arch. Histol. Cytol.
peripheral lesions of the primary olfactory pathway on the 52(suppl.):375–381.
expression of B-50/GAP43 and the olfactory marker protein. Yang, L., Zhang, H., Hu, G., Wang, H., Abate-Shen, C., and
J. Neurosci. Res. 26:31–44. Shen, M. M. (1998). An early phase of embryonic D1x5
Verwoerd, C. D., and van Oostrom, C. G. (1979). Cephalic neural expression defines the rostral boundary of the neural plate.
crest and placodes. Adv. Anat. Embryol. Cell Biol. 58:1–75. J. Neurosci. 18:8322–8330.
Vollrath, M., Altmannsberger, M., Weber, K., and Osborn, M. Yoshihara, Y., Kawasaki, M., Tamada, A., Fujita, H., Hayashi, H.,
(1985). An ultrastructural and immunohistological study of Kagamiyama, H., and Mori, K. (1997). OCAM: A new
the rat olfactory epithelium: unique properties of olfactory member of the neural cell adhesion molecule family related to
sensory cells. Differentiation 29:243–253. zone-to-zone projection of olfactory and vomeronasal axons.
Wang, F., Nemes, A., Mendelsohn, M., and Axel, R. (1998). J. Neurosci. 17:5830–5842.
Odorant receptors govern the formation of a precise topo- Youngentob, S. L., Schwob, J. E., Sheehe, P. R., and Markert, L. M.
graphic map. Cell 93:47–60. (1997). Odorant threshold following methyl bromide-induced
Weiler, E., and Farbman, A. I. (1997). Proliferation in the rat lesions of the olfactory epithelium. Physiol. Behav.
olfactory epithelium: age-dependent changes. J. Neurosci. 62:1241–1252.
17:3610–3622. Yousem, D. M., Turner, W. J., Li, C., Snyder, P. J., and Doty, R.
Weiler, E., and Farbman, A. I. (1999). Mitral cell loss following lat- L. (1993). Kallmann syndrome: MR evaluation of olfactory
eral olfactory tract transection increases proliferation density system. Am. J. Neuroradiol. 14:839–843.
in rat olfactory epithelium. Eur. J. Neurosci. 11:3265–3275. Zaraisky, A. G., Lukyanov, S. A., Vasiliev, O. L., Smirnov, Y. V.,
Whitesides III, J. G., and LaMantia, A.-S. (1996). Differential Belyavsky, A. V., and Kazanskaya, O. V. (1992). A novel
adhesion and the initial assembly of the mammalian olfactory homeobox gene expressed in the anterior neural plate of the
nerve. J. Comp. Neurol. 373:240–254. Xenopus embryo. Dev. Biol. 152:373–382.
Whitlock, K. E., and Westerfield, M. (1998). A transient popula- Zhao, H. and Reed, R. R. (2001). X inactivation of the OCNC1
tion of neurons pioneers the olfactory pathway in the channel gene reveals a role for activity-dependent competition
zebrafish. J. Neurosci. 18:8919–8927. in the olfactory system. Cell 104:651–660.
Woodhall, E., West, A. K., and Chuah, M. I. (2001). Cultured Zheng, C., Feinstein, P., Bozza, T., Rodriguez, I., and
olfactory ensheathing cells express nerve growth factor, brain- Mombaerts, P. (2000). Peripheral olfactory projections are
derived neurotrophic factor, ghia cell line-derived neurotro- differentially affected in mice deficient in a cyclic nucleotide-
phic factor and their receptors. Mol. Brain Res. 88:203–213. gated channel subunit. Neuron 26:81–91.
7

Anatomy and Neurochemistry of the Olfactory Bulb

Igor L. Kratskin
University of Pennsylvania, Philadelphia, Pennsylvania, U.S.A.
Ottorino Belluzzi
University of Ferrara, Ferrara, Italy

The olfactory system is able to recognize a great variety of cited due to space limitations; readers are referred to other
odorous substances and discriminate between chemicals reviews for this information (Halász and Shepherd, 1983;
that have subtle differences in their structural properties. Macrides and Davis, 1983; Brunjes and Frazier, 1986;
Sensory neurons of the olfactory epithelium perceive odor Mori, 1987; Scott and Harrison, 1987; Halász, 1990;
molecules, transduce and encode that information, and Trombley and Shepherd, 1993; Ressler et al., 1994a; Mori
transmit impulse responses to the primary olfactory center, and Yoshihara, 1995; Sullivan et al., 1995; Buck, 1996;
the main olfactory bulb. After processing in the olfactory Shipley and Ennis, 1996; Shepherd and Greer, 1998).
bulb, physiological signals are delivered directly to the sec-
ondary sensory centers in the primary olfactory cortex.
There is “ample evidence that the olfactory bulb is not I. ANATOMICAL ORGANIZATION
merely a ‘ganglion’ in which the olfactory pathway is
synaptically interrupted, but is indeed a centre of great The olfactory bulbs are paired, ovoid-shaped structures
complexity containing associative connections at several forming the rostral end of the telencephalon. In many
levels, intrinsic neuronal circuits of varying length, and a mammals, they occupy the foremost position in the skull
‘centrifugal’ as well as the sensory input” (Nieuwenhuys, and are quite large. In humans and other primates, the
1967). bulbs, displaced by the enlarged cerebrum, are relatively
This chapter focuses on the anatomical organization and small and located under the ventral surface of frontal
the neurotransmitters of the olfactory bulb, with an emphasis lobes. The olfactory bulb is a cortical structure and has a
on afferent projections from brain sources. An overview of characteristic laminar organization (Fig. 1). The first
neural and molecular mechanisms underlying odor coding anatomical descriptions of the olfactory bulb, made in the
in the olfactory bulb is also provided. Most of the informa- second half of the nineteenth century and summarized in
tion comes from rodents; human data are presented, when- the classic book of Ramón y Cajal (1911), were signifi-
ever possible. A number of original papers could not be cantly expanded by subsequent studies.
In the olfactory bulb, like in other brain centers, an
input fiber, a principal cell, and an intrinsic neuron form a
This chapter is dedicated to the memory of Dr. A. A. Bronstein
(1927–1976), an extraordinary olfactory scientist from the Sechenov triad of neuronal elements (Shepherd and Koch, 1998).
Institute of Evolutionary Physiology and Biochemistry, St. Petersburg, Two major sets of input fibers come to the bulb: axons of
Russia. olfactory sensory neurons, or olfactory axons, which trans-

139
140 Kratskin and Belluzzi

form the olfactory nerve layer. Within a single bundle,

olfactory axons are packed very tightly (5–20 nm from
one another), allowing ephaptic interactions between
neighboring axons (Eng and Kocsis, 1987). Olfactory
axons do not branch before entering the bulb, and their
number corresponds to the number of OSNs that, in
rabbits, was estimated to be around 50,000,000 on each
side of the nasal cavity (Allison and Warwick, 1949). In
adult humans, this number may be about 6000,000
(Moran et al., 1982). Unique glial cells, called olfactory
ensheathing cells, surround axon bundles on their way to
the bulb and within the olfactory nerve layer (Doucette,
1991). These cells have common features with astrocytes
and Schwann cells and express a series of neurotrophic
factors (Bartolomei and Greer, 2000; Mackay-Sim and
Chuah, 2000). Ensheathing cells may promote axonal
regeneration after traumatic injury (Bartolomei and
Greer, 2000).

2. Glomerular Layer
Axons of OSNs run from the olfactory nerve layer to
spherical neuropil regions termed olfactory glomeruli. In
different vertebrate species, the glomeruli vary from 30 to
200 m in diameter and constitute the glomerular layer,
one or two glomeruli thick. Each olfactory axon innervates
only a single glomerulus. After entering the glomerulus,
the olfactory axon gives rise to an arbor of branches (mean
branch length 170 m) with terminal boutons and en pas-
Figure 1 Transverse vibratome section of rat olfactory bulb
illustrating the basic laminar organization (staining with Giemsa sant varicosities; the branch arbor occupies about 14% of
dye). The olfactory nerve layer and glomeruli are stained due to the glomerular area (Halász and Greer, 1993; Klenoff and
anterograde labeling of olfactory axons with horseradish peroxi- Greer, 1998). Within glomeruli, olfactory axons make
dase. ONL, olfactory nerve layer; GL, glomerular layer; EPL, synapses onto dendrites of principal and intrinsic cells, so
external plexiform layer; MCL, mitral cell layer; IPL, internal each glomerulus is a complex structure consisting of
plexiform layer; GRL, granule cell layer. Scale bar-50 µm. axonal and dendritic compartments (Kosaka et al., 1997;
Kasowski et al., 1999).
A number of light and electron microscopic studies
mit information about odor molecules, and axons of brain
have described in eloquent detail neurons in the glomeru-
neurons, or centrifugal axons, which exert modulatory
lar layer (Pinching and Powell, 1971a; Macrides and
influences on bulbar microcircuits (Fig. 2A). There are two
Davis, 1983; Halász, 1990). Each glomerulus is surrounded
classes of principal cells in the olfactory bulb: mitral cells
by numerous small somata (long axis 6–8 m) of
and tufted cells (Fig. 2B). Intrinsic neurons, or local
periglomerular (PG) cells. A PG cell dendrite ramifies and
interneurons, fall into three categories: periglomerular
terminates within one or two glomeruli, intermingling with
cells, granule cells, and short axon cells (Fig. 2C).
terminals of olfactory axons and dendrites of principal
cells, whereas a PG cell axon extends across three to five
A. Layers and Neuron Types glomeruli. The population of PG cells comprises neurons
differing in their neurochemical, morphological, and physio-
1. Olfactory Nerve Layer
logical features (Kosaka et al., 1995, 1997; Puopolo and
Unmyelinated axons (mean diameter ~0.3 m) of olfac- Belluzzi, 1998a; Toida et al., 1998, 2000). About 10% of
tory sensory neurons (OSNs) constitute the primary PG cells, composing a chemically distinct neuronal group,
olfactory projection. The axons come to the olfactory have no synapses from olfactory axons (Kosaka et al.,
bulb in discrete bundles that interweave on its surface and 1997; Toida et al., 1998).
Anatomy and Neurochemistry of the Olfactory Bulb 141

sess spines, in contrast to smooth dendrites of tufted cells.

An external tufted cell (long axis 10–15 m) has a short
apical dendrite, terminating within a glomerulus, and one
to three basal dendrites extending immediately below
glomeruli. Some external tufted cells send axons to the
olfactory cortex, whereas others, whose axons terminate
within the bulb, represent intrinsic cells. Intrinsic tufted
cells either have connections in the glomerular layer or
engage in point-to-point reciprocal projections between
opposite regions of the olfactory bulb, forming a topo-
graphically organized “intrabulbar associational system”
(Schoenfeld et al., 1985).
The glomeruli are the most distinctive feature of the
olfactory bulb and illustrate “the principle of grouping
neural elements and synapses into anatomically defined
modules” (Shepherd and Greer, 1998). Glomeruli are not
functionally uniform and play a key role in odor coding in
the olfactory bulb. An example of glomerular specificity is
the so-called “modified glomerular complex,” which is a
group of glomeruli in the caudal dorsomedial part of the
bulb (Teicher et al., 1980; Greer et al., 1982). These
glomeruli have atypical structural features and likely
process information related to a specific odor cue impor-
tant for suckling behavior. Development of the glomeruli
depends upon influences exerted by OSNs (see Chapter 6).
There is a unique plasticity in the olfactory system:
although the OSNs are replaced by newly generated cells
throughout adult life (Graziadei and Monti Graziadei,
1978), a constancy of zonal projections from the olfactory
Figure 2 Schematic representation of afferent fibers, principal
cells, and local interneurons in the olfactory bulb. Layers of the epithelium to glomeruli is maintained (Schoenfield et al.,
bulb are indicated as in Figure 1. (A) ON(m) and ON(l), medial 1994). Extracellular matrix/neuronal cell adhesion mole-
and lateral groups of olfactory axons. Centrifugal fibers originate cules are one of the factors providing the basic guidance of
in the ipsilateral and contralateral anterior olfactory nucleus olfactory axons to the glomeruli (Kafitz and Greer, 1998).
(iAON and cAON), taenia tecta (TT), olfactory cortex (OC),
nucleus of the horizontal limb of the diagonal band (HDB), locus
coeruleus (LC), and raphe nucleus (Ra); pE, pars externa of the 3. External Plexiform Layer
AON; pM, pars medialis of the AON. (B) The axons (a), axon
collaterals, and dendrites (d) of a mitral cell (M), displaced mitral A very dense neuropil, formed by dendrites of principal
or internal tufted cell Md/Ti, middle tufted cell Tm, and external neurons and granule cells, and a relatively low density of
tufted cell Te; LOT, lateral olfactory tract. (C) GI, GII, and GIII cell bodies are characteristic features of the external plexi-
designate three types of granule cells; PG, periglomerular cell. form layer (EPL). Most neurons in the EPL are middle and
Various short axon cells are shown: SA(B), Blanes’ cell; SA(C), internal tufted cells, and some represent short axon cells.
Cajal’s cell; SA(G), Golgi cell; SA(H), Hensen’s cell; SA(S), Middle tufted cells have their somata (long axis 15–20
Schwann cell; SA(V), Van Gehuchten cell. (From Shepherd and m) near the middle of the EPL; each cell gives rise to thin
Greer, 1998.)
basal dendrites and an apical dendrite that terminates
within a single glomerulus. Axons of these cells give off
collaterals and project to the primary olfactory cortex.
PG cells are intermixed with external tufted cells and Internal tufted cells (long axis about 27 m), located in the
short axon cells. A short axon cell has an oval cell body deeper one third of the EPL, are similar in their morpho-
(long axis 12 m in rats), dendrites that ramify between or logy to displaced mitral cells. Importantly, it is within this
around glomeruli, and an axon extending to one to three layer that basal dendrites of principal cells have synaptic
glomeruli. Dendrites of PG cells and short axon cells pos- contacts with peripheral dendrites of granule cells.
142 Kratskin and Belluzzi

4. Mitral Cell Layer and Walters, 1982). Rounded or fusiform somata (long
axis 6–10 m) of granule cells are densely packed in the
The mitral cell layer is thin and contains relatively large
granule cell layer and constitute aggregates containing
somata (long axis 20–33 m) of mitral cells. These cells
three to five cells apiece. In these aggregates, gap junctions
have one apical dendrite (diameter 2–12 m; length
couple granule cells, so the activity of neighboring neurons
200–800 m) that runs through the EPL and terminates
may be synchronized (Reyher et al., 1991). Granule cells
within a single glomerulus. Each mitral cell also has two to
lack axons and have peripheral and deep dendritic
nine basal dendrites (diameter 1–8 m; length up to 1300
processes (Price and Powell, 1970a). Each cell gives rise to
m) that branch and terminate in the EPL, within a field
one relatively thick peripheral dendrite that ramifies and
with a radius of about 900 m. Two types of mitral cells
terminates in the EPL, extending over an area 50–200 m
have been identified (Macrides and Schneider, 1982; Mori
in diameter. One to four fine deep dendrites (length
et al., 1983; Orona et al., 1984). Type I mitral cells, which
50–100 m) terminate within the granule cell layer. All
are more numerous, have basal dendrites in the deep por-
parts of the granule cell, including a cell body, are
tion of the EPL, whereas type II, or displaced, mitral cells,
endowed with appendages called spines. Peripheral and
whose somata lie more superficially, have basal dendrites
basal dendrites have approximately two spines per 10 m
near the middle of the EPL. Myelinated axons of mitral
of dendritic length. Deep to the mitral cell layer, spines
cells (diameter 0.5–3.0 m) give off collaterals, which ter-
always have a postsynaptic location. In the EPL spines par-
minate deep within the bulb, and run to the secondary
ticipate in reciprocal synapses between granule and
olfactory centers. The mitral cell number has been esti-
mitral/tufted cells and are therefore presynaptic as well as
mated to be about 51,000 in persons in their mid-twenties
postsynaptic structures. These spines are also referred to as
(Bhatnagar et al., 1987).
gemmules to emphasize their special position in the synap-
The mitral and tufted cells exhibit multiple differences,
tic contacts (Rall et al., 1966; Price and Powell, 1970a).
including the position of cell bodies, distribution of basal
Three types of granule cells have been distinguished in
dendrites, and the transmitter specificity. These neurons
hamsters (Schneider and Macrides, 1978), rabbits (Mori
also have distinct genetic determinants of cell differenti-
and Kishi, 1982), and mice (Greer, 1987). Type I granule
ation (Greer and Shepherd, 1982). Moreover, the mitral
cells (GI in Fig. 2C) have intermediately positioned somata
and tufted cells are likely connected to particular groups of
and peripheral dendrites terminating at all levels of the
granule cells (see below) and have differing projection pat-
EPL. Type II and type III granule cells (GII and GIII in
tern in the olfactory cortex (Schoenfeld and Macrides,
Fig. 2C) have their cell bodies in the deep and superficial
1984; Schoenfeld et al., 1985; Scott et al., 1985). It thus
parts of the granule cell layer, respectively. Peripheral den-
appears that the two classes of principal cells are involved
drites of deep (GII) and superficial granule cells (GIII)
in segregated and overlapping circuits via intrinsic and
terminate, correspondingly, at the deep and superficial lev-
extrinsic connections (Macrides et al., 1985).
els of the EPL. Only deep and superficial granule cells
have been recognized in rats (Orona et al., 1983). These
5. Internal Plexiform Layer
data suggest that there is a segregation of local microcir-
The internal plexiform layer is immediately subjacent to cuits in the EPL, i.e., granule cells of different types are
the mitral cell layer and represents a thin neuropil zone connected to different types of principal cells. Type II and
containing a few short axon cells. The main elements com- type III granule cells are likely connected to mitral cells
posing the internal plexiform layer are numerous axons and tufted cells, respectively; type I granule cells may
and axon collaterals of mitral and tufted cells and receive signals from both classes of principal neurons.
peripheral dendrites of granule cells. These dendrites are Short axon cells are relatively numerous in the granule
the target for axons of external tufted cells that constitute cell layer. Several types of short axon cells (Fig. 2C) were
the intrabulbar associational system (Schoenfeld et al., identified by the Golgi impregnation technique (Ramón y
1985; Liu and Shipley, 1994). Some axons terminating in Cajal, 1911; Price and Powell, 1970d; Pinching and
the internal plexiform layer originate in neurons of the Powell, 1971a; Schneider and Macrides, 1978) and
basal forebrain and brainstem. immunostaining for calcium-binding protein parvalbumin
(Kosaka et al., 1994), and two of them, Golgi cells and
Blanes’ cells, had their somata in the granule cell layer.
6. Granule Cell Layer
These neurons are intermediate in size between granule
Granule cells are the most numerous neurons in the olfac- and mitral cells (about 15 m in the rat) and have three to
tory bulb, and their number in the rat is said to range from eight dendrites mainly restricted to the layer. In Blanes’
1000,000 to 3000,000 (Meisami and Safari, 1981; Struble cells, only distal parts of the dendrites possess spines; den-
Anatomy and Neurochemistry of the Olfactory Bulb 143

drites of Golgi cells are almost completely spineless. ventricles. Progenitor cells travel via the rostral migratory
Axons of Blanes’ cells and Golgi cells terminate within the stream into the bulb, invade the glomerular and granule cell
granule cell layer and, occasionally, in the internal plexi- layers, and become PG cells and granule cells. As shown in
form layer and in the deep of the EPL. mice, disconnection of the olfactory bulb from the rest of the
brain does not prevent proliferation and differentiation of
B. Convergence Ratios progenitor cells, but redirects them to the anterior olfactory
nucleus and frontal cortex (Jankovski et al., 1998). There is
In the rabbit, each olfactory bulb receives ~50,000,000 electrophysiological evidence that newly generated PG cells
olfactory axons and contains ~2000 glomeruli, 50,000 and granule cells in rats establish synaptic connections with
mitral cells, and 100,000 tufted cells (Allison and existing cells of the bulb neuronal network (O. Belluzzi and
Warwick, 1949). Simple calculations suggest that the con- J. LoTurco, manuscript in preparation). In mutant mice
vergence ratios for olfactory axons are very high in this exhibiting deficits in the migration of bulbar cell precursors,
species: 25,000 onto a single glomerulus, 1000 onto each the width of the granule cell layer was significantly reduced
mitral cell, and 500 onto each tufted cell. The number of and odor discrimination was impaired (Gheusi et al., 2000).
PG cells is approximately 20 times higher than that of The results of this study suggest that new granule cells are
mitral cells. Rough estimates suggest that, in rabbits, a sin- not simply added to the bulb, but replace dying neurons, and
gle glomerulus is composed of 25,000 branching olfactory that granule cell activity is important for olfactory discrimi-
axons and dividing dendrites of 25 mitral cells, 50 tufted nation. We can conclude that processes of cell death and cell
cells, and 500 PG cells. The ratio of granule cells to mitral generation coexist in the olfactory bulb, resulting in contin-
cells is ~50–100:1. In adult rats, the number of OSNs has ual remodeling of local synaptic connections.
been estimated at 20,000,000 (Paternostro and Meisami,
1996a,b), whereas the number of glomeruli is about 2400 D. Synaptic Connections
(Meisami and Sendera, 1993). Thus, the convergence of
olfactory axons onto glomeruli may be about 8000:1 in this
Signal analysis in the olfactory bulb is carried out at two
species. There are ~6000,000 sensory neurons (Moran
anatomical levels using intrinsic cells specific to each
et al., 1982) and ~8000 glomeruli (Meisami et al., 1998) in
level. Input processing occurs in the glomerular layer on
adult humans. This yields a convergence ratio of 750 olfac-
the basis of connections between olfactory axons, principal
tory axons per glomerulus, or one order of magnitude less
neurons, and PG cells. Output control results from interac-
than that in rats.
tions between principal neurons and granule cells in
the EPL. Centrifugal influences may modulate activity
C. Cell Death and Cell Generation
in local microcircuits at both anatomical levels. A number
of studies have investigated synaptic actions underlying
Cell death is inherent in the developing brain, and the
sensory processing in the olfactory bulb (reviewed by
olfactory bulb is not an exception to the rule. Studies in
Shipley and Ennis, 1996; Shepherd and Greer, 1998) (see
the rat indicated that cell death in the layers of mitral and
Chapter 9).
granule cells occurs within 10–15 postnatal days and
that early (on postnatal day one) external naris closure
prolongs this time period up to 60 days (Fiske and 1. Input Processing
Brunjes, 2001). This demonstrates how sensory input
may regulate cell numbers in the bulb. Compensatory Within glomeruli, terminals of olfactory axons containing
reorganization in the bulb following cell death has been round vesicles make Gray type 1 chemical synapses (with
investigated in the mutant mouse strain pcd (Purkinje asymmetric membrane thickenings) (see Gray, 1969) onto
cell degeneration), in which mitral, but not tufted, cells dendrites of mitral/tufted and PG cells (Pinching and
degenerate during early adulthood (Greer, 1987; Greer Powell, 1971a,b,c; White, 1973). The dendrites of principal
and Halász, 1987). In these mice, granule cells, dener- and PG cells are interconnected with one another by
vated from mitral cells, formed new dendrodendritic synapses of opposite directions. Principal cell dendrites
synapses with tufted cells. containing round synaptic vesicles make type 1 synapses,
The olfactory bulb is one of the few brain structures whereas dendrites of PG cells containing pleomorphic
receiving a supply of newly generated cells throughout adult vesicles establish Gray type 2 synapses (with symmetric
life (Altman, 1969; Luskin, 1993, 1998; Lois et al., 1996; membrane thickenings). Serial reconstructions show that
Doetsch et al., 1997). Neuronal precursors are generated 25–40% of these dendrodendritic synapses are arranged as
from stem cells in the subventricular zone lining the lateral reciprocal pairs. Intraglomerular connections form the
144 Kratskin and Belluzzi

synaptic triad, in which the olfactory axon (input element) units that mediate localized inhibition onto restricted
makes excitatory synapses onto principal cell dendrites groups of mitral/tufted cell basal dendrites (Woolf et al.,
(output elements) and dendrites of PG cells (intrinsic ele- 1991a,b). There are about 50–100 granule cells for one
ments). Within this triad, incoming impulses reach PG cells mitral cell, and each granule cell has at least 50 gemmules
directly from olfactory axons and indirectly via excitatory reciprocally connected to mitral/tufted cell dendrites. Such
synapses from principal cell dendrites. Intraglomerular numerous synaptic contacts provide powerful interactions
microcircuits function to process information about a given between principal and granule cells. While it has been gen-
odorant and to generate a pattern of activity associated with erally believed that only granule cells have reciprocal
the quality and intensity of a stimulus. synapse with principal cells, recent studies showed that
Neighboring glomeruli are connected mainly by axons dendrites of mitral/tufted cells in the EPL also participate
of PG cells that make type 2 synapses with apical dendrites in reciprocal synapses with processes of some short axon
of mitral/tufted cells and somata and dendrites of PG cells. cells (Toida et al., 1994, 1996).
In addition, PG cell axons synapse onto dendrites, somata, Synaptic connections in the EPL form the triad that
and initial axon segments of short axon cells. Axons of the includes the principal cell apical dendrite (input element),
latter cells containing pleomorphic synaptic vesicles estab- the soma and basal dendrites of the principal cell (output
lish type 2, presumably inhibitory, synapses with PG cells. elements), and the granule cell gemmule (intrinsic
Tufted cell axon collaterals that form type 1 contacts onto element). This microcircuit is the basis for output control in
tufted cell apical dendrites also provide connections the olfactory bulb. Different types of principal neurons con-
between glomeruli. Both inhibitory and excitatory actions nect to different types of granule cells and have differing
between glomeruli may occur (Shepherd and Greer, 1998). projection sites. This suggests that parallel pathways exist
Interglomerular microcircuits serve to recruit adjacent in the olfactory bulb to convey information about specific
glomeruli responsive to a given odorant and to enhance the characteristics of odor stimuli Dendrodendritic synapses
contrast between glomeruli processing information about provide the substrate for self-inhibition within each path-
dissimilar odors. Spatial clustering of glomeruli of a simi- way and lateral inhibition in parallel pathways.
lar specificity may be due to the lateral inhibition in the Granule cells in the EPL also participate in axodendritic
glomerular layer and inhibitory actions of granule cells in type 1 synapses made by centrifugal axon terminals that
the deeper layers of the bulb. contain round synaptic vesicles and establish synapses on
Intraglomerular and interglomerular microcircuits may granule cell gemmules (Price and Powell, 1970b,c).
be influenced by centrifugal inputs to the olfactory bulb. Granule cells are much more numerous than centrifugal
Centrifugal axons establish synapses with PG cell den- axons, but the terminals of such axons make multiple
drites, somata and dendrites of short axon cells, and occa- synapses onto several gemmules that are clustered around
sionally with dendrites of principal cells (Pinching and them (Price, 1968; Price and Powell, 1970b,c). Therefore,
Powell, 1972). Most of these contacts are restricted to each granule cell peripheral dendrite has at least one
extraglomerular neuropil, but axons from the raphe nuclei synaptic contact with a centrifugal axon (Fig. 3A), some-
terminate both outside and within glomeruli. times very close to a reciprocal synapse (Fig. 3B). In
addition, type 2 synapses, formed by terminals with
pleomorphic vesicles, are present on the shafts of granule
2. Output Control
cell peripheral dendrites, but are never observed on gem-
Contacts between basal dendrites of principal neurons and mules (Price and Powell, 1970b). Axons that make these
gemmules (spines) of granule cell peripheral dendrites are synapses may originate in GABAergic brain neurons.
the predominant type of synaptic connections in the EPL. A centrifugal axon terminal is an input element in the
About 80% of these contacts are organized as reciprocal synaptic triad that also involves a granule cell gemmule and
pairs, in which the synapse from principal to granule cell principal cell dendrite. Central modulation of intrabulbar pro-
is type 1, whereas the synapse from granule to principal cessing and output formation are mediated by granule cells
cell is type 2 (Rall et al., 1966; Price and Powell, 1970b; and may be realized exclusively through a dendrodendritic
Jackowski et al., 1978). Electrophysiological data provide synapse between principal and granule cells. This synapse is
strong evidence that the mitral/tufted-to-granule synapse is the site of a convergence of intrabulbar signals and inputs
excitatory and the granule-to-mitral/tufted synapse is from olfactory and nonolfactory brain structures. Therefore, it
inhibitory (see Mori, 1987). The latter synapse is the sole is characterized as a “multifunctional synapse” (Shepherd and
output of the granule cell. Modeling of information pro- Greer, 1998), and granule cells are considered the “final com-
cessing in granule cell dendritic spines suggests that sub- mon path” for both intrinsic signals and modulatory extrinsic
sets of spines may function as complex and independent influences (Price and Powell, 1970c).
Anatomy and Neurochemistry of the Olfactory Bulb 145

II. ODOR CODING IN THE OLFACTORY BULB

There is a body of evidence suggesting that olfactory

glomeruli are structural and functional units for odor cod-
ing in the olfactory bulb (Kauer et al., 1991; Shepherd,
1991; Kauer and Cinelli, 1993; Mori and Yoshihara, 1995;
Shepherd and Greer, 1998; Mori et al., 1999, 2000; Xu
et al., 2000). The idea that a glomerulus is a functional unit
and different glomeruli may receive information about dif-
ferent odors goes back to early anatomical (Clark and
Warwick, 1946; Allison and Warwick, 1949; Clark, 1951,
1957) and electrophysiological investigations (Adrian,
1950, 1951) of the olfactory system.
Further studies in the field have demonstrated that
(1) each glomerulus receives projections from OSNs
located in many regions of the olfactory epithelium, (2) a
Figure 3 Reconstruction of centrifugal axon terminals making single glomerulus is activated by different odorants, and
synaptic contacts onto granule cell dendritic gemmules (spines) (3) each odorant activates many glomeruli. Methods used
in the external plexiform layer. (A) One centrifugal axon termi- in these studies included horseradish peroxidase tracing
nal (c) establishes synapses with gemmules (g) of five granule (e.g., Jastreboff et al., 1984; Astic and Saucier, 1986) and
cell peripheral dendrites (p) a short distance from the reciprocal
monoclonal antibody immunolabeling of the primary
synapse (r) between the gemmule and mitral cell dendrite (m).
olfactory projection (Schwob and Gottlieb, 1986, 1988;
(B) Termination of a centrifugal fiber (c) on the granule cell gem-
mule, in close proximity to the reciprocal synapse. (From Price Carr et al., 1994; Mori and Yoshihara, 1995; Ring et al.,
and Powell, 1970c.) 1997; Nagao et al., 2000); behavioral testing (e.g., Slotnick
et al., 1987; Lu and Slotnick, 1999; Rubin and Katz, 2001);
analysis of odor-induced changes in c-fos expression
The entire surface of granule cells in the internal (Onoda, 1992; Guthrie et al., 1993; Sallaz and Jourdan,
plexiform and granule cell layers participate in contacts with 1993; Guthrie and Gall, 1995) and [14C]2-deoxyglucose
axons, which derive from both intrabulbar and central uptake in the olfactory bulb (Sharp et al., 1975; Stewart et
sources (Price and Powell, 1970b,c). One group of axon ter- al., 1979; Jourdan et al., 1980; Lancet et al., 1982; Royet
minals contains round vesicles and makes type 1 synapses et al., 1987; Sallaz and Jourdan, 1993; Johnson and Leon,
commonly placed on spines of the dendrites and perikarya of 1996, 2000a,b; Johnson et al., 1998, 1999); volatage-sen-
granule cells. Another group of terminals contains pleomor- sitive dye (Kauer, 1991; Kauer et al., 1991; Kauer and
phic vesicles and makes type 2 synapses mainly onto the Cinelli, 1993; Cinelli et al., 1995) and intrinsic signal
dendritic shafts and somata of granule cells. The axon termi- imaging of odor-evoked neuronal activity in the bulb
nals of both groups also form contacts with somata and den- (Rubin and Katz, 1999, 2001; Uchida et al., 2000;
drites of short axon cells. Intrabulbar axons, which form type Belluscio and Katz, 2001; Meister and Bonhoeffer, 2001);
1 synapses with granule and short axon cells, are the axon and electrophysiological recordings of sensory and bulbar
collaterals of principal neurons (Price and Powell, 1970b). neuron responses to odorants (e.g., Leveteau and
Short axon cells are thought to be the source of intrabulbar MacLeod, 1966; Getchell, 1974; Getchell and Getchell,
axons that make type 2 synapses on granule cells (Price and 1974; Kauer and Moulton, 1974; Getchell and Shepherd,
Powell, 1970b). Results of electrophysiological experiments 1978; Mori and Yoshihara, 1995; Yokoi et al., 1995;
suggest that short axon cells exert inhibitory influences on Kashiwadani et al., 1999).
granule cells (Mori, 1987). Centrifugal axon terminals that In 1991, a large multigene family encoding up to 1000
contain round synaptic vesicles and make type 1 contacts putative olfactory receptor (OR) proteins in rat OSNs was
with dendrites and somata of granule and short axon cells discovered (Buck and Axel, 1991). Homologous multigene
have been described (Price, 1968; Price and Powell, families were then identified in some other vertebrates and
1970b,c). Axons of GABAergic brain neurons presumably in humans (see Lancet and Ben-Arie, 1993; Ressler et al.,
establish type 2 synapses on granule and short axon cells. 1994a; Sullivan et al., 1995; Buck, 1996; Mombaerts,
Centrifugal axons are likely to make synaptic contacts with 1999; Glusman et al., 2000, 2001) (see Chapter 4). This
different types of granule cells, thus modulating bulbar out- discovery, confirmed at the functional level (Zhao et al.,
put signals generated by different types of principal neurons. 1998), provided a vital clue to understanding mechanisms
146 Kratskin and Belluzzi

of odor coding in the olfactory system. In situ hybridiza- at a large distance from the glomeruli activated at lower
tion studies with specific gene probes have shown that (1) concentrations (Johnson and Leon, 2000a).
each OSN may express only one OR, (2) OSNs expressing The gathering of voluminous data has led to a concept
the same OR are randomly scattered within one of four of “odor maps” in the olfactory bulb and allowed formu-
spatial zones of the olfactory epithelium, and (3) OR lation of basic principles underlying their organization at
mRNAs are present in olfactory axons. The latter finding the molecular, cellular, and systems level (Xu et al.,
allowed the investigation of the projection pattern of OSNs 2000). This concept considers olfactory glomeruli as
expressing different ORs. structural and functional modules that extend to the deep
Studies of OR gene-labeled olfactory axons in the of the bulb, involving principal and intrinsic cells associ-
olfactory bulb have demonstrated that (1) OSNs expressing ated with the activated glomerulus. Such multicellular
a given OR project to two individual glomeruli located in modules in the bulb are similar to columns and barrels in
the dorsomedial and ventrolateral portions of the bulb, (2) the cortex. Thus, it appears that glomerular modules syn-
the position of these glomeruli is bilaterally symmetric and thesize, piece by piece, molecular information about
constant in different animals within a species, and (3) there odorant features, providing the basis for the recognition
is a clear correspondence between the number of genes and of odor quality and intensity, and the discrimination
the number of glomeruli identified with each probe between odors. Perceptual reconstitution of odors occurs
(Resssler et al., 1994a,b; Vassar et al., 1994; Buck, 1996; at higher levels of the olfactory system, giving birth to the
Mombaerts et al., 1996). These findings suggested that sensation of smell.
each glomerulus receives input from OSNs expressing a
given OR and that information, broadly distributed in the
olfactory epithelium, “is transformed in the bulb into a III. CENTRIFUGAL INNERVATION
highly organized and spatially stereotyped information
map, which is, in essence, a map of information provided The central nervous system controls and adjusts incoming
by different ORs” (Buck, 1996). The OR protein appears flow and processing of afferent signals via centrifugal projec-
to be an important, but not the sole, determinant in estab- tions to various levels of sensory pathways (Hagbarth, 1960).
lishing such sensory map and maintaining its constancy The olfactory bulb is unique among primary sensory centers
throughout life (Singer et al., 1995; Mombaerts et al., in receiving extraordinary dense centrifugal, or bulbopetal,
1996; Wang et al., 1998; Gogos et al., 2000). inputs (Ramón y Cajal, 1911; Ottoson and Shepherd, 1967;
The results of functional and molecular studies indi- Macrides and Davis, 1983; Kratskin, 1987; Halász, 1990;
cated that a single odorant might activate a number of dif- Shipley and Ennis, 1996). Two major groups of axons project
ferent ORs. This suggested that different ORs recognize to the bulb from brain. One group is comprised of the afferent
different structural features of the odor molecule and map fibers that arise from the primary olfactory cortex, whereas
them onto distinct groups of glomeruli. According to this the other group is comprised of axons originating in nonolfactory
suggestion, a unique combination, or ensemble, of acti- structures of the basal forebrain and brainstem (Fig. 4).
vated glomeruli would encode each odorous chemical.
Systematic studies using 2-deoxyglucose autoradiogra- A. Projections from the Primary Olfactory Cortex
phy provided strong evidence that different odorants are
represented by distinct spatial activity patterns in the The most prominent projection to the olfactory bulb origi-
glomerular layer and that modules of activity within nates in the anterior olfactory nucleus (AON), a structure
these spatial patterns correlate with specific structural that contains about 54% of bulbopetal neurons in the
features of odor molecules (Johnson et al., 1998, 1999; mouse brain (Carson, 1984a). All parts of the AON, except
Johnson and Leon, 2000a,b). These results “are consis- the pars externa, project to both olfactory bulbs; neurons of
tent with a combinatorial mechanism of olfactory coding the pars externa send their axons exclusively to the con-
wherein unitary responses of olfactory receptors to odor- tralateral bulb (Haberly and Price, 1978b; Davis and
ant features would produce spatial patterns of bulbar Macrides, 1981; Luskin and Price, 1983; Schoenfeld and
activity that are characteristic for a given odorant” Macrides, 1984). In the AON of rats, approximately 50%
(Johnson et al., 1988). This mechanism allows for the of bulbopetal neurons have bilateral projections with dif-
discrimination of very subtle structural features that dis- ferent branches of the same axon (Valverde, 1965).
tinguish geometric isomers (Johnson and Leon, 2000b) Afferents from the AON predominantly terminate in the
and enantiomers (Rubin and Katz, 2001). Moreover, it granule cell layer and, to a lesser extent, in the internal
has been found that an increase in the odorant concentra- plexiform layer and glomerular layer. Connections
tion results in stimulation of additional glomeruli, located between the olfactory bulb and the AON pars externa are
Anatomy and Neurochemistry of the Olfactory Bulb 147

topographically organized; distinct sectors of the pars part, and their axons terminate in the granule cell layer
externa receive inputs from restricted areas of the ipsilateral (Haberly and Price, 1978a). In addition, the piriform
bulb and project to corresponding areas in the contralateral cortex is the source of a pathway reaching the olfactory
bulb (Schoenfeld and Macrides, 1984). The dorsal pedun- bulb after a synaptic relay in the AON (Haberly and
cular cortex and ventral taenia tecta contain neurons pro- Price, 1978a; Davis and Macrides, 1981). Projections
jecting to the granule cell layer, internal plexiform layer, from the AON and piriform cortex are found in the olfac-
and glomerular layer of the ipsilateral olfactory bulb. tory bulb of rats at the time of birth (Schwob and Price,
The piriform cortex is another substantial source of 1984).
projections to the olfactory bulb. In mice, this structure Relatively numerous bulbopetal cells (2.4% in the
contains about 36% of bulbopetal brain neurons. Cells mouse brain) are found in the nucleus of the lateral olfac-
projecting to the bulb are located in layers IIb and III of tory tract. Axons arising from this nucleus project to the
the piriform cortex, with the highest density in its rostral ipsilateral and contralateral olfactory bulbs (DeOlmos et
al., 1978; Carson, 1984a), and terminate in the deep part of
the granule cell layer (Davis and Macrides, 1981; Luskin
and Price, 1983). Neurons projecting to the ipsilateral bulb
are also located in the entorhinal cortex, anterior and pos-
terolateral cortical amygdaloid nuclei, and in the peri-
amygdaloid area (DeOlmos et al., 1978; Shipley and
Adamek, 1984).

B. Projections from the Basal Forebrain

and Brainstem

The nucleus of the horizontal limb of the diagonal band

(NHDB) is the major source of bulbopetal axons arising
from nonolfactory brain structures (Price, 1969; Price and
Powell, 1970e; Macrides and Davis, 1983; Shipley and
Ennis, 1996). The NHDB is a component of the basal fore-
brain system that innervates neocortex and hippocampus
and plays an important role in learning and memory. There
are two compartments in the NHDB of rats; the medial
compartment contains small-to-medium-sized cells,
whereas the lateral part is composed of large neurons and
is often referred to as the magnocellular preoptic nucleus
(Záborszky, et al., 1986). Both parts of the NHDB con-
tribute 3.5% of bulbopetal cells found in the mouse brain
(Carson, 1984a). Fibers from the NHDB reach the ipsilat-
eral bulb within the medial forebrain bundle (DeOlmos
et al., 1978; Macrides et al., 1981) and lateral olfactory
tract (Price, 1969; Price and Powell, 1970e) and have
restricted, nonoverlapping projection areas (Luskin and
Price, 1982). Brain lesions and tract-tracing studies indi-
Figure 4 Sources of centrifugal projections to the olfactory cate that NHDB neurons project to the glomerular layer,
bulb: 1, olfactory bulb; 2, anterior olfactory nucleus; 3, dorsal granule cell layer, and EPL (Price, 1968; Price and Powell,
peduncular cortex; 4, ventral taenia tecta or anterior hippocampal 1970c; Godfrey et al., 1980a,b; Macrides et al., 1981).
rudiment; 5, nucleus of the vertical limb of the diagonal band; 6,
Local injections of biotin dextran amine into the medial
nucleus of the horizontal limb of the diagonal band; 7, primary
and lateral compartments of the rat NHDB give rise to ter-
olfactory cortex; 8, lateral preoptic area; 9, lateral hypothalamus;
10, nucleus of the lateral olfactory tract; 11, posterolateral corti- minal arborizations in the glomerular layer and granule
cal nucleus of the amygdala; 12, raphe nuclei; 13, locus cell layer, respectively (Kratskin and Yu, 1996a).
coeruleus. Broken lines show the amygdaloid nuclei and subdivi- A few bulbopetal neurons (about 0.2% in the mouse
sions of the anterior olfactory nucleus. Pathways of centrifugal brain) have been found in the ventral part of the nucleus of
fibers are shaded. (Adapted from DeOlmos et al., 1978.) the vertical limb of the diagonal band, substantia innomi-
148 Kratskin and Belluzzi

nata, and the ventral pallidum (DeOlmos et al., 1978; (Macrides and Davis, 1983; Kratskin, 1987, 1989; Halász,
Carson, 1984a; Záborszky, et al., 1986). A small number of 1990). In bony fishes, amphibians, and reptiles, the olfac-
bulbopetal cells are also located in the lateral and dorso- tory bulb receives afferent fibers from the olfactorecipient
medial hypothalamic areas and in the subthalamic zona regions of the telencephalon and nonolfactory structures of
incerta (Carson, 1984a; Shipley and Adamek, 1984). the forebrain and brainstem (Prasada Rao and Finger,
Axons from these brain sources project to the ipsilateral 1984; Kratskin, 1987; Belekhova et al., 1995; Duchamp-
olfactory bulb. Viret and Duchamp, 1997; Lanuza and Halpern, 1998).
The dorsal and median raphe nuclei are the source of Ultrastructural observations in the frog show that many
the bilateral projection to the olfactory bulb. These centrifugal axons terminate on deep dendrites of granule
unpaired midbrain nuclei contribute 0.5% (about 400 neu- cells (I. Kratskin, J. P. Rio, N. Kenigfest and J. Repérant,
rons) of the total number of bulbopetal cells in mice manuscript in preparation).
(Carson, 1984a). Almost 1300 raphe neurons project to
the rat olfactory bulb (McLean and Shipley, 1987a). Tract- D. Functional Implications
tracing studies have shown that axons of raphe neurons
pass within the medial forebrain bundle. These axons Centrifugal inputs to the olfactory bulb are likely tonic
largely terminate around and within glomeruli and, to a in character (e.g., Paolini and McKenzie, 1997b) and
lesser extent, in the deeper layers of the bulb (Bobillier et may effectively influence bulbar processing by modulat-
al., 1979; McLean and Shipley, 1987a). The locus ing the activity of local interneurons (see Linster and
coeruleus, the paired pontine nucleus, also projects to Gervais, 1996; Linster and Hasselmo, 1997). Reciprocal
both olfactory bulbs. This nucleus contains about 0.4% of connections between the bulb and secondary olfactory
mouse bulbopetal cells (Carson, 1984a). In the rat, up to centers from multiple feedback loops, which may serve
40% of the 1600 neurons in the locus coeruleus send to coordinate signal processing and self-regulation in the
axons to the bulb (Shipley et al., 1985). Fibers from the olfactory system. Inputs from the basal forebrain and
locus coeruleus run in the medial forebrain bundle and ter- brainstem likely exert modulatory influences on the bul-
minate on different parts of granule cells in the granule bar neuronal network and provide interactions between
cell layer and internal plexiform layer (Macrides et al., olfactory and other sensory systems. Bulb output signals
1981; Halász, 1990). are transmitted directly to specific cortical zones, thus
avoiding reticular modulation at the thalamic level of
C. General Characteristics sensory processing. Projections from the raphe nuclei,
which are part of the ascending reticular system, may
The characteristic features of centrifugal innervation of allow for the reticular control over the olfactory input to
the mammalian olfactory bulb are as follows: (1) multiple brain. The projection from the lateral hypothalamus, a
afferent fibers to the bulb originate in both olfactory and structure that receives inputs from the secondary olfac-
nonolfactory brain structures; (2) there is no clear corre- tory centers, complete a complex path that may influ-
spondence between the position of a bulbopetal neuron in ence, for example, the organization of feeding behavior.
the brain and the location of its terminal field in the bulb; The lateral hypothalamic area projects to the NHDB, and
(3) the AON and piriform cortex contain the vast majority “olfactory information does reach the nucleus of origin
(about 90% in the mouse) of bulbopetal neurons; (4) cen- of the olfactory centrifugal fibres, but only after conver-
trifugal axons largely project to the ipsilateral olfactory gence of these different pathways upon the hypothala-
bulb, but the locus coeruleus, nucleus of the lateral olfac- mus and interaction with midbrain and hypothalamic
tory tract, raphe nuclei, and the AON (except for the pars influences” (Price and Powell, 1970f ). Inputs to the
externa) have bilateral projections; (5) axons from the olfactory bulb from the NHDB, whose neuron activity
AON pars externa exclusively project to the contralateral may be modulated by the bulbar output (Paolini and
bulb, and this projection is the only one that is topograph- McKenzie, 1997a; Linster and Hasselmo, 2000), are
ically organized; (6) centrifugal axons mostly terminate believed to play an important role in olfaction (Paolini
on intrinsic neurons; and (7) fibers from all brain sources and McKenzie, 1993, 1996).
contact different parts of granule cells, whereas projec- The function of the olfactory bulb is likely not confined to
tions from the AON and nonolfactory brain structures also the sense of smell. It is possible that, at least in nonhuman
reach interneurons in the glomerular layer. mammals, the bulbs perform not only sensory functions, but
Extensive centrifugal innervation of the primary olfac- are also directly involved in nonspecific, limbic-related mech-
tory center is observed across vertebrate species and has a anisms of arousal and forebrain excitation (Herrick, 1933;
conservative structural and functional organization Cain, 1974; Wenzel, 1974; Shepherd et al., 1981). Diverse
Anatomy and Neurochemistry of the Olfactory Bulb 149

behavioral changes, alterations in learning and memory, and lar neuropil (see Trombley and Shepherd, 1993; Shepherd
impairments of brain transmitter systems are observed in bul- and Greer, 1998). Some mGluRs are involved in postsy-
bectomized rats (e.g., Kelly et al., 1997; Yamamoto et al., naptic effects of Glu, whereas those located on terminals
1997), suggesting a complex function of the olfactory bulbs. of olfactory axons (Kinzie et al., 1997) are “autoreceptors”
This may be one of the reasons why the olfactory bulb has an that may regulate presynaptic release of the transmitter.
enormously rich supply of centrifugal fibers, a feature that GABAB receptors, detected on terminals of olfactory
distinguishes it from other primary sensory centers. axons (Bonino et al., 1999), also represent autoreceptors,
and these definitely modulate transmission at the primary
olfactory synapse (Potapov, 1985; Nickell et al., 1994;
IV. NEUROTRANSMITTERS AND Keller et al., 1998; Aroniadou-Anderjaska et al., 2000).
MODULATORS GABA released from interneurons in the glomerular
region triggers both tonic and stimulus-evoked inhibition
The olfactory bulb “appears to be a veritable cornucopia of of Glu release, thus reducing postsynaptic responses.
putative transmitters and neuroactive peptides” (Macrides Even a single impulse coming from OSNs can evoke
and Davis, 1983), “for the number and variety of which it sufficient GABA release to activate presynaptic GABAB
rivals all other regions of the brain” (Halász and Shepherd, receptors (Aroniadou-Anderjaska et al., 2000). This
1983). The presence of some transmitters in the bulb is regulatory mechanism may serve to shape activity
entirely associated with centrifugal axons, thus contribut- patterns in glomerular modules. D2 dopamine (DA)
ing to its neurochemical diversity. The transmitters and receptors that are located on terminals of olfactory
modulators proposed for bulb neurons and afferent fibers axons (Nickell et al., 1991; Koster et al., 1999) provide
are shown in Figure 5. another possibility for regulating Glu release (Hsia et al.,
1999; Berkowicz and Trombley, 2000). A decrease in
presynaptic Glu release, induced by D2 receptor activa-
A. Olfactory Axons tion, may be due to intraglomerular actions of various bulb
1. Transmitter Glutamate neurons (see below). Thus, different autoreceptors may
modulate the efficacy of sensory input to the olfactory
Glutamate (Glu) is enriched in axon terminals of OSNs bulb.
in the olfactory bulb (Sassoè-Pognetto et al., 1993; Didier
et al., 1994). Electrophysiological studies provided evi-
dence that Glu is a transmitter at excitatory synapses 3. Taurine
between olfactory axons and dendrites of mitral / tufted and The amino acid taurine (Taur) is produced by OSNs
PG cells within the glomeruli (Berkowicz et al., 1994; (Kratskin and Hao, 2001), and terminals of olfactory axons
Bardoni et al., 1996; Ennis et al., 1996; Aroniadou- co-localize Glu and Taur (Didier et al., 1994; Kratskin and
Anderjaska et al., 1997, 1999a). Two types of ionotropic Yu, 1996b). Taur is abundant in the brain and is known to
Glu receptors (GluRs) mediate postsynaptic responses cause neuronal inhibition (see Puopolo et al., 1998;
evoked in mitral cells by stimulation of olfactory axons. Kratskin et al., 2000). Several factors are involved in regu-
The early fast response is due to activation of the AMPA lating Taur release (Oja and Saransaari, 2000). Olfactory
type GluRs, whereas GluRs of the NMDA type mediate the axons, in particular, may release Taur via the gaseous sec-
late long-lasting excitation. Prolonged NMDA GluR-medi- ond messenger nitric oxide, which is present in OSN ter-
ated postsynaptic activity appears to facilitate synaptic minals within the bulb (Broillet and Firestein, 1996), as
integration and plasticity and, thus, may play an important well as upon depolarization and activation of presynaptic
role in olfactory processing and memory (Aroniadou- mGluRs. Application of Taur to a slice preparation of rat
Anderjaska et al., 1997; Ennis et al., 1998). Subunits of olfactory bulb produces strong inhibition of mitral and
various GluRs are expressed in the olfactory bulb, and tufted cells (Puopolo et al., 1998). This action is due to
some of them are located on dendrites within glomeruli direct GABAA receptor activation and does not involve
(Trombley and Shepherd, 1993; Giustetto et al., 1997; glycine receptors, which are expressed by bulb neurons
Shepherd and Greer, 1998; Montague and Greer, 1999). (Trombley and Shepherd, 1993, 1994; Trombley et al.,
1999). Further experiments have shown that Taur does not
influence the membrane potential of PG cells (O.
2. Modulation of Glutamate Release
Belluzzi, M. Puopolo, and I. Kratskin, manuscript in
Several types of metabotropic Glu receptors (mGluRs) are preparation). Such a difference in the effects of Taur may
present in different regions of the bulb, including glomeru- be related to the differential subunit composition of
150 Kratskin and Belluzzi

Figure 5 Neurotransmitters and modulators in the olfactory bulb: ACh, acetylcholine; Carn, carnosine; CCK, cholecystokinin; DA,
dopamine; Enk, enkephalin; Glu, glutamate; 5-HT, serotonin; LHRH, luteinizing hormone releasing hormone; NE, norepinephrine; OMP,
olfactory marker protein; SOM, somatostatin; SP, substance P; Taur, taurine. Small arrows show the direction of synaptic transmission;
solid arrows indicate centrifugal projections to the bulb. (Adapted from Halász and Shepherd, 1983.)

GABAA receptors from mitral/tufted and PG cells (see 4. Olfactory Marker Protein, Carnosine, Zinc
Laurie et al., 1992; Persohn et al., 1992; Fritschy and
Mohler, 1995), implying specific molecular structure OSNs express a specific protein called olfactory marker
requirements of GABAA receptors for Taur sensitivity. In protein (OMP) and the dipeptide carnosine (Margolis
these studies, Taur significantly reduced GluR-mediated et al., 1986). OMP gene deletion in mice alters the ability
excitatory currents evoked in external tufted cells by olfac- of OSNs to generate the electro-olfactogram, suggesting
tory axon stimulation. The GABAB receptor antagonist that the neural activity directed toward the bulb is also
CGP55845A blocked this effect (which could not be decreased (Buiakova et al., 1996). Carnosine and Glu are
ascribed to GABAA receptor activation), suggesting that co-localized in olfactory axon terminals (Sassoè-Pognetto
Taur acts on presynaptic GABAB receptors and decreases et al., 1993), which also contain high levels of zinc (see
Glu release. It is possible that one function of Taur in the Trombley and Shepherd, 1993). Bath application of
olfactory bulb is to moderate the excitability of principal carnosine increases membrane conductance in cultured
cells at both pre-and postsynaptic levels. neurons of the bulb (Kanaki et al., 1997). Furthermore,
Anatomy and Neurochemistry of the Olfactory Bulb 151

carnosine reduces inhibitory actions of zinc on NMDA and and mGluRs are involved in regulating transmitter release
GABA receptor-mediated currents and synaptic transmis- from mitral cell endings.
sion (Trombley et al., 1998). These observations suggest
that carnosine released from olfactory axons may
2. Tufted Cells
modulate the excitability of bulb neurons.
Dopamine (DA) is a putative transmitter in external and
middle tufted cells. Several studies have shown that many
B. Principal Cells tufted cells exhibit immunoreactivities for the enzymes of
DA synthesis tyrosine hydroxylase (TH) and dopadecar-
1. Mitral Cells
boxylase (see Davis and Macrides, 1983; Halász, 1990).
Mitral cells and internal tufted cells contain Glu (e.g., In the olfactory bulb of hamsters, more than 80% of
Ottersen and Storm-Mathisen, 1984; Liu et al., 1989), and TH-containing neurons are external tufted cells. TH-
a high density of GluRs is found in the EPL (Cotman immunostained external tufted cells are also present in
et al., 1987), where principal neurons make numerous the human olfactory bulb (Smith et al., 1991). Type D1
synaptic contacts. Electrophysiological and pharmacologi- and D2 receptors for DA are expressed in the glomerular
cal analyses provide evidence that Glu is a transmitter in layer, EPL, and the mitral cell layer (Coronas et al.,
mitral / tufted-to-granule dendrodendritic synapses and 1997). External naris closure in rats produces a rapid and
suggest that granule cell excitation is mediated predomi- long-lasting decrease in the TH activity and expression of
nantly by NMDA GluRs (Isaacson and Strowbridge, 1998; TH mRNA in the olfactory bulb (Cho et al., 1996), as
Schoppa et al., 1998; Chen et al., 2000; Halabisky et al., well as a reduction in the DA content (Philipot et al.,
2000; Christie et al., 2001; Isaacson, 2001). Calcium 1998). However, high-frequency stimulation of the olfac-
influx through NMDA GluRs may trigger directly GABA tory nerve in rats whose external nares have been closed
release from granule cell dendrites. It has been shown that, results in a partial and temporary increase in the DA lev-
under particular conditions, AMPA GluRs can mediate els. These findings demonstrate that sensory input regu-
synaptic actions of mitral cells (Isaacson, 2001). Indeed, a lates DA production in the olfactory bulb. Recent in vitro
recent immunocytochemical study localized NMDA and experiments have confirmed the notion that sensory-
AMPA GluRs at postsynaptic sites on granule cell dendritic dependent TH expression occurs in bulb neurons and
spines, and revealed the spatial organization of these have shown that NMDA GluRs may mediate the
receptors (Sassoè-Pognetto and Ottersen, 2000). Various influence of sensory input (Puche and Shipley, 1999).
GABAA receptor subunits are found in the glomerular Substance P is found in external tufted cells of the ham-
layer and EPL, where mitral cells have numerous synaptic ster (Burd et al., 1982), and mRNA encoding substance P
contacts (Pirker et al., 2000). is expressed in external tufted cells of rats (Warden and
Several groups of investigators have shown that Young, 1988). In rats, external tufted cells, which consti-
Glu released from both apical and basal dendrites of tute the intrabulbar associational system, are immuno-
mitral/tufted cells causes activation of the same and reactive for cholecystokinin (Liu and Shipley, 1994).
neighboring principal cell dendrites (Aroniadou-
Anderjaska et al., 1999a,b; Isaacson, 1999; Friedman and C. Intrinsic Neurons
Strowbridge, 2000; Salin et al., 2001). The NMDA type
1. Periglomerular Cells
autoreceptors were found to mediate self-excitation of
principal cells at dendritic sites. No synaptic contacts GABA and DA are the most likely transmitters of PG cells
between dendrites of mitral/tufted cells have been (Mugnaini et al., 1984; Halász, 1990; Kosaka et al., 1995;
observed in mammals, so the process of self-excitation is Toida et al., 2000). Studies in rats have shown the coexis-
thought to be nonsynaptic in nature. It may be noted, tence of TH and the GABA-synthesizing enzyme glutamic
however, that an ultrastructural study of the salamander acid decarboxylase (GAD) or TH and GABA in PG cells
olfactory bulb revealed type 1 synapses between dendrites (Kosaka et al., 1985; Gall et al., 1987). Indeed, 30–70% of
of principal cells in the glomerular layer and EPL (Allen immunoreactive cells produce both GABA and DA. Many
and Hamilton, 2000). In addition to ionotropic GluRs, PG cells contain enkephalin and substance P (Davis et al.,
mitral cells express mGluRs, and these are located on 1982; Halász, 1990). Co-localization of enkephalin with
dendrites within glomeruli (van den Pol, 1995; Kinzie et GABA and/or TH (Kosaka et al., 1995) and coexistence of
al., 1997). Activation of mGluRs results in suppression of substance P with GABA and DA (Kosaka et al., 1988)
Glu release from mitral cell terminals (Schoppa and have been observed in PG cells. It is generally believed
Westbrook, 1997). These findings suggest that both GluRs that PG cells are largely inhibitory in nature. It has been
152 Kratskin and Belluzzi

shown that inhibitory synaptic interactions between neigh- 2001). These results suggest that there are granule cells
boring PG cells are mediated by GABAA receptors that contain Glu in synaptic vesicles and exert NMDA
(Puopolo and Belluzzi, 1998b). However, somata and den- GluR-mediated postsynaptic excitation of mitral cells in
drites of PG cells may accumulate high concentrations of the EPL. Whether the same granule cell may release both
C1 (Siklos et al., 1995), thus providing the basis for exci- GABA and Glu is unknown.
tatory actions of GABA observed in the glomeruli
(Rhoades and Freeman, 1990). The lack of odor stimula- 3. Short Axon Cells
tion (without denervation of the bulb) results in an alter-
Short axon cells likely use GABA and DA as neurotrans-
ation in DA, but not GABA, production in the PG cells.
mitters (Halász, 1990). The coexistence of these sub-
This suggests differential regulation of the transmitter-
stances has been reported in the superficial short axon cells
specific phenotype of PG cells by OSN activity–dependent
of rats (Gall et al., 1987). In rats and hamsters, some short
factors (Baker, 1990). On the other hand, the density of
axon cells are immunopositive for enkephalin and somato-
synapses, made by olfactory axons, is greater on TH than
statin (Davis et al., 1982). Presumptive short axon cells
on GABA-immunostained cell processes; therefore,
containing somatostatin, GABA, and DA have been
GABA-containing PG cells may be less sensitive to the
described in the bulb of humans (Ohm et al., 1988, 1990;
lack of sensory input (Bartolomei and Greer, 1993). DA,
Smith et al., 1993).
GABA, and substance P were observed in presumed PG
cells of the human olfactory bulb (Ohm et al., 1990; Smith
et al., 1991, 1993). D. Centrifugal Fibers
1. Serotonin
2. Granule Cells
The only source of serotonin (5-HT) in the bulb is the pro-
Granule cells are the most neurochemically homogeneous jection from the raphe nuclei (McLean and Shipley, 1987a;
intrinsic neurons, and GABA most fully satisfies the tradi- Araneda et al., 1989). Some bulbopetal raphe neurons
tional criteria for transmitter identification, clearly being a co-localize 5-HT and galanin or somatostatin (Araneda
transmitter of granule cells (Halász and Shepherd, 1983; et al., 1999). The density of 5-HT fibers in the glomerular
Halász, 1990). The highest levels of GABA, the greatest layer is two to three times that of any other layer; deple-
activity of its metabolic enzymes, GAD and GABA- tion of 5-HT from axons in the bulb produces shrinkage of
aminotransferase, as well as release of GABA and its glomeruli (Moriizumi et al., 1994). 5-HT-containing
specific uptake, are found in the EPL. The sites of dendro- fibers are present in the glomerular region of the human
dendritic synapses between granule and mitral/tufted cells bulb (Smith et al., 1993). During rat postnatal develop-
have the highest level of 3H-GABA binding and very high ment, the rate of 5-HT fiber arborization exceeds the
GAD activity. Nearly all somata and dendrites of granule growth rate of glomeruli (McLean and Shipley, 1987b). 5-
cells display immunoreactivities for GAD and GABA. HT receptor subtypes 1A and 2A are dominant in the bulb
GAD-immunostained granule cells are also observed in (Wright et al., 1995). While immunostaining localized 2A
the bulb of humans (Ohm et al., 1990). Many granule cells receptors to PG and granule cells (Morilak et al., 1993), in
in the rat and hamster contain enkephalin; granule cell den- situ hybridization studies showed expression of specific
drites immunostained for enkephalin are found in the EPL mRNA in mitral and tufted cells (Pompeiano et al., 1994;
(Davis et al., 1982; Matsutani et al., 1988). Accordingly, a McLean et al., 1995). Surprisingly, no 5-HT receptors
high level of -receptor binding sites is observed in the were found in the glomeruli. An involvement of 5-HT in
EPL (McLean et al., 1986). In the rat granule cell layer, olfactory learning (McLean et al., 1993) and odor dis-
95% of enkephalin-stained cell bodies are immunoreactive crimination in rats (Moriizumi et al., 1994) has been
for GAD, whereas only 11% of GAD-immunostained shown.
somata show immunoreactivity for enkephalin. This sug-
gests that enkephalin-containing neurons represent a sub-
2. Norepinephrine
population of GABAergic granule cells (Kosaka et al.,
1987). Granule cells express NMDA and AMPA GluRs Norepinephrine (NE) is also not intrinsic to the olfactory
(Sassoè-Pognetto and Ottersen, 2000) and a specific type bulb. NE-containing fibers arise from the locus coeruleus
of mGluRs (van den Pol, 1995). Striking results have been and terminate on dendrites and somata of granule cells in
recently obtained using a combination of whole-cell the internal plexiform and granule cell layers (Macrides
recordings in a slice preparation of rat olfactory bulb and et al., 1981; Font et al., 1982; McLean et al., 1989; Halász,
electron microscopic immunocytochemistry (Didier et al., 1990). Postnatal maturation of this input (McLean and
Anatomy and Neurochemistry of the Olfactory Bulb 153

Shipley, 1991) correlates with the development of NE rats causes inhibition of mitral cells and that this effect
influences on the interaction between granule and mitral might be due to granule cell activation (Nickell and
cells (Wilson and Leon, 1988). Both  and  NE receptors Shipley, 1988b). Intracellular recordings, however, demon-
are present the olfactory bulb (Woo and Leon, 1995; strated inhibition of granule cells and facilitation of mitral
Shipley and Ennis, 1996). Locus coeruleus lesion with cells following single- or brief multiple-pulse stimulation
6-hydroxydopamine reduced the NE content of the bulb of the lateral NHDB (Kunze et al., 1991, 1992a,b). These
(Sullivan et al., 1993) and increased the density of 1 and observations were consistent with the activation of an
2 receptors (Woo et al., 1995). About 20% of bulbopetal inhibitory input to granule cells. A subsequent study using
neurons in the locus coeruleus co-localize NE and neu- extracellular recordings showed that suppression of tonic
ropeptide Y (Bouna et al., 1994). neuronal activity in the lateral NHDB by the long-lasting
Only inhibitory actions of NE on mitral cells were found GABA agonist muscimol results in sustained facilitation of
in vivo (Salmoiraghi et al., 1964; McLennan, 1971), where- presumed tufted cells in the EPL and does not influence
as NE-evoked mitral cell excitation was observed in isolat- mitral cell firing (Paolini and McKenzie, 1997b). Changes
ed turtle olfactory bulb (Jahr and Nicoll, 1982). In bulb cell in unit activity, observed in the granule cell layer, suggested
culture, NE suppressed granule-to-mitral cell inhibition complex interactions between granule and tufted cells.
(Trombley and Shepherd, 1992) due to 2 receptor–mediat- Further studies are needed to elucidate the influence of the
ed reduction of mitral cell Ca2 currents (Trombley, 1992). GABAergic input on neurons of the olfactory bulb.
Activation of the locus coeruleus (Jiang et al., 1996) and
application of NE to a bulb slice preparation (Ciombor et al.,
4. Acetylcholine
1999) increased rat mitral cell responses to perithreshold
olfactory nerve stimulation. This suggests that one function There is evidence that acetylcholine (ACh) is a transmitter
of NE in the bulb is to enhance detection of weak odor stim- of bulbopetal neurons. Bulbar cells do not express mRNA
uli. Both sectioning the olfactory peduncle and  antagonist encoding choline acetyltransferase (ChAT), the enzyme of
injection decreased the number of c-fos–containing granule ACh synthesis (Oh et al., 1992), and central deafferenta-
cells in odor-specific areas, suggesting NE-mediated cen- tion of the olfactory bulb almost completely decreases its
trifugal influences on c-fos expression in the bulb (Sallaz ChAT activity (Godfrey et al., 1980b). While several stu-
and Jourdan, 1996). Behavioral studies showed NE modula- dies have shown no bulb neurons immunopositive for
tory influences on formation of specific odor memories ChAT (see Kratskin, 1989; Halász, 1990), some authors
(Royet et al., 1983; Sullivan et al., 1993, 2000). have observed ChAT-stained cells in the rat olfactory bulb
(Ojima et al., 1988; Phelps et al., 1992). However, the
number of these cells was very small, suggesting that ACh
3. GABA
fibers in the bulb mostly originate from extrinsic sources.
The observation of decreased levels of GABA in the bulb The main source of cholinergic afferents to the olfactory
following its central deafferentation (Godfrey et al., 1980a), bulb is the NHDB (Carson, 1984b; Záborszky et al., 1986).
the finding by electron microscopy of 3H-GABA–labeled In the rat NHDB, up to 20% of bulbopetal neurons are
terminals in the granule cell layer (which did not resemble cholinergic, and most of them are located rostrally in the
intrinsic neuron terminals) (Halász, 1990), and early medial compartment of the NHDB. Some cholinergic cells
pharmacological experiments (McLennan, 1971) have in the NHDB give rise to divergent projections to the bulb
pointed to GABA as a transmitter in centrifugal axons. This and hippocampus (Okoyama et al., 1987). Presumably
was definitively established when GABAergic bulbopetal cholinergic bulbopetal neurons are located in the NHDB of
neurons were found in the rat NHDB (Záborszky et al., frogs (Kratskin and Ragimova, 1985).
1986). About 30% of bulbopetal cells in this structure are Cholinergic fibers terminate in the glomerular layer,
GABAergic; most of them are located caudally in the lateral EPL, and, to a lesser extent, in the internal plexiform and
NHDB compartment. A few such cells are also observed in granule cell layers. Ultrastructural studies have shown that
the ventral pallidum, anterior amygdaloid area, piriform ChAT-stained terminals make synapses mainly with den-
cortex, and nucleus of the lateral olfactory tract. drites/somata of intrinsic neurons, i.e., PG cells, short axon
GABAergic centrifugal innervation is likely not specific to cells, and granule cells, and also with external tufted cells
the mammalian bulb, as GABA-stained bulbopetal cells are (Le Jeune and Jourdan, 1993; Kasa et al., 1995).
found in the amphibian brain as well (Kratskin et al., 1991, Importantly, many of these contacts were of the symmetric
1992, 1997). type, which is generally associated with inhibitory synap-
A field potential study suggested that continuous tic actions. Cholinoceptive cells have been studied by light
electrical stimulation (at 10 Hz) of the lateral NHDB in and electron microscopic histochemistry of the ACh-
154 Kratskin and Belluzzi

degrading enzyme acetylcholinesterase (AChE). These bulb, whereas a muscarinic agonist has an opposite action
studies suggest that likely cholinoceptive neurons are short (El-Etri et al., 1999). This suggests that release of NE
axon cells across the bulb and TH-stained tufted cells; a from centrifugal fibers is differentially regulated through
few PG cells, but not granule cells, are also stained for different AChRs. There is a possibility that presynaptic
AChE (e.g., Nickell and Shipley, 1988a; Le Jeune and nAChRs facilitate Glu-mediated synaptic transmission in
Jourdan, 1994). It is possible, however, that cholinoceptive the bulb (Girod et al., 2000).
bulb cells have no AChE or that its activity is present in
neurons that do not receive cholinergic input (Kasa et al.,
5. Other Neuroactive Substances
1996). ChAT-stained fibers, but not cell bodies, are found
in the human olfactory bulb; such fibers form a dense Centrifugal fibers terminating in different regions of the
plexus around glomeruli (S. Arnold, personal communica- olfactory bulb may contain excitatory amino acids, such
tion, 1999). The possibility that alterations in cholinergic as Glu and aspartate, DA, and various peptides. In
innervation of the bulb may be one of the causes of olfac- particular, atypical glomeruli in rats receive a dense
tory dysfunction in Alzheimer’s disease is discussed (e.g., supply of axons displaying immunoreactivity for luteinizing
Kasa et al., 1997; Durand et al., 1998). hormone–releasing hormone (Zheng et al., 1988). Other
Both muscarinic and nicotinic ACh receptors (mAChRs peptides found in bulbopetal fibers include substance
and nAChRs) are expressed in the olfactory bulb (see P, enkephalin, somatostatin, neuropeptide Y, oxytocin,
Shipley and Ennis, 1996). The EPL has the highest concen- vasopressin, and cholecystokinin. The sources of
tration of mAChRs in the brain (Rotter et al., 1979). these projections remain unknown. Most likely, bul-
Intermediate levels of type 1 and 2 mAChRs are observed bopetal neurons co-localize peptides and “classical” trans-
deep to the EPL, whereas the glomerular layer shows the mitters, but the significance of such coexistence is not
lowest density of mAChRs. In contrast, nAChRs are con- understood.
centrated in superficial layers, including the glomerular
layer (Le Jeune et al., 1995). At the ultrastructural level, ACKNOWLEDGMENTS
type 1 and 2 mAChRs are found on dopaminergic and
GABAergic cells that receive input from olfactory axons, The authors wish to acknowledge the support of the
on granule cell gemmules in the EPL, and somata and den- National Institutes of Health (grant DC04083; I. K.) and
drites of deep short axon cells (Crespo et al., 2000). In gen- MURST-Cofin (O. B.).
eral, the laminar distribution of AChRs correlates with that
of ChAT-stained terminals and AChE-stained cells,
suggesting that different AChR types may mediate cholin- REFERENCES
ergic actions at different levels of odor processing. It is also
possible that excitatory and inhibitory effects of ACh on Adrian, E. D. (1950). The electrical activity of the mammalian
bulb neurons are mediated by nAChRs and mAChRs, olfactory bulb. EEG Clin. Neurophysiol. 2:377–388.
Adrian, E. D. (1951). Olfactory discrimination. Ann. Psychol.
respectively (Castillo et al., 1999). Atypical glomeruli
50:107–113.
receive a dense supply of cholinergic axons but fail to Allen, D. M., and Hamilton, K. A. (2000). Ultrastructural identi-
express any AChRs (Zheng et al., 1987; Le Jeune and fication of synapses between mitral/tufted cell dendrites.
Jourdan, 1993; Le Jeune et al., 1995). In development, Brain Res. 860:170–173.
cholinergic input and nAChRs appear much earlier than Allison, A. C., and Warwick, T. T. (1949). Quantitative observa-
mAChRs (Le Jeune et al., 1996). An increase in the number tions on the olfactory system of the rabbit. Brain
of mAChRs (within the first 4 weeks of life) likely coin- 72:186–196.
cides with granule cell maturation. Altman, J. (1969). Autoradiographic and histological studies of
Electrical stimulation of the NHDB reduces the bulb postnatal neurogenesis. IV. Cell proliferation and migration in
field potential evoked by activation of the anterior com- the anterior forebrain, with special reference to persisting neu-
missure, suggesting that bulbopetal input from the rogenesis in the olfactory bulb. J. Comp. Neurol.
137:433–458.
NHDB regulates transmission of olfactory information
Araneda, S., Magoul, R., and Calas, A. (1989). Tracing specific
between hemispheres (Nickell and Shipley, 1993). transmitter pathways in the rat CNS: combination of [3H]
Iontophoresis of ACh produces both facilitation and serotonin retrograde labeling with immunocytochemical
depression of spontaneously active neurons in different detection of endogenous transmitters. J. Neurosci. Meth.
layers of the rat olfactory bulb (Ravel et al., 1990). Local 30:211–218.
infusion of ACh, nicotine, and the irreversible AChE Araneda, S., Gysling, K., and Calas, A. (1999). Raphe serotoner-
inhibitor soman increases the basal NE content of the gic neurons projecting to the olfactory bulb contain galanin or
Anatomy and Neurochemistry of the Olfactory Bulb 155

somatostatin but not neurotensin. Brain Res. Bull. Bonino, M., Cantino, D., and Sassoè-Pognetto, M. (1999).
49:209–214. Cellular and subcellular localization of -aminobutyric acidB
Aroniadou-Anderjaska, V., Ennis, M., and Shipley, M. T. (1997). receptors in the rat olfactory bulb. Neurosci. Lett.
Glomerular synaptic responses to olfactory nerve input in rat 274:195–198.
olfactory bulb slices. Neuroscience 79:425–434. Bouna, S., Gysling, K., Calas, A., and Araneda, S. (1994). Some
Aroniadou-Anderjaska, V., Ennis, M., and Shipley, M. T. noradrenergic neurons of locus coeruleus-olfactory pathway
(1999a). Current-source density analysis in the rat contain neuropeptide. Y. Brain Res. Bull. 34:413–417.
olfactory bulb: laminar distribution of kainate/AMPA-and Broillet, M. -C., and Firestein, S. (1996). Gaseous second mes-
NMDA-receptor-mediated currents. J. Neurophysiol. sengers in vertebrate olfaction. J. Neurobiol. 30:49–57.
81:15–28. Brunjes, P. C., and Frazier, L. L. (1986). Maturation and plasticity
Aroniadou-Anderjaska, V., Ennis, M., and Shipley, M. T. in the olfactory system of vertebrates. Brain Res. Rev. 11:1–45.
(1999b). Dendrodendritic recurrent excitation in mitral Buck, L. B. (1996). Information coding in the vertebrate olfac-
cells of the rat olfactory bulb. J. Neurophysiol. tory system. Annu. Rev. Neurosci. 19:517–544.
82:489–494. Buck, L., and Axel. R. (1991). A novel multigene family may
Aroniadou-Anderjaska, V., Zhou, F. M., Priest, C. A., Ennis, M., encode odorant receptors: a molecular basis for odor recogni-
and Shipley, M. T. (2000). Tonic and synaptically evoked tion. Cell 65:175–187.
presynaptic inhibition of sensory input to the rat olfactory Buiakova, O. I., Baker, H., Scott, J. W., Farbman, A., Kream, R.,
bulb via GABAB heteroreceptors. J. Neurophysiol. Grillo, M., Franzen, L., Richman, M., Davis, L. M.,
84:1194–1203. Abbondanzo, S., Stewart, C. L., and Margolis, F. L. (1996).
Astic, L., and Saucier, D. (1986). Anatomical mapping of the Olfactory marker protein (OMP) gene deletion causes altered
neuroepithelial projection to the olfactory bulb in the rat. physiological activity of olfactory sensory neurons. Proc.
Brain Res. Bull. 16:445–454. Natl. Acad. Sci. USA 93:9858–9863.
Baker, H. (1990). Unilateral, neonatal olfactory deprivation alters Burd, G. D., Davis, B. J., and Macrides, F. (1982). Ultrastructural
tyrosine hydrohylase expression but not aromatic amino acid identification of substance P immunoreactive neurons in the
decarboxylase or GABA immunoreactivity. Neuroscience main olfactory bulb of the hamster. Neuroscience
36:761–771. 7:2697–2704.
Bardoni, R., Magherini, P. C., and Belluzzi, O. (1996). Excitatory Cain, D. P. (1974). The role of the olfactory bulb in limbic mech-
synapses in the glomerular triad of frog olfactory bulb in anisms. Psychol. Bull. 81:654–671.
vitro. NeuroReport 7:1851–1855. Carr, V. M., Murphy, S. P., Morimoto, R. I., and Farbman, A. I.
Bartolomei, J. C., and Greer, C. A. (1993). Synaptic organization (1994). Small subclass of rat olfactory neurons with specific
of immunocytochemically identified GABA and TH pro- bulbar projections is reactive with monoclonal antibodies to
cesses in rat olfactory bulb glomeruli. Soc. Neurosci. Abstr. the HSP70 heat shock protein. J. Comp. Neurol. 348:150–160.
19:125. Carson, K. A. (1984a). Quantitative localization of neurons pro-
Bartolomei, J. C., and Greer, C. A. (2000). Olfactory ensheathing jecting to the mouse main olfactory bulb. Brain Res. Bull.
cells: bridging the gap in spinal cord injury. Neurosurgery 12:629–634.
47:1057–1068. Carson, K. A. (1984b). Localization of acetylcholinesterase-
Bercowicz, D. A., and Trombley, P. Q. (2000). Dopaminergic positive neurons projecting to the mouse main olfactory bulb.
modulation at the olfactory nerve synapse. Brain Res. Brain Res. Bull. 12:635–639.
855:90–99. Castillo, P. E., Carleton, A., Vincent, J. D., and Lledo, P. M.
Bercowicz, D. A., Trombley, P. Q., and Shepherd, G. M. (1994). (1999). Multiple and opposing roles of cholinergic transmis-
Evidence for glutamate as the olfactory receptor cell neuro- sion in the main olfactory bulb. J. Neurosci. 19:9180–9191.
transmitter. J. Neurophysiol. 71:2557–2561. Chen, W. R., Xiong, W., and Shepherd, G. M. (2000). Analysis
Belekhova, M. G., Ozirskaya, E. V., Chkheidze, D. D., and of relations between NMDA receptors and GABA release
Tumanova, N. L. (1995). Amygdalofugal innervation of olfac- at olfactory bulb reciprocal synapses. Neuron 25:625–633.
tory bulb in amphibious turtles. J. Evol. Biochem. Physiol. Cho, J. Y., Min, N., Franzen, L., and Baker, H. (1996). Rapid
31:39–44. down-regulation of tyrosine hydroxylase expression in the
Belluscio, L., and Katz, L. C. (2001). Symmetry, stereotypy, and olfactory bulb of naris0occluded adult rats. J. Comp. Neurol.
topography of odorant representations in mouse olfactory 369:264–276.
bulbs. J. Neurosci. 21:2113–2122. Christie, J. M., Schoppa, N. E., and Westbrook, G. L. (2001).
Bhatnagar, K. P., Kennedy, R. C., Baron, G., and Greenberg, R. A. Tufted cell dendrodendritic inhibition in the olfactory bulb is
(1987). Number of mitral cells and the bulb volume in the dependent on NMDA receptor activity. J. Neurophysiol.
aging human olfactory bulb: a quantitative morphological 85:169–173.
study. Anat. Rec. 218:73–87. Cinelli, A. R., Hamilton, K. A., and Kauer, J. S. (1995).
Bobillier, P., Seguin, S., Degueurce, A., Lewis, B. D., and Salamander olfactory bulb neuronal activity observed by
Pujol, J.-F. (1979). The efferent connections of the nucleus video rate, voltage-sensitive dye imaging: III. Spatial and
raphe centralis superior in the rat as revealed by autoradio- temporal properties of responses evoked by odorant stimula-
graphy. Brain Res. 166:1–8. tion. J. Neurophysiol. 73:2053–2071.
156 Kratskin and Belluzzi

Ciombor, K. J., Ennis, M., and Shipley, M. T. (1999). Duchamp-Viret, P., and Duchamp, A. (1997). Odor processing in
Norepinephrine increases rat mitral cell excitatory responses the frog olfactory system. Prog. Neurobiol. 53:561–602.
to weak olfactory nerve input via alpha-1 receptors in vitro. Durand, M., Coronas, V., Jourdan, F., and Quirion, R. (1998).
Neuroscience 90:595–606. Developmental and aging aspects of the cholinergic inner-
Clark, W. E. le Gros. (1951). The projection of the olfactory vation of the olfactory bulb. Int. J. Dev. Neurosci.
epithelium on the olfactory bulb in the rabbit. J. Neurol. 16:777–785.
Neurosurg. Psychiatry 14:1–10. El-Etri, M. M., Ennis, M., Griff, E. R., and Shipley, M. T. (1999).
Clark, W. E. le Gros. (1957). Inquires into the anatomical basis of Evidence for cholinergic regulation of basal norepinephrine
olfactory discrimination. Proc. R. Soc. Lond. Ser. B 146: 299–319. release in the rat olfactory bulb. Neuroscience 93:611–617.
Clark, W. E. le Gros, and Warwick, T. T. (1946). The pattern of Eng, D. L., and Kocsis, J. D. (1987). Activity dependent changes
olfactory innervation. J. Neurol. Neurosurg. Psychiatry in extracellular potassium and excitability in turtle olfactory
14:1–10. nerve. J. Neurophysiol. 57:740–754.
Coronas, V., Srivastava, L. K., Liang, J. J., Jourdan, F., and Ennis, M., Zimmer, L. A., and Shipley, M. T. (1996). Olfactory
Moyse, E. (1997). Identification and localization of nerve stimulation activated rat mitral cells via NMDA and
dopamine receptor subtypes in rat olfactory mucosa and non-NMDA receptors in vitro. NeuroReport 7:989–992.
bulb: a combined in situ hybridization and ligand binding Ennis, M., Linster, C., Aroniadou-Anderjaska, V., Ciombor, K.,
radioautographic approach. J. Chem. Neuroanat. and Shipley, M. T. (1998). Glutamate and synaptic plasticity
12:243–257. at mammalian primary olfactory synapses. In Olfaction and
Cotman, C. W., Monaghan, D. T., Ottersen, O. P., and Storm- Taste XII, C. Murphy and C. A. Greer (Eds.). Ann. NY Acad.
Mathisen, J. (1987). Anatomical organization of excitatory Sci., New York, pp. 457–466.
amino acid receptors and their pathways. Trends Neurosci. Fiske, B. K., and Brunjes, P. C. (2001). Cell death in the devel-
10:273–280. oping and sensory-deprived rat olfactory bulb. J. Comp.
Crespo, C., Blasco-Ibanez, J. M., Brinon, J. G., Alonso, J. R., Neurol. 431:311–319.
Dominguez, M. I., and Martinez-Guijarro, F. J. (2000). Font, C., Araneda, S., Pujol, J. F., and Bobillier, P. (1982).
Subcellular localization of m2 muscarinic receptors in Biochemical and autoradiographic investigations of the
GABAergic interneurons of the olfactory bulb. Eur. J. retrograde axonal transport of labeled material following
Neurosci. 12:3963–3974. [3H]norepinephrine injection in the olfactory bulb.
Davis, B. J., and Macrides, F. (1981). The organization of cen- Neurochem. Int. 4:569–575.
trifugal projections from the anterior olfactory nucleus, ven- Friedman, D., and Strowbridge, B. W. (2000). Functional role
tral hippocampal rudiment and piriform cortex to the main of NMDA autoreceptors in olfactory mitral cells.
olfactory bulb in the hamster: an autoradiographic study. J. Neurophysiol. 84:39–50.
J. Comp. Neurol. 203:475–493. Fritschy, J. -M., and Mohler, H. (1995). GABAA-receptor hetero-
Davis, B. J., and Macrides, F. (1983). Tyrosine hydroxylase geneity in the adult rat brain: differential regional and cellular
immunoreactive neurons and fibers in the olfactory system of distribution of seven major subunits. J. Comp. Neurol.
the hamster. J. Comp. Neurol. 214:427–440. 359:154–194.
Davis, B. J., Burd, G. D., and Macrides, F. (1982). Localization Gall, C. M., Hendry, S. H. C., Seroogy, K. B., Jones, E. G., and
of methionin-enkephalin, substance P and somatostatin Haycock, J. W. (1987). Evidence for coexistence of GABA
immunoreactivities in the main olfactory bulb of the hamster. and dopamine in neurons of the rat olfactory bulb. J. Comp.
J. Comp. Neurol. 204:377–383. Neurol. 266:307–318.
DeOlmos, J., Hardy, H., and Heimer, L. (1978). The afferent con- Getchell, T. (1974). Unitary responses in frog olfactory epithe-
nections of the main and accessory olfactory bulb formations lium to sterically related molecules at low concentrations.
in the rat: an experimental HRP-study. J. Comp. Neurol. J. Gen. Physiol. 64:241–261.
181:213–244. Getchell, T. V., and Getchell, M. L. (1974). Signal-detecting
Didier, A., Ottersen, O. P., and Storm-Mathisen, J. (1994). mechanisms in the olfactory epithelium: molecular discrimi-
Differential subcellular distribution of glutamate and taurine nation. Ann. NY Acad. Sci. 237:62–75.
in primary olfactory neurons. NeuroReport 6:145–148. Getchell, T. V., and Shepherd, G. M. (1978). Responses of olfac-
Didier, A., Carleton, A., Bjaalie, J. G., Vincent, J.-D., Ottersen, tory receptor cells to steep pulses of odour at different con-
O. P., and Storm-Mathisen, J. (2001). A dendrodendritic centrations in the salamander. J. Physiol. (Lond.)
reciprocal synapse provides a recurrent excitatory connection 282:521–540.
in the olfactory bulb. Proc. Natl. Acad. Sci. USA Gheusi, G., Cremer, H., McLean, H., Chazal, G., Vincent, J. D.,
98:6441–6446. and Lledo, P. M. (2000). Importance of newly generated neu-
Doetsch, F., Garcia-Verdugo, J. M., and Alvarez-Buylla, A. rons in the adult olfactory bulb for odor discrimination. Proc.
(1997). Cellular composition and three-dimensional organiza- Natl. Acad. Sci. USA 97:1823–1828.
tion of the subventricular germinal zone in the adult Girod, R., Barazangi, N., McGehee, D., and Role, L. W. (2000).
mammalian brain. J. Neurosci. 17:5046–5061. Facilitation of glutamatergic neurotransmission by presynap-
Doucette, R. (1991). PNS-CNS transition zone of the first cranial tic nicotinic acetylcholine receptors. Neuropharmacology
nerve. J. Comp. Neurol. 312:451–466. 39:2715–2725.
Anatomy and Neurochemistry of the Olfactory Bulb 157

Giustetto, M., Bovolin, P., Fasolo, A., Bonino, M., Cantino, D., Systems originating in the olfactory peduncle. J. Comp.
and Sassoè-Pognetto, M. (1997). Glutamate receptors in the Neurol. 181:781–808.
olfactory bulb synaptic circuitry: heterogeneity and synaptic Hagbarth, K. (1960). Centrifugal mechanisms of sensory control.
localization of N-methyl-D-aspartate receptor subunit 1 and Ergebn. Biol. 22:47–66.
AMPA receptor subunit 1. Neuroscience 76:787–798. Halabisky, B., Friedman, D., Radojicic, M., and Strowbridge, B. W.
Glusman, G., Bahar, A., Sharon, D., Pilpel, Y., White, J., and (2000). Calcium influx through NMDA receptors directly
Lancet, D. (2000). The olfactory receptor gene superfamily: evokes GABA release in olfactory bulb granule cells.
data mining, classification and nomenclature. Mamm. J. Neurosci. 20:5124–5134.
Genome 11:1016–1023. Halász, N. (1990). The Vertebrate Olfactory System. Chemical
Glusman, G., Yanai, I., Rubin, I., and Lancet, D. (2001). The Neuroanatomy, Function and Development. Akadémiai
complete human olfactory subgenome. Genome Res. Kiadó, Budapest.
11:685–702. Halász, N., and Greer, C. A. (1993). Terminal arborizations of
Godfrey, D. A., Ross, C. D., Carter, J. A., Lowry, O. H., and olfactory nerve fibers in the glomeruli of the olfactory bulb.
Matschinsky, F. M. (1980a). Effect of intervening lesions on J. Comp. Neurol. 337:307–316.
amino acid distributions in rat olfactory cortex and olfactory Halász, N., and Shepherd, G. M. (1983). Neurochemistry of the
bulb. J. Histochem. Cytochem. 28:1157–1169. vertebrate olfactory bulb. Neuroscience 10:579–619.
Godfrey, D. A., Ross, C. D., Herrmann, A. D., and Herrick, C. J. (1933). The function of the olfactory parts of the
Matschinsky, F. M. (1980b). Distribution and derivation of cerebral cortex. Proc. Natl. Acad. Sci. USA 19:7–14.
cholinergic elements in the rat olfactory bulb. Neuroscience Hsia, A. Y., Vincent, J. D., and Lledo, P. M. (1999). Dopamine
5:273–292. depresses synaptic inputs into the olfactory bulb. J. Neurophysiol.
Gogos, J. A., Osborne, J., Nemes, A., Mendelsohn, M., and 82:1082–1085.
Axel, R. (2000). Genetic ablation and restoration of the olfac- Isaacson, J. S. (1999). Glutamate spillover mediates excitatory
tory topographic map. Cell 103:609–620. transmission in the rat olfactory bulb. Neuron 23: 377–384.
Gray, E. G. (1969). Electron microscopy of excitatory and Isaacson, J. S. (2001). Mechanisms governing dendritic
inhibitory synapses. A brief review. Prog. Brain Res. gamma-aminobutyric acid (GABA) release in the rat
31:141–155. olfactory bulb. Proc. Natl. Acad. Sci. USA 98:337–342.
Graziadei, P. P. C., and Monti Graziadei, G. A. (1978). The olfac- Isaacson, J. S., and Strowbridge, B. W. (1998). Olfactory recip-
tory system: a model for the study of neurogenesis and axon rocal synapses: dendritic signaling in the CNS. Neuron
regeneration in mammals. In Neuronal Plasticity, C. W. 20:749–761.
Cotman (Ed.). Raven Press, New York, pp. 131–153. Jackowski, A., Parnavelas, J. G., and Lieberman, A. R. (1978).
Greer, C. A. (1987). Golgi analyses of dendritic organization The reciprocal synapse in the external plexiform layer of the
among denervated olfactory bulb granule cells. J. Comp. mammalian olfactory bulb. Brain Res. 159:17–28.
Neurol. 257:442–452. Jahr, C. E., and Nicoll, R. A. (1982). Noradrenergic modulation
Greer, C. A. and Halász, N. (1987). Plasticity of dendrodendr- of dendrodendritic inhibition in the olfactory bulb. Nature
itic microcircuits following mitral cell loss in the olfactory 297:227–229.
bulb of the murine mutant PCD. J. Comp. Neurol. Jankovski, A., Garcia, G., Soriano, E., and Sotelo, C. (1998).
256:284–298. Proliferation, migration and differentiation of neuronal prog-
Greer, C. A., and Shepherd, G. M. (1982). Mitral cell degenera- enitor cells in the adult mouse subventricular zone surgically
tion and sensory function on the neurological mutant mouse separated from its olfactory bulb. Eur. J. Neurosci.
Purkinje cell degeneration (PCD). Brain Res. 235:156–161. 10:3853–3868.
Greer, C. A., Stewart, W. B., Teicher, M. H., and Shepherd, G. M. Jastreboff, P. J., Pedersen, P. E., Greer, C. A., Stewart, W. B.,
(1982). Functional development of the olfactory bulb and a Kauer, J. S., Benson, T. E., and Shepherd, G. M. (1984).
unique glomerular complex in the neonatal rat. J. Neurosci. Specific olfactory receptor populations projecting to identi-
2:1744–1759. fied glomeruli in the rat olfactory bulb. Proc. Natl. Acad. Sci.
Guthrie, K. M., and Gall, C. M. (1995). Odors increase Fos in USA 81:5250–5254.
olfactory bulb neurons including dopaminergic cells. Jiang, M., Griff, E. R., Ennis, M., Zimmer, L. A., and Shipley, M. T.
NeuroReport 6:2145–2149. (1996). Activation of locus coeruleus enhances the responses
Guthrie, K. M., Anderson, A. J., Leon, M., and Gall, C. M. of olfactory bulb mitral cells to weak olfactory nerve input. J.
(1993). Odor-induced increases in c-fos mRNA expression Neurosci. 16:6319–6329.
reveal an anatomical “unit” for odor processing in olfactory Johnson, B. A., and Leon M. (1996). Spatial distribution of
bulb. Proc. Natl. Acad. Sci. USA 90:3329–3333. [14C]2-deoxyglucose uptake in the glomerular layer of the rat
Haberly, L. B., and Price, J. L. (1978a). Association and olfactory bulb following early odor preference learning.
commisural fiber systems of the olfactory cortex of the rat: J. Comp. Neurol. 376:557–566.
I. Systems originating in the piriform cortex and adjacent Johnson, B. A., and Leon M. (2000a). Molecular representations
areas. J. Comp. Neurol. 178:711–740. of odorants in the glomerular layer of the rat olfactory bulb
Haberly, L. B., and Price, J. L. (1978b). Association and comis- and the effects of stimulus concentration. J. Comp. Neurol.
sural fiber systems of the olfactory cortex of the rat: II. 422:496–509.
158 Kratskin and Belluzzi

Johnson, B. A., and Leon M. (2000b). Odorant molecular length: Keller, A., Yagodin, S., Aroniadou-Anderjaska, V., Zimmer, L.
one aspect of the olfactory code. J. Comp. Neurol. A., Ennis, M., Sheppard, N. F. Jr., and Shipley, M. T.
426:330–338. (1998). Functional organization of rat olfactory bulb
Johnson, B. A., Woo, C. C., and Leon M. (1998). Spatial coding glomeruli revealed by optical imaging. J. Neurosci. 18:
of odorant features in the glomerular layer of the rat olfactory 2602–2612.
bulb. J. Comp. Neurol. 393:457–471. Kelly, J. P., Wrynn, A. S., and Leonard, B. E. (1997). The olfac-
Johnson, B. A., Woo, C. C., Hingco, E. E., Pham, K. L., and Leon M. tory bulbectomized rat as a model of depression: an update.
(1999). Multidimensional chemotopic responses to n- Pharmacol. Ther. 74:299–316.
aliphatic acid odorants in the rat olfactory bulb. J. Comp. Kinzie, J. M., Shinohara, M. M., van den Pol, A. N., Westbrook, G. L.,
Neurol. 409:529–548. and Segerson, T. P. (1997). Immunolocalization of
Jourdan, F., Duveau, A., Astic, L., and Holley, A. (1980). Spatial metabotropic glutamate receptor 7 in the rat olfactory bulb.
distribution of [14C]2-deoxyglucose uptake in the olfactory J. Comp. Neurol. 385:372–384.
bulbs of rats stimulated with two different odours. Brain Res. Klenoff, J. R., and Greer, C. A. (1998). Postnatal development of
188:139–154. olfactory receptor cell axonal arbors. J. Comp. Neurol.
Kafitz, K. W., and Greer, C. A. (1998). Differential expression of 390:256–267.
extracellular matrix and cell adhesion molecules in the olfac- Kosaka, T., Hataguchi, Y., Hama, K., Nagatsu, I., and Wu, J.-Y.
tory nerve and glomerular layers of adult rats. J. Neurobiol. (1985). Coexistence of immunoreactivities for glutamate
34:271–282. decarboxylase and tyrosine hydroxylase in some neurons in
Kanaki, K., Kawashima, S., Kashiwayanagi, M., and Kurihara, K. the periglomerular region of the rat main olfactory bulb: pos-
(1997). Carnosine-induced inward currents in rat olfactory sible coexistence of gamma-aminobutyric acid (GABA) and
bulb neurons in cultured slices. Neurosci. Lett. 231: dopamine. Brain Res. 343:166–171.
167–170. Kosaka, T., Kosaka, K., Heizmann, C. W., Nagatsu, I., Wu, J.-Y.,
Kasa, P., Hlavati, I., Dobo, E., Wolff, A., Joo, F., and Wolff, J. R. Yanaihara, N., and Hama, K. (1987). An aspect of the organi-
(1995). Synaptic and non-synaptic cholinergic innervation of zation of the GABAergic system in the rat main olfactory
the various types of neurons in the main olfactory bulb of bulb: laminar distribution of immunocytochemically defined
adult rats: immunocytochemistry of choline acetyltransferase. subpopulations of GABAergic neurons. Brain Res.
Neuroscience 67:667–677. 411:373–378.
Kasa, P., Karcsu, S., Kovacs, I., and Wolff, J. R. (1996). Kosaka, K., Hama, K., Nagatsu, I., Wu, J.-Y., and Kosaka, T.
Cholinoceptive neurons without acetylcholinesterase activity (1988). Possible coexistence of amino acid (-amino-
and enzyme-positive neurons without cholinergic synaptic butyric acid), amine (dopamine) and peptide (substance
innervation are present in the main olfactory bulb of adult rats. P); neurons containing immunoreactivities for glutamic
Neuroscience 73:831–844. acid decarboxylase, tyrosine hydroxylase and substance P
Kasa, P., Raconczay, Z., and Gulya, K. (1997). The cholinergic in the hamster main olfactory bulb. Exp. Brain Res.
system in Alzheimer’s disease. Prog. Neurobiol. 71:633–642.
52:511–535. Kosaka, K., Heizmann, C. W., and Kosaka, T. (1994). Calcium-
Kashiwadani, H., Sasaki, Y. F., Uchida, N., and Mori, K. (1999). binding protein parvalabumin-immunoreactive neurons in the
Synchronized oscillatory discharges of mitral/tufted cells with rat olfactory bulb. 1. Distribution and structural features in
different molecular receptive ranges in the rabbit olfactory adult rat. Exp. Brain Res. 99:191–204.
bulb. J. Neurophysiol. 82:1786–1792. Kosaka, K., Aika, Y., Toida, K., Heizmann, C. W., Hunziker, W.,
Kasowski, H. J., Kim, H., and Greer, C. A. (1999). Jacobowitz, D. M., Nagatsu, I., Streit, P., Visser, T. J., and
Compartmental organization of the olfactory bulb glomerulus. Kosaka, T. (1995). Chemically defined neuron groups and
J. Comp. Neurol. 407:261–274. their subpopulations in the glomerular layer of the rat main
Kauer, J. S. (1991). Contributions of topography and parallel pro- olfactory bulb. Neurosci. Res. 23:73–88.
cessing to odor coding in the vertebrate olfactory pathway. Kosaka, K., Toida, K., Margolis, F. L., and Kosaka, T. (1997).
Trends Neurosci. 14:79–85. Chemically defined neuron groups and their subpopulations
Kauer, J. S., and Cinelli, A. R. (1993). Are there structural and in the glomerular layer of the rat main olfactory bulb. II.
functional modules in the vertebrate olfactory bulb? Microsc. Prominent differences in the intraglomerular dendritic
Res. Tech. 24:157–163. arborization and their relationship to olfactory nerve termi-
Kauer, J. S., and Moulton, D. G. (1974). Responses of olfactory nals. Neuroscience 76:775–786.
bulb neurons to odor stimulation of small nasal areas in the Koster, N. L., Norman, A. B., Richtand, N. M., Nickell, W. T.,
salamander. J. Physiol (Lond.) 243:717–737. Puche, A. C., Pixley, S. K., and Shipley, M. T. (1999).
Kauer, J. S., Neff, S. R., Hamilton, K. A., and Cinelli, A. R. Olfactory receptor neurons express D2 dopamine receptors.
(1991). The salamander olfactory pathway: visualizing and J. Comp. Neurol. 411:666–673.
modeling circuit activity. In Olfaction: A Model System for Kratskin, I. L. (1987). Centrifugal innervation of the main olfac-
Computational Neuroscience, J. L. Davis and H. tory bulb in vertebrates. In Systems of Sense Organs.
Eichenbaum (Eds.). MIT Press, Cambridge, MA, pp. Morphological and Functional Aspects of Evolution, G. V.
43–68. Gershuni (Ed.). Nauka, Leningrad, pp. 101–120.
Anatomy and Neurochemistry of the Olfactory Bulb 159

Kratskin, I. L. (1989). Neurotransmitters of centrifugal innerva- Le Jeune, H., and Jourdan, F. (1994). Acetylcholinesterase-
tion of the main olfactory bulb. J. Evol. Biochem. Physiol. containing intrinsic neurons in the rat main olfactory bulb:
25:174–187. cytological and neurochemical features. Eur. J. Neurosci.
Kratskin, I. L., and Hao, Y. P. (2001). Expression of a rate-limit- 6:1432–1444.
ing enzyme of taurine biosynthesis in the rat olfactory Le Jeune, H., Aubert, I., Jourdan, F., and Quirion, R. (1995).
mucosa. Soc. Neurosci. Abstr. 27:163. Comparative laminar distribution of various autoradiographic
Kratskin, I. L., and Ragimova, N. G. (1985). Sources of centrifu- cholinergic markers in adult rat main olfactory bulb. J. Chem.
gal innervation of the frog olfactory bulb. A combined histo- Neuroanat. 9:99–112.
chemical study. Doklady Acad. Nauk SSSR 282:747–749. Le Jeune, H., Aubert, I., Jourdan, F., and Quirion, R. (1996).
Kratskin, I. L., and Yu, X. (1996a). Different parts of the diago- Developmental profiles of various cholinergic markers in the
nal band nucleus project to different layers of the olfactory rat main olfactory bulb using quantitative autoradiography.
bulb in the rat. Chem. Senses 21:626–627. J. Comp. Neurol. 373:433–450.
Kratskin, I. L., and Yu, X. (1996b). Taurine in the olfactory bulb. Leveteau, J., and MacLeod, P. (1966). Olfactory discrimination in
Immunocytochemical localization. Soc. Neurosci. Abstr. the rabbit olfactory glomerulus. Science 175:170–178.
22:2021. Linster, C., and Gervais, R. (1996). Investigation of the role of
Kratskin, I. L., Kenigfest, N. B., Repérant, J., and Vesselkin, N. interneurons and their modulation by centrifugal fibers in a
P. (1991). Centrifugal innervation of the frog main olfactory neural model of the olfactory bulb. J. Comput. Neurosci.
bulb by GABA-immunoreactive cells. Doklady Acad. Nauk 3:225–246.
SSSR 319:254–256. Linster, C., and Hasselmo, M. (1997). Modulation of inhibition in
Kratskin, I., Kenigfest, N., Repérant, J., Rio, J. P., and Vesselkin, a model of olfactory bulb reduces overlap in the neural repre-
N. (1992). Presumptive GABAergic feedback input to the frog sentation of olfactory stimuli. Behav. Brain Res. 84:117–127.
olfactory bulb: a double labeling study with retrograde axon- Linster, C., and Hasselmo, M. E. (2000). Neural activity in the
al tracing and GABA immunohistochemistry. Brain Res. horizontal limb of the diagonal band of Broca can be modu-
581:171–174. lated by electrical stimulation of the olfactory bulb and cortex
Kratskin, I. L., Rio, J. P., Kenigfest, N. B., Doty, R. L., and in rats. Neurosci. Lett. 282:157–160.
Repérant, J. (2000). A light and electron microscopic study of Liu, C.-J., Grandes, P., Matute, C., Cuénod, M., and Streit, P.
taurine-like immunoreactivity in the main olfactory bulb of (1989). Glutamate-like immunoreactivity revealed in rat
frogs. J. Chem. Neuroanat. 18:87–101. olfactory bulb, hippocampus and cerebellum by monoclonal
Kratskin, I. L., Yu, X., Suo, L., and Smutzer, G. S. (1997). GABA- antibody and sensitive staining method. Histochemistry
immunoreactive neurons in the medial amygdala project to the 90:427–445.
salamander olfactory bulb. Chem. Senses 22:724–725. Liu, W. L., and Shipley, M. T. (1994). Intrabulbar associational
Kunze, W. A. A., Shafton, A. D., Kemm, R. E., and McKenzie, J. S. system in the rat olfactory bulb comprises cholecys-
(1991). Effect of stimulating the nucleus of the horizontal tokinin-containing tufted cells that synapse onto the
limb of the diagonal band on single unit activity in the olfac- dendrites of GABAergic granule cells. J. Comp. Neurol.
tory bulb. Neuroscience 40:21–27. 346:541–558.
Kunze, W. A. A., Shafton, A. D., Kemm, R. E., and McKenzie, J. S. Lois, C., Garcia-Verdugo, J. M., and Alvarez-Buylla, A. (1996).
(1992a). Olfactory bulb output neurons excited from a basal Chain migration of neuronal precursors. Science 271:
forebrain magnocellular nucleus. Brain Res. 583:327–331. 979–981.
Kunze, W. A. A., Shafton, A. D., Kemm, R. E., and McKenzie, J. S. Lu, X. C., and Slotnick, B. M. (1999). Olfaction in rats with
(1992b). Intracellular responses of olfactory bulb granule extensive lesions of the olfactory bulbs: implications for odor
cells to stimulating the horizontal diagonal band nucleus. coding. Neuroscience 84:849–866.
Neuroscience 48:363–369. Luskin, M. B. (1993). Restricted proliferation and migration of
Lancet, D., and Ben-Arie, N. (1993). Olfactory receptors. Curr. postnatally generated neurons derived from the forebrain sub-
Biol. 3:668–674. ventricular zone. Neuron 11:173–189.
Lancet, D., Greer, C. A., Kauer, J. S., and Shepherd, G. M. Luskin, M. B. (1998). Neuroblasts of the postnatal mammalian
(1982). Mapping of odor-related neuronal activity in the forebrain: their phenotype and fate. J. Neurobiol. 36:221–233.
olfactory bulb by high-resolution 2-deoxyglucose autoradiog- Luskin, M. B., and Price, J. L. (1982). The distribution of axon
raphy. Proc. Natl. Acad. Sci. USA 79:670–674. collaterals from the olfactory bulb and the nucleus of the hor-
Lanuza, E., and Halpern, M. (1998). Efferent and centrifugal izontal limb of the diagonal band to the olfactory cortex,
afferents to the main and accessory olfactory bulb in the snake demonstrated by double retrograde labeling techniques.
Thamnophis sirtalis. Brain Behav. Evol. 51:1–22. J. Comp. Neurol. 209:249–263.
Laurie, D. J., Seeburg, P. H., and Wisden, W. (1992). The distrib- Luskin, M. B., and Price, J. L. (1983). The topographic organiza-
ution of 13 GABAA receptor subunit mRNAs in the rat brain. tion of associational fibers of the olfactory system in the rat,
II. Olfactory bulb and cerebellum. J. Neurosci. 12:1063–1076. including centrifugal fibers to the olfactory bulb. J. Comp.
Le Jeune, H., and Jourdan, F. (1993). Cholinergic innervation of Neurol. 216:264–291.
olfactory glomeruli in the rat: An ultrastructural immunocyto- Mackay-Sim, A., and Chuah, M. I. (2000). Neurtrophic factors in
chemical study. J. Comp. Neurol. 336:279–292. the primary olfactory pathway. Prog. Neurobiol. 62:527–559.
160 Kratskin and Belluzzi

Macrides, F., and Davis, B. J. (1983). The olfactory bulb. In a search for the accessory olfactory bulb. Ann. NY Acad. Sci.
Chemical Neuroanatomy, P. C. Emson (Ed.). Raven Press, 855:708–715.
New York, pp. 391–426. Meister, M., and Bonhoeffer, T. (2001). Tuning and topography in an
Macrides, F., and Schneider, S. P. (1982). Laminar organization odor map on the rat olfactory bulb. J. Neurosci. 21:1351–1360.
of mitral and tufted cells in the main olfactory bulb of the Mombaerts, P. (1999). Molecular biology of odorant receptors in
adult hamster. J. Comp. Neurol. 208:419–430. vertebrates. Annu. Rev. Neurosci. 22:487–509.
Macrides, F., Davis, B. J., Youngs, W. M., Nadi, N. S., and Mombaerts, P., Wang, F., Dulac, C., Chao, S. K., Nemes, A.,
Margolis, F. L. (1981). Cholinergic and catecholaminergic Mendelsohn, M., Edmondson, J., and Axel, R. (1996).
afferents to the olfactory bulb in the hamster: a neuroanatom- Visualizing an olfactory sensory map. Cell 87:675–686.
ical, biochemical and histochemical investigation. J. Comp. Montague, A. A., and Greer, C. A. (1999). Differential distribu-
Neurol. 203:495–514. tion of ionotropic glutamate receptor subunits in the rat olfac-
Macrides, F., Schoenfeld, T. A., Marchand, J. E., and Clancy, A. N. tory bulb. J. Comp. Neurol. 405:233–246.
(1985). Evidence for morphologically, neurochemically and Moran, D. T., Rowley, J. C., III, Jafek, B. W., and Lowell, M. A.
functionally heterogeneous classes of mitral and tufted cells (1982). The fine structure of the olfactory mucosa in man.
in the olfactory bulb. Chem. Senses 10:175–202. J. Neurocytol. 11:721–746.
Margolis, F. L., Kawano, T., and Grillo, M. (1986). Ontogeny of Mori, K. (1987). Membrane and synaptic properties of identified
carnosine, olfactory marker protein and neurotransmitter neurons in the olfactory bulb. Progr. Neurobiol. 29:275–320.
enzymes in olfactory bulb and olfactory mucosa of the rat. In Mori, K., and Kishi, K. (1982). The morphology and physiology of
Ontogeny of Olfaction, W. Breipohl (Ed.). Springer-Verlag, the granule cells in the rabbit olfactory bulb revealed by intra-
Berlin, pp. 107–116. cellular recording and HRP injection. Brain Res. 247:129–133.
McLean, J. H., and Shipley, M. T. (1987a). Serotonergic affer- Mori, K., and Yoshihara, Y. (1995). Molecular recognition and
ents to the rat olfactory bulb: I. Origins and laminar speci- olfactory processing in the mammalian olfactory system.
ficity of serotonergic inputs in the adult rat. J. Neurosci. Prog. Neurobiol. 45:585–619.
7:3016–3028. Mori, K., Kishi, K., and Ojima, H. (1983). Distribution of den-
McLean, J. H., and Shipley, M. T. (1987b). Serotonergic afferents drites of mitral, displaced mitral, tufted, and granule cells in
to the rat olfactory bulb: II. Changes in fiber distribution dur- the rabbit olfactory bulb. J. Comp. Neurol. 219:339–355.
ing development. J. Neurosci. 7:3029–3039. Mori, K., Nagao, H., and Yoshihara, Y. (1999). The olfactory
McLean, J. H., and Shipley, M. T. (1991). Postnatal development bulb: coding and processing of odor molecule information.
of the noradrenergic projection from the locus coeruleus to the Science 286:711–715.
olfactory bulb in the rat. J. Comp. Neurol. 304:467–477. Mori, K., von Campenhause, H., and Yoshihara, Y. (2000). Zonal
McLean, J. H., Shipley, M. T., Nickell, W. T., Aston-Jones, G., organization of the mammalian main and accessory olfactory
and Reyher, C. K. H. (1989). Chemoanatomical organiza- systems. Phil. Trans. R. Soc. Lond. Ser. B 355:1801–1812.
tion of the noradrenergic input from locus coeruleus to the Moriizumi, T., Tsukatani, T., Sakashita, G., and Miwa, T. (1994).
olfactory bulb of the adult rat. J. Comp. Neurol. 285: Olfactory disturbance induced by deafferentation of seroton-
339–349. ergic fibers in the olfactory bulb. Neuroscience 61:733–738.
McLean, J. H., Darby-King, A., Sullivan, R. M., and King, S. R. Morilak, D. A., Garlow, S. J., and Ciaranello, R. D. (1993).
(1993). Serotonergic influence on olfactory learning in the Immunocytochemical localization and description of neurons
neonate rat. Behav. Neural Biol. 60:152–162. expressing serotonin2 receptors in the rat brain. Neuroscience
McLean, J. H., Darby-King, A., and Paterno, G. D. (1995). 54:701–717.
Localization of 5-HT2A receptor mRNA by in situ Mugnaini, E., Oertel, W. H., and Wouterflood, F. F. (1984).
hybridization in the olfactory bulb of the postnatal rat. J. Immunocytochemical localization of GABA neurons and
Comp. Neurol. 353:371–378. dopamine neurons in the rat main and accessory olfactory
McLean, S., Rothman, R. B., and Herkenham, M. (1986). bulbs. Neurosci. Lett. 47:221–226.
Autoradiographic localization of - and -opiate receptors in Nagao, H., Yoshihara, Y., Mitsui, S., Fujisawa, H., and Mori, K.
the forebrain of the rat. Brain Res. 378:49–60. (2000). Two mirror-image sensory maps within domain orga-
McLennan, H. (1971). The pharmacology of inhibition of mitral nization in the mouse main olfactory bulb. NeuroReport
cells in the olfactory bulb. Brain Res. 29:177–184. 11:3023–3027.
Meisami, E., and Safari, L. (1981). A quantitative study of the Nickell, W. T., and Shipley, M. T. (1988a). Two anatomically spe-
effects of early unilateral olfactory deprivation on the number cific classes of candidate cholinoceptive neurons in the rat
and distribution of mitral and tufted cells and of glomeruli in olfactory bulb. J. Neurosci. 8:4482–4491.
the rat olfactory bulb. Brain Res. 221:81–107. Nickell, W. T., and Shipley, M. T. (1988b). Neurophysiology of
Meisami, E., and Sendera, T. J. (1993). Morphometry of rat olfac- magnocellular forebrain inputs to the olfactory bulb in the rat:
tory bulbs stained for cytochrome oxidase reveals that the frequency potentiation of field potentials and inhibition of
entire population of glomeruli forms early in the neonatal output neurons. J. Neurosci. 8:4492–4502.
period. Dev. Brain Res. 71:253–257. Nickell, W. T., Norman, A. B., Wyatt, L. M., and Shipley, M. T.
Meisami, E., Mikhail, L., Baim, D., and Bhatnagar, K. P. (1998). (1991). Olfactory bulb DA receptors may be located on ter-
Human olfactory bulb: aging of glomeruli and mitral cells and minals of the olfactory nerve. NeuroReport 2:9–12.
Anatomy and Neurochemistry of the Olfactory Bulb 161

Nickell, W. T., Behbehani, M. M., and Shipley, M. T. (1994). Paternostro, M. A., and Meisami, E. (1996a). Essential role of
Evidence for GABAB-mediated inhibition of transmission thyroid hormones in maturation of olfactory receptor neurons:
from the olfactory nerve to mitral cells in the rat olfactory an immunocytochemical study of number and cytoarchitec-
bulb. Brain Res. Bull. 35:119–123. ture of OMP-immunopositive cells in developing rats. Int.
Nieuwenhuys, R. (1967). Comparative anatomy of olfactory cen- J. Dev. Neurosci. 14:867–880.
tres and tracts. Prog. Brain Res. 23:1–64. Paternostro, M. A., and Meisami, E. (1996b). Marked restora-
Oh, J. D., Wolff, N. J., Roghani, A., Edwards, R. H., and Butcher, L. L. tion of density and number of mature (knob-bearing)
(1992). Cholinergic neurons in the rat central nervous system olfactory receptor neurons in rats recovering from
demonstrated by in situ hybridization of choline acetyltrans- hypothyroid-induced growth retardation. Dev. Brain Res.
ferase mRNA. Neuroscience 47:807–822. 96:173–183.
Ohm, T. G., Braak, E., and Probst, A. (1988). Somatostatin- Persohn, E., Malherbe, P., and Richards, J. G. (1992).
14-like immunoreactive neurons and fibres in the human Comparative molecular neuroanatomy of cloned GABAA
olfactory bulb. Anat. Embryol. 179:165–172. receptor subunits in the rat CNS. J. Comp. Neurol.
Ohm, T. G., Müller, H., Ulfig, N., and Braak, E. (1990). Glutamic- 326:193–216.
acid-decarboxylase- and parvalbumin-like-immunoreactive Phelps, P. E., Houser, C. R., and Vaughn, J. E. (1992). Small
structures in the olfactory bulb of the human adult. J. Comp. cholinergic neurons within fields of cholinergic axons charac-
Neurol. 291:1–8. terize olfactory-related regions of rat telencephalon.
Oja, S. S., and Saransaari, P. (2000). Modulation of taurine Neuroscience 48:121–136.
release by glutamate receptors and nitric oxide. Prog. Philipot, B. D., Men, D., McCarty, R., and Brunjes, P. C. (1998).
Neurobiol. 62:407–425. Activity-dependent regulation of dopamine content in the
Ojima, H., Yamasaki, T., Kojima, H., and Akashi, A. (1988). olfactory bulbs of naris-occluded rats. Neuroscience
Cholinergic innervation of the main and accessory 85:969–977.
olfactory bulbs of the rat as revealed by a monoclonal anti- Pinching, A. J., and Powell, T. P. S. (1971a). The neuron types of
body against choline acetyltransferase. Anat. Embryol. the glomerular layer of the olfactory bulb. J. Cell Sci. 9:
178:481–488. 305–345.
Okoyama, S., Tago, H., McGeer, P. L., Hersh, L. B., and Kimura, Pinching, A. J., and Powell, T. P. S. (1971b). The neuropil of the
H. (1987). Cholinergic divergent projections from rat basal glomeruli of the olfactory bulb. J. Cell Sci. 9:347–377.
forebrain to the hippocampus and olfactory bulb. Neurosci. Pinching, A. J., and Powell, T. P. S. (1971c). The neuropil of the
Lett. 83:77–81. periglomerular region of the olfactory bulb. J. Cell Sci.
Onoda, N. (1992). Odor-induced fos-like immunoreactivity in the 9:379–409.
rat olfactory bulb. Neurosci. Lett. 137:157–160. Pinching, A. J., and Powell, T. P. S. (1972). The termination of
Orona, E., Scott, J. W., and Rainer, E. C. (1983). Different gran- centrifugal fibers in the glomerular layer of the olfactory bulb.
ule cell populations innervate superficial and deep regions of J. Cell Sci. 10:621–635.
the external plexiform layer in rat olfactory bulb. J. Comp. Pirker, S., Schwarzer, C., Wieselthaler, A., Sieghart, W., and
Neurol. 217:227–237. Sperk, G. (2000). GABAA receptors: immunocytochemical
Orona, E., Rainer, E. C., and Scott, J. W. (1984). Dendritic and distribution of 13 subunits in the adult rat brain. Neuroscience
axonal organization of mitral and tufted cells in the rat olfac- 101:815–850.
tory bulb. J. Comp. Neurol. 226:346–356. Pompeiano, M., Palacios, J. M., and Mengod, G. (1994).
Ottersen, O. P., and Storm-Mathisen, J. (1984). Glutamate- and Distribution of the serotonin 5-HT2 receptor family mRNAs:
GABA-containing neurons in the mouse and rat brain, as Comparison between 5-HT2A and 5-HT2C receptors. Mol.
demonstrated with a new immunocytochemical technique. Brain Res. 23:163–178.
J. Comp. Neurol. 229:374–392. Potapov, A. A. (1985). Baclofen inhibition of synaptic transmis-
Ottoson, D., and Shepherd, G. M. (1967). Experiments and sion in frog olfactory bulb glomeruli. Neirofiziologia
concepts in olfactory physiology. Prog. Brain Res. 17:834–837.
23:114–138. Prasada Rao, P. D., and Finger, T. E. (1984). Asymmetry of the
Paolini, A. G., and McKenzie, J. S. (1993). Effects of lesions in olfactory system in the brain of the winter flounder,
the horizontal diagonal band nucleus on olfactory habituation Pseudopleuronectes americanus. J. Comp. Neurol. 225:
in the rat. Neuroscience 57:717–724. 492–510.
Paolini, A. G., and McKenzie, J. S. (1996). Lesions in the mag- Price, J. L. (1968). The termination of centrifugal fibres in the
nocellular preoptic nucleus decrease olfactory investigation in olfactory bulb. Brain Res. 7:483–486.
rats. Behav. Brain Res. 81:223–231. Price, J. L. (1969). The origin of the centrifugal fibres to the
Paolini, A. G., and McKenzie, J. S. (1997a). Intracellular record- olfactory bulb. Brain Res. 14:542–545.
ing of magnocellular preoptic neuron responses to olfactory Price, J. L., and Powell, T. P. S. (1970a). The morphology of the
brain. Neuroscience 78:229–242. granule cells of the olfactory bulb. J. Cell Sci. 7:91–123.
Paolini, A. G., and McKenzie, J. S. (1997b). Effects of inactiva- Price, J. L., and Powell, T. P. S. (1970b). The synaptology
tion of the magnocellular preoptic nucleus on olfactory bulb of the granule cells of the olfactory bulb. J. Cell Sci.
processing. NeuroReport 8:929–935. 7:125–155.
162 Kratskin and Belluzzi

Price, J. L., and Powell, T. P. S. (1970c). An electron-microscopic on neophobia and learned aversion to a novel food. Behav.
study of the termination of the afferent fibres to the olfactory Brain. Res. 10:297–309.
bulb from the cerebral hemisphere. J. Cell Sci. 7:157–187. Royet, J. -P., Sicard, G., Souchier, C., and Jourdan, F. (1987).
Price, J. L., and Powell, T. P. S. (1970d). The mitral and short Specificity of spatial patterns of glomerular activation in the
axon cells of the olfactory bulb. J. Cell Sci. 7:631–651. mouse olfactory bulb: computer-assisted image analysis of
Price, J. L., and Powell, T. P. S. (1970e). An experimental study 2-deoxyglucose autoradiograms. Brain Res. 417:1–11.
of the origin and the course of the centrifugal fibres to the Rubin, B. D., and Katz, L. C. (1999). Optical imaging of odorant
olfactory bulb in the rat. J. Anat. 107:215–237. representations in the mammalian olfactory bulb. Neuron
Price, J. L., and Powell, T. P. S. (1970f). The afferent connexions 23:499–511.
of the nucleus of the horizontal limb of the diagonal band. Rubin, B. D., and Katz, L. C. (2001). Spatial coding of enan-
J. Anat. 107:239–256. tiomers in the rat olfactory bulb. Nature Neurosci. 4:355–366.
Puche, A. C., and Shipley, M. T. (1999). Odor-induced, activity- Salin, P.-A., Lledo, P. -M., Vincent, J.-D., and Charpak, S. (2001).
dependent transneuronal gene induction in vitro: mediation by Dendritic glutamate autoreceptors modulate signal processing
NMDA receptors. J. Neurosci. 19:1359–1370. in rat mitral cells. J. Neurophysiol. 85:1275–1282.
Puopolo, M., and Belluzzi, O. (1998a). Functional heterogeneity Sallaz, M., and Jourdan, F. (1993). C-fos expression and 2-deoxyglu-
of periglomerular cells in the rat olfactory bulb. Eur. J. cose uptake in the olfactory bulb of odour-stimulated awake
Neurosci. 10:1073–1083. rats. Neuro Report 4:55–58.
Puopolo, M., and Belluzzi, O. (1998b). Inhibitory synapses Sallaz, M., and Jourdan, F. (1996). Odour-induced c-fos expres-
among interneurons in the glomerular layer of rat and frog sion in the rat olfactory bulb: involvement of centrifugal affer-
olfactory bulbs. J. Neurophysiol. 80:344–349. ents. Brain Res. 721:66–75.
Puopolo, M., Kratskin, I., and Belluzzi, O. (1998). Direct Salmoiraghi, G. C., Bloom, F. E., and Costa, E. (1964).
inhibitory effect of taurine on relay neurons of the rat olfac- Adrenergic mechanisms in rabbit olfactory bulb. Am. J.
tory bulb in vitro. NeuroReport 9:2319–2323. Physiol. 207:1417–1424.
Rall, W., Shepherd, G. M., Reese, T. S., and Brightman, M. W. Sassoè-Pognetto, M., and Ottersen, O. P. (2000). Organization of
(1966). Dendrodenritic synaptic pathway for inhibition in the ionotropic glutamate receptors at dendrodendritic synapses in
olfactory bulb. Exp. Neurol. 14:44–56. the rat olfactory bulb. J. Neurosci. 20:2192–2201.
Ramón y Cajal S. (1911). Histologie du Système Nerveux de Sassoè-Pognetto, M., Cantino, D., Panzanelli, P., Verdun di
l’Homme et des Vertébrés, Vol. 2. A. Maloine, Paris. Cantogno, L., Giustetto, M., Margolis, F. L., De Bias, S., and
Ravel, N., Akaoka, H., Gervais, R., and Chouvet, G. (1990). The Fasolo, A. (1993). Presynaptic co-localization of carnosine
effect of acetylcholine on rat olfactory bulb unit activity. and glutamate in olfactory neurones. NeuroReport 5:7–10.
Brain Res. Bull. 24:151–155. Schneider, S. P., and Macrides, F. (1978). Laminar distribution of
Ressler, K. J., Sullivan, S. L., and Buck, L. B. (1994a). A molec- interneurons in the main olfactory bulb of the adult hamster.
ular dissection of spatial patterning in the olfactory system. Brain Res. Bull. 3:73–82.
Curr. Opin. Neurobiol. 4:588–596. Schoenfeld, T. A., and Macrides, F. (1984). Topographic organi-
Ressler, K. J., Sullivan, S. L., and Buck, L. B. (1994b). zation of connections between the main olfactory bulb and
Information coding in the olfactory system: evidence for a pars externa of the anterior olfactory nucleus in the hamster.
stereotyped and highly organized epitope map in the olfactory J. Comp. Neurol. 227:121–135.
bulb. Cell 79:1245–1255. Schoenfeld, T. A., Marchand, J. E., and Macrides, F. (1985).
Reyher, C. K., Lubke, J., Larsen, W. J., Hendrix, G. M., Shipley, Topographic organization of tufted cell axonal projections in
M. T., and Baumgarten, H. G. (1991). Olfactory bulb granule the hamster main olfactory bulb: An intrabulbar associational
cell aggregates: morphological evidence for interperikaryal system. J. Comp. Neurol. 235:503–518.
electrotonic coupling via gap junctions. J. Neurosci. Schoenfeld, T. A., Clancy, A. N., Forbes, W. B., and Macrides, F.
11:1485–1495. (1994). The spatial organization of the peripheral olfactory
Rhoades, B. K. and Freeman, W. J. (1990). Excitatory actions of system of the hamster. Part I: Receptor neuron projections to
GABA in the rat olfactory bulb. Soc. Neurosci. Abstr. 16:403. the main olfactory bulb. Brain Res. Bull. 34:183–210.
Ring, G., Mezza, R. C., and Schwob, J. E. (1997). Schoppa, N. E., and Westbrook, G. L. (1997). Modulation of
Immunohistochemical identification of discrete subsets of rat mEPSCs in olfactory bulb mitral cells by metabotropic gluta-
olfactory neurons and the glomeruli that they innervate. J. mate receptors. J. Neurophysiol. 78:1468–1475.
Comp. Neurol. 388:415–434. Schoppa, N. E., Kinzie, J. M., Sahara, Y., Segerson, T. P., and
Rotter, A., Birsdal, N. J. M., Burgen, A. S. V., Field, P. M., Westbrook, G. L. (1998). Dendrodendritic inhibition in the
Hulme, E. C., and Raisman, G. (1979). Muscarinic receptors olfactory bulb is driven by NMDA receptors. J. Neurosci.
of the central nervous system of the rat. I. Technique for 18:6790–6802.
autoradiographic localization of the binding of (3H)propyl- Schwob, J. E., and Gottlieb, D. I. (1986). The primary olfactory
benzilylcholine mustard and its distribution in the forebrain. projection has two chemically distinct zones. J. Neurosci.
Brain Res. Rev. 1:141–165. 6:3393–3404.
Royet, J.-P., Gervais, R., and Araneda, S. (1983). Effects of local Schwob, J. E., and Gottlieb, D. I. (1988). Purification
6-OHDA and 5,6-DHT injections into the rat olfactory bulb and characterization of an antigen that is spatially
Anatomy and Neurochemistry of the Olfactory Bulb 163

segregated in the primary olfactory projection. J. Neurosci. by the 2-deoxyglucose method. J. Comp. Neurol.
8:3470–3480. 185:715–734.
Schwob, J. E. and Price, J. L. (1984). The development of axon- Struble, R. G., and Walters, C. P. (1982). Light microscopic dif-
al connections in the central olfactory system of rats. J. Comp. ferentiation of two populations of rat olfactory bulb granule
Neurol. 223:177–202. cells. Brain Res. 236:237–251.
Scott, J. W., and Harrison, T. A. (1987). The olfactory bulb: Sullivan, R. M., Wilson, D. A., Lemon, C., and Gerhardt, G. A.
anatomy and physiology. In Neurobiology of Taste and Smell, (1993). Bilateral 6-OHDA lesions of the locus coeruleus
T. E. Finger and W. L. Silver (Eds.). John Wiley & Sons, New impair associative olfactory learning in newborn rats. Brain
York, pp. 151–178. Res. 643:306–309.
Scott, J., Rainer, E., Pemberton, J., Orona, E., and Moudrian, L. Sullivan, R. M., Stackenwalt, G., Nasr, F., Lemon, C., and
(1985). Pattern of olfactory bulb mitral and tufted cell con- Wilson, D. A. (2000). Association of an odor with activation
nections to the anterior olfactory nucleus of the rat. J. Comp. of olfactory bulb noradrenergic -receptors or locus coeruleus
Neurol. 242:415–424. stimulation is sufficient to produce learned approach respons-
Sharp, F. R., Kauer, J. S., and Shepherd, G. M. (1975). Local sites es to that odor in neonatal rats. Behav. Neurosci.
of activity-related glucose metabolism in rat olfactory bulb 114:957–962.
during olfactory stimulation. Brain Res. 98:596–600. Sullivan, S. L., Ressler, K. J., and Buck, L. B. (1995). Spatial pat-
Shepherd, G. M. (1991). Computational structure of the olfac- terning and information coding in the olfactory system. Curr.
tory system. In Olfaction: A Model System for Computational Opin. Genet. Dev. 5:516–523.
Neuroscience, J. L. Davis and H. Eichenbaum (Eds.). MIT Teicher, M. H., Stewart, W. B., Kauer, J. S., and Shepherd, G. M.
Press, Cambridge, MA, pp. 3–42. (1980). Suckling pheromone stimulation of a modified
Shepherd, G. M., and Greer, C. A. (1998). Olfactory bulb. In The glomerular region in the developing rat olfactory bulb
Synaptic Organization of the Brain, G. M. Shepherd (Ed.). revealed by the 2-deoxyglucose method. Brain Res.
Oxford University Press, New York, pp. 159–203. 194:530–535.
Shepherd, G. M., and Koch, C. (1998). Introduction to synaptic Toida, K., Kosaka, K., Hezmann, C. W., and Kosaka, T. (1994).
circuits. In The Synaptic Organization of the Brain, G. M. Synaptic contacts between mitral/tufted cells and GABAergic
Shepherd (Ed.). Oxford University Press, New York, pp. 1–36. neurons containing calcium-binding protein parvalbumin in
Shepherd, G. M., Nowycky, M. C., Greer, C. A., and Mori, K. the rat olfactory bulb, with special reference to reciprocal
(1981). Multiple overlapping circuits within olfactory and synapses between them. Brain Res. 650:347–352.
basal forebrain systems. Adv. Physiol. Sci. 30:263–278. Toida, K., Kosaka, K., Hezmann, C. W., and Kosaka, T. (1996).
Shipley, M. T., and Adamek, J. P. (1984). The connections of the Electron microscopic serial-sectioning/reconstruction study
mouse olfactory bulb: a study using orthograde and retrograde of parvalbumin-containing neurons in the external plexiform
transport of wheat germ agglutinin conjugated to horseradish layer of the rat olfactory bulb. Neuroscience 72:449–466.
peroxidase. Brain Res. Bull. 12:669–687. Toida, K., Kosaka, K., Hezmann, C. W., and Kosaka, T. (1998).
Shipley, M. T., and Ennis, M. (1996). Functional organization of Chemically defined neuron groups and their subpopulations
olfactory system. J. Neurobiol. 30:123–176. in the glomerular layer of the rat main olfactory bulb. III.
Shipley, M. T., Halloran, F. J., and de la Torre, J. (1985). Structural features of calbindin D28K-immunoreactive neu-
Surprisingly rich projection from locus coeruleus to the olfac- rons. J. Comp. Neurol. 392:179–198.
tory bulb in the rat. Brain Res. 329:294–299. Toida, K., Kosaka, K., Aika, Y., and Kosaka, T. (2000).
Siklos, L., Rickmann, M. J., Freeman, W. J., and Wolff, J. R. Chemically defined neuron groups and their subpopulations
(1995). Chloride is preferentially accumulated in a subpopu- in the glomerular layer of the rat main olfactory bulb. IV.
lation of dendrites and periglomerular cells of the main olfac- Intraglomerular synapses of tyrosine hydroxylase-immunore-
tory bulb in adult rats. Neuroscience 64:165–172. active neurons. Neuroscience 101:11–17.
Singer, M. S., Shepherd, G. M., and Greer, C. A. (1995). Trombley, P. Q. (1992). Norepinephrine inhibits calcium currents
Olfactory receptors guide axons. Nature 377:19–20. and EPSPs via a G-protein-coupled mechanism in olfactory
Slotnick, B. M., Graham, S., Laing, D. G., and Bell, G. A. (1987). bulb neurons. J. Neurosci. 12:3992–3998.
Detection of propionic acid vapor by rats with lesions of Trombley, P. Q., and Shepherd, G. M. (1992). Noradrenergic
olfactory bulb areas associated with high 2-DG uptake. Brain inhibition of synaptic transmission between mitral and gran-
Res. 417:343–346. ule cells in mammalian olfactory bulb cultures. J. Neurosci.
Smith, R. L., Baker, H., Kolstad, K., Spencer, D. D., and Greer, 12:3985–3991.
C. A. (1991). Localization of tyrosine hydroxylase and olfac- Trombley, P. Q., and Shepherd, G. M. (1993). Synaptic transmission
tory marker protein immunoreactivities in the human and and modulation in the olfactory bulb. Curr. Opin. Neurobiol.
macaque olfactory bulb. Brain Res. 548:140–148. 3:540–547.
Smith, R. L., Baker, H., and Greer, C. A. (1993). Immunohisto- Trombley, P. Q., and Shepherd, G. M. (1994). Glycine exerts
chemical analyses of the human olfactory bulb. J. Comp. potent inhibitory actions on mammalian olfactory bulb neu-
Neurol. 333:519–530. rons. J. Neurophysiol. 71:761–767.
Stewart, W. B., Kauer, J. S., and Shepherd, G. M. (1979). Trombley, P. Q., Horning, M. S., and Blakemore, L. J. (1998).
Functional organization of rat olfactory bulb analysed Carnosine modulates zinc and cooper effects on amino acid
164 Kratskin and Belluzzi

receptors and synaptic transmission. NeuroReport ergic receptors in the main olfactory bulb of rats. Int. J. Dev.
9:3503–3507. Neurosci. 14:913–919.
Trombley, P. Q., Hill, B. J., and Horning, M. S. (1999). Woolf, T. B., Shepherd, G. M., and Greer, C. A. (1991a). Serial
Interactions between GABA and glycine at inhibitory amino reconstructions of granule cell spines in the mammalian olfac-
acid receptors on rat olfactory bulb neurons. J. Neurophysiol. tory bulb. Synapse 7:181–192.
82:3417–3422. Woolf, T. B., Shepherd, G. M., and Greer, C. A. (1991b). Local
Uchida, N., Takahashi, Y. K., Tanifuji, M., and Mori, K. (2000). information processing in dendritic trees: subsets of spines in
Odor maps in the mammalian olfactory bulb: domain organi- granule cells of the mammalian olfactory bulb. J. Neurosci.
zation and odorant structural features. Nature Neurosci. 11:1837–1854.
3:1035–1043. Wright, D. E., Seroogy, K. B., Lundgren, K. H., Davis, B. M.,
Valverde, F. (1965). Studies on the Piriform Lobe. Cambridge and Jennes, L. (1995). Comparative localization of serotonin
University Press, Cambridge. 1A, 1C, and 2 receptor subtype mRNAs in rat brain. J. Comp.
van den Pol, A. N. (1995). Presynaptic metabotropic glutamate Neurol. 351:357–373.
receptors in adult and developing neurons: autoexcitation in Xu, P., Greer, C. A., and Shepherd, G. M. (2000). Odor maps in
the olfactory bulb. J. Comp. Neurol. 359:253–271. the olfactory bulb. J. Comp. Neurol. 422:489–495.
Vassar, R., Chao, S. K., Sitcheran, R., Nunez, J. M., Vosshall, L. Yamamoto, T., Jin, J., and Watanabe, S. (1997). Characteristics
B., and Axel, R. (1994). Topographic organization of sensory of memory dysfunction in olfactory bulbectomized rats and
projections to the olfactory bulb. Cell 79:981–991. the effects of cholinergic drugs. Behav. Brain Res.
Wang, F., Nemes, A., Mendelsohn, M., and Axel, R. (1998). 83:57–62.
Odorant receptors govern the formation of a precise topogra- Yokoi, M., Mori, K., and Nakanishi, S. (1995). Refinement of
phic map. Cell 93:47–60. odor molecule tuning by dendrodendritic synaptic inhibition
Warden, M. K., and Young, W. S. III. (1988). Distribution of in the olfactory bulb. Proc. Natl. Acad. Sci. USA 92:
cells containing mRNAs encoding substance P and neu- 3371–3375.
rokinin B in rat central nervous system. J. Comp. Neurol. Záborszky, L., Carlsen, J., Brashear, H. R., and Heimer, L.
272:90–113. (1986). Cholinergic and GABAergic afferents to the olfactory
Wenzel, B. M. (1974). The olfactory system and behavior. In bulb in the rat with special emphasis on the projection neurons
Limbic and Autonomic System Research, L. V. DiCara (Ed.). in the nucleus of the horizontal limb of the diagonal band. J.
Plenum Press, New York, pp. 1–40. Comp. Neurol. 243:488–509.
White, E. L. (1973). Synaptic organization of the mammalian Zhao, H., Ivic, L., Otaki, J. M., Hashimoto, M., Mikoshiba, K.,
olfactory glomerulus: new findings including an intraspecific and Firestein, S. (1998). Functional expression of a mam-
variation. Brain Res. 60:299–313. malian odorant receptor. Science 279:237–242.
Wilson, D. A., and Leon, M. (1988). Noradrenergic modulation Zheng, L. M., Ravel, N., and Jourdan, F. (1987). Topography of
of olfactory bulb excitability in the postnatal rat. Dev. Brain centrifugal acetylcholinesterase-positive fibres in the olfac-
Res. 42:69–75. tory bulb of the rat: evidence for original projections in atyp-
Woo, C. C., and Leon, M. (1995). Distribution and development ical glomeruli. Neuroscience 23:1083–1093.
of -adrenergic receptors in the rat olfactory bulb. J. Comp. Zheng, L. M., Caldani, M., and Jourdan, F. (1988). Immunohisto-
Neurol. 352:1–10. chemical identification of luteinizing hormone-releasing
Woo, C. C., Wilson, D. A., Sullivan, R. M., and Leon, M. (1995). hormone-positive fibres and terminals in the olfactory system
Early locus coeruleus lesions increase the density of -adren- of the rat. Neuroscience 24:567–578.
8

Central Olfactory Structures

Thomas A. Cleland and Christiane Linster

Cornell University, Ithaca, New York, U.S.A.

In this chapter we review central olfactory structures (Fig. 1), including the bullfrog (Kemali and Guglielmotti, 1987;
with an emphasis on those receiving direct inputs from the Northcutt and Royce, 1975; Scalia et al., 1991) and snake
olfactory bulb, generally referred to as secondary olfactory (Halpern, 1976). The major secondary olfactory structures
structures (Fig. 2) (see Chapter 7). We will first provide an described in mammals and discussed in this chapter
overview of these secondary olfactory structures, then include (listed roughly rostrocaudally): the anterior olfac-
describe their common organizational principles, provide tory nucleus; a group of rostromedial cortices including
detailed descriptions of the incoming and outgoing projec- the ventral tenia tecta, anterior hippocampal continuation,
tions of each structure, and finally discuss evidence regard- and indusium griseum; the olfactory tubercle; the anterior
ing their putative olfactory functions. and posterior piriform cortices and endopiriform nucleus;
the periamygdaloid cortex and anterior cortical nucleus of
the amygdala; and the lateral entorhinal cortex. Despite a
I. OVERVIEW OF SECONDARY OLFACTORY basic conservation of bulbar projection patterns among
STRUCTURES vertebrates, and particularly among mammals, there are
nonetheless important species differences. Such differ-
Secondary olfactory structures include all areas of the ences may provide insight into the multiple mechanisms
brain to which mitral and tufted cell axons from the olfac- by which different species employ olfactory information to
tory bulb (OB) directly project. This term is synonymous solve similar adaptive problems and ultimately contribute
with the common term primary olfactory cortices (de to an understanding of the respective roles played by the
Olmos et al., 1978) (see also Haberly, 2001; Halasz, 1990; diverse central structures receiving olfactory information.
Price, 1973, 1987; Shipley, 1995); however, the latter term In general, the canonical features of the secondary olfactory
is not used in this chapter, as recognition of the olfactory projections described in this review are derived primarily
bulb as a cortical structure has rendered it ambiguous. The from studies in rat, and secondarily from studies in mouse
centripetal projection patterns of bulbar mitral and tufted and primates; however, data from other species are also
neurons have been described in several mammalian reviewed to emphasize specific points and highlight the
species, among them being the rat (Price, 1973), opossum commonalities and diversity of the vertebrate radiation.
(Meyer, 1981; Scalia and Winans, 1975; Shammah- All secondary olfactory structures are paired, except for
Lagnado and Negrao, 1981), monkey (Turner and interhemispheric commissures; there is no evidence of
Mishkin, 1978; Turner et al., 1978), hamster (Davis et al., asymmetry in the anatomy or function of any of these
1978), rabbit (Broadwell, 1975a), tree shrew (Skeen and areas. The axons of mitral cells and a subset of tufted cells
Hall, 1977), and hedgehog (Radtke-Schuller and Kunzle, emerge from the olfactory bulb, forming the olfactory
2000), as well as in several nonmammalian vertebrates peduncle; this path is also the route of the rostral migratory

165
166 Cleland and Linster

stream, along which new presumptive olfactory bulb neu-

rons migrate throughout life from progenitor cells in the
subventricular zone (Jankovski et al., 1998; Meisami and
Hamedi, 1986). Within the olfactory peduncle, and imme-
diately caudal to the olfactory bulb, lies the anterior olfac-
tory nucleus (AON), actually a cortical structure incorpo-
rating several morphologically diverse subdivisions with
characteristic projection patterns. The AON is predomi-
nantly two-layered (with superficial plexiform and deep
cellular layers) but gradually assumes a trilaminar form
near its caudal extreme adjacent to the anterior
commissure (Halasz, 1990; Valverde et al., 1989). Among

Figure 1 Overview of olfactory structures. Each of the panels

shows a sagittal section through a rat brain at different lateral loca-
tions from medial to lateral. (A) Sagittal section at 0.4 mm lateral
from bregma, showing rostromedial olfactory cortices and the olfac-
tory tubercle. OB: olfactory bulb; AOM, AOP: medial and posteri-
or anterior olfactory nucleus; VTT: ventral tenia tecta; AHC:
anterior hippocampal continuation (also known as dorsal tenia
tecta); Tu: olfactory tubercle; DP: dorsal peduncular cortex; IG:
indusium griseum. CA1, CA3, DG: hippocampus; MO: medial
orbitofrontal cortex; IL: infralimbic cortex; Cg1 and Cg2: cingulate
cortex; ac: anterior commissure; MTN: medial thalamic nuclei; DA Figure 2 Schematic depiction of canonical secondary olfactory
and AH: dorsal and anterior hypothalamic areas; VMH: ventrome- projections. Pathways depicted represent the most commonly
dial hypothalamic nucleus. (B) Sagittal section at 2.4 mm lateral reported connections and are neither exhaustive nor universal. (A)
from bregma. AOL: lateral anterior olfactory nucleus; lo: lateral Projections from the olfactory bulb to secondary olfactory struc-
olfactory tract; Pir: piriform cortex; VO: ventral orbitofrontal cor- tures. Directionality is implicit. (B) Projections from secondary
tex; DEn: dorsal endopiriform nucleus; aca: anterior part of the olfactory structures to the olfactory bulb. Directionality is implicit.
anterior commissure; HDB: nucleus of the horizontal limb of the (C) Associative connections among secondary olfactory structures
diagonal band; LOT: nucleus of the lateral olfactory tract; STh: sub- and output projections to prominent tertiary olfactory structures.
thalamic nuclei; VT: ventral thalamic nuclei. (C). Sagittal section at Directionality denoted by arrows. OB: olfactory bulb; AON: anteri-
3.9 mm lateral from bregma. LO: lateral orbitofrontal cortex; DEn or olfactory nucleus; aPC and pPC: anterior and posterior piriform
and VEn: dorsal and ventral endopiriform nucleus; Pir: piriform cortex; PC: piriform cortex (anterior and posterior); EC: entorhinal cor-
cortex; lo: lateral olfactory tract; CxA: periamygdaloid cortex (also tex; Tu: olfactory tubercle; PAC: periamygdaloid cortex; ACo: ante-
known as the cortex-amygdala transition zone); ACo: anterior cor- rior cortical amygdaloid nucleus; VTT: ventral tenia tecta; AHC:
tical amygdaloid nucleus; PMCo: posteriomedial cortical amyg- anterior hippocampal continuation; IG: indusium griseum; IC:
daloid nucleus: LEnt: lateral entorhinal cortex; CA1, CA2, CA3, insular cortex; OC: orbitofrontal cortex; HT: hypothalamus; Th:
DG: hippocampus. (Adapted from Paxinos and Watson, 1986.) thalamus.
Central Olfactory Structures 167

other possible functions, the AON (via the anterior com- (defined in Haberly, 1985, 1998, 2001) and the lateral
missure) mediates interhemispheric communication entorhinal cortex, derivatives of the lateral pallium, as well
between the olfactory bulbs in mammals. as the transitional periamygdaloid cortex and the anterior
Dorsomedial to the AON, and medial to the lateral cortical nucleus of the corticomedial division of the amyg-
olfactory tract, lie several secondary olfactory structures daloid complex. The piriform cortex is a three-layered
collectively termed the rostromedial olfactory cortices: the allocortex (Haberly and Price, 1978a), incorporating a
indusium griseum (also known as the dorsal hippocampal superficial plexiform layer and two cell body layers, that
continuation or the supracallosal gyrus), the anterior hip- has been extensively studied in the context of olfactory
pocampal continuation (also known as the dorsal tenia function (Haberly, 1985, 1998) (see Chapter 9). Deep to
tecta), and the ventral tenia tecta, which medially adjoins the piriform cortex lies the endopiriform nucleus, which
the caudal AON (Carmichael et al., 1994; Haberly and some scholars regard as layer IV of piriform cortex, either
Price, 1978a,b; Kier et al., 1995; Luskin and Price, 1983b; alone or in conjunction with the deep portion of layer III
Shipley and Adamek, 1984; Wyss and Sripanidkulchai, (Tseng and Haberly, 1989; Valverde, 1965). Within the
1983). The olfactory tubercle is also sometimes included amygdaloid complex, bulbar collaterals from the LOT
in this group (Shipley, 1995) (see below), as is the dorso- innervate the anterior cortical nucleus and periamygdaloid
medial peduncular cortex (Haberly, 1998; Haberly and cortex; the latter adjoins and is sometimes also considered
Price, 1977; Halasz, 1990). The anterior hippocampal part of the piriform cortex (Paxinos and Watson, 1986).
continuation and indusium griseum are structurally While these amygdaloid structures exhibit trilaminar struc-
comparable to, and often considered part of, the tures similar to that of piriform cortex, their layers II and
hippocampal formation (Adamek et al., 1984; Wyss and III are somewhat underdeveloped in comparison (Krettek
Sripanidkulchai, 1983); they probably derive from the and Price, 1978). The lateral entorhinal cortex is the most
medial (limbic) pallium, as does the ventral tenia tecta caudal target of olfactory bulb axons (Davis et al., 1978;
(Kier et al., 1995). Heimer, 1968; Price, 1973; Scalia and Winans, 1975).
Just caudal to the olfactory peduncle, and medial to the Entorhinal cortex, which includes medial, lateral, and
lateral olfactory tract and cortices, lies the distinctive intermediate divisions, has six layers as opposed to the
olfactory tubercle. The olfactory tubercle, like the amyg- three (or four) layers observed in piriform cortex; entorhi-
daloid complex and the basal ganglia, is derived from the nal cortex has thus been considered a transitional cortex
striatal subpallium (Butler and Hodos, 1996; Heimer and between olfactory allocortices and the isocortex. The piri-
Wilson, 1975), though in architecture it varies between form, entorhinal, and periamygdaloid cortices are often
cortical (predominant in the lateral tubercle, adjacent to the collectively termed the lateral olfactory cortices.
lateral olfactory tract) and striatal (predominant in the
medial tubercle) organization (Heimer and Wilson, 1975;
Millhouse and Heimer, 1984). The olfactory tubercle
II. ORGANIZATIONAL PRINCIPLES
caudally adjoins the rostromedial olfactory cortices and
has been grouped together with them by some authors, A. Projections from the Olfactory Bulb
differentiating them collectively from the anterior olfac-
tory nucleus and from the continuous lateral olfactory The diverse cortices receiving direct input from the OB
cortices (Haberly, 2001; Shipley, 1995). However, it may exhibit some fundamental similarities in their architec-
ultimately be preferable to consider the olfactory tubercle tures. Each, for example, consists of a superficial plexi-
(a part of the ventral striatum, with its subpallial and form layer (layer I) and one or more deeply located cell
diencephalic projections and its lack of associative layers. Afferent inputs from the OB and associational
connections with other olfactory cortices) separately from inputs from other regions project to layer I, where they
the rostromedial cortices (with their probable medial synapse with pyramidal cell dendrites as well as with local
pallial derivation and interconnectivity with the hippocampal interneurons. In the lateral cortices, bulbar afferents are
formation). sharply restricted to the most superficial portion of the
Laterally, bulbar mitral and tufted cell axons exit the layer, layer Ia, while associational projections from other
peduncular region following the lateral olfactory tract regions arborize within the deeper portion of this plexi-
(LOT). The LOT is heavily myelinated, though it also form layer, termed layer Ib (see below) (Price, 1973). In
contains numerous unmyelinated fibers which, at least in the anterior hippocampal continuation and indusium
cat, may outnumber the myelinated axons (Price and griseum, in contrast, bulbar afferents and associational
Sprich, 1975; Willey et al., 1983). Collaterals from these projections mix within layer I (Wyss and Sripanidkulchai,
axons enter the anterior and posterior piriform cortices 1983). The AON is organized similarly to the lateral
168 Cleland and Linster

cortices, although in the rhesus monkey, a microsmatic piriform cortex, with one quarter of the neurons studied
primate, afferents from the olfactory bulb to the AON additionally projecting into the olfactory tubercle (Ojima
terminate throughout that structure rather than being et al., 1984) (note that these three structures are to date the
restricted to layer I (Turner et al., 1978). Deeper layers in only ones in which individual mitral cell projections have
secondary olfactory cortices are all cell body layers—one been traced). However, within each of the innervated struc-
in the AON, two in the piriform cortex (or three if the tures, the locations of each neuron’s multiple dense termi-
endopiriform nucleus is included), four in the indusium nal arborizations are highly localized, exhibiting a patchy
griseum and anterior hippocampal continuation—except distribution (Ojima et al., 1984). Consequently, while bul-
for the six-layered entorhinal cortex, which in the nomen- bar projections are clearly diverse, they are also likely to
clature of Amaral and Witter (1995) includes four cellular be highly organized. Furthermore, the collateral architec-
layers and one additional plexiform layer (the lamina ture ensures that multiple, widely spaced secondary olfac-
dissecans). Note that while the entorhinal cortex is com- tory structures could receive nearly identical input patterns
monly regarded as six-layered, these layers are dissimilar from the same set of activated mitral cells (Ojima et al.,
to the six layers of mammalian isocortex. 1984).
Bulbar projections to the rostromedial cortices emerge The LOT in mammals is considered to contain all of the
from the superficial plexiform layer of the AON, which is bulbar axons projecting caudally to the lateral olfactory
continuous with that of the ventral tenia tecta, and extend cortices and olfactory tubercle. However, medial and
dorsally along the midline within the superficial plexiform lateral subdivisions are apparent within the rabbit LOT
layers of the ventral tenia tecta, anterior hippocampal con- which reflect the distinct medial and lateral tracts observed
tinuation, and indusium griseum (Adamek et al., 1984; in some other vertebrates. In rabbit, one type of mitral cell
Shipley and Adamek, 1984). Bulbar afferents to the lateral axon collateral courses through the LOT and terminates
cortices, olfactory tubercle, and amygdaloid complex in within the lateral olfactory cortices and the lateral, more
mammals project caudally via the LOT and the superficial cortically organized portion of olfactory tubercle. A sec-
plexiform layers of these cortices. In the LOT, both the ond type of axon collateral branches from the main axon
number and the diameter of bulbar axons decrease as the within the olfactory bulb, travels through the ventromedial
projections extend further caudally (Price and Sprich, olfactory peduncle, remaining medial to the LOT, and
1975); furthermore, the density of innervation of sec- innervates the medial, more striatally organized portion of
ondary olfactory structures by LOT axon collaterals paral- the olfactory tubercle. While evolutionary divergence
lels the developmental sequence of innervation, being renders it challenging to compare olfactory systems in
greatest near the lateral olfactory tract and sparsest in the specific detail, there are many conserved characters in
medial olfactory tubercle (Schwob and Price, 1984). While nonmammalian tetrapods that can shed light on the prob-
both mitral and tufted cells project to the AON and to the able plesiomorphic organization of these projections. In
rostral piriform cortex and olfactory tubercle (Haberly and the garter snake, Thamnophis sirtalis, bulbar projections
Price, 1977; Schoenfeld and Macrides, 1984), the bulbar are clearly segregated into three tracts: a lateral olfactory
projection to more caudal lateral olfactory cortices tract that projects to lateral (piriform) cortex and rostral
becomes progressively dominated by mitral cells (Haberly amygdala, an intermediate olfactory tract that projects to
and Price, 1977). Other than the short-latency interbulbar the olfactory tubercle, and a medial tract that projects
projection via the pars externa of the AON (Haberly and ipsilaterally to the dorsomedial retrobulbar formation
Price, 1978b), no clear topographic organization of bulbar (Lanuza and Halpern, 1998). These three projections are
projections to secondary olfactory structures is in evi- comparable to the lateral and medial portions of the rabbit
dence: the limited topographical regularity immediately LOT and the projection to the mammalian rostromedial
caudal to the olfactory bulb is lost before the LOT emerges cortices, respectively. In bullfrogs (Rana spp.), two distinct
from the olfactory peduncle (Price and Sprich, 1975), tracts are observed: one lateral tract corresponding to the
small regions of the olfactory bulb project to large sec- LOT (in that it projects to lateral pallium, dorsal striatum
ondary areas while small areas within olfactory cortex including the cortical amygdaloid nucleus, and a ventral
receive projections from widely distributed areas in the portion of dorsal pallium, as does the mammalian LOT)
olfactory bulb (Haberly and Price, 1977), and individual and one medial tract projecting to the medial pallium (cor-
mitral and tufted cells innervate diverse secondary regions responding to the mammalian rostromedial cortices) as
(Luskin and Price, 1982; Scott, 1981) (see Scott et al., well as to multiple septal nuclei (Northcutt and Royce,
1980). Indeed, in rabbit, individual mitral cells project 1975; Scalia et al., 1991). Interestingly, in rats, transection
collaterals into multiple secondary olfactory structures— of either the medial or lateral portions of the olfactory
typically arborizing in both the AON and the anterior peduncle disrupted normal performance in a two-choice
Central Olfactory Structures 169

behavioral test, whereas the medial pathway was required given cortical structure) and associative (connections
in order to mediate normal olfactory arousal in isocortex between different cortices) (Shipley, 1995). In piriform cor-
during slow-wave sleep (Gervais and Pager, 1982). tex, local connections are mediated by a variety of excita-
However, unimpaired olfactory task performance can be tory and inhibitory interneurons as well as pyramidal cell
maintained even in rats with more severe LOT lesions collaterals (Haberly, 1998). Associative connections among
(Slotnick and Berman, 1980; Slotnick and Risser, 1990; olfactory cortical structures are extensive and exhibit a
Slotnick and Schoonover, 1993), suggesting that a greater degree of laminar and regional organization. Most intercor-
understanding of the respective contributions of different tical projections among secondary olfactory structures can
secondary olfactory structures to the performance of sub- be classified into one of two fiber systems according to
tly different behavioral tasks may be necessary. their laminar pattern of termination (Luskin and Price,
Many more species-specific deviations from the canon- 1983a,b). The first of these fiber systems, termed the layer
ical bulbar projection have been described. For example, in Ib fiber system, includes projections from the AON and the
the lesser hedgehog tenrec (Echinops telfairi), reciprocal piriform and entorhinal cortices, which terminate in layers
connections have been observed between the olfactory Ib and often layer III; the projections from each of these
bulb and frontal isocortex, rendering those areas (sulcal different structures are typically concentrated at different
and orbitofrontal cortices) secondary olfactory structures characteristic levels within layer Ib. The second fiber sys-
in this species (Radtke-Schuller and Kunzle, 2000), unlike tem, termed the layer II–deep Ib fiber system, originates
most studied species to date in which orbitofrontal cortex from the dorsal peduncular cortex, ventral tenia tecta, and
is a tertiary recipient of olfactory information (Barbas, periamygdaloid cortex, and terminates in layer II.
1993; Carmichael et al., 1994). Similarly, in mice, the Projections from the anterior cortical nucleus of the amyg-
olfactory bulb directly projects to the insular cortex, an daloid complex arborize throughout layers Ia–Ib. A second
isocortical structure that also typically receives only ter- system of classification is apparent based on the origins of
tiary olfactory input (Shipley and Adamek, 1984; Shipley these projections: projections from layer II pyramidal cells
and Geinisman, 1984). In both macaque monkeys (Macaca tend to project to more caudal sites, whereas pyramidal
spp.) and hedgehog (Erinaceus europaeus), a direct pro- cells in layer III target more rostral sites. Layer II cells of
jection from the olfactory bulb to the nucleus of the anterior piriform cortex also send commissural projections
horizontal limb of the diagonal band of Broca has been contralaterally, though these are limited in number and dis-
described, which has not been described in most species tribution compared to ipsilateral projections (Haberly and
studied (Carmichael et al., 1994; De Carlos et al., 1989), Price, 1978a,b; Luskin and Price, 1983a,b). Interestingly,
and a direct projection from the OB to the supraoptic the olfactory tubercle is the only secondary olfactory struc-
nucleus of the hypothalamus has been reported in rats ture that does not give rise to associational projections
(Smithson et al., 1989). Finally, in the lemur (Microcebus (Haberly and Price, 1978a).
murinus), in addition to the canonical projections, bulbar
fibers also directly and bilaterally innervate the hippocam- C. Feedback Projections to the Olfactory Bulb
pus proper (usually considered a tertiary projection area)
as well as the septum, caudate-putamen, and, via the Excepting the olfactory tubercle and indusium griseum, all
medial forebrain bundle, several hypothalamic nuclei and of the secondary olfactory structures described herein send
two mesencephalic modulatory centers (the locus direct feedback projections to the olfactory bulb
coeruleus and the raphe nuclei) (Mestre et al., 1992). (Carmichael et al., 1994; Davis and Macrides, 1981; Davis
et al., 1978; de Olmos et al., 1978; Haberly and Price,
B. Associational Connections Within and Among 1978a,b; Luskin and Price, 1983b; Shipley and Adamek,
Secondary Olfactory Structures 1984; Wyss and Sripanidkulchai, 1983). Among the
secondary structures served by the LOT, corticobulbar
Extensive connections, termed associational fibers, project feedback projections are heavier from rostral areas (AON
between secondary olfactory cortices; their axons arborize and anterior piriform cortex) than from posterior piriform
in layer I as do the bulbar afferents, although in the lateral cortex and other caudal areas (Shipley and Adamek, 1984)
cortices they are sharply segregated from the afferents, pro- and arise mainly from layer II and III pyramidal cells.
jecting into the deeper portion of that layer (layer Ib) and Most of these feedback projections are thought to termi-
also into the two superficial cell body layers (layers II and nate on granule cells in the olfactory bulb, though some
III) (Luskin and Price, 1983a). These associational connec- extend into the glomerular layer. Most feedback
tions have been grouped into two classes: local (or intrinsic; projections are ipsilateral, with the notable exception of
short connections between neurons in different layers of a those originating in the AON, which project bilaterally or
170 Cleland and Linster

contralaterally. Although the functional roles of these pro- insular cortex do project to the mediodorsal thalamic
jections are not known, it is interesting to note that in rab- nucleus in those species.
bit, there are complex changes in olfactory bulb dynamical
activity when the feedback projections are blocked with a E. Neuromodulatory Inputs
cooling probe (Gray and Skinner, 1988). These changes in
dynamic activities may be at the basis of experimental Neuromodulatory inputs to the secondary olfactory struc-
results showing that isolation of the olfactory bulb from tures described herein arise from four main sources: the
secondary olfactory structures impairs the formation of nucleus of the diagonal band (acetylcholine, GABA), the
epileptiform activity in the olfactory bulb of rabbits, dorsal and medial raphe nuclei (serotonin), the locus
suggesting a dependency on more central structures for the coeruleus (norepinephrine), and the substantia nigra-
induction and maintenance of epileptiform activity in the ventral tegmental area (dopamine). Both cholinergic and
olfactory bulb (Gray et al., 1987). GABAergic neurons from the diagonal band (emerging
primarily from the horizontal limb) project to the olfactory
D. Projections to Tertiary Olfactory Structures bulb and secondary olfactory structures (Gaykema et al.,
1990; Haberly and Price, 1978a; Zaborszky et al., 1986).
Olfactory information is also distributed from sec- The known cellular and synaptic effects of cholinergic
ondary olfactory structures to several other regions of modulation in piriform cortex are detailed in Chapter 9
the brain, including orbitofrontal cortex, insular cortex, (see also Linster and Hasselmo, 2001), and a number of
the mediodorsal, submedial, and anterior nuclei of the researchers have demonstrated the importance of choliner-
thalamus, the hypothalamus, the amygdaloid complex gic modulatory inputs for olfactory learning and discrimi-
and the hippocampus (Barbas, 1993; Carmichael et al., nation (De Rosa and Hasselmo, 2000; Doty et al., 1999;
1994; Cavada and Reinoso-Suarez, 1985; Cavada et al., Hunter and Murray, 1989; Linster et al., 2001; Paolini and
2000; Datiche and Cattarelli, 1996b, Krettek and Price, McKenzie, 1993; Roman et al., 1993). The OB and sec-
1977a; Luskin and Price, 1983b; Price, 1985; Price and ondary olfactory structures also receive noradrenergic,
Slotnick, 1983; Price et al., 1991; Reep and Winans, serotonergic, and dopaminergic modulatory inputs
1982; Smithson et al., 1989; Takagi, 1986). Generally, (Datiche and Cattarelli, 1996a; Datiche et al., 1995; Fallon
corticocortical projections from secondary olfactory and Moore, 1978; Fallon et al., 1978; Jones et al., 1977;
structures originate in more superficially located cell Moore et al., 1978). While the physiological effects of
layers (typically layer II), while corticodiencephalic these other neuromodulators have been studied in olfactory
projections originate from deeper layers [e.g., endopiri- cortex (Gellman and Aghajanian, 1993, 1994; Marek and
form nucleus, polymorphic (medial, striatal) zone of the Aghajanian, 1994, 1995; Sheldon and Aghajanian, 1990,
olfactory tubercle, deep cells within periamygdaloid 1991) (reviewed in Hasselmo, 1995), to our knowledge
and entorhinal cortices] (Price, 1985; Price and only norepinephrine, along with acetylcholine, has been
Slotnick, 1983). Different secondary olfactory struc- related to olfactory learning and memory (reviewed in
tures projecting to common tertiary structures typically Sullivan and Wilson, 1994; Sullivan et al., 1992).
project to discrete subregions; for example, olfactory
projections to the thalamus include both highly conver- F. Chemoarchitecture
gent projections from the lateral olfactory cortices to the
mediodorsal and submedial thalamic nuclei (Price, Although the transmitters used by associational and output
1987; Price and Slotnick, 1983), as well as projections fibers in secondary olfactory structures have not been
from the indusium griseum and anterior hippocampal definitively established, there is strong evidence, derived
continuation to the anterior thalamic nuclei (Wyss and from a variety of experimental techniques, that glutamate
Sripanidkulchai, 1983). Tertiary olfacto-hypothalamic is the principal excitatory neurotransmitter for both affer-
projections arise from the anterior olfactory nucleus, the ent and associative fiber systems (Carnes et al., 1990;
piriform cortex, the olfactory tubercle and the amyg- Fuller and Price, 1988; Fuller et al., 1987; Godfrey et al.,
daloid nuclei (Price et al., 1991). Notably, in rat, the pir- 1980; Hoffman and Haberly, 1993; Jung et al., 1990; Ray
iform cortex provides input by way of the mediodorsal et al., 1992); aspartate has also been suggested as an
thalamic nucleus to the same prefrontal areas to which excitatory neurotransmitter in these structures. Excitatory
it projects directly (Ray et al., 1992); in contrast, no EPSCs are mediated via both AMPA- and NMDA-type
corticothalamic projections from the piriform cortex glutamate receptors; glutamate also acts on metabotropic
have been observed in cat or rabbit (Motokizawa et al., receptors in the piriform cortex. GABA is believed to be
1988), though the olfactory tubercle, amygdala, and the predominant inhibitory neurotransmitter in the
Central Olfactory Structures 171

secondary olfactory cortices (Haberly, 1985), acting on projections enabled contralateral access to previously
postsynaptic GABAA receptors as well as metabotropic obtained, ipsilaterally stored odor preferences. If the ante-
GABAB receptors on both pre- and postsynaptic rior commissure was then sectioned in these older rat pups,
membranes. In addition, several neuropeptides have been this acquired contralateral access to learned preference
identified in neurons of the olfactory bulb and secondary was lost, demonstrating that the odor preference was still
olfactory structures (reviewed in Shipley, 1995). maintained unilaterally (Kucharski and Hall, 1987, 1988).
These data suggest that some form of odotopic cross-
attunement between the paired olfactory bulbs and/or ante-
rior olfactory nuclei occurs as the contralateral projections
III. CONNECTIVITY OF SECONDARY
between the bulbs develop, and further that this putative
OLFACTORY STRUCTURES
cross-attunement does not require further training after
A. Anterior Olfactory Nucleus development but can access previously formed, unilateral
memory traces.
The AON, a subset of which has also been termed anterior The AON is the major source of feedback connections
olfactory cortex (Haberly, 2001), is a laminated structure to the olfactory bulb from any source (Carson, 1984); all
embedded within the olfactory peduncle. The AON has subdivisions of the AON project to both the ipsilateral
been divided into several subregions with distinct architec- and the contralateral olfactory bulb except for pars exter-
tures and connectivities (described by Broadwell, 1975b, na, which projects only to the contralateral olfactory bulb
Davis and Macrides, 1981, de Olmos et al., 1978, Haberly via the anterior commissure (Broadwell, 1975b; Davis
and Price, 1978b, Shipley, 1995, Shipley and Adamek, and Macrides, 1981; Haberly and Price, 1978b). The
1984). It is predominantly two-layered, consisting of a AON also projects to the piriform cortex, olfactory tuber-
superficial plexiform layer containing incoming projection cle, ventral tenia tecta, orbitofrontal cortex, and hypo-
fibers and the apical dendrites of its intrinsic neurons, and thalamus (Barbas, 1993; Luskin and Price, 1983b; Price
a tightly packed cell body layer (Haberly and Price, et al., 1991) and receives projections from several struc-
1978b), but gradually assumes a trilaminar form near its tures including the piriform and entorhinal cortices
caudal extreme adjacent to the anterior commissure (Luskin and Price, 1983b; Wyss, 1981), as well as the
(Halasz, 1990; Valverde et al., 1989). The AON receives CA1 region of the hippocampal formation (van Groen
projections from olfactory bulb mitral and tufted cells in its and Wyss, 1990). While no clear topographical organiza-
superficial plexiform layer, layer Ia (Scott et al., 1985); tion is apparent in most of these projections (e.g., Luskin
bulbar afferents also course along this superficial layer into and Price, 1982; Price and Sprich, 1975), a few are clear-
the superficial plexiform layers of the adjoining ventral ly topographical. Bulbar projections to the AON pars
tenia tecta (medially) and anterior piriform cortex (lateral- externa, and that structure’s projections to the contralat-
ly) (Shipley and Adamek, 1984), the latter forming the eral OB, are both strictly topographical (Schoenfeld and
lateral olfactory tract. In 3-week-old cats, a study of neu- Macrides, 1984; Scott et al., 1985); the pars lateralis may
ropeptide Y–immunoreactive neurons observed in the OB also exhibit some topographical organization (Scott et al.,
and olfactory peduncle revealed a dense contralateral 1985). Finally, the AON receives topographically orga-
projection through the anterior commissure, which was nized inputs from the ventral tenia tecta (Luskin and
dramatically reduced after the first 4 postnatal months. In Price, 1983b).
contrast, ipsilaterally projecting neurons of this type are Direct bulbobulbar contralateral projections that bypass
not substantially reduced in the adult (Sanides-Kohlrausch the AON have also been demonstrated in several species
and Wahle, 1990). Such dense contralateral projections including cat, rabbit, caiman, turtle, and fish, but are
during development could conceivably aid in correlating believed to be absent in others such as rat, mouse, rhesus
the odotopic projections of the two olfactory bulbs with monkey, hamster, guinea pig, frog (Rana spp.), and some
one another. In neonatal rats unilaterally trained on odors lizards (reviewed in Halasz, 1990; Kemali and
(before the development of a functional olfactory anterior Guglielmotti, 1987; Scalia et al., 1991; Shipley and
commissure), the learned odor preference was stored uni- Adamek, 1984; Turner et al., 1978) (see Leveteau et al.,
laterally, as evidenced by the animals’ exhibition of a pref- 1993). However, in frogs (Rana esculenta), primary olfac-
erence for the odor only when presented to the trained tory receptor neurons themselves have been shown to
(ipsilateral) side. Older rat pups, however, exhibited project bilaterally and innervate both olfactory bulbs; this
learned preference when the odor was presented to either contralateral projection is mediated by an interbulbar
side, even if they had been trained before the anterior com- adhesion distinct from the anterior and habenular
missure developed; i.e., the development of contralateral commissures (Leveteau et al., 1992).
172 Cleland and Linster

Little is known about the functional role of the AON for C. Olfactory Tubercle
olfactory processing, save that it mediates most or all bilat-
eral bulbobulbar communication in many species and is The olfactory tubercle in mammals is a prominent bulge on
thus presumably important for the bilateral comparison of the base of the brain just caudal to the olfactory peduncle
olfactory information. Odor responses recorded in rabbit and medial to the lateral olfactory tract; it receives afferent
AON neurons appeared less odor-selective than those input from mitral and tufted cells in the OB (de Olmos et
recorded in the OB (Boulet et al., 1978). When adult rats al., 1978; Heimer, 1968). The olfactory tubercle exhibits a
were trained on simple discrimination tasks, changes in superficial plexiform layer like the lateral and rostromedi-
odor-evoked neural activity as measured by 2-deoxy- al olfactory cortices, but its cellular architecture varies:
glucose (2DG) staining were observed in the AON of the medially, it resembles other striatopallidal complexes,
trained animals compared to their untrained counterparts; whereas laterally (adjoining the piriform cortex) it exhibits
interestingly, no changes in 2DG uptake were observed in a trilaminar cortical organization (Heimer and Wilson,
the piriform cortex in this experiment (Hamrick et al., 1975; Millhouse and Heimer, 1984). However, the olfac-
1993). Enhanced c-fos expression has also been observed tory tubercle differs from the piriform cortex in that it does
in the AON of rats that received forward pairing of odors not send output projections to the OB or to any other sec-
with a foot shock stimulus, demonstrating that odor- ondary olfactory structures (Haberly and Price, 1978a;
induced c-fos expression can be modified through aversive Luskin and Price, 1983b); the outputs of the olfactory
conditioning in the AON as well as in the olfactory bulb tubercle are directed towards the mediodorsal and subme-
(Funk and Amir, 2000). C-fos expression within OB odor- dial nuclei of the thalamus (Price and Slotnick, 1983), the
activated regions was also reduced bilaterally when ventral pallidum, and the nucleus accumbens (Heimer and
centrifugal afferents were severed by unilateral section of Wilson, 1975; Luskin and Price, 1983b), and, in monkeys,
the olfactory peduncle or by application of noradrenergic the orbitofrontal cortex (Barbas, 1993). The inputs and
antagonists within the OB, while 2DG uptake patterns projections to and from the olfactory tubercle can vary
were unaffected (Sallaz and Jourdan, 1993, 1996). These substantially among species; for example, in many
results suggest that the AON is a plastic structure that par- macrosmatic animals (in which the olfactory sense is well
ticipates in olfactory learning along with the olfactory bulb developed), the olfactory tubercle receives copious direct
and other secondary olfactory structures. bulbar input, and cell bridges exist between the olfactory
tubercle and other striatal structures (Butler and Hodos,
B. Rostromedial Olfactory Structures 1996), whereas in humans and other microsmatic primates,
the region of the tubercle receiving afferent input from the
The ventral tenia tecta, anterior hippocampal continuation, OB is greatly reduced (Shipley, 1995).
and indusium griseum receive input from the OB in their Neurons in the olfactory tubercle of rats can respond to
small molecular layers (Adamek et al., 1984; de Olmos electrical stimulation of the OB, as suggested by their
et al., 1978; Levy et al., 1999; Shipley and Adamek, 1984; direct bulbar inputs. Both excitatory and inhibitory
Wyss and Sripanidkulchai, 1983); in addition, the latter responses have been observed and were modulated by the
two cortical structures receive input from the entorhinal application of dopamine (Inokuchi et al., 1987, 1988).
cortex. Because the anterior hippocampal continuation and
indusium griseum have been considered part of the hip- D. Piriform Cortex and Endopiriform Nucleus
pocampal formation, their inputs from the OB have been
suggested to provide a more direct olfactory input to the Of all secondary olfactory structures, the piriform cortex
hippocampus proper than that via the entorhinal cortex (also termed the primary olfactory cortex) has been most
(Adamek et al., 1984). While there are projections from the intensively studied with respect to olfactory function
ventral tenia tecta and anterior hippocampal continuation (Haberly, 1985, 1998, 2001) (see Chapter 9). This three-
to the OB, the indusium griseum does not project back to layered allocortex receives abundant afferent input from
the OB. The ventral tenia tecta additionally receives pro- the OB as well as inputs from other secondary olfactory
jections from, and projects back to, the AON, while the cortices, excepting the olfactory tubercle (Kowianski et al.,
indusium griseum receives additional input from the piri- 1999; Krettek and Price, 1977a; Luskin and Price, 1983b;
form cortex (Adamek et al., 1984; Luskin and Price, Wyss, 1981). Mitral and tufted cells project to the piriform
1983b; Wyss and Sripanidkulchai, 1983). Interestingly, cortex by way of the lateral olfactory tract (LOT) and
connections from the OB to the ventral tenia tecta are arborize exclusively in layer Ia. The piriform cortex also
absent in the microsmatic rhesus monkey (Turner and receives input from the orbitofrontal and insular cortices,
Mishkin, 1978). hippocampal formation, basal forebrain, brainstem,
Central Olfactory Structures 173

thalamus, and hypothalamus (Haberly and Price, 1978a; Winans, 1981; Kowianski et al., 1999; Krettek and Price,
Kowianski et al., 1999) and sends extensive projections 1977b; Kunzle and Radtke-Schuller, 2000; Wyss, 1981). A
back to the OB (Carmichael et al., 1994; de Olmos et al., second superficial corticoid structure within the amyg-
1978; Haberly and Price, 1978a,b; Luskin and Price, daloid complex, adjoining the anterior cortical nucleus, is
1983a). These feedback projections terminate mainly on or the nucleus of the lateral olfactory tract; this structure
near OB granule cells, which are inhibitory to mitral and exhibits a trilaminar structure similar to that of the anteri-
tufted projection neurons in the bulb. Many projections or cortical nucleus, though its interconnectivity with other
from the piriform cortex to other regions have also been secondary olfactory structures is less established.
described (Carmichael et al., 1994; Haberly and Price, Research in rats and in monkeys has shown that, in
1978a; Kowianski et al., 1999; Price et al., 1991; Takagi, awake, behaving animals, neurons in the amygdaloid com-
1986), including projections to many other secondary plex respond selectively to olfactory stimulation. In rats,
olfactory structures as well as to the hippocampal forma- neurons in the basolateral amygdala responded to odors
tion, orbitofrontal and insular isocortices, the amygdaloid (Cain, 1975; Cain and Bindra, 1972), displayed selective
complex, the hypothalamus, and the mediodorsal and odor responses in an odor discrimination task, and rapidly
submedial nuclei of the thalamus. A more detailed review reversed this selectivity during reversal learning
of the known architecture, connectivity and function of the (Schoenbaum et al., 1998). In monkeys, odor selectivity in
piriform cortex is provided in Chapter 9 (see also Haberly, medial amygdalar neurons could be obtained without
1985, 1998, 2001; Linster and Hasselmo, 2001). training (Tanabe et al., 1975). In PET studies of humans,
Deep to the piriform cortex lies the endopiriform nu- aversive odors have been shown to activate the amygdala
cleus, a large group of multipolar cells interconnected with in both hemispheres (Zald and Pardo, 2000). Furthermore,
the overlying cortex (to the extent that it, either alone or in in neonatal rats, lesions of the amygdaloid complex
combination with the deep portion of layer III, is consid- blocked the acquisition of odor preferences in a condi-
ered layer IV of piriform cortex by some authors) (Tseng tioned odor association paradigm, although this impair-
and Haberly, 1989; Valverde, 1965). The function of the ment could be overcome by overtraining (Sullivan and
endopiriform nucleus is unknown; however, studies with Wilson, 1993). In contrast, in adult rats, lesions of either
animal models suggest that it plays an important role in the lateral olfactory tract inputs to the amygdala or of the
temporal lobe epileptogenesis (Behan and Haberly, 1999). amygdala itself did not affect simple odor discrimination
The input and output connections of the endopiriform learning (Slotnick, 1985; Slotnick and Risser, 1990;
nucleus are very similar to those of the piriform cortex Sutherland and McDonald, 1990). Lesions of the bed
(Kowianski et al., 1999), but efferents from the endopiri- nucleus of the stria terminalis (described in Broadwell,
form nucleus lack the precise laminar order of those from 1975b; Krettek and Price, 1978; Turner and Zimmer, 1984)
the piriform cortex and form a heavy caudorostral pathway specifically blocked activation of the hypothalamic
that the piriform cortex lacks (Behan and Haberly, 1999). paraventricular nucleus (PVN), which regulates adreno-
cortical secretion by olfactory stimuli while sparing acti-
E. Periamygdaloid Cortex and the Anterior Cortical vation of the PVN via other sensory modalities (Mor et al.,
Nucleus of the Amygdaloid Complex 1987).

In mammals, axons from the main OB project to the peri- F. Entorhinal Cortex
amygdaloid cortex (considered part of the piriform cortex
by some) (Paxinos and Watson, 1986) and the anterior The lateral portion of the entorhinal cortex is the most cau-
cortical nucleus of the amygdaloid complex. While the dal projection of olfactory bulb axons (Davis et al., 1978;
accessory olfactory bulb also projects to the amygdaloid Heimer, 1968; Price, 1973; Scalia and Winans, 1975).
complex, its target regions are not shared with those of the Entorhinal cortex is divided into medial, lateral, and inter-
main OB (Haberly and Price, 1978a; Krettek and Price, mediate divisions and is commonly categorized into six [or
1978; Luskin and Price, 1983b). These “extended seven; there is some disagreement between the primate and
amygdalar” structures exhibit a characteristic trilaminar rat literatures (Amaral and Witter, 1995)] layers as
structure, although layers II and III are somewhat less opposed to the three (or four) layers seen in piriform cor-
developed than in the piriform cortex (Krettek and Price, tex. Entorhinal cortex has been considered transitional
1978). Olfactory output targets of the periamygdaloid cor- between olfactory allocortices and the isocortex, although
tex and the anterior cortical nucleus include the piriform its six layers do not directly correspond to the six layers of
cortex, entorhinal cortex, infralimbic area, ventral mammalian isocortex. While entorhinal cortex projects
agranular insular area, and perirhinal area (Kevetter and back to the OB, and also to other olfactory cortical
174 Cleland and Linster

structures including the anterior olfactory nucleus, ventral influences which of these structures become most activated.
tenia tecta, indusium griseum, piriform cortex, endopiri- For example, in studies measuring regional cerebral blood
form nucleus, olfactory tubercle, and amygdaloid cortices flow increases using PET, presentation of single odors
(Kowianski et al., 1999; Luskin and Price, 1983a,b; Wyss, increased activity in the piriform, periamygdaloid,
1981), its strongest projection is to the hippocampal orbitofrontal, insular and cingulate cortices as well as in
formation. the thalamus, indicating that olfactory stimuli activate
A number of studies have shown that olfactory stimuli diverse regions throughout the human brain. When sub-
can modulate neural activity in the entorhinal cortex of jects were tested on olfactory discrimination and memory
behaving rats (Chabaud et al., 2000; Kay and Freeman, tasks, additional regions were activated, including the
1998; Mouly et al., 2001). Highly coherent dynamical cerebellum (Savic et al., 2000). In a similar PET study, the
neural responses from piriform cortex and entorhinal cor- orbitofrontal cortex was differentially activated when sub-
tex have been evoked by odor application, and these jects were asked to make judgments about odor presence,
dynamics change in the entorhinal cortex as a function of familiarity, intensity, hedonicity, or edibility; furthermore,
the behavioral relevance of a given odor stimulus that activation was differentially lateralized depending on
(Chabaud et al., 2000; Kay and Freeman, 1998; Mouly the task (Royet et al., 2001). Aversive odor stimuli have
et al., 2001). While these data suggest the functional been shown to produce regional cerebral blood flow
relevance of entorhinal cortex to olfactory processing, increases in orbitofrontal cortex; highly aversive odor
behavioral lesion studies have demonstrated that rats with stimuli additionally evoked such increases in the
posterior sections of the LOT (severing bulbar projections amygdaloid complex (Zald and Pardo, 1997). Finally, in
to the entorhinal and amygdaloid cortices) are not a functional magnetic resonance imaging study differenti-
impaired in odor discrimination tasks (Slotnick and Risser, ating effects of odor stimulation per se from those deriving
1990; Thanos and Slotnick, 1997; Zhang et al., 1998). from motor and other correlates of odor sampling behav-
While these data in turn may superficially suggest that iors, Sobel and colleagues (1998) showed that active sniff-
entorhinal cortex is not a crucial structure for olfactory dis- ing, either in the absence or in the presence of an odor,
crimination learning, it is also clear that entorhinal cortex induced activation in the piriform and the medial and
receives olfactory input not only from the OB, but also posterior orbitofrontal cortices; in contrast, smelling an
from many other secondary olfactory structures; that is, odor, regardless of sniffing activity, induced activation
posterior LOT lesions may not eliminate the participation mainly in the lateral and anterior orbitofrontal cortex (see
of the entorhinal cortex in olfactory stimulus processing. Chapter 12). These results emphasize a crucial caveat to
Finally, it has been reported that short-term memory for imaging and other physiological studies: regions in which
olfactory stimuli in a delayed-nonmatch-to-sample task in activity correlates with or is shown to mediate important
rats can be increased in duration by lateral entorhinal cor- features of an olfactory task are not necessarily chemosensory
tex lesions (Ferry et al., 1996; Wirth et al., 1998); these in nature.
results could of course also be interpreted as a decrease in Physiological and behavioral data from nonhuman ani-
the rats’ ability to extinguish associations likely to be no mals have also offered considerable insight into the func-
longer appropriate due to the passage of time, suggesting tional roles of various secondary olfactory structures in
that such lesions would impair reversal learning. odor acquisition, processing, and memory. Neural
responses evoked or modulated by olfactory stimulation
have been reported in the anterior olfactory nucleus (anes-
IV. FUNCTIONAL ASPECTS thetized rabbit) (Boulet et al., 1978), amygdala (awake,
behaving rats) (Schoenbaum et al., 1998) (monkeys)
The anatomy of the olfactory pathways described in this (Tanabe et al., 1975), orbitofrontal cortex (awake rats)
chapter clearly shows that olfactory processing involves a (Lipton et al., 1999; Ramus and Eichenbaum, 2000,
large number of structures, interconnected with each other Schoenbaum et al., 1998, 2000) (monkeys) (Tanabe et al.,
in complex fashion, and incorporating both feedforward 1975), and in the hypothalamus (awake monkeys) (Karadi
and feedback interactions. As early as in the olfactory bulb, et al., 1989; Tazawa et al., 1987). Several experiments have
feedback projections from more central brain structures shown that odor-evoked neural activity is modified by
influence neural dynamics and are crucial for olfactory experience: in the AON, enhanced c-fos transcription was
learning and processing. Whole-brain imaging studies in observed in rats that had been trained to associate a foot-
humans have shown that multiple, diverse neural structures shock with an odor (Funk and Amir, 2000); similarly,
become activated during tasks involving olfactory stimula- changes in neural activity recorded with [14C]2-deoxyglu-
tion, and furthermore that the nature of the task strongly cose were observed in the AON of rats that had learned a
Central Olfactory Structures 175

simple odor detection task (Hamrick et al., 1993). Odor- not have been eliminated by the LOT lesions. Indeed, in
evoked single-unit responses in the orbitofrontal cortex some types of olfactory memory tasks, lesions of the
and amygdala of awake, behaving rats are modulated by entorhinal cortex can be interpreted to facilitate olfactory
the reward associations that rats learned during the task; in recognition. In an olfactory habituation task, rats with aspi-
addition, the activity of most neurons in both these struc- rative entorhinal cortex lesions displayed recognition of a
tures is also modulated by other task-related events (Lipton previously investigated odor at latencies for which control
et al., 1999; Ramus and Eichenbaum, 2000; Schoenbaum rats did not (Wirth et al., 1998). Similar results have been
et al., 1998, 2000). In the entorhinal cortex, the learning of obtained using conditioned odor aversion, in which
olfactory stimuli is accompanied by changes in neural entorhinal cortex lesions lengthened the time window dur-
dynamics, as measured by local field potentials (Chabaud ing which an association between the odor stimulus and
et al., 2000; Kay and Freeman, 1998; Mouly et al., 2001). the subsequent aversive stimulus could be formed (Ferry
All of these data together demonstrate that an animal’s et al., 1996, 1999). Lesions of the mediodorsal thalamic
experience with and expectations about odors can durably nucleus, in contrast, impaired both acquisition of an odor
alter the odor-evoked response patterns of individual neu- discrimination task and its reversal (Slotnick and Risser,
rons and the overall spatial activity patterns in response to 1990). Finally, lesions of the olfactory inputs to the amyg-
odorants, as well as the dynamics of the interplay of neur- dala did not impair performance on olfactory detection and
al populations. While these effects are readily apparent, it discrimination tasks (Slotnick, 1985); however, Sullivan
remains unclear just what factors are being encoded or and Wilson (1993) reported that bilateral amygdala lesions
what the functional meanings of such changes might be. in neonatal rats affected learned odor preferences and that
Behavioral lesion studies can also yield valuable infor- these effects could be reversed by extensive training.
mation about the putative contributions of various
structures to tasks such as odor detection, identification,
discrimination, responsivity, and learning. In a series of V. CONCLUSION
behavioral lesion studies, Slotnick and colleagues have
shown that deficits in odor detection and learning are Olfactory sensory input pathways diverge immensely after
related to the extent to which the olfactory bulb is discon- emerging from the relative bottleneck of the olfactory
nected from the forebrain. For example, transections of bulb. As reviewed in this chapter, these secondary olfac-
only the lateral olfactory tract, the anterior limb of the tory projections innervate a broad diversity of structures
anterior commissure, or the olfactory tubercle had little deriving from several distinct telencephalic pallial and
effect on performance of a simple odor discrimination subpallial tissues, as well as diencephalic, midbrain, and
task, whereas combined lesions of these structures brainstem structures. Many of these structures are highly
produced severe impairments (Slotnick and Schoonover, interconnected with one another via associative projec-
1992). Interestingly, transections of the lateral olfactory tions, while some are relatively isolated from other
tract, sparing the more medially directed outputs of the secondary olfactory influences; furthermore, each of them
olfactory bulb, had little effect on odor retention, suggest- receives characteristic extrinsic and modulatory inputs
ing that medial olfactory projections can suffice to perform from other regions of the brain. While these structures and
certain olfactory tasks (Slotnick and Berman, 1980). projections are remarkably conserved among vertebrates,
Lesions of either the lateral or medial portions of the olfac- there are also numerous species-specific variations that
tory peduncle impaired rats’ performance in a food odor presumably derive from the divergent adaptive needs of
detection task, and medial lesions specifically impaired each species, both in terms of novel or missing projections
mitral cell responsivity to food odor presentation during and in terms of the relative densities of projection patterns
slow-wave sleep (Gervais and Pager, 1982). among secondary and tertiary olfactory structures. Lacking
More posterior lesions of the lateral olfactory tract, dis- specific knowledge of what purposes most of these struc-
connecting the amygdaloid complex and entorhinal cortex tures serve, or even of the physiological and adaptive tasks
from direct olfactory bulb inputs, had no detectable effects that must be performed by the organism and for which it
on either retention of a previously learned odor detection requires olfactory perceptual information, what framework
task or the acquisition of a simple odor discrimination for analysis is likely to be the most conducive to elucidat-
(Slotnick, 1985; Slotnick and Risser, 1990); however, ing an understanding of these structures over time?
substantial associational projections to the entorhinal It may be counterproductive to think of secondary
cortex from multiple secondary olfactory structures olfactory structures as primarily “olfactory” in nature; in
remained intact under this procedure. Thus, the contribu- particular, it may be misleading to judge such structures
tion of entorhinal cortex activity to the olfactory task may primarily on the basis of the purported odor selectivity of
176 Cleland and Linster

individual neurons, or even of ensembles. Aside from the Behan, M., and Haberly, L. B. (1999). Intrinsic and efferent con-
typically unwarranted assumptions about mechanisms nections of the endopiriform nucleus in rat. J. Comp. Neurol.
that necessarily underlie statistical measures of selectiv- 408(4):532–548.
ity, it is unlikely that ever-increasing specificity is the Boulet, M., Daval, G., and Leveteau, J. (1978). Qualitative and
quantitative odour discrimation by mitral cells as compared to
general goal of all secondary processing. Rather, a func-
anterior olfactory nucleus cells. Brain Res. 142(1):123–134.
tional approach is likely to be stronger: for what various
Broadwell, R. D. (1975a). Olfactory relationships of the telen-
purposes might a given organism require olfactory sen- cephalon and diencephalon in the rabbit. I. An autoradiographic
sory data, and what elements of those data are needed for study of the efferent connections of the main and accessory
the organism to respond adaptively? How precise must an olfactory bulbs. J. Comp. Neurol. 63(3):329–345.
olfactory identification be in order to meet the organism’s Broadwell, R. D. (1975b). Olfactory relationships of the telen-
needs, and what are the probable costs of false-positive cephalon and diencephalon in the rabbit. II. An autoradi-
errors compared with false negatives? Even if maximally ographic and horseradish peroxidase study of the efferent
specific odor identification were prerequisite to all connections of the anterior olfactory nucleus. J. Comp.
decision processes utilizing olfactory information—an Neurol. 164(4):389–409.
unlikely possibility—neural activity based increasingly Butler, A. B., and Hodos, W. (1996). Comparative Vertebrate
Anatomy: Evolution and Adaptation. Wiley-Liss, New York.
on contingency and less on the physical characteristics of
Cain, D. P. (1975). Effects of insulin injection on responses of
the stimulus would be expected as the response cascade
olfactory bulb and amygdala single units to odors. Brain Res.
proceeds beyond primary sensory areas in the brain. In 99(1):69–83.
some tissues, studying how the categorization of different Cain, D. P., and Bindra, D. (1972). Responses of amygdala single
olfactory stimuli changes, for example, may be more units to odors in the rat. Exp. Neurol. 35(1):98–110.
indicative than measuring how theoretically orthogonal Carmichael, S. T., Clugnet, M. C., and Price, J. L. (1994). Central
their representations may be. olfactory connections in the macaque monkey. J. Comp.
In short, a functional approach to understanding the Neurol. 346(3):403–434.
contributions of secondary olfactory structures might be to Carnes, K. M., Fuller, T. A., and Price, J. L. (1990). Sources of
hypothesize an information-processing task to which a given presumptive glutamatergic/aspartatergic afferents to the
structure might contribute, to assess what elements of magnocellular basal forebrain in the rat. J. Comp. Neurol.
302(4): 824–852.
olfactory sensory information would be required in order
Carson, K. A. (1984). Quantitative localization of neurons pro-
for it to fulfill that task, and to predict what cellular and
jecting to the mouse main olfactory bulb. Brain Res. Bull.
network mechanisms would be useful in order to extract 12(6):629–634.
the needed information from the ensemble activity of the Cavada, C., and Reinoso-Suarez, F. (1985). Topographical orga-
mitral/tufted cell projection neurons that innervate it. nization of the cortical afferent connections of the prefrontal
Differences in species—e.g., microsmatic compared to cortex in the cat. J. Comp. Neurol. 242(3):293–324.
macrosmatic, predator compared with prey species—along Cavada, C., Company, T., Tejedor, J., Cruz-Rizzolo, R. J., and
with what is known about the functions and connectivities Reinoso-Suarez, F. (2000). The anatomical connections of the
of different regions within the brain, will likely be useful macaque monkey orbitofrontal cortex. A review. Cereb.
in hypothesizing about the respective utilities of divergent Cortex 10(3):220–242.
secondary olfactory structures and the purposes for which Chabaud, P., Ravel, N., Wilson, D. A., Mouly, A. M., Vigouroux, M.,
Farget, V., and Gervais, R. (2000). Exposure to behaviourally
each may sample the data provided by the mitral/tufted
relevant odour reveals differential characteristics in rat central
ensemble representation.
olfactory pathways as studied through oscillatory activities.
Chem. Senses 25(5):561–573.
Datiche, F., and Cattarelli, M. (1996a). Catecholamine innerva-
REFERENCES tion of the piriform cortex: a tracing and immunohistochemi-
cal study in the rat. Brain Res. 710(1–2):69–78.
Adamek, G. D., Shipley, M. T., and Sanders, M. S. (1984). The Datiche, F., and Cattarelli, M. (1996b). Reciprocal and topo-
indusium griseum in the mouse: architecture, Timm’s histo- graphic connections between the piriform and prefrontal cor-
chemistry and some afferent connections. Brain Res. Bull. tices in the rat: a tracing study using the B subunit of the
12(6):657–668. cholera toxin. Brain Res. Bull. 41(6):391–398.
Amaral, D. G., and Witter, M. P. (1995). Hippocampal formation. Datiche, F., Luppi, P. H., and Cattarelli, M. (1995). Serotonergic
In The Rat Nervous System, G. Paxinos (Ed.), 2nd ed., and non-serotonergic projections from the raphe nuclei to the
Academic Press, San Diego, pp. 443–493. piriform cortex in the rat: a cholera toxin B subunit (CTb) and
Barbas, H. (1993). Organization of cortical afferent input to 5-HT immunohistochemical study. Brain Res. 671(1):27–37.
orbitofrontal areas in the rhesus monkey. Neuroscience Davis, B. J., and Macrides, F. (1981). The organization of cen-
56(4):841–864. trifugal projections from the anterior olfactory nucleus,
Central Olfactory Structures 177

ventral hippocampal rudiment, and piriform cortex to the Gervais, R., and Pager, J. (1982). Functional changes in waking
main olfactory bulb in the hamster: an autoradiographic study. and sleeping rats after lesions in the olfactory pathways.
J. Comp. Neurol. 203(3):475–493. Physiol. Behav. 29(1):7–15.
Davis, B. J., Macrides, F., Youngs, W. M., Schneider, S. P., and Godfrey, D. A., Ross, C. D., Carter, J. A., Lowry, O. H., and
Rosene, D. L. (1978). Efferents and centrifugal afferents of Matschinsky, F. M. (1980). Effect of intervening lesions on
the main and accessory olfactory bulbs in the hamster. Brain amino acid distributions in rat olfactory cortex and olfactory
Res. Bull. 3(1):59–72. bulb. J. Histochem. Cytochem. 28(11):1157–1169.
De Carlos, J. A., Lopez-Mascaraque, L., and Valverde, F. (1989). Gray, C. M., and Skinner, J. E. (1988). Centrifugal regulation of
Connections of the olfactory bulb and nucleus olfactorius neuronal activity in the olfactory bulb of the waking rabbit as
anterior in the hedgehog (Erinaceus europaeus): fluorescent revealed by reversible cryogenic blockade. Exp. Brain Res.
tracers and HRP study. J. Comp. Neurol. 279(4):601–618. 69(2):378–386.
de Olmos, J., Hardy, H., and Heimer, L. (1978). The afferent con- Gray, C. M., Freeman, W. J., and Skinner, J. E. (1987). Induction
nections of the main and the accessory olfactory bulb forma- and maintenance of epileptiform activity in the rabbit olfac-
tions in the rat: an experimental HRP-study. J. Comp. Neurol. tory bulb depends on centrifugal input. Exp. Brain Res.
181(2):213–244. 68(1):210–212.
De Rosa, E., and Hasselmo, M. E. (2000). Muscarinic choliner- Haberly, L. B. (1985). Neuronal circuitry in olfactory cortex:
gic neuromodulation reduces proactive interference between anatomy and functional implications. Chem. Senses 10: 219–238.
stored odor memories during associative learning in rats. Haberly, L. B. (1998). Olfactory cortex. In The Synaptic
Behav. Neurosci. 114(1):32–41. Organization of the Brain, G. M. Shepherd (Ed.). Oxford
Doty, R. L., Bagla, R., and Kim, N. (1999). Physostigmine University Press, New York, pp. 377–417.
enhances performance on an odor mixture discrimination test. Haberly, L. B. (2001). Parallel-distributed processing in olfac-
Physiol. Behav. 65(4–5):801–804. tory cortex: new insights from morphological and physiological
Fallon, J. H., and Moore, R. Y. (1978). Catecholamine innerva- analysis of neuronal circuitry. Chem. Senses 26(5):551–576.
tion of the basal forebrain. III. Olfactory bulb, anterior olfac- Haberly, L. B., and Price, J. L. (1977). The axonal projection pat-
tory nuclei, olfactory tubercle and piriform cortex. J. Comp. terns of the mitral and tufted cells of the olfactory bulb in the
Neurol. 180(3):533–544. rat. Brain Res. 129(1):152–157.
Fallon, J. H., Koziell, D. A., and Moore, R. Y. (1978). Catecholamine Haberly, L. B., and Price, J. L. (1978a). Association and com-
innervation of the basal forebrain. II. Amygdala, suprarhinal missural fiber systems of the olfactory cortex of the rat.
cortex and entorhinal cortex. J. Comp. Neurol. 180(3): J. Comp. Neurol. 178(4):711–740.
509–532. Haberly, L. B., and Price, J. L. (1978b). Association and com-
Ferry, B., Oberling, P., Jarrard, L. E., and Di Scala, G. (1996). missural fiber systems of the olfactory cortex of the rat. II.
Facilitation of conditioned odor aversion by entorhinal cortex Systems originating in the olfactory peduncle. J. Comp.
lesions in the rat. Behav. Neurosci. 110(3):443–450. Neurol. 181(4):781–807.
Ferry, B., Wirth, S., and Di Scala, G. (1999). Functional interac- Halasz, N. (1990). The Vertebrate Olfactory System. Akademia
tion between entorhinal cortex and basolateral amygdala dur- Kiado, Budapest.
ing trace conditioning of odor aversion in the rat. Behav. Halpern, M. (1976). The efferent connections of the olfactory bulb
Neurosci. 113(1):118–125. and accessory olfactory bulb in the snakes, Thamnophis sirta-
Fuller, T. A., and Price, J. L. (1988). Putative glutamatergic lis and Thamnophis radix. J. Morphol. 150(2 Pt. 2):553–578.
and / or aspartatergic cells in the main and accessory olfactory Hamrick, W. D., Wilson, D. A., and Sullivan, R. M. (1993).
bulbs of the rat. J. Comp. Neurol. 276(2):209–218. Neural correlates of memory for odor detection conditioning
Fuller, T. A., Russchen, F. T., and Price, J. L. (1987). Sources of in adult rats. Neurosci. Lett. 163(1):36–40.
presumptive glutamergic/aspartergic afferents to the rat ven- Hasselmo, M. E. (1995). Neuromodulation and cortical function:
tral striatopallidal region. J. Comp. Neurol. 258(3):317–338. modeling the physiological basis of behavior. Behav. Brain
Funk, D., and Amir, S. (2000). Enhanced fos expression within the Res. 67(1):1–27.
primary olfactory and limbic pathways induced by an aversive Heimer, L. (1968). Synaptic distribution of centripetal and
conditioned odor stimulus. Neuroscience 98(3): 403–406. centrifugal nerve fibres in the olfactory system of the rat.
Gaykema, R. P., Luiten, P. G., Nyakas, C., and Traber, J. (1990). An experimental anatomical study. J. Anat. 103(3):
Cortical projection patterns of the medial septum-diagonal 413–432.
band complex. J. Comp. Neurol. 293(1):103–124. Heimer, L., and Wilson, R. D. (1975). The subcortical projections
Gellman, R. L., and Aghajanian, G. K. (1993). Pyramidal cells in of the allocortex: similarities in the neural connections of the
piriform cortex receive a convergence of inputs from hippocampus, the piriform cortex and the neocortex. In
monoamine activated GABAergic interneurons. Brain Res. Perspectives in Neurobiology, M. Santini (Ed.). Raven, New
600(1):63–73. York, pp. 177–193.
Gellman, R. L., and Aghajanian, G. K. (1994). Serotonin2 recep- Hoffman, W. H., and Haberly, L. B. (1993). Role of synaptic
tor-mediated excitation of interneurons in piriform cortex: excitation in the generation of bursting-induced epileptiform
antagonism by atypical antipsychotic drugs. Neuroscience potentials in the endopiriform nucleus and piriform cortex.
58(3):515–525. J. Neurophysiol. 70(6):2550–2561.
178 Cleland and Linster

Hunter, A. J., and Murray, T. K. (1989). Cholinergic mechanisms Kunzle, H., and Radtke-Schuller, S. (2000). Basal telencephalic
in a simple test of olfactory learning in the rat. regions connected with the olfactory bulb in a Madagascan
Psychopharmacology 99(2):270–275. hedgehog tenrec. J. Comp. Neurol. 423(4):706–726.
Inokuchi, A., Mooney, K. E., and Snow, J. B., Jr. (1987). Lanuza, E., and Halpern, M. (1998). Efferents and centrifugal
Dopaminergic modulation of bulbofugal projections in the rat afferents of the main and accessory olfactory bulbs in the
olfactory tubercle. Am. J. Otolaryngol. 8(4):214–218. snake Thamnophis sirtalis. Brain Behav. Evol. 51(1):1–22.
Inokuchi, A., Boran, T. V., Kimmelman, C. P., and Snow, J. B., Jr. Leveteau, J., Andriason, I., and Mac Leod, P. (1992). The bilater-
(1988). Effects of locus ceruleus and olfactory bulb stimula- al bulbar projections of the primary olfactory neurons in the
tion on rat olfactory tubercle neuron activity. Otolaryngol. frog. Exp. Brain Res. 89(1):93–104.
Head Neck Surg. 98(2):116–120. Leveteau, J., Andriason, I., and Mac Leod, P. (1993). Interbulbar
Jankovski, A., Garcia, C., Soriano, E., and Sotelo, C. (1998). reciprocal inhibition in frog olfaction. Behav. Brain Res.
Proliferation, migration and differentiation of neuronal 54(1):103–106.
progenitor cells in the adult mouse subventricular zone surgi- Levy, F., Meurisse, M., Ferreira, G., Thibault, J., and Tillet, Y.
cally separated from its olfactory bulb. Eur. J. Neurosci. (1999). Afferents to the rostral olfactory bulb in sheep with
10(12):3853–3868. special emphasis on the cholinergic, noradrenergic and sero-
Jones, B. E., Halaris, A. E., McIlhany, M., and Moore, R. Y. (1977). tonergic connections. J. Chem. Neuroanat. 16(4):245–263.
Ascending projections of the locus coeruleus in the rat. I. Axonal Linster, C., and Hasselmo, M. E. (2001). Neuromodulation and the
transport in central noradrenaline neurons. Brain Res. 127(1):1–21. functional dynamics of piriform cortex. Chem. Senses
Jung, M. W., Larson, J., and Lynch, G. (1990). Role of NMDA 26(5):585–594.
and non-NMDA receptors in synaptic transmission in rat pir- Linster, C., Garcia, P., Hasselmo, M. E., and Baxter, M. G. (2001).
iform cortex. Exp. Brain Res. 82(2):451–455. Selective loss of cholinergic neurons projecting to the olfactory
Karadi, Z., Oomura, Y., Nishino, H., and Aou, S. (1989). system increases perceptual generalization between similar, but
Olfactory coding in the monkey lateral hypothalamus: behav- not dissimilar, odorants. Behav. Neurosci. 115(4):826–833.
ioral and neurochemical properties of odor-responding neu- Lipton, P. A., Alvarez, P., and Eichenbaum, H. (1999).
rons. Physiol. Behav. 45(6):1249–1257. Crossmodal associative memory representations in rodent
Kay, L. M., and Freeman, W. J. (1998). Bidirectional processing orbitofrontal cortex. Neuron 22(2):349–359.
in the olfactory-limbic axis during olfactory behavior. Behav. Luskin, M. B., and Price, J. L. (1982). The distribution of axon
Neurosci. 112(3):541–553. collaterals from the olfactory bulb and the nucleus of the hor-
Kemali, M., and Guglielmotti, V. (1987). A horseradish peroxi- izontal limb of the diagonal band to the olfactory cortex,
dase study of the olfactory system of the frog, Rana escu- demonstrated by double retrograde labeling techniques.
lenta. J. Comp. Neurol. 263(3):400–417. J. Comp. Neurol. 209(3):249–263.
Kevetter, G. A., and Winans, S. S. (1981). Connections of the cor- Luskin, M. B., and Price, J. L. (1983a). The laminar distribution
ticomedial amygdala in the golden hamster. II. Efferents of of intracortical fibers originating in the olfactory cortex of the
the “olfactory amygdala”. J. Comp. Neurol. 197(1):99–111. rat. J. Comp. Neurol. 216(3):292–302.
Kier, E. L., Fulbright, R. K., and Bronen, R. A. (1995). Limbic Luskin, M. B., and Price, J. L. (1983b). The topographic organi-
lobe embryology and anatomy: dissection and MR of the zation of associational fibers of the olfactory system in the rat,
medial surface of the fetal cerebral hemisphere. Am. J. including centrifugal fibers to the olfactory bulb. J. Comp.
Neuroradiol. 16(9):1847–1853. Neurol. 216(3):264–291.
Kowianski, P., Lipowska, M., and Morys, J. (1999). The piriform Marek, G. J., and Aghajanian, G. K. (1994). Excitation of
cortex and the endopiriform nucleus in the rat reveal gener- interneurons in piriform cortex by 5-hydroxytryptamine:
ally similar pattern of connections. Folia Morphol. 58(1):9–19. blockade by MDL 100,907, a highly selective 5-HT2A recep-
Krettek, J. E., and Price, J. L. (1977a). Projections from the tor antagonist. Eur. J. Pharmacol. 259(2):137–141.
amygdaloid complex and adjacent olfactory structures to the Marek, G. J., and Aghajanian, G. K. (1995). Protein kinase C
entorhinal cortex and to the subiculum in the rat and cat. inhibitors enhance the 5-HT2A receptor-mediated excitatory
J. Comp. Neurol. 172(4):723–752. effects of serotonin on interneurons in rat piriform cortex.
Krettek, J. E., and Price, J. L. (1977b). Projections from the Synapse 21(2):123–230.
amygdaloid complex to the cerebral cortex and thalamus in Meisami, E., and Hamedi, S. (1986). Relative contribution of
the rat and cat. J. Comp. Neurol. 172(4):687–722. brain and peripheral connections to postnatal growth and cell
Krettek, J. E., and Price, J. L. (1978). A description of the amyg- accretion in the rat olfactory bulb. Brain Res. 394(2): 282–286.
daloid complex in the rat and cat with observations on intra- Mestre, N., Petter, A., and Bons, N. (1992). Systematisation of
amygdaloid axonal connections. J. Comp. Neurol. 178(2): the olfactory bulb efferent projections in a lemurian primate:
255–280. Microcebus murinus. J. Hirnforsch. 33(2):173–184.
Kucharski, D., and Hall, W. G. (1987). New routes to early mem- Meyer, R. P. (1981). Central connections of the olfactory bulb in
ories. Science 238(4828):786–788. the American opossum (Didelphys virginiana): a light micro-
Kucharski, D., and Hall, W. G. (1988). Developmental change in scopic degeneration study. Anat. Rec. 201(1):141–156.
the access to olfactory memories. Behav. Neurosci. Millhouse, O. E., and Heimer, L. (1984). Cell configurations in the
102(3):340–348. olfactory tubercle of the rat. J. Comp. Neurol. 228(4):571–597.
Central Olfactory Structures 179

Moore, R. Y., Halaris, A. E., and Jones, B. E. (1978). Serotonin impairment following horizontal diagonal band lesions.
neurons of the midbrain raphe: ascending projections. J. Behav. Neurosci. 107(1):72–81.
Comp. Neurol. 180(3):417–438. Royet, J. P., Hudry, J., Zald, D. H., Godinot, D., Gregoire, M. C.,
Mor, G., Saphier, D., and Feldman, S. (1987). Neural pathways Lavenne, F., Costes, N., and Holley, A. (2001). Functional
that mediate the effects of afferent stimuli on paraventricular neuroanatomy of different olfactory judgments. Neuroimage
nucleus multiunit activity in freely moving rats. J. Neurosci. 13(3):506–519.
Res. 17(4):452–458. Sallaz, M., and Jourdan, F. (1993). C-fos expression and 2-
Motokizawa, F., Yasuda, N., and Shimomura, F. (1988). The piriform deoxyglucose uptake in the olfactory bulb of odour-stimula-
cortex is not a direct olfactory relay to the mediodorsal thalamic ted awake rats. Neuroreport 4(1):55–58.
nucleus in cats and rabbits. Neurosci. Lett. 84(1):41–45. Sallaz, M., and Jourdan, F. (1996). Odour-induced c-fos expres-
Mouly, A. M., Fort, A., Ben-Boutayab, N., and Gervais, R. sion in the rat olfactory bulb: involvement of centrifugal affer-
(2001). Olfactory learning induces differential long-lasting ents. Brain Res. 721(1–2):66–75.
changes in rat central olfactory pathways. Neuroscience Sanides-Kohlrausch, C., and Wahle, P. (1990). Morphology of
102(1):11–21. neuropeptide Y-immunoreactive neurons in the cat olfactory
Northcutt, R. G., and Royce, G. J. (1975). Olfactory bulb projec- bulb and olfactory peduncle: postnatal development and
tions in the bullfrog Rana catesbeiana. J. Morphol. species comparison. J. Comp. Neurol. 291(3):468–489.
145(3):251–267. Savic, I., Gulyas, B., Larsson, M., and Roland, P. (2000).
Ojima, H., Mori, K., and Kishi, K. (1984). The trajectory of Olfactory functions are mediated by parallel and hierarchical
mitral cell axons in the rabbit olfactory cortex revealed by processing. Neuron 26(3):735–745.
intracellular HRP injection. J. Comp. Neurol. 230(1):77–87. Scalia, F., and Winans, S. S. (1975). The differential projections
Paolini, A. G., and McKenzie, J. S. (1993). Effects of lesions in of the olfactory bulb and accessory olfactory bulb in mam-
the horizontal diagonal band nucleus on olfactory habituation mals. J. Comp. Neurol. 161(1):31–55.
in the rat. Neuroscience 57(3):717–724. Scalia, F., Gallousis, G., and Roca, S. (1991). Differential projec-
Paxinos, G., and Watson, C. (1986). The Rat Brain in Stereotaxic tions of the main and accessory olfactory bulb in the frog.
Coordinates. Academic Press, San Diego. J. Comp. Neurol. 305(3):443–461.
Price, J. L. (1973). An autoradiographic study of complementary Schoenbaum, G., Chiba, A. A., and Gallagher, M. (1998).
laminar patterns of termination of afferent fibers to the olfac- Orbitofrontal cortex and basolateral amygdala encode expected
tory cortex. J. Comp. Neurol. 150(1):87–108. outcomes during learning. Nat. Neurosci. 1(2):155–159.
Price, J. L. (1985). Beyond the primary olfactory cortex: olfac- Schoenbaum, G., Chiba, A. A., and Gallagher, M. (2000).
tory related areas in the neocortex, thalamus and hypothalamus. Changes in functional connectivity in orbitofrontal cortex and
Chem. Senses 10:239–258. basolateral amygdala during learning and reversal training.
Price, J. L. (1987). The central olfactory and accessory olfactory J. Neurosci. 20(13):5179–5189.
systems. In Neurobiology of Taste and Smell, T. E. Finger and W. Schoenfeld, T. A., and Macrides, F. (1984). Topographic organi-
L. Silver (Eds.). John Wiley and Sons, New York, pp. 179–204. zation of connections between the main olfactory bulb and
Price, J. L., and Slotnick, B. M. (1983). Dual olfactory represen- pars externa of the anterior olfactory nucleus in the hamster.
tation in the rat thalamus: an anatomical and electrophysio- J. Comp. Neurol. 227(1):121–135.
logical study. J. Comp. Neurol. 215(1):63–77. Schwob, J. E., and Price, J. L. (1984). The development of axonal
Price, J. L., and Sprich, W. W. (1975). Observations on the later- connections in the central olfactory system of rats. J. Comp.
al olfactory tract of the rat. J. Comp. Neurol. 162(3):321–336. Neurol. 223(2):177–202.
Price, J. L., Slotnick, B. M., and Revial, M. F. (1991). Olfactory Scott, J. W. (1981). Electrophysiological identification of mitral
projections to the hypothalamus. J. Comp. Neurol. 306(3): and tufted cells and distributions of their axons in olfactory
447–461. system of the rat. J. Neurophysiol. 46(5):918–931.
Radtke-Schuller, S., and Kunzle, H. (2000). Olfactory bulb and Scott, J. W., McBride, R. L., and Schneider, S. P. (1980). The
retrobulbar regions in the hedgehog tenrec: organization and organization of projections from the olfactory bulb to the piri-
interconnections. J. Comp. Neurol. 423(4):687–705. form cortex and olfactory tubercle in the rat. J. Comp. Neurol.
Ramus, S. J., and Eichenbaum, H. (2000). Neural correlates of 194(3):519–534.
olfactory recognition memory in the rat orbitofrontal cortex. Scott, J. W., Ranier, E. C., Pemberton, J. L., Orona, E., and
J. Neurosci. 20(21):8199–8208. Mouradian, L. E. (1985). Pattern of rat olfactory bulb mitral
Ray, J. P., Russchen, F. T., Fuller, T. A., and Price, J. L. (1992). and tufted cell connections to the anterior olfactory nucleus
Sources of presumptive glutamatergic/aspartatergic afferents pars externa. J. Comp. Neurol. 242(3):415–424.
to the mediodorsal nucleus of the thalamus in the rat. J. Comp. Shammah-Lagnado, S. J., and Negrao, N. (1981). Efferent connec-
Neurol. 320(4):435–456. tions of the olfactory bulb in the opossum (Didelphis marsupialis
Reep, R. L., and Winans, S. S. (1982). Afferent connections of aurita): a Fink-Heimer study. J. Comp. Neurol. 201(1):51–63.
dorsal and ventral agranular insular cortex in the hamster Sheldon, P. W., and Aghajanian, G. K. (1990). Serotonin (5-HT)
Mesocricetus auratus. Neuroscience 7(5):1265–1288. induces IPSPs in pyramidal layer cells of rat piriform cortex:
Roman, F. S., Simonetto, I., and Soumireu-Mourat, B. (1993). evidence for the involvement of a 5-HT2-activated interneu-
Learning and memory of odor-reward association: selective ron. Brain Res. 506(1):62–69.
Sheldon, P. W., and Aghajanian, G. K. (1991). Excitatory responses
180 Cleland and Linster

to serotonin (5-HT) in neurons of the rat piriform cortex: evi- Tazawa, Y., Onoda, N., and Takagi, S. F. (1987). Olfactory input
dence for mediation by 5-HT1C receptors in pyramidal cells and to the lateral hypothalamus of the old world monkey.
5-HT2 receptors in interneurons. Synapse 9(3):208–218. Neurosci. Res. 4(5):357–375.
Shipley, M. T. (1995). Olfactory system. In The Rat Nervous Thanos, P. K., and Slotnick, B. M. (1997). Short-term odor memory:
System, 2nd ed., G. Paxinos (Ed.). Academic Press, San effects of posterior transection of the lateral olfactory tract in
Diego, pp. 899–928. the rat. Physiol. Behav. 61(6):903–906.
Shipley, M. T., and Adamek, G. D. (1984). The connections of the Tseng, G. F., and Haberly, L. B. (1989). Deep neurons in piriform
mouse olfactory bulb: a study using orthograde and retrograde cortex. I. Morphology and synaptically evoked responses
transport of wheat germ agglutinin conjugated to horseradish including a unique high-amplitude paired shock facilitation.
peroxidase. Brain Res. Bull. 12(6):669–688. J. Neurophysiol. 62(2):369–385.
Shipley, M. T., and Geinisman, Y. (1984). Anatomical evidence Turner, B. H., and Mishkin, M. (1978). A reassessment of the
for convergence of olfactory, gustatory, and visceral afferent direct projections of the olfactory bulb. Brain Res.
pathways in mouse cerebral cortex. Brain Res. Bull. 151(2):375–380.
12(3):221–226. Turner, B. H., and Zimmer, J. (1984). The architecture and some
Skeen, L. C., and Hall, W. C. (1977). Efferent projections of the of the interconnections of the rat’s amygdala and lateral peri-
main and the accessory olfactory bulb in the tree shrew allocortex. J. Comp. Neurol. 227(4):540–557.
(Tupaia glis). J. Comp. Neurol. 172(1):1–35. Turner, B. H., Gupta, K. C., and Mishkin, M. (1978). The locus
Slotnick, B. M. (1985). Olfactory discrimination in rats with and cytoarchitecture of the projection areas of the olfactory
anterior amygdala lesions. Behav. Neurosci. 99(5):956–963. bulb in Macaca mulatta. J. Comp. Neurol. 177(3):381–396.
Slotnick, B. M., and Berman, E. J. (1980). Transection of the lat- Valverde, F. (1965). Studies on the Piriform Lobe. Harvard
eral olfactory tract does not produce anosmia. Brain Res. Bull. University Press, Cambridge, MA.
5(2):141–145. Valverde, F., Lopez-Mascaraque, L., and De Carlos, J. A. (1989).
Slotnick, B. M., and Risser, J. M. (1990). Odor memory and odor Structure of the nucleus olfactorius anterior of the hedgehog
learning in rats with lesions of the lateral olfactory tract and (Erinaceus europaeus). J. Comp. Neurol. 279(4):581–600.
mediodorsal thalamic nucleus. Brain Res. 529(1–2):23–29. van Groen, T., and Wyss, J. M. (1990). Extrinsic projections from
Slotnick, B. M., and Schoonover, F. W. (1992). Olfactory path- area CA1 of the rat hippocampus: olfactory, cortical, subcor-
ways and the sense of smell. Neurosci. Biobehav. Rev. 16(4): tical, and bilateral hippocampal formation projections. J.
453–472. Comp. Neurol. 302(3):515–528.
Slotnick, B. M., and Schoonover, F. W. (1993). Olfactory sensi- Willey, T. J., Maeda, G., Schultz, R. L., Seibly, W. S., and
tivity of rats with transection of the lateral olfactory tract. Horowitz, J. M. (1983). The principal projection pathway
Brain Res. 616(1–2):132–137. between the olfactory bulb and the prepyriform cortex in the
Smithson, K. G., Weiss, M. L., and Hatton, G. I. (1989). cat. J. Neurosci. Res. 9(3):253–277.
Supraoptic nucleus afferents from the main olfactory bulb—I. Wirth, S., Ferry, B., and Di Scala, G. (1998). Facilitation of olfac-
Anatomical evidence from anterograde and retrograde tracers tory recognition by lateral entorhinal cortex lesion in rats.
in rat. Neuroscience 31(2):277–287. Behav. Brain Res. 91(1–2):49–59.
Sobel, N., Prabhakaran, V., Desmond, J. E., Glover, G. H., Wyss, J. M. (1981). An autoradiographic study of the efferent
Goode, R. L., Sullivan, E. V., and Gabrieli, J. D. (1998). connections of the entorhinal cortex in the rat. J. Comp.
Sniffing and smelling: separate subsystems in the human Neurol. 199(4):495–512.
olfactory cortex. Nature 392(6673):282–286. Wyss, J. M., and Sripanidkulchai, K. (1983). The indusium gri-
Sullivan, R. M., and Wilson, D. A. (1993). Role of the amygdala seum and anterior hippocampal continuation in the rat. J.
complex in early olfactory associative learning. Behav. Comp. Neurol. 219(3):251–272.
Neurosci. 107(2):254–263. Zaborszky, L., Carlsen, J., Brashear, H. R., and Heimer, L.
Sullivan, R. M., and Wilson, D. A. (1994). The locus coeruleus, (1986). Cholinergic and GABAergic afferents to the olfactory
norepinephrine, and memory in newborns. Brain Res. Bull. bulb in the rat with special emphasis on the projection neurons
35(5–6):467–472. in the nucleus of the horizontal limb of the diagonal band. J.
Sullivan, R. M., Zyzak, D. R., Skierkowski, P., and Wilson, D. A. Comp. Neurol. 243(4):488–509.
(1992). The role of olfactory bulb norepinephrine in early Zald, D. H., and Pardo, J. V. (1997). Emotion, olfaction, and the
olfactory learning. Brain Res. Dev. Brain Res. 70(2):279–282. human amygdala: amygdala activation during aversive
Sutherland, R. J., and McDonald, R. J. (1990). Hippocampus, amyg- olfactory stimulation. Proc. Natl. Acad. Sci. USA 94(8):
dala, and memory deficits in rats. Behav. Brain Res. 37(1):57–79. 4119–4124.
Takagi, S. F. (1986). Studies on the olfactory nervous system of Zald, D. H., and Pardo, J. V. (2000). Functional neuroimaging of the
the Old World monkey. Prog. Neurobiol. 27(3):195–250. olfactory system in humans. Int. J. Psychophysiol. 36(2): 165–181.
Tanabe, T., Iino, M., and Takagi, S. F. (1975). Discrimination of Zhang, Y., Burk, J. A., Glode, B. M., and Mair, R. G. (1998).
odors in olfactory bulb, pyriform-amygdaloid areas, and Effects of thalamic and olfactory cortical lesions on continu-
orbitofrontal cortex of the monkey. J. Neurophysiol. 38(5): ous olfactory delayed nonmatching-to-sample and olfactory
1284–1296. discrimination in rats (Rattus norvegicus). Behav. Neurosci.
112(1):39–53.
9

Sensory Physiology of Central Olfactory Pathways

Donald A. Wilson and Regina M. Sullivan

University of Oklahoma, Norman, Oklahoma, U.S.A.

I. INTRODUCTION help clarify this issue (Singer and Shepherd, 1994), it does
appear that, similar to other sensory systems, odorant stim-
Central sensory pathways construct representations of the uli are broken down into component features, each recog-
external world based on a combination of spatiotemporal nized by a particular receptor, and the problem for the
patterns of receptor neuron input and a running average of remainder of the olfactory pathway is to reconstruct those
internal activity patterns. In most sensory systems, the features into a perceptual whole.
relationship between stimulus energy in the external world In addition to discriminating pure, isolated stimuli, a
and the spatiotemporal pattern of receptor neuron activity problem for all sensory systems is that they must function in
appears relatively straightforward. For example, spatial the real world. Thus, the visual system is able to recognize
relationships of visual stimuli are maintained by spatial a stimulus partially obscured by other objects and the audi-
patterns of visual receptor cell activity in the retina and tory system can interpret speech against a background of
subsequent precise retinotopic projections to visual corti- other noises. Similarly, the olfactory system is able to rec-
cal centers. Similarly, auditory stimulus frequency infor- ognize garlic in the spaghetti sauce even when the odors
mation is extracted by a spatial gradient of frequency from the freshly cut lawn are blowing in the window. On the
sensitivity along the basilar membrane of the cochlea and other hand, that garlic odor is a mixture of many individual
subsequent precise tonotopic projections to auditory corti- molecular components, yet is perceived as a single stimulus.
cal centers. Lateral inhibition along both the visual and This review will focus on what is currently known
auditory sensory pathways helps to more precisely define about the sensory physiology of central olfactory struc-
the specific visual spatial pattern or auditory frequency of tures that allows odorant discrimination and odorant mix-
the initiating stimulus. ture analysis and synthesis to occur (see Chapters 7 and 8
However, how the olfactory system constructs a repre- for the detailed anatomy of these structures). This review
sentation of the external odor world is not so obvious. will focus on terrestrial vertebrates, although striking sim-
Simple analytical chemistry does not appear to be suffi- ilarities with invertebrates will be noted (Christensen and
cient to account for olfactory perception. Molecules that White, 2000). While excellent work has been done on
are structurally very similar may be perceptually very dif- synaptic physiology of this system, much of which utilized
ferent, and vice versa. Furthermore, it is not clear at in vitro preparations (see Haberly, 1998; Shepherd and
present which features of olfactory stimuli the olfactory Greer, 1998; Shipley and Ennis, 1996 for reviews), this
system uses for odorant discrimination (e.g., carbon chain review will focus on in vivo sensory physiology in verte-
length, presence and location of functional groups, molecu- brates and response to odorants.
lar resonant frequency). While further analysis of ligand- The central olfactory system of vertebrates (Fig. 1)
receptor interactions at the olfactory receptor sheet should includes the main olfactory bulb and the primary olfactory

181
182 Wilson and Sullivan

II. MAIN OLFACTORY BULB

A. Glomerular Layer

Olfactory receptor axons synapse onto second-order

olfactory neurons within the main olfactory bulb
(Fig. 2). As described elsewhere in this volume, single
receptor neurons appear to express a single receptor
protein. Receptor neurons expressing the same receptor
protein, while randomly scattered within one of four
zones of the olfactory receptor sheet, converge on two
individual glomeruli within the olfactory bulb, one
located dorsomedially in the bulb and one more ventro-
laterally (Buck, 1996; Mombaerts, 1999). Thus, each of
the approximately 2000 glomeruli of the rodent olfac-
tory bulb is believed to each receive relatively homoge-
neous input from neurons expressing one of the 1000
different receptor proteins. Furthermore, receptor neu-
rons expressing similar or homologous receptor genes
(Tsuboi et al., 1999), and with similar odorant receptive
fields (Bozza and Kauer, 1998), tend to terminate in
neighboring glomeruli, enhancing the possibility of lat-
eral inhibitory interactions between similar molecular
Figure 1 Basic schematized organization of the vertebrate features. The high convergence ratio of olfactory
olfactory system. Circled structures receive direct input from the receptor neurons to mitral cells within a glomerulus
main olfactory bulb. Note that most areas receiving direct input
(1000: 1) significantly amplifies sensitivity of the system
from the main olfactory bulb project back to the bulb. Modulatory
by reducing odorant response threshold in mitral cells
inputs project broadly to all primary olfactory structures,
although there is substantial heterogeneity in laminar density of compared to olfactory receptor neurons (Duchamp-Viret
terminations within each area. Abbreviations: OB, main olfactory et al., 1989)
bulb; AON, anterior olfactory nucleus; Amy, amygdala; PC, piri- Vertebrate olfactory receptor neurons have relatively
form cortex; Ent, entorhinal cortex; PFC, prefrontal cortex; Hyp, broad odorant receptive fields (Duchamp-Viret et al.,
hypothalamus; DMN, dorsomedial nucleus of the thalamus; LC, 1999; Kaluza and Breer, 2000; Malnic et al., 1999; Sato
locus coeruleus, NE, norepinephrine; HLDB, horizontal limb of et al., 1994; Sicard and Holley, 1984). Similarly, individ-
the diagonal band of Broca; ACh, acetylcholine; Raphe, raphe ual main olfactory bulb glomeruli respond to multiple
nucleus; 5-HT, 5-hydroxytryptamine (serotonin). odorants, although each odorant produces a unique
spatial pattern of glomerular activation as determined by
2-deoxyglucose autoradiography (Johnson and Leon,
2000; Johnson et al., 1998, 1999; Jourdan et al., 1980;
cortex (piriform cortex). In mammals, a higher-order Stewart et al., 1979), c-fos immunohistochemistry
olfactory cortex exists, the orbitofrontal/insular cortex. (Guthrie et al., 1993; Sallaz and Jourdan, 1993), and
The thalamic relay to the olfactory orbitofrontal cortex is optical imaging (Joerges et al., 1997; Rubin and Katz,
the dorsomedial nucleus of the thalamus, although a 1999; Uchida et al., 2000). A recent optical imaging
direct projection from the piriform cortex to the study of intrinsic signals in the rat revealed that the spe-
orbitofrontal cortex also exists (see Chapter 8). While cific functional group present in an odorant determined
odorant responses have been examined in a number of the glomerular zone of activation (e.g., anteromedial or
other central brain regions, such as the amygdala (Cain dorsolateral), while more subtle features of the odorant
and Bindra, 1972; Schoenbaum et al., 1999; Tanabe et molecule (e.g., carbon chain length or branching pattern)
al., 1975a) and hypothalamus (Karadi et al., 1989; determined which glomeruli within that zone would be
Kogure and Onoda, 1983; Pfaff and Gregory, 1971; Scott activated (Uchida et al., 2000). This spatial pattern of
and Pfaffmann, 1972), the sensory physiology of the glomerular activation is believed to encode the molecular
main olfactory bulb, piriform cortex, and orbitofrontal features present in the sampled odorant. However, while
cortex will be emphasized here. individual odorant features may be encoded by individual
Sensory Physiology of Central Olfactory Pathways 183

Figure 2 Basic schematized neural connectivity of the main olfactory bulb and piriform cortex. Individual receptors within the olfac-
tory epithelium express one of 1000 different receptor proteins and are randomly scattered within one of four zones, yet receptors express-
ing the same receptor protein converge onto a small number of exclusive glomeruli (three receptor types are labeled A, B, and C in this
example). The receptors are hypothesized to be responsive to individual odorant features, rather than odorant molecules as a whole. Mitral
cells receive receptor input from a single glomerulus (and thus convergent receptor input; e.g., A or B) and project to the piriform cor-
tex. Within the olfactory bulb, interglomerular and interoutput neuron lateral inhibition is mediated by juxtaglomerular and granule cells,
respectively, heightening contrast between similar odorant features. Neurons in the piriform cortex appear to form a combinatorial array,
allowing convergence of multiple odorant features (e.g., AB or ABC) and/or behavioral state/nonolfactory inputs to occur on single neu-
rons. Both the olfactory bulb and piriform cortex receive extensive input from neuromodulatory and nonolfactory inputs.

glomeruli, odorants in a mixture can interact at the Many juxtaglomerular cells express both the inhibitory
receptor level and/or within the glomerular layer to pro- amino acid neurotransmitter GABA and dopamine (Gall
duce odorant mixture specific glomerular activation in et al., 1987; Kosaka et al., 1985). Juxtaglomerular cells
both vertebrates (Bell et al., 1987) and invertebrates respond to odorants with simple depolarizations and
(Cromarty and Derby, 1998; Derby; et al., 1991; Joerges bursts of spikes and may directly mirror olfactory nerve
et al., 1997). Thus, some aspects of odor synthesis may input (Onoda and Mori, 1980; Wellis and Scott, 1990).
occur even before the first central synapse of the olfac- One role of juxtaglomerular cell GABA release may be to
tory pathway. presynaptically inhibit glutamate release from olfactory
Within glomeruli, olfactory receptor axons synapse nerve axons (Aroniadou-Anderjaska et al., 2000; Nickell
onto the primary output neurons of the olfactory bulb, et al., 1994). Mitral/tufted cells also express GABA recep-
mitral cells, as well as onto a second class of output neu- tors (Bowery et al., 1987), and thus, juxtaglomerular cell
rons, tufted cells (Fig. 2). Juxtaglomerular cells, a class of activation could mediate either lateral or feedforward
olfactory bulb interneurons that mediate interglomerular inhibition of these output neurons. In the frog, activation
inhibition, also receive direct olfactory nerve input. of either GABAB receptors or dopamine D2 receptors in
Olfactory receptor axons release the excitatory amino acid the glomerular/external plexiform layers results in a
glutamate from their axon terminals and activate both decrease in mitral cell spontaneous activity with a sparing
NMDA and non-NMDA receptors on second-order neuron of odorant-evoked activity (Duchamp-Viret et al., 1997,
dendrites (Berkowicz et al., 1994; Ennis et al., 1996). 2000). These results suggest that one role of inhibition in
184 Wilson and Sullivan

the glomerular layer may be to increase the signal-to- B. Olfactory Bulb Output Neurons
noise ratio of bulb output and thus odor saliency. In the
rat, dopamine D2 receptors are located on presynaptic In the rat, mitral cells extend an apical dendrite into a single
olfactory receptor cell axons (Coronas et al., 1997; Koster glomerulus, with each glomerulus innervated by approxi-
et al., 1999; Nickell et al., 1991). Stimulation of dopamine mately 25 mitral cells (Fig. 2) (Shepherd and Greer, 1998).
receptors reduces olfactory nerve evoked potentials in Mitral cells respond to olfactory nerve input with both a fast
olfactory bulb (Gurski and Hamilton, 1996; Hsia et al., AMPA receptor–mediated depolarization and a slower,
1999; Nowycky et al., 1983), and more specifically acti- NMDA receptor–mediated depolarization (Berkowicz et al.,
vation of D2 receptors in rat reduces glomerular layer 1994; Ennis et al., 1996). Mitral cell responses to odorants
odorant-evoked spatial patterns of 2-deoxyglucose uptake are generally more complex than the simple responses
(Sallaz and Jourdan, 1992). In contrast, D2 receptor block- described for juxtaglomerular neurons above, reflecting the
ade or reduction in olfactory bulb dopamine content additional circuit processes affecting these cells.
enhances and blurs odorant-specific glomerular activation Intracellular recordings of mitral/tufted cell responses to
(Guthrie et al., 1990) and increases mitral/tufted cell odorants reveal prominent short- and long-latency hyperpo-
responsiveness to odorants (Wilson and Sullivan, 1995). larizations, in addition to the depolarization and evoked
In accordance with these physiological results, the D2 spikes presumably mediated by direct glutamatergic excita-
receptor agonist quipirole reduces odor detection perfor- tion from the olfactory nerve (Hamilton and Kauer, 1985,
mance in a dose-dependent manner (Doty and Risser, 1989; Wellis et al., 1989). Similar multiphasic membrane
1989). Interestingly, systemic injection of the D1 receptor potential responses to odorants have been observed with
agonist SKF38393 enhances odor-detection performance intracellular recordings from invertebrate antennal lobe neu-
(Doty et al., 1988). rons (Christensen et al., 1998).
Juxtaglomerular cell dopamine expression is highly Low-intensity odorant stimulation within the mitral cell
odorant experience dependent. Olfactory bulb dopamine odorant-receptive field evokes a low-amplitude depolariza-
levels increase following brief odorant exposure tion that may be suprathreshold for spike initiation
(Coopersmith et al., 1991), while odorant deprivation (Hamilton and Kauer, 1989; Wellis et al., 1989). In sala-
significantly reduces bulb dopamine content (Brunjes et al., mander, this depolarization is frequently preceded by a
Wilson and Wood, 1992) via an experience-dependent brief hyperpolarization (Hamilton and Kauer, 1989). As
decrease in tyrosine hydroxylase expression (Baker, stimulus intensity increases, the amplitude of the odorant-
1990; Baker et al., 1993; Kosaka et al., 1987; Puche and evoked depolarization increases and latency decreases,
Shipley, 1999). Given the described effects of dopamine resulting in a high-frequency burst of spikes. This burst is
on odorant responses, glomerular layer dopamine may then followed by a second period of hyperpolarization that
function as a form of experience-dependent volume con- can last several hundreds of milliseconds under artificial
trol—during periods of intense odorant stimulation, respiration conditions. As stimulus intensity increases
dopamine may suppress olfactory nerve input, perhaps further, the second period of hyperpolarization begins to
to maintain bulb activity within an optimal dynamic truncate the evoked spike burst, in some cases leading to a
range. During periods of weak odorant stimulation, single, short-latency evoked spike followed by hyperpolar-
dopamine levels fall to enhance sensitivity of the system. ization in response to high-intensity odorant. These mem-
This enhanced sensitivity, however, comes at the price brane potential results correspond well with extracellular
of a decrease in glomerular and mitral/tufted cell odor- spike train recordings in a variety of terrestrial species
ant discrimination (Guthrie et al., 1990; Wilson and (Chaput and Holley, 1985; Duchamp-Viret and Duchamp,
Sullivan, 1995). A strikingly similar dopaminergic 1997; Harrison and Scott, 1986; Imamura et al., 1992;
mechanism of gain control exists in the vertebrate retina. Kauer, 1974; Mair et al., 1982; Mathews, 1972; Meredith,
Dark adaptation leads to changes in dopamine release 1986; Scott, 1977).
and a reduction in lateral inhibition in the retina, which Thus, in response to a single odorant pulse, a triphasic
increases sensitivity but reduces spatial resolution (Daw membrane potential response can be observed in mitral/tufted
et al., 1989). cells. Odorant intensity appears to be encoded by a rate
The olfactory bulb glomerular layer thus creates an code and/or a latency code, with responses to high-intensity
odorant-specific spatial feature map through precise odorants often consisting of a single spike followed by
projection patterns of olfactory receptor axons, while inhi- inhibition. Given the short latency of the initial hyperpolar-
bition in the glomerular layer both acts as an experience- ization, it is assumed to be mediated by juxtaglomerular
dependent gain control and allows sharpening of the neurons in a feedforward manner. The initial depolarization
odorant-specific spatial patterns. is mediated by AMPA and NMDA receptor activation on
Sensory Physiology of Central Olfactory Pathways 185

apical dendritic tufts of mitral cells. There is also recent Odorant stimulation can either enhance the spontaneous
anatomical (Allen and Hamilton, 2000) and physiological patterning of a single cell, or shift cell activity to a dif-
(Aroniadou-Anderjaska et al., 1999; Friedman and ferent phase of the respiratory cycle (Chalansonnet and
Strowbridge, 2000; Isaacson, 1999) evidence for gluta- Chaput, 1998). The respiratory entrainment of activity
matergic mitral-mitral cell excitation and/or autoexcitation. during odorant stimulation is stable over a wide range of
These mitral-mitral cell connections could contribute to the odorant concentrations, despite potential changes in lat-
synchrony observed in odorant responses of neighboring eral inhibition discussed above (Chalansonnet and
mitral cells (Buonviso et al., 1992; Kashiwadani et al., 1999; Chaput, 1998). The effects of active sniffing (i.e., an
Stopfer et al., 1997), which could also contribute to an inten- increase in inhalation rate to 5–10 Hz during exploration
sity code as well as play an important role in odorant quality and arousal) on odorant response patterns has not been
coding, as discussed below. thoroughly examined in vertebrates, although it is
The late-onset, slow hyperpolarization is mediated by assumed to modify odorant access to the receptor sheet
GABAergic granule cell interneurons. Mitral and tufted (Dethier, 1987; Youngentob et al., 1987) and appears to
cells connect with granule cells via dendrodendritic recip- modify granule cell–mediated inhibition in the bulb
rocal synapses along mitral/tufted cell lateral dendrites (Young and Wilson, 1999). More attention has been paid
(Shepherd and Greer, 1998). Glutamate released by to effects of odorant stimulation frequency in inverte-
mitral/tufted cell dendrites excites AMPA and NMDA brates (Christensen and Hildebrand, 1988; Gomez et al.,
receptors on granule cells (Chen et al., 2000; Isaacson and 1999; Loudon and Koehl, 2000; Schneider et al., 1998).
Strowbridge, 1998; Jacobson et al., 1986; Schoppa et al., In sphinx moths stimulated with puffs of odorant at dif-
1998; Trombley and Westbrook, 1990; Wilson et al., ferent rates, antennal lobe neuron response patterns var-
1996), which in turn release GABA back onto mitral/tuft- ied significantly, with some cells able to make discrete
ed cell dendrites. Mitral cell lateral dendrites can extend responses to odorant pulses at stimulation frequencies as
for up to 1 mm around the olfactory bulb, and thus may high as 10 Hz (Christensen and Hildebrand, 1988). Given
contact many granule cells. The granule cells are believed the ubiquity of active sniffing during exploration across
to perform lateral inhibitory functions, with GABAergic animal species (Dethier, 1987), additional research into
synapses on distal lateral dendrites, perhaps primarily the consequences of variations in stimulus frequency on
functioning to reduce backpropagation of spikes along peripheral and central odorant coding seems warranted.
these extended dendrites, rather than directly influencing For example, olfactory cortical targets of mitral cells
spike initiation at the initial segment. Direct evidence for must be able to discriminate between a mitral cell weakly
such lateral inhibitory actions comes from in vitro studies excited by an odorant inhaled at normal respiration rates
showing that inhibitory post-synaptic currents (IPSCs) can (perhaps evoking a short spike train at 10 Hz), from a
be evoked in both mitral cells and tufted cells by electrical mitral cell activated by an intense odorant while sniffing
stimulation of distant glomeruli (Christie et al., 2001). (perhaps evoking a single spike on each inhalation with
Tufted cells are influenced by a more narrow region of inhalations occurring at 10 Hz).
glomerular input (glomerular distances up to 400 ) than Of course, in addition to detecting odorants and encod-
mitral cells (glomerular distances up to 800 ), which have ing odorant intensity, mitral/tufted cells encode odorant
much longer lateral dendrites (Christie et al., 2001). This, quality/identity. Odorant quality appears to be encoded by
along with other structural differences (Ezeh et al., 1993; variations in odorant/molecular receptive fields of individ-
Macrides et al., 1985; Orona et al., 1983, 1984; Scott, ual mitral/tufted cells and spatial clustering of cells with
1981), suggests a potential important functional difference similar receptive fields within the olfactory bulb. As with
between the two principal bulb output neurons, although olfactory receptor neurons (Bozza and Kauer, 1998;
no detailed comparisons of odorant evoked activity have Malnic et al., 1999; Sato et al., 1994), odorant-receptive
been made between these two cell types. fields of mitral/tufted cells are based on responsiveness to
In addition to the phasic nature of the response within molecular features rather than to an odorant as a whole.
a single odorant pulse, single-unit studies in freely Odorants within the receptive field of an individual
breathing animals demonstrate a strong respiratory cycle mitral/tufted cell evoke excitatory/suppressive changes in
modulation of mitral/tufted cell activity (Chalansonnet firing rate, generally in phase with the respiratory cycle, as
and Chaput, 1998; Macrides and Chorover, 1972; Ogawa, described above. Because of the spatial clustering of cells
1998; Onoda and Mori, 1980; Pager, 1985). Mitral/tufted with similar receptive fields and the lateral inhibitory net-
cell spontaneous activity generally oscillates with the res- works described above, however, mitral/tufted cell recep-
piratory cycle, with different cells maximally active at tive fields may be more focused or precise than receptor
different phases of the cycle (inspiration or expiration). neurons.
186 Wilson and Sullivan

Individual mitral/tufted cells respond to many odorants more likely to respond similarly to odorants, while cells
(Duchamp-Viret and Duchamp, 1997; Harrison and Scott, more distant (150 ) are more likely to respond differ-
1986; Imamura et al., 1992; Katoh et al., 1993; Kauer, ently (Buonviso and Chaput 1990; Meredith, 1986; Wilson
1974; Mair et al., 1982; Mathews, 1972; Meredith, 1986; and Leon, 1987). For example, simultaneous recordings
Mori et al., 1992). The receptive field appears to include from pairs of mitral/tufted cells reveal that if a mitral/tufted
odorants that share a similar molecular feature (carbon cell is excited by amyl acetate, most cells within 100 
chain length or functional group), although blend-or of that cell will also be excited, while cells 150  will
mixture-specific neurons have been identified in the most likely be inhibited or nonresponsive (Buonviso and
invertebrate antennal lobe (Vickers et al., 1998). Using a Chaput 1990).
homologous alkane odorant series, cross-habituation stud- Furthermore, cells stimulated simultaneously with
ies demonstrate that habituation of mitral/tufted cell odorant in their receptive fields tend to synchronize their
responses to one odorant within its receptive field signifi- firing (Buonviso et al., 1992; Kashiwadani et al., 1999).
cantly suppresses responses to other receptive field odor- Given that individual odors are composed of many fea-
ants (Wilson, 2000b), strongly suggesting that mitral/tufted tures, each of which activates glomeruli at some distance
cell responses to multiple odorants are mediated by a single from each other, this synchronization of co-active neurons
input. could be critical for binding of the features into perceptual
Odorant receptive fields of mitral/tufted cells appear to be wholes by higher-order neurons (see below). Granule
organized in a roughly center-surround fashion (Meredith, cell–mediated feedback/lateral inhibition is again implicated
1986; Wilson and Leon, 1987). Using a stimulus set of in this synchronization (Bressler and Freeman, 1980;
homologous odorants varying in carbon chain length, Buonviso et al., 1996; Kashiwadani et al., 1999; Rall et al.,
individual mitral/tufted cells are excited by a range of 1966). Similar observations have been made in the inverte-
chain lengths (Imamura et al., 1992; Katoh et al., 1993; Mori brate olfactory system (Laurent, 1999, Wehr and Laurent,
et al., 1992) and inhibited by neighboring longer or shorter 1996). Desynchronizing antennal lobe output neurons with
chain lengths (Yokoi et al., 1995). This inhibitory surround local infusion of GABA antagonists impairs behavioral
is largely due to granule cell mediated lateral inhibition odorant discrimination by honey bees (Stopfer et al., 1997).
and can be reduced by GABA receptor antagonists (Yokoi In summary, mitral and tufted cells express odorant
et al., 1995). receptive fields for molecular features, similar to that
The excitatory region of the receptive field is believed described for olfactory receptor neurons. Receptive field
to be largely dependent on the glomerulus from which that characteristics are largely driven by the specific glomeru-
cell receives its afferent input. Thus, just as there are odor- lus from which the cell derives its afferent input, and thus,
ant-specific spatial patterns of glomerular activation noted the specific receptive field expressed by a mitral/tufted cell
above, there are spatial patterns (or differential spatial is largely dependent on that cell’s location in the olfactory
responsiveness) of mitral/tufted cell odorant–evoked unit bulb. The receptive field appears to include odorants shar-
activity (Imamura et al., 1992; Katoh et al., 1993; Kauer ing a common molecular feature. Cells near to each other
and Moulton, 1974; Mori and Yoshihara, 1995; Wilson and have similar receptive fields and are under lateral inhibi-
Leon, 1988) and local field potential activity (Adrian, tory influences from neighboring glomeruli-output neuron
1953; Freeman and Skarda, 1985; Viana DiPrisco and groups. Odorant responses consist of excitatory-inhibitory
Freeman, 1985). For example, mitral/tufted cells connected sequences, which are significantly shaped by both odorant
to glomeruli in the dorsomedial region of the olfactory intensity and quality. Respiration parses the response into
bulb have receptive fields that include aliphatic acids and 100–500 ms long components depending on respiration
exclude alkanes, while cells in the ventrolateral olfactory rate. Within these respiratory cycles, activity is further
bulb have receptive fields that include alkanes and exclude organized by synchronization of simultaneously firing
aliphatic acids (Imamura et al., 1993; Katoh et al., 1993; mitral/tufted cells.
Mori and Yoshihara, 1995).
In addition to global variation in odorant receptive field C. Modulation and Nonolfactory Responses
characteristics, local circuit interactions produce more
regional variations in odorant receptive fields. Mammalian Although mitral/tufted cells in the main olfactory bulb are
glomeruli are approximately 100–150  in diameter and second-order neurons in the olfactory system, they are
include apical dendrites of around 25 mitral cells (Royet already heavily influenced by both current behavioral state
et al., 1989; Shepherd and Greer, 1998). Mitral/tufted cells and past odorant experience. The olfactory bulb receives
physically near each other, and thus likely to receive input massive centrifugal inputs from a variety of olfactory and
from the same glomerulus (Buonviso et al., 1991a), are nonolfactory structures (Shepherd and Greer, 1998).
Sensory Physiology of Central Olfactory Pathways 187

Centrifugal inputs include acetylcholine (ACh) from the Sullivan, 1995). Olfactory associative conditioning also
horizontal limb of the diagonal band, norepinephrine from modifies subsequent glomerular (Coopersmith and Leon,
the locus coeruleus, and serotonin from the raphe nucleus, 1984; Johnson et al., 1995; Sullivan and Leon, 1986),
as well as strong feedback from olfactory cortical areas mitral/tufted cell (Wilson et al., 1987), granule cell (Woo
(feedback from olfactory cortical areas constitutes 80% of et al., 1996), and local field potential responses (Viana
centrifugal inputs to the bulb) (Haberly, 1998). DiPrisco and Freeman, 1985) to the learned odorant.
One of the initial paradigms demonstrating behavioral Associative learning during early development enhances
state modulation of olfactory bulb odorant responsiveness odorant-specific focal glomerular 2-deoxyglucose uptake
described food-deprivation effects on responses to food to that odorant (Coopersmith and Leon, 1984; Sullivan and
odor (Pager et al., 1972). Multiunit and single-unit record- Leon, 1986). Furthermore, mitral/tufted single units near
ings of mitral/tufted cells in awake rats revealed that these modified glomeruli display enhanced inhibitory
responses to food odor or odors associated with food were responses selectively to the learned odorant, while cells
greater in food-deprived rats than in satiated rats (Pager, distant to those glomeruli do not (Wilson and Leon, 1988;
1974, 1983; Pager et al., 1972). Deprivation state had no Wilson et al., 1987). These changes in olfactory bulb physi-
effect on responses to novel odorants (Pager, 1972, 1983). ology require co-activation of centrifugal noradrenergic
The enhanced responsiveness to food odor in deprived rats input from the locus coeruleus during odorant exposure for
appears to be related to a state-dependent reduction in induction (Sullivan et al., 1989). The mitral/tufted cell
habituation to the food odor mediated by centrifugal inputs response modification has been hypothesized to be due to
to the bulb (Gervais and Pager, 1983). Lesions of centrifu- learning-induced changes in granule cell–mediated den-
gal input to the bulb (olfactory peduncle cut) eliminate the drodendritic inhibition (Wilson and Sullivan, 1994).
deprivation-induced modulation of responses to food odor Similar norepinephrine-dependent, learning-induced
(Pager, 1978). changes have been observed in odorant-evoked spatiotem-
Similar behavioral state or nonolfactory modulation of poral olfactory bulb local field potentials in adult animals
mitral/tufted cell unit activity (Garcia-Diaz et al., 1985; (Viana DiPrisco and Freeman, 1985) and in the accessory
Jiang et al.,1996; Kay and Laurent, 1999; Nickell and olfactory bulb (Brennan and Keverne, 1997)
Shipley, 1988; Potter and Chorover, 1976; Scott, 1977; In summary, despite being the first central relay for
Wilson and Sullivan, 1990) or olfactory local field poten- olfactory information, a variety of nonolfactory signals
tials (Chabaud et al., 2000; Viana DiPrisco and Freeman, converge on olfactory bulb neurons to allow dynamic modu-
1985) has been demonstrated in other paradigms. lation of odorant processing, as well as more permanent
Activation of centrifugal inputs to the main olfactory odorant memories. In fact, even the first synapse of the
bulb can hyperpolarize mitral cells (e.g., anterior commis- olfactory pathway between olfactory receptors and second-
sure) (Nakashima et al., 1978), enhance mitral/tufted cell order neurons is capable of experience-dependent plasticity
spontaneous activity (e.g., norepinephrine) (Wilson and (e.g., LTP) (Ennis et al., 1998) and neuromodulation that
Sullivan, 1991); suppress spontaneous activity (e.g., acetyl- can shape spatial and temporal odorant-response patterns.
choline) (Nickell and Shipley, 1988), or enhance mitral/tufted
cell responsiveness to weak afferent input (e.g., norepineph-
rine) (Jiang et al., 1996). Olfactory bulb output and respon- III. PIRIFORM CORTEX
siveness to odorants, therefore, is under constant dynamic
regulation by centrifugal inputs responsive to behavioral A detailed description of the anatomy and synaptic physi-
state and nonolfactory events. Thus, as in other sensory ology of the piriform is outside the scope of this review,
systems, olfactory bulb responses to odorants in behaving but several excellent reviews exist (Bower, 1991; Haberly,
animals is a reflection not only of odorant quality and 1998; Lynch, 1986). What follows is a brief introduction to
quantity, but also of the context and state of the receiving the functional organization of the piriform cortex followed
animal. by a description of what is known about the sensory phys-
Finally, mitral/tufted cell odorant–response patterns are iology of the piriform cortex.
modulated not only by current conditions, but also past Mitral/tufted cell axons project via the lateral olfactory
odorant experience and olfactory learning. As mentioned tract to the olfactory cortex, which is composed of several
above, periods of reduced odorant stimulation cause a structures including the anterior olfactory nucleus, a major
decrease in glomerular layer dopamine, which, upon sub- source of commissural connections in the olfactory sys-
sequent return of odorant input, enhances glomerular and tem, and the piriform cortex (Fig. 2). While the anterior
mitral/tufted cell responses to odorant at the expense of and posterior regions of the piriform cortex appear to be
odorant discrimination (Guthrie et al., 1990; Wilson and both structurally (Haberly, 1998; Johnson et al., 2000) and
188 Wilson and Sullivan

functionally (Chabaud et al., 2000; Haberly, 1998; Illig encoding individual molecular features could in turn acti-
and Haberly, 2000; Litaudon and Cattarelli, 1995; vate coincidence detecting pyramidal cells of the piriform
Litaudon et al., 1997a; Mouly et al., 1998; Wilson and cortex, each maximally responsive to a particular combi-
Bower, 1992) quite distinct, there are several basic charac- nation of features. Broadly dispersive intracortical associ-
teristics of piriform cortical functional organization that ation fibers further contribute to this associational network
apply to the entire structure. The piriform cortex is a (Haberly, 1998; Haberly and Price, 1978; Johnson et al.,
relatively simple, three-layered cortical structure with 2000)
pyramidal cell bodies arranged in a tight Layer II and more If the combinatorial array model of piriform function is
dispersed in Layer III. Dendrites of both groups of pyra- correct, then odorant-receptive fields of cortical pyramidal
midal cells extend into Layer I, where mitral/tufted cell cells might, at least superficially, appear very similar to
axons terminate on approximately the most distal half. The odorant-receptive fields of mitral/tufted cells, although
proximal half of the dendritic tree receives association and with the two cell classes responding to odorants for differ-
commissural input from other regions of the olfactory cor- ent reasons. That is, as discussed above, a particular odor-
tex. Both the afferent input via the lateral olfactory tract ant may be composed of several features. A mitral cell may
and the commissural/association fibers are glutamatergic, respond to that odorant, and similar odorants, because of
and cortical pyramidal cells express both NMDA and non- the presence of a single feature that dominates the receptor
NMDA receptor types. GABAergic inhibitory interneu- input to that mitral cell. A cortical pyramidal neuron, on
rons are located in both Layers I and III. the other hand, may respond to that odorant, and similar
Similar to mitral/tufted cells, piriform cortex neurons dis- odorants, because of the unique combination of odorant
play both excitatory and inhibitory responses to odorants features present (i.e., it responds to the odor(s) as a whole).
(Haberly, 1969; McCollum et al., 1991; Nemitz and Odorant responses of piriform cortical single units have
Goldberg, 1983; Tanabe, et al., 1975; Wilson, 1998a). been described in several species [frog (Duchamp-Viret
Intracellular recordings reveal somewhat more simple odor- et al., 1996), rat (Haberly, 1969), monkey (Tanabe et al.,
ant-evoked postsynaptic potentials in piriform pyramidal 1975a)] and in both awake (McCollum et al., 1991;
neurons than in mitral cells, although relatively few studies Schoenbaum and Eichenbaum, 1995a) and anesthetized
have been reported to date (Nemitz and Goldberg, 1983; preparations (Haberly, 1969; Giachetti and MacLeod,
Wilson, 1998a, b). In freely breathing rats, odorant stimula- 1975; Nemitz and Goldberg, 1983; Tanabe et al., 1975a;
tion evokes a short-latency large depolarization, in phase Wilson, 2000). In general, similar to mitral/tufted cells,
with the respiratory cycle (Fig. 3) (Wilson, 1998a). This piriform cortical pyramidal cells have broad odorant-
odorant-evoked depolarization can be suprathreshold for receptive fields (Fig. 3) (Tanabe et al., 1975a; Wilson,
evoking spikes, which can reach instantaneous frequencies of 1998a, 2000), although in frog cortex there is also a sub-
over 100 Hz, but generally are within the range of 50–100 population of narrowly tuned cells (Duchamp-Viret et al.,
Hz, which corresponds to the odorant-evoked gamma-fre- 1996). In one of the few direct comparisons of receptive
quency waves recorded in piriform. The respiratory entrained fields between olfactory areas, Tanabe et al. (1975) suggest
depolarization is often bounded by periods of hyperpolariza- that piriform cortex single units are somewhat more highly
tion, which accentuate the responses to each inhalation. tuned (narrow receptive fields) than mitral cells, with
Despite the remarkable precision and topography of the cells in orbitofrontal cortex the most highly tuned—form-
olfactory nerve input to the olfactory bulb glomerular ing a hierarchy of odorant discrimination ability along the
layer, the mitral/tufted cell projection to the piriform cor- primary olfactory pathway (see below).
tex is broadly nontopographic. Projections to the anterior In a more direct test of the combinatorial array
piriform may have some spatial patterning, with individual hypothesis outlined above, a comparison of odorant cross-
axons terminating in small clusters rather than being uni- habituation between mitral/tufted cells and anterior
formly dispersed (Buonviso et al., 1991b Ojima et al.,), but piriform cortex layer II/III single units was made using
in general any one region of the olfactory bulb can project a homologous series of alkane odorants. It was hypothe-
to every region of the piriform cortex and any one region sized that if mitral/tufted cells respond to multiple odor-
of the cortex can receive input from every region of the ants because each of the effective odorants shares a
bulb (Haberly and Price, 1977; Scott et al., 1980). This common feature, then habituation to that feature should
broadly scattered input from a highly spatially ordered reduce responsiveness to all odorants by that cell. Piriform
olfactory bulb has led to models of piriform cortex as a cortex cells, however, should show less cross-habituation
combinatorial array, ideally suited to combine odorant between similar odorants if cortical cells respond to
molecular features into perceptually whole odors. Thus, collections of features, because each odorant would contain
co-activation of spatially dispersed mitral/tufted cells a unique feature ensemble. These precise results were
Sensory Physiology of Central Olfactory Pathways 189

Figure 3 Examples of odorant-receptive fields (A) and an intracellularly recorded odorant response (B) in anterior piriform cortical neu-
rons. The odorant-receptive fields of piriform cortical neurons are similar to those described for both olfactory receptor neurons and mitral
cells, with, for example, responses varying with odorant carbon chain length (A). Receptive fields in piriform cortex are highly dynamic,
with rapidly habituating odorant responses (B). (C) In contrast to mitral/tufted cells in the main olfactory bulb, however, this habituation is
highly odorant-specific. Responses to odorants differing by only 2–4 carbons in length are unaffected in piriform cortex, while mitral/tufted
cells demonstrate more generalized habituation.

obtained in urethane-anesthetized, freely breathing rats Given the incredible diversity of potential odorant fea-
(Fig. 3C) (Wilson, 2000). In addition, anterior piriform tures and odorant mixtures in the world, however, it is
single units showed minimal cross-habituation between unlikely that the synthesis of feature ensembles in the piri-
binary odorant mixtures and their components (Wilson, form cortex is based on innate synaptic connections, but
1998a), further supporting the hypothesis that the piriform rather occurs through olfactory experience-induced synap-
cortex synthesizes feature ensembles into perceptual odor tic plasticity. Experience-dependent perceptual learning
wholes. of this sort is used to explain receptive fields in visual
190 Wilson and Sullivan

inferotemporal cortex for complex objects and faces (Gilbert Cattarelli, 1995; Litaudon et al., 1997a). The anterior piri-
et al., 2001). As a test of the role of experience in cortical form may be further functionally divided into dorsal and
feature synthesis, we have recently demonstrated that ventral regions (Haberly, 1998). These functional distinc-
blockade of piriform cortex cholinergic muscarinic recep- tions presumably arise from the significant variation in
tors with scopolamine during exposure to novel odorants such anatomical features as dominance of lateral olfactory
causes piriform cortex neurons to function as feature- tract input over association fiber input (greatest in the ven-
detectors similar to mitral/tufted cells (Wilson, 2001). tral region of the anterior piriform and least in the poster-
These results are interpreted as a scopolamine blockade of ior piriform) and some differences in cell populations
synaptic plasticity that would normally allow feature (Haberly, 1998) and modulatory inputs (e.g., ACh)
ensembles to be synthesized by the cortical neurons. In (Lysakowski et al., 1987). No studies to date have exam-
fact, scopolamine can also prevent behavioral perceptual ined differences in odorant-receptive fields between anter-
learning—enhanced olfactory acuity—that occurs after ior and posterior piriform neurons, although there is some
exposure to novel odors (Fletcher and Wilson, 2002). evidence that posterior piriform responses to odorants
As described above, lateral inhibition forms a critical may be more plastic than anterior responses (Chabaud
component of odorant-response patterns in mitral/tufted et al., 2000; Litaudon et al., 1997b; Mouly et al., 1998).
cells of the olfactory bulb, shaping both the temporal Synaptic plasticity can be evoked in both afferent and
nature of the response as well as emphasizing the spatial association fiber synapses (Jung et al., 1990; Kanter and
nature of the response inherent in olfactory bulb organiza- Haberly, 1990; Roman et al., 1987; Stripling and Patneau,
tion. While both feedforward and feedback inhibitory cir- 1999; Wilson, 1998b), although some evidence suggests
cuits exist in the piriform cortex (Haberly, 1998; Kanter that association fiber synapses may be under more modu-
et al., 1996; Kapur et al., 1997; Satou et al., 1982; Scholfield, latory control than LOT afferent synapses (Hasselmo and
1978) and membrane hyperpolarization is expressed in Bower, 1992; Hasselmo et al., 1997; Stripling and
cortical neuron response to odorant (Wilson, 1998a), no Patneau, 1999; Tang and Hasselmo, 1994). Together with
investigation of the role of inhibition in cortical odorant the dominance of LOT input and potential spatial patterns
responses has yet been carried out. Lateral inhibition func- of odorant evoked activity in the anterior piriform, these
tions in most sensory systems to enhance existing spatial results suggest that the anterior regions of piriform may
response patterns, allowing one cell (or group of cells) to be more involved in odorant discrimination and the poste-
inhibit neighboring cells with similar receptive fields. This rior piriform more involved in odorant memory and odor-
can enhance contrast and/or signal-to-noise characteristics ant associations (Hasselmo and Barkai, 1995; Litaudon
of the system. If the piriform cortex truly lacks any spatial et al., 1997b).
organization, then the role of lateral inhibition may be dif- Finally, single units in the anterior piriform cortex of
ferent in this system. Several studies have attempted to the rat appear to express spatial receptive fields in addition
detect spatial patterns of evoked activity in the piriform to odorant receptive fields. Single units in the anterior
cortex with limited success (Cattarelli, and Cohen, 1989; piriform cortex can be driven by odorants unilaterally
Cattarelli et al., 1988; Sharp et al., 1977), although some presented to either the ipsilateral or contralateral naris.
of the difficulty may have been due to the rapid odorant Different cells express preferred stimulation sites, with
habituation that occurs in the piriform (Wilson, 1998a). some cells responsive only to ipsilateral stimulation,
Optical imaging of in vivo piriform responses to olfactory some only to contralateral stimulation, some equally
bulb electrical stimulation has shown some spatial speci- responsive to both, and some requiring bilateral stimula-
ficity, with different regions of the bulb activating slightly tion (Wilson, 1997). The convergence of ipsilateral and
different regions of anterior piriform, but with diffuse acti- contralateral inputs in piriform cortex may be involved in
vation of more posterior regions (Litaudon et al., 1997a). maintaining bilateral access to odorant memories
Similarly, a more recent study using well-spaced odorant (Kucharski and Hall, 1987), response amplification
stimuli and c-fos labeling has detected odorant-specific (Bennett, 1968; Klimek et al., 1998), or even stimulus
spatial patterns of activated neurons in the anterior localization (Wilson and Sullivan, 1999). Imaging work
piriform, but not in the posterior piriform (lllig and in humans suggests that the two nares may have some-
Haberly, 2000). what different odorant-tuning characteristics (Sobel
The noted functional difference between the anterior et al., 1999) and that cortical odorant processing is later-
and posterior regions of the piriform cortex has been alized (Zatorre et al., 1992). Thus, commissural pathways
demonstrated with a variety of techniques including local in both humans and rats may play a critical role in central
field potential recordings (Chabaud et al., 1999, 2000; odorant processing, the precise nature of which is yet to
Mouly et al., 1998) and optical imaging (Litaudon and be described.
Sensory Physiology of Central Olfactory Pathways 191

A. Modulation and Nonolfactory Responses single-unit (Schoenbaum and Eichenbaum, 1995a) and
local field potential recordings (Kay and Freeman, 1998).
Odorant responses of anterior piriform cortex neurons are Analysis of oscillatory local field potentials suggest that
extremely dynamic, capable of showing marked habitua- during odorant sampling 12–35 Hz -frequency oscilla-
tion within a few inhalations of an odorant in anesthetized tions travel from rostral (olfactory bulb) to caudal regions
rats (Wilson, 1998a). As described above, this habituation (entorhinal cortex) (Kay and Freeman, 1998; Chapman
is highly odorant specific, thus the cortex can filter out et al., 1998). However, in an odorant-conditioning task prior
background or currently nonsignificant stimuli, while to odorant sampling, these oscillations travel in the reverse
maintaining responsiveness to novel odorants. In awake direction (Kay and Freeman, 1998). Single-unit recordings
rats in an odorant-conditioning task, piriform cortex single in freely moving rats performing an odorant discrimination
units also show a decrease in responsiveness to repeated task similarly show changes in cortical activity during
odorants (McCollum et al., 1991). This rapid, experience- many stages of the odorant-discrimination task in addition
dependent, odorant-specific change in cortical receptive to the odorant-sampling period itself, including during
fields is similar to that reported in other sensory systems preparation for odorant sampling and during receipt of a
(Edeline, 1999) and may contribute to odorant identifica- water reward (Schoenbaum and Eichenbaum, 1995a).
tion and memory. Similar experience-dependent, odorant- Furthermore, piriform odorant responses can be affected
specific decreases in odorant responses of single units in by the current learned hedonic valence of that odorant
the orbitofrontal cortex of primates have also been (Schoenbaum and Eichenbaum, 1995a). This is in contrast
observed, as described below. to the learned changes in olfactory bulb mitral/tufted cell
In the auditory system, both experience-dependent, single-unit responses described above. Learned changes in
stimulus-specific decreases and increases can be observed olfactory bulb responses are specific to learned odorants
within receptive fields of cortical neurons, following habit- but do not encode learned hedonic valence, i.e., learned
uation (Condon and Weinberger, 1991) and associative aversive odorants and learned appetitive odorants are
learning (Weinberger, 1998), respectively. While no direct encoded similarly by the olfactory bulb (Sullivan and
studies of learning-induced changes in piriform cortex sin- Wilson, 1991).
gle-unit odorant-receptive fields have been reported, work Much of this experience- or state-dependent modulation
in two other paradigms suggest that such associative of cortical odorant responses is dependent on centrifugal
changes can occur. Rats can learn to discriminate between inputs to the piriform cortex from neuromodulatory centers
“artificial” odorants induced by focal electrical stimulation such as the horizontal limb of the diagonal band (ACh) and
of different regions of the olfactory bulb (Mouly et al., locus coeruleus (norepinephrine). Cholinergic modulation
1985). Evoked responses in the piriform cortex to these of piriform cortex function has received the most attention
artificial odorants are enhanced as the animal learns this at both the experimental physiological and neural computa-
discrimination (Litaudon et al. 1997; Roman et al., 1987). tion levels. ACh input to the olfactory system plays an
Learning to discriminate real odorants in a similar discri- important role in behavioral odorant memory. Blockade of
mination paradigm enhances 2-deoxyglucose uptake in the ACh muscarinic receptors impairs both associative and
anterior olfactory nucleus in response to the learned odor- nonassociative odorant memory (DeRosa and Hasselmo,
ant (Hamrick et al., 1993). While no learning associated 2000; Hunter and Murray, 1989; Ravel et al., 1994).
2-deoxyglucose uptake changes were detected in the piri- ACh also has a variety of specific effects on piriform
form cortex in this study, any changes may have been physiology (Barkai and Hasselmo, 1994; Hasselmo and
masked by the rapid cortical habituation described above. Bower, 1992; Linster et al., 1999; Zimmer et al., 1999). In
As in the olfactory bulb, piriform cortex odorant vitro physiology has demonstrated that muscarinic recep-
responses can be influenced by behavioral state. The tor agonists reduce piriform cortex pyramidal cell firing
hunger modulation of food odor responses observed in the adaptation (i.e., increase duration of bursts evoked by
main olfactory bulb also occurs in local field potential depolarization) (Barkai and Hasselmo, 1994; Tseng and
responses to food odor in the piriform cortex, although Haberly, 1989), selectively suppress association fiber
largely in the posterior piriform and not in the anterior piri- synaptic activation of pyramidal cells (with minimal effect
form (Chabaud et al., 2000). Similar to the olfactory bulb on LOT afferent synapses) (Hasselmo and Bower, 1992),
multiunit responses, these hunger-induced changes in cor- and enhance associative synaptic plasticity in the piriform
tical responsiveness are specific to food odor (Chabaud cortex (Hasselmo and Barkai, 1995). The muscarinic
et al., 2000). receptor–mediated suppression of association fibers has
Activity in the piriform cortex is also modulated by a been replicated in vivo by stimulation of the horizontal
variety of nonolfactory events, as determined by both limb of the diagonal band to evoke ACh release in piriform
192 Wilson and Sullivan

(Linster et al., 1999; Rosin et al., 1999). Further in vivo projects to the orbitofrontal cortex (Krettek and Price,
work has demonstrated that electrical stimulation of the 1977; Price and Slotnick, 1983). Electrophysiological
horizontal limb of the diagonal band increases excitability (Cinelli et al., 1987) and anatomical (Shipley and
of piriform cortex single units via a cholinergic muscarinic Geinesman, 1984) evidence suggests there may also be a
mechanism (Zimmer et al., 1999). direct projection from the olfactory bulb to the
These physiological effects of ACh on piriform function orbitofrontal/insular cortex in rats. Orbitofrontal cortex
have led to the hypothesis that ACh reduces interference efferents form feedback loops with primary olfactory
between similar patterns of odorant input, thus enhancing structures including the piriform cortex and dorsomedial
odorant discrimination and recognition of previously nucleus of the thalamus (Price et al., 1991). In both rats
learned odorants (Hasselmo, 1995); norepinephrine may and primates, lesions of either the orbitofrontal cortex or
have similar effects in the piriform cortex (Bouret et al., the dorsomedial nucleus of the thalamus impair odorant-
2000; Hasselmo and Bower, 1992). In support of this model, discrimination learning (Eichenbaum et al., 1980;
recent work has demonstrated that the ACh muscarinic McBride and Slotnick, 1997; Staubli et al., 1987; Tanabe
receptor antagonist scopolamine applied to the piriform et al., 1975b; Zatorre and Jones-Gotman, 1991).
cortical surface or systemically injected reduces odorant Single units in the orbitofrontal cortex respond to, and
discrimination by piriform single units as demonstrated by can discriminate between, odorants in both rodents (Onoda
enhanced cross-habituation (Wilson, 2001). Given that et al., 1984; Schoenbaum and Eichenbaum, 1995a) and
stimulation of the olfactory bulb and piriform cortex acti- primates (Rolls and Baylis, 1994; Tanabe et al., 1975a).
vates neurons in the horizontal limb of the diagonal band Odorant discrimination by single units in the orbitofrontal
(Linster and Hasselmo, 2000), odorant stimulation itself cortex (as measured by receptive field size) is improved
can regulate ACh feedback to the cortex and thus modify over that observed in the olfactory bulb and piriform cor-
subsequent coding and plasticity. tex (Onoda et al., 1984; Tanabe et al., 1975a; Yarita et al.,
In summary, the anterior piriform cortex may serve to 1980), and odorant discrimination by ensembles of
synthesize odorant feature input from the olfactory bulb orbitofrontal neurons is improved over single units (Rolls
into perceptual odor wholes. Odorant discrimination with- et al., 1996b; Schoenbaum and Eichenbaum, 1995b).
in the piriform cortex is enhanced compared to mitral/tufted The orbitofrontal cortex also receives inputs from sev-
cells of the olfactory bulb and olfactory receptors. eral sensory systems in addition to the olfactory system,
Extensive association connections within the cortex con- including the gustatory, visual, and somatosensory systems
tribute to and reinforce odorant synthesis as well as allow (Carmichael and Price, 1995; Cavada et al., 2000), as well
associative memory to tie odorants and odorant-related as spatial location information (Lipton et al., 1999). In fact,
experiences together. Experience can produce highly spe- these diverse inputs can converge on single neurons, lead-
cific changes in cortical odorant-receptive fields, with ing to single cells that respond to olfactory, gustatory, visual,
association fibers and the posterior piriform cortex playing or somatosensory stimuli alone or in combination (Rolls
a prominent role in these memory functions. Behavioral and Baylis, 1994; Rolls et al., 1999). Similar to that
state and past experience can shape both piriform odorant described for the piriform cortex, orbitofrontal neurons
responsiveness and general cortical activity through exten- respond to many phases of odorant discrimination behav-
sive centrifugal inputs to the cortex. Finally, the piriform ior, including during preodorant sampling behavior,
cortex is a major source of centrifugal input to the olfac- odorant sampling, and postodorant reward consummation
tory bulb; thus, as in thalamocortical sensory systems, the (Schoenbaum and Eichenbaum, 1995a).
cortex can directly influence its own input via descending While behavioral state and previous olfactory experi-
control of activity in more peripheral structures. ence shape odorant responses in both the olfactory bulb and
piriform cortex, as discussed above, this state-dependent,
associative nature of odorant processing appears highly
IV. ORBITOFRONTAL CORTEX refined in the orbitofrontal cortex. Thus, for example, the
response of most odorant-sensitive orbitofrontal cortex
The major neocortical area processing olfactory informa- neurons to odorants is dependent on taste reward associa-
tion is the orbitofrontal region of the prefrontal cortex, tions of the particular odorant (Critchley and Rolls, 1996a;
which in rats includes the insular cortex. Neuroanatomical Rolls et al., 1996b). That is, if the odorant is associated with
studies have demonstrated that the piriform cortex projects a pleasant sweet taste, the response to that odorant may be
directly to the orbitofrontal cortex (Johnson et al., 2000; greater than if the odorant is associated with an unpleasant
Krettek and price, 1977; Price et al., 1991), as well as to salt taste, or vice versa. These differential responses require
the dorsomedial nucleus of the thalamus, which in turn repeated experience to emerge and thus represent a learned,
Sensory Physiology of Central Olfactory Pathways 193

cross-modal association (Rolls et al., 1996a). Other learned and Yoshihara, 1995; Rolls, 2000; Wilson and Shepherd,
cross-modal associations that influence primate orbitofrontal 1995), leads to the following description of hypothetical
single-unit responses to odorants include vision (Rolls and events that may allow this remarkable feat.
Baylis, 1994) and somatosensation (Rolls et al., 1999). Odorant molecules are broken into informational fea-
Cross-modal associations with odorants have been tures by binding with specific receptors in the nose.
described in both the rat and primate orbitofrontal cortex Interactions between molecules and/or between features
(Lipton et al., 1999; Rolls et al., 1996a). may occur at the receptor level, resulting in unique recep-
As in both the main olfactory bulb and piriform cortex, tor output for some feature combinations (Cromarty and
behavioral state also influences orbitofrontal cortical Derby, 1998; Derby et al., 1991). The contrast between
responses to odorants. Orbitofrontal cortex single-unit features is then sharpened through precise projections to
responses to food odors (or associated food gustatory, the main olfactory bulb glomerular layer and glomerular
visual or somatosensory stimuli) are modulated by hunger layer inhibition. Again, some feature mixing may occur at
(Critchley and Rolls, 1996b). Feeding to satiety selectively the glomerular layer (Joerges et al., 1997; Vickers et al.,
reduces orbitofrontal responsiveness to the odor of that 1998). Thus, the spatial pattern of activity within the olfac-
food (Critchley and Rolls, 1996b). In humans, eating a sin- tory bulb glomerular layer represents the collection of
gle food to satiety selectively reduces pleasant ratings of molecular features, including an initial processing of some
that food odor, although simple exposure to a food odor for feature combinations present in the odorants sampled.
a comparable duration has a similar effect (Rolls and Mitral/tufted cells then project this alphabet of features
Rolls, 1997). Similarly to the single-unit work in monkeys, into the piriform cortex. Based on the current behavioral
odorant pleasantness influences activity in the human state (hunger?) and past experiences (memory of past food
orbitofrontal cortex as determined by PET studies (Royet odors), the representation of some features by mitral/tufted
et al., 2000), and feeding to satiety reduces orbitofrontal cells will be selectively enhanced over others through
cortex activation in response to that food odor in humans olfactory bulb centrifugal modulation.
as determined by fMRI (O’Doherty et al., 2000). The piriform cortex takes the mitral/tufted cell input
These results suggest that odorant coding in the and furthers the process of combining the features into per-
orbitofrontal cortex is similar to that for other sensory ceptually whole odors initiated in the periphery. This is
stimuli processed by prefrontal cortex, namely that performed by the combinatorial anatomy of the piriform
responses to stimuli reflect not only the sensory qualities and past experience with specific combinations of features.
of that stimulus, but also the current and past context of the That is, features that have been associated together in the
stimulus, including sensory and hedonic associations and past will, due to experience-dependent synaptic plasticity
biological significance (Goldman-Rakic, 1987; Kolb, within the piriform cortex, be more effective at driving
1984; Rolls, 2001; Schoenbaum and Eichenbaum 1995a). coincidence detecting piriform cortical neurons. Thus,
Orbitofrontal cortex neurons in turn provide descending rather than random association of odorant features within
feedback to the piriform cortex and olfactory bulb (Cinelli the piriform cortex, past experience will allow some com-
et al., 1987; Haberly, 1998), which can modulate subse- binations of features to be more easily combined and
quent peripheral processing as described above for salient. This role of the piriform cortex in odorant feature
piriform cortex. synthesis is suggested by behavioral data showing that ani-
mals with piriform cortex lesions have difficulty learning
odorant discriminations of complex odorant mixtures, but
V. GENERAL PRINCIPLES not of more simple odorants (Staubli et al., 1987).
Furthermore, piriform cortex neurons can discriminate
We can now return to the original problem stated in the between odorant mixtures and their components, suggest-
introduction that all sensory systems must function in the ing a synthesis of odorant features (Wilson, 1998a). The
real world (Fig. 2). Specifically, the olfactory system must reassembly of molecular features based on past associative
be able to recognize garlic in the spaghetti sauce in the experience within the piriform cortex allows extraction
presence of odor from a freshly cut lawn and to recognize and synthesis of perceptual odor wholes from the collec-
that odor as the perceptual entity of “garlic,” despite it tion of molecular features (i.e., the stimuli garlic and grass
being a mixture of many individual molecular compo- odor are present), in a conceptually similar way to the syn-
nents. Our current understanding of olfactory system thesis of simple visual features into complex visual objects
sensory physiology outlined above, as well as extensive in higher-order visual cortices (Logothetis and Sheinberg,
theoretical and computer modeling work (Freeman, 1981; 1996). In addition, the dynamic receptive fields and
Haberly, 1985; Hasselmo et al., 1990; Lynch, 1986; Mori enhanced odorant discrimination of the piriform cortex
194 Wilson and Sullivan

allows selective filtering of background or currently less input/output function of rat piriform cortex pyramidal cells.
relevant odorants. Again, specific activity patterns will be J. Neurophysiol 72:644–658.
enhanced depending on the behavioral state of the animal. Bell, G. A., Laing, D. G., and Panhuber, H. (1987). Odour mixture
It is hypothesized that within the piriform cortex, identifi- suppression: evidence for a peripheral mechanism in human
and rat. Brain Res. 426:8–18.
cation of the sensory stimulus (what odor is it?) is largely
Bennett, M. H. (1968). The role of the anterior limb of the anter-
complete. It should be noted, however, that piriform cortex
ior commissure in olfaction. Physiol. Behav. 3:507–515.
lesions produce little effect on well-learned odorant dis- Berkowicz, D. A., Trombley, P. Q., and Shepherd, G. M. (1994).
crimination behavior, although they may impair learning to Evidence for glutamate as the olfactory receptor cell neuro-
discriminate novel odorants (Slotnick and Schoonover, transmitter. J. Neurophysiol. 71:2557–2561.
1992; Staubli et al., 1987; Zhang et al., 1998) Bouret, S., Briois, L., Lestienne, R., and Sara, S. J. (2000). Locus
In addition to association of odorant molecular features coeruleus stimulation modulates responses to olfactory stim-
(sensory processing—this is garlic), the piriform cortex uli in piriform cortex. Soc. Neurosci. Abstr. 26:657–35.
and orbitofrontal cortex combine to allow association of Bower, J. M. (1991). Piriform cortex and olfactory object recog-
odorants with sensory context, memory, and hedonic reac- nition. In Olfaction: A Model System for Computational
Neuroscience, J. L. Davis and H. Eichenbaum (Eds.). MIT
tions (perceptual processing—I see food previously asso-
Press, Cambridge, MA, pp. 265–285.
ciated with garlic, I have eaten and enjoy garlic, I am Bowery, N. G., Hudson, A. L., and Price, G. W. (1987). GABAA
hungry for garlic). Through descending connections this and GABAB receptor site distribution in the rat central ner-
perception can influence subsequent peripheral sensory vous system. Neuroscience 20:365–383.
processing by the bulb and anterior piriform cortex. Bozza, T. C., and Kauer, J. S. (1998). Odorant response properties
Efferent connections of the piriform and orbitofrontal cor- of convergent olfactory receptor neurons. J. Neurosci. 18:
tices can then shape behavior appropriate for the given 4560–4569.
stimulus and current internal state. Something smells Brennan, P. A., and Keverne, E. B. (1997). Neural mechanisms of
mammalian olfactory learning. Prog. Neurobiol. 51:457–481.
good, let’s eat.
Bressler, S. L., and Freeman, W. J. (1980). Frequency analysis of
olfactory system EEG in cat, rabbit, and rat. EEG Clin.
Neurophysiol 50:19–24.
ACKNOWLEDGMENTS Brunjes, P. C., Smith-Crafts, L. K., and McCarty, R. (1985).
Unilateral odor deprivation: effects on the development of
The authors wish to acknowledge the support of grants olfactory bulb catecholamines and behavior. Dev. Brain Res.
from NIDCD (DAW), NSF (DAW), and NICHD (RMS). 22:1–6.
Buck, L.B. (1996). Information coding in the vertebrate olfac-
tory system. Ann. Rev. Neurosci. 19:517–544.
REFERENCES Buonviso, N., and Chaput, M. A. (1990). Response similarity to
odors in olfactory bulb output cells presumed to be connected to
the same glomerulus: electrophysiological study using simulta-
Adrian, E. D. (1950). The electrical activity of the olfactory bulb.
neous single-unit recordings. J. Neurophysiol. 63: 447–454.
EEG Clin. Neurophysiol. 2:377–388.
Buonviso, N., Chaput, M. A., and Scott, J. W. (1991a). Mitral cell-to-
Allen, D. M., and Hamilton, K. A. (2000). Ultrastructural identi-
glomerulus connectivity: an HRP study of the orientation of mitral
fication of synapses between mitral/tufted cell dendrites.
cell apical dendrites. J. Comp. Neurol. 307:57–64.
Brain Res. 860:170–173.
Buonviso, N., Revial, M. F., and Jourdan, F. (1991b). The pro-
Aroniadou-Anderjaska, V., Ennis, M., and Shipley, M. T. (1999). jections of mitral cells from small local regions of the olfac-
Dendrodendritic recurrent excitiation in mitral cells of the rat tory bulb: an anterograde tracing study using PHA-L
olfactory bulb. J. Neurophysiol. 82:489–494. (Phaseolus vulgaris leucoagglutinin). Eur. J. Neurosci. 3:
Aroniadou-Anderjaska, V., Zhou, F. M., Priest, C. A., Ennis, M., 493–500.
and Shipley, M. T. (2000). Tonic and synaptically evoked Buonviso, N., Berthommier, F., and Chaput, M. A. (1992).
presynaptic inhibition of sensory input to the rat olfactory bulb Temporal pattern analyses in pairs of neighboring mitral cells.
via GABAB heteroreceptors. J. Neurophysiol. 84: 1194–1203. J. Neurophysiol. 68:417–424.
Baker, H. (1990). Unilateral, neonatal olfactory deprivation alters Buonviso, N., Chaput, M. A., and Berthommier, F. (1996).
tyrosine hydroxylase expression but not aromatic amino acid Similarity of granular-induced inhibitory periods in pairs of
decarboxylase or GABA immunoreactivity. Neuroscience neighboring mitral/tufted cells. J. Neurophysiol. 76: 2393–2401.
36:761–771. Cain, D. P., and Bindra, D. (1972). Responses of amygdala single
Baker, H., Morel, K., Stone, D. M., and Maruniak, J. A. (1993). units to odors in the rat. Exp. Neurol. 35:98–110.
Adult naris closure profoundly reduces tyrosine hydroxylase Carmichael, S. T., and Price, J. L. (1995). Sensory and premotor
expression in mouse olfactory bulb. Brain Res. 614:109–116. connections of the orbital and medial prefrontal cortex of
Barkai, E., and Hasselmo, M. E. (1994). Modulation of the macaque monkeys. J. Comp. Neurol. 363:642–664.
Sensory Physiology of Central Olfactory Pathways 195

Cattarelli, M., and Cohen, L. B. (1989). Optical recording of the responses in the primate orbitofrontal cortex: analysis in an
in vivo piriform cortex responses to electrical stimulation of olfactory discrimination task. J. Neurophysiol. 75:1659–1672.
the lateral olfactory tract in the rat. Chem. Senses 14:577–586. Critchley, H. D., and Rolls, E. T (1996b). Hunger and satiety
Cattarelli, M., Astic, L., and Kauer, J. S. (1988). Metabolic map- modify the responses of olfactory and visual neurons in the
ping of 2DG uptake in the rat piriform cortex using compu- primate orbitofrontal cortex. J. Neurophysiol. 75:1673–1686.
terized image processing. Brain Res. 442:180–184. Cromarty, S.I., and Derby, C.D. (1998). Inhibitory receptor binding
Cavada, C., Company, T., Tejedor, J., Cruz-Rizzolo, R. J., and events among the components of complex mixtures contribute
Reinoso-Suarez, F. (2000). The anatomical connections of the to mixture suppression in responses of olfactory receptor neu-
macaque monkey orbitofrontal cortex. A review. Cerebral rons of spiny lobsters. J. Comp. Physiol. A 183:699–707.
Cortex 10:220–242. Daw, N. W., Brunken, W. J., and Parkinson, D. (1989). The func-
Chabaud, P., Ravel, N., Wilson, D. A., Mouly, A. M., Vigouroux, tion of synaptic transmitters in theretina. Ann. Rev. Neurosci.
M., Farget, V., and Gervais, R (2000). Exposure to behav- 12:205–225.
Derby, C. D., Girardot, M. N., and Daniel, P. C. (1991). Responses
iourally relevant odour reveals differential characteristics in
of olfactory receptor cells of spiny lobsters to binary mixtures.
rat central olfactory pathways as studied through oscillatory
II. Pattern mixture interactions. J. Neurophysiol. 66:131–139.
activities. Chem. Senses 25:561–573.
DeRosa, E., and Hasselmo, M. E. (2000). Muscarinic cholinergic
Chalansonnet, M., and Chaput, M. A. (1998). Olfactory bulb out-
neuromodulation reduces proactive interference between
put cell temporal response patterns to increasing odor con-
stored odor memories during associative learning in rats.
centrations in freely breathing rats. Chem. Senses 23:1–9. Behav. Neurosci. 114:32–41.
Chapman, C. A., Xu, Y., Haykin, S., and Racine, R. J. (1998). Dethier, V. G. (1987). Sniff, flick and pulse: an appreciation of
Beta-frequency (15-35Hz) electroencephalogram activities interruption. Proc. Am. Phil. Soc. 131:159–176.
elicited by toluene and electrical stimulation in the behaving Doty, R. L., and Risser, J. M. (1989). Influence of the D2
rat. Neuroscience 86:1307–1319. dopamine receptor agonist quinpirole on the odor detection
Chaput, M. A., and Holley, A. (1985). Responses of olfactory bulb performance of rats before and after spiperone administration.
neurons to repeated odor stimulations in awake freely-breath- Psychopharmacology 98:310–315.
ing rabbits. Physiol. Behav. 34:249–258. Doty, R. L., Li, C., Bagla, R., Huang, W., Pfeiffer, C., Brosvic,
Chen, W. R., Xiong, W., and Shepherd, G. M. (2000). Analysis of G. M., and Risser, J. M. (1998). SKF 38393 enhances odor
relationships between NMDA receptors and GABA release at detection performance. Psychopharmacology 136:75–82.
olfactory bulb reciprocal synapses. Neuron 25:625–633. Duchamp-Viret, P., and Duchamp, A. (1997). Odor processing in
Christensen, T. A., and Hildebrand, J. G. (1988). Frequency cod- the frog olfactory system. Prog. Neurobiol. 53:561–602.
ing by central olfactory neurons in the sphinx moth Manduca Duchamp-Viret, P., Duchamp, A., and Vigouroux, M. (1989).
sexta. Chem. Senses 13:123–130. Amplifying role of convergence in olfactory system: a com-
Christensen, T. A., and White, J. (2000). Representation of olfac- parative study of receptor cell and second-order neuron sensi-
tory information in the brain. In: The Neurobiology of Taste tivities. J. Neurophysiol. 61:1085–1094.
and Smell, 2nd ed., T.E. Finger, W. L. Silver, and D. Restrepo Duchamp-Viret, P., Palouzier-Paulignan, B., and Duchamp, A.
(Eds). Wiley-Liss, Inc., New York pp. 201–232. (1996). Odor coding properties of frog olfactory cortical neu-
Christensen, T. A., Waldrop, B. R., and Hildebrand, J. G. (1998). rons. Neuroscience 74:855–895.
Multitasking in the olfactory system: context-dependent respons- Duchamp-Viret, P., Coronas, V., Delaleu, J. -C., Moyse, E., and
es to odors reveal dual GABA-regulated coding mechanisms in Duchamp, A. (1997). Dopaminergic modulation of mitral cell
single olfactory projection neurons. J. Neurosci. 18:5999–6008. activity in the frog olfactory bulb: a combined radioligand
Christie, J. M., Schoppa, N. E., and Westbrook, G. L. (2001). Tufted binding-electrophysiological study. Neuroscience 79:203–216.
Duchamp-Viret, P., Chaput, M. A., and Duchamp, A. (1999).
cell dendrodendritic inhibition in the olfactory bulb is dependent
Odor response properties of rat olfactory receptor neurons.
on NMDA receptor activity. J. Neurophysiol. 85:169–173.
Science 284:2171–2174.
Cinelli, A. R., Ferreyra-Moyano, H., and Barragan, E. (1987).
Duchamp-Viret, P., Delaleu, J. -C., and Duchamp, A. (2000).
Reciprocal functional connections of the olfactory bulbs and
GABAB-mediated action in the frog olfactory bulb makes
other olfactory related areas with the prefrontal cortex. Brain
odor responses more salient. Neuroscience 97:771–777.
Res. Bull. 19:651–661. Edeline, J. M. (1999). Learning-induced physiological plasticity
Condon, C. D., and Weinberger, N. W. (1991). Habituation pro- in the thalamo-cortical sensory systems: a critical evaluation
duces frequency-specific plasticity of receptive fields in the of receptive field plasticity, map changes and their potential
auditory cortex. Behav. Neurosci. 105:416–430. mechanisms. Prog. Neurobiol. 57:165–224.
Coopersmith, R., and Leon, M. (1984). Enhanced neural response Eichenbaum, H., Shedlack, K. J., and Eckmann, K. W. (1980)
to familiar olfactory cues. Science 225:849–851. Thalamocortical mechanisms in odor-guided behavior. I.
Coronas, V., Srivastava, L. K., Liang, J. J., Jourdan, F., and Moyse, Effects of lesions of the mediodorsal thalamic nucleus and
E. (1997). Identification and localization of dopamine receptor frontal cortex on olfactory discrimination in the rat. Brain.
subtypes in rat olfactory mucosa and bulb: a combined in situ Behav. Evol. 17:255–275.
hybridization and ligand bind radioautographic approach. Ennis, M., Zimmer, L. A., and Shipley, M. T. (1996). Olfactory
J. Chem. Neuroanat. 12:243–257. nerve stimulation activates rat mitral cells via NMDA and
Critchley, H. D., and Rolls, E. T (1996a). Olfactory neuronal non-NMDA receptors in vitro. NeuroReport 7:989–992.
196 Wilson and Sullivan

Ennis, M., Linster, C., Aroniadou-Anderjaska, V., Ciombor, K., Haberly, L. B. (1998). Olfactory cortex. In The Synaptic
and Shipley, M. T. (1998). Glutamate and synaptic plasticity Organization of the Brain, G.M. Shepherd (Ed.). Oxford
at mammalian primary olfactory synapses. Ann. NY Acad. Sci. University Press, New York. pp. 377–416.
855:457–466. Haberly, L. B., and Price, J. L. (1977). The axonal projection pat-
Ezeh, P. I., Wellis, D. P., and Scott, J. W. (1993). organization of terns of the mitral and tufted cells of the olfactory bulb in the
inhibition in the rat olfactory bulb external plexiform layer. rat. Brain Res. 129:152–157.
J. Neurophysiol. 70:263–274. Haberly, L. B., and Price, J. L. (1978). Associational and commis-
Fletcher, M. L., and Wilson, D. A. (2002). Experience modifies sural fiber systems of the olfactory cortex of the rat. I. Systems
olfactory acuity: Ach-dependent learning decreases behavioral originating in the piriform cortex and adjacent areas. J. Comp.
generalization between similar odorants. J. Neurosci. 22:RC201. Neurol. 178:711–740.
Freeman, W. J. (1981). A physiological hypothesis of perception. Hamilton, K. A., and Kauer, J. S. (1985). Intracellular potentials of
Perspect. Biol. Med. 24:561–592. salamander mitral/tufted neurons in response to odor stimula-
Freeman, W. J., and Skarda, C. A. (1985). Spatial EEG patterns, tion. Brain Res. 338:181–185.
non-linear dynamics and perception: the neo-Sherringtonian Hamilton, K. A., and Kauer, J. S. (1989). Patterns of intracellular
view. Brain Res. Rev. 10:147–175. potentials in salamander mitral/tufted cells in response to odor
Friedman, D., and Strowbridge, B. W. (2000). Functional role of stimulation. J. Neurophysiol. 62:609–625.
NMDA autoreceptors in olfactory mitral cells. Hamrick, W. D., Wilson, D. A., and Sullivan, R. M. (1993). Neural
J. Neurophysiol. 84:39–50. correlates of memory for odor detection conditioning in adult
Gall, C. M., Hendry, S. H. C., Seroogy, K. B., Jones, E. G., and rats. Neurosci. Lett. 163:36–40.
Haycock, J. W. (1987). Evidence for coexistence of GABA Harrison, T. A., and Scott, J. W. (1986). Olfactory bulb responses
and dopamine in neurons of the rat olfactory bulb. J. Comp. to odor stimulation: analysis of response pattern and intensity
Neurol. 266:307–318. relationships. J. Neurophysiol. 56:1571–1589.
Garcia-Diaz, D. E., Aguilar-Baturoni, H. U., Guevara-Aguilar, R., Hasselmo, M. E. (1995). Neuromodulation and cortical function:
and Wayner, M. (1985). Olfactory bulb neurons respond to modeling the physiological basis of behavior. Behav. Brain
gastric distension. Brain Res. Bull. 15:661–664. Res. 67:1–27.
Gervais, R., and Pager, J. (1983). Olfactory bulb excitability Hasselmo, M. E., and Barkai, E. (1995). Cholinergic modulation
selectively modified in behaving rats after local 6-hydroxy- of activity-dependent synaptic plasticity in the piriform cortex
dopamine treatment. Behav. Brain Res. 9:165–179. and associative memory function in a network biophysical
Giachetti, I., and MacLeod, P. (1975). Cortical neuron responses simulation. J. Neurosci. 15:6592–6604.
to odours in the rat. In Olfaction and Taste V, D. A. Denton and Hasselmo, M. E., and Bower, J. M. (1992). Cholinergic suppres-
J. P. Coghlan (Eds.). Academic Press, New York, pp. 303–307. sion specific to intrinsic not afferent fiber synapse in rat piri-
Gilbert, C. D., Sigman, M., and Crist, R. E. (2001). The neural form (olfactory) cortex. J. Neurophysiol. 67:1222–1229.
basis of perpetual learning. Neuron 31:681–697. Hasselmo, M. E., Wilson, M. A., Anderson, B. P., and Bower, J. M.
Gomez, G., Voigt, R., and Atema, J. (1999). Temporal resolution (1990). Associative memory function in piriform (olfactory)
in olfaction III.: Flicker fusion and concentration-dependent cortex: computational modeling and neuropharmacology. Cold
synchronization with stimulus pulse trains of antennular Spring Harbor Symp. Quant. Biol. 55:599–610.
chemoreceptor cells in the American lobster. J. Comp. Hasselmo, M. E., Linster, C., Patil, M., Ma, D., and Cekic, M.
Physiol. A 185:427–436. (1997). Noradrenergic suppression of synaptic transmission
Goldman-Rakic, P. S. (1987). Circuitry of primate prefrontal cortex may influence cortical signal-to-noise ratio. J. Neurophysiol.
and regulation of behavior by representational memory. In 77:3326–3339.
Handbook of Physiology. The Nervous System. Higher Functions Hsia, A. Y., Vincent, J. D., and Lledo, P. M. (1999). Dopamine
of the Brain. Am. Physiol. Soc., Bethesda, MD, pp. 373–417. depresses synaptic inputs into the olfactory bulb. J. Neurophysiol.
Gurski, M. R., and Hamilton, K. A. (1996). Effects of dopamine 82:1082–1085.
and fluphenazine on field potential amplitude in the salaman- Hunter, A. J., and Murray, T. K. (1989). Cholinergic mechanisms is a
der olfactory bulb. Exp. Brain Res. 108:236–246. simple test of olfactory learning in the rat. Psychopharmacology
Guthrie, K. M., Wilson, D. A., and Leon, M. (1990). Unilateral 99:270–275.
olfactory deprivation modifies olfactory bulb function. Illig, K. R., and Haberly, L. B. (2000). Odor-specific regional acti-
J. Neurosci. 10:3402–3412. vation of piriform cortex. Chem. Senses 25:605.
Guthrie, K. M., Anderson, A. J., Leon, M., and Gall, C. (1993). Imamura, K., Mataga, N., and Mori, K. (1992). Coding of odor
Odor-induced increase in c-fos mRNA expression reveal an molecules by mitral/tufted cells in rabbit olfactory bulb. I.
anatomical ‘unit’ for odor processing in olfactory bulb. Proc. Aliphatic compounds. J. Neurophysiol. 68:1986–2002.
Natl. Acad. Sci. 90:3329–3333. Isaacson, J. S. (1999). Glutamate spillover mediates excitatory
Haberly, L. B. (1969). Single-unit responses to odors in the transmission in the rat olfactory bulb. Neuron 23:377–384.
prepyriform cortex of the rat. Brain Res. 12:481–484. Isaacson, J. S., and Strowbridge, B. W. (1998). Olfactory recipro-
Haberly, L. B. (1985). Neuronal circuitry in olfactory cortex: cal synapses: dendritic signaling in the CNS. Neuron 20:
Anatomy and functional implications. Chem. Senses 10: 219–238. 749–761.
Sensory Physiology of Central Olfactory Pathways 197

Jacobson, I., Butcher, S., and Hamberger, A. (1986). An analysis Karadi, Z., Oomura, Y., Nishino, H., and Aou, S. (1989). Octory
of the effects of excitatory amino acid receptor antagonists on coding in the monkey lateral hypothalamus: behavioral and
evoked field potentials in the olfactory bulb. Neuroscience neurochemical properties of odor-responding neurons.
19:267–273. Physiol. Behav. 45:1249–1257.
Jiang, M., Griff, E. R., Ennis, M., Zimmer, L. A., and Shipley, Kashiwadani, H., Sasaki, Y. F., Uchida, N., and Mori, K. (1999).
M. T. (1996). Activation of locus coeruleus enhances the Synchronized oscillatory discharges of mitral/tufted cells with
responses of olfactory bulb mitral cells to weak olfactory different molecular receptive ranges in the rabbit olfactory
nerve input. J. Neurosci. 16:6319–6329. bulb. J. Neurophysiol. 82:1786–1792.
Joerges, J., Kuttner, A., Galizia, C. G., and Menzel, R. (1997). Kauer, J. S. (1974). Response patterns of amphibian olfactory
Representation of odours and odour mixtures visualized in the bulb neurones to odour stimulation. J. Physiol. 243:
honeybee brain. Nature 387:285–288. 695–715.
Johnson, B. A., and Leon, M. (2000) Modular representations of Kauer, J. S., and Moulton, D. G. (1974). Responses of olfactory
odorants in the glomerular layer of the rat olfactory bulb and bulb neurones to odour stimulation of small nasal areas in the
the effects of stimulus concentration. J. Comp. Neurol. salamander. J. Physiol. 243:717–737.
422:496–509. Kay, L. M., and Freeman, W. J. (1998). Bidirectional processing in
Johnson, B. A., Woo, C. C., Duong, H., Nguyen, V., and Leon, M. the olfactory-limbic axis during olfactory behavior. Behav.
(1995). A learned odor evokes an enhanced Fos-like glomeru- Neurosci. 112:541–553.
lar response in the olfactory bulb of young rats. Brain Res. Kay, L. M., and Laurent, G. (1999). Odor- and context-dependent
699:192–200. modulation of mitral cell activity in behaving rats. Nature
Johnson, B. A., Woo, C. C., and Leon, M. (1998). Spatial coding Neurosci. 2:1003–1009.
of odorant features in the glomerular layer of the rat olfactory Klimek, L., Hummel, T., Moll, B., Kobal, G., and Mann, W. J.
bulb. J. Comp. Neurol. 393:457–471. (1998). Lateralized and bilateral olfactory function in patients
Johnson, B. A., Woo, C. W., Hingco, E. E., Pham, K. L., and Leon, with chronic sinusitus compared with healthy control sub-
M. (1999). Multidimensional chemotopic responses to jects. Laryngoscope 108:111–114.
n-aliphatic acid odorants in the rat olfactory bulb. J. Comp. Kogure, S., and Onoda, N. (1983). Response characteristics of
Neurol. 409:529–548. lateral hypothalamic neurons to odors in unanesthetized rab-
Johnson, D. M. G., Illig, K. R., Behan, M., and Haberly, L. B. bits. J. Neurophysiol. 50:609–617.
(2000). New features of connectivity in piriform cortex Kolb, B. (1984). Functions of the frontal cortex of the rat: a com-
visualized by intracellular injection of pyramidal cells sug- parative review. Brain Res. Rev. 8:65–98.
gest that “primary” olfactory cortex functions like “associa- Kosaka, T., Hatagichi, Y., Hama, K., Nagtsu, I., and Wu, J.
tion” cortex in other sensory systems. J. Neurosci. 20: (1985). Coexistence of immunoreativities for glutamate
6974–6982. decarboxylase and tyrosine hydroxylase in some neurons in
Jourdan, F., Duveau, A., Astic, L., and Holley, A. (1980). Spatial the periglomerular region of the rat main olfactory bulb: pos-
distribution of [14C]-2-deoxyglucose uptake in the olfactory sible coexistence of gamma-aminobutyric acid (GABA) and
bulbs of rats stimulated with two different odours. Brain Res. dopamine. Brain Res. 343:166–171.
188:139–154. Kosaka, T., Kosaka, K., Hama, K., Wu, J. Y., and Nagatsu, I.
Jung, M. W., Larson, J., and Lynch, G. (1990). Long-term poten- (1987). Differential effect of functional olfactory deprivation
tiation of monosynaptic EPSPs in rat piriform cortex in vitro. on the GABAergic and catecholaminergic traits in the rat
Synapse 6:279–283. main olfactory bulb. Brain. Res. 413:197–203.
Katoh, K., Koshimoto, H., Tani, A., and Mori, K. (1993). Coding Koster, N. L., Norman, A. B., Richtand, N. M., Nickell, W. T.,
of odor molecules by mitral/tufted cells in rabbit olfactory bulb. Puche, A. C., Pixley, S. K., and Shipley, M. T. (1999). Olfactory
II. Aromatic compounds. J. Neurophysiol. 70:2161–2175. receptor neurons express D2 dopamine receptors. J. Comp.
Kaluza, J. F., and Breer, H. (2000). Responsiveness of olfactory neu- Neurol. 411:666–673.
rons to distinct aliphatic aldehydes. J. Exp. Biol. 203: 927–933. Krettek, J. E., and Price, J. L. (1977). Projections from the amyg-
Kanter, E. D., and Haberly, L. B. (1990). NMDA-dependent daloid complex to the cerebral cortex and thalamus in the rat
induction of long-term potentiation in afferent and association and cat. J. Comp. Neurol. 172:687–722.
fiber systems of piriform cortex in vitro. Brain Res. 525: Kucharski, D., and Hall, W. G. (1987). New routes to early mem-
175–179. ories. Science 238:786–788.
Kanter, E. D., Kapur, A., and Haberly, L. B. (1996). A dendritic Laurent, G. (1999). A systems perspective on early olfactory cod-
GABAA-mediated IPSP regulates facilitation of NMDA- ing. Science 286:723–728.
mediated responses to burst stimulation of afferent fibers in Linster, C., and Hasselmo, M. E. (2000). Neural activity in the
piriform cortex. J. Neurosci. 16:307–312. horizontal limb of the diagonal band of Broca can be modu-
Kapur, A., Pearce, R. A., Lytton, W. W., and Haberly, L. B. (1997). lated by electrical stimulation of the olfactory bulb and cortex
GABAA-mediated IPSCs in piriform cortex have fast and slow in rats. Neurosci. Lett. 282:157–160.
components with different properties and locations on pyrami- Linster, C., Wyble, B. P., and Hasselmo, M. E. (1999). Electrical
dal cells. J. Neurophysiol. 78:2531–2545. stimulation of the horizontal limb of the diagonal band of
198 Wilson and Sullivan

broca modulates population EPSPs in piriform cortex. homologous series of fatty acid odor molecules.
J. Neurophysiol. 81:2737–2742. J. Neurophysiol. 67:786–789.
Lipton, P. A., Alvarez, P., and Eichenbaum, H. (1999). Crossmodal Mouly, A. M., Vigouroux, M., and Holley, A. (1985). On the
associative memory representations in rodent orbitofrontal cor- ability of rats to discriminate between microstimulations of
tex. Neuron 22:349–359. the olfactory bulb in different areas. Behav. Brain Res. 17:
Litaudon, P., and Cattarelli, M. (1995). Piriform cortex late activ- 45–58.
ity revealed functional spatial heterogeneity. NeuroReport Mouly, A. M., Litaudon, P., Chabaud, P., Ravel, N., and
6:1377–1380 Gervais, R. (1998). Spatiotemporal distribution of a late syn-
Litaudon, P., Datiche, F., and Cattarelli, M. (1997a). Optical chronized activity on olfactory pathways following stimula-
recording of the rat piriform cortex activity. Prog. Neurobiol. tion of the olfactory bulb in rats. Eur. J. Neurosci. 10:
52:485–510. 1128–1135.
Litaudon, P., Mouly, A. M., Sullivan, R. M., Gervais, R., and Nakashima, M., Mori, K., and Takagi, S.F. (1978). Centrifugal
Catarelli, M. (1997b). Learning-induced changes in rat piri- influence on olfactory bulb activity in the rabbit. Brain Res.
form cortex activity mapped using multisite recording with 154:301–316.
voltage sensitive dye. Eur. J. Neurosci. 9:1593–1602. Nemitz, J. W., and Goldberg, S. J. (1983). Neuronal responses of
Logothetis, N. K., and Sheinberg, D. L. (1996). Visual object rat pyriform cortex to odor stimulation: an extracellular and
recognition. Ann. Rev. Neurosci. 19:577–621. intracellular study. J. Neurophysiol. 49:188–203.
Loudon, C., and Koehl, M. A. R. (2000). Sniffing by a silkworm Nickell, W. T., and Shipley, M. T. (1988). Neurophysiology of
moth: wing fanning enhances air penetration through and magnocellular forebrain inputs to the olfactory bulb in the rat:
pheromone interception by antennae. J. Exp. Biol. 203: frequency potentiation of field potentials and inhibition of
2977–2990. output neurons. J. Neurosci., 8:4492–4502.
Lynch, G. (1986). Synapses, Circuits and the Beginnings of Nickell, W. T., Norman, A. B., Wyatt, L. M., and Shipley, M. T.
Memory. MIT Press, Cambridge, MA. (1991). Olfactory bulb DA receptors may be located on ter-
Lysakowski, A., Wayner, B. H., Bruce, G., and Hersh, L. B. minals of the olfactory nerve. NeuroReport 2:9–12.
(1989). An atlas of the regional and laminar distribution of Nickell, W. T., Behbehani, M. M., and Shipley, M. T. (1994).
choline acetyltransferase immunoreactivity in rat cerebral Evidence for GABAB-mediated inhibition of transmission
cortex. Neuroscience 28:291–336. from the olfactory nerve to mitral cells in the rat olfactory
Macrides, F., and Chorover, S. L. (1972). Olfactory bulb units: bulb. Brain Res. Bull. 35:119–123.
activity correlated with inhalation cycles and odor quality. Nowycky, M. C., Halasz, N., and Shepherd, G. M. (1983). Evoked
Science 175:84–87. field potential analysis of dopaminergic mechanisms in the
Macrides, F., Schoenfeld, T. A., Marchand, J. E., and Clancy, A. N. isolated turtle olfactory bulb. Neuroscience 8:717–722.
(1985). Evidence for morphologically, neurochemically and O’Doherty, J., Rolls, E. T., Francis, S., Bowtell, R., McGlone, F.,
functionally heterogeneous classes of mitral and tufted cells in Kobal, G., Renner, B., and Ahne, G. (2000). Sensory-specific
the olfactory bulb. Chem. Senses 10:175–202. satiety-related olfactory activation of the human orbitofrontal
Malnic, B., Hirono, J., Sato, T., and Buck, L. B. (1999). cortex. NeuroReport 11:399–403.
Combinatorial receptor codes for odors. Cell 96:713–723. Ogawa, Y. (1998). Firing properties of olfactory bulb neurons
Mathews, D. F. (1972). Response patterns of single units in the during sniffing in rats. Physiol. Behav. 64:755–764.
olfactory bulb of the rat to odor. Brain Res. 47:389–400. Ojima, H., Mori, K., and Kishi, K. (1984). The trajectory of
McBride, S. A., and Slotnick, B. (1997). The olfactory thalamo- mitral cell axons in the rabbit olfactory cortex revealed by
cortical system and odor reversal learning examined using an intracellular HRP injection. J. Comp. Neurol. 230:77–87.
asymmetrical lesion paradigm in rats. Behav. Neurosci. Onoda, N., and Mori, K. (1980). Depth distribution of temporal
111:1273–1284. firing patterns in olfactory bulb related to air-intake cycles.
McCollum, J., Larson, J., Otto, T., Schottler, F., Granger, R., and J. Neurophysiol. 44:29–39.
Lynch, G. (1991). Short-latency single-unit processing in Onoda, N., Imamura, K., Obata, E., and lino, M. (1984).
olfactory cortex. J. Cog. Neurosci. 3:293–299. Response selectivity of neocortical neurons to specific odors
Mair, R. G. (1982). Response properties of rat olfactory bulb neu- in the rabbit. J. Neurophysiol. 52:638–650.
rones. J. Physiol. 326:341–359. Orona, E., Scott, J. W., and Rainer, E. C. (1983). Different granule
Meredith, M. (1986). Patterned response to odor in mammalian olfac- cell populations innervate superficial and deep regions of the
tory bulb: the influence of intensity. J. Neurophysiol. 56: 572–597. external plexiform layer in rat olfactory bulb. J. Comp. Neurol.
Mombaerts, P. (1999). Molecular biology of odorant receptors in 217:227–237.
vertebrates. Ann. Rev. Neurosci. 22:487–509. Orona, E., Rainer, E. C., and Scott, J. W. (1984). Dendritic and
Mori, K., and Yoshihara, Y. (1995). Molecular recognition and axonal organization of mitral and tufted cells in the rat olfac-
olfactory processing in the mammalian olfactory system. tory bulb. J. Comp. Neurol. 226:346–356.
Prog. Neurobiol. 45:585–619. Pager, J. (1974). Selective modulation of the olfactory bulb elec-
Mori, K., Mataga, N., and Imamura, K. (1992). Differential trical activity in relation to the learning of palatability in hun-
specificities of single mitral cells in rabbit olfactory bulb for a gry and satiated rats. Physiol. Behav. 12:189–195.
Sensory Physiology of Central Olfactory Pathways 199

Pager, J. (1978). Ascending olfactory information and centrifugal Roman, F., Staubli, U., and Lynch, G. (1987). Evidence for
influxes contributing to a nutritional modulation of the rat synaptic potentiation in a cortical network during learning.
mitral cell responses. Brain Res. 140:251–269. Brain Res. 418:221–226.
Pager, J. (1983). Unit responses changing with behavioral out- Rosin, J. F., Datiche, F., and Cattarelli, M. (1999). Modulation of
come in the olfactory bulb of unrestrained rats. Brain Res. the piriform cortex activity by the basal forebrain: an optical
289:87–98. recording study in the rat. Brain Res. 820:105–111.
Pager, J. (1985). Respiration and olfactory bulb unit activity in Royet, J. P., Jourdan, F., Ploye, H., and Soucheier, C. (1989).
the unrestrained rat: statements and reappraisals. Behav. Morphometric modifications associated with early sensory
Brain Res. 16:81–94. experience in the rat olfactory bulb. II. Stereological study of
Pager, J., Giachetti, I., Holley, A., and LeMagnen, J. (1972). A the population of olfactory glomeruli. J. Comp. Neurol.
selective control of olfactory bulb electrical activity in relation 289:594–609.
to food deprivation and satiety in rats. Physiol. Behav. Royet, J. P., Zald, D., Versace, R., Costes, N., Lavenne, F., Koenig,
9:573–579. O., and Gervais, R. (2000). Emotional responses to pleasant and
Pfaff, D. W., and Gregory, E. (1971). Olfactory coding in olfac- unpleasant olfactory, visual, and auditory stimuli: a positron
tory bulb and medial forebrain bundle of normal and castrated emission tomography study. J. Neurosci. 20:7752–7759.
male rats. J. Neurophysiol. 34:208–216. Rubin, B. D., and Katz, L. C. (1999). Optical imaging of odorant
Potter, H., and Chorover, S. L. (1976). Response plasticity in representations in the mammalian olfactory bulb. Neuron
hamster olfactory bulb: peripheral and central processes. 23:499–511.
Brain Res. 116:417–429. Sallaz, M., and Jourdan, F. (1992). Apomorphine disrupts odour-
Puche, A. C., and Shipley, M. T. (1999). Odor-induced, activity- induced patterns of glomerular activation in the olfactory
dependent transneuronal gene induction in vitro: mediation by bulb. NeuroReport 3:833–836.
NMDA receptors. J. Neurosci. 19:1359–1370. Sallaz, M., and Jourdan, F. (1993). C-fos expression and
Price, J. L., and Slotnick, B. M. (1983). Dual olfactory representa- 2-deoxyglucose uptake in the olfactory bulb of odour-stimu-
tion in the rat thalamus: an anatomical and electrophysiological lated awake rats. NeuroReport 4:55–58.
study. J. Comp. Neurol. 215:63–77. Sato, T., Hirono, J., Tonoike, M., and Takebayashi, M. (1994).
Price, J. L., Carmichael, S. T., Carnes, K. M., Clugnet, M. C., Tuning specificities to aliphatic odorants in mouse olfactory
Kuroda, M., and Ray, J. P. (1991). Olfactory input to the pre- receptor neurons and their local distribution. J. Neurophysiol.
frontal cortex. In Olfaction: A Model System for Computational 72:2980–2989.
Neuroscience. J.L. Davis and H. Eichenbaum (Eds.). MIT Satou, M., Mori, K., Tazawa, Y., and Takagi, S.F. (1982). Two
Press, Cambridge, MA, pp. 101–120. types of postsynaptic inhibition in pyriform cortex of the rab-
Rall, W., Shepherd, G. M., Reese, T. S., and Brightman, M. W. bit: fast and slow inhibitory postsynaptic potentials.
(1966). Dendrodendritic synaptic pathway for inhibition in J. Neurophysiol. 48:1142–1156.
the olfactory bulb. Exp. Neurol. 14:44–56. Scholfield, C. N. (1978). A barbiturate induced intensification of
Ravel, N., Elaagouby, A., and Gervais, R. (1994). Scopolamine the inhibitory potential in slices of guinea-pig olfactory cor-
injection into the olfactory bulb impairs short-term olfactory tex. J. Physiol. 275:559–566.
memory in rats. Behav. Neurosci. 108:317–324. Schneider, R. W. S., Price, B. A., and Moore, P. A. (1998). Antennal
Rolls, E. T. (2000). The orbitofrontal cortex and reward. Cerebral morphology as a physical filter of olfaction: temporal tuning of
Cortex 10:284–294. the antennae of the honey bee, Apis mellifera. J. Insect Physiol.
Rolls, E. T. (2001). The rules of formation of the olfactory repre- 44:677–684.
sentations found in the orbitofrontal cortex olfactory areas in Schoenbaum, G., and Eichenbaum, H. (1995a). Information cod-
primates. Chem. Senses 26:595–604. ing in the rodent prefrontal cortex. I. Single-neuron activity in
Rolls, E. T., and Baylis, L. L. (1994). Gustatory, olfactory and orbitofrontal cortex compared with that in pyriform cortex.
visual convergence within the primate orbitofrontal cortex. J. Neurophysiol. 74:733–750.
J. Neurosci. 14:5437–5452. Schoenbaum, G., and Eichenbaum, H. (1995b). Information cod-
Rolls, E. T., and Rolls, J. H. (1997). Olfactory sensory-specific ing in the rodent prefrontal cortex. II. Ensemble activity in
satiety in humans. Physiol. Behav. 61:461–473. orbitofrontal cortex. J. Neurophysiol. 74:751–762.
Rolls, E. T., Critchley, H. D., Mason, R., and Wakeman, E. A. Schoenbaum, G., Chiba, A. A., and Gallagher, M. (1999). Neural
(1996a). Orbitofrontal cortex neurons: role in olfactory and encoding in orbitofrontal cortex and basolateral amygdala during
visual association learning. J. Neurophysiol. 75: 1970–1981. olfactory discrimination learning. J. Neurosci. 19:1876–1884.
Rolls, E. T., Critchley, H. D., and Treves, A. (1996b). Schoppa, N. E., Kinzie, J. M., Sahara, Y., Segerson, T. P., and
Representation of olfactory information in the primate Westbrook, G.L. (1998). Dendrodendritic inhibition in the
orbitofrontal cortex. J. Neurophysiol. 75:1982–1996. olfactory bulb is driven by NMDA receptors. J. Neurosci.
Rolls, E. T., Critchley, H. D., Browning, A. S., Hernadi, A., and 18:6790–6802.
Lenard, L. (1999). Responses to the sensory properties of fat Scott, J. W. (1977). A measure of extracellular unit responses to
of neurons in the primate orbitofrontal cortex. J. Neurosci., repeated stimulation applied to observations of the time
19:1532–1540. course of olfactory responses. Brain Res. 132:247–258.
200 Wilson and Sullivan

Scott, J. W. (1981). Electrophysiological identification of mitral orbitofrontal cortex of the monkey. J. Neurophysiol. 38:
and tufted cells and distributions of their axons in olfactory 1284–1296.
system of the rat. J. Neurophysiol. 46:918–931. Tanabe, T., Yarita, H., lino, M., Ooshima, Y., and Takagi, S. F.
Scott, J. W., and Pfaffmann, C. (1972). Characteristics of responses (1975b). An olfactory projection area in orbitofrontal cortex
of lateral hypothalamic neurons to stimulation of the olfactory of the monkey. J. Neurophysiol. 38:1269–1283.
system. Brain Res. 48:251–264. Tang, A. C., and Hasselmo, M. E. (1994). Selective suppression of
Scott, J. W., McBride, R. L., and Schneider, S.P. (1980). The orga- intrinsic but not afferent fiber synaptic transmission by baclofen
nization of projections from the olfactory bulb to the piriform in the piriform (olfactory) cortex. Brain Res. 659:75–81.
cortex and olfactory tubercle in the rat. J. Comp. Neurol. Trombley, P. Q., and Westbrook, G. L. (1990). Excitatory synaptic
194:519–534. transmission in cultures of rat olfactory bulb. J. Neurophysiol.
Sharp, F. R., Kauer, J. S., and Shepherd, G. M. (1977). Laminar 64:598–606.
analysis of 2-deoxyglucose uptake in olfactory bulb and olfac- Tseng, G. F., and Haberly, L. B. (1989). Deep neurons in piriform
tory cortex of rabbit and rat. J. Neurophysiol. 40:800–813. cortex II. Membrane properties that underlie unusual synaptic
Shepherd, G. M., and Greer, C. A. (1998). Olfactory bulb. In The responses. J. Neurophysiol. 62:386–400.
Synaptic Organization of the Brain, G. M. Shepherd (Ed.). Tsuboi, A., Yoshihara, S., Yamazaki, N., Kasai, H., Asai-Tsuboi,
Oxford University Press, New York, pp. 159–203. H., Komatsu, M., Serizawa, S., Ishii, T., Matsuda, Y., Nagawa,
Shipley, M. T., and Ennis, M. (1996). Functional organization of F., and Sakano, H. (1999). Olfactory neurons expressing
olfactory system. J. Neurobiol. 30:123–176. closely linked and homologous odorant receptor genes tend to
Shipley, M. T., and Geinisman, Y. (1984). Anatomical evidence project their axons to neighboring glomeruli on the olfactory
for convergence of olfactory, gustatory, and visceral afferent bulb. J. Neurosci. 19:8409–8418.
pathways in mouse cerebral cortex. Brain Res. Bull. Uchida, N., Takahashi, Y. K., Tanifuji, M., and Mori, K. (2000).
12:221–226. Odor maps in the mammalian olfactory bulb: domain organi-
Sicard, G., and Holley, A. (1984). Receptor cell responses to zation and odorant structural features. Nature Neurosci.
odorants: similarities and differences among odorants. Brain 3:1035–1043.
Res. 232:283–296. Viana DiPrisco, G., and Freeman, W. J. (1985). Odor-related
Singer, M. S., and Shepherd, G. M. (1994). Molecular modeling of bulbar EEG spatial pattern analysis during appetitive condi-
ligand-receptor interactions in te OR5 olfactory receptor. tioning in rabbits. Behav. Neurosci. 99:964–978.
NeuroReport 5:1297–1300. Vickers, N. J., Christensen, T. A., and Hildebrand, J. G. (1998).
Slotnick, B. M., and Schoonover, F. W. (1992). Olfactory pathways Combinatorial odor discrimination in the brain: Attractive and
and the sense of smell. Neurosci. Biobehav. Rev. 16:453–472. antagonist odor blends are represented in distinct combina-
Sobel, N. Khan, R. M., Saltman, A., Sullivan, E. V., and Gabrieli, tions of uniquely identifiable glomeruli. J. Comp. Neurol.
J. D. E. (1999). The world smells different to each nostril. 400:35–56.
Nature 402:35. Wehr, M., and Laurent, G. (1996). Odour encoding by temporal
Staubli, U., Schottler, F., and Nejat-Bina, D. (1987). Role of dor- sequences of firing in oscillating neural assemblies. Nature.
somedial thalamic nucleus and piriform cortex in processing 384:162–166.
olfactory information. Behav. Brain Res. 25:117–129. Weinberger, N. M. (1998) Physiological memory in primary audi-
Stewart, W. B., Kauer, J. S., and Shepherd, G. M. (1979). tory cortex: characteristics and mechanisms. Neurobiol. Learn.
Functional organization of rat olfactory bulb analyzed by the Mem. 70:226–251.
2-deoxyglucose method. J. Comp. Neurol. 185:715–734. Wellis, D. P., and Scott, J. W. (1990). Intracellular response of
Stopfer, M., Bhagavan, S., Smith, B. H., and Laurent, G. (1997). identified rat olfactory bulb interneurons to electrical and odor
Impaired odour discrimination on desynchronization of stimulation. J. Neurophysiol. 64:932–947.
odour-encoding neural assemblies. Nature 390:70–74. Wellis, D. P., Scott, J. W., and Harrison, T. A. (1989). Discrimination
Stripling, J. S., and Patneau, D. K. (1999). Potentiation of late among odorants by single neurons of the rat olfactory bulb.
components in olfactory bulb and piriform cortex requires J. Neurophysiol. 61:1161–1177.
activation of cortical association fibers. Brain Res. 841:27–42. Wilson, D. A. (1997). Binaral interactions in the rat piriform cor-
Sullivan, R. M., and Leon, M. (1986). Early olfactory learning tex. J. Neurophysiol. 78:160–169.
induces an enhanced olfactory bulb response in young rats. Wilson, D. A. (1998a). Habituation of odor responses in the rat
Dev. Brain Res. 27:278–282. anterior piriform cortex. J. Neurophysiol. 79:1425–1440.
Sullivan, R. M., and Wilson, D. A. (1991). Neural correlates of Wilson, D. A. (1998b). Synaptic correlates of odor habituation in
conditioned odor avoidance in infant rats. Behav. Neurosci. the rat anterior piriform cortex. J. Neurophysiol. 80:998–1001.
105:307–312. Wilson, D. A. (2000). Comparison of odor receptive field plastic-
Sullivan, R. M., Wilson, D. A., and Leon, M. (1989). Norepinephrine ity in the rat olfactory bulb and anterior piriform cortex.
and learning-induced plasticity in infant rat olfactory system. J. Neurophysiol. 84:3036–3042.
J. Neurosci. 9:3998–4006. Wilson, D. A. (2001). Scopolamine enhances generalization
Tanabe, T., lino, M., and Takagi, S. F. (1975a). Discrimination of between odor representations in rat olfactory cortex. Learn
odors in olfactory bulb, pyriform-amygdaloid areas, and Memory 8:279–285.
Sensory Physiology of Central Olfactory Pathways 201

Wilson, D. A., and Leon, M. (1987). Evidence of lateral synaptic Wilson, M., and Shepherd, G. M. (1995). Olfactory cortex. In The
interactions in olfactory bulb output cell responses to odors. Handbook of Brain Theory and Neural Networks M. A. Arbib
Brain Res. 417:175–180. (Ed.), MIT Press, Cambridge, MA, pp. 669–673.
Wilson, D. A., and Leon, M. (1988). Spatial patterns of olfactory Woo, C. C., Oshita, M. H., and Leon, M. (1996). A learned odor
bulb single-unit responses to learned olfactory cues in young decreases the number of Fos-immunopositive granule cells in
rats. J. Neurophysiol. 59:1770–1782. the olfactory bulb of young rats. Brain Res. 716:149–156.
Wilson, D. A., and Sullivan, R. M. (1990). Olfactory associative Yokoi, M., Mori, K., and Nakanishi, S. (1995). Refinement of
conditioning in infant rats with brain stimulation as reward. I. odor molecular tuning by dendrodendritic synaptic inhibition
Neurobehavioral consequences. Dev. Brain Res. 53:215–221. in the olfactory bulb. Proc. Natl. Acad. Sci. 92:3371–3375.
Wilson, D. A., and Sullivan, R. M. (1994). Neurobiology of asso- Young, T. A., and Wilson, D. A. (1999). Frequency dependent
ciative learning in the neonate: Early olfactory learning. modulation of inhibition in the rat olfactory bulb. Neurosci.
Behav. Neural Biol. 61:1–18. Lett., 276:65–67.
Wilson, D. A., and Sullivan, R. M. (1995). The D2 antagonist Youngentob, S. L., Mozell, M. M., Sheehe, P. R., and Hornung, D. E.
spiperone mimics the effects of olfactory deprivation on (1987). A quantitative analysis of sniffing strategies in rats per-
mitral/tufted cell odor response patterns. J. Neurosci. forming odor detection tasks. Physiol. Behav. 41: 59–69.
15:5574–5581. Zatorre, R. J., and Jones-Gotman, M. (1991). Human olfactory
Wilson, D. A., and Sullivan, R. M. (1999). Respiratory airflow discrimination after unilateral frontal or temporal lobectomy.
pattern at the rat’s snout and an hypothesis regarding its role Brain. 114:71–84.
in olfaction. Physiol. Behav. 66:41–44. Zatorre, R. J., Jones-Gotman, M., Evans, A. C., and Meyer, E.
Wilson, D. A., and Wood, J. J. (1992). Functional consequences (1992). Functional localization and lateralization of human
of unilateral olfactory deprivation: time-course and age sensi- olfactory cortex. Nature 360:339–340.
tivity. Neuroscience. 49:183–192. Zhang, Y., Burk, J. A., Glode, B. M., and Mair, R. G. (1998). Effects
Wilson, D. A., Sullivan, R. M., Gall, C. M., and Guthrie, K. M. of thalamic and olfactory cortical lesions on continuous olfac-
(1996). NMDA-receptor modulation of lateral inhibition and tory delayed nonmatching-to-sample and olfactory discrimination
c-fos expression in olfactory bulb. Brain Res. 719:62–71. in rats (Rattus norvegicus). Behav. Neurosci. 112:39–53.
Wilson, M. A., and Bower, J. M. (1992). Cortical oscillations and Zimmer, L. A., Ennis, M., and Shipley, M. T. (1999). Diagonal
temporal interactions in a computer simulation of piriform band stimulation increases piriform cortex neuronal excitabil-
cortex. J. Neurophysiol. 67:981–995. ity in vivo. NeuroReport 10:2101–2105.
10

Psychophysical Measurement of Human Olfactory Function,

Including Odorant Mixture Assessment

Richard L. Doty
University of Pennsylvania, Philadelphia, Pennsylvania, U.S.A.
David G. Laing
University of Western Sydney, Sydney, Australia

I. INTRODUCTION mathematical concepts developed in the mid-nineteenth

century by Weber (1834) and Fechner (1860) and by
As can be gleaned from other chapters in this volume, the Thurstone, Stevens, and others in the twentieth century
perception of an odorant depends upon the activation of a (e.g., Anderson, 1970; Stevens, 1961; Thurstone, 1927a,b).
subset of ~1000 olfactory receptor types distributed, in the Tests derived from these traditions include absolute detec-
human, across ~6,000,000 receptor cells. Each receptor tion thresholds (the lowest odorant concentration that can
cell commonly carries only one type of receptor, and the be perceived), differential thresholds (the smallest differ-
relative distribution of the receptor types among the ence in concentration of a given chemical that can be per-
~6,000,000 receptor cells is unknown. Since most odorous ceived), and various indices of suprathreshold sensation
substances found in nature are comprised of more than one magnitude. Most were developed within the theoretical
chemical, a typical stimulus simultaneously activates over- framework of establishing mathematical rules or laws that
lapping subsets or arrays of many olfactory receptor cells. govern the build-up of suprathreshold sensation relative to
From these arrays the nervous system extracts a unitary stimulus intensity. To achieve these ends, well-defined
sensation for a given stimulus, although, for some sub- stimuli (e.g., single chemicals of known chemical purity)
stances, a few major “notes” can be discerned, as is well were usually employed, allowing for straightforward stim-
known to wine and beer connoisseurs. Hence, from one ulus specification.
perspective olfaction is largely a synthetic sensory system, Other trends, however, resulted in the development or
synthesizing a distinct individual sensory sensation from a application of tests more useful in applied settings. For
complex set of chemicals, many of which have an individ- example, eighteenth- and nineteenth- century physicians
ual odor. From another perspective, however, it is an simply presented familiar odorants to patients to see if they
analytical sensory system, capable of extracting from hun- could be identified, usually without insight into prior
dreds of potential sensations a few dominant qualities. psychophysical developments. In the twentieth century rel-
During the last two centuries, numerous tests have been atively sophisticated procedures were developed within the
devised to assess the function of this system. Historically, food industry (e.g., the forced-choice triangle test), where
many of these tests have been modeled on procedural and the need exists for quantifying the discriminability or

203
204 Doty and Laing

acceptability of various product formulations in relation to (4) plastic squeeze bottles (Amoore and Ollman, 1983; Cain
perceived qualitative attributes. Unlike the academic psy- et al., 1988; Doty, 2000; Guadagni et al., 1963), (5) air-dilu-
chophysical traditions, and akin to the clinical traditions, tion olfactometers (Cheesman and Kirkby, 1959; Doty et al.,
the stimuli were multicomponent or chemically complex. 1988b; Kobal and Plattig, 1978; Lorig et al., 1999; Punter,
Although quantitative, the metrics employed in these para- 1983; Walker et al., 1990; Wenzel, 1948), (6) microencapsu-
digms were more operational and rarely linked to simple lated “scratch and sniff” odorized strips (Doty, 1995; Doty
physicochemical properties such as odorant concentration. et al., 1984a; Richman et al., 1992), and (7) bottles from
The present chapter has two major goals. The first is to which blasts of saturated air are presented (Elsberg and
provide the reader with an up-to-date overview of the Levy, 1935) (Fig. 1). In environmental control studies,
quantitative methods available for assessing the sense of mobile units containing olfactometers, odor exposure
smell, regardless of the historical traditions that led to their chambers, analytical equipment, and subject waiting rooms
development. Emphasis is placed on the relative utility of have been employed (e.g., Berglund et al., 1984; Springer,
various approaches for achieving this end. The second goal 1974) (Fig. 2).
is to examine elements of odor mixture perception, includ- In addition to these approaches to the presentation of
ing how well individual components can be discerned. An stimuli, intravenous administration of odorants has
understanding of odor mixture processing is of value in been employed to produce chemosensory sensations.
elucidating how the olfactory system works, as its neural This has been used primarily by Japanese otolaryngol-
architecture seems to be designed to filter or collapse com- ogists in an attempt to determine whether the olfactory
plex arrays of chemical information into distinct, inter- receptors are working when nasal congestion or block-
pretable, and manageable percepts. Although many of the age eliminates or mitigates airflow to the receptor
examples described in this chapter come from clinical region. The assumption underlying this technique is
studies, the tenants of the chapter are broadly applicable to that the stimulus makes its way to the olfactory recep-
settings outside the clinic, including industrial and tors via the bloodstream. Most commonly thiamine
regulatory ones. propyldisulfide (Alinamin) is injected into the median
cubital vein, and recordings of the duration and latency
of the onset of a garlic-like sensation experienced by
II. STIMULUS CONTROL AND the patient are made (see Takagi, 1989, for review).
PRESENTATION Although this procedure may be of value in some cases,
there is some controversy regarding its physiological
In some chemosensory paradigms, extremely accurate basis (i.e., whether the stimulus reaches the receptors
stimulus specification is required, and elaborate via diffusion from nasal capillaries, from lung air, or
olfactometers and other devices for presenting known con- both) (see Maruniak et al., 1983). Furthermore, such
centrations of odorants in specific quantities for various testing is invasive, highly variable, not readily adapt-
durations have been devised (for review, see Prah et al., able to a forced-choice paradigm, and lacks normative
1995). This is particularly true for devices employed in referents.
event-related potential research (see Chapter 11).
In other paradigms, including those related to assessing
olfactory function in patients, it is not necessary to know III. PSYCHOPHYSICAL TEST
the exact number of molecules that enter the nose to make PROCEDURES
the test valid. The key issue in the latter case is that the
odorants are presented in a reliable manner and that norms Today, any procedure that provides a quantitative measure
are available to establish whether a patient’s responses are of sensory function and requires a verbal or conscious overt
normal or abnormal. Thus, accurate clinical assessment of response on the part of the examinee is generally consid-
chemosensory function can be made using surprisingly ered to be a psychophysical procedure. In this section, the
simple stimulus presentation equipment. basic psychophysical paradigms available for measuring
Devices used to present odorants to humans include olfactory function are discussed and examples of their
(1) the draw tube olfactometer of Zwaardemaker (1925, application are provided. The interested reader is referred
1927), (2) glass sniff bottles (Cheesman and Townsend, to other sources for more detailed information about psy-
1956; Doty et al., 1986; Nordin et al., 1998), (3) odorized chophysical methods, including their mathematical founda-
glass rods, wooden sticks, felt-tipped pens, alcohol pads, tions (Ekman and Sjöberg, 1965; Gescheider, 1988;
or strips of blotter paper (Davidson & Murphy, 1997; Guilford, 1954; Köster, 1975; Marks, 1974; Stevens, 1961;
Hummel et al., 1997; Semb, 1968; Toyota et al., 1978), Tanner and Swets, 1954).
Psychophysical Measurement of Human Olfactory Function 205

Figure 1 Procedures for presenting odorants to subjects for assessment. (A) Early draw-tube olfactometer of Zwaardemaker. In this
apparatus, an outer tube, made of rubber or another odorous material, slides along a calibrated inner tube, one end of which is inserted
into the subject’s nostril. When the odorized tube is slid toward the subject, less of its internal surface is exposed to the inspired airstream,
resulting in a weaker olfactory sensation. (B) Sniff bottle. (C) Perfumer’s strip. (D) Squeeze bottle. (E) Blast injection device. The experi-
menter injects a given volume of odor into the bottle and releases the pressure by squeezing a clamp on the tube leading to the nostril,
producing a stimulus pulse. (F) Microencapsulated “scratch-and-sniff ” test. (G) Sniff ports on a rotating table connected to one of the
University of Pennsylvania’s dynamic air-dilution olfactometers.
206 Doty and Laing

contamination by response biases (i.e., the conservatism

or liberalism in reporting the presence of an odor under
uncertain conditions). In addition, they are typically more
reliable and produce lower threshold values (Blackwell,
1953; Doty et al., 1995).
Two types of threshold procedures that have received
the most clinical use are the ascending method of limits
(AML) and the single staircase (SS) procedures. In the
AML procedure, odorants are presented sequentially from
low to high concentrations and the point of transition
between detection and no detection is estimated. Forced-
choice responses are required on each trial. In the SS
method (a variant of the method of limits technique) (see
Cornsweet, 1962), the concentration of the stimulus is
increased following trials in which a subject fails to detect
the stimulus and decreased following trials where correct
detection occurs. In both these procedures, the direction of
initial stimulus presentation is made from weak to strong
in an effort to reduce adaptation effects of prior stimulation
(see Pangborn et al., 1964).
An example of a clinical application of the AML pro-
cedure is provided by Cain (1982a) who used 60-mL glass
sniff bottles to present either water (diluent) or odorant
(n-butanol dissolved in water) to 43 patients with various
degrees of olfactory dysfunction. Four repeated ascending
series were presented to each side of the nose in a two-
Figure 2 (Top) Odor evaluation room of mobile odor evalua-
tion laboratory designed to evaluate responses of panel members
alternative, forced-choice format. This test, which took
to diesel exhaust. (Bottom) Schematic of mobile odor evaluation approximately half an hour per patient to administer,
laboratory. (From Springer, 1974.) demonstrated that the olfactory dysfunction in these cases
was typically bilateral.
An example of the clinical use of a SS procedure comes
from a study that demonstrates loss of olfactory function in
A. Detection and Recognition Threshold Tests early Alzheimer’s disease (Doty et al., 1987). In this experi-
ment, a trial consisted of the presentation of two 100-mL
A popular means for assessing chemosensory function is glass sniff bottles to the patient in rapid succession. One
to establish, operationally, a measure of the lowest con- bottle contained 20 mL of a given concentration of phenyl
centration of a stimulus that can be detected. A qualitative ethyl alcohol dissolved in USP-grade light mineral oil,
odor sensation (e.g., “banana-like”) is rarely perceived at whereas the other contained mineral oil alone. The patient
very low odorant concentrations, where only the faint was asked to report which of the two bottles in a pair pro-
presence of something is noted. The absolute or detection duced the strongest sensation. The first trial was presented
threshold is the lowest odorant concentration where such at a -6.50 log (liquid volume/volume) concentration step.
a presence is reliably detected, whereas the recognition If a miss occurred on any trial before five were correctly
threshold is the lowest concentration where odor quality is completed at that concentration, the process was repeated
reliably discerned. In modern olfactory detection thresh- at 1 log concentration step higher. When five consecutive
old testing, a subject is asked to indicate, on a given trial, correct trials occurred at a given concentration level, the
which of two or more stimuli (e.g., a low concentration staircase was “reversed” and the next pair of trials was pre-
odorant and one or more nonodorous blanks) smells sented at a 0.5 log concentration step lower. From this
strongest, rather than to report whether an odor is per- point on, only one or two trials were presented at each step
ceived or not. Recognition thresholds are obtained in a (i.e., if the first trial was missed, the second was not given
similar manner, but the requirement is to report which one and the staircase was moved to the next higher 0.5 log step
has the target quality. Such “forced-choice” procedures concentration). When correct performance occurred on
are less susceptible than non-forced-choice procedures to both trials, the concentration of the next trial was given at
Psychophysical Measurement of Human Olfactory Function 207

0.5 log unit step lower. The average of the last four of subjects (Yoshida, 1984). More recently, Stevens et al.
seven staircase reversal points served as the threshold esti- (1988) obtained 60 threshold values over the course of 30
mate. Examples of individual data obtained using the SS days from three subjects (20 for butanol, 20 for pyridine,
procedure are shown in Figure 3. and 20 for -phenylethylmethylethylcarbinol). These inves-
In general, threshold values are relative and dependent tigators found that the within-subject variability across
upon such factors as the method of stimulus dilution, vol- test days was as great as the between-subject variability
ume of inhalation, species of molecule, type of psychophys- on a given test day, suggesting to these authors that the large
ical task, and number of trials presented (Pierce et al., 1996). individual differences observed in threshold values are not a
A number of investigators have been struck with the fact that reflection of big differences among stable threshold values of
threshold measures often exhibit considerable intra- and subjects but reflect large day-to-day fluctuations in the test
intersubject variability. For example, in one study of 60 sub- measures. Unfortunately, much of this fluctuation likely
jects, intersubject variation as great as 5 log units was reflects the use of the single ascending detection threshold
reported (Brown et al., 1968). In another, in which a non– technique, in which the apparent limen is traversed only
forced-choice ascending threshold procedure was used (the once. Clearly, test procedures with more trials, such as the SS
Japanese “T&T Olfactometer”), variation on the order of 16 procedure, produce less variable measures and, when employed,
log units was present among groups of 430 –1000 young do not exhibit as marked day-to-day fluctuations.

Figure 3 Data illustrating single-staircase detection threshold determinations. Each plus (+) indicates a correct detection when an odor-
ant versus a blank is presented. Each zero (0) indicates an incorrect report of an odorant. Threshold value (T; vol/vol in light USP grade
mineral oil) is calculated as the mean of the last four of seven staircase reversals. Although the geometric mean is the correct measure,
the arithmetic mean usually provides a close approximation. The o’s and d’s on the abcissa indicate the counterbalancing order of the pre-
sentation sequences for each trial and are read downward (o-odorant presented first, then diluent; d-diluent presented first, then odorant).
In the first reversal point (where five correct sets of pairs occur at the same concentration), the fifth order sequence is determined by the
first o or d of the subsequent column of four order sequences. (From Doty, 1991a.)
208 Doty and Laing

B. Difference Threshold Tests

In classical psychophysics, the smallest amount by which

a stimulus must be changed to make it perceptibly stronger
or weaker is termed a “just noticeable difference,” or JND.
This value is also called a difference or differential thresh-
old (in contrast to an absolute threshold, as described
above). The size of the increment in odorant concentration Figure 4 Hypothetical distributions of signal plus noise (SN) and
(I) required to produce a JND increases as the compari- noise alone (N) plotted on the same axes. When the strength of the
son concentration (I) increases, with the ratio approximat- perceived signal increases, the SN distribution moves to the right,
ing a constant; i.e., I/I
K (Weber’s law) (Weber, 1834). increasing d’, the measure of the distance between the two distrib-
K is a rough index of the sensory system’s sensitivity (i.e., utions in standard deviation units (z-scores). (From Doty, 1976.)
the smaller the K value, the more sensitive the system is to
fine changes in stimulation). However, numerous studies
suggest that K is not a constant, being influenced by the a traditional non–forced-choice detection threshold para-
size of I, particularly at the extremes of the sensory con- digm, the investigator would conclude that these two sub-
tinuum (Doty, 1991a). jects differed in sensitivity to the stimulus, when, in fact,
An example of a brief clinical test used to establish a dif- they only differed in regards to their response biases.
ference threshold is described by Eichenbaum et al. (1983). SDT assumes that a stimulus is imbedded within a
In this test, 10 binary dilutions (in water) of acetone, background of noise. Noise can arise from a variety of
ethanol, almond extract, and lemon extract were presented. sources and can be conceptualized at a number of levels
Initially, the highest and lowest concentrations of a given (e.g., variations in attention, stimulus fidelity, neural firing
odorant were presented and the subject was required to unrelated to the stimulus, fluctuations in distracting
choose the stronger stimulus. Successively stronger stimuli physiological processes). In most cases noise is assumed to
were then paired with the strongest stimulus until, on the be normally distributed (as is done here to simplify discus-
last of the 10 trials, the two samples were identical. sion). Whenever a signal is added to the “noise” (N) distri-
Eichenbaum operationally defined the difference threshold bution, a “signal plus noise” (SN) distribution results. Both
as the lowest concentration for which discrimination up to the N and SN distributions can be placed on the same set
and including the dilution was effortless. of axes, as shown in Figure 4. The measure of the subject’s
sensitivity is the distance between the means of these
C. Signal Detection Tests distributions.
The concept of the response criterion is illustrated for a
Signal detection theory (SDT) differs fundamentally from hypothetical subject in Figure 5 (Doty, 1991a). On any
the approach of sensory measurement inherent in classical given trial, a low-concentration odorant (SN) or a blank
threshold theory. Thus, SDT rejects the notion of a thresh- stimulus (N) is presented, and the subject’s task is to report
old (whether absolute or differential) and focuses on (1) noise whether or not an odor was presented. Reports of “yes” are
and signal plus noise as the milieu of the detection situation represented by the areas under the N and SN curves to the
and (2) the influences of subject expectancies and rewards right of the vertical line depicting the subject’s response
on the detection decision. Signal detection procedures pro- criterion, whereas reports of “no” are indicated by the
vide both a measure of sensory sensitivity and the subject’s areas to the left of this line. In case 1, the subject exhibits
response criterion or bias (Tanner and Swets, 1954). In a very liberal criterion, reporting the presence of an odor
effect, the response criterion is the internal rule used by a on the majority of the SN trials () and on half of the N tri-
subject in deciding whether or not to report detecting a als (). Thus, although correct detection of the odorant
stimulus (e.g., the liberalism or conservatism in reporting a occurred nearly all of the time (), many false alarms ()
sensation under uncertain circumstances). For example, were present. Perhaps in this instance the subject was
two subjects may experience the same subtle degree of sen- rewarded for reporting the detection of an odor and not
sation from a very weak stimulus. One, however, may admonished for making false alarms. In case 2, the subject
report that no sensation was perceived (e.g., perhaps chose a less liberal response criterion. Although fewer cor-
because of lack of self-confidence), whereas the other may rect detections of the odor were made (), fewer false
report the presence of the sensation. In both cases, the stim- alarms were also made (). In case 3, the observer chose a
ulus was perceived to the same degree. However, the two very conservative response criterion, making few false
subjects had different criteria for reporting its presence. In alarms but similarly making fewer correct detections. This
Psychophysical Measurement of Human Olfactory Function 209

determining d for any combination of hit and false-alarm

proportions is to use the table provided by Elliot (1964). In
addition, nonparametric signal detection measures are also
available (Brown, 1974; Frey and Colliver, 1973; Grier,
1971; Hodos, 1970; but see Macmillan and Creelman,
1996), as are methods for testing the parametric assump-
tions of traditional signal detection analysis (Gescheider,
1976; Green and Swets, 1966).
The classical parametric measure of response bias is
termed . Not to be confused with the  in Figure 5,  rep-
resents the ratio, at the criterion point, of the ordinate of
the SN distribution to that of the N distribution. This value
can be easily calculated from the hit and false-alarm rates
by use of ordinate values from the normal curve, as dis-
cussed by Gescheider (1976).
Despite the fact that hundreds of trials have traditionally
been used in signal detection studies, some chemical sens-
es studies have employed far fewer trials, largely out of
practicality considerations. For example, Potter and
Butters (1980) and Eichenbaum et al. (1983) computed
Figure 5 Hypothetical examples of how the response
d using only 30 test trials. Even though such estimates are
criterion can vary when perceptual sensitivity (d) remains con- somewhat unstable (because a test’s reliability is a function
stant. In case 1, a liberal criterion was chosen in which a rela- of its length), they may be less so than typically assumed,
tively large number of false positives occurred [i.e., α, the and there is at least some empirical rationale for the use of
reports of the presence of odor when the blank (N) is presented]. abbreviated signal detection tests. Thus, O’Mahony et al.
In cases 2 and 3 more conservative criteria were chosen, (1979b), in a study of gustatory sensitivity to sodium chlo-
decreasing both the number of false positives (α) and hits ride, found that Brown’s (1974) nonparametric R index
(β). Traditional threshold measures confound the influences of fell, after 40 trials, within 5% of the values obtained after
perceptual sensitivity and the setting of the response criterion. 200 trials in slightly over half the subjects tested. However,
(From Doty, 1976. Copyright © 1976, Academic Press.) an analogous olfactory study has not been performed, and
ideally all of the subjects should evidence such response
stability. For these reasons it is prudent to use as many tri-
would tend to result, for example, when a subject is als as possible in signal detection tasks.
penalized for making false positives and given few rewards
for successful detection of the odor. In all three of these D. Suprathreshold Scaling Procedures
hypothetical cases, the sensitivity (i.e., d) was equivalent,
as indicated by the constant distance between the N and A number of psychological attributes can be assigned to
SN distributions. odors, including strength, pleasantness, and quality.
In a typical olfactory experiment employing SDT, the Although the first of these attributes changes in a system-
subject is presented with a large number of trials of a sin- atic way with stimulus concentration, odorant pleasantness
gle low concentration of odorant interspersed with blank or unpleasantness is more variable and idiosyncratic
trials (Doty et al, 1981; Semb, 1968). Even though the (see Doty, 1975). In regard to odor quality, only in rare
number of blank and odorant trials need not be equivalent, instances is it dramatically altered by changes in
this is commonly the case. The proportion or percent of the suprathreshold odorant concentration. Since the perceived
total odor trials (S) on which a subject reports detecting an intensity of an odorant is a function of its concentration,
odor (the hit rate) is calculated, as is the percent of blank ratings or other measures of perceived intensity have been
trials (N) on which an odor is reported (the false alarm used to evaluate olfactory function. Because the intensity
rate). The parametric sensitivity measure, d, can then be of a stimulus is related to the number of neurons that are
computed by converting the proportions to normal distrib- recruited and the frequency at which they fire, such meas-
ution standard deviation values (z-scores) via a normal ures may relate to the extent of neural damage present in
probability table; d equals the z-score for hits minus the the afferent pathway (Drake et al., 1969). However,
z-score for false alarms. A more convenient procedure for suprathreshold rating or scaling methods appear to be less
210 Doty and Laing

sensitive to olfactory dysfunction than a number of other magnitude of each member of a stimulus set is estimated
tests (e.g., detection threshold tests and tests of odor identi- by using some other sensory modality or cognitive
fication), although they have the advantage of being rela- domain. A key difference between this procedure and rat-
tively brief, easy to administer, and less susceptible than ing scale procedures is that the ratio relations among the
threshold tests to subtle stimulus contamination. Negative intensities of the different stimuli are defined, and the sub-
findings, however, must be conservatively interpreted, as ject’s responses are not confined to categories or a short
in some cases suprathreshold rating scales have completely response line. Continua commonly used in the cross-modal
missed major changes observed by other methods (e.g., the matching task termed magnitude estimation include num-
influences of age on olfactory function) (see Rovee ber (e.g., assigning numbers proportionate to an odor’s
et al., 1975). intensity) and distance (e.g., pulling a tape measure a dis-
Despite the fact that olfactory psychophysicists and tance proportional to an odor’s intensity) (Berglund et al.,
psychometricians have sought to develop psychological 1971; Stevens, 1961). When intensities of sensations from
scales with ruler-like properties (i.e., the so-called ratio two or more modalities are judged on a single common
scale, where distances along the scale have ratio properties scale, the procedure is termed the method of magnitude
and a true zero point is present), the degree to which this is matching. Magnitude estimation and magnitude matching
possible is debatable. Judgments of the intensity of odors are among the most commonly used cross-modal matching
must be viewed as relative, as they are markedly influ- procedures and are discussed in more detail below.
enced by both subject idiosyncrasies and contextual factors In the typical magnitude estimation paradigm, the sub-
(e.g., a moderately intense odor is reported to be more ject assigns numbers relative to the magnitude of the sen-
intense when presented with weak comparison stimuli than sations. For example, if the number 60 is used to indicate
with strong comparison stimuli) (Eyman et al., 1975; the intensity of one concentration of an odorant, a concen-
Helson, 1964). Fortunately, for the purposes of clinical tration that smells four times as intense would be assigned
testing, neither the exact form of the underlying psycho- the number 240. If another concentration is perceived to be
logical scale nor the influences of stimulus context need to half as strong as the initial stimulus, it would be assigned
be of great concern to the examiner, as long as the test pro- the value 30. The examinee can assign any range of
cedures are standardized and it can be demonstrated that numbers to the stimuli, as long as they reflect the relative
the responses on the scaling tasks are reliable and differ- magnitudes of the perceived intensities. In some cases, a
entiate among persons with differing degrees of olfactory standard for which a number has been preassigned (often
function. the middle stimulus of the series) is presented to the sub-
Rating scales can be used to estimate the relative ject in an effort to make his or her responses more reliable.
amount of a psychological attribute perceived by a subject. In other cases, the individual is free to choose any number
In chemosensory assessment, two types are popular: cate- system he or she wishes, as long as the numbers are made
gory scales, where the relative amount of a sensation is sig- proportional to the magnitude of the attribute (the “free
nified by indicating which of a series of discrete categories modulus method”). For example, one subject may choose
best describes the sensation, and line scales (also termed to assign the first stimulus the number 250, whereas
visual analog or graphic scales), where the subject or another may choose to assign this same stimulus the num-
patient indicates the strength of the sensation by placing a ber 5. If a second stimulus is perceived to be 10 times
mark along a line that has descriptors (termed anchors) stronger than the first by each of these individuals, the first
located at its extremes (e.g., very weak–very strong). one would assign the number 2500, whereas the second
Recently, scales have been developed in which logarithmic one would assign the number 50. The important point is
elements have been incorporated into their design in an that the absolute values of the numbers are not important;
effort to overcome ceiling effects and to more closely only the ratios between them are relevant.
mimic ratio-like properties of magnitude estimation (see To obtain an index of suprathreshold function, magni-
below) (e.g., Green et al., 1996; Neely et al., 1992). The tude estimation data are most commonly plotted on log-log
reader is referred elsewhere to discussions of the properties coordinates (log magnitude estimates on the ordinate and
of rating scales, including the influence of category num- log odorant concentrations on the abscissa) and the best
ber on their psychometric properties (Anderson, 1970; line of fit determined using linear regression. The resulting
Doty, 1991b; Guilford, 1954). function, log P
n log  log k, where P
perceived
Intensity matching procedures have also been applied in intensity, k
the Y intercept,
stimulus concentration,
the clinical and other applied settings, with cross-modal and n
the slope, can be represented in its exponential
matching procedures (e.g., magnitude estimation) being form as a power function, P
k n, where the exponent n
the most popular. In cross-modal matching, the relative is the slope of the function on the log-log plot. In olfaction,
Psychophysical Measurement of Human Olfactory Function 211

it is likely that such problems can be minimized by ensur-

ing that the instructions, test procedures, and test stimuli are
carefully standardized and monitored. Comparative assess-
ments of nine-point rating scales, line scales, magnitude
estimation scales, and a hybrid of category and line scales
suggest that, for untrained or mathematically unsophisticated
subjects, category scales and line scales may be superior to
magnitude estimation when such factors as variability,
reliability, and ease of use are considered (Lawless and
Malone, 1986a,b).
Since the magnitude estimation function’s intercept and
height above the origin depend to a large degree on idio-
syncratic differences in the use of numbers and the specific
magnitude estimation method employed (e.g., fixed vs. free
modulus), only its slope has traditionally been used as an
index of sensory function. In an attempt to gain additional
information from the function’s ordinate position, investi-
gators have employed the method of cross-modal
Figure 6 Relationship between perceived magnitude of three
magnitude matching, which provides, at least theoretically,
types of stimuli, as measured by magnitude estimation, and stim-
ulus magnitude. Note that the perceived intensity of the example information about the perceived intensity of stimuli from
odorant increases in a negatively accelerated fashion, indicating a the absolute position of the magnitude estimation function
power function exponent less than 1 (in this case 0.33). (Adapted and corrects, to some degree, for differences among sub-
and modified from Stevens, 1961.) jects in number usage (for a detailed discussion of this
procedure, see Marks et al., 1988). In the most common
application of this method, judgments of the intensity of
n varies in magnitude from odor to odor, but is generally sensations from two modalities (e.g., loudness and odor
less than 1, reflecting a negatively accelerated function on intensity) are made on a common magnitude estimation
linear-linear coordinates (Fig. 6). As noted elsewhere, vari- scale (Marks et al., 1986). Under the assumption that
ous investigators have made modifications in these equa- subjects experience stimuli on one of the continua (i.e.,
tions in an attempt to take into account such factors as loudness) in a similar manner (an assumption that some
threshold sensitivity and adaptation (Doty, 1991a; question), differences among their loudness ratings would
Overbosch, 1986). be expected to reflect differences in number usage. The
It is noteworthy that magnitude estimation, perhaps odor intensity continuum can then be adjusted accordingly.
more so than most other sensory procedures, can be biased Such normalization allows, theoretically, for a direct com-
or influenced in systematic ways by procedural and subject parison of scale values across subjects; thus, if the adjusted
factors (Doty, 1991a; Marks, 1974). The magnitude odor intensity magnitude value for one subject is 10 and for
estimation task is relatively complex in that accurate another subject is 20 at the same concentration level, the
responses to a stimulus require a good memory for the second subject is presumed to experience twice the odor
prior stimulus. If too much time lapses between the pre- intensity as the first subject.
sentation of stimuli, the memory of the prior stimulus
fades. On the other hand, if the trials are spaced too E. Quality Discrimination Tests
closely together, adaptation can distort the relationship.
Not all subjects consistently provide ratio estimates of The most straightforward chemosensory quality discrimina-
stimuli, and a number do not understand the concept of tion test requires individuals to decide whether two stimuli
producing ratios (Baird et al, 1970; Moskowitz, 1977). have the same or different quality. In one scenerio, a series
Furthermore, the magnitude of the exponent is dependent of same-odorant and different-odorant pairs is presented,
on the choice of the stimulus scale (i.e., the units in which and the proportion of pairs that are correctly differentiated is
the stimulus concentration is expressed), although in olfac- taken as the measure of discrimination (O’Mahony, 1979;
tion this is probably of minor consequence (Myers, 1982). O’Mahony et al., 1979; Potter and Butters, 1980;). Variants
The degree to which these and other potential shortcom- on this theme include picking the “odd” stimulus from a set
ings hinder the use of magnitude estimation procedures in from which only the “odd” stimulus differs (e.g., the so-
applied settings, such as the clinic, is not known; however, called triangle test) (Frijters et al., 1980).
212 Doty and Laing

Another form of discrimination test is based on a proce- required to match the stimuli, one by one, to those of a set
dure called multidimensional scaling (MDS). In one variant of identical stimuli. As an example, Abraham and Matha
of this procedure, ratings are made for all possible pairs of (1983) presented subjects with eight vials that contained
stimuli (or selected subsets of pairs) on a line scale four odorants (two vials per odor). The subject’s task was
anchored with descriptors like “completely different vs. to pair up the equivalent two-vial containers. The number
exactly the same,” and the correlation matrix among these of pairs correctly matched on each of two administrations
ratings is subjected to an algorithm that places the stimuli of the test was used by these authors as the test score.
in two- or more dimensional space relative to their per-
ceived similarities (e.g., Schiffman et al., 1981). The G. Quality Identification Tests
process is akin to constructing a map of a country from
a list of distances available between the cities of that coun- Among the most popular procedures for assessing taste
try. Persons with poor discrimination abilities fail to discern and smell function are those that require stimulus quality
differences and similarities among stimuli, as illustrated by identification. Such tests can be divided into three groups:
multidimensional spaces that have no distinct or reliable naming tests, yes/no identification tests, and multiple-
groupings. Because of its time-consuming nature and the choice identification tests. The respective responses
fact that statistical procedures for comparing one person’s required, on a given trial, in these three classes of tests are
MDS space to another’s (or to a norm) are poorly worked (1) to provide a name for the stimulus, (2) to signify
out, MDS has not been used routinely in the clinic. whether the stimulus smells like an object named by the
Interestingly, when subjects are asked to rate the similarity examiner (e.g., does this smell like a rose?), and (3) to
of stimuli that are only indicated to them by name (i.e., the identify the stimulus from a list of names or pictures.
odorants, per se, are never presented), stimulus spaces Odor naming tests in which no response alternatives are
derived by MDS are analogous to those obtained by the provided have been used clinically (e.g., Gregson et al.,
actual use of the odorants (Carrasco and Ridout, 1993; 1981) but are of limited value since many normal individ-
Ueno, 1992). This implies that well-defined imagery, or at uals have difficulty in naming or identifying even familiar
least conceptual representations, exist for odorous stimuli. odors without cues. Yes/no identification tests are much
Recently, Wise and Cain (2000) used a response more useful, since they require a patient to report whether
latency approach to determine the discriminability of or not each of a set of stimuli smells like a particular sub-
unmixed odors and mixed odors. A clear monotonic rela- stance named by the experimenter. Two trials with each
tionship was found between latency and accuracy, with stimulus are usually given, with the correct alternative pro-
latency decreasing with accuracy. In addition, subjects vided on one trial and an incorrect one on the other (e.g.,
required more time and made more errors in discrimina- orange odor is presented and the subject is asked on one
tions between binary mixtures and their unmixed compo- trial whether the odor smells like orange and on another
nents than between the unmixed components. It was trial whether the odor smells like peppermint). Although
concluded that this approach may provide a novel measure such a test requires the patient to keep the percept in mem-
of differences in odor quality, since latency provides infor- ory long enough to compare it with the target word (which,
mation about discriminability. of course, must also be recalled from memory), some of its
proponents argue that it is less influenced by cognitive and
F. Quality Recognition Tests memory demands than multiple-choice identification tests
(see below). Since chance performance on this type of test
Two general classes of quality recognition tests can be is 50% compared to 25% on a four-alternative multiple-
defined. In the first class, the subject is asked whether each choice identification test, its range of discriminability is
stimulus of a presented set is recognized. Identification is lower, and therefore more trials are needed to obtain the
not required. As indicated at the beginning of the chapter, same statistical power as the multiple-choice odor
this procedure is relatively crude, despite the fact that it is identification test.
perhaps the most common means used by neurologists to Numerous multiple-choice odor identification tests
measure olfactory function (Sumner, 1962). In the second have been described in the clinical literature (Cain et al.,
class, a patient is presented with a “target” stimulus and 1983; Doty, 1991b; Doty et al., 1984a; Gregson et al.,
subsequently asked to select the target from a larger set of 1981; Wood and Harkins, 1987; Hummel et al., 1997;
stimuli. The number of correct responses of a series serves Wright, 1987). These tests are conceptually similar and, in
as the test score. the few cases that have been examined, strongly correlated
A variant on this theme is the stimulus matching task, in with one another (Cain and Rabin, 1989; Doty et al.,
which a set of stimuli are provided and the subject is 1994; Wright, 1987). The most widely used of these tests
Psychophysical Measurement of Human Olfactory Function 213

[the University of Pennsylvania Smell Identification Test (termed the target or inspection stimulus or stimulus set)
(UPSIT), commercially termed the Smell Identification and to select, after an interval of time (e.g., 30 sec up to
TestTM, Sensonics, Inc., Haddon Heights, NJ] examines several days), that odorant or set of odorants from foils
the ability of subjects to identify, from sets of four (distracters). Repeated trials may be performed at one or
descriptors, each of 40 “scratch and sniff ” odorants (Fig. more retention intervals for each of several stimuli or sets
1) (Doty, 1995; Doty et al., 1984a,b). The number of cor- of stimuli. In an effort to minimize the rehearsal of verbal
rect items out of 40 serves as the test measure; this value labels reflecting the odor qualities or referents during the
is compared to norms and a percentile rank is determined, delay intervals, the examinee is sometimes asked to per-
depending on the age and gender of the subject (Fig. 1F) form an unrelated task during the retention period, such as
(Doty, 1995). This test has several unique features, includ- counting backwards by twos or threes. The proportion of
ing amenability to self-administration and a means for trials where correct performance occurs is a typical mea-
detecting malingering (see Sec. VI). Furthermore, it is sure derived from such tests.
available in English, French, German, and Spanish ver- The results from an odor memory test must be inter-
sions. The popularity of this test is attested to by the fact preted with caution. Despite attempts to minimize labeling
that hundreds of scientific publications have arisen from of the inspection odor with a familiar word or item on the
its use by investigators from many laboratories and clinics. part of a subject, such labeling undoubtedly occurs, and,
Several odor identification confusion matrix tests have thus, what is being measured across intervals is the mem-
been described that are applicable to clinical settings ory of the label, not the memory of the odor. In other
(Köster, 1975; Wright, 1987). The test that has been most words, once an individual recognizes an odor as that of an
widely applied is that of Wright’s (1987). In his test, each orange, all that has to be remembered over time is the con-
of 10 suprathreshold stimuli is presented to a patient in cept “orange,” not the specific smell of the orange. Later,
counterbalanced order 10 times (100 total trials). The when given stimuli from which to select the earlier per-
response alternatives are the names of the 10 stimuli: ceived odor, the subject simply looks for the smell of an
ammonia, chlorine bleach, licorice, mothballs, peppermint, orange (which has been known for much of his or her life).
roses, turpentine, vanilla, Vicks vapor rub, and vinegar. No In effect, the odor is not what is being uniquely remem-
feedback as to the correctness of the subjects’ responses is bered over the retention interval, only its name or concept.
given. The percentage of responses given to each alternative For this reason, investigators have attempted to employ
for each odorant is determined and displayed in a rectangu- novel, nondescript, and unfamiliar odorants in such tasks.
lar matrix (stimuli making up rows and response alterna- Unfortunately, it is difficult to find target odors and foils
tives making up equivalently ordered columns). Responses that are not readily labeled by subjects as pleasant or
along the negative diagonal therefore represent correct unpleasant, fruity or nonfruity, medicine-like or non– med-
responses, whereas those that fall away from the diagonal icine-like, etc. In general, both short- and long-term odor
represent “confusions.” The percentage of correct recognition is markedly facilitated by verbal encoding
responses is used as the main test measure, although some (Jehl et al., 1997).
of its proponents argue that the confusions (off-diagonal Another point that should be stressed about odor mem-
responses) may provide meaningful clinical information. ory tests is that the performance across the delay intervals
The main limitations of Wright’s confusion matrix are comprises the “memory” component of the task, not the
(1) its long administration time (approximately 45 min) overall test score. Thus, an odor memory test is essentially
and (2) the lack of evidence that the off-diagonal responses an odor discrimination test with varying inspection (delay)
provide any meaningful clinical information (although intervals. If, for example, scores on a nominal odor mem-
such responses may be of value in detecting malingering) ory task differ between two groups (as evidenced by a
(see Kurtz et al., 1999). It would seem that if off-diagonal main group effect in an analysis of variance), then a sig-
responses are to be sensitive to aberrations or distortions nificant interaction term between delay interval and group
seen in most clinical cases, more subtle differences in the must be present for such scores to reflect differences in
response alternatives need to be employed within the odor memory per se. Without an interaction with delay
matrix. Should subtle aberrations be reliably categorized, interval, the difference would reflect discrimination, not
this approach would have considerable clinical value. memory. That being said, a number of examples of clinical
applications of odor memory tests are available from the
H. Memory Tests literature. Unfortunately, convincing evidence for a true
odor memory deficit is lacking in most cases.
In a basic odor recognition memory test, a subject is Campbell and Gregson (1972) developed a test of short-
required to smell an odorant or a small set of odorants term odor memory in which four odors in a row were
214 Doty and Laing

presented and the patient was asked if the fourth, which was (accurately measures what it portends to measure). Related
the same as one of the first three, was equivalent to the first, to a test’s validity are its sensitivity (ability to detect abnor-
second, or third odorant. No delay interval, per se, was malities when present) and specificity (ability to detect
defined between the presentation of the stimuli, but pre- abnormalities with a minimum number of false positives).
sumably the trials were presented closely after one another. Although a test cannot be valid without being reliable, the
Seven three-odor combinations of 12 inspection stimuli reverse is not the case; i.e., a test can be reliable but not
were administered. Patients who had difficulty with this valid. Despite the fact that measures of test reliability and
task were subsequently given two-odor combinations. The validity are available for many medical and psychological
test score was the number of odors that were consistently tests, this is not the case for most olfactory tests. Indeed,
recognized by the subject. This test was shown to be sensi- measures of validity (other than a few intercorrelations
tive to olfactory deficits due to schizophrenia (Campbell among different tests) are extremely rare; hence, in this
and Gregson, 1972), Kallmann’s syndrome (Gregson and chapter studies of reliability are emphasized (for more dis-
Smith, 1981), and Korsakoff psychosis (Gregson et al., cussion on this point, see Schwartz, 1991).
1981). However, it is debatable whether the scores truly The reliability of a test can be determined in several ways.
reflect memory processes per se. First, the test can be administered on two occasions to each
Jones et al. (1975) presented 20 pairs of odorants at member of a group of subjects and a correlation coefficient
0- and 30-second delay intervals to 14 alcoholic Korsakoff computed between the test scores on the two occasions
psychosis patients, 14 alcoholic controls, and 14 nonalco- (termed the test-retest reliability coefficient or the coefficient
holic controls. On a given trial, the subject’s task was to of stability). Second, when parallel forms of a test are avail-
report whether the second stimulus was the same as or dif- able, the two forms can be administered to the same set of
ferent from the first. In the 30-second delay interval, the subjects and a correlation coefficient computed between the
subjects counted backward by threes. Since the Korsakoff two forms. Third, subsections of some types of tests (e.g.,
psychosis patients performed significantly more poorly multiple-item odor identification tests) can be correlated with
than did the control groups at both the 0- and 30-second one another to provide an estimate of test stability. The test is
retention intervals, it is questionable whether odor memory viewed, in this case, as consisting of parallel forms, and the
is the trait being influenced in this case. resulting coefficient, when based upon the correlation of half
More recently, Jones-Gotman and Zatorre (1993) reported of the items with the other half of the items, is termed the
that, in patients having undergone surgical cerebral extirpation split-half reliability coefficient. Since reliability is related to
for control of epilepsy, odor memory impairment was noted test length, as will be noted below, a statistical correction for
between the controls and two of the eight surgical groups test length must be applied to the correlation coefficient
evaluated—namely, those who had received excision from the obtained in this way to provide the correct reliability coef-
right temporal or right orbitofrontal cortices. The memory task ficient for the full test (Guilford, 1954).
consisted of eight target odors and eight new foils, and the yes- The magnitude of a reliability coefficient depends, to
no recognition testing was performed twice after the initial a large degree, on the variation of the test scores of the
testing—20 minutes later and 24 hours later. Although the group upon which it is computed. If all members of a group
authors interpret their findings as evidence of a “right hemi- score exactly the same on a test administered on two test
sphere predominance in odor memory,” their underlying data occasions, the reliability coefficient will not be able to be
do not support the notion that differences in odor memory, per computed, even though, in effect, there is a perfect correla-
se, were present among the groups. Thus, in the overall analy- tion between the test scores on the two occasions. If only a
sis, where the test scores at the various delay intervals were small variation occurs among the subjects, then the reliabil-
evaluated as a function of operative group and delay interval, ity coefficient may be spuriously low. Thus, in assessing
main effects of both of these factors were noted, but no inter- reliability one must have some understanding of the varia-
action between them was present. No interactions with delay tion among the test scores. Also, it should be noted that
interval were noted in any subgroup analyses. Hence, this while a high reliability coefficient indicates that a group of
study suggests that odor discrimination is altered by certain individuals scored similarly relative to one another on a test
cerebral excisions, but not necessarily odor memory. from one test occasion to the other, all of the individual’s
test scores still may be lower (or higher) on the second than
on the first test occasion. In other words, systematic
IV. TEST RELIABILITY changes in the test values can occur which are not reflected
in the reliability coefficient. In such a case, a high reliabil-
The utility of an olfactory test reflects the degree to which ity coefficient is misleading, as the overall stability of the
it is reliable (consistent, dependable, or stable) and valid test may vary systematically over time.
Psychophysical Measurement of Human Olfactory Function 215

Although there is a trend among modern developers of single ascending series n-butanol and PEMEC detection
olfactory tests to assess the reliability of their instruments, thresholds were 0.49 and 0.70, respectively. The reliability
there is a dearth of information on this point in the vast of the detection threshold component of the Sniffin’ Sticks
majority of cases. In general, forced-choice odor identifi- test has been reported to be 0.61 (Hummel et al., 1997).
cation tests with a relatively large number of items evi- Doty et al. (1995) concluded that (1) detection threshold
dence a high degree of reliability (e.g., both the test-retest values are more reliable than recognition threshold values,
and split-half r’s of the 40-item UPSIT are consistently (2) thresholds based upon a single series AML procedure
above 0.90) (Doty et al., 1984a, 1985, 1987, 1995). Shorter are less reliable that thresholds based upon a staircase pro-
identification tests evidence lower reliability. For example, cedure, (3) reversal location within a staircase series has no
the test-retest reliability of the 16-item Scandinavian Odor- influence on reliability, and (4) a clear relationship
Identification Test is 0.79 (Nordin et al., 1998) and that of between reliability and test length (e.g., number of stair-
the 12-item self-administered B-SIT is 0.73 (Doty et al., case reversals) exists. Importantly, in a related study it was
1989). Recently, the reliability of the identification com- found that the threshold measures tended to load on the
ponent of the ‘Sniffin’ Sticks’ test was reported to be 0.73 same principal component in a principal components
(Hummel et al., 1997). analysis as a number of the other test measures evaluated
Since it has been reported that olfactory thresholds vary (e.g., the UPSIT, a yes/no odor identification test, and tests
considerably among individuals and evidence considerable of odor discrimination), suggesting that all of these tests
day-to-day fluctuations within the same individuals measure a common sensory domain (Doty et al., 1994).
(Stevens et al., 1988), one might expect their reliability to
be suspect. Indeed, reliability coefficients for various
threshold tests do vary considerable from study to study,
V. OTHER CONSIDERATIONS
and extremely low reliability coefficients have been noted
in some cases (e.g., Heywood and Costanzo, 1986; Punter, A. Unilateral Versus Bilateral Testing
1983). Nonetheless, particularly in cases where repeated
estimates of the threshold are obtained, respectable reli- Most individuals with chemosensory dysfunction evidence
ability coefficients have been reported. Jones (1955), for the dysfunction bilaterally (Cain and Rabin, 1989). In
example, presented ascending concentrations of n-butanol, cases where unilateral losses are present, they are often
safrol, and n-butyric acid in sniff bottles (with a compari- unnoticed. When time is at a premium, bilateral testing is
sion blank for reference) to 24 college students. The series preferable to unilateral testing since it reflects clinically
were repeated six times for each subject for each stimulus, meaningful deficits. However, there are a number of occa-
and the subjects were required to recognize the substance. sions when unilateral olfactory testing is of considerable
Reliability coefficients, based upon intraclass correlations, value (e.g., in the detection of some types of tumors)
were 0.82, 0.77, and 0.80, respectively, for the three sub- (Doty, 1979), and the ideal assessment of a patient includes
stances. In a study of 40 subjects, Koelega (1979) reported unilateral, as well as bilateral, testing.
test-retest reliability coefficients for a four-alternative Unilateral testing is straightforward. Although it is pos-
forced-choice n-amyl acetate threshold test to be 0.65, sible to present a stimulus to one naris and obtain mainly
0.51, and 0.59 for bilateral, right nostril, and left nostril unilateral stimulation, the possibility of the crossing of
presentations, respectively. In a study of 32 subjects rang- odorant to the contralateral side within the rear of the
ing in age from 22 to 59 years, Cain and Gent (1991) nasopharynx upon exhalation cannot be excluded. Thus, it
reported left nostril:right nostril correlations of 0.68, 0.96, is prudent to close the contralateral naris without distorting
0.86, and 0.83, respectively, for detection thresholds from the septum [e.g., by using a piece of MicrofoamTM tape
single ascending series presentations of butanol, phenyl (3M Corporation, Minneapolis, MN) cut to fit tightly over
ethyl methyl ethyl carbinol (PEMEC), isoamyl butyrate, the borders of the naris] and have the patient exhale
and pyridine. Doty et al. (1995) found test-retest reliability through the mouth after inhaling through the nose (Doty
coefficients for detection thresholds of the six odorants et al., 1992). As in the case when both nares are blocked,
contained within the non–forced choice T&T olfactometer this precaution decreases the likelihood for air to enter the
test series (skatole, isovaleric acid, -undecalactone, blocked nasal chamber via the retronasal route.
-phenyl ethanol, cyclotene) to range from 0.56 to 0.71; Furukawa et al. (1988) noted that 7 of 94 patients (7%)
recognition thresholds coefficients were lower, ranging they examined, all of whom evidenced no bilateral thresh-
from 0.22 to 0.45. The reliability of the single staircase old deficits, evidenced significant unilateral threshold
forced-choice phenyl ethyl alcohol detection threshold was deficits. They reported a similar phenomenon in 6 of 12
found to be 0.88, whereas the reliability coefficients for patients who had had brain surgery. Of 82 consecutive
216 Doty and Laing

nonanosmic patients presenting to the University of smell are age, gender, and smoking habits. Of these three
Pennsylvania Smell and Taste Center with chemosensory factors, age is the most important (for reviews, see Doty,
dysfunction, 14 (i.e., 17%) were observed whose unilateral 1991a; Doty and Snow, 1988; Schiffman, 1993). Indeed,
detection thresholds were discrepant from one another by over the age of 80 years, nearly three out of four persons
at least three orders of magnitude (Doty, unpublished). exhibit marked olfactory dysfunction; half of those
Interestingly, nine of these 14 individuals were anosmic on between the ages of 65 and 80 years evidence such dys-
one side of the nose, even though only three had bilateral function (Doty et al., 1984b). Age-related declines in
detection threshold values that were obviously abnormal. olfactory performance are observed for a variety of olfac-
tory tests, including tests of odor detection threshold, iden-
B. Detection of Malingering tification, discrimination, adaptation, and suprathreshold
odor intensity perception (for reviews, see Corso, 1981;
Because considerable compensation can be available in Doty, 1990; Murphy, 1986; Schiffman et al., 1979;
accident cases for alterations in ability to smell, malinger- Weiffenbach, 1984). In addition, age influences the
ing on chemosensory tasks is not uncommon. It is fre- responsiveness of the nasal mucosa to volatile chemicals
quently suggested in the medical literature that if a patient that produce irritation and other skin sensations (Stevens
cannot readily perceive the vapors from an irritating sub- and Cain, 1986). In general (1) large individual differences
stance presented to the nose, he or she is malingering (e.g., are present in the test scores of older individuals, (2) olfac-
Griffith, 1976). However, this is not a definitive method for tory dysfunction is most evident after the sixth decade of
detecting malingering. Thus, individuals who, on other life, and (3) women, on average, evidence age-related
grounds, are believed to be feigning anosmia usually have declines in odor perception at a later age than do men.
difficulty in denying experiencing the effects of NH4 or The decline in the ability to smell in later life is not
other irritants, particularly since these stimuli often pro- inconsequential. Thus, a disproportionate number of older
duce eye watering, coughing, and other reflexes that are persons die from accidental gas poisoning (Chalke et al.,
manifested overtly. Furthermore, there appears to be con- 1958), and many complain that their food has no flavor
siderable variability among normal individuals in trigemi- (Doty et al. 1984b). The latter phenomenon, which can
nal responsiveness to such stimulants. lead to decreased interest in food, may explain some cases
A more valid approach for detecting cheating on the of age-related nutritional deficiencies. As documented
basis of psychophysical testing is to examine response clinically (e.g., Deems et al., 1991), decreased “taste” per-
strategies of patients on forced-choice tests, since malin- ception during deglutition largely reflects the loss of stim-
gerers often avoid the correct response more often than ulation of the olfactory receptors via the retronasal route
expected on the basis of chance. This is well illustrated by (Burdach and Doty, 1987; Mozell et al., 1969).
responses to the UPSIT. Since the UPSIT is a four-alterna- In general, women of all ages outperform men on tests of
tive forced-choice test, approximately 25% of the test odor identification, detection, discrimination, and suprathresh-
items (i.e., 10) are correctly answered, on average, by an old intensity and pleasantness perception (Cain, 1982b; Doty,
anosmic. The probability of scoring 5 or less on the UPSIT 1986; Doty et al., 1984a; Koelega and Köster, 1974; Le
and not having at least some ability to smell is less than Magnen, 1952). Such differences are present for a wide vari-
5 in 100. The probability of scoring zero on the UPSIT and ety of odorants, including human breath and bodily secretions
having no sense of smell is less than 1 in 100,000. (Doty et al., 1975, 1978b, 1982), and are observed as early as
As noted in Chapter 11, electrophysiological measures such testing can be reliably performed (Doty, 1986). The fact
are now available that distinguish between intranasal stim- that female babies more readily show a preference for odors
ulation of the olfactory and trigeminal systems. Although from their own mothers than do male babies suggests that such
such testing is not possible in all persons, it does allow for sex differences are present at birth and are either inborn or due
a determination as to whether gross responses are present to early developmental sexually dimorphic influences (Makin
in the olfactory system, adding key information as to the and Porter, 1989) (see Chapter 15).
likelihood of malingering. The influence of tobacco smoking on olfactory function
is less marked, on average, than that of age or gender (e.g.,
C. Subject Variables Doty et al., 1984b). This influence, however, is dose-
related and present in both previous and past smokers
The reader should be aware that numerous factors influ- (Frye et al., 1990). Interestingly, cessation from smoking
ence olfactory function in “normal” individuals and that results in some improvement of olfactory function over
these factors can significantly alter the ability to smell. time—improvement that is related to the amount of previ-
Among the variables that meaningfully alter the ability to ous smoking and the duration of such cessation.
Psychophysical Measurement of Human Olfactory Function 217

Both reversible and irreversible changes in smell 1995; Köster and De Wijk, 1991; Stuiver, 1958). First, the
function have been observed following exposure to a wide amount of adaptation induced is a function of the duration
variety of environmental agents, including industrial of exposure and the concentration of the adapting stimulus.
chemicals and dusts (see Chapter 27). In the most exten- Second, the subject’s attention level influences the degree of
sive study on this point, the olfactory function of 731 adaptation. Third, the rate and degree of recovery from
workers at a chemical plant that manufactures acrylates adaptation are a function of the magnitude and duration of
and methacrylates was tested (Schwartz et al., 1989). the adapting stimulus. Fourth, cross-adaptation is most
Decrements in odor identification test scores proportionate commonly asymmetrical. For example, while exposure to
to the estimated dose exposure levels of these acrylates odorant A decreases the perceived intensity of odorant B,
were found. Interestingly, individuals who had never exposure to odorant B may not decrease the exposure to
smoked cigarettes but who had been exposed to acrylates odorant A to the same degree. Fifth, the sensitivity to a given
were six times more likely than their nonexposed counter- odorant is typically reduced more by the exposure to that
parts to evidence olfactory decrements. odorant than to any other odorant. Sixth, in rare instances an
Prior experience with odors, particularly that obtained on odorant may have a larger adapting effect on the sensitivity
taste and smell organoleptic panels, clearly influences meas- to another odorant than it does on itself. Seventh, the sensi-
ures of the ability to smell. For example, repeated testing tivity to an odorant that self-adapts strongly is usually also
within the perithreshold odorant concentration range results reduced strongly by other odorants. Eighth, adaptation of
in decreased thresholds or enhancement of signal detection one side of the nose produces adaptation, albeit less, in the
sensitivity measures (Doty et al., 1981; Engen, 1960; Rabin other side of the nose. Ninth, adaptation to complex odor-
and Cain, 1986; Wysocki et al., 1989); practice with feed- ants (i.e., odorants made up of more than one chemical) is
back influences the ability to name odors (Desor and generally less than adaptation to single-component odor-
Beauchamp, 1974; Engen and Ross, 1973). Interestingly, ants. Finally, adaptation to odorants can be relatively rapid.
the hedonic quality of odorants can be influenced by For example, Aronsohn (1886) found that subjects conti-
repeated exposure, making unpleasant odors less unpleasant nuously exposed to the vapors of lemon or orange oil report-
and pleasant odors less pleasant (Cain and Johnson, 1978). ed complete loss of olfactory sensations, on average, in 3
Assuming that adaptation is not the primary basis for this minutes (range: 2.5–11 min). Recovery occurred in about
phenomenon, affective components of odors appear to the same time required to induce the loss.
habituate somewhat independently of odor intensity.

D. Adaptation VI. THE PERCEPTION OF ODORANT

MIXTURES
Exposure to an odorant, if recent and relatively continuous,
can produce a temporary decrease in its ability to be per- As noted above, a number of modern olfactory tests,
ceived, empirically reflected, for example, by heightened including the UPSIT, employ stimuli that, for the most
detection threshold values or decreased intensity ratings part, are complex mixtures of chemicals, mimicking
(for a review, see Cometto-Muñiz and Cain, 1995). Some stimuli encountered in everyday life. More often than not
chemicals produce a decrement in the perception of other such stimuli are perceived as a unitary gestalt and given a
chemicals (termed cross-adaptation). Fortunately, most name associated with the object or source from which they
modern clinical olfactory tests are either little influenced are known to emanate—cinnamon, pizza, cheese, gasoline,
by adaptation or operationally are standardized in such orange, lemon, walnut, etc. (see Livermore and Laing,
a way that any adaptation that occurs is unlikely to mean- 1998b). There is considerable clinical utility in using such
ingfully influence the test results. For example, the UPSIT tests, since many receptor types are activated. This is in
was designed to minimize adaptation by (1) employing contrast to threshold tests employing single odorants, as
largely multicomponent “natural” odorants, (2) requiring they presumably examine the responses of the olfactory
minimal sampling of each odorant, (3) having verbal, system to a smaller subset of receptors. It has been shown
rather than odorous, response alternatives, (4) ordering the that rodents who have sustained damage to 80–90% of
presentation of odorants such that dissimilar odorants fol- their olfactory receptor cells still retain their ability to
low one another (thereby minimizing cross-adaptation), detect some single odorants. Similarly, odor sensitivity is
and (5) allowing adequate time between the smelling of retained unchanged when large lesions have been made in
each odorant item (Doty et al., 1984a). the bulb. Therefore, from at least a theoretical standpoint,
Several general rules have emerged from studies of adap- major changes in the olfactory system can occur and not be
tation that are worthy of note (Cometto-Muñiz and Cain, detectable by the use of some single odorants. In contrast,
218 Doty and Laing

the perception of mixtures invariably involves inhibitory overall intensity of mixtures. Although the vector model
interactions at the bulb (and possibly other olfactory cen- has received widespread attention (e.g., Berglund, 1974;
ters) that occur through complex neural circuitry. Lesions Berglund and Olsson 1993a; Berglund et al., 1976; Cain,
that disrupt the circuitry are likely to alter the characteris- 1975; Cain and Drexler, 1974; Moskowitz and Barbe,
tic suppression effects observed between odors in mixtures. 1977; Laing et al. 1984; Olsson, 1994), after two decades
Rat data indicate that lesions involving much of the bulb can of investigation its best predictions have been for simple
result in the failure to re-learn a mixture analysis task, com- binary mixtures. Other models for predicting the perceived
pared to their successful retention of odor sensitivity and intensity of simple mixtures have been proposed (e.g., the
ability to discriminate between odor qualities (Slotnick et Strongest Component Model, the U Model, and the UPL
al., 1997). Model; see Laffort and Dravnieks, 1982). Such models are
How is it that mixtures of chemicals end up providing modifications of the vector model, but have not been
a largely unitary perceptual gestalt? How much informa- extended to multicomponent mixtures. The most recent
tion, in terms of discriminating individual components, model in this series was the UPL2 model (Laffort et al.,
can humans obtain from complex mixtures? If one odorant 1989) which incorporated the power function that normal-
suppresses the odor of another (as is seen in the case of ly relates perceived odor intensity to concentration. The
deodorants or room fresheners), how does this relate to the ERM model of Schiet and Frijters (1988) was also based
relative concentrations of the odorants within the mixture? on a power function relating these factors and, although
Are there psychophysical rules or laws explaining mixture applied with some success to simple gustatory mixtures,
relationships? These and other questions related to odorant was not an improvement in the models just described for
mixtures are the basis of the remainder of the chapter. olfactory mixtures. As summarized by Cain et al. (1995),
“the principle by which psychophysical information on
A. Effects of Mixing Odorants on Their Perceived single components reflects itself in a model of interaction
Intensity seems to evade the psychophysical models presented here”
(all the above).
Usually when two single compound odorants are mixed Clearly, none of the aforementioned models adequately
together, the perceived intensity of one or both is altered describe the changes in perceived intensity for all pairs of
substantially, the net result being a lowering of the odorants examined, and none have been demonstrated to
intensity of the components. However, on rare occasions reliably predict the intensity of mixtures containing more
enhancement may occur. In early mixture studies, than two odorants. Booth (1995) provides an interesting
Aronsohn (1886) reported that the odor of camphor was critique on the modeling of odor interactions but provides
neutralized by such odors as gasoline, cologne water, and no firm ground for future studies to proceed. Among
oil of juniper, and Nagel (1897) found that counteraction a number of shortcomings, none of the above models have
between two odorants could result in both being rendered been based upon the receptive and neural processes that
almost odorless. Zwaardemaker (1900), the most famous underlie the perception of mixtures, nor has attention been
of early olfactory scientists, confirmed these observations given to choosing odors that have physicochemical fea-
for a number of mixtures using an olfactometer and tures that might provide some basis for antagonistic inter-
demonstrated that the extent of perceptual interactions actions. Furthermore, these models have provided no
between two odorants was more dependent on their con- insight as to the nature of the sensory processes, and none
centrations than on their qualities. Similar results have adequately predicts the intensity of multicomponent mix-
been reported by others, including Moncrieff (1959) and tures. Present evidence suggests that addition or partial
Jones and Woskow (1964), the latter reporting that the per- addition of the perceived intensities of the components of
ceived intensity of a binary mixture, although less than the mixtures occurs with binary and ternary mixtures; beyond
sum of its component intensities, is more than a simple this number of components neural processes limit intensity
average of the two. addition (Berglund et al., 1976; Laing et al., 1994a;
Zwaardemaker (1930) conceptualized the mutual Moskowitz and Barbe, 1977).
weakening of the perceived intensity of a mixture of two The interactions noted above concern suprathreshold
components as follows: “The two sensations can be imag- concentrations of odorants and provide examples of where
ined as two vectors representing two forces counteracting the sense of smell compresses rather than adds intensity
each other in our intellect.” The interaction between two information. In contrast, additivity of neural input appears
odorants was later formalized by Berglund et al. (1973) in to be inherent in mixtures containing only sub-threshold
a mathematical model that incorporated the application of quantities of odorants (Laska and Hudson, 1991; Laska
vector addition to odor mixtures for the prediction of the et al., 1990). Indeed, in mixtures with only three odorants,
Psychophysical Measurement of Human Olfactory Function 219

the magnitude of the addition was noted by Laska et al. to C. Identification of Components in Odorant
be substantial and to often exceed that obtained from sim- Mixtures
ple summation. Patterson et al. (1993) reported instances
of near-true additivity of subthreshold components and Prior to studies of the abilities of humans to analyze mix-
suggested that additivity may function to enhance sensitiv- tures, informal information from perfumers and flavorists
ity to the typically complex (and often subthreshold) odor suggested that between 5 and 30 components may be iden-
stimuli encountered in everyday life. They noted that the tified in mixtures (D. G. Laing, unpublished data). Over
number of chemicals activating the system could be as the past decade it has become clear that these numbers are
important as the strength of any one of the odorants, pro- an overestimate, as most individuals, including perfumers,
viding a type of “biological economy” of the input. are only able to identify up to 3 or, rarely, 4 components.
An early hint that only a small number of odorants can be
B. Discrimination of Components in Odorant identified in mixtures was apparent in the report by
Mixtures Berglund (1974), who suggested from studies of the add-
ition of the perceived intensities of components, that an
Since, as mentioned earlier, odors are commonly encoun- analytic or additive process occurred up to 3 components,
tered as mixtures in our environment, an important char- whereupon above this number an interactive process pre-
acteristic of the human sense of smell is to discriminate dominated. The latter was apparent as an asymptote in the
differences between mixtures. Discriminating the odors total perceived intensity of a mixture, with little change
of fresh and “off ” milk, ripe and overripe fruit, cork taint occurring as the number of components increased. In
in wine, and various perfumes are examples. In the area accord with this notion, Moskowitz and Barbe (1977)
of pollution control, changes in the complex odor of found that in some instances the overall intensity of 5 com-
sewage provide engineers with an insight as to the part of ponent mixtures was less than that of mixtures with fewer
the treatment process that is not functioning properly; components.
sulfides emanate if the anaerobic process is malfunction- In perhaps the first formal scientific studies on this topic,
ing, and sour, rancid, and acid odors appear if the sludge Laing and Francis (1989) and Livermore and Laing (1996)
treatment is inappropriate. In studies with binary mix- reported that training and experience did not increase the
tures, Rabin (1988) and Rabin and Cain (1989) showed number of components identified with subjects that had
that humans are particularly sensitive to the presence of been trained for a few minutes, 3 weeks, or who were per-
small amounts of odorants that are not normally found in fumers and flavorists, the maximum still being 3–4.
a stimulus. They reported that (1) high familiarity with Varying the task or the odorants resulted in no improvement
a major component and the ability to label it consistently in the number identified; thus, in another study a selective
facilitates the detection of a minor component, (2) the attention procedure was not more efficient than a procedure
minor component is not detected as readily if it is that required subjects to identify as many components as
unfamiliar, and (3) unpleasant stimuli are more possible during an ad lib sampling method (Laing and
detectable than pleasant ones, although the effect was not Glemarec, 1992). Futhermore, this maxima was not altered
as large as the effect of familiarity. Experience, therefore, if the odorants used were those classified by perfumers as
and to a lesser extent pleasantness, improves discrimina- “poor blenders,” i.e., odorants that they used to “stand out”
tion between two single odorants or two mixtures, sug- in mixtures (Livermore and Laing, 1998a). Schiet and
gesting that similar cognitive processes operate with the Frijters (1988), using another approach to this problem,
two types of stimuli. reported that subjects invariably underestimate the number
Although the Rabin studies suggested that humans are of components in mixtures containing up to 4 components.
very sensitive to small changes in an olfactory stimulus, A similar result was obtained by Jellinek and Köster
Laska and Hudson (1992) reported that relatively large (1979), whose subjects found the odor of single chemicals
changes in the composition of mixtures are sometimes to be as complex as that of mixtures.
required for discrimination to occur. Thus, discrimination Clearly, the data of the aforementioned studies indicate
of 3-, 6-, or 12-component mixtures from the same mix- that there is a significant limitation in the olfactory system
tures minus 1 component produced error levels of in the processing of information from more than about
20–40%, with the level depending on the type of odorant 3 odorants. Mixtures of complex odors tend to behave like
that was removed. Accordingly, the dependence on the mixtures of single odorants, with a maximum of about
type of odor removed precluded defining a limit in the 3 being identified in stimuli containing up to 8 complex
ability of humans to discriminate between two complex odors. The possibility that the entry of hundreds, perhaps
mixtures. thousands, of odorants into the nose would produce a
220 Doty and Laing

nonidentifiable smell sensation has not yet eventuated being identified in mixtures containing double this number
(Livermore and Laing, 1998b). of components. Their psychophysical study showed that
1 and zero components were identified in 12- and 15-com-
ponent stimuli, respectively. The fact that the 15-compo-
D. Mechanisms Involved in Odor Mixture
nent stimulus had an odor, albeit not one that could be
Perception
associated with any of the components or an object or
source, indicates that neural input from some or all of the
The limited ability of humans to discriminate and identify
arrays characterizing the components was registered.
odorants in mixtures is likely due to a number of
In light of such observations, it is interesting to note that,
mechanisms. Changes in spatial processing arising from
in the rat, neural images of complex olfactory stimuli,
competition for receptor sites and cells at the periphery,
including rat nest odors comprised of volatiles from urine,
and inhibition in the bulb and at other olfactory centers,
feces, and bodies (Stewart et al. 1979), have shown that the
would be expected to reduce the information within the
number of activated glomeruli is similar to that found with
activated receptor cell arrays, making it difficult to recog-
a simple single odorant such as limonene (Bell et al., 1987).
nize the patterns of activation due to different odorants.
Therefore, spatial processing of single and complex odor-
Since temporal processing favors the first processed odor-
ants involves both peripheral and bulbar interactions that
ant, the initial odorant has the opportunity to act as an
reduce and simplify identification. Accordingly, the olfac-
antagonist towards other odorants at the periphery and to
tory system uses spatial coding to analyze and identify sin-
inhibit neural activity arising from other odorants in the
gle odorants when presented alone and in simple mixtures
bulb. However, if the delivery of different odorants in
and simplifies identification of complex mixtures by com-
ternary or more complex mixtures to the nose is less than
bining the remaining parts of the arrays into a single char-
the time for processing two odorants in working memory,
acteristic array that is associated with the object or source
the latter becomes the ultimate limiting factor as regards
of the stimulus. This interpretation is in agreement with the
the number of odorants identified. These mechanisms are
finding of Jellinek and Köster (1979) that single odors are
discussed in detail below.
perceived to be as complex as those of mixtures.
But there is another aspect to spatial processing. An
1. Spatial Processing
intriguing feature of single odorants and odor mixtures is
As noted at the beginning of this chapter and elsewhere in that they can be characterized by several qualities or
this volume, a given odorant activates unique arrays of “notes” (Laing and Willcox, 1983; Moskowitz and Barbe,
receptor cells in the nose (Kauer, 1991; Mackay-Sim et al., 1977). Hexenal, for example, is described as having
1982), which, in turn, are reflected by patterns of activa- “green” and “fatty” qualities, and ethyl butyrate as “sweet”
tion of glomeruli and mitral/tufted cells in the olfactory and “fruity.” However, when single odorants are compo-
bulb. Different odors produce different arrays that repre- nents of mixtures but cannot be identified, often one or
sent the spatial codes of identification. However, when more of their qualities can be discerned. Recently, Jinks
a mixture of two odorants is sensed and the perception of and Laing (2001) investigated the qualities of binary,
one or both is suppressed to some degree, the arrays repre- ternary, and quaternary mixtures of four dissimilar odor-
senting the two stimuli in the bulb show a reduction in the ants to determine the information about odor quality that
number of glomeruli that are activated (Bell et al., 1987; needs to be retained for identification of the odorant. The
Joerges et al., 1997). If the suppression of one of the odor- data indicated that failure to identify an odorant could
ants is such that it cannot be perceived, little of the normal occur with loss of some but not all of the qualities.
array of activated glomeruli is seen (Bell et al., 1987). The However, failure could also occur when the major qualities
suppression may be due to fewer receptor cells being acti- were present but the ratios of their perceived intensities
vated (Ache et al., 1988; Kurahashi et al., 1994; Simon and were substantially altered. This suggested that a different
Derby, 1995) because of competition by the odorants for smell could be produced using the same qualities but in
the same receptor sites, resulting in less input to the bulb. different ratios. Identification, therefore, was affected by
Suppression can also be caused by lateral inhibition the type and/or the perceived intensity of the qualities of an
between glomeruli or mitral cells in the bulb (Pinching and odorant. These results were interpreted in terms of
Powell, 1971; Shepherd and Greer, 1990; White, 1979). a Configurational Hypothesis of Olfaction, in analogy with
The loss of identity of up to 5 odorants in 8-component the Configurational Hypothesis of Facial Recognition
mixtures (Livermore and Laing, 1998a) prompted Jinks (Enns and Shore, 1997; Rakover and Teucher, 1997). In
and Laing (1999) to propose that the competition and brief, in the case of a face, identification of a person not
inhibition between odorants could result in no odorant only requires certain features to be present in a drawing or
Psychophysical Measurement of Human Olfactory Function 221

photograph, but these features must be in the correct Nevertheless, several qualities can usually be discerned in
proportion to each other. Similarly, identification of an complex aromas, and these are likely to be those that
odorant or a complex mixture requires some of the charac- remain from individual odorants in the mixture which are
teristic qualities in the correct proportions to be perceived. insufficient to identify the latter but contribute to the over-
But what is the neural basis of a quality or “odor note,” all aroma of the complex mixture.
and how is it represented in the spatial code? Thanks to
advances in molecular biological studies of the chemo-
2. Temporal Processing
receptive process, an insight into this problem is possible.
For example, as noted at the beginning of this chapter, it During the 1980s, Getchell et al. (1984) reported that
is commonly accepted that each human receptor cell has odorants can differ by hundreds of milliseconds in the
only one type of receptor (Rawson et al., 1997) and that times they take to activate receptor cells, while Kuznicki
there are ~1000 receptor types, as indicated by the and Turner (1986) showed that humans require different
number of receptor genes (Buck and Axel, 1991). reaction times to recognize the four common tastants.
Stimulation of receptor cells by a single odorant will These findings prompted Laing (1987) to propose that if
result in a variety of cells being activated in accordance the time differences between the activating times of
with the degree or “ease of fit” of the odorant to each odorants at the periphery were maintained as the neural
receptor site type. If the fit is predominantly to two or message traveled through the bulb and other olfactory pro-
three receptor types, they will be the main inputs to the cessing centers in the brain that dealt with memory and
array of glomeruli and mitral/tufted cells activated in the identification, then a “fast” odorant would have a number
bulb. However, the conformations adopted by an odorant of advantages if presented in a mixture with a “slow” odor-
to fit the two to three receptor types will be dictated by the ant. For example, the faster odorant may be more success-
structural features of the odorant and receptor molecule. ful in competition for receptor sites and cells, and being
In one conformation a molecule may be aligned within the first to activate the bulb, it could trigger lateral inhibi-
a receptor site according to its length and functional tion between glomeruli or between mitral cells to further
group, e.g., the 17 receptor for octanal (Araneda et al., reduce neural input from the slower odorant. Accordingly,
2000); in another it may sense a structural feature com- it was predicted that the faster odorant would be the first
mon to a number of odorants, e.g., an 8-carbon chain con- odorant identified in a mixture, the slower odorant would
taining a terminal carbonyl group common to aliphatic incur the greatest suppression of intensity, and the number
aldehydes, acids, esters, and ketones (Imamura et al., of cells and glomeruli in spatial arrays activated by the lat-
1992). Since the overall odors of these latter aliphatic car- ter odorant would be reduced.
bonyl substances are easily discriminable (Laska et al., To investigate the first two of the above predictions,
2000), each odorant must require at least two receptor Laing et al. (1994b) used a specially designed computer-
types to be occupied for this to occur. Accordingly, it is controlled olfactometer, which allowed odorants to be
tempting to suggest that activation of the cells with the delivered together in a mixture or in series separated by
common receptor for these odorants results in an odor intervals as small as 50 ms. By asking subjects which of
quality common to each odorant, while activation of the two odorants was perceived first during a trial and vary-
cells unique to each odorant produces a quality unique to ing the time between delivery of both odorants from 100
each odorant. In addition, the spatial map of each odorant to 600 ms, the processing time difference between them
should show glomeruli or mitral/tufted cells that are acti- was established as that which produced a chance
vated by all four odorants and others that only one of the response, i.e., 50% for the forced-choice yes/no task. The
odorants will activate. From the limited data available, it magnitude of the differences varied from zero to more
is suggested that the conformations an odorant can adopt than a second and was dependent on both the quality and
in different types of receptors defines the important struc- perceived intensity of the odorants, with the latter being
tural features that provide the qualities perceived. This more important. Perceived intensity was also reduced
interpretation suggests that the spatial code for an odorant more for the slower odorant. With both predictions
contains information about molecular structure and odor upheld, the existence of temporal processing and its
qualities. In contrast, the spatial map of complex mixtures implications for mixture perception were demonstrated.
such as chocolate aroma, where none of dozens of odor- A later study (Jinks and Laing, 1999b) confirmed that
ants can be identified, will be composed of input from knowledge of processing time differences allowed pre-
receptor cells representing features of many odorants, and dictions of which odor would be perceived first in other
it may be the location and magnitude of the input to the mixtures. Thus, they showed that when odor A was per-
bulb rather than molecular features that define its identity. ceived before B and B was perceived before C, that A was
222 Doty and Laing

was needed before the usually faster odorant was per-

ceived first (Fig. 8). With the other set, subjects recorded
chance responses even with the 900 ms delay. The mean
responses of subjects, when asked to identify the odorants
in the mixture or delay conditions, showed that this was at
chance level. The result is in agreement with the earlier
studies of Laing and colleagues, who found that few sub-
jects could identify all the components of ternary mixtures.
Overall, the results with ternary mixtures indicated that
a mechanism related to the speed of information retrieval
about the identity or temporal order of the components was
the cause. The most likely candidate appears to be the
inability of olfactory working memory to process the infor-
mation about the identity and order of the first two com-
ponents before neural input from the third began to be
processed. Although it is not fully understood, working
memory is defined as the “system responsible for the tem-
porary storage and manipulation of information, forming
an important link between perception and controlled
action” (Baddeley, 1998). The process of identifying an
Figure 7 Regression lines representing the proportion of trials an
odor in a binary mixture was perceived “first” when presented with
odor within working memory is likely to involve several
a time advantage, disadvantage, or as true mixture (0 ms interval). steps: encoding of the odor by neurons, recalling of the
Arrows and times in boxes indicate when both odorants were per-
ceived first on 50% of trials. (A) Stimulus of coniferan and triethyl-
amine with coniferan being perceived 538 ms before triethylamine;
(B) a stimulus of carvone/triethylamine with carvone perceived first
1739 ms before triethylamine; (C) a stimulus of carvone/coniferan
with carvone perceived first 251 ms before coniferan.

perceived before C, demonstrating that transitivity had

occurred (Fig. 7). However, investigation of temporal
processing in ternary mixtures revealed a substantial
limitation in the ability of humans to indicate which odor
is perceived first and the existence of a third mechanism
that affects perception of components in odor mixtures
(Jinks and Laing, 1999b). Temporal processing of ternary
mixtures and the third mechanism, which is postulated to
involve olfactory working memory, are discussed below.

3. The Role of Memory

The perception of the order of processing odorants in
ternary mixtures, however, has proved to be very difficult
(Jinks and Laing, 1999b). Initial experiments indicated
that subjects recorded chance level responses when asked Figure 8 Proportion of trials (numbers above bars) in which
subjects selected an odor “coming first” in binary and ternary
to indicate which odorant was perceived first or last. To
mixtures and mixtures where the presentation of triethylamine
investigate whether the chance results were due to a limi- was delayed. Conditions: 1, binary mixture of carvone/coniferan;
tation in the capacity of olfactory working memory to 2, ternary mixture of carvone/coniferan/triethylamine; 3,4, and 5,
process both order and identity of the odorants, presenta- ternary mixture with the presentation of triethylamine delayed by
tion of the third odorant was delayed by 300, 600, and 300, 600, and 900 ms, respectively. Open and shaded bars indicate
900 ms. With one of the two sets of three odorants studied, that the means were significantly/not significantly different from
the results indicated that a delay of between 600 and 900 ms 0.5 (chance), respectively.
Psychophysical Measurement of Human Olfactory Function 223

coded representation of the odor from long term memory, REFERENCES

comparison of the two representations, and the judging and
responding to the representations. This type of process has Abraham, A., and Matha, K. V. (1983). The effect of right tem-
been proposed for visual information (Eskandar et al., poral lobe lesions on matching of smells. Neuropsychologia
1992). Indeed, the interference of a third odorant with the 21:277–281.
perception of others is reminiscent of that reported for Ache, B. W., Gleeson, R. A., and Thompson, H. A. (1988).
visual spatial memory, where it was proposed that an Mechanisms for mixture suppression in olfactory receptors of
irrelevant visual stimulus may have obligatory access to the spiny lobster. Chem. Senses 13:425–434.
Amoore, J. E., and Ollman, B. G. (1983). Practical test kits for
a visual store and interfere with the storage and processing
quantitatively evaluating the sense of smell. Rhinology
of other visual spatial information in working memory
21:49–54.
(Toms et al., 1994). Limitations in the capacity of olfactory Anderson, N. H. (1970). Functional measurement and psy-
working memory to process more than two odorants with- chophysical judgment. Psychol. Rev. 77:153–170.
in 600 –900 ms appears to be a major factor limiting the Araneda, R. C., Kini, A. D., and Firestein, S. (2000). The mole-
discrimination and identification of odorants in multicom- cular receptive range of an odorant receptor. Nature Neurosci.
ponent mixtures. Such a finding has implications for the 3:1248–1255.
perception of odorants released during an eating episode Aronsohn, E. (1886). Experimentalle Untersuchungen zur
where many can be released within the processing time Physiologie des Geruchs. Archiv. Physiol. Leipzig. 321–357.
differences cited here, but only a few may be identified. Baddeley, A. (1998). Working memory. C. R. Acad. Sci. Paris,
Controlled release of odorants from different food media Life Sci. 321:167–173.
could, however, allow products to be developed with high Baird, J. C., Lewis, C., and Romer, D. (1970). Relative frequen-
flavor impact. cies of numerical responses in ratio estimation. Percept.
Psychophys. 8:358–362.
Bell, G. A., Laing, D. G., and Panhuber, H. (1987). Odor mixture
VII. CONCLUSIONS suppression: evidence for a peripheral mechanism in human
and rat. Brain Res. 426:8–18.
The present chapter has provided an up-to-date review of Berglund, B. (1974). Quantitative and qualitative analysis of
the psychophysical means for testing the human sense of industrial odors with human observers. Ann. NY Acad. Sci.
237:35–51.
smell and has examined how the human olfactory system
Berglund, B., and Olsson, M. J. (1993a). Odor-intensity interac-
likely integrates information from complex arrays of odor-
tion in binary mixtures. J. Exp. Psychol. Human Percept.
ant chemicals which, individually, would seem to produce Perform. 19:302–314.
conflicting odorous sensations. It is of interest that rela- Berglund, B., and Olsson, M. J. (1993b). Odor-intensity interac-
tively high correlations exist among the scores derived tion in binary and ternary mixtures. Percept. Psychophys.
from nominally distinct olfactory tests, regardless of 53:475–482.
whether they are based upon single- or multicomponent Berglund, B., Berglund, U., Ekman, G., and Engen, T. (1971).
stimuli. Test reliability has been shown to be largely Individual psychophysical functions for 28 odorants. Percept.
a result of test length, irrespective of the nature of the stim- Psychophys. 9:379–384.
uli included in the tests. To what extent tests employing Berglund, B., Berglund, U., Lindvall, T., and Svensson, L. T.
multicomponent odors are superior to ones employing sin- (1973). A quantitative principle of perceived intensity sum-
gle odorants is an empirical issue, although it would seem mation in odor mixtures. J. Exp. Psychol. 100:29–38.
that by sampling more elements of the system, a test Berglund, B., Berglund, U., and Lindvall, T. (1976).
should be more sensitive. Continued efforts to refine the Psychological processing of odor mixtures. Psychol. Rev.
83:432–441.
procedural elements of olfactory tests should help in the
Berglund, B., Berglund, U., Johansson, I., and Lindvall, T.
development of test batteries sensitive to wider ranges of
(1984). Mobile laboratory for sensory air-industrial environ-
olfactory deficits than those that are currently available. ments. In Indoor Air, Vol. 3, B. Berglund, T. Lindvall, and
J. Sumndeel (Eds.). Swedish Council for Building Research,
Stockholm, pp. 467–472.
ACKNOWLEDGMENTS Blackwell, H. R. (1953). Psychophysical. thresholds: experimen-
tal studies of methods of measurement. Bull. Engin. Res. Inst.
This paper was supported, in part, by Grants PO1 DC No. 36. University of Michigan Press, Ann Arbor.
00161, RO1 DC 04278, RO1 DC 02974, RO1 AG 27496, Booth, D. A. (1995). Cognitive processes in odorant mixture
and RO1 AG 08148 from the National Institutes of Health, assessment. Chem. Senses 20:639–643.
Bethesda, MD, to RLD, and an Australian Research Council Brown, J. (1974). Recognition assessed by rating and ranking.
Large Grant to DGL. Br. J. Psychol. 65:13–22.
224 Doty and Laing

Brown, K. S., Maclean, C. M., and Robinette, R. R. (1968). The Davidson, T. M., and Murphy, C. (1997). Rapid clinical evalua-
distribution of the sensitivity to chemical odors in man. Hum. tion of anosmia. The alcohol sniff test. Arch. Otolaryngol.
Biol. 40:456–472. Head Neck Surg. 123:591–594.
Buck, L., and Axel, R. (1991). A novel multigene family may Deems, D. A., Doty, R. L., Settle, R. G., Moore-Gillon,
encode odorant receptors: a molecular basis for odor recogni- V., Shaman, P., Mester, A. F., Kimmelman, C. P., Brightman, V.
tion. Cell 65:175–187. J., and Snow, J. B. Jr. (1991). Smell and taste disorders, a study
Burdach, K. J., and Doty R. L. (1987). The effects of mouth of 750 patients from the University of Pennsylvania Smell and
movements, swallowing, and spitting on retronasal odor per- Taste Center. Arch. Otolaryngol. Head Neck Surg.
ception. Physiol. Behav. 41:353–356. 117:519–528.
Cain, W. S. (1975). Odor intensity, mixtures and masking. Chem. Desor, J. A., and Beauchamp, G. K. (1974). The human capacity
Senses Flav. 1:339–352. to transmit olfactory information. Percept. Psychophys.
Cain, W. S. (1982a). Sumner’s “on testing the sense of smell” 16:551–556.
revisited. Yale J. Biol. Med. 55:515–519. Doty, R. L. (1975). An examination of relationships between the
Cain, W. S. (1982b). Odor identification by males and females: pleasantness, intensity and concentration of 10 odorous
predictions vs. performance. Chem. Senses 7:129–142. stimuli. Percept. Psychophys. 17:492–496.
Cain, W. S., and Drexler, M. (1974). Scope and evaluation of Doty, R. L. (1979). A review of olfactory dysfunctions in man.
odor counteraction and masking. Ann. NY Acad. Sci. Am. J. Otolaryngol. 1:57–79.
237:427–439. Doty, R. L. (1986). Gender and endocrine-related influences
Cain, W. S., and Gent, J. F. (1991). Olfactory sensitivity: reliabil- upon olfactory sensitivity. In Clinical Measurement of Taste
ity, generality, and association with aging. J. Exp. Psychol. and Smell, H. L. Meiselman and R.S. Rivlin (Eds.).
Hum. Percept. Perform. 17:382–391. Macmillan, New York, pp. 377–413.
Cain, W. S., and Johnson, F., Jr. (1978). Lability of odor pleas- Doty, R. L. (1990). Aging and age-related neurological disease:
antness: influence of mere exposure. Perception 7:459–465. olfaction. In Handbook of Neuropsychology, F. Goller and
Cain, W. S., and Rabin, R. D. (1989). Comparability of two tests J. Graftnan (Eds.). Elsevier, Amsterdam, pp. 211–226.
of olfactory functioning. Chem. Senses 14:479–485. Doty, R. L. (1991a). Olfactory system. In Smell and Taste in
Cain, W. S., Gent, J., Catalanotto, F. A., and Goodspeed, R. B. Health and Disease, T. V. Getchell, R. L. Doty, L. M.
(1983). Clinical evaluation of olfaction. Am. J. Otolaryngol. Bartoshuk, and J. B. Snow, Jr. (Eds.). Raven Press, New York,
4:252–256. pp. 175–203.
Cain, W. S., Gent, J. P., Goodspeed, R. B., and Leonard, G. Doty, R. L. (1991b). Olfactory dysfunction in neurodegenerative
(1988). Evaluation of olfactory dysfunction in the Connecticut disorders. In Smell and Taste in Health and Disease, T. V.
Chemosensory Clinical Research Center. Laryngoscope 98: Getchell, R. L. Doty, L. M. Bartoshuk, and J. B. Snow, Jr.
83–88. (Eds.). Raven Press, New York, pp. 735–751.
Cain, W. S. Schiet, F. T., Olsson, M. J., and de Wijk, R. A. (1995). Doty, R. L. (1995). The Smell Identification TestTM Administration
Comparison of models of odor interaction. Chem. Senses Manual, 3rd ed. Sensonics, Inc., Haddon Hts., NJ.
20:625–637. Doty, R. L. (2000). The Smell Threshold TestTM Administration
Campbell, 1. M., and Gregson, R. A. M. (1972). Olfactory short Manual. Sensonics, Inc., Haddon Hts., NJ.
term memory in normal, schizophrenic, and brain damaged Doty, R. L. (2001). Olfaction. Ann. Rev. Psychol. 52:423–452.
cases. Aust. J. Psychol. 24:179–185. Doty, R. L., and Snow, J. B., Jr. (1988). Age-related changes in
Carrasco, M., and Ridout J. B. (1993). Olfactory perception and olfactory function. In Molecular Neurobiology of the
olfactory imagery: a multidimensional scaling analysis. J. Olfactory System, R. L. Margolis and T. V. Getchell (Eds.).
Exp. Psychol. Hum. Percept. Perform. 19:287–301. Plenum Press, New York, pp. 355–374.
Chalke, H. D., Dewhurst, J. R., and Ward, C. W. (1958). Loss of Doty, R. L., Ford, M., Preti, G., and Huggins, G. (1975). Human
sense of smell in old people. Public Health 72:223–230. vaginal odors change in pleasantness and intensity during the
Cheeseman, G. H., and Kirkby, H. M. (1959). An air dilution menstrual cycle. Science 190:1316 –1318.
olfactometer suitable for group threshold measurements. Doty, R. L., Kligman, A., Leyden, J., and Orndorff, M. M.
Quart. J. Exp. Psychol. 11:115–123. (1978). Communication of gender from human axillary odors:
Cheeseman, G. H., and Townsend, M. J. (1956). Further experi- relationship to perceived intensity and hedonicity. Behav.
ments on the olfactory thresholds of pure chemical substances, Biol. 23:373–380.
using the “sniff-bottle method.” Q. J. Exp. Psychol. 8:8–14. Doty, R. L., Snyder, P., Huggins, G., and Lowry, L. D. (1981).
Cometto-Muñiz, J. E, and Cain, W. S. (1995). Olfactory adapta- Endocrine, cardiovascular, and psychological correlates of
tion. In Handbook of Olfaction and Gustation, R. L. Doty olfactory sensitivity changes during the human menstrual
(Ed.). Marcel Dekker, New york, pp. 257–281. cycle. J. Comp. Physiol. Psychol. 95:45–60.
Cornsweet, T. N. (1962). The staircase-inethod in psychophysics. Doty, R. L., Ram, C. A., Green, P., and Yankell, S. (1982).
Am. J. Psychol. 75:485–491. Communication of gender from breath odors: relationship to
Corso, J. F. (1981). Aging Sensory Systems and Perception. perceived intensity and pleasantness. Horm. Behav. 16:
Praeger, New York. 13–22.
Psychophysical Measurement of Human Olfactory Function 225

Doty, R. L., Shaman, P., and Dann, M. (1984a). Development of Enns, J. T., and Shore, D. I. (1997). Separate influences of orien-
the University of Pennsylvania Smell Identification Test: a tation and lighting in the invertedface effect. Percept.
standardized microencapsulated test of olfactory function. Psychophys. 59:23–31.
Physiol. Behav. 32:489–502. Eskandar, E. N., Optican, L. M., and Richmond, B. R. (1992).
Doty, R. L., Shaman, P., Applebaum, S. L .,Giberson, R., Role of inferior temporal neurons in visual memory: II.
Sikorsky, L., and Rosenberg, L. (1984b). Smell identification Multiplying temporal waveforms related to vision and
ability: changes with age. Science 226:1441–1443. memory. J. Neurophysiol. 68:1296–1306.
Doty, R. L., Newhouse, M. G., and Azzalina, J. D. (1985). Eyman, R. K., Kim, P. J., and Call, T. (1975). Judgment error in
Internal consistency and short-term test-retest reliability of category vs magnitude scales. Percept. Motor Skills 40:
the University of Pennsylvania Smell Identification Test. 415–423.
Chem. Senses 10:294–300. Fechner, G. T. (1860). Elemente der Psychophysik. Breitkopf and
Doty, R. L., Gregor, T., and Settle, R. G. (1986). Influences of Härtel, Leipzig.
intertrial interval and sniff bottle volume on the phenyl ethyl Frey, P. W., and Colliver, J. A. (1973). Sensitivity and responsiv-
alcohol olfactory detection threshold. Chem. Senses ity measures for discrimination learning. Learn. Motiv.
11:259–264. 4:327–342.
Doty, R. L., Reyes, P., and Gregor, T. (1987). Presence of both Frijters, J. E. R., Kooistra, A., and Vereijken, P. F. G. (1980).
odor identification and detection deficits in Alzheimer’s dis- Tables of d for the triangular method and the 3-AFC sig-
ease. Brain Res. Bull. 18:597–600. nal detection performance. Percept. Psychophys. 27:
Doty, R. L., Deems, D. A., Frye, R., Pelberg, R., and Shapiro, A. 176–178.
(1988). Olfactory sensitivity, nasal resistance, and autonomic Frye, R. E., Schwartz, B., and Doty, R. L. (1990). Dose-related
function in the multiple chemical sensitivities (MCS) syn- effects of cigarette smoking on olfactory function. JAMA
drome. Arch. Otolaryngol. Head Neck Surg. 114:1422–1427. 263:1233–1236.
Doty, R. L., Agrawal, U., and Frye, R. (1989). Evaluation of the Furukawa, M., Kamide, M., Miwa, T., and Umeda, R. (1988).
internal consistency reliability of the fractionated and whole Importance of unilateral examination in olfactometry. Auris
University of Pennsylvania Smell Identification Test. Percept. Nasus Larynx (Tokyo) 15:113–116.
Psychophys. 45:381–384. Gescheider, G. A. (1976). Psychophysics: Method and Theory.
Doty, R. L., Stem, M. B., Pfeiffer, C., Gollomp, S. M., and Lawrence Erlbaum Associates, Hillsdale, NJ.
Hurtig, H. 1. (1992). Bilateral olfactory dysfunction in early Gescheider, G. A. (1988). Psychophysical scaling. Ann. Rev.
stage treated and untreated idiopathic Parkinson’s disease. Psychol. 39:169–200.
J. Neurol. Neurosurg. Psychiatry 55:138–142. Getchell, T. V., Margolis, F. L., and Getchell, M. L. (1984).
Doty, R. L., Smith, R., McKeown, D., and Raj, J. (1994). Tests of Perireceptor and receptor events in vertebrate olfaction. Prog.
human olfactory function: Principal components analysis sug- Neurobiol. 23:317–345.
gests that most measure a common source of variance. Green, B. G., Dalton, P., Cowart, B., Shaffer, G., Rankin, K., and
Percept. Psychophys. 56:701–707. Higgins, J. (1996). Evaluating the ‘Labeled Magnitude Scale’
Doty, R. L., McKeown, D., Lee W. W., and Shaman, P. (1995). for measuring sensations of taste and smell. Chem. Senses
Test-retest reliability of 10 olfactory tests. Chem. Senses 21:323–334.
20:645–656. Green, D. M., and Swets, J. A. (1966). Signal Detection Theory
Drake, B., Johansson, B., von Sydow, D., and Döving, K. B. and Psychophysics. Wiley, New York.
(1969). Quantitative psychophysical and electrophysiological Gregson, R. A. M., and Smith, D. A. R. (1981). The clinical
data on some odorous compounds. Scand. J. Psychol. assessment of olfaction: differential diagnoses including
10:89–96. Kallmann’s syndrome. J. Psychosomat. Res. 25:165–174.
Eichenbaum, H., Morton, T. H., Potter, H., and Corkin, C. (1983). Gregson, R. A. M., Free, M. L., and Abbott, M. W. (1981).
Selective olfactory deficits in case H. M. Brain 106:459–472. Olfaction in Korsakoffs, alcoholics and normals. Br. J. Clin.
Ekman, G., and Sjöberg, L. (1965). Scaling. Ann. Rev. Psychol. Psychol. 20:3–10.
16:451– 474. Grier, J. B. (1971). Nonparametric indexes for sensitivity and
Elliot, P. B. (1964). Tables of d. In: Signal Detection and bias: computing formulas. Psychol. Bull. 75:424–429.
Recognition by Human Observers, Swets, J. A. (Ed.). Wiley, Griffith, I. P. (1976). Abnormalities of smell and taste. Practitioner
New York, 1964. 217:907–913.
Elsberg, C. A., and Levy, 1. (1935). The sense of smell: I. A new Guadagni, D. G., Buttery, R. G., and Okano, S. (1963). Odour
and simple method of quantitative olfactometry. Bull. Neurol. thresholds of some organic compounds associated with food
Inst. NY 4:4 –19. flavours. J. Sci. Food Agr. 14:761–765.
Engen, T. (1960). Effect of practice and instruction on olfactory Guilford, J. P. (1954). Psychometric Methods. McGraw-Hill,
thresholds. Percept. Motor Skills 10:195–198. New York.
Engen, T., and Ross, B. M. (1973). Long-term memory of odors Helson, H. (1964). Adaptation-Level Theory: An Experimental
with and without verbal descriptions. J. Exp. Psychol. 100: and Systematic Approach to Behavior. Harper and Row, New
221–227. York.
226 Doty and Laing

Heywood, P. G., and Costanzo, R. M. (1986). Identifying nor- Kurtz, D. B., White, T. L., Hornung, D. E. and Belknap, E. (1999).
mosmics: a comparison of two populations. Am. J. Otolaryngol. What a tangled web we weave: discriminating between malin-
7:194–199. gering and anosmia. Chem. Senses 24: 697–700.
Hodos, W. (1970) Nonparametric index of response bias for use Kuznicki, J. T., and Turner, L. S. (1986). Reaction time in the
in detection and recognition experiments. Psychol. Bull. perceptual processing of taste quality. Chem. Senses
74:351–354. 11:183–201.
Hummel, T., Sekinger, B., Wolf, S. R., Pauli, E., and Kobal, G. Laffort, P. (1989). Models for describing intensity interactions in
(1997). ‘Sniffin’ sticks’: olfactory performance assessed by odor mixtures: a reappraisal. In Perception of Complex Smells
the combined testing of odor identification, odor discrimina- and Tastes, D. G. Laing, W. S. Cain, R. L. McBride, and B. W.
tion and olfactory threshold. Chem. Senses 22:39–52. Ache (Eds.). Academic Press, Sydney, pp. 205–223.
Imamura, K., Mataga, N., and Mori, K. (1992). Coding of odor Laffort, P., and Dravnieks, A. (1982). Several models of
molecules by mitral/tufted cells in rabbit olfactory bulb. suprathreshold quantitative olfactory interaction in humans
I. Aliphatic compounds. J . Neurophysiol. 68:1986–2002. applied to binary, ternary and quaternary mixtures. Chem.
Jehl, C., Royet, J. -P., and Holley, A. (1997). Role of verbal Senses 7:153–174.
encoding in short-and long-term odor recognition. Percept. Laing, D. G. (1987). Coding of chemosensory stimulus mixtures.
Psychophys. 59:100–110. Ann. NY Acad. Sci. 510:61–66.
Jellinek, J. S., and Koster, E. P. (1979). Perceived fragrance com- Laing, D. G., and Francis, G. W. (1989). The capacity of humans
plexity and its relation to familiarity and pleasantness. J. Soc. to identify odors in mixtures. Physiol. Behav. 46:809–814.
Cosmet. Chem. 30:253–262. Laing, D. G., and Glemarec, A. (1992). Selective attention and
Jinks, A., and Laing, D. G. (1999a). A limit in the processing of the perceptual analysis of odor mixtures. Physiol. Behav.
components in odour mixtures. Perception 28:395–404. 52:1047–1053.
Jinks, A., and Laing, D. G. (1999b). Temporal processing reveals Laing, D. G., and Willcox, M. E. (1983). Perception of compo-
a mechanism for limiting the capacity of humans to analyze nents in binary odor mixtures. Chem. Senses 7:249–264.
odor mixtures. Cognitive Brain Res. 8:311–325. Laing, D. G., Panhuber, H., Willcox, M. E., and Pittman, E. A.
Jinks, A., and Laing, D. G. (2001). The analysis of odor mixtures (1984). Quality and intensity of binary odor mixtures.
by humans: Evidence for a configurational process. Physiol. Physiol. Behav. 33:309–319.
Behav. 72:51–63. Laing, D. G., Eddy, A., and Best, D. J. (1994a). Perceptual char-
Joerges, J., Kuttner, A., Galizia, C. G., and Menzel, R. (1997). acteristics of binary, trinary and quaternary odor mixtures
Representations of odours and odour mixtures visualized in consisting of unpleasant constituents. Physiol. Behav.
the honeybee brain. Nature 387:285–288. 56:81–93.
Jones, B. P., Moskowitz, H. R., and Butters, N. (1975). Olfactory dis- Laing, D. G., Eddy, A., Francis, G. W., and Stephens, L. (1994b).
crimination in alcoholic Korsakoff patients. Neuropsychologia Evidence for temporal processing of odor mixtures in
13:173–179. humans. Brain Res. 651:317–328.
Jones, F. N. (1955). Olfactory absolute thresholds and their impli- Laska, M., and Hudson, R. (1991). A comparison of the detection
cations for the nature of the receptor process. J. Psychol. thresholds of odor mixtures and their components. Chem.
40:223–227. Senses 16:651–662.
Jones, F. N., and Woskow, M. H. (1964). On the intensity of odor Laska, M., and Hudson, R. (1992). Ability to discriminate
mixtures. Ann. NY Acad. Sci. 116:484–494 between related mixtures. Chem. Senses 17:403–415.
Jones-Gotman, M., and Zatorre, R. J. (1993). Odor recognition Laska, M., and Hudson, R., and Distel, H. (1990). Olfactory sen-
memory in humans: role of right temporal and orbitofrontal sitivity to biologically relevant odors may exceed the sum of
regions. Brain Cognition 22:182–198. component thresholds. Chemoecology 1:139–141.
Kauer, J. S. (1991). Contributions of topography and parallel pro- Laska, M., Ayabe-Kanamura, S., Hubener, F., and Saito, S.
cessing to odor coding in the vertebrate olfactory pathway. (2000). Olfactory discrimination ability for aliphatic odorants
TINS 14:79–85. as a function of oxygen moiety. Chem. Senses 25:189–197.
Koelega, H. S. (1979). Olfaction and sensory asymmetry. Chem. Lawless, H. T., and Malone, G. T. (1986a). The discrimination effi-
Senses Flav. 4:89–95. ciency of common scaling methods. J. Sensory Stud. 1: 85–98.
Koelega, H. S., and Koster, E. P. (1974). Some experiments on sex Lawless, H. T., and Malone, G. T. (1986b). A comparison of rat-
differences in odor perception. Ann. NY Acad. Sci. 237:234–246. ing scales: sensitivity, replicates and relative measurement.
Köster, E. P. (1975). Human psychophysics in olfaction. In Methods J. Sensory Stud. 1:155–174.
in Olfactory Research, D. G. Moulton, A. Turk, and J. W. Le Magnen, J. (1952). Les phenomenes olfacto-sexuels chez
Johnston, Jr. (Eds.). Academic Press, New York, pp. 345–374. I’homme. Arch. Sci. Physiol. 6:125–160.
Köster, E. P., and de Wijk, R. A. (1991). Olfactory adaptation. In Livermore, A., and Laing, D. G. (1996). Influence of training and
The Human Sense of Smell, D. G. Laing, R. L. Doty, and experience on the perception of multicomponent odor mix-
W. Breipohl (Eds.), Springer-Verlag, Berlin, pp. 199–215. tures. J. Exp. Psychol. Human Percept. Perform. 22:267–277.
Kurahashi, T., Lowe, G., and Gold, G. H. (1994). Suppression of Livermore, A., and Laing, D. G. (1998a). The influence of odor
odorant responses by odorants in olfactory receptor cells. type on the discrimination and identification of odorants in
Science 265:118–120. multicomponent odor mixtures. Physiol. Behav. 65:311–320.
Psychophysical Measurement of Human Olfactory Function 227

Livermore, A., and Laing, D. G. (1998b). The influence of chem- O’Mahony, M. (1979). Short-cut signal detection measurements
ical complexity on the perception of multicomponent odor for sensory analysis. J. Food Sci. 44:302–303.
mixtures. Percept. Psychophys. 60:650–661. O’Mahony, M. (1983). Gustatory responses to non-gustatory
Lorig, T. S., Elmes, D. G., Zald, D. H., and Pardo, J. V. (1999). stimuli. Perception 12:627–633.
A computer-controlled olfactometer for fMRI and electro- O’Mahony, M., Gardner, L., Long, D., Heintz, C., Thompson, B.
physiological studies of olfaction. Behav. Res. Meth. Inst. and Davies, M. (1979). Salt taste detection: an R-index
Comput. 31:370–375. approach to signal-detection measurements. Perception
Mackay-Sim, A., Shaman, P., and Moulton, D. G. (1982). 8:497–506.
Topographic coding of olfactory quality: odorant-specific pat- Overbosch, P. (1986). A theoretical model for perceived intensity
terns of epithelial responsivity in the salamander. in human taste and smell as a function of time. Chem. Senses
J. Neurophysiol. 48:584–596. 11:315–329.
Macmillan, N., and Creelman, C. (1996). Triangles in ROC Pangborn, R. M., Berg, H. W., Roessler, E. B., and Webb, A. D.
space: history and theory of “nonparametric” measures of (1964) Influence of methodology on olfactory response.
sensitivity and response bias. Psychonomic Bull. Rev. Percept. Motor Skills 18:91–103.
3:164–170. Patte, R., and Laffort, P. (1979). An alternative model of olfactory
Makin, J. W., and Porter, R. H. (1989). Attractiveness of lactating quantitative interaction in binary mixtures. Chem. Sens. Flav.
females’ breast odors to neonates. Child Dev. 60:803–810. 4:267–274.
Marks, L. E. (1974). Sensory Processes. Academic Press, New Patterson, M. Q., Stevens, J. C., Cain, W. S., and Cometto-Muniz,
York. J.E. (1993). Detection thresholds for an olfactory mixture and
Marks, L. E., Szczesiul, R., and Ohlott, P. (1986). On the cross- its three constituent compounds. Chem. Senses 18:723–734.
modal perception of intensity. J. Exp. Psychol. Hum. Percept. Pierce, J. D., Jr., Doty, R. L., and Amoore, J. E. (1996). Analysis
Perform. 12:517–534. of position of trial sequence and type of diluent on the detec-
Marks, L. E., Stevens, J. C., Bartoshuk, L. M., Gent, J. F., Rifkin, B., tion threshold for phenyl ethyl alcohol using a single staircase
and Stone, V. K. (1988). Magnitude matching: the measurement method. Percept. Motor Skills 82:451– 458.
of taste and smell. Chem. Senses 13:63–87. Pinching, A. J., and Powell, T. P. S. (1971). The neuropil of the
Maruniak, J. A., Silver, W. L., and Moulton, D. G. (1983). glomeruli of the olfactory bulb. J. Cell Sci. 9:347–377.
Olfactory receptors respond to blood-borne odorants. Brain Potter, H., and Butters, N. (1980). An assessment of olfactory
Res. 265:312–316. deficits in patients with damage to prefrontal cortex.
Moncrieff, R. W. (1959). The counteraction of odors. Chem. Neuropsychologia 18:621–628.
Canada 11:66 –72. Prah, J. D., Sears, S. B., and Walker, J. C. (1995). Modern
Moncrieff, R. W. (1963). What is odour? A new theory. Am. approaches to air dilution olfactometry. In: Handbook of
Perfum. Soc. Oil Rev. 54:453– 454. Olfaction and Gustation, R.L. Doty (Ed.). Marcel Dekker,
Moskowitz, H. (1977). Magnitude estimation: notes on what, New York, pp. 227–255.
how, when, and why to use it. J. Food Qual. 3:195–227. Punter, P. H. (1983). Measurement of human olfactory thresholds
Moskowitz, H. R., and Barbe, C. D. (1977). Profiling of odor for several groups of structurally related compounds. Chem.
components and their mixtures. Sens. Process. 1:212–226. Senses 7:215–235.
Mozell, M. M., Smith, B. P., Smith, P. E., Sullivan, R. L., Jr., and Rabin, M. D. (1988). Experience facilitates olfactory quality dis-
Swender, P. (1969). Nasal chemoreception in flavor identifi- crimination. Percept. Psychophys. 44:532–540.
cation. Arch. Otolaryngol. 90:131–137. Rabin, M. D., and Cain, W. S. (1986). Determinants of measured
Murphy, C. (1986). Taste and smell in the elderly. In Clinical olfactory sensitivity. Percept. Psychophys. 39:281–286.
Measurement of Taste and Smell, H. L. Meiselman and R. S. Rabin, M. D., and Cain, W. S. (1989). Attention and learning in
Rivlin (Eds.). Macmillan, New York, pp. 343–371. the perception of odor mixtures. In Perception of Complex
Myers, A. L. (1982). Psychophysical scaling and scales of phys- Smells and Tastes, D. G. Laing, W. S. Cain, R. L. McBride,
ical stimulus measurement. Psychol. Bull. 92:203–214. and B. W. Ache (Eds.). Academic Press, Sydney, pp. 173–188.
Nagel, W. A. (1897). Über Mischgerüche und die Komponenten- Rakover, S. S., and Teucher, B. (1997). Facial inversion effects—
gliederung des Geruchssinnes. Z. Psychol. Physiol. Sinnesorg. parts and whole relationship. Percept. Psychophys. 59:
15:82–101. 752–761.
Neely, G., Ljunggren, G., Sylven, C., and Borg, G. (1992). Rawson, N. E., Gomez, G., Cowart, B., Brand, J. G., Lowry, L. D.,
Comparison between the Visual Analogue Scale (VAS) and Pribitikin, E. A., and Restrepo, D. (1997). Selectivity and
the Category Ratio Scale (CR-10) for the evaluation of leg response characteristics of human olfactory neurons.
exertion. Int. J. Sports Med. 13:133–136. J. Neurophysiol. 77:1606–1613.
Nordin, S., Brämerson, Lidén E., and Bende, M. (1998). The Richman, R. A., Post, E. M., Sheehe, P. R., and Wright, H. N.
Scandinavian odor-identification test: development, reli- (1992). Olfactory performance during childhood. I.
ability, validity, and normative data. Acta Otolaryngol. 118: Development of an odorant identification test for children.
226–234. J. Pediatr. 121:908–911.
Olsson, M. J. (1994). An interaction model for odor quality and Rovee, C. K., Cohen, R.Y., and Shlapack, W. (1975). Life-span
intensity. Percept. Psychophys. 55:363–372. stability in olfactory sensitivity. Dev. Psychol. 11:311–318.
228 Doty and Laing

Schiet, F. T., and Frijters, J. E. R. (1988). An investigation of the Sutton, S., Braren, M., and Zubin, J. (1965). Evoked-potential
equiratio-mixture model in olfactory psychophysics: a case correlates of stimulus uncertainty. Science 150:1187–1188.
study. Percept. Psychophys. 44:304–308. Takagi, S. F. (1989). Human Olfaction. Tokyo Press, Tokyo.
Schiffman, S. S. (1993). Olfaction in aging and medical dis- Tanner, W. P., Jr., and Swets, J. A. (1954). A decision-making the-
orders. In Science of Olfaction, M .J. Serby and K. L. Chobor ory of visual detection. Psychol. Rev. 61:401–409.
(Eds.). Springer-Verlag, New York, pp. 500–525. Thurstone, L. L. (1927a). A law of comparative judgment. Psychol
Schiffman, S., Orlandi, M., and Erickson, R. P. (1979). Changes Rev. 34:273–286.
in taste and smell with age: psychophysical aspects. In Thurstone, L. L. (1927b). Psychophysical analysis. Am. J. Psychol.
Sensory Systems and Communication in the Elderly, J. M. 38:368–369.
Ordy, and K. Brizze (Eds.). Raven Press, New York, pp. Toms, M., Morris, N., and Foley, P. (1994). Characteristics of
227–246. visual interference with visuospatial working memory. Br. J.
Schiffman, S. S., Reynolds, M. L., and Young, F. W. (1981). Psychol. 85:131–144.
Introduction to Multidimensional Scaling: Theory, Methods, Toyota, B., Kitamura, T., and Takagi, S. F. (1978). Olfactory
and Applications. Academic Press, Orlando, FL. Disorders—Olfactometry and Therapy. Igaku-Shoin, Tokyo.
Schwartz, B. S. (1991). The epidemiology of olfactory dysfunc- Ueno, Y. (1992). Perception of odor quality by free image-asso-
tion. In The Human Sense of Smell, D. G. Laing, R. L. Doty, ciation test. Jpn. J. Psychol. 63:256–261.
and W. Breipohl (Eds.). Springer-Verlag, Berlin, pp. 307–334. Walker, J. C., Kurtz, D. B., Shore, F. M., Ogden, M. W., and
Schwartz, B. S., Doty, R. L., Monroe, C., Frye, R. E., and Barker, S. Roynolds, J. H. IV. (1990). Apparatus for the automated mea-
(1989). The evaluation of olfactory function in chemical workers surement of the responses of humans to odorants. Chem.
exposed to acrylic acid and acrylate vapors. Am. J. Public Health Senses 15:165–177.
79:613–618. Weber, E. H. (1834). De Pulsu, Resorptione, Auditu et Tactu:
Semb, G. (1968). The detectability of the odor of butanol. Annotationes Anatomicae et Physiologicae. Koehler, Leipzig.
Percept. Psychophys. 4:335–340. Weiffenbach, J. M. (1984). Taste and smell perception in aging.
Shepherd, G. M., and Greer, C. A. (1990). Olfactory bulb. In The Gerodontology 3:137–146.
Synaptic Organization of the Brain, G. M. Shepherd (Ed.). Wenzel, B. (1948). Techniques in olfactometry. Psychol. Bull.
Oxford University Press, New York, pp. 133–169. 45:231–246.
Simon, T. W., and Derby, C. D. (1995). Mixture suppression White, E. L. (1979). Synaptic organization of the mammalian
without inhibition for binary mixtures from whole cell patch olfactory glomerulus: new findings including an intraspecific
clamp studies of in situ olfactory receptor neurons of the spiny variation. Brain Res. 60:299–313.
lobster. Brain Res. 678:213–224. Wise, P. M., and Cain, W. S. (2000). Latency and accuracy of dis-
Slotnick, B. M., Bell, G. A., Panhuber, H., and Laing, D. G. criminations of odor quality between binary mixtures and
(1997). Detection and discrimination of propionic acid after their components. Chem. Senses 25:247–265.
removal of its 2-DG identified major focus in the olfactory Wood, J. B., and Harkins, S. W. (1987). Effects of age, stimulus
bulb: a psychophysical analysis. Brain Res. 762:89–96. selection, and retrieval environment on odor identification.
Springer, K. (1974). Combustion odors-a case study. In Human J. Gerontol. 42:584–588.
Responses to Environmental Odors, A. Turk, J. W. Johnston, Wright, H. N. (1987). Characterization of olfactory dysfunction.
Jr., and D. G. Moulton (Eds.). Academic Press, New York, pp. Arch. Otolaryngol. Head Neck Surg. 113:163–168.
227–262. Wysocki, C. J., Dorries, K. M., and Beauchamp, G. K. (1989).
Stevens, J. C., and Cain, W. S. (1986). Smelling via the mouth: Ability to perceive androstenone can be acquired by ostensi-
effect of aging. Percept. Psychophys. 40:142–146. bly anosmic people. Proc. Natl. Acad. Sci. 4(86):7976–7978.
Stevens, J. C., Cain, W. S., and Burke, R. J. (1988). Variability of Yoshida, M. (1984). Correlation analysis of detection threshold
olfactory thresholds. Chem. Senses 13:643–653. data for “standard test” odors. Bull. Facul. Sci. Eng. Chuo
Stevens, S. S. (1961). The psychophysics of sensory function. In Univ. 27:343–353.
Sensory Communication, W. A. Roseblith, (Ed.). MIT Press, Zwaardemaker, H. C. (1900). Die kompensation von Geruch-
Cambridge, MA. sempfindungen. Arch. Physiol. Leipzig, pp. 423–432, a trans-
Stewart, W. B., Kauer, J. S., and Shepherd, G. M. (1979). lated in Perf. Ess. Oil Rec. 50:217–221, 1959.
Functional organization of rat olfactory bulb analyzed by Zwaardemaker, H. (1925). L’Odorat. Doin, Paris.
the 2-deoxyglucose technique. J. Comp. Neurol. 185: Zwaardemaker, H. (1927). The sense of smell. Acta Oto-
715–734. Laryngol. 11:3–15.
Stuiver, M. (1958). Biophysics of the sense of smell. Doctoral Zwaardemaker, H. C. (1930). An intellectual history of a physi-
thesis, University of Groningen, The Netherlands. ologist with psychophysical aspirations. In A History of
Sumner, D. (1962). On testing the sense of smell. Lancet Psychology in Autobiography, Vol. 1., C. Murchison (Ed.).
2:895–897. Clarke University Press, Worcester, MA, p. 491.
11

Electrophysiological Measurement of Olfactory Function

Gerd Kobal
University of Erlangen, Erlangen, Germany

I. INTRODUCTION important to remember that methodological approaches

employing nonhumans are usually simply surrogates or
Unlike the situation in other sensory modalities, the field models for the human sense of smell. In some cases, data
of human olfactory electrophysiology is rather poorly obtained from animals or cell cultures do not generalize
developed. For example, in vision, recordings of electro- well to humans. A case in point is a recent study that
retinograms (ERG) or of visual event–related potentials suggests that the distribution of the human olfactory
(VERP) are routinely used diagnostically, in contrast to the epithelium is different from what would be predicted from
situation in olfaction, where analogous potentials are rarely animal investigations (Leopold et al., 2000). It is apparent
measured, even in university medical centers. However, as that the collection of valid information about human
noted in this chapter, a number of laboratories are working olfactory processing requires the use of human subjects.
in this field and have generated an impressive number of The bulk of this chapter is devoted to the most common
publications and a body of useful information for better electrophysiological signal that has been measured to date,
understanding elements of the human olfactory process. namely, the olfactory event–related potential (OERP). Other
Why should one want to obtain electrophysiological or measures that are discussed in detail include the EOG,
other physiological measurements of the olfactory system event-related changes in the background electroencephalo-
in humans? First, there is a general need for more reliable gram (EEG), and signals derived using magnetic source
data in all fields of science. Recording brain potentials, imaging (MSI). A description of findings that have provided
magnetic responses, changes in blood flow, etc., provides new insights into the olfactory system or have raised
information that minimizes or eliminates potential biases questions concerning functional properties and relationships
related to conscious subject responses. For example, in of brain areas activated by odorants is also presented.
some medical cases—particularly those associated with
litigation—malingering may occur and electrophysiologi-
cal assessment can greatly aid in the detection of such II. A BRIEF HISTORY OF ELECTROPHYSIO-
deception. Second, electrophysiological studies may help LOGICAL RESEARCH ON THE HUMAN
to determine the neural structures involved in pathological SENSE OF SMELL
changes in sensory responsiveness, such as in hyposmia or
dysosmia. For example, EOG measurement can be used, in In 1883 Fleischl von Marxow observed that ammonia
some cases, to establish the involvement of the epithelial produced electrical brain potentials when presented to a
receptors in an olfactory deficit. Questions concerning rabbit’s nose (Fleischl von Marxow, 1890). Although
such localization often arise, particularly in relation to Berger (1929) assumed that such potentials could also
pharmacological or surgical interventions. Finally, it is be found in the human EEG, he failed to demonstrate

229
230 Kobal

them. Indeed, it was not until the 1960s that OERPs were potentials are EEG-derived polyphasic signals reflecting
recorded by Finkenzeller (1965) and Allison and Goff activation of cortical neurons which generate electromag-
(1967). Around this same time, the first electrophysiologi- netic fields (Picton and Hillyard, 1988). The more neurons
cal recording from the human nasal mucosa (i.e., the EOG) that are activated or synchronized, the larger the amplitude
was obtained by Osterhammel et al. (1969). This achieve- of the signal obtained at the surface of the scalp. Since the
ment was based upon the earlier animal studies of Hosoya EEG is a noisy signal, which contains activity from many
and Yoshida (1937) and Ottoson (1956). cortical neurons, ERPs need to be extracted from the back-
Although suprisingly little work has subsequently been ground activity. The classical approach to this problem
carried out on the EOG (see Hummel et al., 1996b; involves averaging of individual responses to olfactory
Leopold et al., 2000), research on human odor ERPs stimuli such that random activity would cancel itself out,
continued through the 1970s in various laboratories (e.g., thereby leaving only nonrandom activity. Therefore,
Giesen and Mrowinski, 1970; Herberhold, 1976; stimuli are typically presented repetitively with a steep
Cianfrone and Subiaco, 1978). However, technical diffi- onset (100 ms) in a well-controlled and homogeneous
culties in producing defined olfactory stimuli with steep environment so that the stimulus onset synchronizes the
rise times and in analyzing the large amount of generated activity of as many cortical neurons as possible.
data hindered rapid progress. Moreover, data presented by Three prerequisites must be met to obtain clear and accu-
Smith et al. (1971) suggested that an OERP could not be rate OERPs. First, as noted above, the stimulius must have a
found in patients who had lost their trigeminal sensitivity, steep onset. Although a shallow stimulus onset may lead to a
conceivably discouraging others to continue efforts along sensation, this sensation may not be reflected in an ERP as
these lines. As described in detail below, subsequent stud- the cortical activity “drowns” in background noise. Second,
ies by my group (e.g., Kobal and Plattig, 1978; Kobal, the stimulus needs to be presented repetitively. This requires
1981) found that Smith et al.’s observation was likely due precise temporal control of stimulus onset in the range of
to experimental artifact. As a result of our early studies, a milliseconds as fluctuations in the timing of stimulus onset
new era in the study of human OERPs, based upon will lead to differences in the peak latencies of individual
sophisticated odorant presentation techniques, was born. ERPs (“jitter”). This jitter will lead to the modification/
cancellation of peaks in the averaged response. In addition,
desensitization to repeated stimuli becomes an issue. Finally,
III. EVENT-RELATED POTENTIALS VERSUS
to properly interpret the response it is necessary to know
EVOKED POTENTIALS
whether it is derived from intranasal chemical stimulation of
the trigeminal (CN V) or olfactory (CN I) system.*
It should be noted that event-related potentials, which
reflect high order processing, can be elicited by both exter-
nal and internal stimuli. For example, when a series of
*The recording of an ERP can be compared to the situation in a
stimuli is presented with a constant interstimulus interval,
the omission of one stimulus may trigger an event-related soccer stadium where, in analogy to an EEG electrode, a micro-
phone is positioned over the middle of the field to record all
potential, even though no physical, external stimulus is
sounds. The stadium is filled with thousands of people, similar to
present. This is in contrast to “evoked potentials” (also
the millions of neurons sitting under a recording electrode on the
termed “exogenous” or “obligatory” potentials), which scalp. During long sequences of the game, only noise is recorded
reflect the very early components of the response and are coming from the soccer fans, talking to each other, laughing,
largely independent of a subject’s mental state or arousal commenting on the quality of the game etc. When the game
(Näätänen et al., 1993). The latter have not been recorded becomes more exciting there may be more noise; when it is less
noninvasively in humans to odorants, although they have exciting, people typically become quieter. But when a goal is
been obtained from the olfactory bulb and amygdala scored communication between soccer fans becomes syn-
during surgery (Hughes et al., 1969, 1972; Hughes and chronized with many of them shouting—which compares to the
Andy, 1979a,b; Kobal et al., 1998, unpublished). synchronization of cortical neurons by the sudden onset of a
stimulus. However, similar to the electrical fields of the EEG, it
is a difficult to localize the site where the goal has been scored.
IV. OLFACTORY EVENT–RELATED If the noise comes from the left side this does not necessarily
POTENTIALS mean that the goal has been scored on the left. It may also be due
to the fact that most of the fans of the scoring club sit on the left
A. Stimulation Requirements and Considerations side, but the goal was actually scored on the right side (and this
situation may change during the game, when teams switch sides).
In general, precise stimulus control is crucial when record- Finally, it may also be that the goal is not scored in this stadium
ing event-related potentials. Why is this so? Event-related but elsewhere. Specifically, the fans of one of our two teams may
Electrophysiological Measurement of Olfactory Function 231

(a) (b)

Figure 1 (a) Schematic diagram of the switching device. When the odorant is switched on or off, the subject is unable to discern
turbulences or changes in flow rate or pressure. The temperature and humidity or the carrier gas (air) are closely controlled. (b) The
Burghart OM4/b olfactometer. Left. Subject being presented with odorants and performing a computerized visual attention task. Right.
Data collection module. Center. olfactometer body. (Photo courtesy of the University of Pennsylvania Smell and Taste Center,
Philadelphia, PA.)

How is it possible to produce chemical stimuli that have plus dilution
D), whereas the other contains odorless air
a rectangular shape with rapid onset, that are precisely (control
C). Different odorant concentrations are gener-
controlled in terms of timing, duration, and intensity, and ated by means of air dilution; hence, a preestablished, fully
that do not simultaneously activate sensory systems other odorant-saturated air stream (odorant
O) is mixed with
than olfaction? Based on the principles of air-dilution an odorless air stream (dilution
D). While the sum of the
olfactometry (Prah et al., 1995), such a system was devel- two air streams is always constant (equal to the control air
oped in the late 1970s and refined in the 1980s (Fig. 1) stream C), different O:D ratios produce different stimulus
(Kobal and Plattig, 1978; Kobal, 1981; Kobal and concentrations. A separate system of finely tuned pressure
Hummel, 1988). Odorants are applied intranasally by and vacuum is applied such that, similar to an air curtain,
means of a canula with an inner diameter of 2–3 mm. This a small current of odorless air prevents molecules from O
canula is inserted for approximately 1 cm into the naris tubings to be drawn into other tubes. This cross current
such that its opening lies beyond the nasal valve. allows attaching several different odor lines to the same
Presentation of odorants does not simultaneously activate dilution line. During the interstimulus interval, a precisely
mechano- or thermoreceptors in the nasal mucosa, as odor tuned vacuum draws the odorant-containing air steam from
pulses are embedded in a constantly flowing, humidified the vacinity of the flowing air, ensuring that only odorless
air stream (typically 6–8 L/min). Hence, subjects do not air enters the subject’s nose during this time. Employing
perceive any change in flow rate when the stimulator is this device, it is possible to switch between an odorized air
switched from a no-stimulus to a stimulus condition, and steam and control air in less than 20 msec. Depending on
vice versa. the physicochemical properties of the odorants employed,
In this system, two air streams are directed towards the a switch from one odorant to another can be made in less
outlet of the olfactometer. Both have the same flow rate, than 5 seconds without contamination from the previous
the same temperature, and the same humidity. One con- stimulus.
tains an odorant at a defined concentration (odorant
O The constant airflow directed into a subject’s nose
requires humidification (~80% relative humidity) and a
have brought receivers to listen to the broadcast of a different, stable temperature (36°C), since dry cool air produces nasal
extremely important game; it may be that in this different stadium congestion, mucus discharge, and pain, which can interfere
a goal is scored which turns the situation in the league in favor of with the olfactory process (Mohammadian et al., 1997,
one of our teams. This would then also create a synchronized 1999; Lötsch et al., 1998). The warmed and humidified air
outcry of the fans although nothing much has happened in the stream employed in our studies becomes unnoticable
play we actually observe. In a similar way, pinpointing the source within a few seconds of its introduction; i.e., the subject
of an event-related potential can be a difficult task. adapts to the following air.
232 Kobal

In the commercially available olfactometers based upon whether the observed cortical responses to peripheral
our designs (Burghart Laboratories, Hamburg, Germany), electrical stimulation reflect olfactory or somatosensory
airflow rates are determined by mass-flow controllers activation.
that, along with switching valves, are controlled by
computer. The recording of stimulus-linked EEG segments B. Recording of Olfactory Event-Related Potentials
(or any other physiological measure that can be trans-
formed into electrical currents) is integrated into the As noted earlier, ERPs are due to changes in electrical
same software that controls the olfactometer and stimulus fields generated by large populations of cortical neurons. If
presentation. This equipment also allows the setup of they are recorded from the intact surface of the skull, their
sequences of stimuli with different quality, intensity, amplitudes are very small (50 V), and it becomes
duration, or interstimulus interval. Thus, the recording of necessary to isolate them from the background activity by
the OERP becomes a routine procedure that can be carried averaging and/or filtering. For averaging, a certain number
out by any technician—a fact of particular importance in of stimulus-synchronous EEG records of 1–2 sec duration
clinical applications. are digitized by computer and transformed into a sequence
In contrast to the stimulus presentation procedures of numerical values. Thus, averaging the stimulus-locked
described earlier, some laboratories record OERPs in array of numbers results in visualized waveforms
relation to stimuli that are puffed into the nasal cavity, a representing responses of synchronously reacting cortical
procedure that we do not recommend. What happens when neurons. However, if this technique is to yield reliable
odorants are puffed into the nose? Under these conditions, data, several prerequisites have to be met. For one thing,
it is not only possible to obtain ERPs in anosmic subjects the background activity of the EEG has to be stationary
(Herberhold, 1976; Cianfrone and Subiaco, 1978; and, at the same time, stochastic in regards to the con-
Swandulla, 1986; Sakuma et al., 1996; Bauer and Mott, cealed response. For another, the specific stimulus-induced
1996; Harada et al., 1997), but ERPs in normosmics that response, i.e., the activity determined by the applied stim-
reflect the mixed activation of both the trigeminal and the ulus, has to be stable, especially if there are a number of
olfactory systems. Such combined activity leads to numer- consecutive positive and negative components. Even the
ous interactions at various levels of neuronal processing slightest phase-shifting can easily cause the positive and
(for review, see Hummel and Livermore, 2001) which can- negative components of the event-related responses to can-
not be remedied by simple mathematical procedures. For cel each other, in the same way that can occur with back-
example, the average of responses to individual stimula- ground activity. As noted in the previous section, positive
tion with carbon dioxide and vanillin or hydrogen sulfide and negative waves are only separated by a fraction of a
is significantly different from the response obtained after second. Therefore, the quality of the temporal presentation
stimulation with the binary mixture of carbon dioxide and of the reproducible stimuli has to be excellent.
hydrogen sulfide (Kobal and Hummel, 1988; Livermore et The first positive peak of OERPs typically occurs at
al., 1992). Thus, it is difficult to interpret responses to latencies of 250 msec. This peak is then followed by at
olfactory stimuli contaminated by mechanical stimulation least two other peaks—a major negativity and the late pos-
in patients with olfactory disorders. In addition, since the itive complex. As there has been confusion in the past on
interactions between the trigeminal and olfactory systems the nomenclature of OERPs, some have suggested naming
are difficult to predict, it is misleading to interpret res- each peak according to its polarity and mean latency at
ponses to mixed olfactory or trigeminal stimuli to reflect position Cz (e.g., Evans, 1993; Hummel, 2000). For
predominantly (and, implicitly, more or less exclusively) example, a negativity at a mean latency of 340 msec would
olfactory or trigeminal activation (Geisler and Murphy, be called N340. In this review, peaks are named in the
2000). more conventional fashion to allow for comparisons across
Another approach to establishing OERPs involves the studies—i.e., P1, N1, and P2 (compare with Evans, 1993).
electrical activation of the olfactory epithelium (Sato et al., Using data published in Kobal and Hummel (1988) for
1996; Ishimaru et al., 1997). While this would appear, OERPs to vanillin, these peaks correspond to P383, N484,
at first glance, to be an extremely attractive technique— and P649 using the nomenclature based on peak latencies at
especially when considering the possibility that this would position Cz. It should be noted that the P2 peak has also
more definitively allow for the investigation of the integri- been described as P1 (Prah and Benignus, 1992) or P3
ty of axonal connections from the olfactory receptors to (Lorig, 1993; Barz et al., 1997).
the olfactory bulb—it suffers from the fact that electrical Why is it that OERPs appear later than auditory or visual
stimulation also activates trigeminal nerve endings of the OERPs? Unlike the case in audition, chemical stimulation
olfactory epithelium (Silver, 1991). Hence, it is not clear needs approximately 100–200 msec utilization time at the
Electrophysiological Measurement of Olfactory Function 233

site of the receptors. For comparative purposes, this utiliza- olfactory induced cortical activation (compare Spencer
tion time must be subtracted from the latencies (Getchell et et al., 1999), as exemplified by the odor of nicotine
al., 1984; Hummel et al., 1996b; Leopold et al., 2000). Thus, (Hummel et al., 1992). Nicotine produces odor, burning,
the N1 and P2 OERP peak latencies are, in fact, comparable and stinging at increasing sensations. When presented at a
to the N100 and P200/P300 latencies of audition and other concentration that mainly produces odor, maximum ampli-
sensory modalities. tudes N1 are found at centro-parietal sites. However, when
As briefly mentioned earlier in this chapter, unlike the used in concentrations that produce trigeminally mediated
case for audition or vision, no early ERPs have been sensations, amplitudes are significantly larger at position
recorded in response to olfactory stimuli, only late near- Cz compared to all other recording sites. This relates, at
field ERPs (i.e., responses from cortical neurons) (see least partly, to the different cortical areas that are activated
Kobal and Hummel, 1991, for review). Peaks of the late by olfactory and trigeminal stimuli.
near-field ERPs fall into two groups. Earlier peaks like
N1 (N340) encode exogenous stimulus characteristics to a D. Control of Testing Conditions
larger extent than later, so-called endogenous, compo-
nents. That is, earlier components encode stimulus During OERP recording, stable environmental conditions
intensity or stimulus quality (e.g., “What is the nature of are important. This includes the visual and acoustical
this stimulus?”), whereas later components are more shielding of subjects. Visual shielding can be performed
related to the frequency, or the saliency of the stimulus using drapes or blindfolds; for acoustical shielding, white
(“What is the meaning of this stimulus?”) (Donchin, 1986; noise (a mixture of all frequencies at the same loudness) is
Picton and Hillyard, 1988; Pause et al., 1996b; Krauel typically used. Further, a defined task has proven most
et al., 1998). helpful. Specifically, many labs use a tracking task, where,
for example, subjects are requested to keep a small square
C. Stimulation and Recording Parameters controlled by a joystick inside a larger one, which ran-
domly moves on a screen at a distance of ~1.5–2 m from
Since the frequency spectrum of late near-field ERPs the subject’s eyes. This task fulfills a number of purposes,
ranges between 1 and 8 Hz, both filtering and sampling including stabilization of eye movements (minimizing
frequency must be set accordingly. We prefer low-pass artifacts from eye movements), maintaining vigilance or
filtering at 30 or 70 Hz. As regards the number of averages attention, and providing quantitative assessments of how
per ERP, eight records are considered to be the absolute vigilant or attentive a subject was during various periods of
minimum.* However, some authors have used up to 200 the recording session (e.g., by assessing error rates).
records to average a single response, such as in the record- Typically, a subject’s performance increases during an
ing of late positive components in the so-called oddball experiment. Evaluating tracking performance also makes it
paradigm, where two stimuli are presented with different possible to assess subtle differences in vigilance due to
probabilities of occurrence (Pause et al., 1996b). Given an such experimental manipulations as the administration of
ISI of 30–40 sec in single trial experiments, recording of sedatives or analgesics.
such large numbers of responses requires at least 100 min. During a test session, subjects are instructed to relax
This is impractical and unnecessary in many situations, and sit as still as possible. This requires that they are
given evidence that improvement of the signal-to-noise- comfortably seated in a chair equipped with arm, leg, and
ratio starts to reach a ceiling effect at 40 averages. In addi- head rests. Often, subjects are trained, using simple
tion, the long testing periods can introduce other artifacts, biofeedback, in a specific breathing technique, the
e.g., changing levels of vigilance during recording. velopharyngeal closure, to prevent respiratory flow
OERPs can be recorded from numerous scalp locations. through the nose (Nagel, 1904; Kobal and Hummel, 1989).
Amplitudes exhibit characteristic patterns across scalp This technique, which is mastered in less than 5 min, helps
loci, with a centro-parietal maximum for both amplitudes to prevent respiratory flow through the nose by lifting the
N1 and P2 (compare Lorig et al., 1996; Pause et al., 1996b; soft palate, which is under voluntary control. An alter-
Murphy et al., 1998). This specific topographical distribu- native to velopharyngeal breathing is to present stimuli
tion can be used to differentiate between trigeminal and synchronously with inspiration (Tonoike et al., 1982; Lorig
et al., 1996; Pause et al., 1999; Hummel et al., 2000).
*The use of only eight stimuli—while producing meaningful However, one must be aware that responses obtained under
results—invites noise which, in turn, may only be mitigated by these circumstances are contaminated by the so-called
increasing the number of subjects. Hence, a larger number of tri- “contingent negative variation” (Walter et al., 1964; Tecce,
als is generally recommended. 1972; Torii et al., 1988; Lorig and Roberts, 1990;
234 Kobal

Auffermann et al., 1993). This negativity builds up as a Pause et al., 1997; Covington et al., 1999), the latencies
consequence of the expectation of odorous stimulation, appeared to be more strongly related to changes in stimulus
which is more likely to happen during inspiration. In addi- intensity than were the amplitudes. This observation, which
tion, there is evidence that the processing of olfactory is analogous to what is commonly observed in other sen-
information differs between inspiratory and expiratory sory modalities, suggests that central olfactory processing
phases (Hummel et al., 2000). may be more strongly tuned for discriminating among odor
Considering these extensive prerequisites for an “ideal” qualities than for differentiating odor intensities per se.
session, it becomes clear why many experiments benefit
from a specific adaptation or training session where 2. Stimulus Duration
subjects are thoroughly acquainted with the experimental
OERPs relate to stimulus onset (Fig. 2). Kobal (1981)
procedures. While these efforts optimize recording condi-
presented the mixed olfactory/trigeminal stimuli isoamyl
tions and, thus, the signal-to-noise ratio of the responses,
acetate and eucalyptol at a constant concentration, but at dif-
they may be regarded as “overkill” in situations when
ferent stimulus durations (100, 300, 500, and 700 msec; ISI
subjects are only tested once, e.g., in clinical settings.
40 sec), to a group of volunteers. While odor intensity
Thus, normal mouth breathing, rather than velopharyngeal
ratings increased with increasing stimulus duration, there
closure, may be less stressful to some patients. Stress itself
was no difference between OERPs. This indicated that the
may adversely effect the signal-to-noise ratio.
OERP—like ERPs in other sensory modalities—is predom-
inantly determined by early segments of stimulus onset.
E. Influences of Stimulus Characteristics and
Stimulus Presentation Procedures 3. Relation to Airflow Rate
OERPs are related to the flow rate used to transport
1. Stimulus Intensity
odorants to the olfactory epithelium. This is expected, as
There is controversy as to how OERP amplitudes relate to the OERP depends on the number of odorous molecules
odorant concentration. In rats, it has been shown that
OERP latencies shorten and amplitudes increase with ris-
ing odorant concentrations (Evans and Starr, 1992). In
humans, some studies (e.g., Pause et al., 1996b) have used
stimuli with mixed olfactory-trigeminal properties, such as
linalool or citral (see Doty et al., 1978; Kobal and
Hummel, 1988). Others have employed odorants with little
or no trigeminal activity but have methodological short-
comings (Thiele and Kobal, 1984; Prah and Benignus,
1992; Pause et al., 1997). Still others have suffered from
small sample sizes and the lack of statistical analyses
(Köster, 1965) or too low of flow rates to observe all but
the late OERP positivity (Prah and Benignus, 1992). In one
study (Pause et al., 1997), low concentrations of linalool
were presented in ascending sequence, conceivably super-
imposing adaptation or habituation on the obtained
responses (Köster, 1965; Köster and de Wijk, 1991; Dalton
and Wysocki, 1996).
A recent study employing 15 subjects reported that
OERP amplitudes (both early and late components)
discriminate between different concentrations of vanillin
(Tateyama et al., 1998). A similar finding was observed in
Figure 2 OERP in relation to stimulus duration; example from
a sample of 36 subjects using H2S (a stimulus that, at the
a single subject; OERP obtained at recording position Cz to euca-
concentrations used, activates only CN I afferents)
lyptol stimuli of different duration, but identical concentration
(Hummel et al., 1998b). Thus, N1 and P2 amplitudes (100, 300, 500, and 700 msec; ISI 40 sec). Response shapes do
increased as H2S stimulus concentration increased. In not change significantly, although stimulus duration varies in the
addition, the OERP latencies were decreased in a concen- range of 1:7. The shape of the response depends on the onset of
tration-related manner. In line with previous research (e.g., the stimulus, which is the same in all cases.
Electrophysiological Measurement of Olfactory Function 235

presented during early parts of the stimulus wave (see (Kobal and Hummel, 1989, 1992; Hummel and Kobal
above); thus, both odor concentration and airflow should 1994). As discussed above, this may be a reflection of the
be determinants of the OERP. Kobal (1981) investigated activation of different brain areas by different odorants
this issue in healthy subjects using eucalyptol and linalool (Ayabe-Kanamura et al., 1997; Kettenmann et al., 1997a,b)
(at concentrations of 1287 and 6481 ppm, respectively; ISI and/or their hedonic valence or emotional significance.
40 sec; stimulus duration 200 msec). Airflow varied
between 5 and 277 mL/sec. Both amplitudes and latencies 5. Interstimulus Interval
varied as a function of airflow (Fig. 3).
Although the ISI is a significant determinant of OERPs,
little research has been done in this area. In one study, the
4. Odor Quality
mixed olfactory/trigeminal stimulant eucalyptol (total flow
Attempts to relate differences in the shape of OERPs to 94 mL/sec; 12037 ppm) was used to investigate this issue
differences in odor quality have not been successful (e.g., (Kobal, 1981). ISIs of 12, 22, 32, 42, and 52 sec were
Kobal and Hummel, 1988). However, OERPs to different investigated in 18 healthy subjects ranging in age from 20
odorants exhibit differences in the topographical distribu- to 41 years. An increase of the ISI from 12 to 42 sec was
tion over the scalp, even when normalized with respect to accompanied by a marked increase of amplitudes of the
maximum OERP amplitudes (Hummel et al., 1992). When OERP. Further increase of the ISI had little, if any, effect.
differences in the topographical distribution of OERP It was concluded that an ISI of 40–50 sec is ideal for such
amplitudes and latencies are considered in relation to the studies, as desensitization was minimal and the time
stimulated naris (Kobal and Hummel, 1992), this complex required to collect an adequate number of potentials was
pattern may be used to investigate stimulus quality. reasonable. Using the odorant isoamyl acetate at concen-
Specifically, differences between latencies and amplitudes trations that reportedly do not produce trigeminally medi-
of OERP to various odorous stimuli (e.g., acetaldehyde, ated sensations, Morgan et al. (1997) confirmed the afore-
phenyl ethyl alcohol, hydrogen sulfide, eugenol, and mentioned findings, at least for young subjects (see also
vanillin) were found in relation to lateralized stimulation Murphy et al., 2000).

Figure 3 OERP to eucalyptol (200 msec stimulus duration, ISI 60 sec, 6481 ppm) presented at different flow rates (5, 85, 162, 235,
277 mL/sec; means, standard deviations, n
10). Results are shown for peaks N1 and P2. Correspondingly obtained intensity ratings
(magnitude estimations; in estimation units, EU) are shown on the right. Amplitudes increase and latencies decrease when the flow rate
is increased, i.e., the dose of the odorant is increased. Psychophysical results have a clearer dose dependency, because the response is
related to the integrative evaluation of the total stimulus duration, whereas the OERP differences are only related to the differences in
integrations during the onset of the stimulus.
236 Kobal

While the studies mentioned above investigated the 1998a; Covington et al., 1999). In the most extensive of
optimal ISI for recording of OERPs, it clearly is possible these studies, Murphy et al. (2000) examined the OERP to
to obtain meaningful data using ISIs of 20 sec or less amyl acetate in a group of 140 individuals whose ages
(Durand-Lagarde and Kobal, 1991; Kettenmann et al., spanned a wide range. A linear prolongation of peak laten-
1997b; Hummel et al., 1998a). Krauel (1999) varied the cies, accompanied by a decrease of amplitudes (amplitude
ISI to investigate the potential presence of a mismatch neg- N1P2: 1.5 V/year; amplitude P2: 2.0 V/year;
ativity (Naatanen et al., 1993) in the OERP. Using an ISI latency P2 and P3: 2.0 msec/year), was noted across the
of 15 sec, a negative deflection was found that was not age categories. These and other studies suggest that the
detectable at an ISI of 30 sec. The authors interpreted their changes in the processing of olfactory information may
findings in terms of the presence of a transient storage of appear relatively early in life. For more details see
olfactory information, as reflected by the mismatch nega- Hummel and Kobal (2001).
tivity.
4. Gender
F. Subject Variables that Influence OERPs OERPs reflect differences in olfactory sensitivity in relation
to gender. In study in 35 healthy subjects, ERP amplitudes
1. Arousal or Vigilance to both H2S and vanillin were found to be larger in women
OERPs are related to arousal on different levels. On the than in men (Becker et al., 1993). Evans et al. (1995) found
one hand, they are dependent on the background signal, the peak-to-peak amplitude N1P2 to be larger in women
wheras on the other hand they are related to cognitive fac- than in men.
tors. Using linalool and eugenol in subjects, Krauel et al.
(1998) found that allocation of attention led to a latency G. Reliability of OERP Measures
decrease of early components and a simultaneous increase
of the amplitude of the late positivity of the OERP (com- Despite the multitude of possible sources of variation,
pare to Spence et al., 2000). Similar findings for the OERP given a proper experimental environment the test-retest
to amyl acetate were reported by Geisler and Murphy reliability of OERPs is regarded as high. For example,
(2000). Kobal and Hummel (1988) reported very good repro-
ducibility of OERPs in 13 subjects who were tested on 3
2. Ultradian Rhythms different days using different stimuli (compare Lago et al.,
1998). Also, when measurements of OERPs to H2S (12
Using OERPs evoked by citral, Pause et al. (1996a) found subjects) were repeated before and after application of a
evidence that the processing of olfactory information nonactive (placebo) nasal spray, there was no significant
varies during the menstrual cycle; they investigated sub- difference between olfactory OERP obtained during the
jects with cycle lengths ranging from 21 to 39 days. two consecutive sessions (Hummel et al., 1998c).
During the periovulatory period (monitored by changes in
body temperature), stimuli were perceived as more com-
H. Relation of OERPs to Psychophysical and
plex/novel, as indicated by an increased amplitude P3-1
Neuropsychological Measures
(see also below). A shortening of ERP latencies was noted
near the time of expected ovulation, and a shortening of
OERPs are generally correlated with psychophysical
peak latencies was observed during the follicular phase.
measures of olfactory function. Tateyama et al. (1998)
Consequently, the menstrual cycle should be considered
found in a study of 16 subjects that correlations between
when looking for subtle changes in the perception of odors
olfactory threshold measures and OERP latencies to
using small sample sizes.
vanillin were larger when the highest concentrations had
been used on the OERP determinations. This may relate to
3. Age
the poor detectability of OERP peaks at low odorant
The well-established age-related decrease in chemosen- concentrations. In addition, correlations between latencies
sory function is mirrored in OERP measures (Hummel and thresholds were higher for the early ERP components
et al., 2001). Specifically, the amplitude of the OERP’s (P1 and N1) than for the later positivities. Correlation
late positivity (P2) and the composite amplitudes of the coefficients were generally larger for OERP latencies than
two major peaks (N1, P2) decrease with increasing age, amplitudes. The ability of subjects to identify odors was
whereas a prolongation of N1 peak latencies occurs found to correlate with P2 amplitudes (Hummel et al.,
(Murphy et al., 1994; Evans et al., 1995; Hummel et al., 1998a).
Electrophysiological Measurement of Olfactory Function 237

Recently, Geisler et al. (1999a) reported that both Hummel (1991) includes the recording of responses to
amplitudes and latencies of the late positivity of the OERP olfactory (e.g., hydrogen sulfide and vanillin) and trigemi-
correlated weakly, but significantly, with scores on a nal (e.g., CO2) stimuli (Fig. 4). All three stimulants are
number of neuropsychological tests, including the Trail applied 15 times both to the left and the right nostril of the
Making Test (a test of visual-motor attentional processing; patient. The interstimulus interval is approximately 40 sec.
Reitan, 1971) and the California Verbal Learning Test (a The session, including preparation, lasts for 80 min. The
test of short- and long-term memory, free and cued recall, EEG (filtering 0.1–30 Hz, electrode impedance 10 k) is
and delayed recognition). They also found that the age- recorded from positions of the international 10/20 system
related decrease of OERP amplitudes and the correspond- (Fz, Cz, Pz, and Fp2, referenced to A1A2). EEG records
ing increase of latencies were accompanied by reduced of 2048 ms duration are digitized (sampling frequency
neuropsychological performance. 250 Hz) and averaged in groups according to the three
stimulants and the binasal stimulation. EEG records cont-
I. Applications of OERPs aminated by eye-blinks (Fp2/A1A2) or motor artifacts
are discarded from the average. This procedure (compare
An important application of OERPs is their use in Evans, 1993) has recently been adopted by the working
diagnosing olfactory deficits and, in some cases, detecting group “Olfaction and Gustation” of the German ORL
malingering. A test procedure standardized by Kobal and Society (Hummel, 2000). So far, in all anosmic patients

Figure 4 Clinical olfactometry in a patient suffering from anosmia in the right (re) nasal cavity. Olfactory specific stimulants (vanillin
and hydrogen sulfide) were used to stimulate both nostrils separately. Stimulation of the left (li) nostrils resulted in a rather large response
to vanillin and in a less clear response to H2S, while there were no responses when the right (re) was stimulated. The use of carbon diox-
ide to stimulate specifically fibs of the trigeminal nerve resulted in responses on both sides, although somewhat different in shape and
latency. Complete oflactory deficits on only one side are not that unusual. The patient noticed this deficit in a reduced ability to smell
compared to times when both nostrils were equally sensitive.
238 Kobal

that have been investigated, intranasal trigeminal ERPs patients, although they were, in fact, anosmic upon
could be obtained after stimulation with CO2, although psychophysical testing. Clearly, more research would be
with significantly smaller amplitudes than those of healthy of value on this topic, especially since OERPs appear to
individuals (Kobal, 1982; Hummel et al., 1996a). In con- be ideally suited to investigate patients who may have
trast, no OERPs could be detected after stimulation with difficulties cooperating or paying attention to sensory
the odorants hydrogen sulfide and vanillin (Kobal and stimuli.
Hummel, 1998). Using identical or slightly different pro-
tocols, this technique has been used in numerous labs (e.g., 3. Multiple Sclerosis
Leplow, 1994; Matern et al., 1995; Cui and Evans, 1997;
Hawkes and Shephard, 1998; Mata et al., 1998; Welge- Olfactory function has been reported to be impaired in
Lüssen, 1999; Geisler et al., 1999b; Welge-Lüssen et al., patients with multiple sclerosis (MS). Hawkes (1996)
2000). found one quarter of 45 MS patients to have delayed
OERPs, even though only 15% exhibited a decreased abil-
J. Assessment of Neurodegenerative Disorders ity to identify odors. Similar to the findings reported in PD
(see above), this suggests that OERPs may be more sensi-
As noted in Chapters 23 and 24, a number of neurodegen- tive to olfactory than some psychophysical measures.
erative disorders are associated with olfactory dysfunction.
OERPs have been found to be altered in many such disor- 4. Motor Neuron Disease
ders, as described in detail below. Hawkes et al. (1998) investigated olfactory function in 58
patients with motor neuron disease (MND) in comparison
1. Parkinson’s Disease to 132 controls. According to psychophysical testing with
Using olfactory and trigeminal ERPs, Barz et al. (1997) an odor identification test, olfactory function was slightly
investigated PD patients treated with anti-Parkinson drugs decreased in MND patients (compare Wenning et al.,
(n
13), PD patients who received no pharmacological 1995). OERP were found to be absent in 2 of 15 patients
treatment (n
18), and 38 age- and sex-matched controls. and delayed in 1 of 15 patients investigated.
Odor identification was impaired in PD patients and was
not influenced by treatment with anti-Parkinson drugs, in 5. Other Disorders
accord with earlier studies (e.g., Doty et al., 1988). OERP
Hummel et al. (1995) compared 12 temporal lobe epilepsy
latencies were prolonged in both PD patients taking and
patients with a left-sided focus to 10 such patients with a
not taking anti-Parkinson drugs (compare Hummel et al.,
right-sided focus. In both groups, longer ERP latencies
1993), although the effect was more pronounced in PD
were found following presentation of the trigeminal stim-
patients taking such drugs. In contrast to the OERP, the
ulant CO2 to the left naris than to the right naris. A differ-
intranasal chemosensory trigeminal system was affected
ent pattern emerged for olfactory stimuli. After right-side
neither by the neuronal degeneration seen in PD nor by
stimulation, latencies were prolonged in patients with
treatment with anti-Parkinson drugs (compare Hummel et
right-sided epileptic foci. After left-side stimulation,
al., 1993). Moreover, there is some suggestion that the
latencies were prolonged in patients with left-sided
OERP may be useful in assessing progression of PD, con-
epileptic foci (Fig. 5). Thus, the neocortical processing of
ceivably reflecting an association between a decreased
olfactory, but not trigeminal, information seems to be
number of neurons in the bulb and disease duration (Pearce
affected by functional lesions of the temporal lobe.
et al., 1995; Hummel, 1999). The observation that some
Moreover, analyses revealed nonoverlapping 95% confi-
patients exhibit normal odor identification test scores but
dence intervals for latency N1 when vanillin was applied
delayed OERPs suggests the possibility that OERPs may
to the right nostril. These results are in accord with the
be a more sensitive measure of subclinical dysfunction
notion, derived from other studies, that the right temporal
within the olfactory pathways than some psychophysical
lobe may play a different role in the processing of olfac-
tests (Hawkes et al., 1997).
tory information than the left temporal lobe. Finally, the
data indicated that olfactory information is predominantly
2. Alzheimer’s Disease
processed ipsilaterally to the stimulated nostril (see Doty
Despite considerable evidence that Alzheimer’s disease et al., 1997, for review).
(AD) is associated with smell loss at its earliest stages, OERPs have also been investigated in migraineurs
no formal studies of OERPs have been performed in AD (Grosser et al., 2000). Migraineurs have been found
patients to date. We have observed OERPs in a few AD to exhibit greater ERP amplitudes N1 to trigeminal
Electrophysiological Measurement of Olfactory Function 239

Figure 5 OERP in temporal lobe epilepsy (TLE). Twelve TLE patients with left-sided focus were compared with 10 TLE patients
with a right-sided focus. After right-sided olfactory stimulation latencies were prolonged in patients with right-sided epileptical foci.
When the left nostril was stimulated in patients with a left-sided focus, OERP latencies were prolonged. Latencies of responses induced
by stimulation on the nonaffected sides were in the range of normal controls. Also, the topographical distribution in TLE patients
was altered. While normals have their largest amplitudes of OERPs at Pz, patients had the largest amplitudes at the central site Cz. We
interpret this as a consequence of plasticity effects in TLE resulting in a change of orientation of the underlying equivalent current dipole
(see MSI).

stimulation, supporting the concept of trigeminal hyper- K. Employment of OERPs to Investigate the
excitability. In contrast, OERP amplitudes P1N1 were sig- Cognitive Processing of Odors
nificantly smaller in migraineurs. A leave-one-out
classification procedure on the basis of these two parame- As indicated above, one of the most important areas in the
ters assigned 76% cases correctly. The OERP amplitude application of OERPs is the investigation of the cognitive
discriminated better between groups than trigeminal ERPs, processing of odors. Early investigations of the late
emphasizing the significance of the olfactory system in positivity of the OERP were performed by Durand-
migraine. Other uses of OERP have been in the assessment Lagarde and Kobal (1991). Using the so-called oddball
of patients with multiple chemical sensitivities paradigm, stimuli were presented at an ISI of 6–8 sec such
(MCS)/idiopathic environmental intolerances (IEI) (Otto that a frequent stimulus (e.g., vanillin) alternated with a
and Hudnell, 1993; Dalton and Hummel, 2000; Hummel rare stimulus (e.g., H2S). Subjects were asked to count the
and Livermore, 2001), and Down syndrome (Wetter and occurrence of the rare stimulus (Fig. 6). As with other
Murphy, 1999). Further, OERPs have been used to address sensory modalities (Sutton et al., 1965), a late positivity
chemosensory changes during the course of a common occurred within the evoked potential in response to the rare
cold (Hummel et al., 1998b) or the investigation of drug stimulus (compare to Kobal and Hummel, 1991).
effects on the olfactory system, e.g., local application of Since this early work, a large number of studies has
decongestive agents in the nasal cavity (Hummel et al., focused on the late positivity of the OERP (for review, see
1998c) or the systemic administration of diazepam Pause and Krauel, 2000). Pause at al. (1996b) presented
(Hummel and Kobal, 1992). data suggesting that this component is modulated by
240 Kobal

Figure 6 Cognitive component. Examples from a single subject and a single patient with multiple chemical sensitivity. Late positive
component (P3) in the normal subject to the target stimulus (H2S) compared to the standard stimulus (phenyl ethyl alcohol) probabilities
were p (target)
0.2, p (standard)
0.8. In this case the late positive component of the patient was prolonged and different in shape.
Patients suffering from MCS complain about an unusual hypersensitivity for odors. This is not a result of a lowered olfactory threshold.
Hence, the earlier parts of the OERPs are not enhanced.

stimulus significance and stimulus probability. They found forward (e.g., Verleger, 1988), they all are based on the
that “the P3 component elicited by meaningful stimuli is view that P3 is largely a result of the cognitive processing
so large that the P2 component can be completely of stimuli.
overlapped.” When odorants are presented at ISIs of 30–40 Several major findings from this area of research are
sec, comparable to a single-stimulus paradigm (Polich worth noting: (1) the P3 component is larger when subjects
et al., 1994; Cass and Polich, 1997), it appears that the believe they have perceived the target odor, i.e., for both hits
olfactory stimulus reaches a significant “meaningfulness,” and false alarms (Pause et al., 1996b); (2) odors presented
which elicits a P3 component (Pritchard, 1981; Morgan only rarely elicit larger amplitudes than frequent odors,
et al., 1999) that determines the shape of the OERP and independent of stimulus quality (Krauel et al., 1998; Pause,
largely replaces the P2. In other words, what is frequently 1999), (3) the emotional significance of an odor may
addressed as a “P2 peak” in the OERP (Kobal and contribute to the generation of P3 (Pause et al., 1996b; Pause
Hummel, 1991) may resemble characteristics of a P300 et al., 1997; Krug et al., 2000); and (4) in analogy to
component (Lorig, 1993). According to a hypothesis research in ERPs in other sensory modalities (Spencer et al.,
elucidated by Donchin and Coles (1988), the P3 signifies 1999), the P3 component can be divided into a P3-1 and a
“context updating” related to the maintenance of the P3-2 peak, which differ in their topographical distribution
internal model of the external environment. From this per- (P3-1: fronto-central; P3-2: centro-parietal). P3-1 seems to
spective, the P3 amplitude would reflect the amount of be more related to the novelty and significance of an odor,
change of the model, while the latency would reflect the whereas P3-2 more resembles features of the classical P3,
duration of stimulus evaluation (compare Geisler et al., e.g., its latency relates to the time required for stimulus
1999a). Although alternative hypotheses have been put evaluation (Pause et al., 1996b; Pause and Krauel, 2000).
Electrophysiological Measurement of Olfactory Function 241

V. OLFACTORY MAGNETIC SOURCE IMAGING subjects’ brain and to check them for both anatomical and
physiological plausibility (e.g., Stefan et al., 1990). Aside
A. Description
from the aforementioned basic issues related to the physics
of magnetic fields, the reliability of the estimations is
The general goal of magnetic source imaging (MSI) is to
influenced by external error sources, e.g., magnetic noise
localize generators of magnetic fields measured at the sur-
caused by electronic devices or artifacts caused by move-
face of the scalp (Cohen, 1972). These generators may be
ments, heart activity, etc. (Abraham-Fuchs et al., 1988; for
the number of cerebral neurons (electrically) active at the
review see Hämäläinen et al., 1993).
same time. Unlike positron emission tomography (PET)
and functional magnetic resonance imaging (fMRI), MSI
allows direct assessment of the activity of the neurons B. Applications to Chemosensation
involved in the processing of sensory information.
Assuming that the human head is a spherical volume Employing a whole-head neuromagnetometer, Kettenmann
conductor (Abraham-Fuchs et al., 1988) and electrical et al. (1996) found bilateral activation in the superior
neuronal activity has the property of a current dipole pos- temporal sulcus at approximately 700 msec using vanillin,
sessing an electric field and a magnetic field, it becomes phenyl ethyl alcohol, and hydrogen sulfide—three sub-
possible to calculate the location, the orientation, and the stances with minimal CN V stimulative properties. Another
strength of the current sources creating the magnetic field study confirmed the results obtained with the whole-head
measured at the surface of the sphere (Romani et al., magnetometer by employing a planar 37-channel sensor
1982). In this model, not knowing the number of active array (Kettenmann et al., 1997b) and additionally identified
dipoles, the magnetic field may be generated by one dipole the neuronal generators underlying the earlier components
or several different ones (the “inverse problem,” first of the event-related potentials in the time interval between
discussed by von Helmholtz in 1853). Unfortunately, this 200 and 700 msec following olfactory stimulation.
is the situation in the active brain and therefore a unique Consistent magnetic fields were identified in both
solution cannot be obtained. Additionally, a radial dipole hemispheres following the stimulation of each nostril. In
does not produce a magnetic field that can be measured 60% of the measurements, reproducible dipolar field
with the usual order of the magnetometers. Such dipoles patterns were obtained 226–380 msec after stimulus
are “silent” and cannot be localized. onset, preceding or following the first major positive
On the basis of a measured magnetic field, a dipole posi- electric deflection P1 of the event-related potential
tion is estimated and followed by a nonlinear fit-strategy (ERP). This equivalent current dipole was named ECD I.
such as a Marquardt or Powell algorithm (Marquardt, 1963; In 44% of the measurements a reproducible dipolar
Powell, 1964). Using these procedures, location, strength, distribution was obtained 306–486 msec after stimulus
and orientation of the theoretically determined dipole is onset, corresponding to the ascending or descending
iteratively changed such that the magnetic field produced slope of the N1 component of the ERP, which was named
by this calculated dipole corresponds to the dipole ECD II. In the left hemisphere, this dipole was not
measured by the magnetometer. The statistical procedure identifiable in any of the subjects after stimulation with
leading to the so-called equivalent current dipole is a least- hydrogen sulfide. It was identifiable only in the right
squares search. As with most recording techniques, the hemisphere in 36% of the measurements. The most sta-
accuracy strongly depends on the signal-to-noise ratio of ble dipolar field pattern appeared 518–730 msec after
the measured data, which decreases with increasing depth stimulus onset (in 66% of the measurements; ECD III)
of the dipole within the brain. Also, localization accuracy is corresponding to the P2 component of the ERP (Fig. 7).
better for a dipole centered below the measurement grid Control measurements excluded the possible contamina-
than below the outer part of the measurement grid. With tion of the measured signals by tactile or auditory
regard to the situation in the human brain, the size of this artifacts. Humidified blanks delivered to the nostrils did
error can only be estimated by simulations (Barth et al., not result in MEG or EEG activity.
1986; Janday and O’Connell, 1987; Meijs et al., 1988; Hari In 14% of all the measurements, all three ECDs could
et al., 1988). It should be mentioned that brain-shaped head be identified during one session. ECD I was localized in
models, i.e., realistic head models, have also been used to the area between the superior temporal plane and the
calculate the source of the current dipole (Hämäläinen and parainsular cortex. ECD II was localized in the anterior-
Sarvas, 1987, 1989). central parts of the insula and ECD III was obtained in the
By linking the magnetically defined equivalent current superior temporal sulcus. Intraindividually, spatial differ-
dipoles to the anatomical data provided by MRI, it is ences of localization sites for one type of ECD were less
possible to visualize the location of activated areas in the than 20 mm. Individually, the angle of orientation varied
242 Kobal

Figure 7 Localization of olfactory activation in the temporal lobe and insula. Example from a single subject. Three main components
of the activity could be discriminated in the temporal plane (ECD I), in the insula (ECD II), and in the superior temporal sulcus (ECD
III). Other odorants, such as eugenol or hexenoic acid, activate more intensively the hippocampus and/or parainsular cortex.

between 10° and 30° in all three dimensions. The dipole (Mesulam and Mufson, 1982), in accord with evidence
strength varied between 0.009 and 0.03 mA*mm. from other functional imaging techniques.
In summary, localization results of the olfactory Only a few other groups have studied the olfactory
ERMFs demonstrated that odorants specifically activate system with MSI. Tonoike et al. (1998) employed a whole-
neocortical areas, i.e., during the period of time when the cortex 122-channel biomagnetometer and found genera-
ERP is obtained, areas that are activated include those tors of olfactory magnetic fields in two regions located
between the superior temporal plane and the parainsular fairly asymmetrically near the bilateral frontal deep areas.
cortex, anterior-central parts of the insula, and the superi- The results of Kettenmann and collagues were confirmed
or temporal sulcus. These electro(magneto)physiological by Sakuma et al. (1997) in a study of 14 subjects that used
data confirm the hypothesis that there is a direct connec- pulses of odorant air containing amyl acetate or phenyl
tion between primary olfactory areas and the insular cortex ethyl alcohol presented via a nasal tube Equivalent current
Electrophysiological Measurement of Olfactory Function 243

dipoles (ECDs) were estimated in the regions around the with the affected nostril even when the olfactory cleft
Sylvian fissure, symmetrically in both hemispheres. seems totally blocked with swollen mucosa, polyps,
mucus, or tumor. One possible explanation for such spared
sensory function would be the presence of functioning
VI. THE ELECTRO-OLFACTOGRAM olfactory neuroepithelium located anterior to the usual
boundaries of the olfactory cleft. Based on the topography
Even today the recording of electro-olfactograms (EOGs) of EOG recordings and histological and immunocyto-
from the human nasal mucosa is very difficult. The placing chemical evaluation of tissue from biopsy specimens,
of the electrodes is no easy task, since the intrusion of a Leopold et al. (2000) reported that the olfactory neuroep-
foreign matter into the nose very often leads to sneezing ithelium extends at least 1–2 cm anterior to the usually
and to excessive mucous discharge. Extensive local defined olfactory cleft. Other studies that have harvested
anesthesia has also to be avoided, since it might affect olfactory receptor neurons report that the neuroepithelium
olfactory fibers and render the subject temporarily anosmic may even extend to the anterior and middle parts of the
or hyposmic. This is probably the reason why so few middle turbinate (Restrepo et al., 1993; Thürauf et al.,
publications on this topic exist. 1996).
In an early experiment of two subjects in which coffee- Clearly, more work needs to be done before EOG
saturated air was used as the stimulant, Osterhammel et al. recordings can be meaningfully employed in clinical
(1969) discovered that the negativity recorded from the investigations or in a large numbers of subjects. Since the
olfactory mucosa increased in relation to an incremental EOG represents the input signal into the olfactory channel,
flow rate of the stimulus. Subsequently, Kobal (1981) its recording is fundamental for the interpretation of more
employed the stimulation method described earlier in this centrally generated olfactory responses (OERP, P300,
chapter (which eliminates flow rate-related mechanical or CNV, and subjective ratings) in order to define the site of
thermal artifacts) and the odorants amyl acetate, H2S, and modulatory influences, when phenomena such as adapt-
eugenol in a study of the EOGs of four subjects. The ation or habituation, anosmia or hyposmia, hypersensi-
responses, all of which characteristically showed negative tivity, or parosmia, etc. are investigated.
electrical potentials at the surface of the olfactory mucosa,
were dependent on the concentration of the stimulus.
When stimuli (e.g., H2S) of longer durations were applied, VII. MISCELLANEOUS TECHNIQUES
a temporal integration over a period of time of 10 sec was
observed, a phenomenon that has been reproduced No major progress has been made in recent years in the
(Hummel et al., 1996b). In response to long-duration recording of the contingent negative variation (CNV), the
stimuli, a decrease in intensity estimates was found to be spontaneous electroencephalogram (EEG), the psychogal-
two to three times greater than the decrease of EOG ampli- vanic skin response (PSR), and other types of reflexes.
tudes. Since the EOG reflects changes at the receptor level, Therefore, only studies related to background EEG activity
it can be assumed that the observed decrement in perceived are discussed in this section.
intensity reflects changes in central nervous processes There is a large body of literature on the use of the EEG
rather than to peripheral adaptation, a well-established in the quantification of human olfactory sensations. In the
phenomenon (see Chapter 10). In the olfactory literature 1950s and 1960s, the area of clinical applications was
such a reduction in odor intensity is generally called dominated by Italian scientists. Archilei and Moretti
adaptation. However, most of the time this concept is not (1958) investigated electroencephalographic changes in 30
distinguished from peripheral desensitization, which the subjects after presentation of odorous stimuli. As a rule,
author would prefer to call adaptation, and central desensi- they observed an arousal reaction in response to olfactory
tization, which the author would prefer to call habituation and trigeminal stimulation, i.e., slow EEG waves (theta-
(Thompson and Spencer, 1966). and alpha-band) were replaced by faster activity (beta-
Recently, Leopold et al. (2000) used EOG recordings to band). Similar findings were reported by Moncrieff (1962)
analyze the distribution of the olfactory epithelium. Up and by Motokizawa and Furuya (1973). These results were
that the time of this study, there was general agreement extended by Perbellini and Scolari (1966), who tested 50
that the olfactory epithelium is located high in the nasal patients using pyridine, vanillin, and “essence of rose.”
cavity, predominantly on the dorsal aspects of the nasal They concluded that the method would appear to be
vault, the septum, and the superior turbinate. However, particularly useful in the medico-legal field for the detec-
when caring for patients with obstructed nasal cavities, tion of deception. However, they also observed a number
there is some suggestion that some may be able to smell of cases where no arousal reaction could be recorded,
244 Kobal

although the subjects reported an olfactory sensation. insight into brain function that such pioneers as Berger
Thus, when using this technique only positive responses (1929) would have only dreamed of eight decades ago.
can be viewed as an unambiguous result. In earlier work,
Bartalena and Romeo (1962) had similarly noted, in 24
REFERENCES
subjects, that weak odorous stimuli were unaccompanied
by electroencephalographic changes. Abraham-Fuchs, K., Schneider, S., and Reichenberger, H.
Brandl et al. (1980) performed frequency analyses of (1988). MCG inverse solution: influence of coil size, grid
the EEG before and after administration of pleasant and size, number of coils, and SNR. IEEE Trans. Biomed. Eng.
unpleasant odorants. Significant correlations between a 35(8):573–576.
pattern of EEG spectra and the subject’s hedonic estimates Allison, T., and Goff, W. R. (1967). Human cerebral evoked
were noted. Yoshidan et al. (1989) have also reported, in a responses to odorous stimuli. Electroenceph. Clin.
preliminary study, that the alpha-band recorded at frontal Neurophysiol. 23(6):558–560.
sites appears to be an indicator of hedonic estimates of Archilei, G., and Moretti, E. (1958). Olfattometria ed elettoence-
olfactory stimuli. However, Kobal et al. (1989) were falografi. Valsalva 34:201–207.
unable to find changes of the stimulus-related EEG in rela- Auffermann, H., Gerull, G., Mathe, F., and Mrowinski, D. (1993).
Olfactory evoked potentials and contingent negative variation
tion to hedonic estimates of odorous stimuli, and Klemm
simultaneously recorded for diagnosis of smell disorders.
et al. (1992) reported no direct effect of pleasant or Ann. Otol. Rhinol. Laryngol. 102(1 Pt 1):6–10.
unpleasant odorants on the theta-band. Ayabe-Kanamura, S., Endo, H., Kobayakawa, T., Takeda, T., and
Lorig and Schwartz (1988) investigated intensity- Saito, S. (1997). Measurement of olfactory evoked magnetic
related effects of odorants on the EEG using topographical fields by 64-channel whole-head SQUID system. Chem.
maps of theta activity. They reported changes of scalp Senses 22:214–215 (abstract).
patterns when the subjects were exposed to varying odor Bartalena, G., and Romeo, G. (1962). Olfattometria ematogenata
intensities. Differences were greatest over temporo- elettroencefalograficain sogetti normali. Boll. Mal. O. G. N.
parietal recording positions. Similarly, Van Toller 80:14–23.
(1987) reported intensity-related differences, although he Barth, D. S., Sutherling, W., Broffman, J., and Beatty, J. (1986).
compared the mixed olfactory-trigeminal stimulant ammo- Magnetic localization of a dipolar current source implanted in
a sphere and a human cranium. Electroenceph. Clin.
nia to olfactory stimulants being perceived as less intense.
Neurophysiol. 63(3):260–273.
Moreover, Van Toller and Kendall-Reed (1989) reported Barz, S., Hummel, T., Pauli, E., Majer, M., Lang, C. J., and
differential effects for olfactory versus trigeminal stimuli. Kobal, G. (1997). Chemosensory event-related potentials in
Extending these findings in a follow-up study, Van Toller response to trigeminal and olfactory stimulation in idiopathic
et al. (1983) reported that odors can be distinguished by Parkinson’s disease. Neurology 49(5):1424–1431.
means of changes in the alpha activity recorded at differ- Bauer, L. O., and Mott, A. E. (1996). Differential effects of
ent sites. Interestingly, several authors have observed cocaine, alcohol, and nicotine dependence on olfactory
changes of the EEG in the presence of undetected odors evoked potentials. Drug Alcohol Depend. 42(1):21–26.
(Lorig et al., 1991; Klemm et al., 1992). Becker, E., Hummel, T., Piel, E., Pauli, E., Kobal, G., and
To summarize, analysis of the stimulus-related EEG is Hautzinger, M. (1993). Olfactory event-related potentials in
certainly of value in further understanding the relation of psychosis-prone subjects. Int. J. Psychophysiol. 15(1): 51–58.
Berger, H. (1929). Über das Elektroenkephalogramm des
olfaction to the general state of cortical activity, e.g., how
Menschen. Arch. Psychiatr. Nervenkr. 40:160–179.
odors may modulate states such as arousal, mood, etc. (see Brandl, U., Kobal, G., and Plattig, K.-H. (1980). EEG-correlates
Lorig, 1992). However, although the recording and analysis of olfactory annoyance in man. In Olfaction and Taste VII, H.
of stimulus-related EEG activity appears to be less difficult Van der Starre (Ed.). IRL Press, London, p. 401.
than, e.g., the recording of olfactory event-related poten- Cass, M., and Polich, J. (1997). P300 from a single-stimulus par-
tials, there is still little evidence that these measures are adigm: auditory intensity and tone frequency effects. Biol.
specifically related to activities in the olfactory system. Psychol. 46(1):51–65.
Cianfrone, G., and Subiaco, L. (1978). Olfattometria objettiva
computerizzata (a risposte evocate deficit). Les Cahiers D’O.
VIII. CONCLUSIONS R. L. 13:963–969.
Cohen, D. (1972). Magnetoencephalography: detection of the
At the present point in time, we are facing a new epoch of brain’s electrical activity with a superconducting magnetome-
functional investigations of the brain (see also Chapter 12). ter. Science 175(22):664–666.
Those who first started to use electrophysiological tech- Covington, J. W., Geisler, M. W., Polich, J., and Murphy, C.
niques have the feeling that their eyes have been uncovered (1999). Normal aging and odor intensity effects on the olfacto-
and that—although not yet clear—one has gained a new ry event-related potential. Int. J. Psychophysiol. 32(3):205–214.
Electrophysiological Measurement of Olfactory Function 245

Cui, L., and Evans, W. J. (1997). Olfactory event-related potentials Giesen, M., and Mrowinski, D. (1970). Klinische
to amyl acetate in congenital anosmia. Electroenceph. Clin. Untersuchungen mit einem Impuls-Olfactometer. [Clinical
Neurophysiol. 102(4):303–306. studies with an impulse-olfactometer]. Arch. Klein. Exp.
Dalton, P., and Hummel, T. (2000). Chemosensory function and Ohren Nasen Kehlkopfheilkd. 196(2):377–380.
response in idiopathic environmental intolerance. Occup. Grosser, K., Oelkers, R., Hummel, T., Geisslinger, G., Brune, K.,
Med. 15(3):539–556. Kobal, G., and Lötsch, J. (2000). Olfactory and trigeminal
Dalton, P., and Wysocki, C. J. (1996). The nature and duration of event-related potentials in migraine. Cephalalgia
adaptation following long-term odor exposure. Percept. 20:621–631.
Psychophys. 58(5):781–792. Hämäläinen, M. S., and Sarvas, J. (1987). Feasibility of the
Donchin, E. (1986). Cognitive Psychophysiology and Human homogeneous head model in the interpretation of neuromag-
Information Processing. Guildford Press, New York. netic fields. Phys. Med. Biol. 32(1):91–97.
Donchin, E., and Coles, M. G. H. (1988). Is the P300 a manifestion Hämäläinen, M. S., and Sarvas, J. (1989). Realistic conductivity
of context updating? Behav. Brain Sci. 11:357–428. geometry model of the human head for interpretation of neu-
Doty, R. L., Brugger, W. E., Jurs, P. C., Orndorff, M. A., Snyder, romagnetic data. IEEE Trans. Biomed. Eng. 36(2):165–171.
P., and Lowry, L. D. (1978). Intranasal trigeminal stimulation Hämäläinen, M., Hari, R., Ilmoniemi, R. J., Knutila, J., and
from odorous volatiles: psychometric responses from anosmic Lounasmaa, O. V. (1993). Magnetoencephalography—theory,
and normal humans. Physiol. Behav. 20(2):175–185. instrumentation, and applications to noninvasive studies of the
Doty, R. L., Deems, D. A., and Stellar, S. (1988). Olfactory working human brain. Rev. Mod. Phys. 65:413–498.
dysfunction in parkinsonism: a general deficit unrelated to Harada, H., Kato, T., Eura, Y. S. K., and Suoya, Y. (1977).
neurologic signs, disease stage, or disease duration. Topography of the olfactory evoked responses in patients with
Neurology 38(8):1237–1244. smell disorder. Chem. Senses 22:360 (abstract).
Doty, R. L., Bromley, S. M., Hummel, T. and Moberg, P. (1997). Hari, R., Joutsiniemi, S. L., and Sarvas, J. (1988). Spatial
Laterality in human nasal chemoreception. In Cerebral resolution of neuromagnetic records: theoretical calculations
Asymmetries in Sensory and Perceptual Processing, S. in a spherical model. Electroenceph. Clin. Neurophysiol.
Christman (Ed.). North Holland Publishing Co., Amsterdam, 71(1):64–72.
pp. 497–542. Hawkes, C. H. (1996). Assessment of olfaction in multiple
Durand-Lagarde, M., and Kobal, G. (1991). P300: a new technique sclerosis. Chem. Senses 21:486 (abstract).
of recording a cognitive component in the evoked potentials. Hawkes, C. H., and Shephard, B. C. (1998). Olfactory evoked
Chem. Senses 16:379 (abstract). responses and identification tests in neurological disease.
Evans, W. J. (1993). Suggestion for collection and reporting of Ann. NY Acad. Sci. 855:608–615.
chemoreception (olfactory) event-related potentials. Chem. Hawkes, C. H., Shephard, B. C., and Daniel, S. E. (1997).
Senses 18:751–756. Olfactory dysfunction in Parkinson’s disease. J. Neurol.
Evans, W. J., and Starr, A. (1992). Stimulation parameters and Neurosurg. Psychiatry 62(5):436–446.
temporal evolution of the olfactory evoked potentials in rats. Hawkes, C. H., Shephard, B. C., Geddes, J. F., Body, G. D., and
Chem. Senses 17:61–78. Martin, J. E. (1998). Olfactory disorder in motor neuron disease.
Evans, W. J., Cui, L., and Starr, A. (1995). Olfactory event- Exp. Neurol. 150(2):248–253.
related potentials in normal human subjects: effects of age Helmholtz, C. v. (1853). Über einige Gesetze der Verteilung
and gender. Electroenceph. Clin. Neurophysiol. elektrischer Ströme in körperlichen Leitern, mit Anwendung
95(4):293–301. auf die thierisch elektrischen Versuche. Ann. Phys. Chem.
Finkenzeller, P. (1965). Gemittelte EEG-Potentiale bei 89:353–377.
olfactorischer Reizung. Pflügers Arch. 292:76–85. Herberhold, C. (1976). Typical results of computer-olfactometry.
Fleischl von Marxow, E. (1890). Mitteilung betreffend der Rhinology 14(3):109–116.
Physiologie der Hirnrinde. Zbl. Physiol. 4:537–540. Hosoya, Y., and Yoshida, H. (1937). Über die bioelektrischen
Geisler, M. W., and Murphy, C. (2000). Event-related brain Erscheinungen an der Riechschleimhaut, Jpn. J. Med. Sci. III
potentials to attended and ignored olfactory and trigeminal Biophysics 5:22.
stimuli. Int. J. Psychophysiol. 37(3):309–315. Hughes, J. R., and Andy, O. J. (1979a). The human amygdala. I.
Geisler, M. W., Morgan, C. D., Covington, J. W., and Murphy, C. Electrophysiological response to odorants. Electroenceph.
(1999a). Neuropsychological performance and cognitive Clin. Neurophysiol. 46:428–443.
olfactory event-related brain potentials in young and elderly Hughes, J. R., and Andy, O. J. (1979b). The human amygdala. II.
adults. J. Clin. Exp. Neuropsychol. 21(1):108–126. Neurophysiological correlates of olfactory perception before
Geisler, M. W., Schlotfeldt, C. R., Middleton, C. B., Dulay, M. F., and after amygdalotomy. Electroenceph. Clin. Neurophysiol.
and Murphy, C. (1999b). Traumatic brain injury assessed with 46:444–451.
olfactory event-related brain potentials. J. Clin. Neurophysiol. Hughes, J. R., Hendrix, D. E., Wetzel, N., and Johnston, J. W.
16(1):77–86. (1969). Correlation between electrophysiological activity
Getchell, T. V., Margolis, F. L., and Getchell, M. L. (1984). from the human olfactory bulb and the subjective response to
Perireceptor and receptor events in vertebrate olfaction. Prog. odoriferous stimuli. In Olfaction and Taste III, C. Pfaffmann
Neurobiol. 23(4):317–345. (Ed.). New York, pp. 172–191.
246 Kobal

Hughes, J. R., Hendrix, D. E., Andy, O. J., Wang, C., Peeler, D., Hummel, T., Rothbauer, C., Pauli, E., and Kobal, G. (1998c).
and Wetzel, N. (1972). Correlation between electrophysiolog- Effects of the nasal decongestant oxymetazoline on human
icaland subjective responses to odorants as recorded from the olfactory and intranasal trigeminal function in acute rhinitis.
olfactory bulb, tract and amygdala of waking man. In Eur. J. Clin. Pharmacol. 54(7):521–528.
Neurophysiology Studied in Man, G. Somjen (Ed.). Hummel, T., Strehle, G., and Di Benedetto, M. (2000).
Amsterdam, pp. 260–280. Responses to olfactory and intranasal trigeminal stimuli: rela-
Hummel, T. (1999). Olfactory evoked potentials as a tool to tion to the respiration cycle. Chem. Senses 25:605 (abstract).
measure progression of Parkinson’s Disease. In Focus on Hummel, T., Heilmann, S., Murphy, C. (2001). Age-related
Medicine—New developments in the Drug Therapy of changes of chemosensory functions. In Olfaction, Taste, and
Parkinson’s Disease, Vol. 14, T. Chase, and P. Bedard (Eds.). Smell, C. Rouby (Ed.). Cambridge University Press,
Blachwell Science, Oxford, pp. 47–53. Cambridge (in press).
Hummel, T. (2000). Standards für die Ableitung chemosen- Ishimaru, T., Shimada, T., Sakumoto, M., Miwa, T., Kimura, Y.,
sorisch evozierter Potentiale zur klinischen Diagnostik von and Furukawa, M. (1997). Olfactory evoked potential produced
Riechstörungen. HNO 48:481–485. by electrical stimulation of the human olfactory mucosa. Chem.
Hummel, T., and Kobal, G. (1992). Chemosensory evoked Senses 22(1):77–81.
potentials. In Chemical Signals in Vertebrates, R. L. Doty, and Janday, B. S., and O’Connell, J. (1987). Locating current sources
D. Müller-Schwarze (Eds.). Plenum Press, New York, within skulls using magnetometry. Radiography
pp. 565–569. 53(612):287–291.
Hummel, T., and Kobal, G. (1994). Chemosensory event-related Kettenmann, B., Jousmaki, V., Portin, K., Salmelin, R., Kobal, G.,
potentials: effects of dichotomous stimulation with eugenol and Hari, R. (1996). Odorants activate the human superior
and dipydril. In Olfaction and Taste XI, K. Kurihara, temporal sulcus. Neurosci. Lett. 203(2):143–145.
N. Suzuki, and H. Ogawa (Eds.). Springer Verlag, Berlin, Kettenmann, B., Hummel, C., Stefan, H., and Kobal, G. (1997a).
pp. 659–663. Multiple olfactory activity in the human neocortex identified
Hummel, T., and Kobal, G. (2001). Olfactory event-related by magnetic source imaging. Chem. Senses 22(5):493–502.
potentials. In Methods & New Frontiers in Neuroscience, Kettenmann, B., Stefan, H., and Kobal, G. (1997b). Differences
Methods in Chemosensory Research. S. A. Simon (Ed.) CRC in magnetoencephalographically identified sources of cortical
Press, Boca Raton, FL, (in press). olfactory activity after stimulation with different odorants.
Hummel, T., and Livermore, A. (2001). Intranasal chemosensory Chem. Senses 22:217–218 (abstract).
function of the trigeminal nerve and aspects of its relation to Klemm, W. R., Lutes, S. D., Hendrix, D. V., and Warrenburg, S.
olfaction. Int. Arch. Env. Health (in press). (1992). Topographical EEG maps of human responses to
Hummel, T., Livermore, A., Hummel, C., and Kobal, G. (1992). odors. Chem. Senses 17:347–361.
Chemosensory event-related potentials in man: relation to Kobal, G. (1981). Elektrophysiologische Untersuchungen des
olfactory and painful sensations elicited by nicotine. menschlichen Geruchssinns. Thieme Verlag, Stuttgart.
Electroenceph. Clin. Neurophysiol. 84(2):192–195. Kobal, G. (1982). A new method for determiantion of the olfac-
Hummel, T., Kobal, G., and Mokrusch, T. (1993). Chemosensory tory and the trigeminal nerve’s dysfunction: olfactory (OEP)
evoked potentials in patients with Parkinson’s disease. In New and chemical somatosensory (CSEP) evoked potentials. In
Developments in Event-Related Potentials, H. J. Heinze, T. Event-Related Potentials in Children, A. Rothenberger (Ed.).
Münte, and G. R. Mangun (Eds.). Birkhäuser Verlag, Boston, Elsevier Biomedical Press, Amsterdam, pp. 455–461.
pp. 275–281. Kobal, G., and Hummel, C. (1988). Cerebral chemosensory
Hummel, T., Pauli, E., Schuler, P., Kettenmann, B., Stefan, H., and evoked potentials elicited by chemical stimulation of the
Kobal, G. (1995). Chemosensory event-related potentials in human olfactory and respiratory nasal mucosa.
patients with temporal lobe epilepsy. Epilepsia 36(1):79–85. Electroenceph. Clin. Neurophysiol. 71(4):241–250.
Hummel, T., Barz, S., Lötsch, J., Roscher, S., Kettenmann, B., Kobal, G., and Hummel, T. (1989). Brain response to chemical
and Kobal, G. (1996a). Loss of olfactory function leads to a stimulation of the trigeminal nerve in man. In Chemical
decrease of trigeminal sensitivity. Chem. Senses 21(1):75–79. Senses—Irritation, Vol. 2, B. G. Green, J. R. Mason, and M.
Hummel, T., Knecht, M., and Kobal, G. (1996b). Peripherally R. Kare (Eds.). Marcel Dekker, New York, pp. 123–139.
obtained electrophysiological responses to olfactory stimula- Kobal, G., and Hummel, T. (1991). Olfactory evoked potentials in
tion in man: electro-olfactograms exhibit a smaller degree of humans. In Smell and Taste in Health and Disease, T. V.
desensitization compared with subjective intensity estimates. Getchell (Ed.). Raven Press, New York, pp. 255–275.
Brain Res. 717(1–2):160–164. Kobal, G., and Hummel, T. (1992). Olfactory evoked potential
Hummel, T., Barz, S., Pauli, E., and Kobal, G. (1998a). activity and hedonics. In Fragrance: the Psychology and
Chemosensory event-related potentials change with age. Physiology of Perfume, T. S. Van, and G. H. Dodd (Eds.).
Electroenceph. Clin. Neurophysiol. 108(2):208–217. Elsevier Applied Science, London, New York, pp. 175–194.
Hummel, T., Rothbauer, C., Barz, S., Grosser, K., Pauli, E., and Kobal, G., and Hummel, T. (1998). Olfactory and intranasal
Kobal, G. (1998b). Olfactory function in acute rhinitis. Ann. trigeminal event-related potentials in anosmic patients.
NY Acad. Sci. 855:616–624. Laryngoscope 108(7):1033–1035.
Electrophysiological Measurement of Olfactory Function 247

Kobal, G., and Plattig, K.-H. (1978). Methodische Anmerkungen Lötsch, J., Ahne, G., Kunder, J., Kobal, G., and Hummel, T.
zur Gewinnung olfaktorischer EEG-Antworten des wachen (1998). Factors affecting pain intensity in a pain model based
Menschen (objektive Olfaktometrie). Z. EEG-EMG upon tonic intranasal stimulation in humans. Inflamm. Res.
9(3):135–415. 47(11):446–450.
Kobal, G., Hummel, T., and Pauli, E. (1989). Correlates of hedonic Marquardt, D. W. (1963). An algorithm for least-squares estima-
estimates in the olfactory evoked potential. Chem. Senses tion of nonlinear parameters. J. Soc. Ind. Appl. Math.
14:718 (abstract). 11:431–441.
Köster, E. P. (1965). Adaption, recovery and specificity of olfactory Mata, J., Jimenez, J., and Postigo, A. (1998). Potenciales evoca-
receptors. Rec. Laryngol. Otol. Rhinol. 86:880–894. dos corticales olfatorios. [Olfactory cortical evoked poten-
Köster, E. P., and de Wijk, R. A. (1991). Olfactory adaption. In tials]. Acta Otorrinolaringol Esp. 49(5):385–387.
The Human Sense of Smell, D. G. Laing, R. L. Doty, and W. Matern, G., Matthias, C., and Mrowinski, D. (1995).
Breipohl (Eds.). Springer-Verlag, Berlin, pp. 199–215. Olfaktorisch evozierte Potentiale (OEP) und Contingent
Krauel, K. (1999). Interstimulus interval and mismatch negativity Negative Variation (CNV) bei der Begutachtung von
in the chemosensory event-related potential. Poster at the 25th Riechstörungen. [Olfactory evoked potentials and contingent
meeting of the German Society for Psychophysiological negative variation in expert assessment of disordered sense of
Methods and Their Applications. smell]. Laryngorhinootologie 74(2):118–121.
Krauel, K., Pause, B. M, Sojka, B., Schott, P., and Ferstl, R. Meijs, J. W., Peters, M. J., Boom, H. B., and Lopes da Silva, F.
(1998). Attentional modulation of central odor processing. H. (1988). Relative influence of model assumptions and
Chem. Senses 23(4):423–432. measurement procedures in the analysis of the MEG. Med.
Krug, R., Plihal, W., Fehm, H. L., and Born, J. (2000). Selective Biol. Eng. Comput. 26(2):136–142.
influence of the menstrual cycle on perception of stimuli with Mesulam, M. M., and Mufson, E. J. (1982). Insula of the old
reproductive significance: an event-related potential study. world monkey. I. Architectonics in the insulo-orbito-temporal
Psychophysiology 37(1):111–122. component of the paralimbic brain. J. Comp. Neurol.
Lago, P., Broich, G., Cecchini, A. P., Sandrini, G., Guizzetti, G., 212(1):1–22.
Callieco, R., Zambarbieri, D., and Nappi, G. (1998). Mohammadian, P., Hummel, T., Lötsch, J., and Kobal, G. (1997).
Trigeminal evoked potentials in man: a new olfactory Bilateral hyperalgesia to chemical stimulation of the nasal
stimulation device. Funct. Neurol. 13(2):135–140. mucosa following unilateral inflammation. Pain
Leopold, D. A., Hummel, T., Schwob, J. E., Hong, S. C., Knecht, 73(3):407–412.
M., and Kobal, G. (2000). Anterior distribution of human Mohammadian, P., Schaefer, D., Hummel, T., and Kobal, G.
olfactory epithelium. Laryngoscope 110(3 Pt 1):417–421. (1999). Experimentally induced nasal irritation. Rhinology
Leplow, B. (1994). Objective assessment of “olfactory” and 37(4):175–178.
“trigeminal” perception of odors in patients with lesions of Moncrieff, R. W. (1962). Effect of odours on EEG records.
the olfactory system. J. Psychophysiol. 8:53–54. Perfum. Essent. Oil. Rec. 53:757–760.
Livermore, A., Hummel, T., and Kobal, G. (1992). Chemosensory Morgan, C. D., Covington, J. W., Geisler, M. W., Polich, J., and
event-related potentials in the investigation of interactions Murphy, C. (1997). Olfactory event-related potentials: older
between the olfactory and the somatosensory (trigeminal) males demonstrate the greatest deficits. Electroenceph. Clin.
systems. Clin. Neurophysiol. 83:201–210. Neurophysiol. 104(4):351–358.
Lorig, T. S. (1992). Cognitive and non-cognitive effecs of odor Morgan, C. D., Geisler, M. W., Covington, J. W., Polich, J., and
exposure: electrophysiological and behavioral evidence. In Murphy, C. (1999). Olfactory P3 in young and older adults.
Fragrance—The Psychology and Biology of Perfume, S. Van Psychophysiology 36(3):281–287.
Toller and G. H. Dodd (Eds.) Elsevier Applied Science, Motokizawa, F., and Furuya, N. (1973). Neural pathway associ-
London, pp. 161–173. ated with the EEG arousal response by olfactory stimulation.
Lorig, T. S. (1993). Event-related potentials to odor stimuli. Bull. Electroenceph. Clin. Neurophysiol. 35(1):83–91.
Psychon. Soc. 31:131–134. Murphy, C., Nordin, S., de Wijk, R. A., Cain, W. S., and Polich,
Lorig, T. S., and Roberts, M. (1990). Odor and cognitive alteration J. (1994). Olfactory-evoked potentials: assessment of young
of the contingent negative variation. Chem. Senses and elderly, and comparison to psychophysical threshold.
15:537–545. Chem. Senses 19(1):47–56.
Lorig, T. S., and Schwartz, G. E. (1988). Brain and Odor: II. EEG Murphy, C., Wetter, S., Morgan, C. D., Ellison, D. W., and
activity durng nose and mouth breathing. Psychobiology Geisler, M. W. (1998). Age effects on central nervous system
16:285–287. activity reflected in the olfactory event-related potential.
Lorig, T. S., Huffmann, E., DeMartino, A., and DeMarco, J. Evidence for decline in middle age. Ann. NY Acad. Sci.
(1991). The effects of low concentrationodors on EEG activity 855:598–607.
and behaviour. J. Psychophysiol. 5:69–77. Murphy, C., Morgan, C. D., Geisler, M. W., Wetter, S.,
Lorig, T. S., Matia, D. C., Peszka, J., and Bryant, D. N. (1996). Covington, J. W., Madowitz, M. D., Nordin, S., and Polich, J.
The effects of active and passive stimulation on chemosensory M. (2000). Olfactory event-related potentials and aging:
event-related potentials. Int. J. Psychophysiol. 23(3):199–205. normative data. Int. J. Psychophysiol. 36(2):133–145.
248 Kobal

Näätänen, R., Paavilainen, P., Tiitinen, H., Jiang, D., and Alho, K. Restrepo, D., Okada, Y., Teeter, J. H., Lowry, L. D., Cowart, B.,
(1993). Attention and mismatch negativity. Psychophysiology and Brand, J. G. (1993). Human olfactory neurons respond to
30(5):436–450. odor stimuli with an increase in cytoplasmic Ca2. Biophys. J.
Nagel, W. A. (1904). Einige Bemerkungen über nasales 64(6):1961–1966.
Schmecken. Ztschr. Psychol. 25:268. Romani, G. L., Williamson, S. J., and Kaufman, L. (1982).
Osterhammel, P., Terkildsen, K., and Zilstorff, K. (1969). Biomagnetic instrumentation. Rev. Sci. Instrum.
Electro-olfactograms in man. J. Laryng. 83:731–733. 53(12):1815–1845.
Otto, D. A., and Hudnell, H. K. (1993). The use of visual and Sakuma, K., Nakashima, K., and Takahashi, K. (1996). Olfactory
chemosensory evoked potentials in environmental and evoked potentials in Parkinson’s disease, Alzheimer’s disease
occupational health. Environ. Res. 62(1):159–171. and anosmic patients. Psychiatry Clin. Neurosci. 50(1):35–40.
Ottoson, D. (1956). Analysis of the electrical activity of the Sakuma, K., Kakigi, R., Kaneoke, Y., Hoshiyama, M., Koyama,
olfactory epithelium. Acta Physiol. Scand. Suppl. 122:1–83. S., Nagata, O., Takeshima, Y., Ito, Y., and Nakashima, K.
Pause, B. M. (1999). Body odor evoked potentials of self and (1997). Odorant evoked magnetic fields in humans. Neurosci.
non-self in humans. Genetica 104:285–294. Res. 27(2):115–122.
Pause, B. M., and Krauel, K. (2000). Chemosensory event-related Sato, M., Kodama, N., Sasaki, T., and Ohta, M. (1996). Olfactory
potentials (CSERP) as a key to the psychology of odors. Int. evoked potentials: experimental and clinical studies. J.
J. Psychophysiol. 36(2):105–122. Neurosurg. 85(6):1122–1126.
Pause, B. M., Sojka, B., Krauel, K., Fehm Wolfsdorf, G., and Silver, W. L. (1991). Physiological factors in nasal trigeminal
Ferstl, R. (1996a). Olfactory information processing during chemoreception. In Chemical Senses—Irritation, Vol. 2, B. G.
the course of the menstrual cycle. Biol. Psychol. 44(1):31–54. Green, J. R. Mason, and M. R. Kare (Eds.). Marcel Dekker,
Pause, B. M., Sojka, B., Krauel, K., and Ferstl, R. (1996b). The New York, pp. 21–37.
nature of the late positive complex within the olfactory event- Smith, D. B., Allison, T., Goff, W. R., and Principato, J. J. (1971).
related potential (OERP). Psychophysiology 33(4):376–384. Human odorant evoked responses: effects of trigeminal or
Pause, B. M., Sojka, B., and Ferstl, R. (1997). Central processing
olfactory deficit. Electroenceph. Clin. Neurophysiol.
of odor concentration is a temporal phenomenon as revealed
30(4):313–317.
by chemosensory event-related potentials (CSERP). Chem.
Spence, C., Kettenmann, B., Kobal, G., and McGlone, F. P.
Senses 22(1):9–26.
(2000). Selective attention to the chemosensory modality.
Pause, B. M., Krauel, K., Sojka, B., and Ferstl, R. (1999). Is odor
Percept. Psychophys. 62(6):1265–1271.
processing related to oral breathing? Int. J. Psychophysiol.
Spencer, K. M., Dien, J., and Donchin, E. A. (1999).
32(3):251–260.
Componential analysis of the ERP elicited by novel events
Pearce, R. K., Hawkes, C. H., and Daniel, S. E. (1995). The ante-
using a dense electrode array. Psychophysiology
rior olfactory nucleus in Parkinson’s disease. Mov. Disord.
36(3):409–414.
10(3):283–287.
Stefan, H., Schneider, S., Abraham-Fuchs, K., Bauer, J., Feistel,
Perbellini, D., and Scolari, R. (1966). L’elettroencefalo-olfat-
tometria. Contributo clinico. [Electroencephalo-olfactometry. H., Pawlik, G., Neubauer, U., Rohrlein, G., and Huk, W. J.
Clinical contribution]. Ann. Laringol. Otol. Rinol. Faringol. (1990). Magnetic source localization in focal epilepsy.
65(4):421–429. Multichannel magnetoencephalography correlated with mag-
Picton, T. W., and Hillyard, S. A. (1988). Endogenous event-relat- netic resonance brain imaging. Brain 113(Pt 5):1347–1359.
ed potentials. In EEG—Handbook, Vol. 3, T. W. Picton (Ed.). Sutton, S., Braren, M., and Zubin, J. (1965). Evoked-potential
Elsevier, Amsterdam. correlates of stimulus uncertainty. Science 150:1187–1188.
Polich, J., Eischen, S. E., and Collins, G. E. (1994). P300 from a Swandulla, D. (1986). Einige Aspekte der klinischen Anwendung
single auditory stimulus. Electroencephalogr. Clin. olfaktorisch evozierter Potentiale. FAU Erlangen-Nürnberg,
Neurophysiol. 92(3):253–261. Erlangen-Nürnberg.
Powell, M. J. D. (1964). An efficient method for finding the min- Tateyama, T., Hummel, T., Roscher, S., Post, H., and Kobal, G.
imum of a function of several variables without calculating (1998). Relation of olfactory event-related potentials to
derivatives. Computer J. 7:155–162. changes in stimulus concentration. Electroenceph. Clin.
Prah, J. D., and Benignus, V. A. (1992). Olfactory evoked Neurophysiol. 108(5):449–455.
responses to odorous stimuli of different intensities. Chem. Tecce, J. J. (1972). Contingent negative variation (CNV) and
Senses 17:417–425. psychological processes in man. Psychol. Bull. 77(2):73–108.
Prah, J. D., Sears, S. B., Walker, J. C. (1995). Modern approaches Thiele, V., and Kobal, G. (1984). Vergleich der objektiven and
to air dilution olfactometry. In R. L. Doty (Ed.). Handbook of subjektiven Methoden olfaktometrischer Bestimmungen—
Olfaction and Gustation, 1st Edition. Marcel Dekker, New Beispiel Schwefelwasserstoff. Schriftenreihe der
York, pp. 227–255. Landesanstalt für Immissionsschutz des Landes NW
Pritchard, W. S. (1981). Psychophysiology of P300. Psychol. 59:41–47.
Bull. 89(3):506–540. Thompson, R. F., and Spencer, W. A. (1966). Habituation: a
Reitan, R. M. (1971). Trail making test results for normal and model phenomenon for the study of neuronal substrates of
brain-damaged children. Percept. Mot. Skills 33(2):575–581. behavior. Psychol. Rev. 73:16–43.
Electrophysiological Measurement of Olfactory Function 249

Thürauf, N., Gjuric, M., Kobal, G., and Hatt, H. (1996). Cyclic associated with the odour of 5-alpha-androsten-3-one. Biol.
nucleotide-gated channels in identified human olfactory Psychol. 16(1–2):85–107.
receptor neurons. Eur. J. Neurosci. 8(10):2080–2089. Verleger, R. (1988). Event-related potentials and cognition: a
Tonoike, M., and Kurioka, Y. (1982). Precise measurements of critique of the context undating hypothesis and an alternative
human olfactory evoked potentials for odorant stimuli interpretation of P3. Behav. Brain Sci. 11:343–356.
synchronizied with respirations. Bull. Electroenceph. Lab. Walter, W. G., Cooper, R., Aldrige, V. J., McCallum, W. C., and
46:62–73. Winter, A. L. (1964). Contingent negative variation: an
Tonoike, M., Yamaguchi, M., Kaetsu, I., Kida, H., Seo, R., and electric sign of sensorimotor association and expectancy in
Koizuka, I. (1998). Ipsilateral dominance of human olfactory the human brain. Nature 203:380–384.
activated centers estimated from event-related magnetic fields Welge Lüssen, A. (1999). Chemosensorisch evozierte Potentiale.
measured by 122-channel whole-head neuromagnetometer Anwendung und Bedeutung im klinischen Alltag.
using odorant stimuli synchronized with respirations. Ann. NY [Chemosensory evoked potentials. Applications and signifi-
Acad. sci. 855:579–590. cance in routine clinical practice]. HNO 47(5):453–455.
Torii, S., Fukuda, H., Kanemoto, H., Miyanhi, R., Hamazu, Y., Welge Lüssen, A., Kobal, G., and Wolfensberger, M. (2000).
and Kawasaki, M. (1988). Contingent negative variation Assessing olfactory function in laryngectomees using the
(CNV) and the psychological effects of odour. In Perfumery, Sniffin’Sticks test battery and chemosensory evoked poten-
S. Van Toller, and G. H. Dodd (Eds.). Chapman and Hall, tials. Laryngoscope 110(2 Pt 1):303–307.
London, pp. 107–121. Wenning, G. K., Shephard, B., Hawkes, C., Petruckevitch, A.,
Van Toller, S. (1987). Brain electrical activity mapping during the Lees, A., and Quinn, N. (1995). Olfactory function in atypical
smelling of odours and trigeminal stimulants. Psychophysiology parkinsonian syndromes. Acta Neurol. Scand. 91(4):247–250.
24:619. Wetter, S., and Murphy, C. (1999). Individuals with Down’s
Van Toller, S., and Reed, M. K. (1989). Brain electrical activity syndrome demonstrate abnormal olfactory event-related
topographical maps produced in response to olfactory and potentials. Clin. Neurophysiol. 110(9):1563–1569.
chemosensory stimulation. Psychiatry Res. 29:429–430. Yoshidan, T., Saito, S., Iida, T., Yamamura, M., and Kanamura, S.
Van Toller, C., Kirk Smith, M., Wood, N., Lombard, J., and Dodd, (1989). Effect of odors on frequency fluctuation of brain-
G. H. (1983). Skin conductance and subjective assessments waves. Chem. Senses 14:311 (abstract).
 12

Functional Neuroimaging of Human Olfaction

Noam Sobel, Bradley N. Johnson, and Joel Mainland

University of California, Berkeley, California, U.S.A.

David M. Yousem
The Johns Hopkins University School of Medicine, Baltimore, Maryland, U.S.A.

I. INTRODUCTION rently still lagging behind invasive methods in both spatial

and temporal resolution, are mostly devoid of these draw-
Scientific advance is largely dependent on methodological backs, and therefore promise to be the method of the future
and technological advance. Neurobiology is currently in the study of neural systems.
experiencing a number of methodological advances of rev- The signal recorded with noninvasive functional neu-
olutionary proportions, not the least of which is the advent roimaging can be either the direct electrical product of
of methods for noninvasive functional neuroimaging. One neural activity or an indirect product of this neural activity,
may be able to indirectly infer a tremendous amount about such as metabolic products or blood flow. Different meth-
how a neural system functions by both observing and ods of neuroimaging have been developed to utilize these
quantifying its input (stimuli) and output (behavior). For different sources of signal, each offering specific advan-
the study of olfaction, the latter relates to olfactory psy- tages and disadvantages. The most commonly used of
chophysics (reviewed in Chapter 10 of this volume). But to these methods in humans are as follows:
completely understand the function of a neural system, one
1. Electroencephalography (EEG) and evoked poten-
must also have methods to directly measure and record its
tials (EPs). These methods directly measure neural
neural activity. For the case of olfaction, one would want
electrical activity. EEG and EPs can be noninva-
to reliably measure where in the brain olfactory informa-
sively obtained, and offer high temporal resolution.
tion is processed, and how this information is processed.
These methods, however, provide very poor spatial
resolution (the application of these methods to the
A. Functional Neuroimaging study of human olfaction is covered in detail in
Chapter 10 of this volume).
To date, most methods for measuring neural activity were 2. Magnetoencephalography (MEG). This method
invasive. The disadvantages of an invasive versus a nonin- measures the electromagnetic fields produced by
vasive method are numerous. An invasive method, by def- neural activity. MEG is noninvasive and offers
inition, (1) alters (usually harms) the system that it sets out high temporal resolution combined with moderate
to measure, (2) often involves anesthesia that in itself alters spatial resolution. Unlike the electrical currents
patterns of neural activity, (3) inflicts suffering on animals, recorded in EEG, the magnetic fields are not dis-
and (4) usually can not be used to study humans. Methods torted by brain, skull, and scalp. MEG, however,
of noninvasive functional neuroimaging, although cur- can only detect sources producing fields oriented

251
252 Sobel et al.

parallel to the surface of the skull. MEG also usu- Table 1 Range of Temporal and Spatial Resolution for
ally requires a concomitant MRI scan upon which Commonly Used Methods of Human Functional Neuroimaging
to register the MEG signals to anatomical space. Method Temporal resolution Spatial resolution
3. Single-photon emission computed tomography
(SPECT), positron emission tomography (PET), EEG ~1 msec ~10 mm
MEG ~1 msec ~5 mm
and magnetic resonance spectroscopy (MRS).
SPECT 60 sec ~6 mm
These three methods measure the biochemical
PET ~45 sec ~4 mm
components of neural transmission, for example, fMRI ~3–5 sec ~1 mm
the distribution and density of a particular labeled
neurotransmitter, metabolite, blood flow, or glu- Source: Adapted from Volkow et al., 1997.
cose utilization. These methods currently offer
rather poor temporal and spatial resolution but
give access to a type of information not available and disadvantages of each method. We then discuss the
with other imaging methods, namely the highly application of these methods to the study of human olfac-
sensitive measuring of biochemical processes—a tion, reporting on the current state of the art. Considering
PET camera can detect picomolar changes in that functional imaging of olfaction is in its infancy, we
labeled compounds. Both PET and SPECT are will relate in this chapter to work using all of the above
somewhat invasive in that they require administer- methods. We will, however, concentrate on work using the
ing to subjects a radioactive isotope and exposure blood flow–dependent methods of PET and fMRI, as those
to ionizing radiation. MRS is noninvasive, but the are the most commonly used methods, and the latter is also
application of this method to humans is only in its the method currently used by these authors.
early stages. PET and SPECT can also be used as
an indirect measure of regional levels of neural B. A Brief History of Blood Flow Neuroimaging
activity through measuring either glucose metabo-
lism or blood flow. One of the main advantages of The foundations of modern blood flow–dependent neuro-
PET is the absence of ferromagnetic or suscepti- imaging lay in the careful observations made in 1885 by
bility artifacts at the skull base which plague the Italian physiologist Angelo Mosso (reviewed in Posner
fMRI studies of the olfactory system. Structural and Raichle, 1998; Raichle, 1998; Volkow et al., 1997). As
MRI scans often accompany PET and SPECT anyone can observe by feeling the undeveloped skull of a
studies to provide a more detailed anatomical newborn, the brain appears to pulsate in unison with heart
localization of activity. rate and respiration. This pulsation is later obscured by the
4. Functional magnetic resonance imaging (fMRI). adult developed skull. Mosso had the opportunity to study
This method is a relatively recent modification to this phenomenon in an adult patient named Bertino, who,
high-resolution structural magnetic resonance due to head injury, had a skull defect that enabled simple
imaging (covered in Chapter 28 of this volume) that recording and quantifying of this pulsation over the frontal
provides an indirect measure of neural activity lobes of the brain. Mosso noted that the rate of pulsation
through measuring blood flow. fMRI has high spa- on the surface of Bertino’s brain increased during the per-
tial resolution, fair temporal resolution, and is formance of specific cognitive tasks, such as performing a
totally noninvasive. It does not require any other mathematical multiplication task. Furthermore, this
concomitant scanning, so intermodality registration increase in rate of pulsation occurred independently of
issues are not a problem. It is these advantages— heart rate and blood pressure as measured on Bertino’s
combined with relatively high and rapidly growing arm. These findings suggested that measuring brain blood
availability—that have made fMRI the most rapidly flow might serve as a method to measure brain activity.
developing method of functional neuroimaging. Mosso’s measurements in a human were later corrob-
Table 1 provides a comparison of temporal and spa- orated by measurements Roy and Sherington made in
tial resolution of the above widely used methods of animals (1890). Roy and Sherington describe a mecha-
human neuroimaging. nism that automatically modulates the supply of blood to
local areas of increased activity in the brain. This rela-
In this chapter we aim to acquaint the nonspecialist tionship between local increases in brain blood flow and
with the above modern methods of human functional local increases in neural activity was further solidified in
neuroimaging, to briefly explain the source of the meas- a second famous early human case study published by
ured signal, and to stress some of the potential advantages John Fulton in 1928. Fulton reported on a patient known
Functional Neuroimaging of Human Olfaction 253

as Walter K., in whom a failed attempt at correcting an The period that is best mapped is typically one minute fol-
arteriovenous malformation in the occipital lobe resulted lowing injection, and a typical PET experiment may con-
in a skull defect overlying the primary visual cortex. As sist of 12 such injections. The resulting blood flow map
the patient himself could later notice, the increased sound can then be overlaid on a structural image of the brain,
of blood flowing through the blood vessels could be such as that obtained with MRI. The above describes the
heard from this location during performance of specific use of PET to follow blood flow as a measure of neural
visual tasks. The physician could also hear, and record, activity. Other tracers are commonly used in PET to follow
this sound, simply by placing a stethoscope over the skull glucose metabolism as a measure of neural activity.
defect. Fulton reports that when the patient opened his
eyes, there was only a moderate increase in the sound of D. fMRI
blood flow. However, when the patient engaged in a
demanding visual task, such as reading, there was a much Like PET, fMRI is also an indirect index of neural activity
larger increase in the sound of blood flow. The sound that measures blood flow, but rather than depend on an
would reliably increase 20–30 seconds after the begin- introduced tracer or contrast agent, fMRI depends on an
ning of a task and would return to baseline within 2 min- intrinsic contrast agent—hemoglobin. The principals that
utes. The above pioneering studies in which the underlie MRI were discovered independently by Block
relationship between local increases in neural activity (1946) and Purcell and colleagues (1946) and were devel-
and local increases in blood flow was established laid the oped into an imaging method by Lauterbur (1973)
foundation for the later development of PET and fMRI (reviewed in Raichle, 1998). In an fMRI experiment, the
functional neuroimaging techniques. subject is placed within a strong static homogeneous mag-
netic field. In such a field, the nuclei of elements in tissue
C. PET that have an odd atomic weight, such as hydrogen, behave
as little spinning magnets that tend to align their macro-
Developments in animal autoradiography made by Kety, scopic spin axes with that of the external field. A brief
Sokoloff, Landau, and colleagues (Landau et al., 1955) pulse of radio frequency magnetic fields can then be used
enabled the quantification of the relationship between glu- to tip the orientation of the spinning protons/magnets away
cose metabolism, blood flow, and neural activity. These from that of the external field. A given radio-frequency of
measurements enabled Lassen and colleagues to develop pulse will tend to selectively tip a given atomic species.
the application of an array of scintillation detectors to Because of the preponderance of protons and their high
measure regional changes in blood flow in the human brain relative sensitivity, fMRI experiments selectively excite
(Ingvar and Risberg, 1965; Lassen et al., 1978). These the protons in water. Following the pulse, these protons
functional measurements of regional blood flow were later will then tend to realign in parallel with the external field
combined with structural x-ray computed tomography (relax), while emitting energy at the same frequency as the
(Hounsfield, 1973) to result in the modern PET scan excitation. This emitted energy is the source of the signal
(Frackowiak et al., 1980; Raichle et al., 1983; Ter- in MRI. The strength of the emitted energy increases with
Pogossian et al., 1975). The PET scan is dependent on the strength of the static magnetic field. The relaxation of
the radioactive decay of positrons emitted from unstable the protons orientation back to that of the static field is
neutron-deficient atoms such as 15O. At the beginning of a referred to as T1 relaxation, and the relaxation of proton’s
PET experiment, this, or another, unstable tracer is injected spin from the orientation transverse to the static field is
into the subjects’ bloodstream. The tracer accumulates in referred to as T2 relaxation. A third type of relaxation,
the brain at concentrations directly proportional to regional which takes into account tissue and field inhomogenities,
blood flow, that is, regions with increased blood flow have is referred to as T2*. These differences in relaxation time
higher concentrations of the tracer. The unstable tracer are the source of contrast in the structural MR image.
then decays (the half-life of 15O is 123 seconds), emitting But how do such differences in relaxation time give rise
positrons that are annihilated by negatively charged elec- to a neurofunctional measure? As described by Fox et al.
trons present in tissue. The energy from this annihilation is (1986, 1988), local changes in neural activity induce local
emitted in the form of two photons that leave the brain in changes in the amount of oxygen in tissue or, more
exactly opposite directions from the point of annihilation. specifically, in the ratio of oxyhemoglobin to deoxyhemo-
By surrounding the brain with an array of radiation detec- globin. This change in ratio is counterintuitive in its direc-
tors coupled through coincidence circuits, it is possible to tion. One might predict that a local increase in neural
detect and map the exact location of the annihilation in the activity would induce a local increase in neurometabolic
brain, or in other words, create a map of brain blood flow. products, namely deoxyhemoglobin. But the dynamics of
254 Sobel et al.

neural activity–induced blood flow are such that the local list of questions on localization of olfactory function that
ratio of oxyhemoglobin to deoxyhemoglobin in fact can be addressed using functional neuroimaging, but the
increases by a few percent over baseline. This finding, following are examples of the type of questions that can be
combined with a previous finding by Pauling and Coryell addressed:
(1936) that changing the amount of oxygen carried by
1. Piriform cortex, one of the main components of the
hemoglobin changes the degree to which hemoglobin dis-
primary olfactory cortex (Price, 1990), is a cytoar-
turbs the magnetic field, suggested that MR relaxation
chitechtonic definition. Therefore, piriform cortex
times, specifically T2* relaxation, would differ in accor-
cannot be delineated in vivo based on structural
dance with the amount of deoxyhemoglobin in blood. Such
imaging information alone. Functional imaging,
use of MRI to measure sensory task–induced local changes
however, may enable in vivo delineation of human
in brain blood flow in humans was first successfully
piriform cortex based on odorant-induced activa-
reported by Ogawa and colleagues (1992) and has since
tion. A question such as this can best be addressed
become standard practice. This method is referred to as
using fMRI, thanks to its high spatial resolution.
fMRI, and the signal is referred to as BOLD (blood oxy-
2. Secondary olfactory cortex is considered to reside
gen level dependent). Other fMRI methodologies exist that
primarily within the orbitofrontal gyri of the ventral
utilize gadolinium-based contrast agents to measure blood
frontal lobe. Which of the four gyri that primarily
flow or tag arterial blood with magnetic pulses to demon-
comprise the orbitofrontal complex is indeed part
strate increased flow to an activated state.
of human olfactory cortex, and the specific func-
tional role played by this cortical region in olfac-
II. APPLYING IMAGING TO THE STUDY OF tory processing, are currently not well understood.
OLFACTION: WHAT QUESTIONS CAN BE Questions such as these could best be addressed
TESTED WITH THESE METHODS? using either fMRI or PET.
3. What other brain regions are involved in olfaction,
Functional neuroimaging is primarily used to address ques- and what are their respective contributions to olfac-
tions within two major frameworks: questions on broad tory function? Whereas the previous localization
localization of function and questions on temporal and spa- question consists of refining a crude but existing
tial properties of function. The distinction between a broad picture of the human olfactory neural substrates,
question on localization of function versus a question on functional imaging can also be used to probe for
spatial properties of function can best be delineated by a novel regions of activity, namely, regions not previ-
relevant example: asking where in the brain is primary ously known to subserve olfaction. Questions such
olfactory cortex is an example of the former, whereas ask- as the above could best be addressed using PET,
ing is there a spatial component in the encoding of odors in where whole brain data are regularly obtained
primary olfactory cortex is an example of the latter. This (whole brain fMRI has also been studied using
distinction represents more than just a difference in sensi- thicker and more numerous anatomical sections,
tivity and scale of measurement; it is asking “where” versus but the fine anatomy of the primary olfactory cortex
asking “how.” That said, this distinction does not represent may be obscured when looking more globally).
mutually exclusive types of information, as indeed part of
understanding “how” a neural subsystem functions is
understanding “where” in the brain it functions. B. Questions on Properties of Function

Functional neuroimaging can be used to address questions

A. Questions on Broad Localization of Function:
regarding both the spatial and temporal properties of
Functional Mapping of the Human Olfactory
neural activity. The following are examples of the types of
System
questions that can be addressed:
Mapping consists of elucidating which brain regions are 1. Is there a spatial component in the encoding of odor
primarily responsible for carrying out specific tasks. This identity at the cortical level? A consistent strategy
type of question represents a dominant paradigm in neuro- employed by the brain to encode sensory informa-
science prevalent since the work of Franz Joseph Gall and tion is spatial mapping. The body surface is
Paul Broca (reviewed in Harrington, 1995; Harris, 1995). somatotopically represented in somatosensory cor-
This paradigm has dominated the field of human func- tex. Auditory frequencies are tonotopically mapped
tional neuroimaging. We can not here offer an exhaustive on the cochlea, and this spatial organization is
Functional Neuroimaging of Human Olfaction 255

maintained in auditory cortex (Gulick et al., 1989). as what is the structural and functional neuroarchi-
Visual representations are retinotopically organized tecture that underlies the special interaction
on the retina, and this spatial organization is between odor and mood, and odor and memory.
maintained in primary visual cortex (Holmes, Questions such as this could best be addressed
1918; Horton and Hoyt, 1991). Although most using either fMRI or PET.
cytoarchitectual evidence suggests no odortopically
organized projection from the bulb to primary
olfactory cortex (Price and Sprich, 1975), recording III. TECHNICAL CONSIDERATIONS IN
of activity in the rat piriform cortex made with opti- FUNCTIONAL NEUROIMAGING OF
cal imaging suggests that the piriform is divided HUMAN OLFACTION
into several functionally heterogeneous regions
(Litaudon et al., 1997). Whereas odortopy is possi- The application of functional neuroimaging to the study of
ble but unlikely in primary olfactory cortex, it has human olfaction consists of an unfortunate compounding
been in fact demonstrated to some extent in of the technical difficulties inherent to both the field of
secondary olfactory cortex in monkeys (Tanabe olfaction and the field of neuroimaging.
et al., 1975). The nature and rules underlying this
secondary cortical odortopy are currently unknown. A. Stimulus Generation
Functional neuroimaging in humans promises to be
an ideal tool to address this issue, because when Generating olfactory stimuli is a complex task as it is, but
studying the relationship between patterns of brain this complexity is increased in the MRI environment,
activity and odorants in humans rather than ani- where one cannot introduce ferrous materials due to the
mals, there is the added dimension of the percept strong magnetic field. Thus, valves, canisters, tubing and
that humans can readily verbally report. Questions piping, etc., that are part of the olfactory stimulus–gener-
such as the above could best be addressed with ating apparatus must all be either nonferrous or at a dis-
fMRI thanks to its high spatial resolution. tance from the subject. Aside from this added
2. What are the temporal properties of odor encoding complication, the rules that apply to stimulus generation
at various levels of processing? Temporal param- in the imaging environment are similar to those in other
eters that can be probed range from standard mea- experimental designs (see Chapters 10 and 11) (see also
sures, such as latency and duration of responses, to Prah et al., 1995). Because imaging studies usually con-
complex measures such as potential information- sist of comparing two olfactory conditions (e.g., odorant
encoding (odorant encoding) in the phase and/or versus no odorant, high concentration versus low concen-
ordering of neural activity. Temporal measures can tration, etc.), it is imperative that the difference between
also be used to probe the nature of interaction these conditions is restricted to the process of interest
between two or more regions processing olfactory alone. Any additional sources of variance, such as valve
information simultaneously. Questions such as the noise, airflow rate change, airflow temperature change,
above could best be addressed using either MEG or humidity change, that are associated with one condition
single-trial fMRI thanks their high temporal resolu- and not the other may contaminate the imaging result. (A
tion. detailed description of simple, yet sufficient, olfactome-
3. What are the differences in neural processing that ters for fMRI studies can be found in Sobel et al., 1997,
correlate with differences in performance? This and Lorig et al., 1999.)
may include elucidating the neural substrates The necessary quality and accuracy of the stimulus-
underlying differences in olfactory performance generating equipment to be used in any given study is
between groups such as men and women. In addi- related to the type of questions asked. For example, con-
tion, functional neuroimaging may be used to probe sidering the questions previously described, one may study
patients with abnormalities of the olfactory system, the emotional response to an odor using relatively simple
be it anosmia, hyposmia, parosmia, olfactory hallu- methods of odorant generation; however, to study the tem-
cinations, or hyperosmia. In this realm fMRI is par- poral aspects of the neural response to odors, one must use
ticularly useful because of the structural a higher quality stimulus-generating device. In this regard,
discrimination it provides, as well as its sensitivity one of the major current controversies regarding the
to pathological alterations in anatomy. methodology of imaging human olfaction is whether sub-
4. Functional neuroimaging can be used to probe jects should sniff or not sniff during the experiment. The
questions regarding higher olfactory function, such advantages of not sniffing are in reducing the risk of
256 Sobel et al.

motion-related artifacts as well as reducing the non–odor- obscure transient activity at the beginning of odorant expo-
ant-related sources of activation, such as the somatosen- sure. These problems are reduced if multiple odorants are
sory or motor components of the sniff. In turn, natural used during a single scan period, but this method can prove
olfaction consists of sniffing. The sniff is in itself part of problematic due to inconsistencies in psychoperceptual
the olfactory percept, modulating patterns of activity in characteristics of different odorants. Adaptation may still
olfactory cortex, as well as the quantity (Laing, 1983; occur even with multiple odorants in the form of cross-
Sobel et al., 2000a) and type (Sobel et al., 2000b) of odor- adaptation, whereby exposure to one odorant decreases the
ant sampled. Thus, to fully characterize olfactory process- response to a second odorant (Cometto-Muñiz and Cain,
ing, ideally both sniffing and nonsniffing paradigms 1995; Köster and de Wik, 1991) (see Chapter 10). The
should be employed, with close monitoring of respiration. advantage of a long temporal window is that precise tim-
In the nonsniffing experiments, such monitoring is neces- ing of odorant presentation is less important. While ERP,
sary to assure that odorants did not induce an unwanted MEG, and some fMRI paradigms require sophisticated
automatic respiratory response (Jackson, 1976; Warren olfactometer presentation, less precise stimulus presenta-
et al., 1992, 1994), and in sniffing experiments such mon- tion techniques can be used with PET.
itoring is necessary so that the sniff itself can later be fac- The sensation of smell is often the result of stimulation
tored out in the statistical analysis of the functional in more than just the olfactory nerve. For example,
imaging data. A procedure that achieves this is as follows: trigeminal nerve stimulation also contributes to the sensa-
subjects are instructed to sniff by either a visual or auditory tion of some smells (reviewed in Chapter 47 of this vol-
cue that appears at a predetermined rate. Subjects are fur- ume). For this reason, and in order to elucidate the
ther instructed to maintain their sniff for the duration of the separate neural substrates that may underlie trigeminal
projected message or tone, typically set to 800 msec. Thus, versus pure olfactory perception, it is imperative that
sniffing can be equalized for all conditions, e.g., sniffs of experimenters be aware of the extent of trigeminal stimu-
odorant versus sniffs of no odorant. Finally, the olfac- lation potentially present in the stimuli used. Doty and
tometer can be coupled to a pneumatotachograph offering colleagues (1978) studied congenitally anosmic patients
a precise measurement of airflow rate, duration, and vol- to identify stimuli that can be detected by these patients
ume of each sniff during the entire experiment. This and thus presumably have trigeminal components.
enables both post hoc and on-line assurances that the sniffs Yousem and colleagues confirmed the results of this
were indeed equal across all conditions and can thus be behavioral approach using functional neuroimaging.
subtracted as a factor in later statistical analysis. In sniff- When they tested congenital anosmics with fMRI and
ing designs, the use of a bite bar to prevent sniff-related olfactory nerve stimulants, no activation was seen, and
head motion is imperative. when anosmic patients with Alzheimer’s disease were
An alternative odor presentation paradigm, suggested given olfactory nerve odorants, again no stimulation was
by Cerf-Ducastel and Murphy (2001), avoids many of the observed (Yousem et al., 1997a, 1998).
logistical challenges presented by using an olfactometer
in the MRI environment. They presented odorants in C. Imaging Parameters
aqueous solution to the oral cavity and showed that corti-
cal activation via retronasal olfaction is qualitatively simi- It is an unfortunate coincidence that the olfactory regions
lar to activation due to airborne stimuli. This method may of the brain, namely the ventral portions of the temporal
be especially suitable for studying flavor, or for direct and frontal lobes, are the regions most susceptible to arti-
comparisons of olfaction and gustation. fact in fMRI. This is because the MRI image is highly sus-
ceptible to distortion at sharp borders of signal intensity.
B. Choice of Odorants A common source for such sharp borders is the interface of
tissue to air-filled cavities, and such cavities abound
Considering that functional neuroimaging studies consist around the olfactory regions of the brain (i.e., the paranasal
of prolonged exposure to odorants, odorants with minimal sinuses, the petrous apices, and the temporal bones).
toxicity should be chosen. To this end, employing odorants Various strategies can be employed to address this com-
that have been safely used for extensive periods of time in mon source of artifact (Yang et al., 1997), and here we will
human psychophysical studies would seem preferable. The describe only a few of the very basic imaging parameters
latter strategy will also readily enable comparison of that are conducive to minimizing such artifacts. First, it is
results across studies and methods. advised to use as thin a slice as possible during acquisition
Studies of areas sensitive to rapid habituation must also of the data. This reduces the probability of averaging areas
take into account the fact that long temporal windows may of very different signal intensity within the same voxel.
Functional Neuroimaging of Human Olfaction 257

minimizing artifacts in the ventral temporal region are only

the very first basic approach and should be augmented by
various methods of image postprocessing (e.g., Yang et al.,
1997).
Another technique for performing fMRI for olfaction
has been advocated by Levy and colleagues. They use a
spoiled T1-weighted gradient echo (FLASH) scan utilizing
gadolinium injections to demonstrate the increased blood
flow to the activated area. This is analogous to a perfusion
scan; however, the T1-weighting, ultra-short echo time,
and spoiling reduces the susceptibility artifact that inter-
feres with BOLD fMRI at the skull base (Levy et al., 1997,
1998 a,b, 1999). The limitation of this technique is that
very few slices are allowed per acquisition due to longer
scan times when not using echoplanar or spiral imaging.
Regarding comparison of imaging methods, it is important
to note that PET is not susceptible to these artifacts in the
ventral portions of the brain.

D. Experimental Design and Statistical Analysis

There are currently two standard designs for functional

Figure 1 The olfactory slice: the oblique orientation at which neuroimaging experiments: the block design and the sin-
we recommend obtaining functional MR data from the olfactory gle-trial design (also referred to as event-related design)
system. Obtaining data at this orientation reduces ventral-tempo-
(Fig. 2). The block design consists of extended alternating
ral–related artifacts. The data can then be displayed at this, or any
other, orientation. epochs of a control and experimental condition, for exam-
ple, an odorant condition versus a no-odorant condition.
This type of design is best suited for methods with rela-
tively low temporal resolution, such as PET. A typical PET
This is advised despite the signal-loss attributed to using study may consist of two blocks, each lasting a minute. A
thinner slices. Second, it has proven helpful to acquire the typical fMRI block design study may consist of blocks
data at an oblique orientation such as seen in Figure 1. For ranging between 20 and 40 seconds in duration that are
a standard coronal or horizontal acquisition to contain the repeated four to six times within an experiment. In con-
olfactory regions, it would necessarily also contain the trast, a single-trial design consists of random presentations
paranasal sinuses and temporal bone. By contrast, this of brief experimental and control epochs. This type of
oblique orientation enables covering the entire olfactory design is best suited for methods with higher temporal
regions, with only minimal coverage of these bony and air- resolution such as MEG or fMRI. A typical fMRI single-
filled structures. An added advantage of this slice orien- trial study may consist of 40 experimental and 40 control
tation is that both primary and secondary olfactory cortex stimuli, with stimuli presented at 16-second intervals in a
appears within the same slice, which is very convenient for randomized fashion. The single-trial design has several
both analysis and presentation. Finally, what has proved as advantages over the block design in the study of human
helpful in some efforts at imaging the olfactory regions is olfaction. In the single-trial design, one negates the poten-
using a specific method, or pulse sequence, for sampling the tial expectation of an upcoming experimental (odorant)
three-dimensional imaging space (referred to as k-space). condition, and more importantly, one can minimize the
Specifically, the commonly used fMRI method referred to habituation that is an inevitable occurrence during a pro-
as echo-planer imaging (EPI) samples the imaged space in longed odorant epoch in a block design experiment.
a zigzag trajectory. An alternative method, referred to as Once functional imaging data are obtained using either of
the spiral trajectory, or spiral sequence (Glover and Lai, the above experiment designs, they are subjected to statistical
1998), samples the imaged space in a spiral fashion. For analysis. This analysis typically consists of parsing the data
reasons beyond the scope of this chapter, the spiral into a matrix of voxels representing the anatomical area that
sequence appears to be less susceptible to the artifacts typ- was imaged. The time course of activity, or signal, in each
ical to the ventral temporal region. The above strategies for such voxel is then correlated with the time course of the task
258 Sobel et al.

Figure 2 Typical time-course of odorant generation. Whereas a block design experiment consists of alternating extended epochs of
odorant presence and odorant absence, the single trial design consists of random presentations of brief odorant or no-odorant epochs.

(e.g., odorant presence vs. odorant absence). Voxels in which than structural, neuroimaging. We will concentrate on studies
the frequency of the signal is significantly correlated with the addressing properties of basic sensory processing. For func-
frequency of the task are considered to be involved in the tional neuroimaging studies of higher olfactory processing,
neural processing underlying that task. For purposes of dis- such as odor memory or semantic processing of odors, we
play, such voxels are commonly color-coded and overlaid on refer the interested reader to the work of Royet et al. (1999),
a gray-scale anatomical image in order to depict the region of Dade et al. (1998, 2001), and Savic et al. (2000). For a review
activity. In cases where only one experimental and one con- on functional imaging of taste, we refer the interested reader to
trol block are defined, as in most PET studies, the above the work of Small et al. (1997 a,b, 1999).
process amounts to subtracting the activation pattern in the
control block (e.g., no-odorant) from activation in the experi- A. Olfactory Epithelium and Bulbs
mental block (e.g., odorant), in order to reveal activation
related to the process of interest alone. The analysis of these When an odorant enters the human nasal passages, it travels
data may be performed for individual subjects, individual a distance of about 7 cm, crosses a mucous barrier, and is
stimuli, or groups of subjects and stimuli, and several differ- then transduced at olfactory receptors that line the upper
ent software packages are available as share-ware for the per- nasal cavity on a sheet termed the olfactory epithelium (see
forming of such analysis. It is beyond the scope of this Chapter 2). The odorant-induced neural signal then pro-
chapter to exceed this simple overview on how functional gresses along the processes of bipolar sensory neurons to
neuroimaging data are analyzed. For a more complete review the first synapse within the olfactory bulb. The location of
of this topic, the interested reader is referred to Bandettini the human olfactory epithelium and bulbs amid air-filled
et al. (1993), Friston et al. (1994, 1996), Buckner et al. cavities, combined with the small size of the olfactory bulbs,
(1996), Zarahn et al. (1997), Aguirre et al. (1997), Josephs renders them inaccessible to current human functional neu-
and Henson (1999), and the references therein. roimaging methods. We have, in fact, tried several times to
acquire functional data from the olfactory bulbs with fMRI,
at both 1.5T and 3T magnetic field strengths, but have not
IV. STATE-OF-THE-ART succeeded in obtaining an artifact-free image. For this rea-
son, under this heading we will deviate from the main scope
In a comprehensive chapter on the neuroanatomy of the of this review, namely human studies, and report on some
human olfactory system, Joseph Price (1990) reviews the imaging studies of the olfactory bulb in animals.
olfactory structures in an order that follows the neuroanatomi- Yang and colleagues (1998) used a 7T fMRI magnet to
cal flow chart of olfactory processing, starting from peripheral study odorant-induced activation in the olfactory bulbs of
and extending to central structures. Here we will follow the rats. The 7T field was utilized to maximize spatial resol-
same neuroanatomical flow chart, reviewing functional, rather ution that was set at 220  220  1000 m. Following
Functional Neuroimaging of Human Olfaction 259

stimulation with iso-amyl acetate, these authors reported

highly significant activations at the level of individual, or
small groups of, glomeruli. These activations were repro-
ducible both across and within animals. These findings
corroborate previous findings with other methods suggest-
ing that there is a spatial component to the encoding of
odors at the level of the olfactory bulb (reviewed in Buck,
1996). This notion has recently received further support in
an imaging study by Rubin and Katz (1999). These authors
used optical imaging in rats to find specific patterns of
glomerulus activity that relate to specific odorant-features
and odorant-concentrations. Optical imaging, however, is
an imaging method not currently applicable to noninvasive
study of adult humans, and is therefore not reviewed in this
chapter. (For more on optical imaging, see Grinvald, 1992;
Malonek and Grinvald, 1997. For more on the application
of optical imaging to study the cortical processing of olfac-
tion in rodents, see Litaudon et al., 1997.)

B. Primary Olfactory Regions

The primary olfactory cortex is currently defined as the

Figure 3 Odorant-induced activation revealed by PET. Odorant-
regions that receive direct projections from the olfactory bulb.
induced activation is seen bilaterally in the region of the piriform
cortex and unilaterally in the right orbitofrontal cortex. (Image
1. Piriform Cortex courtesy of R. Zatore, M. Jones-Gotman, and colleagues.) (See
color insert.)
Price (1990) refers to the piriform cortex as “the largest
and most distinctive olfactory cortical area in most mam-
mals.” The piriform cortex is at the end of the lateral olfac- induced activation in piriform cortex using fMRI (Sobel
tory tract, inhabiting a small portion of both frontal and et al., 2000c). Initially, they found that the somatosensory
temporal lobes at the ventral junction of the two. stimulation induced by sniffing nonodorized air was suffi-
Paradoxically, several attempts to visualize odorant- cient in itself to induce significant activation in the ventral
induced fMRI activation in human piriform cortex have temporal regions (Sobel et al., 1998a). They then hypo-
yielded, at best, only weak activations (e.g., Fulbright thesized that this sniff-delineated region is primary olfac-
et al., 1998; Koizuka et al., 1994; O’Doherty et al., 2000; tory cortex and that the effect of odorants may be
Sobel et al., 1998a, 1999; Yousem et al., 1997, 1999 a,b). measurable within this sniff-delineated region (Sobel et al.,
MEG studies have also not reported piriform cortex acti- 2000c). To test this possibility, subjects were first scanned
vation (Kettenmann et al., 1996; 1997; Kobal and while sniffing, and a region of interest (ROI) was drawn
Kettemann, 1999; Sakuma et al., 1997). PET studies have out of the sniff-activated region. Activity in this region was
had varying levels of success at imaging odorant-induced then measured in separate scans with odorants. It was
activation in piriform cortex, ranging from significant acti- found that odorants induce a rapid onset and short-lived
vation (Fig. 3, see color plate) (Dade et al., 1998; Savic response in about 8% of this region—an area that appeared
et al., 2000; Zatorre et al., 1992) to minimal activation to correspond well with the location of the human piriform
(Zald and Pardo, 1997) to no activation at all (Dade et al., cortex, as suggested in recent atlases (Mai et al., 1997).
2001; Rouby et al., 2000). These differences in the results This response decreased rapidly throughout each individ-
with PET are likely not related to differences in methodol- ual 40-second odorant epoch of a block-design study and
ogy, as most of these attempts at imaging piriform cortex also decreased in initial amplitude from block to block
activation with PET are in fact out of the careful work of over a 4-block experiment (Fig. 4). By using statistical
the same laboratory, headed by Robert Zatorre and measures that took this habituation into account, they
Marilyn Jones-Gotman in Montreal. were able to consistently and reliably measure fMRI
Recently, Sobel and colleagues have shed some light on odorant-induced activation in piriform cortex (Fig. 5; see
this paradox regarding the inability to record odorant- color plate) (Sobel et al., 2000c). Considering that event-
260 Sobel et al.

related experimental designs could minimize habituation,

the above findings suggest that an event-related design
would best fit for measuring odorant-induced activation in
piriform cortex. Recently, Poellinger et al. (2001) probed
temporal differences that may underlie habituation. Using
an event-related fMRI design, they showed that piriform
cortex, entorhinal cortex, and the amygdala exhibit short
phasic increases in signal followed by a prolonged
decrease in signal below baseline. In contrast, the
orbitofrontal cortex exhibited a sustained increase in activ-
ity over nearly 60 seconds of odor presentation.
The rapid habituation of odorant-induced activity
shown in piriform cortex explains why it was not previ-
ously evident in many fMRI and PET studies, but why
was such activity evident in yet other PET studies? One
possibility, discussed in detail by Sobel et al. (2000c), is
related to the differences in statistical analysis commonly
used in PET and fMRI studies. The PET analysis may
lend significant weight to the early transient odorant-
induced activity, whereas commonly used fMRI analysis
Figure 4 Time course of odorant-induced activity in human piri- packages may have obscured this phenomenon. A second
form cortex: the averaged time course from piriform cortex (8% of potential source for the difference between different PET
the region responsive to sniffing alone) in 8 subjects exposed to studies may be related to subject behavior. In the PET
the odorant vanillin. The response shows a rapid habituation studies, subjects knew which block would be an odorant
within the first 40 seconds of odorant presence as well as a con- block and which block would contain a diluent only. It is
tinued habituation throughout the 320 seconds of the experiment. quite conceivable that when knowingly presented with a

Figure 5 Odorant-induced activation revealed by fMRI. Odorant-induced activation in piriform cortex and additiona olfactory regions.
The activation is a composite of 8 subjects stimulated with the odorant vanillin. Activation was analyzed using statistical methods that took
into account the rapid habituation in piriform cortex. (From Sobel et al., 2000c.) (See color insert.)
Functional Neuroimaging of Human Olfaction 261

foil, subjects made less of an effort to scan the olfactory studies (e.g., Birbaumer et al., 1998; Royet et al., 2000;
environment or, in other words, sniff. As previously Zald and Pardo, 1997). The amygdala plays a major role
shown (Sobel et al., 1998a, 2000c), sniffing alone is suffi- in processing of emotionally significant stimuli. For this
cient to induce activity in the ventral temporal regions, reason, Zald and Pardo used an aversive sulfide cocktail to
and therefore it could be that the difference between the study the odorant-induced response in the amygdala. The
odorant and no-odorant conditions was in fact partially a sulfide odorants produced significant bilateral amygdala
difference in sniffing between conditions in these studies. activation (Fig. 6; See color insert). By contrast, pleasant
Although the above suggestions may indeed underlie the odorants did not induce as significant activation in the
between-study variability in piriform cortex activation, it amygdala (Zald and Pardo, 2000). This dissociation of
is still nevertheless evident that the rules underlying piri- amygdaloid activation induced by unpleasant but not
form cortex activity are complex. This activity appears pleasant odorants may be related to the findings that
to significantly change throughout continued stimulation, seizures originating in the amygdala often induce olfac-
as well as change as a result of previous experience, as tory hallucinations of unpleasant but not pleasant odors
suggested in other animals as well (Stopfer and Laurent, (Andy, 1967; Chitanodh, 1966). The latter relationship,
1999). Now that we can consistently image such activity however, must be addressed with caution, as it remains to
using either the method that accounts for habituation be determined if the increased amygdala response was
(Sobel et al., 2000c) or the event-related design in fMRI, related to odorant hedonic value per se, unconfounded by
we should be able to delineate human piriform cortex odorant intensity or odorant species.
in vivo as well as better characterize the rules underlying The amygdala activation induced by unpleasant odor-
piriform cortex activity. ants shows a tendency towards asymmetry, whereby
increased unpleasantness is associated with increased left
amygdala activation as measured in right-handed sub-
2. Anterior Olfactory Nucleus and Olfactory Tubercle
jects. This asymmetry was evident in a PET study (Zald
These two small structures that receive direct projections and Pardo, 2000) (Fig. 7; see color plate) and in a pilot
from the olfactory bulbs are not well defined in the fMRI study (Prabhakaran et al., 1999), but not in an
human (Price, 1990). However, significant odorant- fMRI study by Birbaumer et al. (1998) that compared
induced activation in the area of these structures is odorant-induced amygdala activation in social phobics
consistently measured (Sobel et al., 2000c). Figure 6 and nonphobic controls.
(Fig. 6; see color plate) depicts activation induced by the
odorants vanillin, decanoic acid, propionic acid, and
4. Entorhinal Cortex
valeric acid in composite images of eight subjects. All
four odorants induced significant activation in this The rostral portion of the entorhinal cortex receives
region. Although it would appear that this activation is a direct projection from the olfactory bulbs (Price,
primarily in the region of the anterior olfactory nucleus, 1990). Reporting on odorant-induced activation in the
one cannot structurally delineate these bodies on the entorhinal cortex is somewhat neglected, we think for
MRI image and therefore cannot rule out the possibility the following reason: often, activation in the amygdala
that the source of this activation was in fact the olfactory will also cover a portion of neighboring entorhinal cor-
tubercle or other neighboring structures. These activa- tex. However, because the amygdala is usually a specific
tion images suggest that these regions may play a con- target of interest, researchers tend to report on “amyg-
siderable role in processing olfactory information in dala and neighboring cortex.” The same is true for acti-
humans, in spite of the fact that they are not very well vation that occurs in the hippocampus, which also often
defined in the human from a structural and/or cytoarchi- covers parts of the entorhinal cortex. Thus, although
tectural point of view. What role these structures play in entorhinal cortex odorant-induced activation is com-
human olfactory processing, and how this role is carried monly seen (as reported in Sobel et al., 1998a; Zald and
out, remains unknown. Pardo, 2000), it is not carefully analyzed. In this, we as
an olfactory imaging community are falling into an
unfortunate, yet common, trap. By concentrating our
3. Amygdala
efforts on some regions where we expect something
The anterior cortical nucleus of the amygdala and the important is happening vis-á-vis olfactory processing
periamygdaloid cortex receive direct projections from the (e.g., amygdala), we may be overlooking areas that are
olfactory bulbs (Price, 1987). Odorant-induced amygdala perhaps less trendy but nevertheless are of equal or
activation has indeed been reported in both PET and fMRI greater importance in olfactory function.
262 Sobel et al.

Figure 6 Odorant-induced activation in the region of the anterior olfactory nucleus and olfactory tubercle. The top row shows com-
posite images of activation from 8 subjects in slice 5 of the olfactory acquisition (see Fig. 1). All odorants induced significant acti-
vation in the region of the anterior olfactory nucleus and olfactory tubercle (green arrows). To assist in localization of this, the
average centroid of the activation was transferred to the corresponding point on a standard coronal plane acquisition, using a three-
dimensional cross-referencing program (red circle). The centroid of activation was within a coronal plane 2 mm anterior to the ante-
rior commisure, in the region of the anterior olfactory nucleus (AON), bordering with the olfactory tubercle (Tu), the medial tip of
the frontal portion of the piriform cortex (PirT), and the lateral tip of the diagonal band nucleus (Db). The activated region occa-
sionally spanned down to the uncus (Un) and occasionally down to the area of the basomedial amygdaloid nucleus. Additional acti-
vations seen in this slice are in the claustrum (Cl), peri-insular region (Ins), cingulate gyrus (Cg), and inferior (IFG), and middle
(MFG) fontal gyri. (See color insert.)

C. Secondary Olfactory Regions (Brewer et al., 1998; Gabrieli et al., 1997), but no neuro-
imaging studies of olfaction have focused on hippocampal
1. Hippocampus
activity. Odorant-induced hippocampal activity, however,
The hippocampus receives an olfactory input from the has been reported in several human neuroimaging studies
entorhinal cortex (Price, 1990). The hippocampus has been using either PET, SPECT, or fMRI (e.g., Malaspina et al.
the focus of several human neuroimaging studies of memory 1998; Savic et al., 2000; Small, 1997; Sobel et al., 2000c)
Functional Neuroimaging of Human Olfaction 263

induced by the compounds remains unclear. The existence

of human pheromones is still very controversial (Sobel and
Brown, 2001) (see Chapter 17), but the compounds used
by Savic et al. clearly differ from more traditional odorants
in the pattern of hypothalamic activation they induced.

3. Thalamus
The thalamus receives olfactory projections from the piri-
form cortex, periamygdaloid cortex, entorhinal cortex, and
olfactory tubercle (Price, 1985). Most of these projections
are thought to synapse onto the mediodorsal thalamic nuclei,
and some onto the submedial nucleus. In monkeys, odorant-
induced electrophysiological activity has been recorded from
both of these thalamic sites (Benjamin and Jackson, 1974;
Russchen et al., 1987). In contrast to recordings from the
Figure 7 Odorant-induced activation in the amygdala. PET hypothalamus, odorant-induced responses in the thalamus
activation induced by unpleasant odorants in the amygdala and appear to be broadly tuned (Yarita et al., 1980). Such thala-
neighboring regions. Activation can be seen in the amydala mic activation is seen in human functional neuroimaging
(Amg) bilaterally and in the left claustrum (Cl) and insula (I). studies (e.g., Savic et al., 2000; Sobel et al., 2000c), but this
(Image courtesy of D. Zald and colleagues.) (See color insert.) activation has yet to be carefully characterized. It is, however,
our impression, that the thalamic odorant-induced activation
in humans is medial, yet more anterior than that recorded in
monkeys (Sobel et al., 1999). This impression awaits the
2. Hypothalamus scrutiny of a more careful specific characterization of thala-
The hypothalamus receives extensive olfactory input, most mic odorant-induced patterns of activation in humans.
prominently from the piriform cortex and anterior olfac-
tory nucleus, and also from the amygdala (Price, 1990). 4. Orbitofrontal Cortex
Electrophysiological studies in monkeys suggest narrowly
tuned odorant-responsive cells in the hypothalamus The orbitofrontal cortex is considered the main site of
(Takagi, 1986; Tazawa et al., 1987). Hypothalamic secondary olfactory processing in humans. This cortical
odorant-induced activation has been reported in some region, located on the ventral portion of the frontal lobe, in
human neuroimaging studies (e.g., Rouby et al., 2000; fact evolved as part of a series of concentric rings around
Savic et al., 2001; Sobel et al., 1999). Rouby et al. (2000) primary olfactory cortex (Carmichael et al., 1994; Zald and
were the first to use imaging to address the specific role of Kim, 1996a,b). The orbitofrontal cortex receives both
the hypothalamic relay in human olfactory processing. direct and indirect input from primary olfactory cortex.
Using PET, these authors measured greater hypothalamic Direct projections from piriform cortex form a transsynap-
activity during pleasantness judgments than during inten- tic input to the posterior orbitofrontal region (Price, 1990),
sity judgments of the same odorants. The authors speculate and indirect projections from primary olfactory cortex
that the hypothalamus may be involved in the affective reach the orbitofrontal cortex via the thalamus (Yarita
processing of olfactory information that requires access to et al., 1980). In monkeys, cells with selective odorant-spe-
information about internal state. cific responses have been characterized in orbitofrontal
Using PET, Savic et al. (2001) recently measured robust cortex (Critchley and Rolls, 1996; Rolls et al., 1996;
sex-specific activation of the hypothalamus. Smelling an Tanabe et al., 1975), and in humans, orbitofrontal lesions
androgen-like compound produced activation in the hypo- lead to impairments in odorant discrimination and identifi-
thalamus of women, but not men; smelling an estrogen- cation (Jones-Gotman and Zatorre, 1988; Zatorre and
like compound activated the hypothalamus of men, but not Jones-Gotman, 1991).
women. This surprising pattern of activation is likely due In contrast to the previously described difficulties in
to the nature of the odorants used. Both compounds are imaging odorant-induced activation in primary olfactory
considered by some as putative human pheromones, and cortex, odorant-induced activation in orbitofrontal cortex
the hypothalamus mediates a wide variety of so-called has been routinely reported in both PET and fMRI studies.
pheromonal effects in other animals, but the specific effect The localization of activation within the orbitofrontal gyri
264 Sobel et al.

is quite consistent across these studies and methods same group of subjects exhibited greater left than right
(reviewed in Zald and Pardo, 2000; Zatorre and Jones- orbitofrontal activation when exposed to pleasant odorants
Gotman, 2000). As stressed by Zald and Pardo (2000), as well, suggesting that the leftwards shift in orbitofrontal
the orbitofrontal cortex should not be treated as an cortex activation that they witnessed earlier is not solely
homogeneous region vis-á-vis olfactory processing. the result of the hedonic value of the odorants used.
Electrophysiological studies in monkeys suggest a distinc- Furthermore, Rouby et al. (2000) used PET to compare
tion between lateral and central olfactory zones in posterior activation induced by odorant pleasantness ratings versus
orbitofrontal cortex (Takagi, 1986; Tanabe et al., 1975; odorant intensity ratings and also found predominantly
Yarta et al., 1980). Human neuroimaging studies have right orbitofrontal activation for both types of olfactory
reported odorant-induced activation in various parts of the assessments. The only significant difference in brain acti-
orbitofrontal cortex, most commonly in the intermediate vation induced by these two tasks was in the hypothala-
and posterior orbital region, and in the lateral orbitofrontal mus, where greater activation was measured following
gyrus (Sobel et al., 1998a, 2000c; Zald and Pardo, 1997), pleasantness judgments. Royet et al. (2001) asked subjects
and less commonly in the medial orbitofrontal region to make judgments about the presence (odor detection),
(Royet et al., 1999; Small et al., 1997). Review of the liter- intensity, pleasantness, familiarity, and edibility of various
ature suggests a trend whereby tasks related to olfactory odorants. They found that left orbitofrontal cortex activity
memory induce activation more medially, and tasks related increased significantly during the pleasantness and famil-
to olfactory hedonics induce activation more laterally, but iarity tasks. Right orbitofrontal cortex activity increased in
this distinction is far from concrete, and more work is nec- all five tasks but was highest during familiarity judgments
essary in order to elucidate which orbitofrontal area is pri- and lowest during the detection task. Considering the high
marily involved in which olfactory task. degree of variability regarding orbitofrontal cortex lateral-
Lesions to the right orbitofrontal cortex lead to greater ity, the rules underlying asymmetry remain unclear.
olfactory impairment than lesions to the left orbitofrontal The orbitofrontal cortex is also the site of extensive
cortex (Jones-Gotman and Zatorre, 1993). This functional cross-modal integration (Rolls, 1996). An example of the
asymmetry implied by lesion findings is indeed reflected orbitofrontal integration of smell and taste is the electro-
in patterns of odorant-induced activation as measured with physiological findings in monkeys suggesting that
neuroimaging. Specifically, activation is usually more sig- orbitofrontal neurons decrease their response to the odors
nificant in and occurs in a larger portion of the right than
left orbitofrontal cortex. This asymmetry was first shown
Women Men
in the pioneering PET study reported by Zatorre and col-
leagues in 1992, and later replicated with fMRI (Royet
et al., 1999; Sobel et al., 1998a). Similar asymmetry has
been reported in additional frontal regions (Savic and
Gulyas, 2000; Yousem et al., 1999a,b). Lesion studies have
also suggested greater right than left orbitofrontal cortex
involvement in olfactory memory tasks (Jones-Gotman
and Zatorre, 1993), and this is indeed also reflected in
greater right than left activation in olfactory memory neu- Subject group Left frontal Right frontal Left temporal Right
roimaging studies (Dade et al., 1998; Savic et al., 2000). voxels voxels voxels temporal
activated activated activated voxels
Some evidence suggests that the asymmetry in activated
orbitofrontal cortex activation seen in most imaging stud- Female 157 730 303 465
Male 19 152 55 31
ies may be odorant-specific. Ratio female to 8.3 4.8 5.5 15.0
Using PET, Zald and colleagues found that odorants male
with strong negative hedonic characteristics appear to
break down the functional coupling between the right and Figure 8 Odorant-induced activation is greater in women than
left orbitofrontal cortex (Zald et al., 1998) and induce in men. The composite activation of maps of eight right-handed
women are compared with those of eight right-handed men, given
greater activation in the left than right orbitofrontal cortex
the same olfactory stimuli in an fMRI experiment at 1.5 Tesla.
(Royet et al., 2000; Zald and Pardo, 1997). A similar sug- The women’s group-averaged activation maps showed up to eight
gestion has recently been made for additional frontal times more activated voxels than did those of men for specific
regions as well (Fulbright et al., 1998). Although the pos- regions of the brain. The difference was most striking in the right
sibility of hedonic-based laterality of olfactory activation temporal (peri-insular) regions. (From Yousem et al., 1999b.)
is appealing, Zald and Pardo (2000) later noted that the (See color insert.)
Functional Neuroimaging of Human Olfaction 265

of food that had been eaten to satiety (Critchley and Rolls, These regions include the frontal polar area (Dade et al.,
1996). O’Doherty and colleagues (2000) conducted a repli- 2001), superior frontal gyrus (Dade et al., 2001; Fulbright
cation of this finding in humans using fMRI. They found et al., 1998; Royet et al., 1999; Sobel et al., 2000c; Yousem
that the activation induced by the odor of banana, but not by et al., 1999a), middle frontal gyrus (Dade et al.,
the odor of vanillin, was significantly decreased after eating 2001; Fulbright et al., 1998; Malaspina et al. 1998; Sobel
banana to satiety. The site of decreased activation appeared et al., 2000c), and inferior frontal gyrus (Dade et al., 2001;
to be variable within the orbitofrontal region. Fulbright et al., 1998; Malaspina et al. 1998; Royet et al.,
Odorant-induced activation has been consistently 1999; Sobel et al., 2000c), which was also activated in
witnessed in additional regions within the frontal lobe. response to a subthreshold stimulus (Sobel et al., 1999).

Figure 9 Odorant-induced activation is greater in young than in older subjects. A full-brain acquisition of a young (top) versus older
(bottom) subject, given the same olfactory stimuli in an fMRI experiment at 1.5 Tesla. Significantly greater activation is seen in the
younger versus the older subject. (From Yousem, 1999a.) (See color insert.)
266 Sobel et al.

The specific role that each of these different frontal regions described). These include various parietal lobe activations
plays in olfactory processing remains unclear. (Dade et al., 2001; Malaspina et al. 1998; Royet et al.,
Interestingly, Yousem and colleagues (1999b) reported that 1999; Sakuma et al., 1997; Savic et al., 2000; Yousem
the extant of frontal odorant-induced activation was eight et al., 1999a), the precentral gyrus (Dade et al., 2001;
fold greater in women than in men (Fig. 8; see color plate) Fulbright et al., 1998; O’Doherty et al., 2000), superior
and also age-dependant—greater in young than in old temporal gyrus (Kettenmann et al., 1996; Kobal and
adults (Fig. 9; see color plate) (Yousem et al., 1999a). Kettenmann, 2000; Malaspina et al. 1998; Sakuma et al.,
Although this can be taken to suggest that the role frontal 1997), inferior temporal gyrus (Malaspina et al., 1998),
regions play in olfaction may be related to the advantage in cingulate gyrus (Dade et al., 2001; Fulbright et al., 1998;
olfactory performance often attributed to women over men Levy et al., 1997; O’Doherty et al., 2000; Royet et al.,
(e.g., Doty et al., 1985), it remains unclear if this lesser 1999; Savic et al., 2000; Sobel et al., 2000c; Yousem et al.,
frontal activation represents less frontal processing of the 1999), occipital lobe (Royet et al., 1999; Savic et al., 2000;
same input or is simply a reflection of lesser input from the Yousem et al., 1999b), and cerebellum (Qureshy et al.,
peripheral olfactory structures. 2000; Savic et al., 2000; Small et al., 1997; Sobel et al.,
1998b; Yousem et al., 1997). The odorant-induced acti-
vation in most of these regions has not been carefully
5. Insula
characterized; thus, it remains unclear to what extent these
In the rat, a direct projection from piriform cortex to activations are genuinely odorant-induced or perhaps
agranular insular cortical areas has been described (Price, related to additional resources supporting performance in
1985). It is also assumed that olfactory information reaches olfactory tasks. A case in point may be the activation seen
the insular area via a thalamic relay (Price, 1990). Electrical in cingulate gyrus, which more likely reflects the atten-
responses to olfactory stimuli have been recorded from tional demands generated by performing any task, olfactory
insular areas in monkeys (Takagi, 1986), and odorant- or other, rather than odorants per se. One such region not
induced insular activation is indeed commonly measured in previously associated with olfactory processing that has
functional neuroimaging studies of human olfaction using received some direct experimental attention is the cerebel-
both PET and fMRI (Fulbright et al., 1998; O’Doherty et lum, as discussed below.
al., 2000; Savic et al., 2000; Sobel et al., 2000c). Activation
has been reported in both the anterior and posterior portions
1. Cerebellum
of the insula, and in some cases activation has been
reported as borderline between the insula and claustrum The cerebellum is a large brain structure located at the back
(Savic et al., 2000; Zatorre et al., 1992). The role of the of the brain that in the human contains more neurons than
insular region in olfactory processing remains unknown, the rest of the brain combined (Williams and Herrup,
but it may be related to the integration of olfactory and 1988). The cerebellum has classically been considered as
other sensory information. Taste, for example, has been primarily a motor control organ (Ito, 1984) (for alternative
shown by functional neuroimaging to also be represented in views, see Bower, 1997; Ivry, 1997). Cerebellar functions
the insular region (Kinomura et al., 1994; Small et al., in visual- and auditory-related tasks have been extensively
1997, 1999). Fulbright et al. (1998) found that the extent of described (Huang and Liu, 1991; Stein and Glickstein,
fMRI activation in the left insula was correlated with the 1992). Recently, Qureshy et al. (2000) provided evidence
hedonic ratings of the odorant isovaleric acid whereby acti- that the cerebellum is also active during cognitive processes
vation was greater when subjective ratings of unpleasant- in olfaction. In one task subjects were asked to name odors
ness were more intense, thus suggesting an insular role in (naming task), and in another subjects were asked to com-
the hedonic assessment of odors. The perisylvian region has pare the presented odorant with a previously memorized
also been noted by Yousem et al. (1999a,b) to be a large odorant to determine if the odors were the same (matching
area of odorant-induced activation, which again shows gen- task). Both tasks activated the right posteromedial and left
der and age variability (Fig. 8). anterolateral portions of the cerebellum.
As olfaction is a sensory process largely dependent upon
D. Regions Not Previously Associated with the the fine motor process of sniffing (Laing, 1983; Le Magnen,
Olfactory System 1945; Mozell et al., 1983; Rehn, 1978), the cerebellum may
also play a role in motor control of olfaction. Sniffing plays
Odorant-induced activation has been consistently recorded a major role not only in transport of the olfactory stimulus
with both fMRI and PET from several brain regions not (Hahn et al., 1994), but also in patterns of neural activity in
previously associated with olfactory processing (in addi- primary olfactory cortex in the human (Sobel et al., 1998a,
tion to the various frontal lobe regions previously 2000c). A fine reciprocal interaction persists whereby sniff-
Functional Neuroimaging of Human Olfaction 267

ing strategy and timing modulate odorant intake and, in (Laing, 1983; Sobel et al., 2000a). Maintaining this
turn, odorant intake content modulates further sniffing. For inverse proportionality calls for an accurate rapid feed-
example, in response to increasing odorant concentration, back mechanism that monitors the sensory input (odor
there is a decrease in sniff volume (Laing, 1983; Sobel et concentration) and modulates the motor output (sniff
al., 2000a; Youngentob et al., 1987). volume). Cerebellar maintenance of such feedback
Cerebellar involvement in respiration (Colebatch mechanisms has been extensively described for tactile
et al., 1991; Mansfeld and Tyukody, 1936) suggests that information, as well as for vision and audition, and here
sniff-motor/sensory circuits may be in part controlled by we suggest the same cerebellar function in olfaction. In
the cerebellum. Thus, considering that odor content this capacity, the cerebellum could be subserving main-
affects sniffing, odor content information may also be tenance of the Teghtsoonian model of olfactory size con-
relayed to the cerebellum. Preliminary reports using stancy (Teghtsoonian et al., 1978).
fMRI, suggesting that odorants may indeed activate the
cerebellum (Small et al., 1997; Sobel et al., 1997b;
Yousem et al., 1997), merited a more careful examina-
tion of this possibility. To this end, we studied cerebellar V. CONCLUDING REMARKS
patterns of activation in response to various concentra-
tions of the pure olfactant vanillin and the strongly A. The Naive Observer
trigeminal odorant propionic acid (Sobel et al., 1998b)
(Fig. 10, see color plate). The odorants vanillin and pro- Current efforts at functional neuroimaging of human olfac-
pionic acid both induced significant activation, primarily tion have measured odorant-induced activation in most of
in the posterior lateral hemispheres. Activation was con- the classically defined olfactory regions of the brain and
centration dependent, greater following stimulation with thereby have further validated this new research tool.
higher concentration odorants (Fig. 10). By contrast, the However, visual inspection of these imaging efforts leaves
action of sniffing nonodorized air induced significant one with the following striking impression: if we were to
activation in the anterior cerebellum, primarily in the display the results of all these studies before a person who
central lobule. Cerebellar activation in response to odor- knows a lot about imaging and nothing at all about olfac-
ants has now been reported in several imaging studies tion and ask this person to pick out the brain regions
using both PET and fMRI (Qureshy et al., 2000; Savic responsible for olfactory processing, a surprising result
et al., 2000; Small et al., 1997; Sobel et al., 1998b; may occur. A naïve observer would likely pick one of sev-
Yousem et al., 1997; Zatorre and Jones-Gotman, 2000). eral potential regions based on the extensive odorant-
Before these imaging studies, it was unknown whether induced activation in those regions (e.g., the superior
the cerebellum contributes in any way to olfactory pro- temporal gyrus), but probably none of those regions would
cessing. What may be the role of the cerebellum in olfac- be the piriform cortex, since this region exhibits only a
tion? The following is a working hypothesis: sniff very small activation. This state of affairs has significant
volume is inversely proportional to odor concentration implications for the imaging community regarding how it
treats functional imaging results and significant implica-
tions for the olfaction community regarding how it views
cortical processing of olfaction. On one hand, it is impor-
tant to keep in mind that the size and significance of an
activation are not a linear reflection of its importance or
centrality in sensory processing. For example, there is no
doubt that piriform cortex plays a major role in olfactory
processing, even though it took quite an effort to even mea-
sure this activity with functional neuroimaging, and even
when now consistently measured it is not very great in
extent. Thus, one does not want to be an overly naïve
observer. On the other hand, ridding oneself of preconcep-
Figure 10 Odorant-induced activation is concentration-depen- tions when approaching the results of olfactory neu-
dent in the cerebellum. Activation induced by low, intermediate, roimaging studies may provide new insights into the
and high concentrations of propionic acid in the lateral posterior cortical processing of olfaction. Thus, approaching the
portion of the cerebellum. Greater odorant concentration induced data like an uninformed, naïve observer is at times quite
greater activation in this, but not other, cerebellar regions. (From beneficial, a point in case being the recently revealed cere-
Sobel et al., 1998b.) (See color insert.) bellar role in olfactory processing.
268 Sobel et al.

B. What Is Primary About Primary Olfactory primary olfactory cortex in the human based on functional
Cortex? Time to Add a Functional Definition criteria. Such functional criteria may be a hierarchy of
functional capacity, such as the levels of processing crite-
Human primary olfactory cortex is generally considered to ria described above, or in turn, such functional criteria may
be composed of several structures and areas that inhabit be a temporal hierarchy, i.e., primary olfactory cortex can
the ventral junction of frontal and temporal lobes (Allison, be defined as those areas that first process information
1954; Eslinger et al., 1982; Haberly and Price, 1978; from the olfactory bulb, from a temporal rather than purely
Jones-Gotman et al., 1997; Price, 1973, 1990). A cortical anatomical connectivity standpoint. Either of these poten-
region processing sensory information is generally classi- tial methods for functional classification of cortical regions
fied as “primary” when it is the first cortical region to can now be applied to the human olfactory system by using
receive input from the peripheral receptors. In olfaction, functional neuroimaging.
the peripheral receptors synapse only onto the olfactory A first step in this direction has recently been made
bulb. The olfactory bulb then synapses onto extensive cor- by Savic et al. (2000), who used PET to study activation
tical regions. For example, anterograde tracers placed into induced by olfactory tasks hierarchically organized in
the olfactory bulb of macaque monkeys labeled axons in terms of demands. Savic and colleagues reported a
the following cortical areas: the anterior olfactory nucleus, task-specific recruitment of cortical regions subserving
piriform cortex, ventral tenia tecta, olfactory tubercle, olfactory tasks, starting from olfactory detection (amyg-
anterior cortical nucleus of the amygdala, periamygdaloid dala-piriform, orbitofrontal, cingulate, thalamus), dis-
cortex, and olfactory division of the entorhinal cortex crimination of odor intensity (additional recruitment of
(Carmichael et al., 1994). Additional studies, although insula and cerebellum), discrimination of odor quality
controversial, suggest some projections from the olfactory (additional recruitment of prefrontal cortex, frontal oper-
bulb directly to frontal lobe (Cinelli et al., 1987; Shipley culum, caudate, and subiculum), and, finally, odor
and Adamek, 1984). Are all these regions that receive recognition memory (additional recruitment of temporal
direct bulbar projections, including the possible frontal and parietal regions). This PET study, however, was
portions, to be considered primary olfactory cortex? If so, designed to address global levels of processing and not
what value for us as a field is there in this definition that to specifically extract regions involved in odorant detec-
encompasses almost all the cortical olfactory areas and, in tion alone. To do the latter, one must use extremely low
fact, a rather large portion of the brain! Human primary odorant concentrations that are above chance detection
olfactory cortex as it is currently defined relates to a rather threshold, but under chance identification threshold.
large group of dispersed structures, some of which may When using high-concentration odorants, subjects auto-
have very distinct roles in olfactory processing (e.g., matically also perform odorant identification, and the
amygdala vs. piriform cortex). activation associated with detection alone cannot be
An alternative approach to classifying a cortical region discerned.
as “primary” would be to add functional to structural cri- We would like to here put forth a suggestion for a
teria. One such potential criterion would reflect levels of functional criterion defining primary olfactory cortex that
processing. By this approach, primary olfactory cortex is in fact a combination of the traditional neuroanatomi-
would be the cortical region responsible for the extraction cal definition, combined with the above hierarchical
of the earliest features in the olfactory stimulus space, for functional criteria of processes and time. In our view, one
example, olfactory detection without any further odor-con- might consider the cortical regions that both receive
tent processing. One may note that such a functional clas- direct projections from the olfactory bulb and are acti-
sification was possible even before the advent of functional vated by sniffing alone (i.e., with no odorant content) as
neuroimaging, through the interpretation of human lesion primary olfactory cortex. The sniff is the attentional spot-
outcome. The effect of ventral temporal lesions on olfac- light of olfaction, and it appears to prepare a specific sub-
tory processing, however, has proven to be quite variable set of olfactory cortex for the upcoming olfactory input.
(Doty et al., 1997; Eskenazi et al., 1983, 1986; Eslinger It is the directing of this olfactory searchlight that we
et al., 1982; Henkin et al., 1977; Jones-Gotman and consider the most primary olfactory task and therefore
Zatorre, 1988; Rausch and Serafetinides, 1975). suggest that it reflect primary olfactory cortex. The above
Furthermore, human lesion case studies usually cannot is our suggestion for a functional definition of primary
offer precise localization, nor do they enable studying olfactory cortex. We can, however, imagine several alter-
normal olfactory processing. In contrast, functional neu- native functional criteria for defining primary olfactory
roimaging is devoid of such concerns. Thus, we could now cortex and hope that our fellow neuroimagers of olfaction
use functional neuroimaging to redefine and localize will tackle this task.
Functional Neuroimaging of Human Olfaction 269

C. Final Word Birbaumer, N., Grodd, W., Diedrich, O., Klose, U., Erb, M.,
Lotze, M., Schneider, F., Weiss, U., and Flor, H. (1998). fMRI
This review chapter has been written at an awkward point reveals amygdala activation to human faces in social phobics.
in the time of functional neuroimaging of human olfaction. Neuroreport 9(6):1223–1226.
Block, F. (1946). Nuclear introduction. Physiol. Rev.
Functional MRI of human sensory processing was only
70:460–474.
first achieved 9 years ago (Ogawa et al., 1992). The first
Bower, J. M. (1997). Control of sensory data acquisition. Int. Rev.
effort at fMRI of human olfaction was only 7 years ago Neurobiol. 41:489–513.
(Koizuka et al., 1994). Thus, the number of human olfac- Brand, G., Millot, J. L., and Henquell, D. (2001). Complexity of
tory neuroimaging studies is quite limited. That said, this olfactory lateralization processes revealed by functional imag-
very small number of studies has in fact already given rise ing: a review. Neurosci. Biobehav. Rev. 25(2):159–166.
to some new insights into olfactory processing, and first Brewer, J. B., Zhao, Z., Desmond, J. E., Glover, G. H., and
attempts have now been made by Levy and others to intro- Gabrieli, J. D. E. (1998). Making memories: brain activity
duce functional neuroimaging to the olfactory clinical that predicts how well visual experience will be remembered.
arena (Crespo-Facorro et al., 2001; Levy et al., 1998 a,b, Science 281(5380):1185–1187.
1999; Moberg et al., 1999; Turetsky et al., 2000; Yousem Buck, L. B. (1996). Information coding in the vertebrate olfac-
tory system. Annu. Rev. Neurosci. 19:517–544.
et al., abstract). Furthermore, considering the current
Buckner, R. L., Bandettini, P. A., O’Craven, K. M., Savoy, R. L.,
explosion in proliferation of fMRI machines in research
Petersen, S. E., Raichle, M. E., and Rosen, B. R. (1996).
institutions around the world, if we were to wait for the Detection of cortical activation during averaged single trials
next edition of the Handbook of Olfaction and Gustation in of a cognitive task using functional magnetic resonance imag-
order to introduce the topic of functional neuroimaging of ing. Proc. Natl. Acad. Sci. USA 93(25):14878–14883.
olfaction, it would be considerably too late. Thus, we opted Carmichael, S. T., Clugnet, M. C., and Price, J. L. (1994). Central
to be a bit early rather than quite late and hopefully have olfactory connections in the macaque monkey. J. Comp.
given the reader some insight into how functional neu- Neurol. 346(3):403–434.
roimaging works, how it has been applied to the study of Cerf-Ducastel, B., and Murphy, C. (2001). fMRI activation in
human olfaction, and in what way it may be applied in the response to odorants orally delivered in aqueous solutions.
future. The methodology of functional neuroimaging is Chem. Senses 26(6):625–637.
Chitanondh, H. (1966). Stereotaxic amygdalotomy in the treat-
improving at an exceptionally fast rate. One of the most
ment of olfactory seizures and psychiatric disorders with
exciting recent developments is the use of simultaneous
olfactory hallucination. Confin. Neurol. 27(1):181–196.
MEG and fMRI, offering the temporal resolution of the Cinelli, A. R., Ferreyra-Moyano, H., and Barragan, E. (1987).
former combined with the spatial resolution of the latter Reciprocal functional connections of the olfactory bulbs and
(e.g., Dale et al., 2000). This and other technical develop- other olfactory related areas with the prefrontal cortex. Brain
ments lead us to predict that functional neuroimaging Res. Bull. 19:651–661.
methods will revolutionize our understanding of cortical Colebatch, J. G., Adams, L., Murphy, K., Martin, A. J.,
processing of olfaction over the next few years. Lammertsma, A. A., Tochon-Danguy, H. J., Clark, J. C.,
Friston, K. J., and Guz, A. (1991). Regional cerebral blood
flow during volitional breathing in man. J. Physiol. (Lond)
REFERENCES 443:91–103.
Cometto-Muñiz, J. E., and Cain, W. S. (1995) Olfactory adapta-
Aguirre, G. K., Zarahn, E., and D’Esposito, M. (1997). Empirical tion. In Handbook of Olfaction and Gustation, R. L. Doty
analyses of BOLD fMRI statistics. II. Spatially smoothed data (Ed.). Marcel Dekker, New York, pp. 257–282.
collected under null-hypothesis and experimental conditions. Crespo-Facorro, B., Paradiso, S., Andreasen, N. C., O’Leary, D.
Neuroimage 5(3):199–212. S., Watkins, G. L., Ponto, L. L., and Hichwa, R. D. (2001).
Allison, A. C. (1954). The secondary olfactory areas in the Neural mechanisms of anhedonia in schizophrenia: a PET
human brain. J. Anat. 88:481–488. study of response to unpleasant and pleasant odors. JAMA
Andy, O. J. (1967). The amygdala and hippocampus in olfactory 286(4):427–435.
aura. Electroencephalogr. Clin. Neurophysiol. 23(3):292. Critchley, H. D., and Rolls, E. T. (1996a). Olfactory neuronal
Bandettini, P. A., Jesmanowicz, A., Wong, E. C., and Hyde, J. S. responses in the primate orbitofrontal cortex: analysis in
(1993). Processing strategies for time-course data sets in func- an olfactory discrimination task. Neurophysiology
tional MRI of the human brain. Magn. Reson. Med. 30: 161–173. 75(4):1659–1672.
Benjamin, R. M., and Jackson, J. C. (1974). Unit discharges in Critchley, H. D., and Rolls, E. T. (1996b). Hunger and satiety
the mediodorsal nucleus of the squirrel monkey evoked by modify the responses of olfactory and visual neurons in the
electrical stimulation of the olfactory bulb. Brain Res. primate orbitofrontal cortex. Neurophysiology 75(4):
75(2):181–191. 1673–1686.
270 Sobel et al.

Dade, L. A., Jones-Gotman, M., Zatorre, R. J., and Evans, A. C. Glover, G. H., and Lai, S. (1998). Self-navigated spiral fMRI:
(1998). Human brain function during odor encoding and interleaved versus single-shot. Magn. Reson. Med.
recognition. A PET activation study. Ann. NY Acad. Sci. 39:361–368.
855:572–574. Grinvald, A. (1992). Optical imaging of architecture and function
Dade, L. A., Zatorre, R. J., Evans, A. C., and Jones-Gotman, M. in the living brain sheds new light on cortical mechanisms
(2001). Working memory in another dimension: functional underlying visual perception. Brain Topogr. 5(2):71–75.
imaging of human olfactory working memory. Neuroimage Gulick, W. L., Gescheider, G. A., and Frisna, R. D. (1989).
14(3):650–660. Hearing: Physiological Acoustics, Neural Coding, and
Dale, A. M., Liu, A. K., Fischl, B. R., Buckner, R. L., Belliveau, Psychoacoustics. Oxford University Press, Oxford, United
J. W., Lewine, J. D., and Halgren, E. (2000). Dynamic statis- Kingdom.
tical parametric mapping: combining fMRI and MEG for Haberly, L. B., and Price, J. L. (1978). Association and commis-
high-resolution imaging of cortical activity. Neuron sural fiber systems of the olfactory cortex of the rat. J. Comp.
26(1):55–67. Neurol. 178:711–740.
Doty, R. L., Applebaum, S., Zusho, H., and Settle, R. G. (1985). Hahn, I., Scherer, P. W., and Mozell, M. M. (1994). A mass trans-
Sex differences in odor identification ability: a cross-cultural port model of olfaction. J. Theor. Biol. 167(2):115–128.
analysis. Neuropsychologia 23(5):667–672. Harrington, A. (1995). Unfinished business: models of laterality
Doty, R. L., Bromley, S. M., Hummel, T, and Moberg, P. (1997). in the nineteenth century. In Brain Asymmetry, R. J., Davidson
Laterality in human nasal chemoreception. In Cerebral and K. Hugdahl (Eds.). Bradford Press, pp. 3–29.
Asymmetries in Sensory and Perceptual Processing, S. Harris, L. J. (1995). The corpus callosum and hemispheric com-
Christman (Ed.). Amsterdam: North Holland Publishing Co., munication: An historical survey of theory and research. In
1997, pp. 497–542. Hemispheric Communication, F. L. Kitterle. (Ed.). Lawrence
Eskenazi, B., Cain, W. S., Novelly, R. A., and Friend, K. B. Erlbaum, pp. 1–61.
(1983). Olfactory functioning in temporal lobectomy patients. Henkin, R. I., Comiter, H., Fedio, P., and O’Doherty, D. (1977).
Neuropsychologia 21(4):365–374. Defects in taste and smell recognition following temporal
Eskenazi, B., Cain, W. S., Novelly, R. A., and Mattson, R. (1986). lobectomy. Trans. Am. Neurol. Assoc. 102:146–150.
Odor perception in temporal lobe epilepsy patients with and Holmes, G. (1918). Disturbances of vision by cerebral lesions.
without temporal lobectomy. Neuropsychologia 24(4): Br. J. Ophthal. 2:353–384.
553–562. Horton, J., and Hoyt, W. F. (1991). The representation of the
Eslinger, P. J., Damasio, A. R., and Van Hoesen, G. W. (1982). visual field in human striate cortex. Arch. Ophthalmol.
Olfactory dysfunction in man: anatomical and behavioral 109:816–824.
aspects. Brain Cogn. 1:259–285. Hounsfield, G. N. (1973). Computerized transverse axial scan-
Fox, P. T., and Raichle, M. E. (1986). Focal physiological uncou- ning (tomography). 1. Description of system. Br. J. Radiol.
pling of cerebral blood flow and oxidative metabolism during 46(552):1016–1022.
somatosensory stimulation in human subjects. Proc Natl Acad Huang, C. M., and Liu, G. L. (1991). Auditory responses in the
Sci USA 83(4):1140–1144. posterior vermis of the cat: the buried cerebellar cortex. Brain
Fox, P. T., Raichle, M. E., Mintun, M. A., and Dence, C. (1988). Res. 553(2):201–205.
Nonoxidative glucose consumption during focal physiologic Ingvar, G. H., and Risberg, J. (1965). Influence of mental activ-
neural activity. Science 241(4864):462–464. ity upon regional cerebral blood flow in man. Acta Neurol.
Frackowiak, R. S., Lenzi, G. L., Jones, T., and Heather, J. D. Scand. Suppl. 14:183–186.
(1980). Quantitative measurement of regional cerebral blood Ivry, R. (1997). Cerebellar timing systems. Int. Rev. Neurobiol.
flow and oxygen metabolism in man using 150 and positron 41:555–573.
emission tomography: theory, procedure, and normal values. Jackson, R. T. (1976). Nasal-cardiopulmonary reflexes: a role of
Comput. Assist. Tomogr. 4(6):727–736. the larynx. Ann. Otol. Rhinol. Laryngol. 85(1 Pt 1):65–70.
Friston, K. J., Jezzard, P., and Turner, R. (1994). Analysis of Jones-Gotman, M., and Zatorre, R. J. (1988). Olfactory identifi-
functional MRI time-series. Hum. Brain Mapp. 1:153–171. cation deficits in patients with focal cerebral excision.
Friston, K. J., Williams, S., Howard, R., Frackowiak, R. S., and Neuropsychologia 26(3):387–400.
Turner, R. (1996). Movement-related effects in fMRI time- Jones-Gotman, M., and Zatorre, R. J. (1993). Odor recognition
series. Magn. Reson. Med. 35(3):346–355. memory in humans: role of right temporal and orbitofrontal
Fulbright, R. K., Skudlarski, P., Lacadie, C. M., Warrenburg, S., regions. Brain Cogn. 22(2):182–198.
Bowers, A. A., Gore, J. C., and Wexler, B. E. (1998). Jones-Gotman, M., Zatorre, R. J., Cendes, F., Olivier, A.,
Functional MR imaging of regional brain responses to pleasant Andermann, F., McMackin, D., Staunton, H., Siegel, A. M.,
and unpleasant odors. Am. J. Neuroradiol. 19(9):1721–1726. and Wieser, H. G. (1997). Contribution of medial versus
Fulton, J. F. (1928). Observations upon the vascularity of the lateral temporal-lobe structures to human odour identification.
human occipital lobe during visual activity. Brain 51:310–320. Brain 120:1845–1856.
Gabrieli, J. D. E., Brewer, J. B., Desmond, J. E., and Glover, G. Josephs, O., and Henson, R. N. (1999). Event-related functional
H. (1997). Separate neural bases of two fundamental memory magnetic resonance imaging: modelling, inference and
processes in the human medial temporal lobe. Science optimization. Phil. Trans. R. Soc. Lond. B Biol. Sci.
276(5310):264–266. 354(1387): 1215–1228.
Functional Neuroimaging of Human Olfaction 271

Kettenmann, B., Jousmaki, V., Portin, K., Salmelin, R., Kobal, Mai, J. K., Assheuer, J., and Paxinos, G. (1997). Atlas of the
G., and Hari, R. (1996). Odorants activate the human superior Human Brain. Academic Press, New York.
temporal sulcus. Neurosci. Lett. 203(2):143–145. Malaspina, D., Perera, G. M., Lignelli, A., Marshall, R. S., Esser,
Kettenmann, B., Hummel, C., Stefan, H., and Kobal, G. (1997). P. D., Storer, S., Furman, V., Wray, A. D., Coleman, E.,
Multiple olfactory activity in the human neocortex identified Gorman, J. M., and Van Heertum, R. L. (1998). SPECT imag-
by magnetic source imaging. Chem. Senses 22:493–502. ing of odor identification in schizophrenia. Psychiatry Res.
Kinomura, S., Kawashima, R., Yamada, K., Ono, S., Itoh, M., 821:53–61.
Yoshioka, S., Yamaguchi, T., Matsui, H., Miyazawa, H., Itoh, Malonek, D., and Grinvald, A. (1997). Vascular regulation at sub
H., et al. (1994). Functional anatomy of taste perception in the millimeter range. Sources of intrinsic signals for high resol-
human brain studied with positron emission tomography. ution optical imaging. Adv. Exp. Med. Biol. 413:215–220.
Brain Res. 659(1–2):263–266. Mansfeld, G., and Tyukody, V. (1936). Atemzentrum und
Kobal, G., and Kettenmann, B. (1999). Cerebral representation of Narkose. Arch. Int. Pharmacodyn. 54:219.
odor perception. Adv. Neurol. 81:221–229. Moberg P. J., Agrin R., Gur R. E., Gur R. C., Turetsky B. I., and
Koizuka, I., Yano, H., Nagahara, M., Mochizuki, R., Seo, R., Doty R. L. (1999) Olfactory dysfunction in schizophrenia:
Shimada, K., Kubo, T., and Nogawa, T. (1994). Functional a qualitative and quantitative review. Neuropsycho-
imaging of the human olfactory cortex by magnetic resonance pharmacology 21(3):325–340.
imaging. J. Otorhinolaryngol. Relat. Spec. 56:273–275. Mosso, A. (1881). Ueber den Kreislauf des Blutes im
Kopala, L. C., Good, K. P., Torrey, E. F., and Honer, W. G. (1997) Menschlichen Gehirn. Verlag von Veit, Leipzig.
Olfactory function in monozygotic twins discordant for Mozell, M. M., Hornung, D. E., Leopold, D. A., and Youngentob,
schizophrenia. Am. J. Psychiatry 155:134–136. S. L. (1983). Initial mechanisms basic to olfactory perception.
Köster, E. P., and de Wik, R. A. (1991). Olfactory adption. In The Am. J. Otolaryngol. 4(4):238–245.
Human Sense of Smell, Laing, D. G., Doty R. L., Breiphol, W. Nordin S., and Murphy C. (1996) Impaired sensory and cognitive
(Eds.). Springer-Verlag, Berlin, pp. 199–215. olfactory function in questionable Alzheimer’s disease.
Laing, D. G. (1983). Natural sniffing gives optimum odor per- Neuropsychology 10:113–119.
ception for humans. Perception 12(2):99–117. O’Doherty, J., Rolls, E. T., Francis, S., Bowtell, R., McGlone, F.,
Landau, W. M., Freygang, W. H. J., Roland, L. P., Sokoloff, L., Kobal, G., Renner, B., and Ahne, G. (2000). Sensory-specific
and Kety, S. S. (1955). The local circulation of the living satiety-related olfactory activation of the human orbitofrontal
brain: Values in the unanesthetized and anesthetized cat. cortex. Neuroreport 11(4):893–897. (Corrected and repub-
Trans. Am. Neurol. Assoc. 80:125–129. lished article originally printed in Neuroreport 2000 Feb 7;
Lassen, N. A., Ingvar, D. H., and Skinhoj, E. (1978). Brain func- 11(2):399–403.)
tion and blood flow. Sci. Am. 239(4):62–71. Ogawa, S., Tank, D. W., Menon, R., Ellermann, J. M., Kim, S. G.,
Lauterbur, P. C. (1973) Image formation by induced local inter- Merkle, H., and Ugurbil, K. (1992). Intrinsic signal changes
actions. Examples employing nuclear magnetic resonance. accompanying sensory stimulation: functional brain mapping
Nature 242:190–191. with magnetic resonance imaging. Proc. Natl. Acad. Sci. USA
LeMagnen, J. (1945). Etude des facteurs dynamiques de l’excita- 89(13):5951–5955.
tion olfactive. L’annee Psychol. 44–45:77–89. Pauling, L., and Coryell, C. D. (1936). The magnetic properties
Levy, L. M., Henkin, R. I., Hutter, A., Lin, C. S., Martins, D., and and structure of hemoglobin, oxyhemoglobin, and caron-
Schellinger, D. (1997). Functional MRI of human olfaction. monoxyhemoglobin. Proc. Natl. Acad. Sci. USA 22:210–216.
J. Comput. Assist. Tomogr. 21(6):849–856. Poellinger, A., Thomas, R., Lio, P., Lee, A., Makris, N., Rosen, B.
Levy, L. M., Henkin, R. I., Lin, C. S., Hutter, A., and Schellinger, R., and Kwong, K. K. (2001). Activation and habituation in
D. (1998a). Increased brain activation in response to odors in olfaction—an fMRI study. Neuroimage 13(4):547–560.
patients with hyposmia after theophylline treatment demon- Posner, M. I., and Raichle, M. E. (1998). The neuroimaging of
strated by fMRI. J. Comput. Assist. Tomogr. 22(5):760–770. human brain function. Proc. Natl. Acad. Sci. USA
Levy, L. M., Henkin, R. I., Lin, C. S., Hutter, A., and Schellinger, 95(3):763–764.
D. (1998b). Mapping brain activation to odorants in patients Prabhakaran, V., Sobel, N., Desmond, J., Glover, G., Goode, R.
with smell loss by functional MRI. J. Comput. Assist. Tomogr. L., Sullivan, E., and Gabrieli, J. D. E. (1999). Hedonic value
22(1):96–103. of odors is reflected in the level of activation in the amygdala
Levy, L. M., Henkin, R. I., Lin, C. S., and Finley, A. (1999). (abstr). Neuroimage 5:s198.
Rapid imaging of olfaction by functional MRI (fMRI): iden- Prah, J. D., Sears, S. B., and Walker, J. C. (1995). Modern
tification of presence and type of hyposmia. Comput. Assist. approaches to air dilution olfactometry. In Handbook of
Tomogr. 23(5):767–775. Olfaction and Gustation, R. L. Doty (Ed.). Marcel Dekker,
Litaudon, P., Datiche, F., and Cattarelli, M. (1997). Optical New York, pp. 227–256.
recording of the rat piriform cortex activity. Prog. Neurobiol. Price, J. L. (1973). An autoradiographic study of complementary
52(6):485–510. laminar patterns of termination of afferent fibers to the olfac-
Lorig, T. S., Elmes, D. G., Zald, D. H., and Pardo, J. V. (1999). A tory cortex. J. Comp. Neurol. 150:87–108.
computer-controlled olfactometer for fMRI and electrophysi- Price, J. L. (1985). Beyond the primary olfactory cortex: olfac-
ological studies of olfaction. Behav. Res. Methods Instrum. tory-related areas in the neocortex, thalamus and hypothala-
Comput. 31(2):370–375. mus. Chem. Senses 10:239–258.
272 Sobel et al.

Price, J. L. (1990). Olfactory system. In The Human Nervous Sakuma, K., Kakigi, R., Kaneoke, Y., Hoshiyama, M., Koyama,
System, G. Paxinos (Ed.). Academic Press, San Diego, pp. S., Nagata, O., Takeshima, Y., Ito, Y., and Nakashima, K.
979–1001. (1997). Odorant evoked magnetic fields in humans. Neurosci.
Price, J. L., and Sprich, W. W. (1975): Observation on the lateral Res. 27(2):115–122.
olfactory tract of the rat. J. Comp. Neurol. 162:321–336. Savic, I., and Gulyas, B. (2000). PET shows that odors are
Price, J. L., Russchen, F. T., and Amaral, D. G. (1987). The lim- processed both ipsilaterally and contralaterally to the stimu-
bic region. II. The amygdaloid complex. In Handbook of lated nostril. Neuroreport 11(13):2861–2866.
Chemical Neuroanatomy, vol. 5, A. Bjorklund, T. Hokfelt, and Savic, I., Gulyas, B., Larsson, M., and Roland, P. (2000).
L. W. Swanson (Eds.). Elsevier, Amsterdam, pp. 279–388. Olfactory functions are mediated by parallel and hierarchical
Purcell, E. M., Torry, H. C., and Pound, R. V. (1946). Resonance processing. Neuron 26(3):735–745.
absorption by nuclear magnetic moments in a solid. Physiol. Savic, I., Berglund, H., Gulyas, B., and Roland, P. (2001).
Rev. 69:37 Smelling of odorous sex hormone-like compounds causes
Qureshy A., Kawashima R., Imran M. B., Sugiura M., Goto R., sex-differentiated hypothalamic activation in humans. Neuron
Okada K., Inoue K., Itoh M., Schormann T., Zilles K., and 31(4):661–668.
Fukuda H. (2000) Functional mapping of human brain in Shipley, M. T., and Adamek, G. D. (1984). The connections of the
olfactory processing: a PET study. J. Neurophysiol. mouse olfactory bulb: a study using orthograde and, retro-
84(3):1656–1666. grade transport of wheat germ agglutinin conjugated to horse-
Raichle, M. E. (1998). Behind the scenes of functional brain radish peroxidase. Brain Res. Bull. 12:669–688.
imaging: a historical and physiological perspective. Proc. Small, D. M., Jones-Gotman, M., Zatorre, R. J., Petrides, M., and
Natl. Acad. Sci. USA 95(3):765–772. Evans, A. C. (1997a). A role for the right anterior temporal
Raichle, M. E., Martin, W. R., Herscovitch, P., Mintun, M. A., lobe in taste quality recognition. J. Neurosci.
and Markham, J. (1983). Brain blood flow measured with 17(13):5136–5142.
intravenous H2(15)O. II. Implementation and validation. J. Small, D. M., Jones-Gotman, M., Zatorre, R. J., Petrides, M., and
Nucl. Med. 24(9):790–879. Evans, A. C. (1997b). Flavor processing: more than the sum
Rausch, R., and Serafetinides, E. A. (1975). Specific alterations of its parts. Neuroreport 8(18):3913–3917.
of olfactory function in humans with temporal lobe lesions. Small, D. M., Zald, D. H., Jones-Gotman, M., Zatorre, R. J.,
Nature 255(5509):557–855. Pardo, J. V., Frey, S., and Petrides, M. (1999). Human cortical
Rehn, T. (1978). Perceived odor intensity as a function of air flow gustatory areas: a review of functional neuroimaging data.
through the nose. Sens. Proc. 2(3):198–205. Neuroreport 10(1):7–14.
Rolls, E. T. (1996). The orbitofrontal cortex. Phil. Trans. R. Soc. Sobel, N., and Brown, W. M. (2001). The scented brain: phere-
Lond. B Biol. Sci. 351(1346):1433–1444. monal responses in humans. Neuron 31(4):512–514.
Rolls, E. T., Critchley, H. D., and Treves, A. (1996). Sobel, N., Prabhakaran, V., Desmond, J., Glover, G., Sullivan, E.,
Representation of olfactory information in the primate and Gabrieli, J. D. E. (1997a). A method for functional mag-
orbitofrontal cortex. Neurophysiology 75(5):1982–1996. netic resonance imaging of olfaction. J. Neurosci. Meth.
Roy, C. S., and Sherrington, C. S. (1890). On the regulation of the 78:115–121.
blood supply of the brain. J. Physiol. (London) 11:85–108. Sobel, N., Prabhakaran, V., Desmond, J., Glover, G., Sullivan,
Royet, J. P., Koenig, O., Gregoire, M. C., Cinotti, L., Lavenne, E., and Gabrieli, J. D. E. (1997b). Separate cerebellar
F., Le Bars, D., Costes, N., Vigouroux, M., Farget, V., Sicard, components subserve sniffing and smelling. Society for
G., Holley, A., Mauguiere, F., Comar, D., and Froment, J. C. Neuroscience 27TH Annual Meeting, Part 2, Abstract 806.
(1999). Functional anatomy of perceptual and semantic pro- 15, p. 2077.
cessing for odors. J. Cogn. Neurosci. 11(1): 94–109. Sobel, N., Prabhakaran, V., Desmond, J., Glover, G., Goode, R.
Royet J. P., Zald D., Versace R., Costes N., Lavenne F., Koenig O., L., Sullivan, E., and Gabrieli, J. D. E. (1998a). Sniffing and
and Gervais R. (2000). Emotional responses to pleasant and smelling: separate subsystems in the human olfactory cortex.
unpleasant olfactory, visual, and auditory stimuli: a positron Nature 392:282–286.
emission tomography study. J. Neurosci. 20(20):7752–7759. Sobel, N., Prabhakaran, V., Hartley, C. A., Desmond, J., Zhao, Z.,
Royet, J. P., Hudry, J., Zald, D. H., Godinot, D., Gregoire, M. C., Glover, G., Gabrieli, J. D. E., and Sullivan, E. (1998b).
Lavenne, F., Costes, N., and Holley, A. (2001). Functional Odorant-induced and sniff-induced activation in the cerebel-
neuroanatomy of different olfactory judgments. Neuroimage lum of the human. J. Neurosci. 18:8990–9001.
13(3):506–519. Sobel, N., Prabhakaran, V., Hartley, C. A., Desmond, J. E.,
Rubin, B. D., and Katz, L. C. (1999). Optical imaging of odorant Glover, G. H., Sullivan, E. V., and Gabrieli, J. D. E. (1999).
representations in the mammalian olfactory bulb. Neuron 3: Blind smell: brain activation induced by an undetected air-
499–511. borne chemical. Brain 122:209–217.
Russchen, F. T., Amaral, D. G., and Price, J. L. (1987). The affer- Sobel, N., Khan, R. M., Hartley, C. A., Sullivan, E. V., and
ent input to the magnocellular division of the mediodorsal Gabrieli, J. D. E. (2000a). Sniffing longer rather than stronger
thalamic nucleus in the monkey, Macaca fascicularis. Comp. to maintain olfactory detection threshold. Chem. Senses
Neurol. 256(2):175–210. 25(1):1–8.
Functional Neuroimaging of Human Olfaction 273

Sobel, N., Khan, R. M., Saltman, A., Sullivan, E. V., and Gabrieli, Youngentob, S. L., Mozell, M. M., Sheehe, P. R., and Hornung, D.
J. D. E. (2000b). The world smells different to each nostril. E. (1987). A quantitative analysis of sniffing strategies in rats
Nature 402(6757):35. performing odor detection tasks. Physiol. Behav. 41(1): 59–69.
Sobel, N., Prabhakaran, V., Zhao, Z., Desmond, J. E., Glover, G. Yousem, D. M., Williams, S. C., Howard, R. O., Andrew, C.,
H., Sullivan, E. V., and Gabrieli, J. D. E. (2000c). Time course Simmons, A., Allin, M., Geckle, R. J., Suskind, D., Bullmore,
of odorant-induced activation in the human primary olfactory E. T., Brammer, M. J., and Doty, R. L. (1997a). Functional
cortex. J. Neurophysiol. 83(1):537–551. MR imaging during odor stimulation: preliminary data.
Stein, J. F., and Glickstein, M. (1992). Role of the cerebellum in Radiology 204:833–838.
visual guidance of movement. Physiol. Rev. 72(4): 967–1017. Yousem, D. M., Williams, S. C. R., Simmons, A., Doty, R. L., and
Stopfer, M., and Laurent, G. (1999). Short-term memory in olfac- Kroger, H. (1997b). Functional magnetic resonance imaging
tory network dynamics. Nature 402(6762):664–668. using olfactory stimulants. International Symposium on Smell
Takagi, S. F. (1986). Studies on the olfactory nervous system of and Taste, San Diego, CA, July 7–13.
the Old World monkey. Prog. Neurobiol. 27(3):195–250. Yousem, D. M., Geckle. R. J., and Doty, R. L. (1998).
Tanabe, T., Iino, M., and Takagi, S. F. (1975). Discrimination of Differences between olfactory and trigeminally mediated
odors in olfactory bulb, pyriform-amygdaloid areas, and simulants on FMRI studies. ASNR, Philadelphia, May 20.
orbitofrontal cortex of the monkey. J. Neurophysiol. 38(5): Yousem, D. M., Maldjian, J. A., Hummel, T., Alsop, D. C.,
1284–1296. Geckle, R. J., Kraut, M. A., and Doty, R. L. (1999a). The
Tazawa, Y., Onoda, N., and Takagi, S. F. (1987). Olfactory input effect of age on odor-stimulated functional MR imaging. Am.
to the lateral hypothalamus of the old world monkey. J. Neuroradiol. 20:600–608.
Neurosci. Res. 4(5):357–375. Yousem, D. M., Maldjian, J. A., Siddiqi, F., Hummel, T., Alsop,
Teghtsoonian, R., Teghtsoonian, M., Berglund, B., and Berglund, D. C., Geckle, R. J., Bilker, W. B., and Doty, R. L. (1999b).
U. (1978). Invariance of odor strength with sniff vigor: an Gender effects on odor-stimulated functional magnetic reso-
olfactory analogue to size constancy. J. Exp. Psychol. Hum. nance imaging. Brain Res. 818:480–487.
Percept. Perform. 4(1):144–152. Zald, D. H., and Kim, S. W. (1996a). Anatomy and function of
Ter-Pogossian, M. M., Phelps, M. E., Hoffman, E. J., and the orbital frontal cortex, I: Anatomy, neurocircuitry; and
Mullani, N. A. (1975). A positron-emission transaxial tomo- obsessive-compulsive disorder. Neuropsychiatry Clin.
graph for nuclear imaging (PETT). Radiology 4(1): 89–98. Neurosci. 8(2):25–38.
Turetsky, B. I., Moberg, P. J., Yousem, D. M., Doty, R. L., Arnold, Zald, D. H., and Kim, S. W. (1996b). Anatomy and function of
S. E., and Gur R. E. (2000). Reduced olfactory bulb volume the orbital frontal cortex, II: Function and relevance to obses-
in patients with schizophrenia. Am. J. Psychiatry sive-compulsive disorder. J. Neuropsychiatry Clin. Neurosci.
157(5):828–830. 8(3):249–261.
Volkow, N. D., Rosen, B., and Farde, L. (1997). Imaging the liv- Zald, D. H., and Pardo, J. V. (1997). Emotion, olfaction, and the
ing human brain: magnetic resonance imaging and positron human amygdala: amygdala activation during aversive olfac-
emission tomography. Proc. Natl. Acad. Sci. USA 94(7): tory stimulation. Proc. Natl. Acad. Sci. USA 94: 4119–4124.
2787–2788. Zald, D. H., and Pardo, J. V. (2000). Functional neuroimaging of
Warren, D. W., Walker, J. C., Drake, A. F., and Lutz, R. W. the olfactory system in humans. Int. J. Psychophysiol. 36(2):
(1992). Assessing the effects of odorants on nasal airway size 165–181.
and breathing. Physiol. Behav. 51(2):425–430. Zald, D. H., Donndelinger, M. J., Pardo, J. V. (1998).
Warren, D. W., Walker, J. C., Drake, A. F., and Lutz, R. W. Elucidating dynamic brain interactions with across-subjects
(1994). Effects of odorants and irritants on respiratory behav- correlational analyses of positron emission tomographic
ior. Laryngoscope 104(5 Pt 1):623–626. data: the functional connectivity of the amygdala and
Williams, R. W., and Herrup, K. (1988). The control of neuron orbitofrontal cortex during olfactory tasks. J. Cerebr. Blood
number. Ann. Rev. Neurosci. 11:423–453. Flow Metab. 18: 896–905.
Yang, Q. X., Dardzinski, B. J., Li, S., Eslinger, P. J., and Smith, Zarahn, E., Aguirre, G. K., and D’Esposito, M. (1997). Empirical
M. B. (1997). Multi-gradient echo with susceptibility inho- analyses of BOLD fMRI statistics. I. Spatially unsmoothed
mogeneity compensation (MGESIC): demonstration of fMRI data collected under null-hypothesis conditions. Neuroimage
in the olfactory cortex at 3.0 T. Magn. Reson. Med. 37: 5(3):179–197.
331–335. Zatorre, R. J., and Jones-Gotman, M. (1991). Human olfactory
Yang, X., Renken, R., Hyder, F., Siddeek, M., Greer, C. A., discrimination after unilateral frontal or temporal lobectomy.
Shepherd, G. M., and Shulman, R. G. (1998). Dynamic map- Brain 114(Pt 1A):71–84.
ping at the laminar level of odor-elicited responses in rat Zatorre, R. J., Jones-Gotman, M., Evans, A. C., and Meyer, E.
olfactory bulb by functional MRI. Proc. Natl. Acad. Sci. USA (1992). Functional localization and lateralization of human
95(13):7715–7720. olfactory cortex. Nature 360:339–340.
Yarita, H., Iino, M., Tanabe, T., Kogure, S., and Takagi, S. F. Zatorre R. J., Jones-Gotman, M., and Rouby C. (2000). Neural
(1980). A transthalamic olfactory pathway to orbitofrontal mechanisms involved in odor pleasantness and intensity judg-
cortex in the monkey. J. Neurophysiol. 43(1):69–85. ments. Neuroreport 11(12):2711–2716.
13

Structure–Odor Relationships: A Modern Perspective

Luca Turin
University College, London, United Kingdom

Fumiko Yoshii
Niigata University, Niigata, Japan

I. INTRODUCTION necessarily shared by others. It is striking how few

experiments in which odorants are applied to biological
This review is intended as an introduction for nonspecial- preparations take into account the perceived odor of the
ists to structure–odor relationships (SOR), and as a molecules. We hope that biologists will realize that, once
critique of the field rather than a compendium. The a vocabulary is agreed upon, odor is as reliable a sensation
perspective will be that of biology rather than fragrance as pitch or color.
chemistry. In other words, we are more interested in what
SORs tell us about the mechanisms of human olfaction
than about the synthetic chemistry of odorants. We believe II. THE CURRENT STATE OF SORS
that the recent advances (see Mombaerts, 1999a, for
review) that followed Buck and Axel’s 1991 discovery of Chemists have, by design and by accident, been producing
odorant receptors will someday make odorant design a odorants since the dawn of organic chemistry 200 years
rational process. In the meantime, we want to highlight a ago, and a vast database of odorants and their correspond-
few salient findings which we feel a successful SOR the- ing odor profiles has built up. This seems a good place to
ory must account for, in the hope that this will help state what is perhaps the most surprising fact of SORs: no
researchers design experiments to elucidate the mystery two odorants have ever been found to have exactly the
of primary olfactory reception. same odor. Despite figures often mentioned in the litera-
A perennial difficulty of structure–odor relationships ture of “a few thousand,” as far as we know, the resolution
has been that both structure and odor have proved hard to of the human olfactory system is infinite.
pin down. Considered as a structure-activity problem, The field of fragrance synthesis, though still small in
olfaction is several orders of magnitude more complicated comparison to, say, pharmaceuticals, is an $8 billion
than its conventional pharmacological counterparts industry dominated by a few large firms: in alphabetical
because there are many more structures and a vast number order, Dragoco (Germany), Firmenich (Switzerland),
of odors. There is also an additional problem: as a sensa- Givaudan-Roure (Switzerland), Haarmann & Reimer
tion, olfaction does not seem to enjoy the same status as, (Germany), International Flavors and Fragrances
say, vision. Most biologists—in fact, most people not (United States), Quest (United Kingdom), and Takasago
directly involved with fragrances or flavors—seem to (Japan). Each of these firms has a library of tens of thou-
think that odor sensation is “subjective” and not sands of odorants. Understandably, most of these data

275
276 Turin and Yoshii

Figure 1 (Left) Ethyl citronellyl oxalate, a molecule possessing a macrocyclic musk odor but linear in shape. (Right) A macrocyclic
musk, cyclopentadecanolide. Shape-based theories assume that the linear musk assumes a conformation close to that of the macro-
cyclic when binding to the receptor, hence the similarity in odor.

are proprietary and not available to the scientific is unknown, make it very difficult to apply conventional
community. quantitative structure-activity relationships. QSARs have
Nevertheless, many hundreds of odorants have been proved very useful in many areas of pharmacology (Balbes
described in the literature, and their SORs have been et al., 1994; Dearden and James, 1998). They work best
extensively reviewed, most recently by Rossiter (1996). when the structure of the site to which the molecule binds
Most reviews of SORs are collections of disparate facts is known exactly from crystallographic measurements.
with no unifying theme save a basic postulate: odor must Then the full force of computational chemistry can be
be related to molecular structure. The search for a predic- brought to bear on designing molecules. Some studies
tive theory based on this assumption has been frustrating: have attempted to calculate both the three-dimensional
Bedoukian (1966) stated that “it is not possible to predict structure of the receptor and its interaction with odorants
the odor of a substance with any degree of accuracy.” (Floriano et al., 2000; Singer, 2000). These studies will
McCartney (1968) felt that “the difficulties in the way of undoubtedly become increasingly useful as our knowledge
uncovering the connection [between structure and odor] of receptor structure increases and modeling techniques
have been very great.” Hornstein and Teranishi (1967) con- become more realistic. In the meantime, most of the work
sidered the results of such searches “disappointing.” More proceeds by examining the structures of the odorants
recently, Frater et al. (1998) described the state of SORs as alone. It is not clear how many odorants have been
“sorry.” Indeed Sell (1999) recently suggested that there designed using QSAR alone, or even as a principal tool to
may be no connection at all between structure and odor, guide synthesis. Fragrance companies are reluctant to dis-
and that the wiring from receptor to brain may be arbitrary. cuss the subject. Perhaps the best indication of this is that
The reader interested in getting a feel for the fascinating new odorant synthesis in the firms still proceeds by trial
regularities and irregularities of the structure–odor map is and error. It is our impression that QSAR has strong com-
referred to the excellent review by Boelens (1974) and the petition, particularly from combinatorial chemistry tech-
monograph by Ohloff (1991). niques that now make it easier to synthesize large numbers
Attempts have been made to accommodate discrepant of molecules.
structure–odor relationships by a process known as con-
formational analysis (Yoshii et al. 1994). This involves
exploring the space of conformations adopted by the odor- III. WHAT MAKES AN ODORANT?
ant molecule when deformed away from its energy mini-
mum. The fraction of configuration space allowed depends The general requirements for an odorant are that it should
on the energy arbitrarily assigned to molecular motions. be volatile, hydrophobic, and have a molecular weight less
The value of conformational analysis is unclear since it is than approximately 300 daltons. Ohloff (1994) has stated
usually a directed process in which the molecule is bent that the largest known odorant is a labdane, with a
purposely to resemble another odorant. An example of this molecular weight of 296. The first two requirements make
is given by the study of linear musk citronellyl oxalate physical sense, for the molecule has to reach the nose* and
(Yoshii et al. 1994), whose lowest-energy conformer may need to cross membranes. The size requirement
resembles a macrocyclic musk. At room temperature,
however, the linear musk must also also explore a vast
range of conformations, which resemble dozens of other *Note that some hydrophobic compounds of low volatility can
odorants (Fig. 1). reach the nose from the bloodstream. The garlicky smell of IV
The complexity of structure–odor relationships, and the thiopental is perceived by anesthesia subjects seconds before they
fact that the three-dimensional structure of the receptor site lose consciousness.
Structure–Odor Relationships 277

Figure 2 Comparison of molecular size between a benzenoid musk (left) derived from acetophenone and its sila counterpart (right)
in which the central carbon atom in the t-butyl groups has been replaced with Si. The carbon musk is a strong odorant, the sila musk
odorless.

appears to be a biological constraint. To be sure, vapor A. Odor Descriptors and Odor Profiles
pressure (volatility) falls rapidly with molecular size, but
that cannot be the reason why larger molecules have no Odor descriptors are the words that come to mind when
smell, since some of the strongest odorants (e.g., some smelling a substance. The more generally understood the
steroids) are large molecules. In addition, the cut-off is words are, the more useful they are as descriptors. An
very sharp indeed: for example, substitution of the slightly untrained observer may use, for example, “Grandma’s
larger silicon atom for a carbon in a benzenoid musk caus- linen cupboard” as an accurate descriptor, whereas the pro-
es it to become odorless (Wrobel and Wannagat, 1982d) fessional would be more analytical and say woody (the
(Fig. 2). cupboard), musky (the linen), or camphoraceous (the
A further indication that the size limit has something to mothballs). Note that these descriptors may be applied to a
do with the chemoreception mechanism comes from the single, pure odorant. Nevertheless, odor description
fact that specific anosmias become more frequent as mol- always works by analogy since there is no objective alter-
ecular size increases. At the “ragged edge” of the size native. Odor description seems to have acquired the repu-
limit, subjects become anosmic to large numbers of mole- tation of being arcane, even fanciful, perhaps in part as a
cules. An informal poll among perfumers, for example, has result of the hoopla surrounding fine wines and fragrances.
elicited the fact that most of them are completely anosmic In practice, it is easy for any observer, after a little train-
to one or more musks (e.g., Galaxolide®; MW 244.38) or, ing, to use the standard descriptors of fragrance chemistry.
less commonly, ambergris odorants such as Ambrox® or Accordingly, almost all the examples in this review are
the larger esters of salicylic acid (Fig. 3). chosen from among those commercially available, and we
One can probably infer from this that the receptors can- urge the interested reader to obtain some of them and
not accommodate molecules larger than a certain size and check the odor. Anosmias aside, outright disagreements
that this size is genetically determined (Whissel-Buechy and between observers are, in our experience, rare. One excep-
Amoore, 1973) and varies from individual to individual. tion is Karanal® an ambergris odorant that is perceived as
animalic by some observers (C. Sell, personal communi-
cation) (Fig. 4). Another is trans-2-hexenal, perceived as
green by some (Arctander, 1991) and bitter almond by oth-
ers (Ohloff, 1994).
The much more common and oft-quoted cases of per-
ceptual disagreements, e.g., phenylacetic acid, are prob-
ably due to ambiguity, not difference. Phenylacetic acid
Figure 3 Two molecules that are occasionally odorless to smells both of honey and of fresh urine. When asked to
humans-galaxolide (MW 244.38) and Ambrox (MW 236.40). use either descriptor, subjects will opt for one or the
278 Turin and Yoshii

provide researchers with samples. For those wishing to

delve more deeply into the subject, Arctander’s hand-
book (Arctander, 1994) lists thousands of molecules and
their odor profiles and represents a mine of reliable and
largely untapped information on SORs. Unfortunately,
the chemical structure drawings in Arctander are anti-
quated and often unclear, and the book contains no
descriptor index. In addition, two companies
Figure 4 Two molecules whose odor appears to differ between
observers. Karanal (left) is a woody-amber to most observers but (Leffingwell and Boelens) offer independent informa-
smells unpleasantly urinous to some. Trans-2-hexenal (right) is tion on fragrances and flavors at www.leffingwell.com
described in the literature either as a green (Arctander, 1994) or and www.xs4all.nl/~bacis.
bitter almonds (Ohloff, 1994) odorant. To the authors, it smells of
bitter almonds.

B. Some Odor Categories and Their

Representative Molecules, Chosen to
other without hesitation. When asked whether the other
Illustrate Structural Diversity
descriptor might also apply, however, they will always
agree that there is a honey or urine “side” to the smell.
1. Musk
This is not so strange when one considers a color analo-
gy. Ask a group whether an appropriate shade of Musk is perhaps the most famous of all odor categories
turquoise is blue or green, and you may get half giving because of its universal inclusion in fragrance and its
each answer. This does not mean that they perceive it exotic origin in the secretions of the musk deer. In fact,
differently. because of expense and legislation, musks have been
The reader wishing to become familiar with odorants synthetic for a long time. Musk odor descriptors might be
and their descriptors can peruse Aldrich’s Flavors and “smooth, clean, sweet, and powdery.” The molecules that
Fragrances catalog in which odorants are listed by chem- possess this odor character are exceptionally diverse in
ical type and by principal descriptor. Kits of esters and structure. Macrocyclic musks contain a 15- to 18-carbon
heterocycles are also available from the same firm, cycle closed either by a carbonyl or by a lactone and smell
which provide an excellent introduction to the raw data similar but fresher and more natural, often with fruity over-
of SORs, i.e., structure and odor. It is unfortunate that tones (cyclopentadecanolide, ambrettolide). Nitro musks,
the vast majority of commercial odorants are not repre- discovered originally as a byproduct of explosives chem-
sented in catalogs of chemical suppliers familiar to the istry, smell sweeter and are reminiscent of old-fashioned
biologist. Nevertheless, fragrance firms will on request barbershop smells (Fig. 5).

Figure 5 Representatives from five chemical classes that yield musk odors: (1) androst-16-en-3-ol, a steroid musk; (2) ambrettolide,
a macrocyclic musk; (3) Musk Bauer, a nitro musk; (4) Tonalid, a tetralin musk; (5) Traseolide, an indane musk.
Structure–Odor Relationships 279

2. Ambergris
Originally derived from concretions spat out by whales
and aged in the sun, ambergris odorants smell nothing like
natural ambergris tincture, which has a weak animalic
marine smell. The smell of ambergris odorants was once
aptly described to us by a chemist-perfumer as “glorified
isopropanol.” Ambergris odorants are of interest to the stu-
dent of SORs because they provide an interesting combi-
nation of very closely related smells with widely different
structures: amberketal, timberol, karanal, and cedramber Figure 8 Some examples of green odorants. Clockwise from
top left: cis-3-hexenol, ligustral, nonadienal, and ethyl-
are close enough that a perfumer will occasionally mistake methoxypyrazine.
them for each other (Fig. 6).

3. Camphoraceous
Camphoraceous (mothball) notes are seldom used in per-
fumery, but they are of interest of SORs because they
formed the basis for one of the early attempts at smell clas-
sification by Amoore (1971). Camphor, cyclooctane, cine-
ole are good examples of camphoraceous smells and smell Figure 9 Two bitter-almond odorants: benzaldehyde and
rather similar to each other (Fig. 7). hydrogen cyanide.

4. Green
This category includes cut grass, fresh green bean notes
with a sharp, almost aggressive feel. Diverse compounds 5. Bitter Almonds
possess this descriptor, ranging from classic grassy notes
This easily recognized category is interesting to students of
of cis-3-hexenol and ligustral to the cucumber peel of
SORs because it includes a small molecule (HCN) that
nonadienal and the bell-pepper green note of some
smells metallic, not almond-like, to a large fraction of
pyrazines (Fig. 8).
observers (Fig. 9). Benzaldehyde, nitrobenzene, and trans-
2-hexenal (but see above) are good examples.

6. Other Categories
Many other categories such as musty, spicy, aldehydic, lac-
tonic, indolic, and marine exist, each with subdivisions. It
must be emphasized that the odor categories above are
merely convenient descriptors and only cover a very small
fraction of odor “space.” In fact, especially when one steps
Figure 6 Two ambergris odorants: timberol (left) and cedram- out of perfumery materials proper into smells noticed by
ber (right). chemists in the course of organic and inorganic syntheses,
the most frequent descriptor appears to be sui generis, i.e.,
a smell associated with nothing in particular.

IV. PLAUSIBLE THEORIES OF ODOR

Many theories of SORs have been proposed in the past

(reviewed in Moncrieff, 1951), but advances in biological
understanding, not least the discovery of odorant receptors,
have gradually ruled them out. Leaving aside the pes-
Figure 7 Three camphoraceous odorants: (left) 1,8-cineole, simistic view outlined above, according to which there
(center) camphor, and (right) cyclooctane. may be no relationship between structure and odor, there
280 Turin and Yoshii

appear to be two possible types of SOR theory left stand- about the number of “subsites” (odotopes) and their vari-
ing. One is based on fragments of molecular shape or ability, combined with calculations of the energetics of
odotopes (Mori and Shepherd, 1994), the other on molec- binding. Assuming a very low affinity of 105 M 1 for
ular vibrations (Turin, 1996, 2002). odorant binding, they arrive at the conclusion that in order
to achieve recognition, 300–1000 receptors are needed, in
A. Shape-Based Theories: Odotopes line with current estimates of receptor number (see Sec. VI
for further discussion of this point). Higher affinities, more
Most enzyme-substrate and receptor-ligand binding relies on consistent with olfactory thresholds, lead to greater still
molecular recognition between protein and ligand. receptor numbers.
Recognition depends on interactions that can be either attrac-
tive or repulsive (Davies and Timms, 1998). All attractive
chemical interactions are ultimately electrostatic in nature, B. Vibration Theories
whether they occur between fixed charges, dipoles, induced
dipoles, or atoms able to form weak electron bonds (e.g., The idea that the nose operates as a vibrational spectro-
hydrogen bonds). Repulsive interactions can be electrostatic scope was first proposed by Dyson (1938) and later taken
or quantum-mechanical (electron shell exchange repulsion). up and refined by Wright (1982). What makes it attractive
Almost every change in molecular structure (with some in principle is that vibrational spectra share three proper-
exceptions, described below) alters the set of surface features ties with human olfaction: (1) no two molecular spectra are
capable of forming such attractive or repulsive interactions exactly alike, particularly in the aptly named “fingerprint
and thus affects what we loosely call molecular shape. region”; (2) many functional groups are easily identified
The range of known molecular recognition mechanisms by their specific vibrational frequencies (and by smell, see
in biology is vast. At one extreme might be a vast set of below); and (3) a system utilizing a physical property as
immune- type receptors, each able to bind to a single odor- basic as vibration will be ready for never-before-smelled
ant molecule. At the other end of the spectrum, some bind- molecules, i.e., does not depend on a repertory of existing
ing sites such as those of odorant-binding proteins or expected structures. In that sense, it does not rely on
(Bianchet et al., 1996), albumins (Curry et al., 1999) and molecular recognition.
cytochromes P450 (Lawton and Philpot, 1993) are rather Several difficulties beset vibration theories and ulti-
nonspecific. When odorant receptors were first identified, mately caused their demise 20 years ago:
their large number was taken by some as evidence for
immune-like recognition. However, in vivo and, more 1. Enantiomers, which have identical vibrational
recently, in vitro studies have shown (Firestein et al., 1993; spectra in solution, sometimes have different odors
Duchamp-Viret et al., 1999; Malnic et al., 1999) that, with (see Boelens and van Gemert, 1993). Wright coun-
one notable exception (Wetzel et al., 1999), receptors tered this by emphasizing that while laboratory spec-
respond to more than one odorant, suggesting that they troscopes were achiral, and thus unable to distinguish
detect the presence not of the whole molecule but of a par- between enantiomers, a protein receptor would be
tial structural feature thereof, hence odotopes. intrinsically chiral and would thus respond different-
According to odotope theory, the smell of a molecule is ly to enantiomers. A modified version of this argu-
then due to the pattern, i.e., the relative excitation of a ment is described in Section V.C.
number N of receptors to which it binds. Even if one 2. No mechanism was ever found for a plausible pro-
assumes that receptors are only on or off, this scheme gives tein-based spectroscope, infrared optics being obvi-
considerable combinatorial room. Consider, for instance, a ously out of the question.
molecule having 20 exposed atoms and assume that each 3. Wright assumed that receptors were mechanical
odotope involves three of these. A binary (on-off) one- vibration sensors and that the receptors in the nose
odotope recognition system would then be able to detect would only be able to feel vibrations excited by
1140 molecules. If odotopes involved four atoms, the num- thermal motions at body temperature. He then
ber would rise to 4850, etc. Combining odotopes, and restricted his search for correlations between struc-
adding to this basic scheme a measure of intensity of exci- ture and odor to the region below 600 cm 1. These
tation for each receptor, clearly enables it to detect a vast were somewhat unconvincing and appeared to have
number of odorants. If the large number of odorant recep- little predictive value.
tors is taken to represent odotope categories, the combina-
torial possibilities become astronomical. The situation changed a few years ago with the proposal
A more sophisticated argument has been made by that electron tunneling might be a plausible mechanism
Lancet et al. (1993). They make plausible assumptions enabling proteins to act as vibrational spectroscopes.
Structure–Odor Relationships 281

C. A Biological “Spectroscope” tunneling can occur only if there is an energy level in the
source with energy E above that in the sink. In other
Inelastic electron tunneling spectroscopy (IETS) is a words, tunneling occurs only when a molecular vibra-
nonoptical form of vibrational spectroscopy (Jaklevic and tional energy E matches the energy difference between the
Lambe, 1966; Hansma, 1982; Adkins and Phillips, 1985). energy level of the donor and the energy level of the
It relies on the interaction between electrons tunneling acceptor. The receptor then operates as a spectrometer,
across a narrow gap between metallic electrodes. When the which allows it to detect a single well-defined energy, E.
gap is empty, tunneling electrons cross the gap at constant If there are several vibrational modes, which one(s) get
energy, and the tunneling current is proportional to the excited will depend on the relative strengths of the
overlap between filled and empty electronic states in the coupling. That may be expected to depend on, among
metals. If a molecule is present in the gap, tunneling elec- other things, the partial charges on the atoms and the rel-
trons will be scattered by the partial charges on the mole- ative orientation of the charge movements with respect to
cule’s constituent atoms and lose energy to the molecule the electron tunneling path (Fig. 10).
by exciting one of its vibrational modes. When this hap- Unlike conventional IETS, “biological IETS” does not
pens, electrons can follow an indirect path, first exciting involve scanning of the energy range, which would proba-
the molecular vibration and then tunneling to the second bly be unfeasible in a biological system. Instead, the range
metal at a lower energy. The new tunneling path causes an of vibrational energies is covered piecewise by a series of
increase in the conductance of the junction. receptors tuned to different energies. The energy range is
Metallic conductors are absent in biology, but elec- limited only by the emf (reducing power) of the electron
tron transfer is ubiquitous. Doing IETS with proteins source. An estimate of biological reducing power is 500
(Fig. 10) would involve addition and removal of elec- mV (1 e V
8086 cm 1) (Frausto da Silva and Williams,
trons at well-defined energy levels on either side of an 1993), which means that the entire vibrational range to
odorant-sized (300 daltons) binding site, which serves 4000 cm 1 could be sampled. To cover the vibrational
as the tunneling gap. On one side of the gap, a donor site spectrum, several receptor classes would be required, each
with occupied donor levels is present, while an acceptor tuned to a different segment of the vibrational spectrum.
site with empty acceptor levels is on the other side of the A small number might be sufficient, much as three
tunneling gap. If there is nothing between the electron pigments with broad, partially overlapping absorption
source (donor) and sink (acceptor), then for direct tun- spectra suffice for color vision. One essential feature of the
neling to occur there must be an (occupied) energy level biological spectrometer is its relatively poor resolution.
in the source that matches the energy of an (empty) state A biological system must work at ambient or body tem-
in the sink. perature, i.e., around 300 K. Donor and acceptor levels
If there is a molecule between the electron source and across the tunneling gap will therefore have a minimum
electron sink, and if that molecule vibrates, then indirect width of 2kT (艐400 cm 1). The range 0–4000 cm 1

Figure 10 Schematic of the proposed transduction mechanism: the receptor protein accepts electrons from a soluble electron donor
(NADPH). When the receptor binding site is empty (top), electrons are unable to tunnel across the binding site because no empty levels
are available at the appropriate energy. The disulfide bridge between the receptor and its associated G-protein remains in the oxidized
state. When an odorant (here represented as an elastic dipole) occupies the binding site (bottom), electrons can lose energy during tun-
neling by exciting its vibrational mode. This only happens if the energy of the vibrational mode equals the energy gap between the filled
and empty levels. Electrons then flow through the protein and reduce the disulfide bridge via a zinc ion, thus releasing the G-protein for
further transduction steps.
282 Turin and Yoshii

could thus be covered by 10 or so receptor types. A simi- What could make the SH infallibly distinctive as an
lar arrangement exists in the other spectral senses, vision odotope, as compared to the OH group? Partial charge,
and hearing, in which broadly tuned receptors classes bond length, bond angle, and atom size are somewhat dif-
cover segments of the complete spectrum. ferent between R-SH and R-OH, but it is hard to see how
these can be detected with absolute reliability by, say, an
amino acid side chain in the presence of thermal motion. A
V. ODOTOPES VS. VIBRATIONS: HOW THEY
more distinctive property of sulfur lies in the energy of its
FIT THE FACTS
lone pair orbitals, as witnessed by the specificity with
which it forms complexes with certain metals. If a metal is
In a field as vast and amorphous as that of SORs, observa-
involved, then it becomes hard to explain that (1) dimethyl
tions can be found to lend support to almost any theory. In
sulfide has no thiol character and (2) a thioether (–S–) link
what follows, we shall therefore try to stick to observations
can often replace a –C
C– with very little change in
that are potentially able to disprove one or the other of the
smell and no sulfuraceous character (Boelens and van
two contenders.
Gemert, 1993b) (Fig. 11).
The same problem applies to other smellable functional
A. Smelling Chemical Groups
groups and can be stated more generally: if functional
groups are odotopes, then they are so small as to only be
A fact that has, in our opinion, received too little attention
able to form one or two interactions, e.g., hydrogen bonds,
from olfaction researchers is the ability of humans to
with odotope receptors. Their small size will similarly
detect the presence of functional groups with great relia-
restrict the number of repulsive interactions. Therefore
bility (see Klopping, 1971, for review). The case of thiols
small molecules should bind with various degrees of affin-
(–SH) is familiar, but other chemical groups such as
ity to many odotope receptors, and small molecules should
nitriles (–CN), isonitriles (–NC), oximes (–NOH), nitro
have similar odors, particularly at high concentrations
groups (NO2), and aldehydes (C – O(H)) can be reliably
(Klopping, 1971). That is not the case: small molecules
identified once the odor character the functional group
like methylnitrile and methylnitrate smell distinctively dif-
character confers is known. When nitriles are used as
ferent at all concentrations. Indeed, still smaller ones—
chemically stable replacement for aldehydes, they impart a
e.g., ozone, sulfur hexafluoride, and carbon disulfide—
metallic character to any smell: cumin nitrile smells like
also have this property.
metallic cumin (cuminaldehyde), citronellyl nitrile smells
like metallic lemongrass (citronellal), and nonadienylni-
2. Functional Groups and Vibrational Theory
trile smells like metallic cucumber (nonadienal). Oximes
give a green-camphoraceous character, isonitriles a flat By contrast, the distinctive smell of functional groups is a
metallic character of great power and unpleasantness, nitro natural feature of a vibrational theory. Above 1800
groups a sweet-ethereal character, etc. Remarkably, even wavenumbers, IR absorption lines are diagnostic of the
bonds between atoms can be detected: the acetylenic C–C stretch frequencies of diatomic functional groups. The alde-
triple bond of -ynes imparts an isothiocyanate-like mus- hyde-nitrile replacement rule can be understood from the
tard-like smell to molecules that is clearly recognizable— closeness of their stretch vibration. Similarly, the similarity
for example, in acetylene and in methyloctynoate. in smell between acetylenic bonds and isothiocyanates can
be explained by their respective stretch frequencies.
1. Functional Groups as Odotopes
An odotope theory can explain these regularities only by
assuming that the functional group is an odotope. In the
older structure–odor literature, this used to be described as
electronic factors (as opposed to steric). The idea was that,
given that many functional groups were similar in size, the
recognition mechanism must somehow be sensitive to the
fine structure of the electron distribution (orbital energies,
charge density, etc.) of the functional group. This seem-
ingly reasonable notion runs into problems on closer
examination. Figure 11 Replacing a C
C bond with a sulfur atom does not
Consider, for instance, the SH group in methanethiol. change odor character, suggesting that “electronic” properties of
Alcohols never smell of sulfur, whereas thiols always do. sulfur are not sufficient for molecular recognition.
Structure–Odor Relationships 283

Figure 12 The dependence of the sulfuraceous character on molecular vibrations and atomic partial charges, as predicted by a
vibrational theory. Decaborane (left) smells sulfuraceous, and its terminal B-H bonds have a stretch frequency of ~2500 wavenumbers.
In triethylamine-borane (middle), the B-H stretch is shifted to 2300 wavenumbers and the sulfuraceous smell is no longer present. In
p-carborane (right) the near-neutral partial charges make the SH bond odorless. (See color insert.)

The clearest example so far is that of boranes. The ter- Objection 3, by contrast, can be answered rather sim-
minal B–H bond in boranes has a stretch frequency ply. If terminal BH groups activate the same odotope as
whose range overlaps with that of thiols. Turin (1996) SH, then all BH-containing compounds should have a
therefore predicted that boranes should smell sulfura- sulfuraceous character. This is not the case: as was
ceous, despite the complete absence of similarity, both pointed out to one of us (LT) by R. H. Biddulph (per-
structurally and chemically, between boron and sulfur. A sonal communication), triethylamine-borane does not
comparison between borane and thiol smells is best made smell sulfuraceous. Remarkably, the vibrational frequen-
using decaborane.* Decaborane smells strongly of boiled cy of the BH bond in triethylamine-borane is shifted
onion, a typical SH smell. Curiously, its smell was downward by 200 wavenumbers, i.e., out of the thiol
described as “chocolate-like” (chocolate contains some range. Another instance is that of the three isomers of
thiols) in papers reporting its synthesis, which may carborane, which smell camphoraceous, though o-carbo-
account for the fact that the similarity was not noticed rane has a faint onion-like (sulfuraceous) smell. The rea-
earlier. Other, less stable boranes share this sulfuraceous son for this is not yet clear, but their extraordinary chem-
smell character (Fig. 12). ical stability is consistent with a low polarity of the B-H
There are three possible nonvibrational interpretations bond, and this would tend to reduce the intensity of the
of this finding: (1) boranes do not in fact smell of sulfur, BH stretch vibration to the point where it may be no
(2) the similarity in smell between B–H and S–H, while longer detectable.
real, is pure coincidence, and (3) BH and SH activate the
same odotope receptor by some unknown mechanism, 3. Hindered Functional Groups
despite the difference in shape. In answer to objection 1,
we advise the interested reader not to take the authors’ Molecules could in principle be designed to settle the issue
word and to smell decaborane observing due precautions. of whether functional groups are perceived as odotopes or
Objection 2 is harder to answer, because the odds against by their vibrations. Suppose, for example, that a functional
such a thing happening, while large, are impossible to cal- group possessing a distinctive odor was present in a mole-
culate exactly. Predictions are rare in SOR theories, and cule, but was buried in such a way as to be inaccessible to
this is a conspicuously successful one. molecular recognition. Because tunneling electrons pene-
trate the molecule, the vibrational theory would predict
that it should still smell, whereas odotope theory would
not. The ideal molecule in this respect would include, say,
*This experiment requires caution: although stable at room tem- an SH group within its innards, completely shielded from
perature in air, decaborane is reported to be highly toxic and has touch. Such a molecule does not yet exist and may be
a high vapor pressure. It is therefore best to open the container in impossible to construct given the maximum size require-
a fume cupboard, close it again after a few minutes, and smell the ment for odorants.
very small amount of decaborane condensed on the outside of the Sterically hindered phenols provide a first approxima-
cap. One of us (LT) has been doing this periodically for some tion to this goal. The presence of an OH group on a
time with no apparent ill effects. substituted benzene ring gives the molecule a distinctive
284 Turin and Yoshii

Figure 13 Space-filling models of 2,4-(left) and 2,6-di-t-butyl phenols. These two molecules smell equally phenolic despite the
OH group’s being accessible in one and sterically hindered in the other.

“phenolic” odor, which the corresponding benzene does

not have. Once again, if one assumes that the OH group is
an odotope, then making it less accessible to molecular
recognition should silence its smell. This idea is easily
tested by comparing the smell of di-tert-butyl derivatives
of phenol, which are readily available commercially. The
results go against the odotope theory. 2,6-Di-tert-butyl
phenol, in which the OH group is strongly hindered, smells
as phenolic as, say, the 2,4 derivative in which it is more
accessible (Fig. 13).
It may be argued that the OH group is insufficiently Figure 14 Electron-density maps of ferrocene (left) and nicke-
buried in this molecule and remains accessible to some locene (right) with electrostatic potential mapped onto the sur-
face. Structure, electron density and potential surfaces calculated
molecular interaction. Designing molecules with buried
by semiempirical methods using Spartan with PM3 parameters.
functional groups—for example, the trimethylsilyl analogs Red is more negative. There are small differences in ring spacing
of phenols and thiophenols—could, in principle, settle this and charge distribution, whereas the odor of these two molecules
question. is radically different: ferrocene smells spicy-camphoraceous;
nickelocene smells oily-chemical. (See color insert.)
B. Isosteric Molecules

A strong test of vibrational vs. odotope theories would be et al., 1985, 1993; Wrobel and Wannagat, 1982a–d, 1983).
the odor comparison of molecules identical in atom com- Because of the high polarity and consequent instability of
position, shape, weight, electron distribution, and all other the Si-H group, only C atoms linked to four carbons can be
physical properties, differing only in vibrations. That is, of substituted in this fashion. The geometry of Si–C bonds is
course, an unattainable ideal, but one can come quite close, tetrahedral. Though very similar in overall geometry, sila
either by element substitution (Ni for Fe inside a metal- compounds will differ somewhat from the parent carbon
locene, Si for C) or by isotope substitution (D for H) in compound. The Si-C bond is 1.8 Å long, as compared to
normal odorants. 1.5 Å for a typical C–C bond, and the Si–C bond is more
polar. By contrast, the vibrations of the molecule will be
1. Metallocenes markedly altered by the Si and Ge substitution. For exam-
ple, the Si-C stretch vibration is around 650 wavenumbers
Ferrocene and nickelocene have very similar structures
instead of 1000 (Fig. 15).
and very different smells. Vibrationally, the main differ-
In all cases, there was some change in odor, though it
ence is in the internal movements of the metal ion between
was sometimes subtle and sometimes striking. For
the rings (Fig. 14).
example, sila substitution in linalool “light and refreshing,
floral-woody odor with a citrusy note” (Wrobel and
2. Sila Compounds
Wannagat, 1982a) to give sila-linalool makes it “more
Silicon (and in some cases Ge and Sn) can replace carbon hyacinth-like, sweeter.” Similarly, sila-terpineol smells
in odorants (Mundstedt and Wannagat, 1985; Wannagat more muguet-like and less lilac-like than the parent
Structure–Odor Relationships 285

In summary, the results on sila and germa compounds

are consistent with both theories. Odotope theory does not
adequately account for the very large difference in smell
between C and Si, and especially between Si and Ge com-
Figure 15 Three representative examples of molecules in pounds. Neither theory adequately explains why odor dif-
which Si replacement for C causes a marked change in odor. Left: ferences should be so marked in some cases and weak in
Sila linalool; center: sila terpineol; right: sila cyclocitral. others.

3. Isotope Substitution
compound, but sila-carvomenthene smells “similar” to the
parent carbon compound. Interestingly, though the largest Isotope substitution is in principle the best way to make
jump in size and other properties occurs between C and Si, perfectly isosteric compounds differing “only” in molecu-
Ge derivatives are again different in odor from both. Sila- lar vibrations. The “only” in the sentence above illustrates
cyclocitral smells “camphoraceous, sweet, earthy with a the fact that, as Wade has pointed out in his comprehensive
green tea note,” whereas the parent compound smells review of isotope effects in biology, there are in fact subtle
“minty, turpentine-like” (Mundstedt and Wannagat, 1985). differences in the physical and chemical properties of iso-
In her comprehensive review of SORs, Rossiter (1996) topes as compared to the parent compound (Wade, 1999).
summarized these results by saying: “Those examples Their hydrophobicity will be slightly different because of
where the odor of the sila analogue is similar to that of the the small difference in size and polarizability of the elec-
carbon counterpart are interesting anomalies for . . . vibra- tron cloud surrounding the heavier nuclei. In addition, the
tional theories.” These data could also be interpreted as range of conformations that the compound will explore
interesting anomalies for odotope theories. The reader during thermal motion will be different, because the
interested in exploring the differences in odor between sila altered masses respond differently to thermal excitations.
and parent compounds can readily obtain 1,1-dimethyl-1- Nevertheless, these effects are small: isotope separations
silacyclohexane (艐cyclopentamethylene dimethylsilane) on chromatography columns require long elution times,
and its parent compound 1,1-dimethyl cyclohexane from and the lowest energy conformation (i.e., molecular shape)
either Aldrich or Lancaster (United Kingdom). The differ- will in all cases be unaffected by isotope substitutions. By
ence in smell between the two compounds is striking. The contrast, effects on molecular vibrations can be large: sub-
odor profiles, assessed by a professional perfumer, are as stitution of D for H reduces the X–H stretch frequencies by
follows: 1,1-dimethylcyclohexane, camphoraceous, with a a factor of approximately 兹苶2—for CH, for example, from
faint sweet fruity, powdery background; 1,1-dimethyl-1- 3000 to 2200 wavenumbers.
silacyclohexane, intense, chemical-green note reminiscent Effects of isotope substitution (deuterium for hydrogen)
of cis-3-hexenol, with a faint camphoraceous background on animal olfaction have been known for a long time. Hara
(Fig. 16). (1977) showed that fish could reliably distinguish deuter-
ated glycine from the parent compound. Meloan and
collaborators have performed a remarkable series of stud-
ies (Meloan et al., 1988; Kuo, 1982; Scriven, 1984;
Havens, 1993; DeCou, 1993) in which they showed that
insects could distinguish between isotopes. For example,
cyclohexanone is a powerful cockroach repellent, whereas
deuterated cyclohexanone is inactive.
No human counterpart of these effects was appreciated
until it was reported that deuterated acetophenone could be
distinguished from the parent compound by smell (Turin,
1996). These experiments were performed on a gas
chromatograph using a smelling port to eliminate the pos-
sibility that impurities might be responsible for the smell
difference. The difference in smell to trained observers was
Figure 16 The calculated structures of two commercially subtle, but definite.
available compounds with similar shape and very different We have recently found a more striking isotope odor
odors. Left: 1,1-Dimethylcyclohexane; right: 1,1-dimethyl- difference in dimethyl sulfide. Arctander describes the
sila–cyclohexane. odor of dimethyl sulfide as “repulsive, sharp, green,
286 Turin and Yoshii

cabbage-like” at high concentrations. Dimethylsulfide-d6

clearly smells cleaner, more truffle-like without the gassy
cabbage-like note of the parent compound. This is a par-
ticularly easy experiment to replicate because (1) both
dimethyl sulfide and dimethylsulfide-d6 are safe to smell
(despite its unpromising descriptors, DMS is a perfumery
raw material!) and available at very high purity from
Figure 17 The enantiomers of carvone. (S)-() Carvone (left)
Aldrich, and (2) dimethylsulfide is a very strong odorant,
smells of caraway; (R)-() carvone (right) smells of spearmint.
so impurities will be unlikely to influence the overall odor.
We urge interested readers to do the experiment. The anti-
symmetric and symmetric C–S stretch vibrations are identical. By contrast, if the IR absorption of a regular
shifted from 710 and 654 wavenumbers, respectively,* to solid (crystal) is probed with polarized light, then the
670 and 608 wavenumbers. spectrum depends on the relative orientation of the mol-
Finally, one of us (LT) has obtained a sample of deuter- ecular dipoles in the crystal to the plane of light
ated decaborane. Unlike those of thiols, the terminal polarization.
hydrogens of boranes are not readily exchangeable. This Turin (1996) has argued that a “biological
allows one to test whether the stretch frequency of boranes spectroscope” resembles the latter case, and that the smell
is genuinely necessary to their sulfurous smell. Fully of carvone can be explained by a polarization effect. In a
deuterated decaborane ( 90% D) smells distinctively dif- tunneling mechanism for detection of molecular vibra-
ferent, more mustard-like, pungent, and less sulfuraceous tions, the odorant is bound in a fixed orientation in the
than its H counterpart. receptor and is probed by tunneling electrons which are
In summary, available evidence from isotope experi- polarized, i.e., deflected in specific directions by the
ments appears to be inconsistent with odotope theory and odorant. Strong dipoles (the C
O group in the case of
in broad agreement with vibrational theory. In order for the carvone) will be most likely to show polarization effects. It
odotope theory to apply, one would have to postulate addi- could be that in mint carvone, the C
O is not detected
tional factors to be involved. For example, a very high dif- because it is wrongly oriented. One would then expect that
ferential sensitivity of the odotope receptors to small “adding back” the carbonyl vibration by smelling simulta-
changes in odorant hydrophobicity might account for the neously a small carbonyl-bearing odorant (e.g., acetone,
results. Alternatively, one might suppose that the fact that butanone) with mint carvone would change the smell from
different low-lying vibrational states with energies around mint to caraway, which it does.* It remains to be seen
kT (艐240 wavenumbers) are excited in the deuterated whether similar experiments can be devised for other enan-
odorant will cause its average conformation to be slightly tiomer pairs.
different and thereby cause a difference in odor. There is at Interestingly, the smell of enantiomers poses a very
present no evidence for either mechanism. serious problem for odotope theory, although this fact
seems to have received no attention. The reason is analo-
C. Enantiomers gous to the problem with odorless molecules discussed
above. Suppose that an odorant is probed by several differ-
Most enantiomeric pairs of odorants smell identical, but ent odotope receptors. Each of those receptors is likely to
there are many examples of enantiomer pairs that smell be chiral to some extent, because it is near-impossible to
completely different (Boelens and van Gemert, 1993b). engineer a specific protein-binding site without chirality.
The best known of these outside fragrance chemistry are R A good indication of this comes from drug-receptor
and S carvone: R carvone smells of mint, S carvone of car- interactions, where drug enantiomers almost always have
away (Fig. 17). different actions (Hutt, 1998). Consider now the case of
Differences in smell between enantiomers have in the two enantiomers with identical smells. To account for this,
past been considered strong evidence against vibrational odotope theory needs to make one of two assumptions.
theories of olfaction, because solution IR spectra of Either the enantiomer binds equally well and with the
enantiomers probed with unpolarized light are of course same affinities to the n (chiral) odotope receptors or a

*The demonstration is easy to perform: mix three parts of

butanone with two parts of mint carvone and smell immediately,
*Computed ab initio using a 3-21G* basis set and corrected to because the butanone evaporates rapidly. The mint smell is gone,
0.9 of the calculated value. replaced by a good approximation to caraway.
Structure–Odor Relationships 287

completely different set of odotope receptors with opposite carboxylic acids, alcohols, bromo-carboxylic acids, and
chirality are wired in the same fashion to give the same dicarboxylic acids with carbon chains of varying lengths.
pattern of nerve excitation. Both are unlikely. A remarkable pattern emerges: the matrix of receptor
responses to odorants is sufficiently complex that even
D. The Evidence from Receptor Expression Studies: with this small number of receptors, an odorant in the list
Patterns of Receptor Activation can be identified from the pattern of receptors it is capable
of activating. They conclude that “different odorants are
In the past few years, following the early lead of Raming recognized by distinct combinations of receptors” and
et al. (1993), several receptor expression studies have been interpret this data according to an “odotope” model,
published. The advantage of receptor expression is that the though they do not use the term and do not specify which
response of a single receptor type to different odorants can odotopes may be involved.
be assessed directly. The results so far are still somewhat There is, however, another pattern in their data, namely
contradictory, and the field is evolving very rapidly. Zhao that in each series the number of receptor types that
et al. (1998) have expressed receptors in olfactory neurons respond increases as one lengthens the carbon chain. This
and found a broad response spectrum. Their particular immediately suggests that the spectrum of responses may
receptor subtype was optimally stimulated by heptanal, be related in part to the hydrophobicity of the odorant. The
less so by aldehydes of a shorter or longer chain length, latter can easily and accurately be calculated as logP,
consistent with an odotope-based model. Breer et al. where P is the (calculated) partition coefficient between
(1998), Kaluza and Breer (2000), and Touhara et al. (1999) octanol and water. When the number of receptor types acti-
also found that different receptors had relatively broad vated (a crude measure of the potency of the odorant) is
ligand specificity. These results are in agreement with the plotted against logP (Fig. 20), a new pattern is evident.
in vivo responses of olfactory receptor neurons (Firestein First, the number of receptors activated is roughly propor-
et al., 1993; Duchamp-Viret et al., 1999). tional to logP for each series of odorants. This suggests
By contrast, Krautwurst et al. (1998) reported greater that partition into a hydrophobic site, possibly the receptor
specificity in odorant receptor responses, and Wetzel et al. itself, governs “affinity” within a series. The different
(1999) reported that a receptor only responded to one series appear to have different efficacies, however: dicar-
odorant (helional) at very low concentrations but not to the
closely related molecule piperonal (Fig. 18). These
remarkable results differ from those reported by other
groups, including in vivo studies (see Firestein et al.,
1993). If confirmed and extended, they would suggest that
odorant-receptor interactions are far more specific than has
been hitherto supposed.
To date, however, the most comprehensive set of data
comes from the elegant work of Malnic et al. (1999), in
which 14 receptor types were expressed and their responses
to a set of 19 odorants compared. Figure 19 illustrates the
spectrum of response of the 14 different receptor types to
19 different odorants. The odorants are arranged in series:

Figure 18 Helional (left) and the related molecule piperonal.

A recent study (Wetzel et al. 1999) has suggested that helional Figure 19 Pattern of responses (black circles) of different
alone, out of 100 odorants, can activate an olfactory receptor, olfactory receptors (columns) to different odorants (rows). (From
with even closely related molecules being 1000 times less potent. Malnic et al., 1999.)
288 Turin and Yoshii

Figure 20 Response of expressed olfactory receptors to a variety of odorants. When the number of different receptor classes acti-
vated (ordinate) is plotted against the water-octanol partition coefficient (log P, abscissa), it becomes clear that a determining factor
in molecular selectivity is hydrophobicity. When the data are corrected for scattering intensity in a vibrational mechanism (right),
the correlation improves. Weak responses obtained at 100 M (small circles in original figure) were treated as 0.5. (From data in
Malnic et al., 1999.)

boxylics, while considerably less soluble in octanol than right-hand graph is obtained. The curves now follow
alkanols, are clearly more potent. roughly the same linear relationship. This suggests that the
Such a pattern would be expected from a vibrational data of Malnic et al. (1999) are equally consistent with a
mechanism. In this model the affinity of the odorant for mechanism invoking odotopes and by a vibrational mech-
the receptor is governed by logP because the receptor anism involving the physical quantities logP and S.
site is hydrophobic. The efficacy, by contrast, is gov-
erned by the electron-tunneling cross section of the mol-
VI. WHY ARE THERE SO MANY RECEPTORS?
ecule, i.e., its ability to scatter electrons. That in turn is
proportional to
The large number of receptor sequences found in humans
S
q2·x2 (347 at the last count, see Zozulya et al., 2001) is often
taken as evidence in favor of molecular recognition mech-
where q and x are, respectively, the calculated electrostatic anism based on shape. This is not necessarily so. First, if—
partial charges and atom displacements for each vibra- as seems likely—odorants bind to a specific binding site in
tional mode, and the summation is carried out over all the receptor, then the variability needed to accommodate
vibrational modes. In other words, the larger the charges different odorants of MW 300 daltons should be
on the component atoms and the bigger their displace- restricted to a dozen or so neighboring amino acids which
ments, the stronger the odorant will be. This makes sense are in direct contact with the odorant (Floriano et al. 2000).
when interpreting Figure 20 (left): the least potent odor- Second, a large number of receptors would be more con-
ants are the alkanols (partial charge largely on OH), then sistent with highly specific responses (one receptor, one
come the carboxylics (partial charges on the acid group), odorant). Most of the evidence points to broadly tuned
then the bromocarboxylics (additional charge from the receptors, which removes the need for a large number. By
C-Br bond), and finally the dicarboxylics (two sets of acid way of comparison, several thousand colors can be distin-
group charges).
The calculated values of S for octanol, octanoic acid, *Partialcharges (electrostatic fit) and atom displacements were
bromooctanoic acid, and octanedioic acid are 1.38, 3.50, calculated for the lowest homolog in the series using semiempir-
3.97, and 7.48, respectively.* When the logs of these val- ical methods with AM1 parameters (Mac Spartan, Wavefunction,
ues are used to correct the left-hand graph in Figure 20, the Inc.).
Structure–Odor Relationships 289

guished using the relative intensity of signals coming from i.e., that a molecule can be described legitimately as, say,
only three types of broadly tuned retinal cones. Third, green–weak or green–powerful. When traced back to its
some of the variability seems to be related to the develop- intellectual roots, this idea originates from pharmacology,
mental role receptors play in guiding olfactory receptor where a ligand, irrespective of its affinity for the receptor,
neurons to the correct place in the bulb (Wang et al., 1998). may have a low or high efficacy.
Olfactory receptor-like proteins have been found in nonol- This seems reasonable enough, but is actually quite
factory tissues and may serve a general developmental hard to reconcile with odotope theory. For a molecule to be
purpose (Dreyer, 1998). It is worth bearing in mind that in odorless, it would have to be simultaneously odorless to all
higher vertebrates, including humans, some of the the odotope receptors that it binds to. While this is possi-
olfactory receptor neurons renew themselves throughout ble in principle, it would be more likely to occur with small
life, which may require developmental clues. molecules bearing few odotopes, and therefore binding to
A large number of receptors is also expected from a few receptors, and gradually less likely as molecular size
vibrational mechanism. An idealized receptor would have increases. What is observed is precisely the opposite: with
an odorant-binding site that is as unspecific as possible, some exceptions (see below), odorless molecules appear
analogous to the cuvette of an ordinary spectrometer. At a only as molecular size nears its upper limit, i.e., when the
molecular level of course this cannot be achieved, because number of possible odotopes available for binding is at its
the cuvette needs to be molecule-sized and will thus maximum.
always incorporate some element of selectivity and chiral- An odotope theory modified to account for this might
ity. Conversely, if it were made large enough, say the size include a size selectivity filter in each receptor, such that
of a lipid droplet, in order to accommodate all odorants, it the molecule has to fulfill two criteria to be odorant: fit in
would be too large to allow electron tunneling to occur. A the filter and bind to the receptor. The difficulty with this
biological spectroscope that wishes to accommodate a ad hoc hypothesis is that it then requires perfect uniformity
broad range of odorants therefore needs a large variety of in the size of the selectivity filters. Were this not the case,
odorant-binding pockets. one would expect the most frequent outcome to be not an
Thus, both a shape-based and a vibrational theory may odorless musk but a different smell altogether if only a
require a large number of receptors, and the main differ- subset of odotope receptors are still able to smell it. This is
ence is in the way the receptors are wired. The arguments not the case—musks are either odorant or odorless to dif-
set out in Lancet et al. (1993) (see Sec. VI.A), which ferent subjects without change in smell character.
enable them to calculate the number of receptors required A vibrational theory has the opposite problem, namely,
to achieve a target affinity for a large set of ligands, apply that no molecule that has nonzero partial charges on its
equally well to odotope and vibrational theories. In a component atoms should be odorless, since all molecules
shape-based theory the receptors are wired by odotope, have a vibrational spectrum. This agrees (Turin, 1996)
whereas in a vibrational theory the receptors are wired by with the fact that all small molecules are odorous, except
spectral class. All receptors binding molecules of different for the ones with either very weak or zero charge (e.g.,
shapes but probing the same part of the vibrational spec- hydrogen or oxygen gas) and those for which all the vibra-
trum would be expected, say, to project to the same part of tions are below kT, since they will be indistinguishable
the olfactory bulb. There is some evidence that different from thermal noise. In a vibrational theory, the odorless
parts of the rat olfactory epithelium respond to the pres- character of a molecule can arise only from the fact that it
ence of different functional groups (Scott et al., 1996, does not bind to a receptor.
1997), but it is not clear whether the differences follow The problem of odor behavior near the size cutoff also
odotopes or vibrations. Not enough is known about the applies to a vibrational theory, but in a less extreme fashion
relationship between bulbar responses and either shape or than with odotopes. First of all, there need to be fewer recep-
vibration to decide the issue at the moment. tor types, ideally only enough to cover the vibrational spec-
trum piecewise. Turin (1994) has estimated their minimum
number to be 10. Second, the odorant-binding site could in
VII. THE PUZZLE OF ODORANT INTENSITY principle be completely nonspecific, since all that is
required is that the vibrations of the odorant be probed ade-
A. Odorless Molecules quately by the spectroscopic receptor. Indeed, all receptor
sites could be identical in structure and differ only in the
The question of odorant intensity (strong vs. weak) as dis- segment of the spectrum that they probe, which makes it
tinct from odor character (the sum of descriptors) raises easier to understand why they would all have the same size
issues of unexpected subtlety. Odotope theories implicitly cutoff. How this might be reconciled with the large number
assume that odorant intensity is part of the odor character, of odorant receptors found has been discussed above.
290 Turin and Yoshii

B. Weak and Strong Odorants

There are few reliable data sets on threshold detection

values in the literature (van Gemert, 2000). The largest
data set appears to have been laboriously collected over
many years by the fragrance firm Givaudan-Roure. Their Figure 21 Two extremes of odorant polarity: maltol and
odor-value chart, which would be of considerable interest undecanal.
to researchers, is (understandably) proprietary. Even when
differences in volatility are taken into account, odorants In a vibrational theory, efficacy has a different interpreta-
seem to differ in intensity by at least eight orders of mag- tion. All molecules vibrate, and spectral intensities are likely
nitude. As was discussed above, odotope and vibrational to differ by only a factor of 20 or so between the smallest
theories differ crucially in how they account for this fact. molecule with the weakest charges (e.g., methane) and a
Odotope theories regard odorless, weak, and strong large molecule with large partial charges (e.g., a nitro musk).
odorants, respectively, as inactive, weak, and strong To account for the several orders of magnitude in odorant
agonists. In other words, in an odotope theory, odorant intensity, a vibrational theory must therefore assume that the
intensity is in part related to the efficacy with which the strongest odorants simply bind most tightly to the receptors,
molecule activates the olfactory receptors, not necessarily i.e., that efficacy is proportional to affinity. There is,
just to its affinity for the receptors. The difference between however, a physicochemical difficulty in accounting for the
affinity and efficacy has been elegantly summed up by vast range in intensity of odorants simply by assuming that
Colquhoun (1998): the affinity of a drug for its receptor is the strong ones bind more tightly. To be sure, odorants dif-
“simply the microscopic equilibrium constant for binding fer greatly in polarity, as reflected in their water-octanol
to the inactive state.” Efficacy is “the set of all the other partition (logP). LogP varies by six orders of magnitude: for
equilibrium constants which describe the transduction example, maltol and undecanal—respectively, polar and
events that follow the initial binding reaction.” For hydrophobic strong odorants—have calculated logP values
example, a molecule that binds best to the active receptor of 0.07 and 3.20 (Kantola et al., 1991) (implemented in
conformation will favor the conformation change from Spartan, Wavefunction, Inc.) (Fig. 21).
inactive to active (agonist), whereas one that binds tightly However, many very strong odorants, such as diacetyl
to the inactive state will be an antagonist. and vanillin, are relatively polar. Clearly, some interaction

Figure 22 The strong (left) and “odorless” (right) isomers of lyral bound to a zinc ion via the carbonyl oxygen lone pair and the pi orbital
on the double bond. In the weak isomer, the geometry is unfavorable to zinc binding, as reflected by the angle formed by the two bonds to
zinc: 102° for the strong isomer, 75° for the weak one. Structures determined ab initio using Spartan software with 3-21G(*) parameters.
(See color insert.)
Structure–Odor Relationships 291

other than hydrophobic partition is required to account for Kollmann and his collaborators (Kuntz et al., 1999)
their intensity. have pointed out that there are three strategies to achieve
high-affinity binding: perfect steric fit, e.g., biotin-avidin,
C. Structural Correlates of Odor Intensity unlikely with a set of molecules as diverse as odorants; lots
of hydrogen bonds, difficult to achieve simultaneously
It has long been known empirically that certain structural with high-volatility hydrophobic molecules; and metal
features of molecules tend to make them stronger odorants. binding, which we believe is the strategy evolution has fol-
Moncrieff (1967) has listed many of these. The clearest lowed to achieve high-strength odorants.
correlates seem to be: (1) polar functional groups (OH,
C
O, CN, SH, –O–, etc.) increase intensity; (2) unsatu- D. Zinc Binding Is a Good Predictor of Odorant
ration generally increases intensity; (3) steric shielding of Intensity
a functional group decreases intensity; and (4) when two
hydrogen bond acceptors are present, the odorant is A large number of olfactory receptor sequences have now
stronger when they are close to each other (Ohloff bifunc- been published, and new ones appear every month. While
tional rule) (Ohloff, 1994). Taken together, these rules sug- many of these sequences may be pseudogenes
gest that odorants may be binding to some ligand that has (Mombaerts, 1999b; Rouquier et al., 2000; Zozulya et al.,
a high affinity for double bonds and lone pairs. Ohloff’s 2001), it is now possible to form an accurate impression of
bifunctional rule (see, e.g., Fig. 22) is particularly interest- their relationship to other known 7-TM receptors.
ing because it applies to a large number of structurally Recently, a thorough study by Skoufos (1999) has shown
unrelated odorants and it is consistent with both functional that one of the most conserved regions (Region 3 at the
groups binding to the same ligand. We propose that a zinc cytoplasmic end of TM helix 6) corresponds to the zinc-
ion coordinated to the receptor protein may function as a binding site proposed by Turin (1996). Interestingly, the
ligand for odorants, as was suggested independently by histidine that binds the zinc is completely conserved
Turin (1996) and more recently by Rakow and Suslick (Zozulya et al., 2001). This is what would be expected if
(2000). the zinc-binding site were essential to the operation of the

Figure 23 A sample of strong-weak isomer pairs: 1, methylanthranilate; 2, eudesmol, 3, neron, 4, p-menthane derivatives, 5, caparrapi
oxide, 6, iridanes. In every case, the strong isomer (left) is a bidentate ligand for zinc, whereas the weak isomer has unfavorable geome-
try for zinc- binding. (From Ohloff, 1994.)
292 Turin and Yoshii

receptor, and in particular if it were the odorant-binding

site itself. Remarkably, Sheikh et al. (1999) have shown
that if two histidine zinc-binding sites are engineered at the
cytoplasmic end of helices 3 and 6, then the presence
of zinc prevents receptor activaion in two different types
of 7-TNM receptors, suggesting that relative movement of
helices 3 and 6 is essential.
Furthermore, there is a good deal of circumstantial evi-
dence linking zinc with olfaction and gustation. Turin
(1996) pointed out that many strong odorants possessed
structural features capable of bidentate binding to a metal
ligand. This was recently confirmed (Suslick, 2000) by
colorimetric measurements of odorant binding to metallo-
porphyrins. This binding accounts in a straightforward Figure 24 Strong (left) and weak (right) isomers of muscone
(top) and mayol (bottom), illustrating that differences in odor
fashion for the fact that a hydrophobic zinc salt, zinc rici-
intensity cannot always be ascribed to the accessibility of one or
noleate, is a very effective deodorant. There is also a good more metal-coordinating groups.
deal of circumstantial evidence linking zinc with olfaction
and gustation. Zinc deficiency, either dietary (Alpers,
1994) or caused by treatment with histidine (Henkin et al., sity. If one accepts the notion that a intensity is not an odor
1975), thiocarbamides (Erikssen et al., 1975), or captopril character, but reflects the ability of the odorant to bind to the
(Zumkley et al., 1985), is unique in causing a complete and receptor, then it becomes clear that what these studies are
rapidly reversible anosmia. probing is the size and shape of the binding site.
We propose that this notion can be usefully extended by
including pi-bonding from double bonds, triple bonds, and VIII. SUMMARY AND CONCLUSIONS
cyclopropane rings (Bader, 1990) as possible metal lig-
ands. To test this, an unbiased data set is required. Ohloff’s In summary, it seems fair to say that if the ultimate goal of
(1990) review of strong-weak stereoisomer pairs provides a theory is predictive power, then both odotopes and vibra-
such a set, since it was selected without this theory in mind tion still fall short. Neither theory, when faced with a novel
and the odor threshold data are reliable. A good example is molecule, is yet able to predict reliably what its odor char-
provided by double-bond isomers of lyral (Fig. 22). When acter will be. Vibrational theory is conspicuously success-
the three-dimensional structure of these molecules is cal- ful at explaining the fact that we smell functional groups
culated and bidentate binding to zinc between the carbonyl even when sterically hindered and in accounting for differ-
oxygen and the double bond is included (Fig. 22), it ences in smell between isotopes. Odotope theory explains
becomes clear that only the strong isomer (mol) can bind neither.
to zinc in this fashion. By contrast, vibrational theory is intrinsically unable to
The same idea explains intensity differences between explain differences in the intensity of different odorants, or
many other isomer pairs described by Ohloff. Figure 23 which members of a set of related odorants will be
illustrates some of these instances. Examples 1, 4, and 6 odorless. We propose as a working hypothesis, to be tested
follow Ohloff’s bifunctional rule; the remainder involve a by further experiment, that odor character is determined by
double bond and a functional group. molecular vibrations and odor intensity is determined
There are, however, many exceptions to this rule, almost entirely by molecular shape.
namely those molecules for which enantiomers, isomers, We agree with Beets (1957) that, although the present
or diastereomers have different intensities without there theories may be incomplete, “we need not consider the
being more than one functional group capable of binding question of whether a relationship exists between structure
to zinc. Examples of this are muscone and Mayol. Clearly, and odor. . . . The only question is whether it is simple
this can have little to do with zinc binding and must be due enough to be detectable with our limited intellectual and
to steric interactions within the receptor site (Fig. 24). technical means.”
Many of the best correlations obtained between structure The fact that after several decades of experimental
and “odor” are actually done in such a way that what is being investigations, the basic mechanism by which odors are
tested is the effect of structure on odor intensity rather than detected remains open to question shows that there is much
on odor character. For example, QSAR studies of musks work to be done. At the present rate of discovery, is to be
(Yoshii, 1991, 1992; based on the data of Wood, 1968–1970) expected that the answer to these questions may come in
have been conspicuously successful in predicting odor inten- time for the next edition of this handbook.
Structure–Odor Relationships 293

REFERENCES differential odor responses of a mouse olfactory receptor. Proc.

Natl. Acad. Sci. USA 97:10712–10716.
Adkins, C. J., and Phillips, W. A. (1985). Inelastic electron tun- Frater, G., Bajgrowicz, J. A., and Kraft, P. (1998). Fragrance chem-
nelling spectroscopy J Phys C Solid State Phys 18:1313–1346. istry. Tetrahedron 54:7633–7703.
Alper, D. H. (1994). Zinc and deficiencies of taste and smell. Fraùsto da Silva, J. J. R., and Williams, R. J. P. (1993). The
JAMA 272(16):1233–1234. Biological Chemistry of the Elements. Clarendon Press, Oxford.
Amoore, J. E. (1971). Chemical Senses. Springer-Verlag, Berlin. Hansma, P. J. (Ed.). (19XX). Tunelling Spectroscopy. Plenum
Arctander, S. (1994). Perfume and Flavor Chemicals. Carol Press, New York.
Stream, IL, Allured Publishing. Hara, J. (1977). Olfactory discrimination between glycine and
Bader, R. F. W. (1990). Atoms in Molecules: A Quantum Theory. deuterated glycine in fish Experientia 33:618–619.
Oxford, Oxford University Press. Havens, B. R. (1993). The applications of deuterated sex pheromone
Balbes, L. M., Mascarella, S. W., and Boyd, D. B (1994). A per- mimics of the American cockroach, Periplaneta americana L, to
spective of modern methods in computer-aided drug design. the study of Wright’s vibrational theory of olfaction. Ph.D. the-
Rev. Comput. Chem. 5:337–379. sis, Kansas State University, Manhattan, Kansas.
Beets, M. G. J. (1957). Structure and odour in molecular structure Henkin, R. I., Patten, B. M., Re, P. K., and Bronzert, D. A. (1975).
and organoleptic quality. S.C.I Monograph No. 1. Society of A syndrome of acute zinc loss. Cerebellar disfunction, mental
Chemical Industry, London, pp. 54–90. changes, anorexia, and taste and smell dysfunction. Arch.
Bianchet, M. A., Bains, G., Pelosi, P., Pevsner, J., Snyder, S. H., Neurol. 32(11):745–751.
Monaco, H. L., and Amzel, L. M. (1996). The three-dimen- Hornstein, I., and Teranishi, R. (1967). Flavor chemistry. Chem.
sional structure of bovine odorant binding protein and its mech- Engin. News 93.
anism of odor recognition. Nat. Struct. Biol. 3(11):934–939. Hutt, A. J. (1998). Drug chirality and its pharmacological conse-
Boelens, H. (1974). Relationship between the chemical structure of quences. In Introduction to the Principles of Drug Design and
compounds and their olfactive properties. Cosmet. Perfum. Action, 3rd ed., Smith, H. J. (Ed.). Harwood Academic Publishers.
89:1–7. Jaklevic, R. C., and Lambe, J. (1966). Molcular vibration spectra
Boelens, M. H., and van Gemert, L. J. (1993a). Volatile character- by electron tunnelling. Phys. Rev. Lett. 17:1139–1140.
impact sulfur compounds and their sensory properties. Perfum. Kaluza, J. F., and Breer, H. (2000). Responsiveness of olfactory
Flavor. 18:29–37. neurons to distinct aliphatic aldehydes. J. Exp. Biol.
Boelens, M. H., and van Gemert, L. J. (1993b). Sensory properties 203:927–933.
of optical isomers. Perfum. Flavor. 18: Nov–Dec. Kantola, A., Villar, H. O., and Loew, G. H. (1991). Atom based
Colquhoun, D. (1998). Binding, gating, affinity and efficacy: the parametrization for a conformationally dependent hydrophobic
interpretation of structure-activity relationships for agonists and index. J. Comput. Chem. 12:681–669.
of the effects of mutating receptors. Br. J. Pharmacol. Klopping, H. L. (1971). Olfactory theories and the odors of small
125:923–947. molecules. J. Agric. Food. Chem. 19:999–1004.
Curry, S., Brick, P., and Franks, N. P. (1999). Fatty acid binding to Krautwurst, D., Yau, K. W., and Reed, R. R. (1998). Identification
human serum albumin: new insights from crystallographic of ligands for olfactory receptors by functional expression of a
studies. Biochim. Biophys. Acta 1441(2–3):131–140. receptor library. Cell 95(7):917–926.
Davies, R. H., and Timms, D. (19XX). Intermolecular forces and Kuntz, I. D., Chen, K., Sharp, K. A., and Kollmann, P. A. (1999).
molecular modelling. In Introduction to the Principles of Drug The maximal affinity of ligands. Proc. Natl. Acad. Sci. USA
Design and Action, 3rd ed., Smith, H. J. (Ed). Harwood 96:9997–10002.
Academic Publishers. Kuo, C. K. (1982). The effects of deuterating an attractant of the
Decou, D. F. (1993). The study of Wright’s theory of olfaction with American cockroach, Periplaneta americana L, as a test for the
trans-Z-hexen-l-al and related deuterated compounds. Ph.D. frequency theory of olfaction. M. S. thesis, Kansas State
thesis, Kansas State University, Manhattan, Kansas. University, Manhattan, Kansas.
Dreyer, W. J. (1998). The area code hypothesis revisited: Olfactory Lancet, D., Sadovsky, E., and Seidemann, E. (1993). Probability
receptors and other related transmembrane receptors may func- model for molecular recognition in biological repertoires:
tion as the last digits in a cell surface code for assmebling Significance to the olfactory system. Proc. Natl. Acad. Sci. USA
embryos. Proc. Natl. Acad. Sci. USA 95:9072–9077. 90:3715–3719.
Duchamp-Viret, P., Chaput, M. A., and Duchamp, A. (1999). Odor Lawton, M. P., Philpot, R. M. (1993). Molecular genetics of the
response properties of rat olfactory receptor neurons. Science flavin-dependent monooxygenases. Pharmacogenetics 3(1):
284(5423):2171–2174. 40–44.
Dyson, G. M. (19XX). The scientific basis of odour. Chem. Ind. Malnic, B., Hirono, J., Sato, T., and Buck, L. B. (1999).
57:647–651. Combinatorial receptor codes for odors. Cell 96:713–723.
Erikssen, J., Seegard, E., Naess, K. (1975). Side-effect of thiocar- McCartney, W. (1968). Olfaction and Odours. Springer-Verlag,
bamides. Lancet 1(7900):231–232. Berlin.
Firestein, S., Picco, C., and Menini, A. (1993). The relation Meloan, C. E., Wang, V.-S., Scriven, R., and Kuo, C. K. (1988).
between stimulus and response in olfactory receptor cells of the Testing Wright’s theory of olfaction with deuterated com-
tiger salamander. J. Physiol. 468:1–10. pounds. In Frontiers of Flavor, Proceedings of the 8th
Floriano, W. B., Vaidehi, N., Goddard, W. A. III, Singer, M. S., and International Flavor Conference, Elsevier, Amsterdam, pp.
Shepherd, G. M. (2000). Molecular mechanisms underlying 29–48.
294 Turin and Yoshii

Mombaerts, P. (1999a). Molecular biology of odorant receptors in Wade, D. (1999). Deuterium isotope effects on noncovalent interac-
vertebrates. Annu. Rev. Neurosci. 22:487–509. tions between molecules. Chem. Biol. Interact. 117(3): 191–217.
Mombaerts, P. (1999b). Odorant receptor genes in humans. Curr. Wang, F., Nemes, A., Mendelsohn, M., and Axel, R. (1998). Odorant
Opin. Genet. Dey. 9(3):315–320. receptors govern the formation of a precise topographic map.
Moncrieff, R. W. (1967). The Chemical Senses, 3rd ed. CRC Press, Cell 93(1):47–60.
Cleveland. Wang, V. S. (1981). The isolation and identification of volatile com-
Mori, K., and Shepherd, G. M. (1994). Emerging principles of mol- pounds in wheat which induce oviposition and aggregation
ecular signal processing by mitral/tufted cells in the olfactory response in the the flour beetle. M.S. thesis, Manhattan, Kansas,
bulb. Semin. Cell Biol. 5(1):65–74. Kansas State University.
Munstedt, R., and Wannagat, U. (1985). Sila perfumes and isosteric Wannagat, U., Munstedt, R., and Harder, U. (1985). Sila perfumes
compounds of perfumes 8.Alpha formylpropenilsilanes—model and isosteric compounds of perfumes 9 Sila beta ionones and
substances for sila beta cyclocitral. Liebigs Annal. Chem. related compounds. Liebigs Annal. Chem. 5:950–958.
5:944–949. Wannagat, U., Damrath, V., Huch, V., Veith, M., and Harder, U.
Ohloff, G. (1994). Scent and Fragrances: the Fascination of Odors
(1993). Sila-Riechstoffe und Riechstofffisostere XII.
and Their Chemical Perspectives. Springer-Verlag, Berlin.
Geruchsvergleiche homologer Organoelementverbindungen der
Rakow, N. A., Suslick, K. S. (2000). A colorimetric sensor array for
vierten Hauptgruppe (C, Si, Ge, Sn). J. Organometall. Chem.
odour visualization. Nature 406(6797): 710–713.
443:153–165.
Raming, K., Krieger, J., Strotmann, J., Boekhoff, I., Kubick, S.,
Wetzel, C. H., Oles, M., Wellerdieck, C., Kuczkowiak, M.,
Baumstark, C., and Breer, H. (1993). Cloning and expression of
odorant receptors. Nature 28;361(6410):353–356. Gisselmann, G., and Hatt, H. (1999). Specificity and sensitivity
Rossiter, K. J. (1996). Structure-odor relationships. Chem. Rev. of a human olfactory receptor functionally expressed in human
96:3201–3240. embryonic kidney 293 cells and Xenopus Laevis oocytes. J.
Rouquier, S., Blancher, A., and Giorgi, D. (2000). The olfactory Neurosci. 19(17):7426–7433.
receptor gene repertoire in primates and mouse: evidence for Whissell-Buechy, D., and Amoore, J. E.(1973). Odour-blindness to
reduction of the functional fraction in primates. Proc. Natl. Acad. musk: simple recessive inheritance. Nature 245(5421): 157–158.
Sci. USA 97:2870–2874. Wood, T. F. (1968–1970). Chemistry of the aromatic musks I–IX
Scott, J. W., Davis, L. M., Shannon, D., Kaplan, C. (1996). Relation (nine papers). The Givaudanian, I: Jan., 3–8 (1968); II: March,
of chemical structure to spatial distribution of sensory responses 3–6 (1968); III: May, 5–10 (1968); IV: June, 6–8 (1968); V:
in rat olfactory epithelium. J. Neurophysiol. 75(5):2036–2049. Sept., 5–9 (1968); VI: Dec., 5–7 (1968); VII: Jan., 5–9 (1969);
Scott, J. W., Shannon, D. E., Charpentier, J., Davis, L. M., and VIII: Sept., 6–10 (1969); DC: March–Apr., 3–7 (1970).
Kaplan, C. (1997). Spatially organized response zones in rat Wright, R. H. (1982). The Sense of Smell. Boca Raton, FL, CRC
olfactory epithelium. J. Neurophysiol. 77(4):1950–1962. Press.
Scriven, R. (1984). Studies of natural and synthetic repellents and Wrobel, D., and Wannagat, U. (1982a). Sila perfumes 2: sila
attractants for the American cockroach, Periplaneta americana, linalool. Liebigs Annal. Chem. 4:734–738.
I. Ph.D. thesis, Kansas State University, Manhattan, Kansas. Wrobel, D., and Wannagat, U. (1982b). Sila substituted perfumes. 4.
Sell, C. (1999). Chemoreception. In The Chemistry of Fragrances, Sila derivatives of some musk scents. J. Organometall. Chem.
Pybus, D. H, and Sell, C. (Eds.). Royal Society of Chemistry 225:203–210.
Paperbacks, pp. 216–226. Wrobel, D., and Wannagat, U. (1983). Sila perfumes. 5. Sila terpi-
Sheikh, S. P., Vilardarga, J. P., Bartanski, T. J., Lichtarge O., Iiri, T., neol and related compounds. Liebigs Annal. Chem. 2:211–219.
Meng, E. C., Nissenson, R. A., and Bourne, H. R. (1999). Similar Wrobel, D., Wannagat, U., and Harder, U. (1982). Some methoxysi-
structures and shared switch mechanisms of the 2-adrenoceptor lanes, disiloxanes and digermoxanes with effectivenesss of odor:
and the parathyroid hormone receptor. J. Biol. Chem. sila-substituted perfumes 3. Monatsh. Chemie 113:381–388.
274:17033–17041. Yoshii, F., Hirono, S., and Moriguchi, I. (1994). Relations between
Singer, M. S. (2000). Analysis of the molecular basis for octanal the odor of (R) ethyl citronellyl oxalate and its stable conforma-
interactions in the expressed rat 17 olfactory receptor. Chem. tions. Quant. Structure–Activity Relat. 13(2):44–147.
Senses 25(2):155–165.
Yoshii, F., Hirono, S., Liu, Q., and Moriguchi, I. (1991). Quantitative
Skoufos, E. (1999). Conserved Sequence motifs of olfactory recep-
structure-activity relationships between odoriferous and odorless
tor-like proteins may participate in upstream and downstream
musks. Chem. Senses 16:319–328.
signal transduction. Recept. Channels 6:401–413.
Yoshii, F., Hirono, S., Liu, Q., and Moriguchi, I. (1992). Three-
Touhara, K. Sengoku, S., Inaki, K., Tsuboi, A. Hirono, J., Sato, T.,
Sakano, H., and Haga, T. (1999). Functional identification and dimensional structure model for benzenoid musks expressed by
reconstitution of an odrant receptor in single olfactory neurons. computer graphics. Chem. Senses 17:573–582.
Proc. Natl. Acad. Sci. USA 96:4040–4045. Zhao, H., Ivic, L., Otaki, J. M., Hashimoto, M., Mikoshiba, K., and
Turin, L. (1996). A spectroscopic mechanism for primary olfactory Firestein, S. (1998). Functional expression of a mammalian
reception. Chem. Senses 21(6):773–791. odorant receptor. Science 279(5348):237–122.
Turin, L. (2002). A method for calculating odor character from mol- Zozulya, S., Echeverri, F., and Nguyen, T. (2001). The human ofac-
ecular structure. J. Theor. Biol. (in press). tory receptor repertoire Genome Biol. 2(6): 0018.1–0018.12.
Van Gemert, L. J. (2000). Compilations of odour threshold values in Zumkley, H., Bertram, H. P., Vetter, H., Zidek, W., and Losse, H.
air and water. Available from Bacis (http://www.xs4all.nl/ (1985). Zinc metabolism during captopril treatment. Horm.
~bacis). Metab. Res. 17:256–258.
14

Olfactory System Cybernetics: Artificial Noses

Krishna C. Persaud
University of Manchester Institute of Science and Technology, Manchester, United Kingdom

I. INTRODUCTION the study of messages as a means of controlling machinery

and society, the development of computing machines and
The remarkable capabilities of biological chemosensory other such automata, certain reflections upon psychology and
systems in detecting, recognizing, and discriminating com- the nervous system, and a tentative new theory of scientific
plex mixtures of chemicals, together with rapid advances in method.” The developments in “electronic nose” technology
understanding how these systems operate, has stimulated now herald the birth of “olfactory system cybernetics.” The
the imagination and interest of many researchers and com- motives arise from (1) the perceived limitations of traditional
mercial organizations. Such stimulation has led to the devel- analytical chemistry and instrumentation in classification of
opment of electronic analogues. An artificial sensing system gas mixtures or odors; (2) applications where the gestalt of a
that emulates the human sense of smell is of upmost need in defined mixture of chemical species may be important in per-
a number of fields, including the food, flavor, beverage, and ception, applications where it is important to separate subjec-
cosmetic industries, as well as environmental protection tive and objective assessments; and (3) the commercial drive
industries and governmental agencies. Although the human to achieve devices capable of operating rapid, on-line process
sense of smell has been used for centuries in applied indus- measurement and control in areas of foods, beverages, chem-
trial settings in quality control and other processes, it is ical industries, and waste management. Another motive dri-
liable to variation from illness and other factors, including ving research is the attempt to create biomimetic devices that
subject age, gender, and training. Moreover, there are lin- emulate aspects of biological sensory systems. The deve-
guistic limitations in communicating odor experiences lopments described in this chapter were based on the coales-
among individuals. While, as noted in other chapters of this cence of several evolving scientific disciplines where key
volume, the human sense of smell is exquisitively sensitive developments and concepts were evolving and converging in
and can provide reliable estimates of odorant intensity and the last four decades, leading to a veritable explosion in the
quality, there can be considerable variability of response last decade. Today, electronic noses abound in various config-
among untrained individuals to different odors and odor urations and are used in specific applications, but “artificial
concentrations. An instrument that could perform simple noses” do not yet exist.
odor discrimination and provide an accurate indication of
odor intensity with less variation than that observed in
human responses would be very useful in modern industry. II. KEY CONCEPTS
Cybernetics was defined by Norbert Wiener (1948) as “the
theory of control in engineering, whether human or animal or Some of the key ideas and developments inspiring the mul-
mechanical”, including “not only the study of language but tidisciplinary evolution of olfactory cybernetics are

295
296 Persaud

worthwhile noting. The survival of animal life in complex, Thom (e.g., Thom, 1970, 1972), and (3) concepts of neural
changing environments requires the use of sophisticated mechanisms based on a model of the brain as a spatio-tem-
sensory systems to detect, classify, and interpret patterns poral lattice of nonlinear processing elements (e.g.,
of input stimulation. This involves the development of a Freeman, 1974, 1991).
coding mechanism by which a certain pattern of stimula- Data from a number of sources, including electrophysiol-
tion may be described. Such a code may be defined as a set ogy, chemical structure-activity studies, psychophysics, and
of symbols that can be used to represent patterns of organ- molecular genetics, support the view that a finite number of
isation and the set of rules that governs the selection and olfactory receptor classes exists and that a given class of
use of these symbols. Sensory coding mechanisms in bio- receptor cells responds to a range of molecular entities, with
logical systems would appear to project some representa- some overlap among classes (e.g., Amoore, 1962; Beets,
tion of a pattern at a high level of the nervous system, the 1978; Boelens, 1974; Buck, 1997a, Chess et al., 1992;
resulting neural activity being related to the previous Mombaerts et al., 1996) Molecular parameters important for
experience with regard to this pattern or associated pat- determining an olfactory response are varied but likely
terns (Uttal, 1973). The rapid evolution of research into include the adsorption and desorption energies of the mole-
artificial intelligence, leading to devices that classify and cule from air to a receptor interface, partition coefficients,
interpret patterns, has resulted in fundamental mathemati- electron donor/acceptor interactions (depending on the polar-
cal understanding of how patterns may be encoded and izability of the molecule), and molecular size and shape.
classified by biological systems. The seminal papers set- The rapid advances in neurobiology, electrophysiology,
ting out the key requirements in the design of parallel fea- and biochemistry that were occurring concurrently in the
ture detection systems with learning capabilities (e.g., 1960s, leading to an understanding of the convergent nature
Minsky and Papert, 1969; Rosenblatt, 1962; Selfridge, of the olfactory system (Shepherd, 1998; see Chapters 4–10).
1959), together with recent advances in microelectronics, This allowed models of the sensory systems for the instinc-
have led to the employment of some of these features in tive recognition of patterns, including olfaction and taste, to
“intelligent toys” for children. Some of the fundamental be constructed, as exemplified by Deutsch (1967) based on
principles of pattern classification that seem to be common an idealized organization of the cortex of the brain. It is use-
to biological and artificial systems include “template ful to recapitulate such an approach. A compound is taken to
matching,” whereby the pattern to be classified is contain primary elements or stimuli to which the system can
compared with a set of templates, one for each class, the respond, e.g., compound X contains the elements A, B, and
closest match determining the classification, and “feature C. A complex matrix of neurons make the system concentra-
detection systems” in which a number of measurements tion-independent. These “amplitude-matching neurons” con-
are taken on the input pattern and the resulting data are sist of a set of neurons, each of which is set to trigger at a
combined to reach a decision. These systems may involve different concentration. The output of these neurons converge
either a sequential approach, in which information from to the common gate of an output neuron, so that the output
the evaluation of some features is used to decide which from this neuron is concentration independent. Another
features to evaluate next, or a parallel approach, where matrix of neurons is proposed that finds the ratios of the com-
information about all features are evaluated at the same ponents A/B and B/C. These produce an output to the sens-
time with no weight being placed on any particular feature. ory pattern output neurons, where the output of the stimulus
A characteristic of the activity of the nervous system is ratio neurons are recognised as unique to compound X. The
that it is persistent throughout the life of the organism and system is inherently flexible and has room for excitatory and
so is in some sense stable. However, it also appears to have inhibitory control at each stage, making it an attractive propo-
a stochastic nature that is essential—a nervous system that sition for olfactory cybernetic development. This understand-
is stereotyped would not exhibit adaptive behavior in a ing that receptors with partial specificity could in principle be
changing environment. Animals have goal-driven behavior used to achieve reliable discrimination was later exploited by
that works reliably in an unpredictable world. The afore- Persaud and Dodd (1982) to achieve a functional device
mentioned methods of pattern classification can be elabo- using just three broad specificity sensors.
rated so that stochastic behavior is achieved, and many From the discussion above, it may be recognized that
models that may be applicable to the sensory systems of four main problem areas could be early identified in the
animals have been developed. Such models date from sev- design of a ‘model’ olfactory system:
eral sources, including (1) the early work of Turing (1952),
who suggested that an initially chaotic or inhomogeneous 1. The selection of an effective set of parameters for
state is moved by random fluctuations towards a more sta- the description of the patterns in question. Where
ble state, (2) concepts of catastrophe theory expressed by an odor is concerned, the problems to be addressed
Olfactory System Cybernetics 297

are which characteristics of the molecule can be

measured, and how to encode the resulting pattern.
2. The selection of the decision procedure—how is
the receptor output to be categorized?
3. The selection of a procedure for processing the para-
meters chosen to represent the pattern so as to opti-
mize the parameter values. This increases the
resolution of the system so that certain representative
pattern classes may be differentiated from others.
4. The selection of “hardware” required for the simula-
tion of the transduction, coding and pattern recogni-
tion activities of the olfactory system.
Figure 1 Elements of an electronic nose system. Odor is sam-
pled and presented to individual elements of a sensor array simul-
III. ARRAY-BASED ODOR SENSORS taneously. Signals transduced from the sensors are converted to
digital format, normalized, and presented to a pattern-recognition
The discriminatory power of a small sensor array lies in engine (PARC) that outputs some parameter that is a determining
the utilization of cross-sensitivities between sensor ele- characteristic of the incoming odor. This may be discrimination
ments. The responses of the individual sensors, each pos- from other known odor classes or a mapping to some sensory
sessing a slightly different response towards the sample quality determined by a human panel.
odors, when combined by suitable mathematical meth-
ods, can provide enough information to discriminate
between sample odors. These systems have been given processing systems of biological systems. Tanyolaç and
the terminology “electronic nose” and consist of an array Eaton (1950) explored the surface tension changes of a liq-
of chemical sensors possessing broad specificity, coupled uid when volatile molecules were adsorbed, while the
to electronics and software that allow feature extraction advent of the transistor stimulated interest in utilizing ger-
of salient data for further analysis, together with pattern manium as a chemically sensitive transducer material.
recognition giving identification of sample odor. Among the pioneers who attempted to emulate functional
Software techniques and material science are important properties of the olfactory system electrically were
aspects of the development of the system. Advancement Dravnieks and Trotter (1965), who measured contact sur-
in software signal processing techniques, coupled with face potentials of various materials in the presence of
pattern recognition, enable optimum usage of sensor adsorbed volatiles, Wilkens and Hartmann (1964), who
responses. The specificity and sensitivity of existing measured redox potentials when various chemicals were
chemical sensors are constantly being developed, as well adsorbed onto electrodes, and Buck et al. (1965), who inves-
as new materials. tigated conductivity changes in solids due to interactions
Typically an electronic nose consists of three elements: between adsorbed odorant molecules and charge carriers in
a sensor array which is exposed to the volatiles, conversion the solid. However, it was not until the commercial avail-
of the sensor signals to a readable format, and software ability of gas sensors based on metal oxides in the 1970s
analysis of the data to produce characteristic outputs related (Figaro Inc., Japan) that it become practical to test some of
to the odour encountered (Fig. 1). In order to discrimi- the conceptual ideas, leading, for example, to the work of
nate between samples, the output from the sensor array Persaud and Dodd (1982). The general architecture of this
may be interpreted via a variety of methods such as pattern early device, although based on an array of only three sen-
recognition algorithms, principal component analysis, dis- sors and the somewhat limited knowledge of the time of
criminant function analysis, cluster analysis, and artificial how olfactory processing occurs, forms the basis for most of
neural networks. the instrumentation used today. Such devices have become
known as electronic noses (Gardner and Bartlett, 1994) and
A. Sensor Technology consist of “an instrument, which comprises of an array of
electronic-chemical sensors with partial specificity, and an
Early research in the field was hampered by the lack of suit- appropriate pattern recognition system, capable of recognis-
able sensor materials capable of emulating the functional ing simple or complex odours” such as shown in Figure 1.
characteristics of the olfactory system, as well as by the lim- A large number of sensor technologies are now avail-
ited understanding of the transduction and information able that are applicable to construction of sensor arrays for
298 Persaud

Figure 2 Common sensor technologies used in electronic noses: quartz crystal microbalance, metal oxide sensor, conducting polymer.

electronic nose applications. Figure 2 illustrates three com- doping agent. The physical and chemical mechanisms by
mon sensor types: quartz crystal microbalance, metal which gases can transduce signals are relatively well
oxide, and conducting polymer. The sensors must meet key understood (Kohl, 1989, 1991, 1996). Sensors have been
design parameters for the system. These include sensitivity, developed for the detection (down to the ppm level) of a
speed of operation, cost, size, manufacturability, the ability range of target molecules including H2, CO, NH3, H2S,
to operate in diverse environments, and immunity to cont- NOx, SOx, ethanol, and hydrocarbons.
amination or deactivation (i.e., “poisoning”). The sensors As an alternative to the commercially available metal
must adsorb large numbers of molecules of a particular oxide sensors, several research groups have fabricated thin
species to produce a measurable effect on the sensor that film SnO2 arrays using planar microelectronic technology.
can be transduced into a signal. Potentially these could have a number of advantages
including a reduction in size and lower power consumption.
1. Metal Oxide Sensors An elegant variation on this device technology is a sen-
sor array with a gradient of temperature across the surface
Many researchers have chosen commercially available
as well as a membrane sputtered on the surface (Menzel
metal oxide sensors such as Taguchi Gas Sensors (TGS)
and Goschnick, 2000). These arrays are now commercially
(Figaro Inc., Japan) or Capteur Sensors (Capteur, UK) as
manufactured and incorporated into electronic nose devices
the core sensing element in their investigation of array-
(Ehrmann, 1998).
based odor detectors. These devices consist of an electric-
ally heated ceramic pellet onto which a thin porous film of
2. Quartz Resonator and Surface Acoustic Wave
SnO2 doped with various precious metals has been
Devices
deposited. The doped SnO2 behaves as an n-type semicon-
ductor and the chemisorption of oxygen at the surface Quartz resonator gas sensors consist of a piezoelectric
results in the removal of electrons from the conduction quartz crystal oscillator coated with a sensing membrane.
band. Gases interact with the surface adsorbed oxygen and Typically a quartz disk is sandwiched between two elec-
thereby affect the conductivity of the SnO2 film. The trodes. The adsorption of volatile molecules onto the
devices are run at elevated temperatures (typically membrane results in a decrease in the resonant frequency
300–400°C) to achieve rapid response/recovery times and due to the increased mass. This frequency shift can be
to avoid interference from water. This results in relatively used as the sensor output, and the device response can be
high power consumption. The response characteristics can varied by using different membrane materials (Nanto,
be tailored by varying the operating temperature and the 1997).
Olfactory System Cybernetics 299

Surface acoustic wave (SAW) and bulk acoustic wave advantages include high sensitivity (ppm), high selectiv-
(BAW) devices consist of interdigitated electrodes fabri- ity, and ease of integration with other electronics. A
cated onto a piezoelectric substrate (e.g., quartz) onto potential disadvantage is that the odorant molecules must
which a thin film coating of a selective material is deposited. penetrate the transistor gate, and the devices are not yet
An applied radio frequency voltage produces a Rayleigh available commercially. Variations incorporating con-
surface acoustic wave (i.e., a surface oscillation). ducting polymers as chemical sensing surfaces have been
Adsorption of odors onto the coating increases its mass devised (Hatfield et al., 2000).
and perturbs the wave, leading to a shift in frequency. To
compensate for pressure and temperature effects the 6. Optical Sensors
sample sensor is usually connected to a reference SAW
device and the frequency difference is detected. As A fiber optic sensor for gas sensing is a conventional optical
with quartz resonator devices the coating material deter- fiber typically coated with a coating, which interacts with
mines the selectivity. SAW devices, however, can be oper- the odorant molecules. The coating may be a fluorescent
ated at higher frequencies, which reportedly results in dye. An optical pulse is applied to the sensor and is adsorbed
improved sensitivity (Rapp and Reibel, 1996; Yang, et al. by the coating. The interaction of the odorant molecules and
2000). The advantages of SAWs and BAWs include high fluorescent dyes produces a frequency shift in the returned
selectivity, high sensitivity, stability over wide tempera- fluorescent signal. The returned signal is then analyzed to
ture ranges, low response to humidity, and good repro- determine the properties of the odorant molecules.
ducibility. The disadvantage is the complexity in the Porphyrins and other materials may also be used as the
interface electronics. active adsorbent material in optical sensing (Di Natale et al.,
2000). More recently, microscopic polymer beads impreg-
3. Electrochemical Sensors nated with a fluorescent dye have been incorporated into the
ends of optical fibers, and these have proven to be extremely
Electrochemical sensors are widely used for detection of sensitive to vapors, and “optical electronic noses” have been
specific chemical analytes. Typically a redox reaction is developed. Imaging optical fibers in conjunction with two-
induced to occur at the anode or cathode of an electro- dimensional detectors such as CCD cameras have been used
chemical cell when a specific gas is dissolved in the to fabricate array sensors. These sensors contain spatially
electrolyte. Commercial sensors are made by a variety of separated photopolymers containing analyte-sensitive fluo-
companies such as City Technology Ltd, UK. A variety rescent indicators on an imaging fiber tip. Spatial resolution
of amperometric or coulometric devices are applicable to of the indicators is maintained through the imaging fiber
sensor arrays for electronic noses (Stetter, 1995). array and projected onto a CCD detector (Dickinson et al.,
1996, 1998, 1999; Walt et al., 1995).
4. MOSFET Sensors
MOSFET’s (metal oxide semiconductor field effect tran- 7. Mass Spectrometry–Based Devices
sistors) have been used as gas detectors, the vapor “Virtual” chemical sensors based on a mass spectrometric
producing a shift in the capacitance-voltage characteristic. approach have been developed for multicomponent ana-
MOSFET sensor behavior can be modified with coatings lysis of organic vapors. The sensing principle is based on
of zeolite of various pore sizes or by careful attention to a the injection of a complex sample headspace into a mass
gas-sensing layer in close proximity to the gate of the spectrometer, creating a mass spectrometric pattern of the
MOSFET, resulting in a change of capacitance on expos- unresolved gaseous mixtures. After selecting particular
ure to the gas. Using arrays, patterns can be generated for fragment ions, the resulting reduced mass spectrum of the
a range of solvent volatiles and chemicals such as ammo- sample is treated with pattern recognition (software) typ-
nia and hydrogen (Gardner, 1998; Stetter et al., 2000). ically based on principal components analysis or neural
Some “Electronic Nose” companies now commercially networks (Dittmann and Nitz, 2000; Dittmann et al., 1998,
use such sensors. 1999, 2000). This methodology has been commercialized
by a number of companies such as Agilent Inc., USA HKR
5. ChemFETS Sensor Systeme (Germany), and Alpha MOS (France).
A chemical field effect transistor (ChemFET) is a tran-
8. Carbon Black/Polymer Composite Materials
sistor with the gate electrode coated with a selective
coating. This coating adsorbs odorant molecules, which Cyrano Sciences Inc. USA have commercialized portable
changes the conductivity across the transistor’s gate. The instruments based on 32 carbon black/polymer composite
300 Persaud

materials. The carbon black forms the conducting phase of 1. The sensors show rapid adsorption and desorption
the sensor and is dispersed into an insulating organic poly- kinetics at room temperature.
mer. When the polymer comes into contact with an organ- 2. The sensor elements feature low power consump-
ic vapor it swells, causing a change in the electrical tion (in the order of microwatts) as no heater ele-
resistance of the sensor. An electrical potential is applied ment is required.
across each sensor so that the resistances may be recorded. 3. The structure of the polymer can be closely corre-
The use of different polymers, such as those used in gas lated to specificity towards particular classes of
chromatography columns, gives the sensors a range of sen- chemical compounds.
sitivities to a variety of chemicals (Lewis and Freund, 4. The sensors are resilient to poisoning by com-
1996; Lewis and Severin, 1997). pounds that would normally inactivate some inor-
ganic semiconductor type sensors.
9. Organic Materials
10. Electronic Nose Hybrid Technology
Much research has been carried out on sensors based on
metal-substituted phthalocyanines, each acting as a simple Neither the manufacturer nor the user is restricted to using
chemoresistor. Their applications have been limited to only one type of technology, and no one technology is yet
detection and measurement of gases such as NO2 and H2S generally applicable to generic odour sensing. Hence the
(Cranny and Atkinson, 1992), and they have suffered from use of modular sensor systems, or integrated sensor sys-
poor individual sensor response and reproducibility. tems containing different sensor technologies, is a growing
On the other hand, the unique electrical properties of trend (Kohl, 1997; Stetter et al., 2000).
organic conducting polymers, derived from aromatic and
heteroaromatic materials, have led to a large amount of B. Sensor Output and Data Processing
research and the application of these materials in different
areas. Since 1979, when Diaz et al. (1979) first prepared Figure 3a shows a commercial instrument, the Aromascan
polypyrrole as a freestanding film, thousands of publica- A32S (Osmetech plc), based on a conducting polymer
tions have appeared and many researchers have studied sensor array (Fig. 3b). This was launched in 1994, and sev-
conducting polymer gas sensors based on resistance eral hundred units were sold.
changes in thin film structures. Configurations have been No matter which sensor technology is utilized, the raw
used to measure the shifts in the work function caused by data response of each sensor in the array to a volatile sam-
the adsorption of a range of organic volatiles. The response ple such as shown in Figure 4a is normalized as shown in
of the polypyrrole film (i.e., the magnitude and the sign of Figure 4b, and the resulting pattern is used as descriptor for
the work function shift) is determined by the electrochem- discrimination between different samples. Sensing systems
ical deposition conditions and, in particular, the electro- have been developed where individual elements in an array
lyte/solvent system used. Measurements have also been show broad and overlapping selectivities to chemical
made of the change in the optical absorption spectra result- species, each sensor element responding more selectively to
ing from exposure to organic vapors. These data, together certain groups of chemicals. This approach has the
with the work function shifts, suggest a small but advantage that the array can respond to many thousands of
reversible charge transfer (either donor or acceptor) when chemical species due to the broad selectivity of the adsor-
a gas is adsorbed at the polymer surface. bent surfaces. On the other hand, extremely selective infor-
Gas sensors using polypyrrole films deposited as an over- mation for discrimination between adsorbed chemical
layer onto quartz resonator devices have been investigated by species or mixtures can be obtained by analysis of the
Slater and coworkers (Slater and Watt, 1991). In addition to cross-sensitivities between sensor elements. The relative
monitoring the mass loading affect of adsorbed volatiles, a responses between sensor elements produce patterns that
simultaneous measurement of conductivity changes was serve as unique fingerprints that may be used as odor
made on a separate device. A range of volatiles has been stud- descriptors. Algorithms to eliminate noise are needed in
ied including NH3, methanol, cyclohexane, acetone, and H2S. some systems, since there are often uncontrolled factors
Since 1985, we have concentrated on development of present in measurement data due to effects such as day-to-
conducting polymers as odor-sensing devices, and many day variations in ambient temperature and instrument sig-
materials have been synthesized and characterized for odor nal drift, which can introduce systematic error into the data.
transduction (Persaud and Pelosi, 1985; Persaud et al., The aim of many pattern recognition techniques is to
1996b). The reasons for choosing conducting polymers as identify similarities and regularities present in the data. One
odor sensor elements are as follows: method is cluster analysis, which attempts to find natural
Olfactory System Cybernetics 301

represent the characteristics of the data set. Various meth-

ods are available to accomplish this, either by considering
only a subset of the original variables or by creating more
efficient representative set of new variables. The creation
of new variables can be approached in a number of ways;
two of these are projection and mapping.
Projection is more common and involves using a
weighted linear combination of the original variables to
derive a new, more compact data set that contains nearly
the same informational content as the original variables.
Clearly this involves a certain amount of loss of data, and
this is not ideal since key features may be discarded. Thus,
techniques have been developed to minimize informa-
tion loss. A commonly used projection technique is princi-
pal components analysis (PCA) (Hotelling, 1933). The
purpose of the PCA transformation is to rotate the old
coordinate system in a direction so that the new system
will have most of the relevant information aligned along a
(a) few new axes. The majority of the new axes will carry a
very small proportion of the total information and could be
disregarded without too much loss. It attempts to
maximize the variance information present in data in the
minimum number of mutually orthogonal dimensions.
Graphically, PCA distorts the axes to conform to axes that
contain a maximum of variance information. One way to
describe this would be to consider looking at the data from
a different direction in space. Algebraically, this can be
performed as a simple linear transformation in any number
of dimensions (Fig. 5).
Mapping is a similar technique to projection, but the data
transformations are nonlinear and attempt to preserve key
properties of the data (such as the distances between points),
while reducing the number of dimensions. Other chemo-
metric techniques applied include partial least squares
(PLS), discriminant analysis (DA), and discriminant factor-
ial analysis (DFA). DFA is a multivariate technique that
determines a set of variables that best discriminates one
(b)
group of objects from another. For statistical pattern recog-
Figure 3 (a) Commercial electronic nose system—Aromascan
nition algorithms to be successful, there must be some crite-
A32S. (Courtesy Osmetech plc). (b) Conducting polymer sensor ria for them to base the allocation of different classes, and
array (32 sensors) used in the system. this is usually based on cluster analysis. A given cluster of
points representing a class has a center of gravity and a stan-
classifications in data. Computers are usually used as an aid dard deviation distance that specifies the location and spread
in the cluster analysis of data of more than three dimensions of the cluster. These values are obtained during the training
(since it is difficult for humans to visualize such vector phase of the algorithm, with replicate examples that are
spaces).The centers of clusters are represented by a set of known to be members of a given class.
coordinates, and these are called codebook vectors. Artificial neural networks (ANNs), which have been
A typical chemometric goal in electronic nose applica- used to analyze complex data and to recognize patterns in
tions is to find variables to separate known groups and, in many other disciplines (Rumelhart et al., 1986), have
particular, to be able to classify a gas or mixture of gases shown promising results in recognition of volatile com-
according to various properties. A primary aim of chemo- pounds and odors in electronic nose applications. When an
metrics is to reduce the number of dimensions used to ANN is combined with a sensor array, the number of
302 Persaud

Figure 4 (a) Raw data response from a sensor array made from conducting polymers. Each sensor responds to the same concentration
of analyte with different sensitivity. By normalizing the response across the entire array, a pattern (b) is obtained that is a descriptor of
the analyte presented to the sensor array.

detectable chemicals is generally greater than the number the hardware counterpart. In both cases the network con-
of unique sensor types. The advantages of ANNs over sists of a series of processing nodes arranged into different
other methods of pattern recognition are numerous. Their layers and, as such, mimics the neurons in the brain. Each
performance is less affected by noisy or incomplete data. element is relatively simple, and is typically two function
Another advantage is that they do not require linearly sep- blocks, as shown in Figure 6a.
arable data sets. All neural networks consist of many interconnected
Many ANN configurations and training algorithms have nodes, each having several inputs xi and a single output y.
been used to build electronic noses, including (1) back- Weights wi are associated with each input, and the weighted
propagation-trained feed-forward networks, (2) self-orga- sum S is calculated according to the formula:
nizing maps (SOMs) (Kohonen, 1989), (3) learning vector n
quantizers, (4) fuzzy ARTmaps, (5) Hamming networks, S
兺 wi xi
i
0
(1)
(6) Boltzmann machines, and (7) Hopfield networks.
Classification systems applicable to electronic noses are The weighted sum passes through some thresholding func-
reviewed by Horner (1995). Modern neural networks may tion, which is a nonlinear in-output transfer function. In
be either realized in hardware or a simulation program run- some systems, if the value of S exceeds the threshold func-
ning on a computer. The software simulation, although tion, then the output is 1. If S is less, then the output is 0.
generally slower, is less expensive and more flexible than This is termed hard limiting.
Olfactory System Cybernetics 303

Figure 5 A principal component analysis of replicate samples of different analytes. Each point represents a multidimensional pattern
from 32 sensors projected onto two dimensions.

The Perceptron was invented in the 1950s. It has an input derivative of the activation function. is the learning para-
and an output layer and contains multiple processing ele- meter,  is the momentum parameter.
ments (PEs). Perceptrons can solve pattern recognition prob- If one considers the activation function from Eq. (1) and
lems only if a hyperplane can separate the classes. A generalizes it, then it sums the product of the output from
feedforward network has the ability to separate complex each unit below and the weight with which it is connected
data. Such a network is termed a multilayer perceptron to the current unit.
(MLP) if it uses the gradient descent method to learn. The
冢兺w o 冣
m
multilayer perceptron is an extended version of the percep- oi
f ji i (2)
tron (see Fig. 6b). It contains inner (or hidden) layers, and the i
1

structure can be related to the neuron/synapse structure of the The simplest and most common output measure is the dif-
brain. It is these elements that store the learned responses. ference between the output from a unit oi and the desired
The choice of the number of layers and the number of ele- value di for that unit. The error derivative for the output unit
ments in each layer is crucial, since too few elements can is calculated as:
cause slow learning and inaccurate responses, but too many
ei
f ' (oi) (di  oi) (3)
elements can cause redundant copying of the input layers.
We can define the following terms. A single unit will be so that the error is now expressed as the derivative of the
referred to by the indices i and j attached to the symbols activation function. Errors may now be backpropagated to
listed below. Unit j may be on any layer of the network, but the hidden layer, where the error on each hidden unit in
where both i and j are used, unit j will be closer to the out- layer i is calculated as the summed product of the error
put layer than unit i. The values n and m denote the num- derivatives ej of the units in row j above and the strengths
ber of units in the current layer. wji denotes the strength of of the weights connected to them.
the weight from unit i to unit j on the next layer. The n

weight change to be made is denoted by Wji. oj denotes ei
f '(oi) 兺 ejwji

j
1
(4)
the output from unit j. vj is the input to unit j. dj is the
desired output from unit j. ej is the error derivative on unit The weights may then be updated over the whole net-
j. f(•) is the network activation function, and f ’(•) is the work shown in Figure 6b, the change in weight from unit i
304 Persaud

Figure 6 (a) Functional block of a node in a neural network comprising a summation of several inputs followed by a nonlinear trans-
form. (b) Architecture of a multiplayer perceptron, a commonly used neural network architecture.

to unit j is calculated as the product of the learning rate, the IV. APPLICATIONS
error derivative for unit j and the output from unit i:
wji
 ej oi Electronic noses are in commercial use for a wide variety
(5)
of odor and volatile compound applications. Their most
where unit i is in the layer below unit j. popular applications currently are in food processing, envi-
Hence by presenting exemplar patterns to the input ronmental monitoring, medical diagnostics, process con-
nodes of a neural network, it may be trained so that only trol, and fragrance development.
specified output nodes are activated for a given class of
pattern.
The two most commonly used approaches for electronic A. Food Industry Applications
noses are backpropagation-trained feed-forward networks
and SOMs (Ziegler et al., 1998). Backpropagation is a Traditionally, food quality was assessed by panels of human
supervised algorithm that must be trained with labeled experts and through application of analytical chemistry
odors. It learns the relationship between the sensor values methodology. These are costly and time-consuming and can
and the given odor labels. A SOM is an unsupervised algo- be difficult to calibrate and establish sustained reliability.
rithm that does not require predetermined odor classes for Electronic noses do not replace sensory panels, but they have
training. It essentially performs clustering of the data into useful roles in the food processing industry, especially where
similar groups based on the measured attributes or features simple comparisons need to be made. A host of applications
that serve as inputs to the algorithm. It can be thought of as tested include quality assessment in food production, inspec-
way of projecting multiple dimensions (often each dimen- tion of food quality by odor, control of food cooking processes,
sion represents a different sensor output or a feature inspection of fish, monitoring fermentation processes, check-
extracted from the sensor array) onto a two-dimensional ing for rancidity and spoilage, establishing fruit ripeness, ver-
output allowing the user to visualize the groupings and rela- ifying sources of juice concentrates, and inspecting
tionships of the odors or chemical volatile compounds. packaging material for malodors (Ali and O’Hare, 1997;
Olfactory System Cybernetics 305

Figure 7 Use of a neural network to predict aging of biscuit samples. As the biscuits age, the fats oxidize and the measured odor pat-
tern changes. The neural network is then used to map these changes so that the lifetime of the product can be predicted in terms of weeks
of storage life. Each point corresponds to an odor-measurement pattern, and these are mapped to time from the fresh control sample by
the neural network.

Aparicio et al., 2000; Boerjesson et al., 1996; Brezmes et al., an electronic nose. Odors coming from body fluids can
2000). Figure 7 illustrates an application using conducting indicate metabolic problems as well as infections.
polymer sensor arrays in investigating the shelf life of bis- Other extremely interesting biological applications
cuits where oxidation of fats cause changes in smell and taste are now being developed. These include monitoring
of the product prior to the onset of rancidity. In this case a mammalian cell growth (Bachinger et al., 2000), DNA
neural network–based prediction system was developed detection (Clausen-Schaumann, et al., 2000), and growth
using radial basis function neural network architecture. of bacterial cell cultures (Gardner et al., 1998).

B. Medical Diagnostic Applications C. Environmental Applications

Smell used to be a common diagnostic tool in medicine, Environmental applications of electronic noses increasingly
and physicians were trained to use their sense of smell include agricultural malodor applications. Malodors
during their medical training. In modern times, odor diag- emanating from cow and pig slurries are an increasing
nostics have been relegated to secondary status as a diag- source of environmental pollution, as well as an odor nui-
nostic method only applicable in primative settings. sance to human populations in the vicinity. Many substances
Electronic noses now offer the potential of a robust analyt- produced during the anaerobic digestion of feces have very
ical approach to odor measurement for medical diagnostics low human olfactory thresholds and so are perceived as
(Gardner et al., 2000; Gibson et al., 1997). Electronic nose odor nuisances at very low concentrations in air. These
technology has been used to examine odors emitted from include volatile fatty acids, p-cresol, amines, sulfhides,
the body such as from breath, wounds, and body fluids, disulfide, mercaptans, and many heterocyclic compounds. It
and to identify possible problems, such as bacterial vagi- is now possible to correlate sensor responses to odor mea-
nosis (Chandiok et al., 1997). Breath analysis can be used surements derived from olfactometry using a human panel
to diagnose gastrointestinal probems, sinus problems, (Misselbrook 1997; Persaud et al., 1996 a, b).
infections, diabetes, and liver problems. Infected wounds Monitoring of indoor air quality is also an important
and tissues emit distinctive odors that can be detected by application (Hathcock, Jr., 1999). A hybrid system was
306 Persaud

used aboard the Russian space station MIR to monitor air Amoore, J. E. (1967). Specific anosmia: a clue to the olfactory
quality within the station (Persaud et al., 1999). code. Nature 214:1095–1098.
The problems of monitoring industrial chemical haz- Aparicio, R., Rocha, S. M., Delgadillo, I., and Morales, M. T.
ards may be addressed by array sensing technology (Huby, (2000). Detection of rancid defect in virgin olive oil by the
electronic nose. J. Agric. Food Chem. 48:853–860.
1999; Khopkar, 1998; Stetter et al., 1984). Contaminating
Baby, R. E., Cabezas, M., and Walsoe de Reca, E. N. (2000).
residues of insecticides (lindane and synthetic pyrethroids)
Electronic nose: a useful tool for monitoring environmental
and products from the manufacture of leather and plastic contamination. Sens. Actuators B B69:214–218.
products (phenols, nitrobenzene, anilines) are often Bachinger, T., Riese, U., Eriksson, R. K., and Mandenius, C. F.
offloaded into streams or rivers, despite legislation and in (2000). Electronic nose for estimation of product concentra-
disregard of the danger they pose to the health of the pop- tion in mammalian cell cultivation. Bioproc. Eng.
ulation and the survival of fish and flora. Electronic nose 23:637–642.
devices may prove promising in such monitoring applica- Beets, M. J. G., (1978). Structure-Activity Relationships in
tions (Baby et al., 2000). Human Chemoreception. Applied Science Publishers Ltd.,
London.
Boelens, H. (1974). Relationship between the chemical structure
D. Other Applications of compounds and their olfactive properties. Cosmet. Perfum.
89:1–7.
Many interesting applications are being developed, and the Boerjesson, T., Ekloev, T., Jonsson, A., Sundgren, H., and
state of the art is reviewed in Gardner and Persaud (2000). Schnuerer, J. (1996). Electronic nose for odor classification of
They include fruit ripening, determining the origin of olive grains. Cereal Chem. 73:457–461.
oil, spoilage of meat and fish, grain quality, fermentation Brezmes, J., Llobet, E., Vilanova, X., Saiz, G., and Correig, X.
monitoring, city pollution, and thousands of others. It is clear (2000). Fruit ripeness monitoring using an electronic nose.
Sens. Actuators B B69:223–229.
that the perceived usefulness of such technology is high.
Buck, L. B. (1997a). Information coding in the olfactory system.
J. Neurochem. 69:S210.
Buck, L. B. (1997b). Molecular mechanisms of odor and
V. THE FUTURE pheromone detection in mammals. Mol. Biol. Cell 8:739.
Buck, T. M., Allen, F. G., Dalton, M. (1965). Selection of chem-
With any new technology, there are many unanticipated ical species by surface effects on metals and semiconductors.
problems. The field of odor sensing is a multidisciplinary In Surface Effects in Detection. J. I. Bregman and A.
area that attracts academic and industrial scientists from a Dravnicks (Eds.) Spartan Books Inc., Washington, D. C.
wide range of fields. Perceived problems that need further Chandiok, S., Crawley, B. A., Oppenheim, B. A., Chadwick, P.
development include sampling methodology, transferability R., Higgins, S., and Persaud, K. C. (1997). Screening for bac-
of databases from one sensor array to another, repeatability terial vaginosis: a novel application of artificial nose technol-
ogy. J. Clin. Pathol. 50:790–791.
and reproducibility over the long term, the need for stan-
Chess, A., Buck, L. B., Dowling, M. M., Axel, R., and Ngai, J.
dardization of test protocols, and improved sensor technolo-
(1992). Molecular-biology of smell—expression of the multi-
gy. All of these areas are in active development, and it would gene family encoding putative odorant receptors. Cold Spring
appear that there is a long-term future for this technology. At Harbor Symp. Quant. Biol. 57:505–516.
this point, many devices exist that are applicable to specific Clausen-Schaumann, H., Rief, M., and Seitz, M. (2000).
tasks. The electronic nose does not yet emulate a human Artificial noses sniff DNA. ChemPhysChem 1:89–90.
nose, and the development of an artificial nose, where the Cranny, A. W. J., and Atkinson, J. K. (1992). The use of pattern
output code from the sensor array can be easily correlated to recognition techniques applied to signals generated by a
human sensory perception, is still some time away. multi-element gas sensor array as a means of compensating
for poor individual element response. In Sensors and
Sensory Systems for an Electronic Nose, J. W. Gardner and
P. N. Bartlett (Eds.). Kluwer Academic Publishers,
REFERENCES Dordrecht.
Deutsch, S. (1967). Models of the Nervous System. John Wiley &
Ali, Z., and O’Hare, L. (1997). Analyzing food flavors, Chem. Sons, New York.
Rev. (Deddington, U.K.) 6:2–7. Di Natale, C., Salimbeni, D., Paolesse, R., Macagnano, A., and
Amoore, J. E. (1962a). The stereochemical theory of olfaction. 1. D’Amico, A. (2000). Porphyrins-based opto-electronic nose
Identification of the seven primary odours. Proc. Sci. Sec. for volatile compounds detection. Sens. Actuators B B65:
Toilet Goods Assoc. 37:1–12. 220–226.
Amoore, J. E. (1962b). The stereochemical theory of olfaction. 2. Diaz, A. F., Kanazawa, K. K., and Gardini, G. P. (1979).
elucidation of the stereochemical properties of the olfactory Electropolymerisation of pyrrole. J. Chem. Soc. Chem.
receptor sites. Proc. Sci. Sec. Toilet Goods Assoc. 37:13–23. Commun. 635–636.
Olfactory System Cybernetics 307

Dickinson, T. A., White, J., Kauer, J. S., and Walt, D. R. (1996). Horner, G. (1995). Signal processing for chemosensor arrays.
A chemical-detecting system based on a cross-reactive optical Tech. Mess. 62:166–172.
sensor array. Nature (London) 382:697–700. Hotelling, H. (1933). Analysis of a complex of statistical vari-
Dickinson, T. A., White, J., Kauer, J. S., and Walt, D. R. (1998). ables into principal components. J. Ed. Psychol., 24:417–520.
Current trends in ‘artificial-nose’ technology. Trends Huby, F. J. (1999) Strategies for solving odor problems related to
Biotechnol. 16:250–258. synthetic industrial chemicals. In Electron. Noses Sens. Array
Dickinson, T. A., Michael, K. L., Kauer, J. S., and Walt, D. Based Syst., Proc. Int. Symp. Olfaction Electron. Nose, 5th. J.
R. (1999). Convergent, self-encoded bead sensor arrays in R. Stetter and W. R. Penrose (Eds.)
the design of an artificial nose. Anal. Chem. 71: Khopkar, S. M. (1998). Novel devices for fast monitoring of pol-
2192–2198. lutants. Res J. Chem Environ. 2:15–16.
Dittmann, B., and Nitz, S. (2000). Strategies for the development Kohl, D. (1989). Surface processes in the detection of reducing
of reliable QA/QC methods when working with mass spec- gases with tin dioxide-based devices, Sens Actuators
trometry-based chemosensory systems. Sens. Actuators B 18:71–113.
B69:253–257. Kohl, D. (1991). Chemical sensor materials. Mater. Sci. Monogr.
Dittmann, B., Nitz, S., and Horner, G. (1998). A new chemical 66D:2445–2456.
sensor on a mass spectrometric basis. Adv. Food Sci. Kohl, D. (1996). Fundamentals and new applications of semicon-
20:115–121. ductor gas sensors. Vide: Sci., Tech. Appl. 39–44.
Dittmann, B., Horner, G., and Nitz, S. (1999). Application of a Kohl, D. (1996) Semiconductor and calorimetric sensor devices
new sensor system on the basis of mass spectrometry (MS- and arrays, In Handb. Biosens. Electron. Noses. Sect 41. Ed.
sensor) in food analysis. Lebensmittelchemie 53:11. E. Kress Rogers Press.
Dittmann, B., Zimmermann, B., Engelen, C., Jany, G., and Nitz, Kohonen, T. (1989). Self Organization and Associative Memory,
S. (2000). Use of the MS-sensor to discriminate between dif- 3rd ed. Springer-Verlag, New York.
ferent dosages of garlic flavoring in tomato sauce. J. Agric. Lewis, N. S., and Freund, M. S. (1996). Sensors arrays for
Food Chem. 48:2887–2892. detecting analytes in fluid. Patent 9630750, California
Dravnieks, A., and Trotter, P. J. (1965). Polar vapour detection Institute of Technology, Pasadena
based on thermal modulation of contact potential. J. Sci. Lewis, N. S., and Severin, E. (1997). Sensors for detecting ana-
Instruments 42:624–627. lytes in fluid, Patent 9807024, California Institute of
Ehrmann, S. (1998). KAMINA, the electronic nose from Technology, Pasadena.
Karlsruhe, Germany. A gas sensor system for mass products. Menzel, R., and Goschnick, J. (2000). Gradient gas sensor
Mess Tec 6:304–306. microarrays for on-line process control—a new dynamic clas-
Freeman, W. J. (1974). Dynamic patterns of brain cell assemblies. IV. sification model for fast and reliable air quality assessment.
Mixed systems. Oscillating fields and pulse distributions. Pulse- Sens. Actuators B B68:115–122.
wave problems. Neurosci. Res. Program Bull. 12: 102–107. Minsky, M., and Papert, S. (1969). Perceptrons: An Introduction
Freeman, W. J. (1990). Nonlinear dynamics in olfactory informa- to Computational Geometry. MIT Press, Cambridge, MA.
tion processing. In Proceedings of a Conference held at Misselbrook, T. H., Hobbs, P. J., and Persaud, K. C. (1997). Use
Wellesley College, May 17–18, J. Davis and H. Eichenbaum of an electronic nose to measure odour concentration follow-
(Eds.). MIT Press, Cambridge, MA, pp. 225–249. ing application of cattle slurry to grassland, J. Agri. En. Re.
Gardner, J. W. (1998). Sensors update (volume 1): H Baltes, W 66:213–220.
Gopel and J Hesse. Meas. Sci. Technol. 9:717–718. Mombaerts, P., Wang, F., Dulac, C., Vassar, R., Chao, S. K.,
Gardner, J. W., and Persaud, K. C. (2000). Electronic Noses and Nemes, A., Mendelsohn, M., Edmondson, J., and Axel, R.
Olfaction 2000. Institute of Physics, Bristol. (1996). The molecular biology of olfactory perception, Cold
Gardner, J. W., Craven, M., Dow, C., and Hines, E. L. (1998). The Spring Harbor Symp. Quant. Biol. 61:135–145.
prediction of bacteria type and culture growth phase by an Nanto, H. (1997). Electronic nose. Odor sensor using quartz res-
electronic nose with a multi-layer perception network. Meas. onator microbalance, Nyu Seramikkusu 10:47–52.
Sci. Technol. 9:120–127. Persaud, K., and Dodd, G. (1982). Analysis of discrimination
Gardner, J. W., Shin, H. W., and Hines, E. L. (2000). An elec- mechanisms in the mammalian olfactory system using a
tronic nose system to diagnose illness. Sens. Actuators B model nose, Nature 299: 352–355.
B70:19–24. Persaud, K. C., and Pelosi, P. (1985). An approach to an artificial
Gibson, T. D., Prosser, O., Hulbert, J. N., Marshall, R. W., nose. Trans. Am. Soc. Artif. Organs 31:29–32.
Corcoran, P., Lowery, P., Ruck-Keene, E. A., and Heron, S. Persaud, K. C., Khaffaf, S. M., Hobbs, P. J., and Sneath, R. W.
(1997). Detection and simultaneous identification of microor- (1996a). Assessment of conducting polymer odour sensors for
ganisms from headspace samples using an electronic nose. agricultural malodour measurements, Chem. Senses 21: 495–505.
Sens. Actuators B B44: 413–422. Persaud, K. C., Payne, P. A., Khaffaf, S. M., Dowdeswell, R. M.,
Hatfield, J. V., Covington, J. A., and Gardner, J. W. (2000). Hobbs, P. J., Misselbrook, T. H., and Sneath, R. W. (1996b).
GasFETs incorporating conducting polymers as gate materi- Application of conducting polymer odour sensing arrays to
als. Sens. Actuators B B65:253–256. malodour monitoring, In IEE Computing and Control
Hathcock, S. L., Jr. (1999). Sensor array technology applied to Division Digest, IEE, London, pp. 6/1–6/4.
indoor air quality: objectively measuring the subjective. Adv. Persaud, K. C., Pisanelli, A. M., Szyszko, S., Reichl, M., Horner,
Filtr. Sep. Technol. 13B:705–707. G., Rakow, W., Keding, H. J., and Wessels, H. (1999). A smart
308 Persaud

gas sensor for monitoring environmental changes in closed Spills Conf. Proc., Prev., Behav., Control Cleanup Spills Waste
systems: results from the MIR space station, Sensors Sites. April 1984, Nashville, TE, p. 83 ISBN-0-86587-064-0.
Actuators B-Chemical 55:118–126. Tanyolaç, N. N., and Eaton, J. R. (1950). Study of odors. J. Am.
Rapp, M., and Reibel, J. (1996). Gas analysis with sensor sys- Pharm. Assoc. 39:10.
tems. A route to the electronic nose? Nachr. Chem., Tech. Lab. Thom, R. (1970). Topological models in biology. In Towards a
44:1088–1092. Theoretical Biology.—3: Prolegomena, an IUBS Symposiun,
Rosenblatt, F. (1962). Principles of Neurodynamics. Spartan Waddington, C. H. (Ed.). Edinburgh University Press,
Books, Washington, DC. Edinburgh, pp. 89–116.
Rumelhart, D. E., Hinton, G. E., and Williams, R. J. (1986). Thom, R. (1972). Stabilitite Structurelle et Morphogenese.
Parallel Distributed Processing: Explorations in the Addison Wesley, New York.
Microstructures of Cognition. Vol. 1: Foundations, MIT Press, Turing, A. (1952). The chemical basis of morphogenesis, Phil.
Trans. Roy. Soc. London B 237:32–72.
Cambridge, MA.
Uttal, W. R. (1973). The Psychobiology of Sensory Coding.
Selfridge, O. G. (1959). Pandemonium: a paradigm for learning. In
Harper & Row, New York.
The mechanisation of thought processes, HMSO, London, UK.
Walt, D. R., Dickinson, T., White, J., Kauer, J., Johnson, S.,
Shepherd, G. M. (1998). The Synaptic Organization of the Brain,
Engelhardt, H., Sutter, J., and Jurs, P. (1998). Optical sensor
Oxford University Press, New York.
arrays for odor recognition, Biosensors Bioelectronics. 13:
Shepherd G. (Ed.) (1998). The Synaptic Organization of the 697–699.
Brain, 44th Edition. Oxford University Press, Oxford. Walt, D. R., Dickinson, T. A., Healey, B. G., Kauer, J. S., and
Slater, J. M., and Watt, E. J. (1991). Examination of ammonia- White, J. (1995). Fiber optic array sensors as an architecture
poly(pyrrole) interactions by piezoelectric and conductivity for an artificial nose, Proc. SPIE-Int. Soc. Opt. Eng.
measurements, Analyst 116:1125. 2508:111–116.
Stetter, J. R. (1995). Amperometric electrochemical gas sensors: Wiener, N. (1948). Cybernetics: or Control and Communication
description and applications, NIST Spec. Publ. 865: 61–64. in Animal and Machine. MIT Press.
Stetter, J. R., Strathmann, S., McEntegart, C., Decastro, M., and Yang, Y. M., Yang, P. Y., and Wang, X. R. (2000). Electronic nose
Penrose, W. R. (2000). New sensor arrays and sampling sys- based on SAWS array and its odor identification capability,
tems for a modular electronic nose, Sens. Actuators B B69: Sens. Actuators B B66: 167–170.
410–419. Ziegler, C., Gopel, W., Hammerle, H., Hatt, H., Jung, G.,
Stetter, J. R., Zaromb, S., Penrose, W. R., Findlay, M. W., Jr., Laxhuber, L., Schmidt, H. L., Schutz, S., Vogtle, F., and Zell,
Otagawa, T., and Sincali, A. J. (1984). Portable device for A. (1998). Bioelectronic noses: a status report. Part II,
detecting and identifying hazardous vapors. In Hazard. Mater. Biosensors Bioelectronics 13:539–571.
15

Olfaction and the Development of Social Behavior

in Neonatal Mammals

Richard H. Porter
Institut National de la Recherche Agronomique/Centre National de la Recherche Scientifique, Nouzilly, France
Benoist Schaal
Centre Européen des Sciences du Goût, Dijon, France

I. INTRODUCTION than unfamiliar or unrelated animals. As discussed in

the present chapter, volatile chemicals play an important
The survival and development of newborn mammals role in the mediation of early interactions among con-
depends upon complex coordinated interactions with the specifics, as well as in the development of social discrimi-
mother and, at least in some species, other conspecifics, nation and preferences (see also Chapter 17). Although the
including the father and siblings. Although neonates can- primary focus of this chapter is the influence of biological
not live without the resources provided by their mother, odors on adaptive behavioral responses and social prefer-
they are not simply passive recipients of maternal care. ences of human infants, studies of other species are cited
Beginning shortly after birth, mammalian young exhibit when appropriate to illustrate or clarify particular points.
species-typical behaviors that contribute to their own well- The term “olfaction” is used to refer to the perception of
being. Mothers in particular are highly sensitive to signals odorants even when the underlying chemosensory systems
that communicate the physiological and motivational state have not been identified precisely. Thus, the possible
of their infants. For example, when removed from their involvement of chemical senses other than the olfactory
nest, mouse and rat pups emit ultrasonic “distress” cries system per se (e.g., trigeminal nerve, vomeronasal organ,
that attract their mother and thereby enable her to retrieve nervus terminalis) is subsumed under this label.
them (Allin and Banks, 1972; Noirot, 1966; Zippelius and
Schleidt, 1956). Isolated lambs bleat, similarily facilitating
II. INFLUENCE OF CONSPECIFIC CHEMICAL
parent-offspring reunion and nursing (Walser et al., 1984).
SIGNALS ON THE BEHAVIOR OF NEONATES
In a reciprocal manner, adults communicate via various
sensory modalities with their newborns, who, in turn, dis- A. Nonhuman Mammals
play overt responses to their parents (or cues that they
produce) that are critical for survival. Empirical studies of the responses of newborn mammals to
The ability of mammals to discriminate between indi- naturally occurring olfactory cues have been conducted
viduals or members of different social categories is evident primarily with Norway rats and a few other easy to main-
at an early age (Colgan, 1983; Hepper, 1991; Holmes, tain and breed species of rodents and lagomorphs. Prior to
1991). Suckling young of many species rapidly develop the weaning, pups of such species are continuously exposed to
ability to recognize their own mother and respond prefer- an array of odors emanating from urine, feces, and spe-
entially to her as compared to other females. Moreover, lit- cialized glands of their mother and littermates, which per-
termates—or other classes of close kin —interact differently meate the nest area.

309
310 Porter and Schaal

1. Arousal The role of maternal olfactory cues in neonates’ feeding

behavior can also be assessed by directly manipulating the
When young rodents are deprived of maternal odors, an
odorant source. Thorough cleansing of the nipples of anes-
increase in distress calling, as well as exploratory or loco-
thetized lactating female rats results in reduced sucking
motive behaviors, typically occurs (e.g., Schapiro and
efficiency (Hofer et al., 1976; Teicher and Blass, 1976).
Salas, 1970). Such behaviors are mitigated by reintroduc-
When thermal, tactile, or olfactory characteristics of the
tion of the mother or by contact with air that has been
nipple region are altered, only the latter treatment
passed over the mother or her soiled bedding, although in
adversely effects nipple attachment (Blass et al., 1977).
rats soiled bedding is effective only in young pups (ages
Initial nipple attachment by rat pups is guided by the odor
5–13 days) (Hofer and Shair, 1980; Oswalt and Meier,
of saliva or amniotic fluid spread by the mother on her ven-
1975; Randall and Campbell, 1976; Shair et al., 1997). Rat
trum while grooming during parturition (Blass and
pups rendered anosmic by destruction of the olfactory
Teicher, 1980).
receptors with nasal infusions of zinc sulfate solution con-
Nipple localization or attachment by suckling rabbits
tinue to display heightened rates of ambulation even when
also depends upon maternal olfactory cues. In contrast to
exposed to their mother (Hofer, 1975, 1976), reiterating the
rodents, female rabbits return to their nest only once a day
importance of olfactory cues in suppressing such behavior.
for a single brief nursing period (typically lasting several
minutes) (Hudson and Distel, 1982; Zarrow et al., 1965).
2. Physical Attraction/Orientation Once the mother positions herself over the litter, the pups
Lactating female rats produce chemical signals that are must quickly find a nipple and begin to suck, or risk star-
attractive to suckling young. The pioneering research by vation. Anosmic rabbit pups exhibit anomalies in nipple
Leon established that the emission of such signals by nurs- attachment and feeding behavior during this critical time
ing females, as well as their pups’ physical orientation in (Schley, 1977). Related research has revealed that rabbit
the direction of those cues, are strongest from the time that pups are attracted to an odorous substance carried in the
the offspring become capable of independent locomotion mothers milk and concentrated at the base of the nipples
until they are weaned (Leon and Moltz, 1971, 1972). Pups (Coureaud et al., 2000; Hudson and Distel, 1983; Keil et al.,
of that age move preferentially toward the odor of their 1990; Schley, 1981).
own mother or the odor of a strange lactating female,
rather than an empty goal box or the scent of a nulliparous 4. Huddling
female. The mother therefore appears to emit an “olfactory
tether” that guides pups to her or the nest area and keeps In litter-bearing mammals, huddling by littermates is a
them from wandering away. Subsequent studies have conspicuous form of early social interaction. Altricial rat
found evidence of similar olfactory attractants produced by pups reduce heat loss and oxygen consumption by main-
lactating females of other species, including house mice taining bodily contact with their siblings and actively
(Breen and Leshner, 1977) and spiny mice (Acomys cahir- exchanging positions within the huddle (Alberts, 1978b).
inus) (Porter and Doane, 1976). The sensory control of huddling behavior changes over the
first 2 weeks after birth. During the first week, huddling
preferences are determined primarily by the temperature of
3. Nipple Localization and Sucking
the target stimulus. In one study, 5-day-old pups spent
As early as the first nursing bout, neonates are active par- more time huddled with a warm, fur-covered tube than
ticipants in the feeding process. Successful ingestion of with an anesthetized rat (Alberts and Brunjes, 1978).
mother’s milk requires locating, grasping, and sucking a However, the reverse preference was observed from day 10
nipple. Olfactory signals are critical in this process. Thus, onwards. An anesthetized gerbil was as effective as anoth-
nipple localization and attachment and maintenance of er rat in eliciting huddling responses by 5- to 10-day-old
weight are seriously disturbed in rat and mouse pups pups, but there was a clear conspecific preference begin-
following olfactory bulbectomy or zinc sulfate–induced ning on day 15.
anosmia (Cooper and Cowley, 1976; Risser and Slotnick, As pups become older, olfaction plays an increasingly
1987; Singh et al., 1976; Teicher et al., 1978). Because important role in huddling choices. Whereas zinc sul-
mothers displayed no observable differences in their fate–induced anosmia appears to have no effect on
responses to anosmic versus normal offspring, problems in younger pups’ huddling with an anesthetized cagemate,
finding the nipple appear to reflect alterations in the pups’ the same treatment markedly disrupts such huddling on
capacities rather than inadequate maternal behavior day 10 (Alberts, 1978a). Contrary to untreated control
(Cooper and Cowley, 1976; Singh and Tobach, 1975). pups, 2-week-old anosmic animals spend as much time in
Olfaction and Social Behavior in Neonatal Mammals 311

contact with a heated inanimate object as with an anes- that only a rudimentary structure subsists in the older fetus
thetized target rat (Alberts and Brunjes, 1978). Egyptian or newborn. However, much more anatomical investiga-
spiny mouse (Acomys cahirinus) neonates likewise select tion is needed before we have a clear picture of the status
huddling partners based upon their odor phenotypes, but of the vomeronasal organ in the perinatal period (see
congregate in indiscriminate clumps after their nares are Chapter 46). A final potential chemoreceptive system is
flushed with zinc-sulfate solution (Porter et al., 1978). present in the fetal nasal passages, the terminal nerve,
which sends free nerve endings to the anterior nasal sep-
tum and to the olfactory mucosa (Brown, 1987;
B. Human Neonates Oelschlager et al., 1987). The chemosensory functions of
this structure remain unexplored in humans (but see
1. Early Development and Responsiveness of the
Chapter 48 for such functions in other forms).
Chemoreceptive Systems
The relative contributions of these four different
At least four chemoreceptor subsystems are located within chemoreceptive systems to smell sensation is unclear.
the human nose, all of which differentiate very early dur- However, it is generally accepted that the olfactory system
ing prenatal ontogeny—the main olfactory system (CN I), is tuned to detect low concentrations of volatile odorants,
trigeminal system (CN V), vomeronasal organ, and the ter- whereas the trigeminal system is mainly sensitive to the
minal nerve (CN O). The major developmental events of irritative effects of higher intensity chemical stimulation. It
these systems are summarized below to highlight their is questionable whether the vomeronasal and terminal sys-
potential functions from the fetal period onwards (for tems are functional in humans. The reader is referred to
reviews, see Doty, 1991, 2001; Schaal, 1988a; Schaal and Chapters 46–48 for a more detailed discussion of verte-
Orgeur, 1992) (see also Chapters 46–48). brate studies of these systems.
The main olfactory system develops during the first Nasal chemoreceptors develop in a prenatal environment
trimester of gestation (see Chapter 6). At the most periph- that carries a wealth of potential chemosensory stimuli
eral level, mature-appearing ciliated olfactory neurorecep- (Schaal and Orgeur, 1992; Schaal et al., 1995d). Odorous
tors are seen as early as week 11 of gestation (Pyatkina, compounds enter the amniotic fluid via transcutaneous
1982), and histological studies between gestational months water transfer of tracheal and gut wastes, especially the
5–9 reveal that the fetal olfactory epithelium is well devel- ever-increasing urination of the fetus. The volume and com-
oped by that time (Nakashima et al., 1984). Olfactory neu- position of these substances fluctuate throughout gestation
roreceptors grow clusters of axons toward the forebrain, and even display daily cycles (Abramowich, 1981).
forming a visible olfactory nerve by 7–8 weeks. The shape Additional sources of such olfactory stimuli include the
of the olfactory bulb is evident around 6–8 weeks (Bossy, mother’s metabolic activity, immunogenetic constitution,
1980), and its stratified internal structure is seen from and diet (Hauser et al., 1985; Mennella et al., 1995).
week 10 onwards (Humphrey, 1940). Development of Ultrasonography allows direct visualization of the flow
more central structures involved in olfactory processing of fluids through the fetal respiratory tract (Logvinenko,
has not been well described, but it is clear from the behav- 1990). Swallowing and inhalation movements observed by
ioral and psychophysical data presented below that they such methods increase the volume of amniotic fluid that
are functional, at least in a primitive sense, prenatally. comes into direct contact with fetal chemoreceptors. It is
The trigeminal system, the free nerve endings of which interesting to note that fetuses inhale much more fluid than
mediate somesthesic and common chemical sensations, is they swallow (Duenholter and Pritchard, 1976), suggesting
visible in 4-week-old embryos and has been demonstrated that nasal chemoreceptors may be intensely stimulated.
to respond to tactile stimulation between 7.5 and 10 gesta- Thus, before birth, each neonate is likely to be exposed to
tional weeks (Brown, 1974). The vomeronasal or a unique profile of dietary aromas and other compounds
Jacobson’s organ is well developed in the human fetus, related to feto-maternal metabolism. The manner in which
appearing as small symmetrical invaginations in the nasal these early environmental conditions contribute to the
septum near the nare opening (Moran et al., 1995). shaping of chemosensory function (sensitivity, prefer-
Sensory-like cells in this structure are detected at 5–13 ences) is detailed below.
gestational weeks (Bossy, 1980; Ortmann, 1989). The sec- The functional onset and abilities of the nasal structures
ondary central structures associated with the vomeronasal are difficult to investigate directly in the human fetus for
organ, the accessory olfactory bulbs, can be clearly obvious ethical and practical reasons. However, data from
localised between 5 and 18 weeks. After this time, regres- two indirect research strategies suggest that they are
sion or reorganization of the accessory olfactory bulbs involved in early sensory processing by the fetal and
occurs in some individuals. Humphrey (1940) suggests neonatal brain. One of these strategies assesses the
312 Porter and Schaal

responsiveness of premature human infants shortly after had a noxious odor should treat this problem by consuming
birth to gain insight into functional chemosensory abilities “fragrant wine and sweet food.” More recently, Charles
of fetuses of equivalent gestational age. Limited data from Darwin (1877) remarked that his one-month-old son
a study by Sarnat (1978) suggest that infants born after 24 behaved as if he “perceived his mother’s bosom when three
(or fewer) weeks of gestation are only weakly responsive or four inches from it.” Darwin expressed doubt that the
to nasally administered chemical stimuli, whereas those of baby’s response was based upon visual cues, but specu-
at least 28 weeks gestational age exhibit reliable behavio- lated that he may have been “guided through smell.”
ral responses to odors. However, since menthol was the It is somewhat surprising that experimental evidence in
odorant used in this study, both trigeminal and olfactory support of Darwin’s hypothesis was published almost
afferents may have been activated. More recent experi- a full century after the anecdotal description of his son’s
ments suggest that at 31–37 weeks gestation, premature perception of his mother’s “bosom.” While observing
infants detect and discriminate among lower intensity mothers and infants during breastfeeding, Macfarlane
odorants that activate primarily either the olfactory or the (1975) also noticed that neonates turned their head
trigeminal system (nonanoic acid and cineole, respect- towards the mother’s breast before there was any physical
ively), but not both (Pihet et al., 1996, 1997). contact with it and made no obvious eye movement in that
A second research strategy examines responsiveness direction. Based upon these preliminary observations, he
of the fetus to chemical stimuli. This approach involves developed a method for systematically testing infants’ ori-
animal models and complements related experiments entation responses to various combinations of paired odor
with human premature infants, in that it addresses the stimuli. Infants were placed individually on their back in
issue of whether chemosensory systems can function in a cot, with two stimulus pads suspended next to the baby’s
a liquid medium. Fetal rats receiving intra-oral infusions face, one touching each cheek (Fig. 1). Two successive
of citral (lemon scent) during their last day of gestation tests were conducted using the same pair of odor stimuli,
display sharp increases in gross motor activity and with their left-right positions reversed after the first trial to
changes in heart rate (Smotherman et al., 1991), possibly control for any directional bias. When presented simultan-
mediated via stimulation of the trigeminal nerve (Allen, eously with a pad that had been worn on their mother’s
1929; Doty, 1995). Likewise, intranasal infusions of breast and a clean control pad, 17 of 20 breast-fed infants
odorous solutions of citral or methylthiazoline elicit dif- (2–7 days old) spent more time oriented to the former
ferential heart-rate variations in near-term ovine fetuses; stimulus, thereby indicating that they perceived and were
while control saline does not trigger significant changes, attracted to the maternal breast odor.
citral induces a weak accelerative effect and foul-
smelling methylthiazoline a sharp decelerative response
(Schaal et al., 1991). Very similar results were obtained
with mouse fetuses (Coppola and Millar, 1997).
Although the specific sensory system(s) mediating such
prenatal responses to distinct chemical cues remains to
be elucidated, these results indicate that shortly before
birth murine and ovine fetuses are able to detect and dis-
criminate between odorants even when they are presented
in aqueous solutions. There is no biological basis for
assuming that the same physicochemical processes do
not apply to the human case.

2. Nipple Localization/Sucking
Discussions of the presumed significance of breast and
milk odors produced by nursing mothers can be found in
documents dating to antiquity (reviewed by Fildes, 1986).
For example, an Egyptian papyrus manuscript written in
the sixteenth century B.C. suggests that bad breast milk
smells like fish, while good milk can be recognized by its
manna-like odor. Seventh century A.D. writings by Paul of Figure 1 Apparatus for testing neonates’ directional choice
Aegina recommended that nursing mothers whose milk response to two simultaneously presented odorized pads.
Olfaction and Social Behavior in Neonatal Mammals 313

Following Macfarlane’s pioneering research, a number babies were exposed to the odors of their mother’s breast
of studies using similar testing procedures found that olfac- (Meza et al., 1998; Russell, 1976). To assess whether
tory cues emanating from the breasts of lactating women chemical cues associated with the maternal nipple/areola
are particularly attractive to young infants. For example, region are implicated in spontaneous feeding behavior in
2-week-old bottle-fed infants oriented longer to a breast the natural context, Varendi et al. (1994) laid infants in a
pad from an unfamiliar nursing mother than to the axillary prone position between the mother’s breasts after one
odor of this same woman or to breast odors from a nonpar- breast had been thoroughly washed to eliminate (or at
turient woman (Makin and Porter, 1989). Responses to an least reduce) naturally occurring odors. Observations
axillary pad from a lactating woman did not differ from commenced 5–13 minutes postpartum and continued
those to an odorless control pad. Another study found that until the infant found a nipple and began to suck vigor-
neonates of this same age, who had been exclusively bottle ously—with no assistance from the mother. From the
fed since birth, spent more time turned towards a pad con- total sample of 30 infants, 22 selected the unwashed
taining a lactating female’s breast odor than a pad scented breast for their initial feeding bout (Fig. 2). Since the
with their familiar formula (Porter et al., 1991). Thus, even washing procedure had no effect on the surface tempera-
though the infants had no prior direct contact with such ture of the nipple/areola, the infants had most likely
breast odor and had repeatedly been exposed to the odor of responded to olfactory differences between the washed
their familiar formula in the reinforcing context of feeding, vs. unwashed breasts. This experiment was later replicat-
the breast odor elicited more interest. ed with a sample of 2- to 3-day-olds with similar results;
Further observations of newborn infants reveal that significantly more babies spontaneously grasped and
they are active participants in the nursing process and that sucked from the mother’s unwashed nipple/areola rather
maternal odors contribute to successful early nipple than the alternative breast that had been cleansed of its
attachment and sucking. Widstrom and colleagues (1987) natural scent (Varendi et al., 1997) (see Fig. 2).
described the spontaneous activity of infants placed on Additional evidence of infants’ sensitivity to odors
their mother’s bare chest immediately after delivery. A when breastfeeding is found in accounts of neonates’ reac-
recurring sequence of behavior was observed, including tions to strong odorants applied directly to the mothers’
hand-to-mouth movements, sucking movements of the nipples. In experiments conducted during the nineteenth
mouth and tongue, rooting, nipple attachment, and effec- century, neonates refused breasts treated with strong-
tive sucking. Heightened nonnutritive sucking move- smelling substances such as diluted petroleum, Asa
ments were similarly observed in other studies when foetida, and succinic acid (Kroner, 1882; Preyer, 1885).

Figure 2 Influence of breast and amniotic fluid (AF) odors on neonates’ spontaneous nipple preferences. ** p  0.01; *p  0.05.
(Adapted from Varendi et al., 1994, 1996, 1997.)
314 Porter and Schaal

When similar odors were presented at their nostrils, suck- who had recently given birth—relative to a clean control
ing infants responded by releasing the nipple or withdraw- gown (Sullivan and Toubas, 1998). In another study, the
ing (Garbini, 1896). The salience of odors in this context is effects of different naturally occurring odors on early
also suggested by anecdotal reports of disturbed sucking spontaneous crying were compared (Varendi et al., 1998).
patterns by neonates whose mothers had recently con- During a 60-minute test session commencing 30 minutes
sumed highly spiced/flavored foods or beverages. A survey after birth, babies who were exposed to the odor of their
of French infant-care manuals found that mothers are com- own amniotic fluid spent significantly less time crying
monly advised to refrain from consuming highly odorous than did babies with no odor exposure. In contrast, within
(or odor-producing) dietary items such as garlic, onions, each of the 15-minute test intervals, babies who were pre-
cabbage, leeks, asparagus, spices, alcohol, and coffee (De sented with their mother’s breast odor cried more than
Perceval and Lallemand, 1980). those in the control condition. Although a clear explana-
More recent scientific investigations have substantiated tion of the opposite effects of amniotic fluid and breast
the influence of maternal diet on the odor of breast milk. odors cannot be given at present, the calming effect of
Members of an adult sensory panel judged the perceived amniotic fluid may reflect prenatal familiarization with
intensity of the odor of mothers’ milk as being “most that substance, as discussed below. On the other hand (as
strong” or “more like garlic” 2 hours after the donor had seen above), odors emanating from the breasts of lactating
ingested garlic capsules (Mennella and Beauchamp, 1991). females are attractive to newborn infants and help guide
Nursing infants likewise appeared to detect the changes in them to an appropriate food source. Therefore, since the
their mother’s milk since they remained attached to the nip- babies in the latter experiment had continual exposure to
ple for a longer period and displayed higher sucking rates breast odors but were not able to locate a nipple and suck,
per feed when it smelled of garlic. The flavor of garlic they may have become disturbed or aroused. Macfarlane
transferred to breast milk did not seem to be strongly aver- (1975) similarly reported that 2- to 10-day-old breastfed
sive to these infants since there was no consistent effect on babies displayed increasing activity and crying after their
the amount of milk that they consumed. Additional com- nose remained in contact with a maternal breast pad.
mon substances that have been found to flavor breast milk
after being consumed by lactating women include vanilla, C. Ontogenetic Mechanisms
mint, cheese, and alcohol (Mennella, 1995).
Because neonates of a wide range of mammalian species
respond preferentially to particular conspecific odors
3. Arousal
shortly after birth (before they have the opportunity to gain
Aside from their influence on nipple localization and suck- significant postnatal chemosensory experience), such dis-
ing, maternal odors also appear to affect more generally criminative responsiveness may be termed “inborn.”
newborns’ motor activity and arousal. In one study, expos- Although the underlying bases of such olfactory prefer-
ing 2- to 10-day-old infants to either their mother’s breast ences have not been well elucidated, genetic, maturational,
or neck odor resulted in a reduction of arm and head move- and experiential factors have all been implicated. For
ments (relative to the responses to a clean control pad) example, selective responsiveness to odors could reflect
(Schaal, 1986). The effect of odors on autonomic activa- genetically mediated expression of olfactory receptors, as
tion was assessed in another study by recording infants’ illustrated by specific hyposmia in mice or in our own
respiratory rates (Soussignan et al., 1997). When a cotton species (Pourtier and Sicard, 1990; Whissel-Buechy and
swab moistened with milk from an unfamiliar lactating Amoore, 1973) (see also Chapter 16). Moreover, matur-
woman was placed near their nostrils, breast-fed infants ational changes in olfactory detection thresholds have been
evinced greater changes in their breathing rate (relative to described over the first several days after birth (Alberts and
control trials) than did formula-fed babies of the same age. May, 1980; Lipsitt et al., 1963). As noted below, investiga-
Human neonates that are separated from their mother tions of the role of experience in the development of early
(or caregiver) typically show high rates of crying that are odor preferences have often focused on whether such
abruptly curtailed when maternal bodily contact is reestab- responses are dependent on exposure or learning.
lished (Christensson et al., 1995). Crying in this context The biological substrates bearing odor cues attractive to
appears to be functionally analogous to the separation dis- newborns may be placed into two broad categories: those
tress vocalizations described in nonhuman mammalian to which the neonate’s chemoreceptors had obvious
young. One-day-old crying infants more rapidly stopped contact in utero, and those to which they were apparently
crying when they had access to the odor of a hospital gown not directly exposed prior to postnatal testing. Fetal fluids
that had been worn by their mother—or another woman examplify the first category of odor mixtures that are
Olfaction and Social Behavior in Neonatal Mammals 315

alluring to newborns. As discussed previously, rat pups are research also indicates that newborns of nonmammalian,
initially attracted to amniotic fluid that the parturient dam amniote animals (including species of amphibians, reptiles
had spread on her own abdomen (Teicher and Blass, 1977). and birds; e.g., Burghardt, 1971; Hepper and Waldman,
Whereas washed nipples were not attractive to pups, they 1992; Porter and Picard, 1998; Sneddon et al., 1998) like-
became so when painted with amniotic fluid from unfa- wise retain some “olfactory images” from their larval or
miliar parturient females, thereby indicating that neonatal embryonic chemical environment.
rats respond positively to any sample of such fluid, regard- Biological substrates to which there has been no obvi-
less of the donor female. Nonetheless, newborn rats dis- ous prior exposure are more varied. Newly born lambs dis-
play a preference for their own amniotic fluid over that play heightened autonomic and behavioral activation when
from another litter, which is evidence that they can dis- first exposed to the intense odor of ewe’s inguinal gland
criminate between these two olfactory substrates (Hepper, secretions (Vince and Ward, 1984). This arousal response
1987). Related studies with precocial newborn ungulates was greater for inguinal wax than for maternal wool or
found significant preferences for amniotic fluid over water milk and for the mother’s wax as compared with that from
[pigs (Parfet and Gonyou, 1991) and lambs (Schaal et al., an unfamiliar ewe. During their initial sucking attempts,
1995b)] and for own versus alien amniotic fluid [lambs neonatal piglets’ attraction to the nipple line is mediated to
(Schaal et al., 1995b)]. a great extent by “specific ventral substances” produced by
Early postnatal attraction to the odor of amniotic fluid the sow (Morrow-Tesch and McGlone, 1990). Another
suggests that the fetus becomes familiar with that smell example of such attractive cues is the odor that releases the
and retains a memory of it after birth. Evidence that odor stereotyped nipple-searching pattern in newborn rabbits.
learning, in fact, can occur in utero has been obtained This stimulus, which is emitted on the lactating doe’s
using a conditioned aversion paradigm involving prenatal abdomen in the vicinity of the nipples (Coureaud and
exposure to an artificial odorant (injected in the amniotic Schaal, 2000; Coureaud et al., 2000; Hudson and Distel,
fluid) paired with a noxious stimulus (intraperitoneal 1983), directs localization and oral grasping of the nipple.
injection of lithium chloride) (Smotherman, 1982; A functionally similar volatile compound or mixture is
Stickrod et al., 1982). In choice tests conducted after birth, also present in the doe’s milk, since pups respond to this
rat pups in the prenatal treatment condition, but not control substance with immediate orientation, searching head
pups, avoided the training odor. The aversions acquired in movements, and attempts to seize the object bearing the
this manner appear to be stimulus-specific since they did milk sample (Keil et al., 1990).
not generalize to other odorants. Furthermore, as the post- Because the neonates in the latter series of studies had
natal test ensured that the stimulus was presented exclu- typically been separated from their mother at birth, the
sively in a gaseous phase, the conditioned aversion was reported olfactory preferences are unlikely to be the result
likely mediated, at least in part, by olfaction. Pedersen and of prolonged postnatal experience. Nonetheless, there are
Blass (1982) employed a less intrusive manipulation in several alternative developmental mechanisms that could,
another study, whereby fetuses were merely exposed to an in theory, underlie the manifestation of early attraction to
artificial odorant injected into the amniotic sac (see also specific chemical stimuli by these newborn organisms.
Chotro and Molina, 1990; Smotherman, 1982). At birth Neonatal chemoreceptors might be attuned by prenatal
these pups were fostered onto untreated females to elimi- experience to detect somewhat similar cues involved in the
nate any possibility of odor transfer through the biological guidance of behavioral patterns crucial for survival (see
mother’s milk. One hour after birth, pups that had been below). Furthermore, brief postnatal exposure, often in
exposed to citral prenatally attached rapidly to unwashed association with the reinforcing effects of the mother’s first
nipples in a citral-odorized atmosphere but did not seize a nurturing activities (e.g., licking, nursing, warm and soft
nipple in the absence of that scent. The expression of this contact, vocalizations), could be sufficient to bias later
early preference was dependent, however, on a short hedonic responsiveness to those scents.
period of postnatal exposure to citral, contingent with tac- What evidence is there that human neonates remember
tile or pharmacological activation of the pups. Positive and make use of olfactory information that they acquired
results have also been reported when the fetal environment during their amniotic life? First, human newborns turn
of lambs (Nolte et al., 1995; Schaal et al., 1995b, 1995c), their head preferentially towards the odor of their familiar
rats (Hepper, 1988a), and rabbits (Bilko et al., 1995) was amniotic fluid when given a choice between that substance
manipulated by feeding females aromas throughout preg- and distilled water (Schaal et al., 1995a). Moreover, like
nancy. The structural, functional, and ecological requisites young rats and lambs, neonates of our own species demon-
for prenatal perception of chemosensory stimuli thus seem strate a head-orientation preference for the odor of their
to be verified during the late stages of gestation. Recent own amniotic fluid when it is contrasted with unfamiliar
316 Porter and Schaal

fluid in a two-choice test (Schaal et al., 1998). The fact that were observed in the biologically relevant context of
the infants in these experiments had their nasal passages breastfeeding. When one of a mother’s breasts was moist-
cleared and were washed within minutes after birth and ened with amniotic fluid, and the other not, significantly
that even bottle-fed infants (who were not reexposed to more babies selected, minutes after birth, the moistened
putative amniotic-like compounds that may be carried in breast over the non-moistened breast (Varendi et al., 1996)
breast milk) preferentially responded to their own amnio- (Fig. 2). Such early orientation to the odor of amniotic
tic fluid, supports the conclusion that they had become fluid may have been adaptive thoroughout the evolutionary
familiar with that odor within the uterus or during the birth history of our species, since women presumably handled
process. their babies during and immediately after expulsion from
Infants in a second series of studies were tested with the the birth canal. The mother’s hands would then have been
odor that they had encountered in their amniotic environ- soiled with the birth fluids, which in turn would have been
ment simultaneously paired with a highly salient postnatal transferred to her breasts when she first attempted to nurse
odor: colostrum or milk. Facing such a choice, 2-day-old her newborn infant. Attraction to the odor of amniotic fluid
breast-fed infants did not display differential head orienta- would therefore have facilitated nipple localization. In a
tion or mouthing movements (Marlier et al., 1997, 1998a) subsequent experiment, there was no reliable difference in
(Fig. 3). By 4 days of age, however, infants spent more the number of babies who spontaneously sucked from a
time oriented towards the odor of their mother’s milk than naturally scented breast, versus one treated with amniotic
to the scent of amniotic fluid (Marlier et al., 1998a). It can fluid, at 1–4 days of age (Varendi et al., 1997). However,
be concluded from these results that (1) breast-fed infants babies who had selected the AF-treated breast at that time
remain attracted during a limited postnatal period toward showed a strong preference for the untreated breast when
odor stimuli from the uterine environment, (2) amniotic given the same choice several days later (Fig. 2). This
and colostral odours initially have similar hedonic proper- developmental shift from an initial orientation to a prena-
ties, and (3) after repeated exposure to milk during the first tal odor followed by a relative preference for a postnatal
3–5 days, its odor is preferred over that of amniotic fluid. odor likely reflects differential exposure to those cues. At
Finally, a pattern of results similar to those outlined birth, amniotic fluid was more familiar than breast or
above was obtained in a study in which human infants colostrum/milk odors and therefore elicited preferential
responses. Over the first several days postpartum, howev-
er, babies had increasing reinforced contact with breast
odors but were no longer exposed to amniotic fluid (see
also Marlier et al., 1997, 1998a,b; Schaal, 1988b; Schaal
and Orgeur, 1992.).
Neonatal attraction to amniotic fluid may therefore be
explained by the continuity of odor information acquired
prenatally. But the means by which odorous compounds
not previously directly encountered (e.g., colostrum, milk,
exocrine secretions) acquire positive hedonic properties
that are evident within minutes or hours after birth remains
unknown. Amniotic fluid, colostrum, milk, and exocrine
secretions originate to at least some extent from the mater-
nal blood stream, and might therefore share a degree of
chemical resemblance reflecting maternal diet, genotype,
or metabolism. The aromas of specific food items ingested
by the pregnant and lactating mother are indeed known to
be transferred to her amniotic fluid and lacteal secretions.
Hepper (1988a) and Schaal et al. (1995b) observed prefer-
ential—or less aversive—responses to odors by rat pups or
lambs born to females whose diet contained those same
Figure 3 Mean relative duration (SEM) of head orientation in
five groups of breast-fed infants in a series of two-choice tests aromas during gestation. Moreover, food preferences of
pairing the odor of their own amniotic fluid (open bars) and the weanling rats can be traced back to the flavor composition
odor of their mother’s lacteal secretions (black bars). For age of their mother’s diet while she was nursing them (Capretta
groups 1–5, the number of infants is 9, 20, 12, 16, and 9, respec- and Rawls, 1974; Galef and Henderson, 1972; Mainardi
tively. *p  0.01. (Adapted from Marlier et al., 1997.) et al., 1989).
Olfaction and Social Behavior in Neonatal Mammals 317

The adaptive significance of transnatal olfactory con- to their own mother within the first day or two after birth.
tinuity was assessed further in rabbits. Pregnant females In contrast, individual recognition of mother and littermate
were fed flavored fodder and their pups were fostered at siblings may develop more gradually in altricial neonates
birth to females fed either the “familiar” maternal diet or that are confined to a nest until weaning (e.g., ground
a novel diet (Coureaud et al., 1998a,b). Those pups that squirrels).
experienced prenatal-postnatal chemosensory discontinu- Across taxonomic groups, social recognition tends to be
ity were less successful in obtaining colostrum than were mediated by the same sensory systems that are otherwise
agemates exposed to transnatal continuity. These findings most salient for communication. Thus, terrestrial mammals
are particularly significant when the natural history of the such as rodents, ungulates, and carnivores rely to a great
species is considered; female rabbits typically nurse their extent on olfactory phenotypes to distinguish between indi-
offspring only during a single brief period (3–5 min) viduals of their own species (reviewed by Halpin, 1986;
every 24 hours. Therefore, continuity of chemical signals Hepper, 1991). Although the sense of smell of primates is
before and after birth enhances the likelihood that the presumed to be less salient than vision and audition for the
pups will obtain the nutrients necessary for survival. regulation and coordination of social interactions, odor
From the data reviewed above, it appears that new- cues have been found to provide a sufficient basis for indi-
born mammals are able to detect olfactory similarity vidual recognition in a number of primate species (Colgan,
between chemicals present in their amniotic fluid and 1983; Halpin, 1986), including humans (discussed further
those externalized by the mother following parturition. below).
The chemosensory overlap between amniotic and lacteal
fluids is less well documented in our own species than in A. Neonates’ Discriminative Responses to Maternal
other mammals; however, there is evidence of transfer of Olfactory Signatures
the flavors carried in the pregnant and lactating human
mother’s diet into her amniotic fluid (Hauser et al., 1985; After initially establishing that neonates are attracted to
Mennella et al., 1995) and milk (e.g., Mennella and their mother’s breast odor in the absence of alternative
Beauchamp, 1996). olfactory cues, Macfarlane (1975) conducted a further
series of two-choice preference tests to determine
whether breastfeeding infants recognize their own
III. THE ROLE OF OLFACTORY CUES IN mother’s unique olfactory signature. The subject babies
INDIVIDUAL RECOGNITION were exposed simultaneously to a breast pad that had
been worn by their mother and a breast pad from another
Animals do not interact in an indiscriminate or random woman who was also nursing her own infant of the same
manner with other members of their own species. This is age. Six- and 10-day-old neonates spent significantly
readily apparent in the exclusive relationship that develops more time oriented towards their own mother’s breast
between mammalian infants and their nursing mothers and odor, thereby indicating that they were able to discrimi-
in the encounters between mates, or members of the same nate between the two odor stimuli and preferred that of
subgroup, that often differ noticeably from those involving their own mother. Odors of secretions collected from the
unfamiliar individuals. Following parturition, females of nipple and areola region of lactating females are nonirri-
many species quickly restrict their caregiving behavior to tating and of weak intensity, and therefore unlikely to
their own newborn offspring and reject or even attack alien stimulate the trigeminal nerve (Beauchamp et al., 1991;
young that approach them and attempt to suck. Young like- Doty, 1991; Schaal, 1988a). Additional data support the
wise tend to respond preferentially to their mother and conclusion that the infants in Macfarlane’s study dis-
littermates as compared to other adult females or age- played a positive hedonic response to their mother’s
mates. Based upon such discriminative social interactions, breast odor rather than avoiding the unfamiliar woman’s
it can be inferred that animals are capable of recognizing scent. Babies who had been breast-fed since birth orient-
individuals or distinct classes of conspecifics. The age at ed preferentially to an unfamiliar lactating woman’s
which mother and sibling recognition first becomes breast odor when paired with that same female’s axillary
evident varies across species and is correlated with their odor or a clean control pad (Porter et al., 1992). It will
natural history and the rate of development of sensory and also be recalled that bottle-fed infants similarly spent
locomotor capabilities of the young. For example, pre- more time turned towards a breast pad from a lactating
cocial young that live in groups composed of a large num- woman than in the direction of various alternative scents,
ber of females and their offspring (e.g., sheep and other including that of their familiar formula (Porter et al.,
ungulates) typically begin to respond in a selective manner 1991). Moreover, in Macfarlane’s original experiment,
318 Porter and Schaal

maternal breast pads elicited longer head orientation than known genotypes. Females derived from wild populations
did odorless control pads. can discriminate between odors of soiled bedding material
A subsequent experiment employing a variation of from males heterozygous for one of the recessive t alleles
Macfarlane’s olfactory discrimination apparatus found (/t) and the scent of males homozygous for the wild-type
evidence that breast-fed infants can discriminate their allele (/) (Lenington and Egid, 1985). Urinary and
mother’s individually unique breast odor at 3 days post- whole-body odors of inbred mice that are genetically iden-
partum (Schaal et al., 1980). Reduced movements of the tical except for genes of the major histocompatibility com-
head and arms were observed when their nose came into plex (MHC) can also be distinguished by conspecifics
contact with their mother’s breast pad compared to either (Yamaguchi et al., 1981; Yamazaki et al., 1999) as well as
a clean control pad or one that had been worn by an human subjects (Gilbert et al., 1986). In addition, (uniden-
unfamiliar lactating female. Russell (1976) likewise tified) genes from other autosome and sex chromosome
noted differences in the sucking and orienting responses loci are known to contribute to the individual odor types of
of 6-week-olds when their nursing mother’s breast pad mice and rats (Boyse et al., 1991; Schellinck et al., 1993).
was held under their nostrils, rather than a strange mother’s Given the polygenic contributions to olfactory signa-
pad or one moistened with cow’s milk. Infants at an tures, one would expect a positive correlation between the
even younger age (~24 hours) increased their rate of degree of genetic relatedness of individuals and the simi-
mouthing movements to a greater extent when exposed larity of their odor types. In accordance with this hypoth-
to the scent of their mother’s hospital gown (area that esis, sheep (Romeyer et al., 1993), Turkish hamsters (Heth
had been in contact with her breast/axillary region) than et al., 1999), and rats (Hepper, 1983) appear capable of
during trials with another mother’s gown (Sullivan and detecting olfactory resemblance between individual con-
Toubas, 1998). specifics that are closely related to one another (kin). A
In the above research on maternal-odor recognition, similar relationship between genotype and human body
babies consistently showed discriminative responses to odor is evident from several lines of research (Porter,
olfactory cues emanating from their own mother’s breasts. To 1999). Over 4 decades ago, Kalmus (1955) reported that
determine whether individually distinctive maternal odors highly trained tracking dogs did not distinguish between
are produced at other bodily sites, 2-week-old infants were the scents of identical twins when the samples were pre-
tested for their head-turning responses to their mother’s axil- sented successively rather than simultaneously. Related
lary odors (Cernoch and Porter, 1985). Gauze pads that had experiments have since determined that dogs could learn to
been taped in the armpit of their nursing mother before the respond differentially to the odors of dizygotic (fraternal
tests elicited longer directional head orientation than did axil- twins) fed identical diets and between monozygotic twins
lary pads from either nonparturient women or unfamiliar lac- whose diets were distinctly different (Hepper, 1988b). On
tating mothers. It therefore appears that the olfactory the other hand, when monozygotic twins had been eating
signatures that enable infants to recognize their own mother the same diet, the dogs showed no signs of discriminating
are not restricted solely to the nipple/areola region. between their scents.
Likewise, human subjects more accurately discriminated
B. Underlying Basis of Individual Olfactory between the (hand) odors of two unrelated individuals of
Signatures the same sex than the odors of identical twins or full sib-
lings (Wallace, 1977). Once again, odors of monozygotic
An individual’s characteristic body odor (“olfactory signa- twins were easier to discriminate when the paired stimulus
ture”) ultimately reflects a complex interaction between individuals were eating different diets. Moreover, closer
genetically mediated factors (e.g., metabolic and endocrine matches were found in the chromatography patterns of
processes, density and distribution of skin glands) and vari- sweat samples obtained from identical twins than in those
ous environmental influences (Schaal and Porter, 1991). It from nonkin pairs (Mc Cormick et al., 1995; Sommerville
is common knowledge that the consumption of strongly et al., 1990).
flavored or highly spiced substances may alter the scent of In a series of studies concerning olfactory recognition
the skin surface, faeces, urine, or breath. Micro-organisms of newborn infants, several parents commented that the
also play an important role in the production of bodily scent of their baby reminded them of other family mem-
scents; many secretions and excretions are odorless unless bers (i.e., the neonate’s other parent or an older sibling)
they are processed by resident microflora. (Porter et al., 1983, 1986). These anecdotal claims of
Perhaps the clearest evidence that minor genetic differ- detectable “familial odors” served as the basis for an
ences may result in discernible differences in olfactory experiment in which adult subjects attempted to match the
phenotypes is provided by research with house mice of body odors of mothers and their 3 to 8-year-old offspring
Olfaction and Social Behavior in Neonatal Mammals 319

(Porter et al., 1985). Stimulus mothers and children wore a 2 weeks postpartum (Cernoch and Porter, 1985). These
t-shirt for 3 consecutive nights; they were also given the results differ markedly from those of bottle-fed infants
same brand of soap to use during that period and of the same age, who spent no more time oriented to
instructed to avoid using deodorants, perfumes, or other their own mother’s axillary pad than to an axillary pad
scented toilet articles. During the tests, subjects sniffed a from either a nonparturient woman or an unfamiliar (non-
t-shirt that had been worn by one of the stimulus children lactating) mother. Thus, whereas breast-fed infants recog-
(standard) and then attempted to identify by scent alone nized their mother’s axillary odor, there was no indication
the shirt that had been worn by that child’s mother that the bottle-feeders did so. It follows, therefore, that
(included in an array of shirts worn by four different prenatal learning sensus stricto cannot account for 2-week-
women). The same subjects were given the opposite task in old’s discriminative responses to their nursing mothers’
the other test trial, i.e., asked to select the child’s shirt olfactory signatures. That is, if the fetus becomes familiar
whose odor most closely resembled that of a standard with the unique odor of its mother and subsequently
mother. In both series of tests, odors of mothers and chil- remembers that chemical cue at 2 weeks postpartum, bot-
dren were correctly matched significantly more often than tle-feeders should also show evidence of recognizing their
expected by chance alone. mother’s axillary odor. Nevertheless, a possible influence
While these data indicate that mothers and offspring of prenatal experience on the development of maternal-
bear a detectable olfactory resemblance to one another, odor recognition cannot be completely excluded; fetal
their overlapping odors could reflect shared environmental familiarization could facilitate postnatal learning or per-
factors (e.g., similar diets, identical ambient household haps even provide a sufficient basis for neonates to dis-
odors) rather than genetic determinants. Therefore, to criminate their mother’s olfactory signatures within a brief
assess these alternative hypotheses, additional odor-matching period after birth.
tests were conducted with t-shirts worn by individuals who Although the differing reactions of the bottle- and breast-
were not genetically related (husbands and wives) but lived fed infants to maternal scents cannot be fully explained at
in the same house and shared meals. Because the accuracy present, they most likely reflect differential patterns of
rate of matching the odors of husbands and wives did not mother-infant interactions in these two feeding categories.
differ from chance expectations, environmental similarity When sucking at the breast, an infant’s nostrils are kept in
was not sufficient for the spouses to acquire similar body close physical proximity to the mother’s skin surface and
odors. Taken together, these data suggest that the perceived concomitant olfactory cues for an extended period of time.
olfactory resemblance of mothers and their offspring was On the other hand, because bottle-feeders do not necessarily
mediated at least partially by the shared portion of their experience the same degree of routine exposure to the moth-
genomes. er’s bare flesh, they have less opportunity to become famil-
Little is known regarding the identity of specific gene loci iarized with her characteristic odor phenotype. Thus, one
that may affect human olfactory signatures. Nonetheless, as would expect maternal odor recognition to develop more
discussed above for mice, preliminary data indicate that rapidly in breastfeeders. A similar lack of exposure and
human body and urinary odors are also influenced by the opportunity for learning could account for (2-week-old)
MHC and both human and rat subjects are capable of detect- breast-fed infants’ failure to respond discriminatively to
ing olfactory resemblance among adults with similar MHCs axillary odors from their father (Cernoch and Porter, 1985).
(Eggert et al., 1999a; Wedekind and Furi, 1997). Moreover, a Experiments in which mothers wore artificial scents pro-
number of genetically based metabolic disorders, including vide further evidence of learned recognition and hedonic
phenylketonuria and sweaty feet syndrome, are characterized preference for odors associated with breastfeeding. At 1
by strong odors that can be valuable diagnostic aids (Cone, and 2 weeks postpartum, infants responded preferentially
1968; Liddell, 1976). to the familar perfume scent that their mothers had applied
to their breasts before each nursing bout (Schleidt and
C. Ontogenetic Mechanisms of Individual Olfactory Genzel, 1990). In a similar manner, exposure to a pad pre-
Recognition viously taped to their mother’s neck, and bearing an “obvi-
ous” scent of her perfume, resulted in reduced motor
Converging lines of empirical evidence indicate that activity by neonates (Schaal, 1986).
infants develop the ability to recognize their mother’s Research with rodents demonstrates that specific odorants
olfactory signature through a process of familiarization may acquire a positive hedonic value during early infancy as
and learning. Earlier in this chapter, we remarked that a function of mere exposure (Caza and Spear, 1984; Porter
breast-fed infants discriminate between axillary odors and Etscorn, 1974). These animal data served as the impetus
from their own mother versus another lactating woman at for an analogous study of the effects of early odor exposure
320 Porter and Schaal

on newborn human infants (Balogh and Porter, 1986). On the including social interactions among adults (see also
day of birth, a pad treated with the essential oil of ginger or Chapter 17). An early example involved treating rat pups
cherry was taped inside each infant’s cot and remained there and their mothers with cologne for the first 30-days post-
throughout a familiarization period lasting approximately 23 partum (Marr and Gardner, 1965). In adult breeding tests,
hours. During that time there was no readily identifiable con- rats from the early cologne-exposure group were less sex-
ventional reinforcement associated with the odor exposure. ually responsive than controls when paired with a normal-
Following the familiarization period, the scented pad was odor individual of the opposite sex. Similar results were
removed from the cot and all babies were tested (~45 min- obtained in a series of experiments by Mainardi and col-
utes later) for their head-turning reponses to a cherry versus leagues (1965); as adults, female mice that had been raised
a ginger odor pad. Thus, in each instance one of these odors by perfumed parents preferred males bearing that same
was “familiar” (the one to which the infant had been previ- scent, while control females spent more time with untreat-
ously exposed) and the other novel. A reliable preference for ed males. In a more recent experiment, male rat pups
the familiar odor was displayed by female infants in this test, sucked from females that had lemon scent (citral) applied
but males demonstrated a marked right-turning bias that may directly to their nipples and vagina (Fillion and Blass,
have overridden any odor-mediated response. In later exper- 1986). When sexually mature, these males ejaculated more
iments using a similar odor-exposure paradigm, however, quickly when paired with a receptive female treated with
both male and female infants oriented preferentially in the the familiar lemon odor as compared to tests with an
direction of a familiar exposure odor (Davis and Porter, unscented partner. Social experience in the nest has been
1991). found to have enduring effects on the behavioral respons-
Since mere exposure is sufficient for babies to recognize es of golden hamsters to conspecific odors (Todrank et al.,
subsequently an artificial odorant, it is likely that this same 1999). Adult males demonstrated that they remembered
process plays a role in the development of breast-fed the odors of their individual littermates following a 9-
neonates’ recognition of the maternal olfactory signature. month separation period beginning at weaning.
Furthermore, in the nursing context, the characteristic scent of Fostering of newborn young onto substitute parents is a
the mother is associated with reinforcers such as milk intake, common technique for assessing the role of the early rear-
physical contact, and warmth, which should enhance the ing environment in the development of social preferences.
learning process. Indeed, one-day-old infants more readily Rodents that are reared by females of a different species
developed discriminative behavioral responsiveness to an arti- often respond more positively to natural odors from their
ficial odorant when it was paired with reinforcing tactile stim- foster species than do control animals (McCarty and
ulation (Sullivan et al., 1991). Colostrum ingestion may be a Southwick, 1977; Porter et al., 1977). In some instances,
particularly reinforcing component of breastfeeding. In cross-species fostering results in heightened adult prefer-
lambs, gastric infusion of colostrum is as effective as a normal ences for the foster species’ odor along with reduced
sucking bout in mediating the development of recognition and responsiveness to conspecific chemical cues (Huck and
preference for their mother (Goursaud and Nowak, 1999). Banks, 1980; Quadagno and Banks, 1970). Mice and rats
During the first hours following birth, babies appear to be discriminate between the odors of conspecifics that are
physiologically prepared to learn rapidly their mother’s olfac- genetically identical except for difference in the major
tory signature (Porter and Winberg, 1999). The locus ceruleus histocompatibility complex (MHC) (Brown et al., 1987;
has been reported to be highly active during the perinatal Eggert et al., 1999b). In choice experiments, male mice of
period (Lagercrantz and Slotkin, 1986), and plasma levels of at least some inbred strains tend to prefer females whose
norepinephrine (NE) are 20–30 times greater within the first MHC differs from their own (Yamazaki et al., 1976). Such
few postnatal hours than afterwards (Lagercrantz, 1996). NE preferences for nonself MHC appear to result from early
and associated arousal of the locus ceruleus have recently familial (olfactory) “imprinting.” Males that had been
been implicated in olfactory learning in nonhuman mammals raised until weaning by foster parents with a different
(Brennan et al., 1990; Kendrick et al., 1992; Leon, 1992). MHC mated preferentially with females that shared their
own MHC rather than that of their parents (Yamazaki et al.,
1988). These males therefore avoided mating with females
IV. INFLUENCE OF EARLY OLFACTORY whose olfactory phenotype resembled the familiar foster
EXPERIENCE ON ADULT SOCIAL parent odor. However, a somewhat different pattern of
PREFERENCES results was obtained in subsequent experiments with addi-
tional inbred strains of mice; the effect of the rearing envi-
Numerous studies have documented long-term effects of ronment on MHC-based mating preferences was greater
infantile olfactory experience on the behavior of mammals, for females than males (Arcaro and Eklund, 1999). Early
Olfaction and Social Behavior in Neonatal Mammals 321

familial odor imprinting also appears to account for the The extent to which early (pre- or postnatal) exposure
avoidance of males with the foster parent MHC genotype to odors per se may continue to affect subsequent hedonic
that is shown by wild-derived female mice maintained responses to chemical stimuli by human adults—or even
under semi-natural conditions (Penn and Potts, 1998). older children—has yet to be elucidated. In practice, this
As discussed above for newborn infants, olfactory sig- question would be very difficult to assess since it would
nals have also been implicated in social discrimination and be necessary to restrict pretest experience with particular
preferences by older humans. Recognition of individuals scents to a relatively brief period of perinatal develop-
by their odor signature is well documented in children and ment. This type of long-term longitudinal research design
adults. Despite considerable intersubject variability, tests might be possible with laboratory animals (see above), but
of self-identification by olfactory cues have generally it is less appropriate for human subjects. Despite these
yielded positive results (Hold and Schleidt, 1977; Lord and practical limitations, recent experiments provide some ini-
Kasprzak, 1989; Mallet and Schaal, 1998; Russell, 1976). tial insight into the durability of the memory of perinatal
Beginning within hours after delivery, mothers reliably chemosensory experience. Infants show behavioral evi-
identify the odor of their own neonate when asked to sniff dence of remembering odors after a 2-week interval.
the soiled t-shirts or heads of an array of stimulus babies Within the first 2–3 days postpartum, babies were exposed
(Kaitz et al., 1987; Porter et al., 1983; Russell et al., 1983; to an artifical odorant (cherry or ginger) for approximately
Schaal et al., 1980). Moreover, mothers and fathers distin- 22 hours (Davis and Porter, 1991). The odorant was
guish between the distinctive olfactory signatures of their removed at the end of the exposure session, and subjects
two (full sibling) offspring, and children (3–8 years old) had no further contact with that scent until tests conducted
accurately identified their sibling’s t-shirt by odor cues 2 weeks later. At that time, the infants spent reliably
alone (Porter and Moore, 1981). Schaal et al., (1980) more time oriented towards a pad treated with the expo-
reported that a significant majority of 45- to 58-month-old sure odor than in the direction of a novel scent, thereby
children preferred the odor of a t-shirt that had been worn indicating that they remembered the familiar odorant.
by their own mother over that of an unfamiliar woman. Contrary to these data, however, Schleidt and Genzel
Overall, adults of both sexes did better than expected by (1990) reported that 4-week-old infants did not respond
chance in tests for their ability to recognize the body odor discriminatively to a perfume (rose oil) that their mothers
of their sexual partner (Hold and Schleidt, 1977). had worn on their breasts during the first 2 weeks after
At present, there is no clear evidence that human delivery—even though strong preferences had been exhib-
males or females produce odor signals that function as ited at 1 and 2 weeks. These contradictory results could
general sexual attractants in a manner analogous to what reflect procedural differences. Schleidt and Genzel used a
is seen in a number of other mammals. Nonetheless, par- conditioning paradigm in which the mother’s perfume
ticular odors (e.g., perfumes, scented soaps, the charac- was initially associated with breastfeeding. Thus, the sub-
teristic body odor of individuals) may be arousing or sequent 2 weeks when mothers no longer perfumed them-
attractive because of their association with a sexual con- selves may have served as an extinction period resulting in
text (Kirk-Smith and Booth, 1987). Several research reduced responsiveness to the training odor. In contrast,
teams have been investigating hedonic responses to the odorant in the mere-exposure experiment was not
MHC-correlated odors and their possible influence on associated with any obvious reinforcer, nor was there any
mating preferences in our own species. Female extinction phase. Other methodological differences were
university students rated the body odors of MHC-dis- the type of odor stimuli, the length of pretest exposure, the
similar men more positively than the odors of men age at testing, and the age during the 2-week interval
whose MHC was similar to their own and also claimed without odor exposure.
that the former scents reminded them more of their own There are few reports concerning hedonic responses to
mate (Wedekind et al., 1995). Although it is not known odors by infants older than 2–3 weeks. This is due, at least
whether humans indeed select sexual partners based in part, to procedural difficulties that arise when attempting
upon MHC-based odors, preferences for mates whose to assess olfactory perception and preferences at this age.
MHC differs from one’s own would presumably be adap- The head-orientation test described above is less adequate as
tive. The rate of spontaneous abortions has been reported babies become more active and their motor capacities
to be positively correlated with MHC homozygosity increase. Physiological measures and patterns of nonnutri-
between mates (Beer et al., 1981). Furthermore, because tive sucking have been used to assess early olfactory sensi-
of their MHC polymorphism, offspring of parents with tivity and discrimination, but their utility for investigating
dissimilar MHCs might be able to resist a wider range of odor hedonics has not been clearly demonstrated
pathogens (Apanius et al., 1997). (Beauchamp et al., 1991). Within the first months after birth,
322 Porter and Schaal

babies begin to explore and manipulate toys, and recent Beauchamp, G. K., Cowart, B., and Schmidt, H. J. (1991).
research indicates that this is a fruitful context for studying Development of chemosensory sensitivity and preferences. In
the influence of odors on behavioral responses (Schmidt and Smell and Taste in Health and Disease, T. V. Getchell, R. L. Doty,
Beauchamp, 1989). A sample of 6- to 13-month-old infants L. M. Bartoshuk, and J. B. Snow, Jr. (Eds.). Raven Press, New
York, pp. 405–416.
reacted differently to toys scented with ethanol or vanilla as
Beer, A. E., Quebbeman, J. F., Ayers, J. W. T., and Haines, R. F.
compared to an unscented toy (Mennella and Beauchamp,
(1981). Major histocompatibility complex antigens, maternal
1998). Interestingly, babies’ responses to the scented toys and paternal immune responses, and chronic habitual abor-
were correlated with maternal consumption of products with tions in humans. Am. J. Obstet. Gynecol. 141:987–997.
similar odors. Infants who lived with one or two alcoholic Bilko, A., Altbacker, V., and Hudson, R. (1995). Transmission of
parents spent more time mouthing an alcohol-scented toy food preference in the rabbit: the means of information trans-
than did babys of nonalcoholic parents, and infants whose fer. Physiol. Behav. 56:907–912.
mothers frequently ate vanilla-flavored food spent more Blass, E. M., and Teicher, M. H. (1980). Suckling. Science
time looking at a toy treated with that odorant. It is likely 210:15–22.
that the infants became familiar with vanilla or alcohol Blass, E. M., Teicher, M. H., Cramer, C. P., Bruno, J. P., and
scents/flavors in the home, possibly in their mothers’ milk. Hall, W. G. (1977). Olfactory, thermal and tactile controls of
suckling in preauditory and previsual rats. J. Comp. Physiol.
However, the authors of this study were careful to point out
Psychol. 91:1248–1260.
that clear causal effects cannot be deduced from their correl-
Bossy, J. (1980). Development of olfactory and related structures
ational data and that any long-term consequences of such in staged human embryos. Anat. Embryol. 161:225–236.
early flavor experience remain unknown. Boyse, E. A., Beauchamp, G. K., Yamazaki, K., and Bard, J.
(1991). Genetic components of kin recognition in mammals.
In kin Recognition, P. G. Hepper (Ed.). Cambridge University
REFERENCES Press, Cambridge, pp. 148–161.
Breen, M. F., and Leshner, A. I. (1977). Maternal pheromone: a
Abramovich, D. R. (1981). Interrelation of fetus and amniotic demonstration of its existence in the mouse (Mus musculus).
fluid. In Obstetrics and Gynecology Annual, R. D. Wynn Physiol. Behav. 18:527–529.
(Ed.). Appleton-Century-Crofts, New York. Brennan, P., Kaba, H., and Keverne, E. B. (;1990). Olfactory
Alberts, J. R. (1978a). Huddling by rat pups: multisensory con- recognition: a simple memory system. Science 250:
trol of contact behavior. J. Comp. Physiol. Psychol. 1223–1226.
92:220–230. Brown, J. W. (1974). Prenatal development of the human chief
Alberts, J. R. (1978b). Huddling by rat pups: group behavioral sensory trigeminal nucleus. J. Comp. Neurol. 156:307–336.
mechanisms of temperature regulation and energy conserva- Brown, J. W. (1987). The nervus terminalis in insectivorous bats
tion. J. Comp. Physiol. Psychol. 92:231–245. and notes on its presence during human ontogeny. Ann. NY
Alberts, J. R., and Brunjes, P. C. (1978). Ontogeny of thermal and Acad. Sc. 519:184–200.
olfactory determinants of huddling in rats. J. Comp. Physiol. Brown, R. E., Singh, P. B., and Roser, B. (1987). The major
Psychol. 92:897–906. histocompatibility complex and the chemosensory recognition
Alberts, J. R., and May, B. (1980). Ontogeny of olfaction: devel- of individuality in rats. Physiol. Behav. 40:65–73.
opment of the rat’s sensitivity to urine and amyl acetate. Burghard, G. M. (1971). Chemical-cue preferences of newborn
Physiol. Behav. 24:965–970. snakes: influence of prenatal maternal experience. Science
Allen, W. F. (1929). Effect on respiration, blood pressure, and 171:921–923.
carotid pulse of various inhaled and insufflated vapors when Capretta, P. J., and Rawls, L. H. (1974). Establishment of a flavor
stimulating one cranial nerve and various combinations of preference in rats: Importance of nursing and weaning experi-
cranial nerves. Am. J. Physiol. 87:319–325. ence. J. Comp. Physiol. Psychol. 86:670–673.
Allin, J. T., and Banks, E. M. (1972). Functional aspects of ultra- Caza, P. A., and Spear, N. E. (1984). Short-term exposure to an
sound production by infant albino rats, Rattus norvegicus. odor increases its subsequent preference in preweanling rats :
Anim. Behav. 20:175–185. a descriptive profile of the phenomenon. Dev. Psychobiol.
Apanius, V., Penn, D., Slev, P. R., Ruff, L. R., and Potts, W. K. 17:407–422.
(1997). The nature of selection on the major histocompatibil- Cernoch, J. M., and Porter, R. H. (1985). Recognition of mater-
ity complex. Crit. Rev. Immunol. 17:179–224. nal axillary odors by infants. Child Dev. 56:1593–1598.
Arcaro, K. F., and Eklund, A. (1999). A review of MHC-based Chotro M. G., and Molina J. C. (1990). Acute ethanol contam-
mate preferences and fostering experiments in two congenic ination of the amniotic fluid during gestational day 21: post-
strains of mice. Genetica 104:241–244. natal changes in alcohol responsiveness in rats. Dev.
Balogh, R. D., and Porter, R. H. (1986). Olfactory preferences Psychobiol. 23:535–547.
resulting from mere exposure in human neonates. Infant Christensson, K., Cabrera, T., Christensson, E., Uvnas-Moberg, K.,
Behav. Dev. 9:395–401. and Winberg, J. (1995). Separation distress call in the human
Olfaction and Social Behavior in Neonatal Mammals 323

neonate in the absence of maternal body contact. Acta determinant of the feeding preferences of weaning rat pups.
Paediatr. 84:468–473. J. Comp. Physiol. Psychol. 78:213–219.
Colgan, P. (1983). Comparative Social Recognition. John Wiley Garbini, A. (1896). Evoluzione del senso olfattivo nella infanzia.
& Sons, New York. Arch. Antropol. Etnol. (Firenze) 26:239–286.
Cone, T. E., Jr. (1968). Diagnosis and treatment: some diseases, Gilbert, A. N., Yamazaki, K., Beauchamp, G. K., and Thomal, L.
syndromes, and conditions associated with an unusual odor. (1986). Olfactory discrimination of mouse strains (Mus mus-
Pediatrics 41:993–995. culus) and major histocompatibility types by humans (Homo
Cooper, A. J., and Cowley, J. J. (1976). Mother-infant interaction sapiens). J. Comp. Psychol. 100:262–265.
in mice bulbectomized early in life. Physiol. Behav. Goursaud, A. P., and Nowak, R. (1999). Colostrum mediates the
16:453–459. development of mother preference by newborn lambs.
Coppola, D. M., and Millar, L. C. (1997). Olfaction in utero: Physiol. Behav. 67:49–56.
behavioral studies of the mouse fetus. Behav. Proc. 39:53–68. Halpin, Z. T. (1986). Individual odors among mammals: origins
Coureaud, G., and Schaal, B. (2000). Attraction of newborn rab- and functions. Adv. Study Behav. 16:39–70.
bits to abdominal odors of adult conspecifics differing in sex Hauser, G. J., Chitayat, D., Berns, L., Braver, D., and
and physiological state. Dev. Psychobiol. 36:271–281. Muhlhauser, B., (1985). Peculiar odours in newborns and
Coureaud, G., Schaal, B., Langlois, D., and Perrier, G. (2001). maternal prenatal ingestion of spicy foods. Eur. J. Pediatr.
Orientation response of newborn rabbits to odours emitted by 144:403.
lactating females: Relative effectiveness of surface and milk Hepper, P. G. (1983). Sibling recognition in the rat. Anim. Behav.
cues. Anim. Behav. 61:153–162. 31:1177–1191.
Coureaud, G., Schaal, B., Orgeur, P., and Hudson, R. (1998a). A Hepper, P. G. (1987). The amniotic fluid: an important priming
test of transnatal olfactory continuity in the rabbit. Dev. role in kin recognition. Anim. Behav. 35:1343–1346.
Psychobiol. 33:370. Hepper P. G. (1988a). Adaptive fetal learning: prenatal exposure
Coureaud, G., Schaal, B., Orgeur, P., Hudson, R., Lebas, F., and to garlic affects postnatal preferences. Anim. Behav.
Coudert, P. (1998b). Perinatal odour disruption impairs 36:935–936.
neonatal milk intake in the rabbit. Adv. Ethol. 32:102. Hepper, P. G. (1988b). The discrimination of human odour by the
Darwin, C. (1877). A biographical sketch of an infant. Mind dog. Perception 17:549–554.
7:285–294. Hepper, P. G. (Ed.) (1991). Kin Recognition. Cambridge
Davis, L. B., and Porter, R. H. (1991). Persistent effects of early University Press, Cambridge.
odor exposure on human neonates. Chem. Senses 16:169–174. Hepper, P. G., and Waldman, B. (1992). Embryonic olfactory
De Perceval, G. D., and Lallemand, S. (1980). L’Art learning in frogs. Quart. J. Exp. Psychol. B. 44:179–197.
d’Accomeder les Bébés. Editions du Seuil, Paris. Heth, G., Todrank, J., and Johnston, R. E. (1999). Similarity in
Doty, R. L. (1991). Olfactory function in neonates. In The Human the qualities of individual odors among kin and species in
Sense of Smell, D. G. Laing, R. L. Doty and W. Breipohl Turkish (Mesocricetus brandti) and Golden (Mesocricetus
(Eds.). Springer-Verlag, Berlin, pp. 155–165. auratus) hamsters. J. Comp. Psychol. 113:321–326.
Doty, R. L. (1995). Intranasal trigeminal chemoreception. In Hofer, M. A. (1975). Studies on how early maternal separation
Handbook of Olfaction and Gustation, R. L. Doty (Ed.). produces behavioral change in young rats. Psychosom. Med.
Marcel Dekker, New York, pp. 821–833. 37:245–264.
Doty, R. L. (2001). Olfaction. Ann. Rev. Psychol. 52:423–452. Hofer, M. A. (1976). Olfactory denervation: its biological and
Duenholter J. H., and Pritchard J. A. (1976). Fetal respiration: behavioral effects in infant rats. J. Comp. Physiol. Psychol.
quantitative measurements of amniotic fluid inspired near 90:829–838.
term by human and rhesus fetuses. Am. J. Obstet. Gynecol. Hofer, M. A., and Shair, H. (1980). Sensory processes in control
125:306–309. of isolation-induced ultrasonic vocalization by 2-week-old
Eggert, F., Luszyk, D., Haberkorn, K., Wobst, B., Vostrowsky, O., rats. J. Comp. Physiol. Psychol. 94:271–279.
Westphal, E., Bestmann, H. J., Muller-Ruchholtz, W., and Hofer, M. A., Shair, H., and Singh, P. (1976). Evidence that
Ferstl, R. (1999a). The major histocompatibility complex and maternal ventral skin substances promote suckling in infant
the chemosensory signalling of individuality in humans. rats. Physiol. Behav. 17:131–136.
Genetica 104:265–273. Hold, B., and Schleidt, M. (1977). The importance of human
Eggert, F., Muller-Ruchholtz, W., and Ferstl, R. (1999b). odour in non-verbal communication. Z. Tierpsychol.
Olfactory cues associated with the major histocompatibility 43:225–238.
complex. Genetica 104:191–197. Holmes, W. G. (1991). Parent-offspring recognition in mammals:
Fildes, V. A. (1986). Breasts, Bottles and Babies. Edinburgh a proximate and ultimate perspective. In Mammalian
University Press, Edinburgh. Parenting, N. A. Krasnegor and R.S. Bridges (Eds.). Oxford
Fillion, T. J., and Blass, E. M. (1986). Infantile experience with University Press, New York, pp. 441–460.
suckling odors determines adult sexual behavior in male rats. Huck, U. W., and Banks, E. M. (1980). The effects of cross-
Science 231:729–731. fostering on the behaviour of two species of North American
Galef, B. G., and Henderson, P. W. (1972). Mother’s milk: a lemmings, Dicrostonyx groenlandicus and Lemmus trimu-
324 Porter and Schaal

cronatus: I. Olfactory preferences. Anim. Behav. Mainardi, D., Marasan, M., and Pasquali, A. (1965). Causation
28:1046–1052. of sexual preferences of the house mouse. The behaviour of
Hudson, R., and Distel, H. (1982). The pattern of behaviour of mice reared by parents whose odour was artificially altered.
rabbit pups in the nest. Behaviour 79:255–271. Atti Soc. Italiana di Scienze Naturali Museo Civico Milano
Hudson, R., and Distel, H. (1983). Nipple location by newborn 104:325–338.
rabbits: Behavioural evidence for pheromonal guidance. Mainardi, M., Poli, M., and Valsecchi, P. (1989). Ontogeny of
Behaviour 85:260–275. dietary selection in weanling mice: effects of early experience
Humphrey, T. (1940). The development of the olfactory and the and mother’s milk. Biol. Behav. 14:185–194.
accessory olfactory formations in human embryos and Makin, J. W., and Porter, R. H. (1989). Attractiveness of lactating
fetuses. J. Comp. Neurol. 73:431–468. females’ breast odors to neonates. Child Dev. 60:803–810.
Kaitz, M., Good, A., Rokem, A. M., and Eidelman, A. I. (1987). Mallet, P., and Schaal, B. (1998). Rating and recognition of
Mothers’ recognition of their newborns by olfactory cues. peers’ personal odours in nine-year-old children: an
Dev. Psychobiol. 20:587–591. exploratory study. J. Gen. Psychol. 125:47–64.
Kalmus, H. (1955). The discrimination by the nose of the dog of Marlier, L., Schaal, B., and Soussignan, R. (1997). Orientation
individual human odours and in particular of the odours of responses to biological odours in the human newborn. Initial
twins. Br. J. Anim. Behav. 5:25–31. pattern and postnatal plasticity. C. Roy. Acad. Sci. (Paris) Life
Keil, W., von Stralendorff, F., and Hudson, R. (1990). A behav- Sci. 320:999–1005.
ioral bioassay for analysis of rabbit nipple-search pheromone. Marlier, L., Schaal, B., and Soussignan, R. (1998a). Neonatal
Physiol. Behav. 47:525–529. responsiveness to the odor of amniotic and lacteal fluids: A
Kendrick, K. M., Levy, F., and Keverne, E. B. (1992). Changes in test of perinatal chemosensory continuity. Child Dev.
the sensory processing of olfactory signals induced by birth in 69:611–623.
sheep. Science 256:833–836. Marlier, L., Schaal, B., and Soussignan, R. (1998b). Bottle-fed
Kirk-Smith, M. D., and Booth, D. A. (1987). Chemoreception in neonates prefer an odor experienced in utero to an odor
human behaviour: experimental analysis of the social effects experienced postnatally in the feeding context. Dev.
of fragrances. Chem. Senses 12:159–166. Psychobiol. 33:133–145.
Kroner, T. (1882). Über Sinnesempfindungen des Neugeborenen. Marr, J. N., and Gardner, L. E., Jr. (1965). Early olfactory experi-
Breslauer ärztl. Ztschr. 4:37–58. ence and later social behavior in the rat: preference, sexual
Lagercrantz, H. (1996). Stress, arousal and gene activation at responsiveness, and care of young. J. Genetic Psychol.
birth. News Physiol. Sci. 11:214–218.
107:167–174.
Lagercrantz, H., and Slotkin, T. A. (1986). The stress of being
McCarty, R., and Southwick, C. H. (1977). Cross-species foster-
born. Sci. Am. (April): 92–102.
ing: effects on the olfactory preferences of Onychomys tor-
Lenington, S., and Egid, K. (1985). Female discrimination of
ridus and Peromyscus leucopus. Behav. Biol. 19:255–260.
male odors correlated with male genotype at the T-locus: a
McCormick, J. P., Sommerville, B. A., Eggert, F., Wobst, B., and
response to T-locus or H-2 locus variability. Behav. Genet.
Broom, D. M. (1995). In Chemical Signals in Vertebrates, VII,
15:53–67.
R. Apfelbach and D. Muller-Schwarze (Eds.). Pergamon
Leon, M. (1992). Neuroethology of olfactory preference devel-
Press, Oxford, pp. 539–542.
opment. J. Neurobiol. 23:1557–1573.
Mennella, J. A. (1995). Mother’s milk: a medium for early flavor
Leon, M., and Moltz, H. (1971). Maternal pheromone: discrim-
experiences. J. Hum. Lact. 11:39–45.
ination by pre-weanling albino rats. Physiol. Behav. 7:
Mennella, J. A., and Beauchamp, G. K. (1991). Maternal diet
265–267.
Leon, M., and Moltz, H. (1972). The development of the alters the sensory qualities of human milk and the nursling’s
pheromonal bond in the albino rat. Physiol. Behav. behavior. Pediatrics 88:737–744.
8:683–686. Mennella, J. A., and Beauchamp, G. K. (1996). The human
Liddell, K. (1976). Smell as a diagnostic marker. Postgrad. Med. J. infants’ response to vanilla flavors in mother’s milk and for-
52:136–138. mula. Infant Behav. Dev. 19:13–19.
Lipsitt, L. P., Engen, T., and Kaie, H. (1963). Developmental Mennella, J. A., and Beauchamp, G. K. (1998). Infants’ explo-
changes in the olfactory threshold of the neonate. Child Dev. ration of scented toys: effects of prior experience. Chem.
34:371–376. Senses 23:11–17.
Logvinenko, A. V. (1990). Recording of fetal respiratory move- Mennella, J. A., Johnson, A., and Beauchamp, G. K. (1995).
ments based on studying the shifting of the amniotic fluid in Garlic ingestion by pregnant women alters the odor of amni-
the trachea by the Doppler effect. Akush Ginekol. (Moskow) otic fluid. Chem. Senses 20:207–209.
7:38–41. Meza, C. V., Powell, N. J., and Covington, C. (1998). The influ-
Lord, T., and Kazprzak, M. (1989). Identification of self through ence of olfactory intervention on non-nutritive sucking skills
olfaction. Percept. Mot. Skills 69:219–224. in a premature infant. Occup. Ther. J. Res. 18:71–83.
Macfarlane, A. (1975). Olfaction in the development of social Moran, D. T., Monti-Bloch, L., Stensaas, L. J., and Berliner, D. L.
preferences in the human neonate. In Parent-Infant (1995). Structure and function of the human vomeronasal
Interaction, R. Porter and M. O’Connor (Eds.). (Symp. 33). organ. In Handbook of Olfaction and Taste, R. L. Doty (Ed.).
Elsevier, New York, pp. 103–113. Marcel Dekker, New York, pp. 793–820.
Olfaction and Social Behavior in Neonatal Mammals 325

Morrow-Tesch, J., and McGlone, J. J. (1990). Sources of mater- Porter, R. H., Cernoch, J. M., and McLaughlin, F. J. (1983).
nal odors and the development of odor preferences in baby Maternal recognition of neonates through olfactory cues.
pigs. J. Anim. Sci. 68:3563–3571. Physiol. Behav. 30:151–154.
Nakashima, T., Kimmelman, C. P., and Snow, J. B (1984). Porter, R. H., Cernoch, J. M., and Balogh, R. D. (1985). Odor sig-
Structure of human fetal and adult olfactory neuro-epithelium. natures and kin recognition. Physiol. Behav. 34:445–448.
Arch. Otolaryngol. 110:641–646. Porter, R. H., Balogh, R. D., Cernoch, J. M., and Franchi, C.
Noirot, E. (1966). Ultrasons et comportements maternels chez les (1986). Recognition of kin through characteristic body odors.
petits rongeurs. Ann. Soc. Roy. Zool. Belg. 95:47–56. Chem. Senses 11:389–395.
Nolte, D. L., Provenza, F. D., Callan, R., and Panter, K. E. (1995). Porter, R. H., Makin, J. W., Davis, L. B., and Christensen, K. M.
Garlic in the ovine fetal environment. Physiol. Behav. (1991). An assessment of the salient olfactory environment of
52:1091–1093. formula-fed infants. Physiol. Behav. 50:907–911.
Oelschläger, H. A., Buhl E. A., and Dann J. F. (1987). Porter, R. H., Makin, J. W., Davis, L. B., and Christensen, K. M.
Development of the nervus terminalis in mammals including
(1992). Breast-fed infants respond to olfactory cues from their
toothed whales and humans. In The terminal nerve (nervus ter-
own mother and unfamiliar lactating females. Infant Behav.
minalis). Structure, Function, Evolution, L. S. Demski and
Dev. 15:85–93.
M. Schwanzel-Fukuda (Eds.). Ann. NY Acad. Sci. NY Academy
Pourtier, L., and Sicard, G. (1990). Comparison of the sensitivity
of Sciences, NY, pp. 447–464.
of C57BL/6J and AKR/J mice to air-borne molecules of iso-
Ortmann, R. (1989). Über Sinneszellen am fetalen vomeronasalen
Organ des Menschen. Hals Nase Ohren 37: 191–197. valeric acid and amylacetate. Behav. Genet. 20:499–509.
Oswalt, G. L., and Meier, G. W. (1975). Olfactory, thermal, and Preyer, W. (1885). Die Seele des Kindes (French translation).
tactual influences on infantile ultrasonic vocalization in rats. Alcan, Paris.
Dev. Psychobiol. 8:129–135. Pyatkina, G. A. (1982). Development of the olfactory epithelium
Parfet, K. A. R., and Gonyou, H. W. (1991). Attraction of new- in man. Zschr. Mikrosk. Anat. Forsch. 96:361–372.
born piglets to auditory, visual, olfactory and tactile stimuli. Quadagno, D. M., and Banks, E. M. (1970). The effect of recip-
J. Anim. Sci. 69:125–133. rocal cross fostering on the behaviour of two species of
Pedersen, P. E., and Blass, E. M. (1982). Prenatal and postnatal rodents, Mus musculus and Baiomys taylori ater. Anim. Behav.
determinants of the 1st suckling episode in albino rats. Dev. 18:379–390.
Psychobiol. 15:349–355. Randall, P. K., and Campbell, B. A. (1976). Ontogeny of behav-
Penn, D., and Potts, W. (1998). How do major histocompatibility ioral arousal in rats: effects of maternal and sibling presence.
complex genes influence odor and mating preferences? Adv. J. Comp. Physiol. Psychol. 90:453–459.
Immunol. 69:411–436. Risser, J. M., and Slotnick, B. M. (1987). Nipple attachment and
Pihet, S., Schaal, B., Bullinger, A., and Mellier, D., (1996). An survival in neonatal olfactory bulbectomized rats. Physiol.
investigation of olfactory responsiveness in premature new- Behav. 40:545–549.
borns. Infant Behav. Dev. (ICIS issue): 676. Romeyer, A., Porter, R. H., Poindron, P., Orgeur, P., Chesne, P.,
Pihet, S., Mellier, D., Bullinger, A., and Schaal, B., (1997). and Poulin, N. (1993). Recognition of dizygotic and monozy-
Réponses comportementales aux odeurs chez le nouveau-né gotic twin lambs by ewes. Behaviour 127:119–139.
prématuré: étude préliminaire. Enfance 1:33–46. Russell, M. J. (1976). Human olfactory communication. Nature
Porter, R. H. (1999). Olfaction and human kin recognition. 260:520–522.
Genetica 104:259–263. Russell, M. J., Mendelson, T., and Peeke, H. V. S. (1983).
Porter, R. H., and Doane, H. M. (1976). Maternal pheromone in Mothers’ identification of their infant’s odors. Ethol.
the spiny mouse (Acomys cahirinus). Physiol. Behav.
Sociobiol. 4:29–31.
16:75–78.
Sarnat, H. B. (1978). Olfactory reflexes in the newborn infant.
Porter, R. H., and Etscorn, F. (1974). Olfactory imprinting result-
J. Pediatr. 92:624–626.
ing from brief exposure in Acomys cahirinus. Nature
Schaal, B. (1986). Presumed olfactory exchanges between
250:732–733.
mother and neonate in humans. In Ethology and Psychology,
Porter, R. H., and Moore, J. D. (1981). Human kin recognition by
olfactory cues. Physiol. Behav. 27:493–495. J. Le Camus and J. Cosnier (Eds.). Privat, I. E. V., Toulouse,
Porter, R. H., and Picard, M. (1998). Effects of early odor expo- pp. 101–110.
sure in domestic chicks. Reprod. Nutr. Dev. 38:441–448. Schaal, B. (1988a). Olfaction in infants and children: develop-
Porter, R. H., and Winberg, J. (1999). Unique salience of mater- mental and functional perspectives. Chem. Senses
nal breast odors for newborn infants. Neurosci. Biobehav. Rev. 13:145–190.
23:439–449. Schaal, B. (1988b). Discontinuité natale et continuité chimiosen-
Porter, R. H., Deni, R., and Doane, H. M. (1977). Responses of sorielle: modéles animaux et hypothéses pour lhomme.
Acomys cahirinus pups to chemical cues produced by a foster Année Biol. 27:1–41.
species. Behav. Biol. 20:244–251. Schaal, B., and Orgeur, P. (1992). Olfaction in utero: can the
Porter, R. H., Wyrick, M., and Pankey, J. (1978). Sibling recog- rodent model be generalized? Quart. J. Exp. Psychol.
nition in spiny mice (Acomys cahirinus). Behav. Ecol. 44B:245–278.
Sociobiol. 3:61–68. Schaal, B., and Porter, R. H. (1991). “Microsmatic humans”
326 Porter and Schaal

revisited: the generation and perception of chemical signals. Singh, P. J., Tucker, A. M., and Hofer, M. A. (1976). Effects of
Adv. Study Behav. 20:135–199. nasal ZnSO4 irrigation and olfactory bulbectomy on rat pups.
Schaal, B., Montagner, H., Hertling, E., Bolzoni, D., Moyse, A., Physiol. Behav. 17:373–382.
and Quichon, R. (1980). Les stimulations olfactives dans les Smotherman, W. P. (1982). Odor aversion learning by the rat
relations entre l’enfant et la mère. Reprod. Nutr. Dev. fetus. Physiol. Behav. 29:769–771.
20:843–858. Smotherman, W. P., Robinson, S. R., Ronca, A. E., Alberts, J. R.,
Schaal, B., Orgeur, P., Lecanuet, J. P., Locatelli, A., Granier- and Hepper, P. G. (1991). Heart rate response of the rat fetus
Deferre, C., and Poindron, P. (1991). Nasal chemoreception in and neonate to a chemosensory stimulus. Physiol. Behav.
utero: first data in the fetal sheep. C. Roy. Acad. Sci. (Paris), 50:47–52.
Série III, Life Sci. 113:319–325. Sneddon, H., Hadden, R., and Hepper, P. G. (1998).
Schaal, B., Marlier, L., and Soussignan, R. (1995a). Chemosensory learning in the chicken embryo. Physiol.
Responsiveness to the odour of amniotic fluid in the human Behav. 64:133–139.
neonate. Biol. Neonate 67:397–406 Sommerville, B. A., Green, M. A., and Gee, D. J. (1990). Using
Schaal, B., Orgeur, P., and Arnould, C. (1995b). Olfactory pref- chromatography and a dog to identify some of the compounds
erences in newborn lambs: possible influence of prenatal in human sweat which are under genetic control. In Chemical
experience. Behaviour 132:351–365. Signals in Vertebrates, V. D. W. MacDonald, D. Muller-
Schaal, B., Orgeur, P., Desage M., and Brazier, J.L., (1995c). Schwarze, and S. E. Natynczuk (Eds.). Oxford University
Transfer of the aromas of the pregnant and lactating mother’s Press, Oxford, pp. 634–639.
diet to fetal and neonatal environments in the sheep. Chem. Soussignan, R., Schaal, B., Marlier, L., and Jiang, T. (1997).
Senses 20:93–94. Facial and autonomic responses to biological and artificial
Schaal, B., Orgeur, P., and C. Rognon, C. (1995d). Odor sensing in olfactory stimuli in human neonates: re-examining early
the human fetus: anatomical, functional and chemo-ecological hedonic discrimination of odors. Physiol. Behav. 62:745–758.
bases. In Prenatal Development, A Psychobiological Perspectiv, Stickrod, G., Kimble, D. P., and Smotherman W. P. (1982). In
J. P. Lecanuet, N. A. Krasnegor, W. A. Fifer, and W. utero taste odor aversion conditioning of the rat. Physiol.
Behav. 28:5–7.
Smotherman (Eds.). Lawrence Erlbaum, Hillsdale, NJ, pp.
Sullivan, R. M., and Toubas, P. (1998). Clinical usefulness of
205–237.
maternal odor in newborns: soothing and feeding preparatory
Schaal, B., Marlier, L., and Soussignan, R. (1998). Olfactory
responses. Biol. Neonate 74:402–408.
function in the human fetus: evidence from selective neonatal
Sullivan, R. M., Taborsky-Barba, S., Mendoza, R., Itano, A.,
responsiveness to the odor of amniotic fluid. Behav. Neurosci.
Leon, M., Cotman, C. W., Payne, T. F., and Lott, I. (1991).
112:1438–1449.
Olfactory classical conditioning in neonates. Pediatrics
Schapiro, S., and Salas, M. (1970). Behavioral response of infant
87:511–518.
rats to maternal odor. Physiol. Behav. 5:815–817.
Teicher, M. H., and Blass, E. M. (1976). Suckling in newborn
Schellinck, H. M., Monahan, E., Brown, R. E., and Maxson, S. C.
rats: eliminated by nipple lavage, reinstated by pup saliva.
(1993). A comparison of the contribution of the major histo-
Science 193:422–425.
compatibility complex (MHC) and Y chromosomes to the dis- Teicher, M. H., and Blass, E. M. (1977). First suckling response
criminability of individual urine odors of mice by Long-Evans in the newborn albino rat: the roles of olfaction and amniotic
rats. Behav. Genet. 23:257–263. fluid. Science 198:635–636.
Schleidt, M., and Genzel, C. (1990). The significance of mother’s Teicher, M. H., Flaum, L. E., Williams, M., Eckhert, S. J., and
perfume for infants in the first weeks of their life. Ethol. Lumia, A.R. (1978). Survival, growth and suckling behavior of
Sociobiol. 11:145–154. neonatally bulbectomized rats. Physiol. Behav. 21: 553–561.
Schley, P. (1977). Die Ausschaltung des Geruchsvermögens und Todrank, J., Heth, G., and Johnston, R. E. (1999). Social interac-
sein Einfluss auf das Saugverhalten von Jungkaninchen. Berl. tion is necessary for discrimination between and memory for
Muench. Tieraerztl. Wschr. 90:382–385. odours of close relatives in golden hamsters. Ethology
Schley, P. (1981). Geruchssinn und Saugverhalten bei 105:771–782.
Jungkaninchen. Kleintier Praxis 26:261–263. Varendi, H., Porter, R. H., and Winberg, J. (1994). Does the new-
Schmidt, H. J., and Beauchamp, G. K. (1989). Sex differences in born baby find the nipple by smell? Lancet 344:989–990.
responsiveness to odors in 9-month-old infants. Chem. Senses Varendi, H., Porter, R. H., and Winberg, J. (1996). Attractiveness
14:744. of amniotic fluid odor:evidence of prenatal olfactory learn-
Shair, H. N., Masmela, J. R., Brunelli, S. A., and Hofer, M. A. ing? Acta Paediatr. 85:1223–1227.
(1997). Potentiation and inhibition of ultrasonic vocalization Varendi, H., Porter, R. H., and Winberg, J. (1997). Natural odour
by rat pups: regulation by social cues. Dev. Psychobiol. preferences of newborn infants change over time. Acta
30:195–200. Paediatr. 86:985–990.
Singh, P. J., and Tobach, E. (1975). Olfactory bulbectomy and Varendi, H., Christensson, K., Porter, R. H., and Winberg, J.
nursing behavior in rat pups (Wistar DAB). Dev. Psychobiol. (1998). Soothing effect of amniotic fluid smell in newborn
8:151–164. infants. Early Human Dev. 51:47–55.
Olfaction and Social Behavior in Neonatal Mammals 327

Vince, M. A., and Ward, T. M. (1984). The responsiveness of Yamaguchi, M., Yamazaki, K., Beauchamp, G. K., Bard, J.,
newly born clun-forest lambs to odour sources in the ewe. Thomas, L., and Boyse, E. A. (1981). Distinctive urinary
Behaviour 89:117–127. odors governed by the major histocompatibility locus of the
Wallace, P. (1977). Individual discrimination of humans by odor. mouse. Proc. Natl. Acad. Sci. USA 78:5817–5820.
Physiol. Behav. 19:577–579. Yamazaki, K., Boyse, E. A., Mike, V., Thaler, H. T., Mathieson, B. J.,
Walser, E. S., Walters, E., and Ellison, J. (1984). Observations of Abbott, J., Boyse, J., Zayas, Z. A., and Thomas, L. (1976).
ewes and lambs in the field. Behaviour 91:190–203. Control of mating preferences in mice by genes in the major
Wedekind, C., and Furi, S. (1997). Body odour preferences in men histocompatibility complex. J. Exp. Med. 144: 1324–1335.
and women : do they aim for specific MHC combinations or Yamazaki, K., Beauchamp, G. K., Kupniewski, D., Bard, J.,
simply heterozygosity ? Proc. R. Soc. Lond. B 264: 1471–1479. Thomas, L., and Boyse, E. A. (1988). Familial imprinting
Wedekind, C., Seebeck, T., Bettens, F., and Paepke, A. J. (1995). determines H-2 selective mating preferences. Science
MHC-dependent mate preferences in humans. Proc. R. Soc. 240:1331–1332.
Lond. B 260:245–249. Yamazaki, K., Singer, A., and Beauchamp, G. K. (1999). Origin,
Whissell-Buechy, D., and Amoore, J. E. (1973). Odour-blind- functions and chemistry of H-2 regulated odorants. Genetica
ness to musk: simple recessive inheritance. Nature 104:235–240.
242:271–276. Zarrow, M. X., Denenberg, V. H., and Anderson, C. O. (1965).
Widstrom, A. M., Ransjo-Arvidson, A. B., Christensson, K., Rabbit: frequency of suckling in the pup. Science
Matthiesen, A. S., Winberg, J., and Uvnas-Moberg, K. (1987). 150:1835–1836.
Gastric suction in healthy newborn infants. Acta Paed. Scand. Zippelius, H. M., and Schleidt, W. M. (1956). Ultraschallaute bei
76:566–572. jungen Mausen. Naturwissenschaften 43:508–551.
16

Genetics of Olfactory Perception

Nancy L. Segal
California State University, Fullerton, California, U.S.A.

Tari D. Topolski
University of Washington, Seattle, Washington, U.S.A.

I. BEHAVIORAL-GENETIC APPROACH The present chapter endeavors to fulfill the following

TO OLFACTORY CHARACTERISTICS objectives: (1) present an overview of behavioral-genetic
methods available for examining genetic and environmental
Interest in the nature and bases of olfactory function in contributions to olfactory characteristics, (2) review selected
human and nonhuman populations has stimulated consid- findings from nonhuman animal studies that have considered
erable research activity (Kohl and Francoeur, 1995; Griff genetic influences on olfactory behavior, and (3) survey find-
and Reed, 1995; Kodis, 1998; Laurent, 1999). Studies of ings from twin and family studies at the juncture of olfactory
individual differences in olfactory characteristics have perception and kin recognition. Twin and family studies are
focused on measures related to experience (Castle, Van emphasized given the research interests of the authors.
Toller and Milligan, 2000), age (Murphy et al., 2000; Kline
et al., 2000), health (Wszolek and Markopoulou, 1998)
(see Chapter 22), smoking (de Jong et al., 1999; Davies II. BEHAVIORAL-GENETIC RESEARCH
and Davies, 1999), and sex (Gangestad and Thornhill, METHODS AND DESIGNS
1998; Yousem et al., 1999). Genetic influences on olfac-
tory measures have, however, received relatively less atten- Behavioral genetics is concerned with identifying genetic
tion. For example, overviews by Wysocki and Beauchamp and environmental factors underlying individual differ-
(1991) and Segal and Topolski (1995) referenced only a ences in behavior. The three major designs used in human
few twin and family studies of olfaction, in contrast research are twin, family, and adoption studies. Nonhuman
with the larger accumulation of nonhuman research. The research includes strain studies (studies of animals inbred
ensuing years have witnessed some additional twin and for at least 20 generations) and selection studies (studies of
family studies in this area, yet they remain few in number. animals selected and bred for a trait of interest). Molecular
Neglect of a behavioral-genetic perspective in olfactory biological techniques promise to identify DNA sequences
research is unfortunate because this approach has demon- associated with phenotypic variation (Plomin et al., 2002).
strated genetic effects across a wide range of human psy-
chological and physiological characteristics (Segal, A. Biology of Twinning
1999a). A behavioral-genetic approach to the study of odor
1. Types of Twins
identification and preference can contribute substantially
to our understanding of variation in normal and abnormal Monozygotic (MZ or identical) twins result when a fertilized
olfactory perception. egg (zygote) divides during the first 2 weeks postconception

329
330 Segal and Topolski

(Bryan, 1992). MZ co-twins are essentially genetic dupli- misleading estimates of genetic and environmental influ-
cates, yet various sources of influence can interfere with this ences on phenotypes (Segal, 1999a). Comparative analysis
plan (see Segal, 1999a and references therein). of co-twins’ multiple blood group systems provides accu-
Nondisjunction (failure of chromosomes to separate properly rate results, especially when combined with physical
during cell division) can produce MZ twins discordant for measures (Lykken, 1978). Researchers are now relying
Turner’s syndrome (X0) or Down syndrome (trisomy 21). increasingly on DNA profile analysis as a more accurate
Differential X chromosome inactivation can result in MZ and less expensive procedure. Some laboratories analyze
female twins discordant for X-linked traits, such as color- twins’ DNA patterns from buccal smears, cells obtained by
blindness and fragile X syndrome. Zygotic division delayed gently scraping the inner cheek (Richards et al., 1993).
until after day 8 is thought to underlie physical and anatomi- A recent, noninvasive “swish and spit” technique obtains
cal reversals (e.g., handedness or hair whorl) in approximately buccal cells by having subjects rinse with mouthwash and
25% of MZ twins. Splitting of the fertilized egg is also asso- expectorate into a saline solution (Hayney et al., 1996).
ciated with MZ twins’ increased frequency of midline abnor- Samples can be prepared at home and forwarded to labora-
malities (e.g., spina bifida and symmelia); in 80% of these tories in special kits. Physical resemblance questionnaires
cases only one twin is affected (Bomsel-Helmreich and can be reliably substituted for laboratory methods if the
Mufti, 1995). Physical and behavioral differences between latter are precluded (Segal, 1999a).
MZ twins are also associated with chorion type (Prescott et
al., 1999), unequal fetal blood supply (Bryan, 1992), and birth B. Twin Study Logic and Methods
order (Boggess and Chisholm, 1997).
Dizygotic (DZ or fraternal) twins are the product of the The logic of the classic twin design was first described by
separate fertilization of two eggs by two spermatozoa. DZ Sir Francis Galton (1875): “It is, that their history affords
twins, consequently, share the same genetic relationship as means of distinguishing between the effects of tendencies
nontwin siblings or 50% of their genes, on average, by received at birth and of those that were imposed by the cir-
descent. DZ twin pairs may be same-sex or opposite-sex. cumstances of their after lives; in other words, between the
These two types of DZ twins have been assumed to occur effects of nature and nurture.” The embryological bases of
with equal frequency, but an excess of same-sex male pairs twinning were not firmly established until the early 1900s,
has been suggested (Boklage, 1985). DZ twins’ genetic and yet Galton correctly reasoned that meaningful interpreta-
environmental differences are responsible for behavioral tions of twin data required reference to the distinction
and physical differences between them. Unusual variations between twin types. It is, therefore, surprising that some
of DZ twinning include superfecundation (conception of investigators fail to use standard methods for zygosity
DZ twins following separate coital acts during the same diagnosis or omit discussion of their procedures in pub-
menstrual cycle) and superfetation (multiple conceptions lished work (see Segal, 1986, 1999b).
separated by three to 4 weeks) (see Segal, 1999a). Some Sources of influence on twins’ similarities and differ-
superfecundated twins have different fathers, situations ences include genetic factors, shared environmental fac-
eventuating in extraordinary custody suits (Ambach et al., tors, and nonshared environmental factors. Greater trait
2000). These processes are presumed to be rare, but this resemblance between MZ co-twins, relative to DZ co-
may reflect lack of medical detection and documentation. twins, demonstrates a genetic contribution to individual
Twinning rates have risen considerably in recent years. differences in that trait. Greater trait resemblance between
Between 1980 and 1997, the number of twins born to women MZ twins reared together, relative to MZ twins reared
between the ages of 40 and 44 years increased by 63% apart, indicates shared environmental influence on that
(National Center for Health Statistics, 1999). This change is trait. Case studies can also be useful for associating envi-
tied largely to new assisted reproductive technologies (ART) ronmental effects and MZ within-pair differences, leading
involving hormonal treatments and/or transfer of multiple to more systematic studies.
embryos to women’s uteruses (Hecht and Magoon, 1998). A fundamental assumption in twin research is that trait-
A surprise finding is that ART has also increased the MZ twin- relevant environmental influences function similarly for
ning rate (albeit, less dramatically), possibly by altering the MZ and DZ twins. This concept is revisited periodically,
early environment of the developing embryo (Hecht, 1995). yet there is slim evidence of meaningful associations
between twins’ treatment and their behavioral outcomes
(Hettema, Neale and Kendler, 1995; LaBuda, Svikis and
2. Determination of Twin Type
Pickens, 1997). Other criticisms of twin studies include
Correct classification of twin type is an essential step in the primary biases (twins’ unique prenatal events) and recruit-
research process. This is because misclassification yields ment biases (excess MZ and female twin representation in
Genetics of Olfactory Perception 331

volunteer studies. Discussion and resolution of these issues procedure was used. All the MZ twin pairs, but only
have been addressed in several sources (see Plomin et al., 61% of the DZ twin pairs, were concordant for sensitiv-
1997; Prescott, Johnson and McArdle, 1999; Segal, ity/insensitivity to androstenone, demonstrating genetic
1999a). influence. In contrast, genetic influence on sensitivity to
Twin resemblance for continuous traits (e.g., height pyridine was not detected. The authors commented that the
or IQ) is expressed as an intraclass correlation. Twin experimental method employed in their study may have
similarity for age and sex can spuriously inflate estimates been unable to identify a genetic component associated
of genetic effects, so data are age- and sex-corrected prior with pyridine sensitivity.
to analysis (McGue and Bouchard, 1984). Twin resem- Gross-Isseroff et al. (1992) detected genetic influence
blance for discrete characteristics (e.g., blood type or men- on androstenone threshold in a study of 17 MZ and 15 DZ
tal disorder) is expressed as a concordance rate twin pairs. A significant genetic effect was additionally
(Gottesman, 1991; McGue, 1992). A phi coefficient which found for sensitivity to isoamyl acetate, but not for sensi-
incorporates population incidence and familial resem- tivity to citral or eugenol. These investigators suggested
blance for “either-or traits” is also available (Plomin, that Hubert et al.’s (1980) failure to detect a genetic com-
1990). Recent advances in analysis of twin data include ponent may reflect choice of odorants and/or the relatively
discriminant function (DF) multiple regression (DeFries older age of the participants.
and Fulker, 1985) and biometrical modeling procedures Segal et al. (1995) administered an odor-detection thresh-
(Neale and Cardon, 1992; Nance et al., 1998). old test (perfume-grade phenyl ethyl alcohol, or PEA) to
MZ and DZ twin pairs in California. Genetic influence on
PEA sensitivity was not detected. Twins also completed the
III. TWIN RESEARCH ON OLFACTORY
University of Pennsylvania Smell Identification Test
SENSITIVITY
(UPSIT), a self-administered, standardized test of odor iden-
A. Scientific Studies tification (Doty, 1995). A genetic effect on this measure was
indicated by the somewhat higher MZ (MZ r1
0.31; N

Hubert et al. (1980, 1981) compared detection sensitivity 45 pairs) than DZ intraclass correlation (r1
0.15; N
37
thresholds for acetic acid, isobutyric acid, and 2-sec-butyl- pairs). A larger genetic effect for males than for females was
cyclohexanone in a sample of 51 MZ male twin pairs and also observed.
46 DZ male twin pairs. Evidence of heritable variation in Similarity in odor preference ratings for the 40 UPSIT
odor sensitivity was not detected for any of the three sub- items were compared for MZ and DZ twins (Topolski,
stances. Interestingly, concordance for a specific anosmia 1993). Genetic influence was observed on the weak-strong
was not detected among any of the twin pairs. Relatively dimension for items classified as spicy, flowery, and
reduced sensitivity to isobutyric acid was shown by indi- burned and on the unpleasant-pleasant dimension for items
viduals who smoked, were lighter in weight, or who infre- classified as flowery. Both MZ and DZ twins showed gen-
quently consumed alcoholic beverages. However, these erally low correlations in ratings for foul and resinous
correlates of odor sensitivity accounted for only a very items. The latter result supports the Schleidt et al. (1988)
modest portion of the variance. “preprogrammed survival” theory, which asserts that
Forrai et al. (1981) assessed genetic influence on detection of environmental hazards fosters survival.
ketone-smelling ability (acetone and methylethylketone, or Kopala et al. (1998) compared UPSIT scores for 12 MZ
MEK) using 87 MZ twin pairs and 61 DZ twin pairs. twin pairs discordant for schizophrenia and 12 healthy
Genetic effects for acetone, but not for MEK, were sug- controls. The score for the combined twin group was sig-
gested. nificantly lower than that of the control group, and affected
Ward et al. (1983) administered odor detection and odor and unaffected MZ co-twins did not differ from one
discrimination tests to 14 MZ twin pairs and 6 DZ twin another. The researchers suggested that genetic factors
pairs who were discordant for Parkinson’s disease. contribute to cerebral dysfunction as assessed by odor
Olfactory function was inferior in affected twins, relative identification ability.
to their unaffected co-twins, in 13 of the 14 MZ twin pairs. Finkel (2000) studied odor identification and cognitive
It was suggested that the observed olfactory impairment functioning in 86 MZ twin pairs (31 reared apart and 55
was acquired, rather than inherited. reared together) and 141 DZ twin pairs (72 reared apart
Wysocki and Beauchamp (1984) compared resem- and 69 reared together). Moderate heritabilities were
blance for sensitivity to androstenone and pyridine in 17 derived for four odor functioning measures, although only
MZ twin pairs and 21 DZ twin pairs. An ascending con- those for odor identification (0.29) and intensity (0.25)
centration, two-sample (odorant vs. blank) forced-choice were significant; heritabilities for odor detection and
332 Segal and Topolski

pleasantness were not. A verbal component was found to Doty et al. (1984) documented age differences in olfac-
be highly correlated with odor identification. Age did not tory sensitivity, as measured by cross-sectional studies of
contribute substantially to associations between olfactory performance on the UPSIT. Longitudinal investigations
measures and cognitive abilities. using MZ and DZ twin pairs would provide new perspec-
tives on this behavior and other age-related changes and
B. Selected Case Studies: continuities (Plomin et al., 2001).
Kallmann’s Syndrome

Individuals affected with Kallmann’s syndrome show 2. Twin-Family Design

hypogonadotropic hypogonadism, eunuchoidal features, Families composed of MZ twins, their spouses, and chil-
and anosmia or hyposmia. This disorder is associated with dren are referred to as MZ half-sibling families, a model
a defect in the synthesis and / or release of lutenizing hor- derived from nonhuman studies (Gottesman and
mone-releasing hormone (LHRH) (Dark, 1997–1998). Bertelsen, 1989). The children of MZ twins are geneti-
Hipkin et al. (1990) compared olfactory function in MZ cally equivalent to half-siblings because they share a
male twins discordant for Kallmann’s syndrome. The genetically identical parent (twin mother or twin father).
affected twin was anosmic, while the unaffected twin was Furthermore, twin parents share the same genetic rela-
hyposmic. Reasons for the incomplete expression of this tionship (50%) with their nieces and nephews as with
disorder in these twins were unknown. The twins’ parents their own children. This unique family constellation
and sister showed normal olfactory function. enables behavioral and physical comparisons between
Gasztony et al. (1997) detected hyposmia with hypo- co-twins, nontwin spouses, twins and spouses, parents
gonadotropic hypogonadism in a DZ female twin and and their own children (who share an environment), par-
diagnosed it as Kallmann’s syndrome. Kallmann’s syn- ents and nieces/nephews (who do not share an environ-
drome underlines the significance of smell in sexual devel- ment), siblings, “half-siblings,” “half-brothers,” and
opment through progenitor cells in the olfactory placode. “half-sisters.”
This is because cells of the hypothalamus that secrete This informative twin design has never been used in
luteinizing hormone–releasing hormone arise from these human olfactory research. The many types of relation-
cells. ships that can be generated offer numerous opportunities
to explore hypotheses and predictions concerning the
C. Other Twin Research Designs and transmission of olfactory characteristics. Twin-family
Applications in Olfactory Research designs also promise to highlight the role of olfactory
cues in human kin recognition (Segal, 1999a) (also see
There are approximately 10 variants of the classic twin below).
design reviewed in Segal (1990, 1999a). Several are sum-
marized below because of their potential relevance to
olfactory researchers. 3. Twins Reared Apart
MZ twins reared apart (MZA) offer a direct estimate of
1. Longitudinal Twin Study genetic effects when twins are separated early in infancy
Longitudinal studies sample behavior at selected periods and raised in separate homes chosen at random (McGue
during the life span to record age-related developmental and Bouchard, 1998). The study of DZ twins reared
changes. Longitudinal twin designs offer additional apart (DZA) enables additional tests of possible genetic
opportunities to examine relative genetic and environ- interactions. Information on olfaction is unavailable
mental contributions to the timing and expression of in three early studies of twins reared apart (Newman
characteristics. Greater similarity in the level and con- et al., 1937; Shields, 1962; Juel-Nielsen, 1965). Reared
tour of MZ than DZ twins’ profiles suggests genetic apart twins were, however, studied in conjunction with
influence on the developmental progress of the trait(s) reared together twins in Finkel et al.’s study cited above.
under study. This area of research, now termed chrono- A small subset of participants in the Minnesota Study of
genetics, was anticipated in Galton’s 1875 paper. Twins Reared Apart, directed by Dr. Thomas J.
Incorporating both twins and their singleton siblings into Bouchard, Jr., at the University of Minnesota completed
a longitudinal study improves the efficiency of this the UPSIT and olfactory preference questionnaire
research design (see Wilson, 1983). New sophisticated administered to twins at CSUF. Such data should reveal
methods for analyzing longitudinal twin data have been how different rearing environments affect twin resem-
applied (Nance et al., 1998). blance in olfactory characteristics.
Genetics of Olfactory Perception 333

IV. FAMILY AND ADOPTION METHODS compounding the difficulty in interpretation. The roles of
AND APPLICATIONS IN OLFACTORY age and various environmental factors were emphasized as
RESEARCH affecting olfactory sensitivity to cyanide.
Whissell-Buechy and Amoore (1973) analyzed sensitiv-
According to quantitative genetic theory, the magnitude of ity to musk using 109 Caucasian families. Insensitivity to
resemblance between relatives for continuous traits (traits this odor was observed in 36 families, with males and
influenced by multiple genes) should vary as a function of females being equally affected. The results suggested simple
their genetic relatedness. Specifically, if genetic factors recessive autosomal inheritance for odor blindness to musk.
influence a trait, full brothers and sisters should show Wysocki and Beauchamp (1991) reported a study exam-
greater resemblance than half-brothers and half-sisters. ining the mode of inheritance of androstenone and pyridine
Quantitative traits are also influenced by environmental in 67 biological families. An X-linked pattern of transmis-
factors. Biological relatives living in a common environ- sion was suggested for androstenone, while genetically
ment may display similarities or differences due to either influenced sensitivity to pyridine was not indicated.
genetic or environmental effects. Genetic variance may be
decomposed into several components, such as additive B. Adoption Designs
variance and epistatic variance. Environmental variance
can be divided into shared and nonshared components. If Adoption studies include biological relatives raised apart
family resemblance is not detected for a given trait, this or unrelated individuals raised together. Resemblance
indicates that neither common genetic nor common envi- between biological relatives raised apart is associated with
ronmental factors are importantly influencing that trait shared genes in the absence of correlated, trait-relevant
(Plomin et al., 2001). rearing environments. Similarities between unrelated indi-
viduals living together are explained by common environ-
A. Family Designs mental factors. Methodological difficulties associated with
adoption designs, such as selective placement (i.e., con-
Family studies provide opportunities to trace the mode of gruence between features of the biological and nonbiolog-
genetic transmission of a trait across generations. This task ical families), have been described in Segal (1997, 2000).
is feasible in the case of single-gene traits that are highly Most adoption research has focused on IQ resemblance
resistant to environmental effects (e.g., color-blindness, car- between pairs of adopted away children and their biologi-
ried on the X chromosome), but is complex in the case of cal parents, adoptive parents and children, and adoptive
continuous traits. This is because parents transmit both siblings. Adoption studies have never been used to exam-
genes and environments to their offspring, thus confounding ine variation in olfactory sensitivity and preference, but are
genetic and environmental sources of variance. Other prob- well-suited to this purpose. A new adoption design focus-
lematic features of family studies include age differences ing on same-age unrelated siblings (UST-SAs or “virtual-
among relatives such that the developmental levels and twins”) may offer superior estimates of shared
experiences of family members may differ substantially. environmental influences. This is because these unique
Brown and Robinette (1967) tested cyanide sensitivity siblings enter their home at the same time and share more
in 2885 European school children and 86 parent pairs by a developmental experiences than siblings differing in age
serial dilution technique. The participants were organized (Segal, 2000).
into several categories of relatives (e.g., same-sex twins, Published or ongoing adoption studies of olfactory
same-sex siblings, same-sex parent and offspring). A sim- function are unavailable. Such work would uniquely con-
ple pattern of genetic inheritance was not indicated; rather, tribute to the literature in this field (Segal and Topolski,
the findings appeared to be more compatible with a famil- 1995). Adopted individuals have been included in studies
ial environmental factor that differentially affected males examining odor recognition and emotional closeness
and females. The correlations increased from siblings to among siblings, reviewed below.
same-sex siblings to same-sex twins, a finding interpreted
as suggesting a genetic effect similar to that of complex C. New Research Directions
physiological traits; however, the number of twin pairs was
small (N
20) and zygosity was not assessed. (It also The introduction of modern molecular genetic techniques
appears that MZ and DZ twin pairs were combined, thus has paved the way for new analyses of normal and abnor-
precluding more meaningful appraisal of these data.) Other mal olfactory functioning (Breer et al., 1996). Jones and
same-sex family members (e.g., father-son pairs) were Reed (1989) identified an olfactory neuron specific pro-
more highly correlated than the same-sex sibling pairs, tein (Golf) that appears to mediate olfaction. The mRNA
334 Segal and Topolski

encoding Golf was expressed in olfactory neuroepithelium, knowledge of the molecular bases of olfaction, however,
but not in several other tissues. The authors noted that a genetic factors underlying olfactory variation in drosophila
variant form of the G protein had been associated with are not well understood (Mackay et al., 1996). Attempts to
pseudohypoparathryoidism (PHP), a condition in which unravel the intricacies of the drosophila olfactory system
impaired olfactory functioning is observed. Recently, Doty include studies of flies exposed to various mutagens, as
et al. (1997) questioned the link between olfactory dys- well as studies of variability present in natural populations
function in PHP and and Gs alpha protein deficiency, sug- (Alcorta and Rubio, 1989). Available research includes
gesting that other mehanisms may be responsible. Buck inter- and intrastrain comparisons, selection studies, and
and Axel (1991) indicated that isolation of odorant recep- molecular cloning of olfactory genes. Only selected find-
tors and understanding of their specificity, diversity, and ings from this body of work are reviewed in the present
expression will enhance knowledge of olfactory percep- chapter; more detailed reviews are provided by Griff and
tion. These investigators cloned and characterized 18 Reed (1995) and Carlson (1996).
members of a multigene family that may encode a Five separate strains of Drosophila melanogaster, taken
diversity of odorant receptors. They suggested that a pos- from different locations, were exposed to ethyl alcohol,
sible tandem arrangement of genes “provides a template acetic acid, lactic acid, ethyl acetate, and n-butyraldehyde
for recombination events” that may be associated with (Fuyama, 1976). Interstrain differences in attraction to all
increased diversity within this system. substances, except acetic acid, were demonstrated. Fuyama
Current work suggests associations between specific (1978) also demonstrated variability in olfactory response
genes, olfactory function, and cognitive traits. Graves et al. to ethyl acetate among 40 lines, homozygous for the second
(1999) found that individuals who were anosmic at base- chromosome, chosen from a natural population of
line and who had at least one APOE-epsilon4 allele were Drosophila melanogaster. Olfactory response to 5%
at 4.9 times the risk for cognitive decline, relative to nor- ethanol was observed among lines of D. melanogaster
mosmics lacking APOE-epsilon4 alleles. Clearly, findings selected for increased knockdown resistance to ethanol
from twin, family, and adoption studies of olfactory func- (Hoffman and Cohan, 1987). Ethanol was used as a meta-
tion will enrich, and be enriched by, ongoing and future bolic resource to a greater extent among selected lines than
molecular genetic analyses. Related reviews of develop- among unselected lines, while unselected lines showed
ments in this exciting area are provided by Anholt (1991) greater attraction to ethanol. Based on these findings, the
and Mombaerts (1999) and in Chapters 4, 22, and 23. authors suggested that behavioral alteration may be “non-
adaptive,” such that flies tolerating ethanol more efficiently
are less attracted to it. Genetic variability in response to
V. NONHUMAN STUDIES ethyl alcohol and acetaldehyde was detected among isofe-
male lines representing natural drosophila populations cho-
Nonhuman studies enable experimental control over genetic sen from homogeneous and heterogeneous habitats
and environmental factors underlying behavior (Plomin (Alcorta and Rubio, 1989). This study enabled analysis of a
et al., 2001). As indicated in Sec. II, two forms of genetic link between genetic and environmental variation. It was
control can be achieved. In addition, selected aspects of the expected that relatively increased sensitivity and reduced
environment may be manipulated to identify important geno- variability of response would characterize inhabitants of
type-environment interactions. Alternatively, environments homogeneous environments and that in situations with
may be controlled in order to highlight genetic effects. scarce resources, a prompt response would facilitate acqui-
Another advantage of nonhuman research is the relatively sition of food, mates, etc. Observed responses to ethyl alco-
brief life spans of the test animals, allowing observation of hol (but not to acetaldehyde) confirmed these expectations.
behavioral transmission across many generations. Pruzan and Bush (1977) showed that both wild-type
(B) and mutant (bB) melanogaster larvae display prefer-
A. Drosophila ences for odors of their own strain under a variety of stimu-
lus conditions. Greater olfactory sensitivity was displayed
Many insects are highly dependent upon olfactory cues for by the wild-type larvae. Other investigators have also doc-
facilitating the identification of food sources, mates, and umented inter- and intraspecific variation among different
other resources essential to their survival (Fuyama, 1976, strains of drosophila in response to cues provided by lar-
1978). Research interest in genetic variation underlying vae (see Hoffman and Parsons, 1986; Monte et al., 1989).
olfactory response in drosophila has been considerable, Lilly and Carlson (1990) defined and characterized the
largely owing to progress in drosophila neurogenetics smellblind (sbl) locus. They showed that two alleles (iden-
during the 1960s (Ayyub et al., 1990). Despite increasing tified in independently isolated mutants) are associated
Genetics of Olfactory Perception 335

with a common defect in larval olfactory response, as well described (Greer, 1991). Perturbation of the olfactory bulb
as similarity in chemosensory and visual behaviors. They in the staggerer mutant has been associated with poor per-
asserted that these findings suggest specificity, rather than formance on an olfactory associative learning task, but not
generality, in neural or motor dysfunction for these on an olfactory habituation task (Deiss and Baudoin,
mutants. 1999). This olfactory deficit may partially explain abnor-
Studies of olfactory correlates of mutations at X-linked malities in the social and sexual behavoirs of these mice
loci have been reported. Helfand and Carlson (1989) (Deiss and Baudoin, 1997). Mice with PCD show degen-
showed that the origin of specific anosmia to benzaldehyde erating fibers in the lateral olfactory tract. These animals
can be genetically mapped to a small region of the X chro- offer opportunities to study olfaction using developmental
mosome near the pentagon locus. Ayyub et al. (1990) neurobiological and instrumental genetic methods. Finally,
demonstrated that mutations in five of six X-linked loci members of the Balb/c strain are missing a synapse from
were associated with partial anosmias in reponse to alde- the olfactory receptor axon onto the periglomerular cell
hydes or acetate ester. Ayer and Carlson (1991) reported dendrite. The consequences of this feature for processing
that the abnormal chemosensory jump 6 (acj6) is respon- odors are also unresolved. Additional information about
sible for “both reduction in electroantennogram amplitude these interesting genotypes is available in Greer (1991).
and diminished behavioral response, as if reduced antennal Several studies have documented genetic differences in
responsivesness to odorant is responsible for abnormal olfactory characteristics between strains, although the
chemosensory behavior in the mutant.” mechanisms underlying these observations require further
Molecular cloning of olfactory genes offers an informa- investigation. A sampling of these studies is discussed
tive approach to understanding olfactory functioning in below. Some analyses offer models for studying specific
Drosophila melanogaster. Several genetic screens have olfactory characteristics in humans. Recent advances in
identified olf mutants causing altered responses to odor- molecular-genetic techniques and findings relevant to
ants (Griff and Reed, 1995). Hasan (1990) describes the olfaction in the mouse are summarized in Taylor (1991).
molecular cloning of the olfactory gene olfE, which influ- Wysocki et al. (1977) used a conditioned aversion tech-
ences response to benzaldehyde in larve and in adults. nique and odors considered to be primary or complex in
Mutations in specific X-linked genes have been shown to humans to assess relative odorant insensitivity among male
affect response to benzaldehyde. It is noted, however, that mice from several inbred strains. This approach was
mutations in a specialized neural circuit that mediate adult guided by a general theory of olfaction in which specific
response to this substance may be implicated. The molecu- anosmias are associated with different profiles of odor
lar character of such genes is, therefore, of interest. The insensitivities in humans. Olfactory deficiency among C57
olfE gene appears to have at least two transcripts, one of mice was confined to isovaleric acid; C57 mice were dis-
which may be directly involved in olfactory function. The tinguishable from AKR mice in that a general olfactory
possibility that the gene may be associated with functions deficit was not indicated. That the anosmia observed
other than olfaction was also raised. The olf A, olf B, and among the C57 mice may be analogous to a specific anos-
olfF mutants are also defective in detecting aldehydes, mia in humans was suggested. The olfactory deficit in C57
whereas the olf C mutant is defective in detecting acetate mice observed by Wysocki et al. (1977) was subsequently
esters and olf D mutants are defective in responding to all confirmed by Pourtier and Sicard (1990). The latter recog-
tested odorants (Griff and Reed, 1995). Futher efforts nized, however, that individual variations in animal behav-
along these lines are needed to elucidate the roles played ior may be associated with variations in motivational state
by specific alleles with respect to olfactory characteristics. or physiological functioning. Evidence that the olfactory
mucosa and olfactory bulb of C57BL/6J mice are
B. Mouse Genetics activated by high concentrations of isovaleric acid was also
cited in their study. Pourtier and Sicard (1990) thus inter-
The mouse has been used in behavioral-genetic research preted their findings as indicating a partial, rather than a
for many years (Plomin et al., 2001). Studies applying a complete, deficit in isovaleric acid sensitivity. Greer
genetic approach to understanding the developmental neu- (1991) has commented that the response to increased con-
robiology and organization of the mouse olfactory system centrations of isovaleric acid suggests a role for subsets of
are, however, scarce (Greer, 1991). This may reflect the receptors less sensitive to this substance. In contrast, an
fact that only three genotypes associated with morpholog- absence of response (as reported by Wysocki et al., 1977)
ical anomalies of the olfactory system [neurological would suggest an absence of a relevant subset of receptors.
mutant, staggerer (also called reeler); Purkinje cell degen- More recently, Griff and Reed (1995) assessed the
eration (PCD); and the inbred strain Balb/c] have been genetic transmission of specific anosmia to isovaleric acid
336 Segal and Topolski

(concentrations of 105–106) using progeny from largely associated with Hamilton’s (1964) theory of kin
C57BL/6J and DBA/2J matings. An autosomal recessive selection. His premise was that natural selection should
pattern of inheritance for specific anosmia to this odorant favor alleles predisposing individuals to behave in ways
was observed. that favor the transmission of those alleles into subsequent
Genetic differences in olfactory-mediated choice generations. Alleles influencing individuals to favor oth-
behavior were shown to occur at day 10 in rat pups (Koski, ers likely to carry replicas of these alleles is an indirect
et al., 1977). (The nostrils of mouse pups are not fully means by which these alleles achieve future representa-
formed until days 5–8 after birth). Translocation animals tion. This reasoning suggests the presence of mechanisms
with a Wv marker (which has been associated with slight that facilitate recognition of close kin and identification of
macrotic anemia) showed significantly less olfactory- kin and nonkin. Inclusive fitness refers to the effects of
mediated choice behavior relative to nontranslocation sib- genes on the survival and reproductive success of an indi-
lings, marker animals, and nonmarker siblings. The vidual, plus the effects of genes on the survival and repro-
possibility that environmental factors associated with the ductive success of individuals with whom alleles are
test situation (e.g., handling and stress) may have influ- shared. Inclusive fitness is greater when individuals are
enced behavior was, however, not eliminated. more closely related genetically than when they are more
The effects of olfactory bulbectomy on olfactory behav- distantly related.
iors have been of interest. Wysocki et al. (1978) assessed
the impact of bulbectomy on conditioned taste aversion A. Nonhuman Animal Studies of Kin
using different strains of housemice (C57BL/6J and Recognition
AKR /J and their reciprocal F1 hybrids). Genotypic differ-
ences in preference for saccharin were displayed prior to The literature on kin recognition in animals refers to a wide
bulbectomy. Following conditioning, bulbectomy was variety of organisms (Sherman and Holmes, 1985; Pfennig
shown to markedly disrupt the conditionability of C57 and Sherman, 1995; Hepper and Cleland, 1998–1999).
mice, with little effect on the conditionability of AKR Proposed mechanisms by which kin recognition may occur
mice. F1 hybrids showed an intermediate response. The include green beard alleles, recognition alleles, and pheno-
effects of chemically induced peripheral damage to the typic matching systems. The green beard effect refers to
olfactory mucosa on exploratory behavior have also been signs that would both indicate relatedness between individ-
assessed in mice (Van Abeelen and Crusio, 1985). In this uals and predispose these individuals to act altruistically
study, obstruction of animals’ olfactory information-pro- toward others possessing the same sign (Dawkins, 1989).
cessing abilities, by means of intranasal zinc sulfate irriga- This mechanism is not considered likely, given the low
tion, reduced exploratory behavior due to interference with probability of the same gene being associated with the label
novelty detection. and with the tendency toward altruistic behavior.
Recognition alleles refer to the display of a phenotypic
marker and the ability to detect this marker in other organ-
VI. STUDIES OF OLFACTION AND KIN isms. Phenotypic matching refers to familiarity with one’s
RECOGNITION own phenotype and recognition of this phenotype in others.
Kin recognition includes not only identifying a signal, but
The identification and recognition of conspecifics, as well also acting on it, posing complex questions regarding
as olfactory communication of characteristics such as gen- underlying mechanisms (Hepper and Cleland, 1999).
der and social and reproductive status by olfactory cues, The effects of rearing conditions on kin relations, as
are well documented in the nonhuman animal literature assessed by olfactory preference, have also been investi-
(Brown, 1995; Heth, et al., 1999). It is proposed that gated. Males and females from two mouse species were
organisms possess “olfactory signatures,” or stable, salient either reared by natural parents or were cross-fostered
odors cues allowing recognition by relatives (Porter, (McCarty and Southwick, 1977). Mice who were cross-
1998–1999). Mateo and Johnston (2000) recently showed fostered showed reduced attraction to homospecific odors,
that golden hamsters use their own phenotype as a refer- as shown by preference for the soiled bedding materials of
rent for recognition of kin (see also Chapter 15). adult members from the rearing species. Studies of kin
The possibility that attraction and recognition in recognition have also included full sibling and half-sibling
humans may be partly mediated by olfactory cues, even if pairs. Werner et al. (1987), using a side-choice paradigm,
less efficiently than by visual or auditory means, has gen- demonstrated preference by green hatchling iguanas for
erated considerable research interest (Wells, 1987; Porter, unfamiliar siblings over familiar nonsiblings, following
1998–1999; Segal, 1999c). The roots of this interest are physical contact with pure kin groups.
Genetics of Olfactory Perception 337

A compelling series of studies using lambs offer addi- lambs share genetically influenced odors, enabling
tional insights into processes of kin recognition. Porter their mothers to identify them.
et al. (1991) showed that ewes preferred their own familiar 4. Identical twins (partial contact with first-born
lamb to the co-twin of this lamb who was raised apart in twin): As in the fraternal twin experiment, first-
isolation. Nevertheless, the ewe showed significant prefer- born twins were placed in separate chambers in
ence for the twin lamb that had been taken from her, as their mothers’ presence, while second-born twins
compared with an isolated alien lamb or an alien lamb were isolated. Several hours later, ewes favored
raised by its own mother until the time of testing. It was the identical twins much more than alien lambs,
suggested that the fraternal co-twin lambs’ odors, while dif- but, more importantly, they treated first-born and
ferent, may have been somewhat similar, enabling discrimi- second-born identical twins alike. Ewes bearing
nation and identification of these siblings by the mother. Of identical twins apparently recognized the second-
course, the source of the first-borns’ odors could have born’s odor, based on their exposure to the first-
reflected genetically based scents, acquired scents from born’s odor. Since these lambs share identical
sucking or being licked by its mother, or both. Furthermore, genes they would be expected to exude similar
some twins were opposite-sex, so co-twin distinctions may scents. Of course, lack of discrimination between
have been facilitated by the sex difference. Subsequent twins in this experiment may have partly reflected
experimental manipulations, using 9–29 mother-twin pairs, the mother’s inability to apply her own scent to
separated out these various influences (Romeyer et al., first-born twins because physical contact was
1993). These experiments are described below: denied. Twin lambs were preferred over alien
lambs, as expected.
1. Same-sex fraternal twins (full contact with first-
born twin): First-born lambs remained with their In these experiments, first-borns were preferred because
mothers, while second-born lambs were cleansed they remained with mothers while second-borns were iso-
of amniotic fluid and reared in isolation from lated. There was nothing intrinsic to being a first-born twin
birth. Ewes preferred their familiar first-born that would have affected ewes’ preferences. Finally, we can
twins to their second-born twins and preferred only wonder how well the ewe whose cell led to Dolly, the
their second-born twins to unfamiliar unrelated cloned lamb, would have identified her daughter—the ewe
(alien) lambs. These findings suggested that twin was no longer alive when Dolly was born. Their common
lambs shared genetic and environmental features genetically based scents may have helped this process, but
allowing their mothers to find their separated sec- the absence of acquired odors from the birth process or
ond-borns. from sucking or licking may have impeded it.
2. Identical twins (full contact with first-born twin): The possibility that anatomical and physiological vari-
Identical twin lambs were created by artificially ation associated with genetic variation in the major histo-
cleaving fertilized eggs and implanting them in an compatibility complex (MHC) explains distinctive odor
unrelated ewe. As in the fraternal twin experiment, compounds has been raised (Boyse et al., 1991; Eggert
first-born lambs remained with their mothers, while et al., 1998–1999a). Lenington and Egid (1985) observed
second-born lambs were reared in isolation. Ewes the ability of wild-type female mice to distinguish
showed greater acceptance of their first-born than between males with differing T-locus genotypes by
second-born twin lambs, possibly because first- means of olfactory cues. A subsequent study by these
borns had acquired their mothers’ odors from her investigators demonstrated that males also rely on olfac-
urine, saliva, or milk during their interactions. tory cues to distinguish between genotypes at the T locus,
Mothers also favored their own isolated lamb over but that the genetic basis differs between males and
an alien lamb. females (Egid and Lenington, 1985). The role of the X
3. Same-sex fraternal twins (partial contact with first- and Y chromosomes in the chemosensory identity of
born twin): Following delivery, first-born twins were mice of differing genotypes has also been examined
immediately placed in separate chambers in their (Yamazak et al., 1987). X and Y chromosomes each con-
mothers’ presence, allowing maternal access to ferred an individual odor related to the genotype, but
infants’ odors but preventing physical contact. were less salient than the MHC.
Second-born twins were housed in isolation. Several Olfactory cues associated with the MHC have also been
hours later, ewes preferred their first-born twins over shown to affect mate choice and frequency of pregnancy
their second-born twins, as well as their second-born blocks in inbred strains of mice, thus promoting heterozy-
twins over alien lambs. These findings supported the gosity in the MHC (Egger et al., 1996). Brown (1995)
researchers’ earlier conclusion that fraternal twin observed that congenic mouse and rat strains differing in
338 Segal and Topolski

Class I and Class II regions of the MHC produce different DZ twin adolescents and young adults wore t-shirts for three
urinary odors contributing to mate selection and parental consecutive nights while refraining from using perfume or
recognition. However, it was also found that dietary differ- other cosmetics. Judges sniffed a target t-shirt, then
ences affect individual odors more significantly than attempted to identify the “relative” from an array of three t-
differences at one MHC locus. These results suggest that shirts. (These three shirts belonged to the co-twin and to two
urinary odors reflect an interaction among the MHC, com- unrelated twins of the same age and sex.) Contrary to expec-
mensal bacteria, and dietary substances. A recent study of tation, the proportion of correct identifications for MZ twins
gerbils showed that the behaviors of individuals toward kin did not significantly exceed that of DZ twins. Judges’ lack of
and nonkin vary with the situation and age and sex of the relatedness and/or familiarity with the twins may be associ-
interactants. These findings raise the possibility that the ated with the nonsignificant findings. Most previous studies
MHC may differentially affect kin recognition and kin- demonstrating accuracy in identification of individuals by
correlated behavior (Hepper and Cleland, 1999). odor cues from garments have used genetically related or
How the immune system affects body odors and ulti- socially familiar judges. Twins’ dietary practices may have
mately social relations remain important questions. Future also affected the findings. For example, older twins showed
studies assessing the mediation of social preferences by significantly greater differences in spicy food intake than
olfactory cues among genetically related and unrelated younger twins and received fewer correct ratings.
animals, reared together and apart in diverse settings, will, Porter (1987) noted that to “assess the influence of famil-
hopefully, clarify the underlying mechanisms. The role of iarization on the development of recognition of particular
prenatal learning of olfactory cues also requires more seri- kin, one would ideally like to test individuals with the sig-
ous consideration (Hepper, 1999). Recent advances in natures of relatives with whom there has been no prior con-
gene mapping should assist in the identification of genes tact.” MZA and DZA twins are well suited to such analyses.
relevant to these processes. Spouses and children (who have lived with one twin, but not
the other) and unrelated individuals could be requested to
B. Twin Studies of Kin Recognition distinguish between the odors of reared apart twins. In addi-
tion, reunited co-twins might be asked to identify the odor
Porter (1987) suggested that if “humans actually have of the co-twin from an array of odors (Segal, 1999a). Similar
genetically-determined biochemical ‘fingerprints’ that analyses could be conducted using various other pairs of
serve as a basis for individual odors . . . one would expect a biological and adoptive relatives. Practical aspects of con-
correlation between genetic similarity and odor similarity.” ducting these studies may, however, be difficult to resolve.
Several twin studies have addressed this interesting issue.
Wallace (1977) conducted a series of experiments involving C. Family Studies of Kin Recognition
exposure of research participants to the palm odors of two
unrelated females, MZ female co-twins on the same diet, An increasing number of studies are demonstrating olfac-
and MZ female co-twins on different diets. Subjects were tory recognition by relatives living together. Macfarlane
more successful in distinguishing between the unrelated (1975) found that infants as young as 6 days of age pre-
females than between the twins. Subjects could, however, ferred the breast pad worn by their own mother to that of
more easily discriminate between twins eating different another mother, as measured by the amount of time the
diets than between twins eating similar diets. infant turned toward the pad. Cernoch and Porter (1985)
Galton (1875) was the first investigator to question found that 4-day-old breast-fed infants, but not bottle-fed
whether dogs could distinguish between MZ twins on the infants, were sensitive to differences between the under-
basis of odor cues. It was some time later that Kalmus (1955) arm odors of their mothers and unfamiliar females. Makin
demonstrated failure by dogs to discriminate between MZ co- and Porter (1989) observed that bottle-fed female infants
twins when presented with twins’ odors as part of a retrieval display a nonspecific preference for the breast odors of lac-
task. Dogs could, however, differentiate between MZ co- tating females. The role of olfactory cues in the mother-
twins’ odors in a tracking task. Hepper (1988) showed that infant relationship is further summarized in Porter and
dogs could discriminate between the odors of both MZ and Schaal (1995) and Porter (1998–1999) and in Chapter 15.
DZ co-twins in a matching-to-sample experiment; MZ co- Porter and Moore (1981) found that children were cor-
twins were living apart and followed different dietary prac- rectly identified by siblings and by mothers from odors on
tices. Dogs were unable to distinguish between infant MZ their clothing. Other studies have demonstrated that new
co-twins living in the same home and fed identical diets. mothers can correctly distinguish their own infants from
A twin study of olfaction and kin recognition was com- unrelated infants by odor (Russell et al., 1983; Porter et al.,
pleted in our laboratory at CSUF (Segal et al., 1995). MZ and 1983) and that fathers, grandmothers, and aunts can identify
Genetics of Olfactory Perception 339

the odors of garments worn by newborn relatives (Porter on olfactory characteristics. The classic MZ-DZ twin
et al., 1986). Porter et al. (1986) also showed that adults dis- comparison and its variants offer a wealth of new ways for
criminated between the odors of garments worn by siblings considering the nature and bases of olfactory traits and their
from whom they had been separated for 1–30 months and an role in human social relationships.
age- and sex-matched stranger. In contrast, research partici- Future research should address the role of genetic fac-
pants could not match the odors of spouses from the odors tors in age-related changes in olfactory sensitivity.
of t-shirts worn for 3 consecutive nights (Porter, Cernoch Alteration of the nuclei of the supporting and sensory cells
and Balogh, 1985). Individuals can, however, recognize of the olfactory epithelium has been associated with aging
their own odor (Schleidt, 1980; Lord and Kasprzak, 1989) (Naessen, 1971), although marked individual differences
and the odor of their spouse (Schleidt, 1980). have been reported (Doty and Snow, 1988). Olfactory neu-
Recent work extends these themes. A small sample of rons are unique in that they are continually being replen-
full siblings, half-siblings, step-siblings, and some of their ished during the life span of the organism. Fully mature
mothers participated in a study of odor recognition and neurons maintain their juvenile nature, a process believed
emotional closeness (Gall, 1999). Full siblings and biologi- to be under genetic control. However, evidence that rearing
cal mothers correctly chose the t-shirt worn by their sibling in a filtered air environment increased the life span of
or child, respectively, while half-siblings, step-siblings, and olfactory cells in mice (Hinds et al., 1984) led to increased
their parents were less able to make these discriminations. interest in environmental influences on olfactory neuro-
Biological mothers correctly identifying their offspring genesis. Current evidence supports the view that mitosis in
indicated a significant preference for their children’s olfac- the olfactory epithelium is influenced by genetic factors
tory cues. In contrast, correct identification was not associ- but may be modified by environmental events (see also
ated with preference for the co-sibling’s odor in the three Chapter 5). Twin and adoption studies might help resolve
sibling groups. It was suggested that this finding may these issues. Future studies focusing on human and non-
reflect reduced reliability among the young respondents. human genotypes associated with olfactory impairment
Contrary to expectation, warmth/closeness ratings were not would be welcome. Such work may improve understand-
elevated among either full or half-siblings who correctly ing of olfactory dysfunction (as in the case of Kallmann’s
identified their co-sibling. Self-reports, supplemented by syndrome) and direct us toward better treatments for olfac-
parental assessments and naturalistic observations, may tory difficulties.
offer a more complete picture of siblings’ social relations. New olfactory studies of kin recognition are also needed.
Variations on the studies reviewed above would further Nicolaides (1974) has noted the large number of factors
refine knowledge of the role of olfactory cues in human kin affecting the composition of fatty acids produced by the skin
recognition. The unique relationships generated by MZ and the consequent rarity of two individuals producing the
half-sibling families would help disentangle genetic and same substances in identical proportions. More recent find-
environmental effects on olfaction in the context of kin ings are defining intriguing topics for further study. Identical
recognition. In addition, the comparative accuracy with twins’ sweat samples showed greater resemblance than
which mothers are able to identify biological children those of unrelated pairs as revealed by gas chromatography
whom they have reared versus biological children given (McCormick et al., 1995). In addition, Eggert et al.
away for adoption and / or children they have adopted (1998–1999b) found that human urine odors and the profile
would clarify the contribution of shared genes and envi- of volatile components of urine are associated with the
ronments to kin recognition, based upon olfactory cues. MHC. Most importantly, the profile of some specific com-
Individuals might also be requested to discriminate ponents and volatiles constitutes MHC-related odor signals
between biological and nonbiological siblings. An intri- in humans.
guing research design would request MZA twins to iden- Wedekind et al. (1995) reported that the MHC affects
tify the reared apart twin and an adoptive sibling (with both human body odors and preferences and that female
whom the rearing environment was shared), although prac- preferences are influenced by their hormonal status.
tical considerations may render such studies unfeasible. Interestingly, females preferred the odors of males whose
MHC differed from theirs. Consistent with this work, Ober
et al. (1997) and Genin et al. (2000) found fewer HLA haplo-
VII. SUMMARY
type matches than expected among Hutterite couples.
A. Conclusions and Future Research Directions Significant negative assortative mating was confirmed after
considering parental origin of shared haplotypes. The con-
Twin studies have not been fully exploited as research clusion was that mate choice was influenced by HLA haplo-
tools for investigating genetic and environmental influences types with avoidance of spouses whose haplotypes match
340 Segal and Topolski

one’s own. Preference for mates differing in selected genetic Boggess, K. A., and Chisholm, C. A. (1997). Delivery of the non-
factors may help avoid the unfavorable behavioral and phys- vertex second twin; A review of the literature. Obstet. Survey
ical consequences associated with inbreeding. In contrast 52:728–735.
with Ober et al.’s results, Hedrick and Black (1997) found no Boklage, C. E. (1985). Interactions between opposite-sex dizy-
evidence of negative assortative mating for MHC alleles gotic fetuses and the assumption of Weinberg difference
method epidemiology. Am. J. Human Genet. 37:59–105.
using South Amerindian couples, given that HLA-sharing
Bomsel-Helmreich, O., and Mufti, W. A. (1995). The mechanism
proportions were close to random mating expectations.
of monozygosity and double ovulation. In Multiple
In a related study, Wedekind et al. (1997) found that males Pregnancy: Epidemiology, Gestation and Perinatal Outcome,
and females’ pleasantness ratings of six different shirts cor- L. G. Keith, E. Papiernik, D. M. Keith, and B. Luke (Eds).
related negatively with the MHC similarity between smeller Parthenon, New York, pp. 25–40.
and t-shirt wearer. More revealing, perhaps, raters were Boyse, E. A., Beauchamp, G. K., Yamazaki, K., and Bard, J.
reminded of their own current or former partners when sniff- (1991). Genetic components of kin recognition in mammals.
ing t-shirts whose wearers’ MHC alleles generally differed In Kin Recognition, P.G. Hepper (Ed.). Cambridge University
from theirs. The general consistency of findings across stud- Press, Cambridge, pp. 148–161.
ies (albeit a small number) suggests that MHC similarity and Breer, H., Wanner, I., and Strotmann J. (1996). Molecular genet-
dissimilarity contributes to body odor preferences. Further ics of mammalian olfaction. Behav. Genet. 26:209–219.
Brown, K. S., and Robinette, R. R. (1967). No simple pattern of
work along these lines would add depth and dimension to
inheritance in ability to smell solutions of cyanide. Nature
current understanding of mate choice.
215:406–408.
Advances in decoding the human genome will no doubt Brown, R. E. (1995). What is the role of the immune system
also contribute to what we know and can know about in determining individually distinct body odours? Int.
human olfaction and related characteristics. This informa- J. Immunopharmacol. 17:655–661.
tion could be used to prevent loss of olfactory sensitivity Bryan, E. M. (1992). Twins and Higher Multiple Births: A Guide
with aging. It may also highlight the relevant genetic fac- to Their Nature and Nurture. Edward Arnold, London.
tors mediating social attraction, yielding novel insights Buck, L., and Axel, R. (1991). A novel multigene family may
into the fabric of human relations. encode odorant receptors, A molecular basis for odor recogni-
tion. Cell 65:175–187.
Carlson, J. R.(1996). Olfaction in drisophila: from odor to behav-
ACKNOWLEDGMENTS ior. Trends Gen. 12:175–180.
Castle, P. C., Van Toller, S., and Milligan, G. J. (2000). The effect
Studies by the authors that are referenced in this chapter of odour priming on cortical EEG and visual ERP responses.
were supported by awards from the Fragrance Research Int. J. Psychohysiol. 36:123–131.
Fund, Ltd. (now Olfactory Research Fund, Ltd.), New Cernoch, J. M., and Porter, R. H. (1985). Recognition of mater-
York, NY, and California State University, Fullerton (Dr. nal axillary odors by infants. Child Dev. 61:178–190.
Dark, G. (1997–1998). On-line Medical Dictionary. Academic
Nancy L. Segal), and by a grant from NIDCD,
Medical Publishing and Cancer Web.
PO1DC00161 (Dr. Richard L. Doty). Research assistance
Davies, C. W., and Davies, S. (1999). Prediction of olfactory
was provided by Kathleen W. Brown, Ph.D., Linda Araki, response based on age, gender and smoking. J. Med. Eng.
M.A., Dinah G. Gitlin, B.A., Steven M. Wilson, Ph.D., Technol. 23:73–76.
Jennifer Trevitt, M.A., and Cesar Gasca. Dawkins, R. (1989). The Selfish Gene, 2nd ed. Oxford University
Press, Oxford.
DeFries, J. C., and Fulker, D. W. (1985). Multiple regression
REFERENCES analysis of twin data. Behav. Genet. 13:467–473.
Deiss, V. C. (1997). Hyposmia for butanol and vanillin in mutant
Alcorta, E., and Rubio, J. (1989). Intrapopulational variation of staggerer mice. Physiol. Behav. 61:209–213.
olfactory responses in Drosophila melanogaster. Behav. Deiss, V., and Baudoin, C.(1999). Olfactory learning abilities in
Genet. 19:285–299. staggerer mutant mice. C. R. Acad. Sci. III 322:467–471.
Ambach, E., Parson, W, and Brezinka, C. (2000). de Jong, N., Mulder, I., de Graaf, C., and van Staveren, W. A.
Superfecundation and dual paternity in a twin pregnancy (1999). Impaired sensory functioning in elders: the relation
ending with placental abruption. J. Foren. Sci. 45: 181–183. with its potential determinants and nutritional intake.
Anholt, R. R. H. (1991). Odor recognition and olfactory trans- J. Gerontol. (A. Biol. Sci. Med Sci.) 54:B324–331.
duction: The new frontier. Chem. Senses 16:421–427. Doty, R. L. (1995). The Smell Identification TestTM Administration
Ayer, R. K., and Carlson, J. (1991). acj6: A gene affecting olfac- Manual, 3rd ed. Sensonics, Inc., Haddon Heights, NJ.
tory physiology and behavior in Drosophila. Proc. Natl. Acad. Doty, R. L., and Snow, J. B. Jr., (1988). Age-related alterations in
Sci. 88:5467–5471. olfactory structure and function. In Molecular Neurobiology of
Genetics of Olfactory Perception 341

the Olfactory System, F. L. Margolis and T. V. Gretchell (Eds.) interaction with apolipoprotein E epsilon4 status. Neurology
Plenum Publishing Corporation, New York, pp. 355–373. 53:1480–1487.
Doty, R. L., Shaman, P., Applebaum, S. L., Giberson, R., Greer, C. A. (1991). Genetics and the neurobiology of olfactory
Siksorski, L. and Rosenberg, L. (1984). Smell identification bulb circuits. In Chemical Senses, Vol. 3, Genetics of
ability: changes with age. Science 226:1441–1443. Perception and Communications, C. J. Wysocki and M. R.
Doty, R. L., Fernandez, A. D., Levine, M. A., Moses, A., and Kare (Eds.). Marcel Dekker, Inc., New York, pp. 291–315.
McKeown, D. A. (1997). Olfactory dysfunction in type I Griff, I. C., and Reed, R. R. (1995). The genetics of olfaction.
pseudohypoparathyroidism: dissociation from Gs alpha pro- Curr. Opin. Genet. Dev. 5:657–661.
tein deficiency. J. Clin. Endocrinol. Metab. 82:247–250. Gross-Isseroff, R., Ophir, D., Bartana, A., Voet, H., and Lancet,
Eggert, F., Holler, C., Luszyk, D., Muller-Ruchholtz, W., and D. (1992). Evidence for genetic determination in human twins
Ferstl, R.(1996). MHC-associated and MHC-independent uri- of olfactory thresholds for a standard odorant. Neurosci. Lett.
nary chemosignals in mice. Physiol. Behav. 59:57–62. 141:115–118.
Eggert, F., Muller-Ruchholtz, W., and Ferstl R. (1998–1999a). Hamilton, W. D (1964). The genetical evolution of social behav-
Olfactory cues associated with the major histocompatibility ior. J. Theoret. Biol. 7:1–52.
complex. Genetica 104:191–197. Hasan, G. (1990). Molecular cloning of an olfactory gene from
Eggert, F., Luszyk, D., Haberkorn, K., Wobst, B., Vostrowky, O., Drosophila melanogaster. Proc. Natl. Acad. Sci. 87:
Westphal, E., Bestmann, H. J., Muller-Ruchholtz, W., and 9037–9041.
Ferstl, R. (1998–1999b). The major histocompatibility com- Hayney, M. S., Poland, G. A., and Lipsky, J. J. (1996). A nonivasive
plex and the chemosensory signalling of individuality in ‘swish and spit’ method for collecting nucelated cells for HLA
humans. Genetica 104:265–173. typing by PCR in population studies. Hum. Hered. 46: 108–11.
Egid, K., and Lenington, S. (1985). Responses of male mice to Hecht, B. R. (1995). The impact of assisted reproductive tech-
odors of females: Effects of T- and H-2-locus genotype. nology on the incidence of multiple gestation. In Multiple
Behav. Gen. 15:287–295. Pregnancy: Epidemiology, Gestation and Perinatal Outcome,
Finkel, D., Pedersen, N. L., and Larsson, M. (2001). Olfactory L. G. Keith, E. Papiernik, D. M. Keith, and B. Luke (Eds).
functioning and cognitive abilities: A twin study. J. Gerontol. Parthenon, New York, pp.175–190.
B. Psychol. Soc. Sci. 56:226–233. Hecht, B. R., and Magoon, M. W. (1998). Can the epidemic of
Forrai, G., Bankovi, G., Szabados, T., and Papp, E. S. (1981).
iatrongenic multiples be conquered? Clin. Obstet. Gvnecol.
Ketone compound smelling ability: study in Hungarian twins.
41:126–137.
Acta Med. Acad. Sci. Hung. 38:153–158,
Hedrick, P. W., and Black, F. L. (1997). HLA and mate selection:
Fuyama, Y. (1976). Behavior genetics of olfactory responses in
no evidence in South Amerindians. Am. J. Hum Genet., 61:
Drosophila. I. Olfactory and strain differences in Drosophila
505–511.
melanogaster. Behav. Genet. 6:407–420.
Helfand, S. L., and Carlson, J. R. (1989). Isolation and charac-
Fuyama, Y. (1978). Behavior genetics of olfactory responses in
terization of an olfactory mutant in Drosophila with a chemi-
Drosophila. II. An odorant-specific variant in a natural popu-
cally specific deficit. Proc. Natl. Acad. Sci. 86:2908–2912.
lation of Drosophila melanogaster. Behav. Genet. 8:399–414.
Hepper, P. G. (1988). The discrimination of human odour by the
Gall, J. A. (1999). Odor recognition and emotional closeness
among siblings: A study of kin selection. Ph.D. dissertation, dog. Perception 17:549–554.
Wayne State University, Detroit, MI. Hepper, P. G. (1990). Fetal olfaction. In Chemical Signals
Galton, F. (1875). The history of twins as a criterion of the relative in Vertebrates V, D. W. Macdonald, D. Muller-Schwarze, and
powers of nature and nurture. J. Anthropol. Inst. 5: 391–406. S. E. Natynzcuk (Eds.) Pergamon Press, Oxford, pp. 282–288.
Gangestad, S. W., and Thornhill, R. (1998). Menstrual cycle vari- Hepper, P. G., and Cleland, J. (1998–1999). Developmental
ation in women’s preferences for the scent of symmetrical aspects of kin recognition. Genetica 104:199–205.
men. Proc. R. Soc. Lond. B: Biol. Sci. 265:927–933. Heth, G., Todrank, J., and Johnston, R. E. (1999). Similarity in
Gasztonyi, Z., Paulin, F., Siklosi, G., and Czeizel, E. (1997). the qualities of individual odors among kin an dspecies in
Therapeutic importance of the diagnosis of Kallmann syn- Turkish Mesocricetus brandti and golden Mesocricetus aura-
drome. Orv. Hetil. 138:2529–2532. tus hamsters. J. Comp. Psychol. 113:321–326.
Genin, E., Ober, C., Weitkamp, L., and Thompson, G. (2000). Hettema, J. M., Neale, M. C., and Kendler, K. S. (1995). Physical
A robust test for assortative mating. Eur. J. Human Genet. similarity and the equal-environment assumption in twin stud-
8: 119–124. ies of psychiatric disorders. Behav. Genet. 25:327–335.
Gottesman, I. I. (1991). Schizophrenia Genesis: The Origins of Hinds, J. W., Hinds, P. L., and McNelly, N. A. (1984). An auto-
Madness. W. H. Freeman and Co., New York. radiographic study of the mouse olfactory epithelium:
Gottesman, I. I., and Bertelsen, A. (1989). Confirming unex- Evidence for long-lived receptors. Anat. Rec. 210:375–383.
pressed genotypes for schizophrenia. Arch. Gen. Psychiatry. Hipkin, L. J., Casson, I. F., and Davis, J. C. (1990). Identical twins
46:867–872. discordant for Kallmann’s syndrome. J. Med. Genet. 27: 198–199.
Graves, A. B., Bowen, J. D., Rajaram, L., McCormick, W. C., Hoffman, A. A. (1991). Chemical cues from conspecifics
McCurry, S. M. Schellenberg, G. D., and Larson, E. B. and resource response variation in drosophila. In Chem.
(1999). Impaired olfaction as a marker for cognitive decline: Senses, Vol. 3, Genetics of Perception and Communications,
342 Segal and Topolski

C. J. Wysocki and M. R. Kare (Eds.). Marcel Dekker, Inc., Makin, J. W., and Porter, R. H. (1989). Attractiveness of lactating
New York, pp. 147–167. females’ breast odors to neonates. Child Dev., 60:803–810.
Hoffman, A. A., and Cohan, F. M. (1987). Olfactory responses of Macfarlane, A. (1975). Olfaction in the development of social pref-
Drosophila melanogaster selected for knockdown resistance erences in the human neonate. In Parent-Infant Interaction (Ciba
to ethanol. Behav. Genet. 17:307–312. Foundation Symposium 33) R. Porter and M. O’Connor (Eds.).
Hoffman, A. A., and Parsons, P. A. (1986). Inter- and intra-specific Elsevier/Excepta/New Holland, Amsterdam, pp. 103–117.
variation in the response of Drosophila melanogaster and Mackay, T. F., Hackett, J. B., Lyman, R. F., Wayne, M. L., and
D. stimulans to larval cues. Behav. Genet. 16:295–306. Anholt, R. R. (1996). Quantitative genetic variation of odor-
Hubert, H. B., Fabsitz, R. R., Feinleib, M., and Brown, K. S. guided behavior in natural population of Drosophila
(1981a). Olfactory sensitivity in humans: Genetic versus envi- melanogaster. Genetics 144:727–735.
ronmental controls. Science, 208: 607–609. Mateo, J. M., and Johnston, R. E. (2000). Retention of social
Hubert, H. B., Fabsitz, R. R., Brown, K. S., and Feinleib, M. recognition after hibernation in Belding’s ground squirrels.
(1981b). Olfactory sensitivity in twins. In Twin Research, Vol. 3, Animal Behav. 59:491–499.
Epidemiological and Clinical Studies, L. Gedda, P. Parisi, and McCarty, R., and Southwick, C. H. (1977). Cross-species foster-
W.E. Nance (Eds.). Alan R. Liss, Inc., New York, pp. 97–103. ing: effects on the olfactory preference of Onychomys torridus
Jones, D. T., and Reed, R. R. (1989). Golf: an olfactory neuron- and Peromyscus leucopus. Behav. Biol. 19:255–260.
specific-G protein involved inodorant signal transduction. McCormick, J. P., Sommerville, B. A., Eggert, F., Wobst, B., and
Science 244:790–795. Broom, D. M. (1995). In Chemical Signals in Vertebrates VII,
Juel-Nielsen, N. (1966). Individual and environment, a psychi- R. Apfelbach, D. Muller-Schwarze (Eds.). Pergamon Press,
atric-psychological investigation of monozygotic twins reared Oxford, pp. 539–541.
apart. Psychiatr. Neurol. Scand., Monograph Supplement 183. McGue, M. (1992). When assessing twin concordance, use the
Kalmus, H. (1955). The discrimination by the nose of the dog of probandwise not the pairwise rate. Schiz. Bull. 18:171–176.
individual human odours and in particular the odours of twins. McGue, M., and Bouchard, T. J., Jr. (1984). Adjustment of twin
Animal Behav. 3:25–31. data for the effects of age and sex. Behav. Genet. 14:325–343.
Kline, J. P., Blackhart, G. C., Woodward, K. M., Williams, S. R., McGue, M., and Bouchard, T. J., Jr. (1998). Genetic and envi-
and Schwartz, G.E. (2000). Anterior electroencephalographic ronmental influences on human behavioral differences. Ann.
asymmetry changes in elderly women in response to a pleas- Rev. Neurosci. 21:1–24.
ant and an unpleasant odor. Biol. Psychiatry 52: 241–250. Mombaerts, P. (1999). Seven-transmembrance proteins as odor-
Kodis, M. (1998). Love Scents. Dutton, New York. ant and chemosensory receptors. Science 286:707–711.
Kohl, J. V., and Francoeur, R. T. (1995). The Scent of Eros: Murphy, C., Morgan, C. D., Geisler, M. W., Covington, J. W.,
Mysteries of Odor in Human Sexuality. Continuum Madowitz, M. D., Nordin, S., and Polich, J. M. (2000).
Publication Group, New York. Olfactory event-related potentials and aging: normative data.
Kopala, L. C., Good, K. P., Torrey, E.F., and Honer, W.G. (1998). Int. J. Psychophysiol. 36:133–145.
Olfactory function in monozygotic twins discordant for schiz- Naessen, R. (1971). An inquiry on the morphological character-
ophrenia. Am. J. Psychiatry 155:134–136. istics and possible changes with age in the olfactory region of
Koski, M. A., Dixon, L. K., and Fahrion, N. (1977). Olfactory- man. Acta Otolaryngol. 71:49–62.
mediated choice behavior in mice: developmental and genetic Nance, W. E., Bodurtha, J., Eaves, L. J., Hewitt, J., Maes, H.,
aspects. Behav. Biol. 19:324–332. Segrest, J., Meyer, J., Neale, M., and Schieken, R. (1998).
LaBuda, M. C., Svikis, D. S., and Pickens, R. W. (1997). Twin Models for the longitudinal genetic analysis of same-age
closeness and co-twin risk for substance use disorders: twins: applications to HDL cholesterol. Twin Res. 1:3–8.
Assessing the impact of the equal environment assumption. Neale, M. C., and Cardon, L. R. (1992). Methodology for Genetic
Psychiatr. Res. 70:155–164. Studies of Twins and Families. Kluwer Academic Publishers
Laurent, G. (1999). A systems perspective on early olfactory cod- B. V., Dordrecht.
ing. Science 286:723–728. Newman, H. H., Freeman, F. N., and Holzinger, K. J. (1937).
Lenington, S., and Egid, K. (1985). Female discrimination of male Twins, A Study of Heredity and Environment. University of
odors correlated with male genotype at the T locus: a response Chicago Press, Chicago.
to T-locus or H-2 locus variability? Behav. Genet. 15:53–67. Nicolaides, N. (1974). Skin lipids, their biochemical uniqueness.
Lilly, M., and Carlson, J. (1990). smellblind: a gene required for Science 186:19–27.
drosophila olfaction. Genetics 124:293–302. Ober, C., Weitkamp, L. R., Cox, N., Dytch, H., Kostyu, D., and
Lord, T., and Kasprzak, M. (1989). Identification of self through Elias, S. (1997). HLA and mate choice in humans. Am.
olfaction. Percept. Motor Skills 69:219–224. J. Hum. Genet. 3:497–504.
Lykken, D. T. (1978). The diagnosis of zygosity in twins. Behav. Pfennig, D. W., and Sherman, P. W. (1995). Kin recognition. Sci.
Genet. 8:437–463. Am. 272:98–103.
Lykken, D. T., Bouchard, T. J., Jr., McGue, M., and Tellegen, A. Plomin, R.(1990). Nature and Nurture: An Introduction to
(1990). The Minnesota Twin Family Registry: some initial Behavioral Genetics. Brooks/Cole Publishing Company,
findings. Acta Genet. Med. Gemellol. 39:35–70. Pacific Grove, CA.
Genetics of Olfactory Perception 343

Plomin, R., DeFries, J. C., Craig, I. W., and McGuffin, P. (2002). Segal, N. L. (1990). The importance of twin studies for individual
Behavioral genetics in the postgenomic era. APA Books: differences research. J. Counsel. Dev. 68:612–622.
Washington, D.C. Segal, N. L. (1997). Same-age unrelated siblings: a unique test of
Plomin, R., DeFries., J. C., McClearn, G. E., and McGuffin, P. within-family environmental influences on IQ similarity.
(1997). Behavioral Genetics, 4th ed. Worth Publishers, New J. Ed. Psychol. 89:381–390.
York. Segal, N. L. (1999a). Entwined Lives: Twins and What They Tell
Porter, R. H. (1987). Kin recognition: functions and mediating Us About Human Behavior. Dutton, New York.
mechanisms. In Sociobiology and Psychology: Ideas, Issues Segal, N. L. (1999b). Silent partners: twins with selective
and Applications, C. Crawford, M. Smith and D. Krebs mutism. Twin Res. 22:235–239.
(Eds.). Lawrence Erlbaum Assoc., Hillsdale, NJ, pp. 175–203. Segal, N. L. (1999c). Kin recognition by olfactory cues: What can
Porter, R. H. (1998–1999). Olfaction and human kin recognition. twins tell us? Twin Res. 2:293–298.
Genetica 104:259–163. Segal, N. L. (2000). Virtual twins: new findings on within-family
Porter, R. H., and Moore, J. D. (1981). Human kin recognition by environmental influences on intelligence. J. Ed. Psychol. 92,
olfactory cues. Physiol. Behav. 27:493–495. 442–448.
Porter, R. H., and Schall, B. (1995). Olfaction and development Segal, N. L., and Topolski, T. D. (1995). The genetics of olfactory
of social preferences in neonatal organisms. In Handbook of perception. In Handbook of Clinical Olfaction and Gustation.
Clinical Olfaction and Gustation, R.L. Doty (Ed.). Marcel R. L. Doty (Ed.). Marcel Dekker, Inc., New York, pp. 323–343.
Dekker, Inc., New York, pp. 299–321. Segal, N. L., Grimes-Hillman, M., and Topolski, T. D.(1995a).
Porter, R. H., Cernoch, J. M., and McLaughlin, F. J. (1983). Twin study of genetic relatedness and odor similarity. Aroma-
Maternal recognition of neonates through olfactory cues. chol. Rev. 4:2, 10–12.
Physiol. Behav. 30:151–154. Segal, N. L., Topolski, T. D., Wilson, S. M., Brown, K. W., and
Porter, R. H., Cernoch, J. M., and Balogh, R. D. (1985). Odor Araki, L.(1995b). Twin analysis of odor identification and
signatures and kin recognition. Physiol. Behavi. 34: perception. Physiol. Behav. 57:605–609.
445–448. Sherman, P. W., and Holmes, W. G. (1985). Kin recognition:
Porter, R. H., Balogh, R. D., Cernoch, J. M., and Franchi, C.
issues and evidence. In Experimental Behavioral Ecology and
(1986). Recognition of kin through characteristic body odors.
Sociobiology. In Memoriam Karl von Frisch 1886–1982, B.
Chem. Senses 11:389–395.
Holldobler and M. Lindauer (Eds.). Sinauer, Sutherland, pp.
Porter, R. H., Levy, F., Poindron, P. Litterio, M., Schaal, B., and
437–460.
Beyer, C. (1991). Individual olfactory signatures as major
Shields, J. (1962). Monozygotic Twins: Brought Up Apart and
determinants of early maternal discrimination in sheep. Dev.
Together. Oxford University Press, London.
Psychobiol. 24:151–158.
Siddiqi, O. (1991). Olfaction in drosophila. In Chem. Senses, Vol. 3,
Pourtier, L., and Sicard, G. (1990). Comparison of the sensitivity
Genetics of Perception and Communications, C. J. Wysocki and
of C57BL/6J and AKR/J mice to airborne molecules of isova-
M. R. Kare (Eds.). Marcel Dekker, Inc., New York, pp. 79–96.
leric acid and amyl acetate. Behav. Genet. 20:499–509.
Prescott, C. A., Johnson, R. C., and McArdle, J. J. (1999). Taylor, B. A. (1991). Current methods in mouse genetics.
Chorion type as a possible influence on the results and inter- In Chem. Senses, Vol. 3, Genetics of Perception and
pretation of twin study data. Twin Res. 2:244–249. Communications, C. J. Wysocki and M. R. Kare (Eds.).
Pruzan, A., and Bush, G. (1977). Genotypic differences in larval Marcel Dekker, Inc., New York, pp. 1–11.
olfactory discrimination in two Drosophila melanogaster Topolski, T. D. (1993). A twin study of olfactory perception. M.
strains. Behav. Genet. 7:457–464. A. thesis, California State University, Fullerton.
Richards, B., Skoletsky, J., Shuber, A. P., Balfour, R., Stern, R. Van Abeelen, J. H. F., and Crusio, W. E. (1985). Zinc-induced
C., Dorkin, H. L., Parad, R. B., Witt, D., and Klinger, K. W. peripheral anosmia and mouse exploratory behavior: A quan-
(1993). Multiplex PCR amplification from the CFTR gene titative-genetic analysis. Behav. Genet. 15:583.
using DNA prepared from buccal brushes/swabs. Hum. Mol. Wallace P. (1977). Individual discrimination of humans by odor.
Genet. 2:159–163. Physiol. Behav. 19:577–579.
Romeyer, A., Porter, R. H., Poindron, P., Orgeur, P., Chesné, P., Ward, C. D., Hess, W. A., and Calne, D. B. (1983). Olfactory
and Poulin, N. (1993). Recognition of dizygotic and mono- impairment in Parkinson’s disease. Neurology 33:943–946.
zygotic twin lambs by ewes. Behaviour 127:119–139. Wedekind, C., and Furi, S. (1997). Body odour preferences in
Russell, M . J., Mendelson, T., and Peeke, H. V. S. (1983). Mother’s men and women; do they aim for specific MHC combinations
identification of their infant’s odors. Ethol. Sociobiol. 4: 29–31. or simply heterozygosity? Proc. R. Soc. Lond. B Biol Sci. 264:
Schleidt, M. (1980). Personal odor and non-verbal communica- 1471–1779.
tion. Ethol. Sociobiol. 1:225–231. Wedekind, C., Seebeck, T., Bettens, F., and Paepke, A. J. (1995).
Schleidt, M., Neumann, P., and Morishita, H. (1988). Pleasure and MHC-dependent mate preferences in humans. Proc. R. Soc.
disgust, memories and associations of pleasant and unpleasant Lond. B Biol. Sci. 260:245–249.
odours in Germany and Japan. Chem. Senses 13:279–293. Wells, P. (1987). Kin recognition in humans. In Kin recognition
Segal, N. L. (1986). MZ or DZ? Not even their hairdresser knows in animals, D. J. C. Fletcher and C. D. Michener (Eds.).
for sure. J. Foren. Sci. 31:10–11. Wiley, New York, pp. 395–415.
344 Segal and Topolski

Werner, D. I., Baker, E. M., Gonzales, E., and Sosa, I. R. (1987). Wysocki, C. J., Whitney, G. and Tucker, D. (1977). Specific anos-
Kinship recognition and grouping in hatchling green iguanas. mia in the laboratory mouse. Behav. Genet. 7:171–188.
Behav. Ecol. Sociobiol. 21:83–89. Wysocki, C. J., Nyby, J., and Whitney, G. (1978). Conditioned
Whissell-Buechy, D., and Amoore, J. E. (1973). Odour-blindness taste aversions: genotype by olfactory bulbectomy interac-
to musk: simple recessive inheritance. Nature 242:271–273. tions. Behav. Genet. 8:119.
Wilson, R. S. (1979). Analysis of longitudinal twin data. Acta Wszolek, Z. K., and Markopoulou, K. (1998). Olfactory dys-
Genet Med. Gemellol. 28:93–105. function in Parkinson’s disease. Clin. Neurosci. 5:94–101.
Wysocki, C. J., and Beauchamp, K. (1984). Ability to smell Yamazaki, K., Beauchamp, G. K., Matsuzaki, O., Bard, J.,
androstenone is genetically determined. Proc. Natl. Acad. Sci. Thomas, L., and Boyse, E. A. (1987). Participation of the X
81:4899–4902. and Y chromosomes in the individual chemosensory identity
Wysocki, C. J., and Beauchamp, G. K. (1991). Individual dif- of mice according to genotype. Behav. Genet. 17:644.
ferences in human olfaction. In Chem. Senses, Vol. 3, Yousem, D. M., Maldjian, J. A., Siddiqi, F., Hummel, T., Alsop,
Genetics of Perception and Communications, C. J. Wysocki D. C., Geckle, R. J., Bilker, W. B., and Doty, R. L. (1999).
and M. R. Kare (Eds.). Marcel Dekker, Inc., New York, pp. Gender effects on odor-stimulated functional magnetic reso-
353–373. nance imaging. Brain Res. 13:480–487.
17

Mammalian Pheromones: Fact or Fantasy?

Richard L. Doty
University of Pennsylvania, Philadelphia, Pennsylvania, U.S.A.

I. INTRODUCTION in a more comprehensive paper to appear in the primary

literature (Doty, 2003).†
The focus of this chapter is on a putative class of stimuli
commonly termed “pheromones.” Pheromones are said to
II. HISTORY OF THE PHEROMONE TERM
differ from other chemical stimuli in having been evolved
to transfer specific information among conspecifics crit-
In 1932, the entomologist Bethe distinguished between
ical for sexual, agonistic, and other forms of social behav-
hormones secreted within the body (“endohormones”) and
ior, as well as for altering the reproductive processes of
hormones excreted outside the body (“ectohormones”),
the recipient (e.g., the age of puberty, the timing of estrus,
dividing the latter chemicals into those with intraspecific
and ova implantation). Although nearly a half-century
and those with interspecific effects (termed homoiohor-
has elapsed since the pheromone term was coined, few
mones and alloiohormones, respectively) (Bethe, 1932). A
substances considered to be pheromones have been
quarter century later, Karlson and Lüscher (1959) replaced
chemically identified and there is still a legitimate
the term homoiohormone with the term pheromone,
question as to the utility of the pheromone concept in
defining pheromones as “substances which are secreted to
describing or explaining chemically mediated social
the outside by an individual of the same species, in which
behaviors and endocrine responses of vertebrates. How is
they release a specific reaction, for example, a definite
one to identify a pheromone? Are there generally accepted
behavior or a developmental process.” These authors
criteria for making such an identification? Does evoking
distinguished between pheromones acting via olfaction
the pheromone concept add to our understanding of the
and those acting via oral or ingestive routes, with the
behaviors or endocrine responses of interest? The goal of
former producing immediate “releasing” responses (e.g.,
this chapter is to explore these issues. Largely because of
initiating and guiding the flight of the male silk worm
space limitations, this review is not inclusive. A broader
moth, Bombyx mori, to the female) and the latter produ-
set of examples, as well as an exploration of the question
cing delayed endocrine or reproductive responses (e.g., the
as to whether humans possess pheromones,* is presented
caste-determining and reproduction-inhibiting substances

*The reader is referred to a number of recent critiques of studies

of menstrual synchrony and the concept of human pheromones †Many of the general principles described here could be applied
(Arden and Dye, 1998; Schank, 2000, 2001; Strassmann, 1999; to nonmammalian species as well, including, in some cases, a
Whitten, 1999; Wilson, 1992; Wöhrmannj-Repenning, 2000). number of invertebrates.

345
346 Doty

of many social insects). Although numerous concerns and although three phenomena seemed to be candidates for
qualifications subsequently arose regarding the usefulness primer pheromonal mediation in mammals: the blockage
of this term within insects and other arthropods, especially of implantation in recently mated mice as a result of
given their remarkable capability to learn and respond to a exposure to unfamiliar males or their urine (Bruce, 1960a),
wide range of odors, the pheromone term nonetheless was the tendency towards anestrous or pseudopregnancy in
found useful in describing many chemically mediated grouped female mice separated from males or their odors
social and endocrine responses in invertebrates. (Lee and Boot, 1955), and the release from anestrous
The first use of the term pheromone in describing among grouped mice induced by noncastrate male mice or
mammalian—indeed vertebrate—behavior was in the their odors (Whitten, 1956). It was not until the late 1960s
1960s (Parkes and Bruce, 1961; Whitten, 1966; Wilson or early 1970s, however, that claims for the chemical
and Bossert, 1963). In an influential review appearing isolation of any type of mammalian pheromone were
in Science, Parkes and Bruce (1961) reiterated Bethe’s made. The initial pheromones reported to be isolated were
general dichotomy and noted that some “chemical all of the releasing type, and included (1) substances
messengers” act within an individual (e.g., hormones and within the vaginal secretions of rhesus monkeys that were
“other excitatory substances,” such as CO2), whereas said to elicit copulatory behaviors in males (Curtis et al.,
others (i.e., “pheromones”) act between individuals via 1971; Keverne and Michael, 1971; Michael and Keverne,
ingestion, absorption, or sensory receptors. These writers 1968, 1970a,b; Michael et al., 1971), (2) agents within the
concluded that “Endocrinology has flowered magnifi- tarsal scent glands of male black-tailed deer that elicited
cently in the last 40 years; exocrinology is now about to licking by females (Brownlee et al., 1969; Müller-
blossom.” The generalized pheromone concept was further Schwarze, 1971; Müller-Schwarze et al., 1974), (3) a
popularized by the entomologist E. O. Wilson in a 1963 material from the midventral scent gland of Mongolian
Scientific American article on the topic (Wilson, 1963). In gerbils that received investigation from other gerbils
this paper, Wilson explicitly set the tone for conceptualiz- (Thiessen et al., 1974), and (4) two steroids from the
ing the nature of both priming and releasing mammalian submaxillary salivary glands of boars that lowered the
pheromones. In the case of mammalian releasing threshold for pressure-induced lordosis in female pigs
pheromones, he focused on musks and noted: (namely, 5-androsten-16-en-3-one and its related
alcohol) (Melrose et al., 1971).
Pheromones that produce a simple releaser effect—a
single specific response mediated directly by the cen-
tral nervous system—are widespread in the animal III. DEFINITIONS AND REDEFINTIONS OF
kingdom and serve a great many functions....Although PHEROMONES
two of the six—the mammalian scents muskone and
civetone—have been known for some 40 years and are While Karlson and Lüscher (1959) provided clear-cut cri-
generally assumed to serve a sexual function, their teria for establishing whether a chemical is a pheromone,
exact role has never been rigorously established by it is instructive to see how dictionaries, textbooks, and
experiments with living animals. In fact, mammals scientists working in the field of animal and human odor
seem to employ musklike compounds, alone or in communication have defined pheromones and the degree
combination with other substances, to serve several to which such definitions are congruent with one another
functions: to mark territories, to assist in territorial and with Karlson and Lüscher’s original definition. For a
defense and to identify the sexes.* term to have scientific meaning, it must be defined in a
Up to that time, no reports of the chemical identification consistent and operational manner, as exemplified by such
of putative mammalian pheromones had appeared, terms as hormones, neurotransmitters, and quarks. As will
be seen in this section, there is considerable variation as to
what authoritative sources view as pheromones.
*The caption of the table in which the chemical structures of cive-

tone and muskone were shown read as follows: “Six sex

A. Dictionary Definitions
pheromones including the identified sex attractants of four insect
species as well as two mammalian musks generally believed to be
sex attractants. The molecular weight of most sex pheromones The Random House College Dictionary (1975) defines
accounts for their narrow specificity and high potency.” Note that the term pheromone as follows: “n. Biochem. any of a class
Wilson was classifying agents as pheromones without even of hormonal substances secreted by an individual and stimu-
knowing what effects, if any, they have on behavior or endocrine lating a physiological or behavioral response from
function. an individual of the same species [ Gk phér(ein) (to) bear
Mammalian Pheromones 347

-0- (HOR)MONE].” By this definition, pheromones B. Textbook Definitions

are equivalent to hormones and produce physiological or
behavioral responses in the conspecific recipient. Webster’s Unlike dictionaries, textbook definitions seem to be more
New Collegiate Dictionary (1999) and related Webster likely to explictly imply that a pheromone is a type of
dictionaries, on the other hand, do not specifically require olfactory stimulus. As with dictionaries, some texts have a
a pheromone to be a hormone and confine the elicited brief and simple definition of pheromones, whereas others
response to a behavioral one: “n. a chemical substance that provide more elaboration. At one extreme is the simple
is produced by an animal and serves esp. as a stimulus to definition provided by Bear et al.’s basic neuroscience
other individuals of the same species for one or more textbook (1996), where a pheromone is defined as “an
behavioral responses.” By this definition, the animal must olfactory stimulus used for chemical communication
produce the chemical, not simply serve as an intermediary among individuals.” This definition is quite similar to that
in its transport. Dorland’s Illustrated Medical Dictionary of Konrad Lorenz’s student, Eibl-Eibesfeldt, who defines
(1981) similarly uses a less restrictive definition: “a sub- phermones in his classic 1970 textbook, Ethology (Eibl-
stance secreted to the outside of the body by an individual Eibesfeldt, 1970), without the requirement of olfaction
and perceived (as by smell) by a second individual of the being involved: “[Chemical] substances that are effective in
same species, releasing a specific reaction or behavior in intraspecific communication are called pheromones.” In
the percipient.” Again, an endocrine influence is not both of these cases, pheromones are not explicity equated
specifically noted, although perhaps evoking the term to hormones. Groves and Schlesinger, in their Introduction
perception implies mental awareness on the part of the to Biological Psychology (1979), also do not equate
recipient. In contrast, Stedman’s Medical Dictionary pheromones with hormones, although they, like Bear et al.,
(1999) employs a more traditional hormone-based defin- assume pheromones require olfactory mediation.
ition, although (1) perception is again required, (2) a Furthermore, they imply their existence in humans: “Many
pheromone is viewed as a subclass of ectohormones, and animals and insects use odors to mark territories, to attract
(3) the general nature of the response is defined: “A type of mates, to signal sexual receptivity, and for a variety of other
ectohormone secreted by an individual and perceived by a communications. In many instances, the receptors for these
second individual of the same species, thereby producing a chemical communicators are highly specific and respond to
change in the sexual or social behavior of that individual.” remarkably low concentrations of them. These chemical
Oxford University Press’s Dictionary of Science (1999) messengers are termed pheromones and are used by many
indicates that pheromones are externally secreted hor- organisms including human beings . . . .” The hormone-like
mones (i.e., ectohomones), although the door is left open nature of pheromones is emphasized by Campbell in his
in terms of species specificity. Rather specific mention is popular general biology textbook (1996), although he is
made of the nature of the typical chemicals involved and more tentative regarding human pheromones:
their explicit presence in mammals: “pheromone (ectohor-
Pheromones are chemical signals that function much
mone). A chemical substance emitted by an organism into
like hormones, with one important exception; instead
the environment as a specific signal to another organism,
of coordinating the parts of a single animal’s body,
usually of the same species. Pheromones play an important
pheromones are communication signals between
role in the social behavior of certain animals, especially
animals of the same species. Pheromones are often
insects and mammals. They are used to attract mates, to
classified according to their functions as mate attrac-
mark trails, and to promote social cohesion and coordin-
tants, territorial markers, or alarm substances, to name
ation in colonies. Pheromones are usually highly volatile
a few. Pheromones are small, volatile molecules that
organic acids or alcohols and can be effective at minute
disperse easily into the environment and, like hor-
concentrations.”
mones, are active in minute amounts . . . . Compared to
It is noteworthy that none of these definitions restrict the
most animals, humans do not have well developed
application to invertebrates and that, by implication, the term
olfactory senses, but it is interesting to speculate
is to be applied to a broad spectrum of animals, typically
whether we use pheromones to communicate. Some
implying externally secreted hormones. Only one of these
indirect evidence suggests we do—for example, cases
definitions indicates that pheromones are mediated by the
of women living together for several months, as in
olfactory system, albeit nonexclusively, although the need for
dormitories, convents, or prisons, who begin to have
the pheromones to be perceived may imply this implicitly.
synchronous menstrual cycles.
Species specificity seems to be the most common generally
held requirement. It is obvious that among dictionaries there Clearly, the latter definition—similar to that employed by
is considerable variation as to what, in fact, is a pheromone. Oxford University Press’s Dictionary of Science noted in
348 Doty

the previous section—implies that pheromones mediate which, on contact with a member of the same species,
their effects through the olfactory system and are relatively evokes behavioral and/or physiological responses.
small volatile molecules, precluding the possibility that Theoretically, such substances may act not only by
some compounds believed by some to be pheromones, olfaction but also by ingestion. In mammals, however, they
such as musks, may have relatively low volatility and are believed to act mainly via the sense of smell.” Aron
could work through the vomeronasal organ or through (1979) follows the original definition of Karlson and
taste. Although in this case pheromones are not considered Lüscher (1959), but further states, “A sensory stimulus
equivalent to hormones, they “function like hormones.” whose action is prevented by either olfactory bulb removal
Some authors seem to feel that pheromones come pri- or peripheral anosmia should be considered a pheromone
marily from exocrine glands, although they do not exclude in nature until we obtain further information on the com-
the possibility that they may come from—or are equivalent pounds involved.” Izard and Vandenbergh (1982) employ a
to—urine or urine-based chemicals. For example, Dember simple definition of pheromone that maintains the concept
and Jenkins (1970), in their introductory psychology text- of species specificity but is rather indescript: “Pheromones
book, note the following: are chemical messages secreted by one animal that cause a
specific reaction in another individual of the same
One special class of odorants has been of especial
species.” Meredith (1998), moving far afield from trad-
interest recently to biologists and animal psycho-
itional definitions, defines pheromones as “chemicals used
logists. These are odorous substances secreted by an
for mutually beneficial chemosensory communication
animal from special glands and left as an “odor trail”
between members of a species.” This redefinition, which is
wherever the animal goes. Such substances are called
reminiscent of a proposal by Rutowski (1981), has some
pheromones; they are known to be secreted by a wide
attractive attributes, although it lacks specificity and a
variety of species, ranging from ants to gerbils.
number of phenomena traditionally assumed to be mediated
Analogous to pheromones in function, if not exactly
by pheromones may not necessarily provide mutual bene-
like them physiologically, are odorous substances,
fit, which itself is difficult to establish operationally (e.g.,
such as urine, excreted by animals and left at various
the elicitation of agonistic responses that may decrease the
places in the territory over which they travel. One
mating success of one participant, blocking of the preg-
function of pheromones and similar odorants is to
nancy of an earlier male, delaying puberty).
“mark off” an animal’s territory. . . . Another function
In an introduction to a biblography on mammalian
of such odorants is that of a trail-maker. . . .
pheromones, Whitten and Champlin (1972) provided a
state-of-the-art description of what pheromones were
C. Definitions Employed by Scientists Working in the believed to be at that time and added the observation that
Chemical Senses chemicals that are truly pheromones are most likely found
in only one of the sexes in a given species. They also
Martin (1980) defined pheromones as “isolated chemicals suggested that some chemicals are “co-pheromones.” They
shown to be relatively species-specific which elicit a clear stated:
and obvious behavioral or endocrinological function and
which produce effects involving a large degree of genetic [Pheromones are] substances, produced by one
programming, influenced little by experience.” This defin- member of a species, which influence other members
ition, which is reasonably scientific and operational, of the same species. They may be behavioural
embodies the key elements of the original definition of pheromones and evoke rapid behavioural reactions
Karlson and Lüscher (1959), but is atypical of many of the through nervous pathways or they may be primer
definitions used by working scientists. Indeed, it appears pheromones and induce relatively slow endocrine
that most workers, particularly those studying mammals, responses. There will, of course, be some overlap in
have not confined the pheromone term to its implicit, if not this classification because behavioural responses
explicit, meaning, often redefining the term to meet their may eventually follow endocrine changes and some
own specific conceptions. Shorey, in his 1976 book Animal responses such as alarm and stress will involve
Communication by Pheromones, defined pheromones as neuroendocrine pathways and may be intermediate
follows: “Pheromones are either odors or taste substances in time sequence . . . . Identification of an odorous
that are released by organisms into the environment, where substance from a glandular or other secretion does
they serve as messages to others of the same species.” not prove that the substance is a pheromone. If it is
Signoret (1976) noted, “The term ‘pheromone’ has been limited to one sex as for example, muscone in the pod
widely used for any substance produced by an individual of the male musk deer, then the case is considerably
Mammalian Pheromones 349

stronger. If, however, as is reported for civetone it is seem most profitable to restrict the use of this term
produced by both sexes and by related species then [pheromone] to situations where there seems to be a rea-
one must look for another function. Civetone has sonable probability of isolating one or at least a restricted
been used for centuries as a fixative in perfume mixture of compounds that could, in turn, be synthesized
manufacture. For this purpose it intensifies and pro- and whose actions could then be reconfirmed experiment-
longs more subtle and ephemeral components of a ally,” a task that could be quite difficult a priori. Bronson
perfume. Civetone may perform this same function also stressed that any response to a pheromone “should
for the civet and perhaps it could be considered a serve a reasonable biological function in a natural popula-
co-pheromone. tion.” Importantly, he noted, “The probability of a high
Pheromones are important in animal reproduction degree of specificity in pheromones, however, argues
and some evidence has been provided to show that against the widespread use of the more typical androgen
they may also function in humans. If they do func- metabolites as pheromones.”
tion in man, they could produce more regular cycles It is noteworthy that one of the pioneers of the
and thus aid in delineating the safe period. pheromone field, Hilda Bruce, indicated in 1970 that “the
Alternatively if perception of sex odors is dependent meaning of the word ‘pheromone’ has been extended to
on ovarian status this too could be used to define the include chemical communication in a broader sense and in
safe period (Vierling and Rock, 1967). all species.” She stated that “The term now connotes secre-
tions which convey information of many kinds from one
Although Whitten and Champlin maintained a distinction individual to others and evoke specific behavioural and
between the two basic classes of pheromones, with the physiological reactions in the recipients.” Nonetheless, she
responses to a behavioral pheromone being rapid and the operationally differentiated between priming and releasing
responses to a primer pheromone being slow, they argued pheromones. To her, releaser pheromones were to be
that the pathways through which pheromones exerted recognized by “An immediate and reversible response
themselves could be quite variable: [that] operated directly through the central nervous system,
It is probable that most mammalian pheromones will e.g. recognition, or through rapidly acting neurochemical
act through the olfactory system which includes the channels, as exemplified by the milk-ejection reflex
vomeronasal organ (Jacobson, 1811) and the acces- (Cross and Harris, 1952),” and primer pheromones by an
sory olfactory bulbs. It is, however, possible that “exteroceptive response implicating the anterior pituitary
other pathways such as taste may be used or they gland. This type is slow to develop, demanding prolonged
may even be absorbed through the respiratory or stimulation which initiates a chain of physiological
alimentary mucosae. The pheromones may be effects in the recipient.” Interestingly, she added a third
produced in secretions of the genital organs, the skin pheromone class, which she termed “imprinting
glands or occur in the urine, faeces, or the expired pheromones,” to explain modification of later adult behav-
air. These substances should be volatile but we do ior by the presence of chemical stimuli within the suckling
not know what type of sensation they produce in the environment.* She cited two examples of such pheromones:
recipient animal. Nor do we know if primer In mice (Mainardi et al., 1965) and in rats (Marr and
pheromones are perceived as odors. The pheromones Gardner, Jr., 1965), social behaviour in the adult
concerned with reproduction will most likely be may be modified by olfactory experience during the
under endocrine control and the sensitivity to the period of suckling. Female mice reared in the
pheromone may be influenced by gonadal status absence of the father show a loss of discrimination in
(LeMagnen, 1951). sexual selection when adult, and the same deficiency
develops if the olfactory atmosphere of the nest is
In an insightful early review, Bronson (1971) specific-
artificially altered by spraying the parents everyday
ally stated that no mammalian pheromone had yet been
with perfume (Parma violet). Young rats reared from
isolated and expressed the need to use the pheromone term
only in situations where, in fact, it was likely that some
chemical agent could be identified.* He noted, “It would *Although in the text she indicates that “mammalian olfactory
pheromones belong to all three types” and provides a separate
*There is a contradiction in this review regarding the isolation of category both in her outline and in the text for imprinting
pheromones, as he states that “ . . . the reasonably well isolated pheromones, the last sentence of her paragraph above, which has
mammalian pheromone, that of the tarsal gland of black-tailed been omitted for clarity, curiously reads: “This aspect of primer
deer, is apparently a decided mixture of compounds.” pheromone activity [i.e., imprinting] merits further attention.”
350 Doty

birth to about four weeks of age in the olfactorily here, although some workers have suggested that
artificial atmosphere, scented either with Yardley’s vomeronasal responses may be unconscious (Lloyd-
Red Roses cologne or with oil of wintergreen, also Thomas and Keverne, 1982; Meredith, 1991), and several
showed modifications when adult. theorists have expressed the view that pheromones are
unconsciously mediated by the vomeronasal organ
A number of subsequent authors have assumed that (Belluscio et al., 1999; Dulac and Axel, 1998). Recently,
certain steroids are unequivocally mammalian pheromones Savic et al. (2001) incorrectly noted that “The pheromones
based upon their seemingly direct effects on sex-related are, according to the original [i.e., Karlson and Lüscher]
behaviors. Among the most common of such steroids is definition, volatile [italics mine] compounds secreted into
androstenone, a constituent of boar saliva and other the environment (in sweat, urine) by one individual of a
secretions, which produces the “boar taint” of meat from species . . . .” In fact, volatility was not a part of the original
uncastrated boars and lowers the threshold for lordosis in definition. These authors continue, “Also, it has recently
sexually experienced sows. Thus, Pause et al. (1999) note been reported that a putative pheromone receptor gene is
that “Androstenone has been studied extensively for its expressed in human olfactory mucosa (Rodriguez et al.,
pheromonal properties. In the boar, it has been proven to 2000). These data raise the question whether there are
be a pheromone, because it is produced in the boar testes, compounds that via the nasal mucosa activate the human
and the sow responds to its smell with the mating stance hypothalamus in a sex-specific mode. If so, such com-
(lordosis). Evidence for androstenone being a pheromone pounds would fulfill one important criterium [sic] to qual-
in humans comes from an experiment carried out by Kirk- ify as candidates for putative pheromones in humans.” To
Smith and Booth (1980). They sprayed androstenone onto my knowledge, neither sex specificity nor the activation of
a seat in a dentist’s waiting room and observed 840 people the hypothalamus are widely held criteria for pheromones,
for their seat preferences. The authors could show that although, as noted above, Whitten et al. (1972) believed
more women but fewer men used the odorized seat than that sex specificity might add to the proof that a substance
expected by chance.” Clearly, Pause et al.’s criteria for a is a pheromone, and some subsequent workers have
pheromone (origin in testes, influences on choice of employed such specificity in their definitions of
seating in a dental office) differ from those inherent in pheromones.
most pheromone definitions. In a thoughtful article, Johnston (2000) proposed a
Several relatively recent definitions of pheromone are classification scheme that operationally distinguishes
worthy of note. Buck (2000) reiterates the commonly held between pheromonal and nonpheromonal stimuli on the
view of innateness, stating that “pheromones elicit basis of chemical complexity. He suggests that the term
programmed neuroendocrine changes and innate behav- “chemical signal” be used as a generic term for chemical
iors, suggesting a need for a very precise recognition compounds or mixtures released into the environment that
process.” Stern and McClintock (1998), in a paper serve intraspecific behavioral or physiological functions.
purportedly being the first demonstration of a human Johnston reserves the term “pheromone” for chemical
pheromone (which, in fact, was not identified chemically), signals that employ a single compound and the term
added a new requirement for the pheromone definition— “pheromone blend,” which has been employed in some
namely, that pheromones are airborne. These investigators insect studies, for mixtures of a small number of
state: “Pheromones are airborne chemical signals that are compounds that are maximally effective when they occur
released by an individual into the environment and which in precise ratios. He further suggests that the term “mosaic
affect the physiology or behaviour of other members of the signal” or “odor mosaic” be used to refer to mixtures of
same species.” Using this definition, the many substances large numbers of compounds in which many components
that have been previously deemed pheromones found in are important for producing the full effect. Johnston indi-
aquatic environments or transferred into the vomeronasal cates that “This scheme has the advantage of preserving
organ via liquid media would not be considered the traditional use of the terms ‘pheromone’ and
pheromones, nor would secretions ingested by insects and ‘pheromone blend,’ while adding a third important catego-
other forms that alter cast or endocrine state. Moreover, ry to encompass signals that have previously been given
Stern and McClintock indicated that “Here we investigate little attention or regulated to a second-class status.” While
whether humans produce compounds that regulate a spe- indeed Johnston’s terminology does not significantly per-
cific neuroendocrine mechanism in other people without turb the status quo, it is debatable whether it is an advan-
being consciously detected as odours (thereby fulfilling tage to maintain the use of terms that implicitly or explic-
the classic pheromone definition).” It is not clear what itly define the nature of the interaction to most biologists
“classic” definition of pheromone is being referring to as hormone-like. Moreover, his suggested distinctions are
Mammalian Pheromones 351

difficult to apply a priori, and no differentiation is made chemoreceptive organ for the processing of pheromones”
between learned and unlearned responses. Johnston (Takigami et al., 1999). Hence, “the vomeronasal organ has
correctly points out that single molecules are rarely used as attracted increasing attention as it is thought to mediate the
sex attractants even in insects, presumably reflecting the stereotyped behavioral and neuroendocrine responses to
need for species-specific signals that are more easily estab- chemicals commonly known as ‘pheromones’ ” (Liman,
lished by selected ratios of given agents. It is of interest 1996). As articulated by Matsunami and Buck (1997):
that, by Johnston’s criteria, the term pheromone would
Pheromones are intraspecific chemical signals found
rarely be applied even to insects.
throughout the animal kingdom. They regulate popu-
A number of molecular biologists who work in the
lations of animals by inducing innate behaviors and
chemical senses have recently defined the vomeronasal
stereotyped changes in physiology (Karlson et al.,
organ as “the pheromone receptor,” assuming that the
1959; Wilson, 1963; Sorensen, 1996). Pheromones
main olfactory system is relegated to the perception of
can serve as cues for overcrowding, impending
nonpheromones (Belluscio et al., 1999; Buck, 2000; Dulac
danger, reproductive status, gender, or dominance. In
et al., 1998; Tirindelli et al., 1998).* Thus, Dulac and Axel
rodents, a variety of pheromone effects have been
(1995) wrote that:
reported. These include effects on estrus and the
Mammals possess an olfactory system of enormous onset of puberty as well as the induction of mating
discriminatory power. Humans, for example, are and aggressive behaviors (Halpern, 1987; Wysocki
capable of recognizing thousands of discrete odors. and Meredith, 1987; Novotny et al., 1990; Singer,
The perception of odors in humans is often viewed 1991) . . . . The detection of pheromones is mediated
as an aesthetic sense, a sense capable of evoking by the olfactory system. However, sensory neurons
emotion and memory, leading, to measured thoughts that detect pheromones are typically segregated from
and behaviors. Smell, however, is also the primal those that detect volatile odorants (Keverne, 1983;
sense. In most species, odors can elicit innate and Wysocki et al., 1987; Novotny et al., 1990;
stereotyped behaviors that are likely to result from Hildebrand and Shepherd, 1997). In mammals, sen-
the nonconscious perception of odors. These sory neurons in the nasal olfactory epithelium (OE)
different pathways of olfactory sensory processing detect volatile odorants and some pheromones, while
are thought to be mediated by two anatomically those in the accessory olfactory organ, called the
and functionally distinct olfactory sensory organs, vomeronasal organ (VNO), are thought to be spe-
the main olfactory epithelium (MOE) and the cialized to detect pheromones.
vomeronasal organ (VNO).
While it has been known for years that the VNO
This revolutionary approach takes the definition of responds electrophysiologically to chemicals not generally
pheromone to an entirely new level, particularly since the viewed as “pheromones” (e.g., Hatanaka, 1992; Meredith,
receptors of main and vomeronasal systems appear to share 1982,1991), apparently this has only recently been realized
no homology. Hence, according to this perspective, by some investigators (e.g., Sam et al., 2001). This observa-
pheromone detection would seem to have evolved separately tion, however, throws into question the aforementioned
from nonpheromonal odorant detection (Dulac et al., notion that pheromones can be distinguished from
1998). Although other molecular biologists working in this nonpheromones on the basis of whether the VNO is
field have maintained a less bipartite distinction, nonethe- activated, decreasing the alluring clarity of this distinction
less the vomeronasal organ is typically viewed as “a and making the VNO more like the main olfactory system
in reflecting a broader assessment of environmental
agents. Thus, Sam et al. (2001) reported, in a recent Nature
*It should be noted that the effects of the one isolated agent that paper, that “The prevailing view of the mammalian olfac-
seems to best meet traditional definitions of phermones, tory system is that odorants are detected only in the nasal
androstenone, are mediated in the sow through the main olfactory olfactory epithelium, whereas pheromones are generally
system, not the vomeronasal system. Perhaps the claim of the
detected in the vomeronasal organ. Here we show that
vomeronasal organ as the pheromone detector led Hines (1997) to
vomeronasal neurons can actually detect both odorants
provide the following definition of pheromones in a Science maga-
zine perspectives article: “a special subset of olfactory signals [that and pheromones. This suggests that in mammals, as in
are] not perceived consciously. . . . These molecules, often fatty insects, odorous compounds released from plants or other
acids or steroids, are secreted by animals, then detected by other animal species may act as ‘semiochemicals’—signaling
animals, of the same species, where they regulate such basic func- molecules that elicit stereotyped behaviours that are
tions as mating, the timing of the estrous cycle, and aggressiveness.” advantageous to the emitter or to the receiver.”
352 Doty

Whatever the situation with the VNO, it is clear from generalizations is the tendency to think of mammalian
the information reviewed above that the term pheromone communication in terms of simple stimulus-response
means different things to different people and has been systems. For example, it is now relatively common
redefined over and over in an attempt to fit a range of usage to refer to “aggression-promoting” (or “elicit-
chemically mediated behaviors and endocrine responses ing”) and “aggression-inhibiting” pheromones in
into a common mold. While the term itself is intuitive and mice (e.g., Lee and Griffo, 1974; Mugford and
catchy, it clearly has many of the same problems as the Nowell, 1972). The obvious implication of this
term instinct, with which it is closely allied, leading to sig- terminology is the existence of two simple urinary
nificant dangers. As Lehrman pointed out in his classic compounds which unequivocally either release or
1953 paper, “A Critique of Konrad Lorenz’s Theory of inhibit a stereotyped aggressive response. Mam-
Instinctive Behavior,” labeling behaviors as instinctual— malian social behavior simply does not work that way
while perhaps gratifying—produces an either/or dichotomy except at the purely reflexive level.
with significant associated pitfalls. For example, such
Bronson went on to point out that the nervous system of
labeling tends to preclude the need to study developmental
the mouse not only contains many more neurons than that
or experiential factors associated in the fruition of a given
of an insect, but differs significantly in terms of degree of
behavior, oversimplifying even its genetic underpinnings
encephalization, the numbers of associative neurons, and the
(Beach, 1955; Lehrman, 1953).
flexibility afforded to the mediated behaviors. He indicated
that the releasing pheromone concept is a valuable tool for
IV. EARLY CONCERNS WITH THE CONCEPT describing the “often relatively simple, stimulus-response
OF RELEASING PHEROMONES IN systems of such organisms,” but is questionable for mice and
MAMMALS other mammals. He continued:
Most insect pheromones are usually single
As early as the late 1960s, a number of influential biologists
compounds or simple mixtures, typically secreted
voiced concern about the utility of describing chemicals
by restricted glands, and normally evoking stereo-
involved in the social behavior of mammals as “releasing
typed responses even under totally inappropriate
pheromones,” reflecting their awareness that mammalian
circumstances. Thus many of the standard tests for
behavior is not reflexive in the same way as the behaviors of
insect attractants have relied upon copulatory
many invertebrates. In 1968 Bronson suggested that the
behavior in response to scented filter paper,
term “signaling” should replace the term “releaser,” and in
repeated exposures in many cases providing little
1973 Whitten and Champlin suggested that “behavioral”
habituation of the response (Birch, 1974). It is
should serve as the substitute, as employed in Whitten’s def-
difficult to imagine a male mouse attempting
inition described above. Subsequent investigators suggested
copulation with a scented filter paper let alone
replacing the releasing pheromone term with such terms as
doing so repeatedly, and, by extension, it is
“social odors” (Brown, 1979), “homeochemic substances”
exceedingly difficult to apply the simple releaser
(Martin, 1980), or “semiochemicals” (Albone, 1984).
concept to much of mammalian social behavior,
Bronson (1976) clearly articulated the problem in a
whether elicited in part by odors or not.
statement 27 years ago:
Additionally, experience is a profound modifier to
It is perhaps unfortunate that interest in mammalian mammalian social behavior. There have actually
chemical communication blossomed at a time when been relatively few attempts to examine the role of
the study of insect pheromones was already a experience in odor-induced responses in mammals.
sophisticated field of research. Thus Whitten (1966) Where investigated, however, the results usually
introduced the primer-releaser dichotomy to mam- have indicated a potent role for experience. Thus
malian workers and Bronson (1968) amended it only species identification apparently can be easily
slightly by arguing that the term signaling was a more manipulated by odors early in the life of mammals
appropriate modifier than releaser for a nonprimer (e.g., Carter and Marr, 1970; Mainardi et al., 1965;
pheromone, given the variable, experience-oriented Marr and Lilliston, 1969) and adult sexual expe-
behavior of mammals.* The unfortunate side of such rience is a strong determinant of response to sex
odors (e.g., Caroom and Bronson, 1971; Carr
*To the author’s knowledge, Whitten was not the first to introduce et al., 1965, 1966). One wonders at this point
the pheromone term to mammals, as indicated earlier in this whether the pheromone concept, so useful in insect
review. behavior and physiology, should be bastardized to
Mammalian Pheromones 353

the point where it is used to cover situations in without being concerned about such matters or identifying
mammalian behavior where usually complex odors the chemicals involved, such distinctions are academic.
evoke highly variable responses which are easily In an attempt to address the issue of learning, Müller-
modified by experience. Schwarze (1977) sought to add still another class of
pheromones—so-called “informer” pheromones. The gen-
In the same book in which Bronson voiced his
esis of this suggestion was that some chemical signals in
concerns about the concept of releasing pheromones,
mammals are “ . . . stored in the memory and can be recalled
Beauchamp et al. (1976) cast doubt on the utility of the
later in a variety of contexts.” Unlike the terms releasing
pheromone concept altogether, noting that “. . . we ques-
and priming, this term did not catch on, however, and it
tion the current usefulness of the term ‘pheromone’ in
was apparently not used even by Müller-Schwarze in
describing the influences of biological secretions and
subsequent papers. However, his point—that learning is
excretions upon mammalian reproductive behaviors and
important in odor communication—is a fundamental
suggest that the uncritical use of this term has led to a
one that throws into question the general validity of the
number of misconceptions in the interpretation and
traditionally conceived pheromone concept, as will be
conduct of mammalian behavioral research.” They listed
pointed out later in this chapter.
what the implicit or explicit criteria for a chemical to be
Sorensen and Stacey (1999), while maintaining the
termed a pheromone seemed to be up to that time—a list-
belief that natural selection is focused on specific chem-
ing that, at first glance, provided an operational basis for
icals, nonetheless point out the practical problems with
determining whether a chemical was, in fact, a
the traditional pheromone concept in terrestrial verte-
pheromone. Their list of criteria was as follows:
brates:
Species specificity Presumably, chemical stimuli are predisposed to
A well-defined behavioral or endocrinological function as social signals because they are ubiqui-
function tous and discriminated with great sensitivity and
A large degree of genetic programming specificity. Because of this specificity, organisms
The involvement of only one or at most a relatively detect only a portion of the myriad compounds
few compounds surrounding them. Pheromonal systems have proven
Uniqueness of the isolated compounds or small set of challenging to study because it has been difficult to
compounds in producing the behavioral or predict which of the many chemicals released by
endocrinological response organisms might have pheromonal activity.
Terrestrial vertebrates appear to have evolved,
These authors examined all extant claims of isolation of
repeatedly and independently, a variety of sex
mammalian pheromones up to that time and pointed out
pheromones with no clear common precursors. Few
that none tested or met even half of these criteria. Indeed,
of their sex pheromones have been identified and no
only one criterion, that of assumed chemical simplicity for
general theoretical framework has emerged to
the isolated product, was met by all of the isolated
systematically address either the diversity of sex
substances. They concluded: “It would appear to us that
pheromone systems or the evolutionary processes
the labeling of a compound as a pheromone, when it has
that might have created them.
not been demonstrated to meet a well-defined set of
operational criteria, is problematic if the pheromone term
is to have any meaning beyond that of being synonymous V. CHEMOSENSORY LEARNING USUALLY
with a ‘chemical’.” OVERRIDES CHEMOSENSORY GENETICS
However, even these criteria, for the most part, are dif-
ficult to employ. Thus, how many species need to be tested As noted in the previous sections, most adherents to the
before species specificity can be assumed? What is meant pheromone concept view pheromones as species-specific
by a “large degree” of genetic programming—i.e., how hormone-like agents that differ fundamentally from the
can degrees of genetic programming be operationally plethora of other environmental chemicals that are sensed
determined? How many chemicals must be tested before by organisms. In general, pheromones are viewed as being
the uniqueness of the isolated stimuli can, in fact, be little influenced by learning and are more or less genetic-
ascertained? Answers to such basic questions are needed, ally fixed. However, even if some odors are, in fact, inher-
however, if one is operationally establish what, in fact, is ently more preferable to animals than others, does one
a pheromone. Of course, if authors simply label behaviors have to infer that an innate pheromone is the basis for this
or endocrine responses as being pheromonally mediated preference or is essential for the initiation of exploration or
354 Doty

sniffing of a scent?* In contrast to this perspective, a case tasks, so long as olfactory, rather than visual, stimuli were
can be made that learning is involved in establishing the employed. Thus, Jennings and Keffer (1969) and Nigrosh
meaning of most odorous chemicals to mammals. et al. (1975) found excellent interproblem transfer over a
Moreover, learning seems to be a key component in such series of two-odor discrimination problems in the rat, such
classic examples of “primer” pheromones as the strange that errorless performance on subsequent reversals was
male odor pregnancy block. As described in this section, commonly attained (Slotnick, 2000). These and other
mammals appear to learn such important information as studies demonstrated that rats can learn to identify and
their mother’s odor, the odor of their species, the odor of discriminate among large numbers of odors and can
their offspring, the odor of their social group (e.g., deme), remember whether they were reinforced or not reinforced
the odor of a fecund sexual partner, the odor of the domin- for each of these odors in a test series (Slotnick et al.,
ant or subordinant conspecific, and the odor of familiar 1991). Slotnick (2002) states:
conspecifics, distinguishing them from strangers. In these
Functional studies have overcome many of the
cases, can the stimuli involved be considered pheromones?
technical difficulties of controlling vapor stimuli and
Evidence that rodents exhibit a remarkable capability to
demonstrate that, with odor cues, rats display highly
learn and employ odors in complex “cognitive tasks”
efficient learning rivaling that of primates. In short,
appeared in the early 1970s. Prior to these studies, the
the evidence indicates that rats can ‘think with their
comparative intelligence of a number of animals was
noses’ and have the neural machinery to do so. This
assessed using “learning sets” or “learning to learn” para-
evidence, combined with advances in the molecular
digms based upon visual responses, one of several
biology of olfaction (Mombaerts et al., 1996), has
approaches for assessing animal cognition.† In general, the
resulted in a renaissance in research on olfaction and
relative performance of various species on such tasks was
to the surprising and occasionally controversial sug-
found to mirror the “phylogenetic scale” or scala naturae
gestion that the rodent olfactory system could serve
so often employed in studies of comparative anatomy, i.e.,
as a model for neurobiological studies of cognition
primates  other mammals  birds  reptiles  amphib-
(Reid and Morris, 1993; Slotnick, 1994).
ians  fish  insects (Bitterman, 1965; for critique of this
concept, see Hodos and Campbell, 1969). However, when In light of such observations, a strong argument can be
the sensory specializations of different forms were taken made that, just as mammalian photoreceptors have not
into account, it became apparent that rodents such as the evolved specifically for detecting mothers, fathers, or
rat could perform essentially as well as primates on such Ferrari automobiles, so too mammalian olfactory receptors
have not evolved specifically to detect the odors of mothers,
fathers, or the exhaust smells of Ferrari automobiles.
This lack of specificity extends to the smells of individual
*Of course, it does not necessarily follow from this argument that conspecifics, even though their long-term identification,
preprogrammed specific responses cannot exist, or that such like the visual detection of Ferraris, can occur as long as
responses would necessarily be unmodifiable by experience or
learning at some point intervenes. Thus, while the olfactory
other factors. The point being made is that one should not assume
system, like the visual system, can provide specific infor-
that such preprogrammed responses are the norm.
† In a learning set paradigm, the subject is given a series of mation about the physical nature of the environment, the
discrimination problems to solve, the first of which may require specificity is largely dependent or interdependent
many trials to learn. Over a series of sessions, however, the abil- upon experience. It is noteworthy that olfactory detection
ity to solve new problems dramatically improves, suggesting that thresholds of rats for perfluorocarbons—agents never
the animal has learned “rules” or “concepts” underlying the task. encountered during their phylogeny—are at the same level
For example, in a three-item reversal task, a monkey may be of magnitude as thresholds for many organic chemicals
given two circles and a square as the first problem, being rein- presumably encountered during ancestral evolution,
forced for choosing the square. After attaining high performance reiterating the notion that evolution has not focused on
on this task, the monkey is then given two squares and a circle, the detection of specific chemicals, but on the provision of
with reinforcement given for choosing the circle. While initially
a sensory system that is flexible and sensitive to the
the tendency of the animal is to choose the square, at some point
detection of even de novo agents (Marshall et al., 1981).
he chooses the circle. The next reversal would be like the first,
usually with a different spatial configuation to control for spatial
preferences, and so on. In this case, the animal is acquiring the A. Prenatal Learning
concept of “oddity” and at some point learns to choose the odd
stimulus on a new set without ever having been reinforced for any It is important to be aware that the influences of experience
of the stimuli employed in the new task. on establishing the social significance of some odors can
Mammalian Pheromones 355

occur even before birth. Thus, the olfactory system of many B. Neonatal Learning
mammals, including humans, is functional in utero and
intrauterine learning can manifest itself in postpartum life. In a manner conceivably analogous to the visual and audi-
Evidence for prenatal function includes observations that tory imprinting processes of birds, many odors are learned
premature human infants exhibit discriminative responses during early periods of the developing mammal.* While
among low concentrations of odorants presented to them mere exposure to odors can produce learning in some
(Pihet et al., 1997; Sarnat, 1978), and rat fetuses transferred instances, odor preferences are reinforced in the suckling
from the abdominal cavity of their mothers into saline with- environment by the warmth of the mother and the licking
out interruption of the maternal blood supply exhibit of her pup, even independently of milk reinforcement.
increased activity, altered heart rate, and facial wiping Indeed, simply pairing an artificial odor with a warm
responses to odorants (Smotherman and Robinson, 1987, surface or with tactile stroking is sufficient to establish
1990). Evidence that experiences with odors before birth conditioned olfactory preferences in rat pups (Alberts and
can influence behaviors later in life comes from many Brunjes, 1978; Alberts and May, 1984; Dominguez et al.,
sources. For example, human fetuses learn odors related to 1999). In general, neonates detect and find attractive the
their pregnant mother’s diet (Schaal et al., 2000), and such odorous components of amniotic fluid, particularly those
intrauterine learning can be reinforced in the nursing of their own mothers (e.g., Hepper, 1987; Schaal et al.,
situation, where flavors ingested by the mother can be 1998; Teicher and Blass, 1977), likely reflecting intra-
transmitted via the mother’s milk (Galef and Henderson, uterine experience, as noted above, and possibly explain-
1972; Galef and Sherry, 1973; Mennella and Beauchamp, ing their attraction to nipple-related odors and other
1991a,b, 1996). Rat pups exposed to citral in utero maternal secretions around the time of birth (Schaal et al.,
attach, postpartum, to washed citral-scented nipples and not 1994). Such attraction aids in guidance to the nipple and
to normal unwashed nipples (Pedersen and Blass, 1982). alters their general motor activity and arousal (for review,
Offspring of pregnant rats receiving an infusion of an see Porter and Schaal, 2000). Parturient females of many
odorant into the amniotic fluid and made sick by lithium species engage in self-grooming that deposits saliva and
chloride injected into the mother avoid postnatally the odor amniotic fluid on their ventra and nipple regions, and these
to which they had been exposed (Smotherman, 1982; secretions carry, in large part, the chemical message that
Stickrod et al., 1982). If no toxic agent is administered to directs the first suckling episode of the newborn (Teicher
the mother, then a postnatal preference for the prenatally and Blass, 1976; Teicher et al., 1977). Even human infants,
exposed odorant may appear in later life, particularly if that who preferentially exhibit head orientations towards
same odorant is present in the early perinatal period maternal breast odors within the first few minutes of life,
(Nishiazaka et al., 1993; Pedersen et al., 1983). In some are likely influenced by prior experience with amniotic
cases postnatal odor preferences can be induced by simply fluid. As noted by Porter and Winberg (1999), “the role of
feeding the pregnant mother the target odorant (Hepper, maternal olfactory signals in the mediation of early breast-
1988; Schaal et al., 1995). For example, rabbit pups whose feeding is functionally analogous to that of nipple-search
mothers were fed aromatic juniper berries during pregnancy
(such berries are part of the rabbits’ natural diet) prefer
juniper at weaning, even if raised after birth by a foster
mother fed standard laboratory food (Bilko et al., 1994). *It should be emphasized that odors are not necessarily unique,
This preference for juniper, which is not seen in controls, is in that early learning via all of the senses occurs in mammals and
still present months later, even without additional juniper many other forms, including many insects and birds (Beach and
experience (Hudson and Distel, 1999). The magnitude of Jaynes, 1954). Cross-fostering can even influence visual social
the summated electrical potential at the surface of the olfac- preferences in some mammals. For example, male sheep and
tory epithelium (the electro-olfactogram) in response to goats cross-fostered to the opposite species, unlike their normally
juniper, obtained from the epithelia of sacrified rabbits reared counterparts, show a nearly exclusive preference for faces
whose mothers were fed the juniper leaves, is larger than of females of their foster species; cross-fostered females also
exhibit, relative to normals, an increased preference for the cross-
that of rabbits whose mothers were not fed the leaves, sug-
fostered species’ female faces, although their preferences are
gesting neural changes at the level of the olfactory receptors
more-or-less equally divided among the faces of the genetic and
(Hudson et al., 1999). This observation is in accord with cross-fostered species (Kendrick et al., 2001). These investigators
other studies demonstrating exposure-induced alterations in concluded, “these results provide strong evidence that social and
peripheral olfactory physiology (Coopersmith and Leon, sexual preferences are primarily determined by maternal and
1984, 1986; Wang et al., 1993; Youngentob and Kent, social rather than genetic influences even in mammals and that
1995). effects are stronger and more durable in males than in females.”
356 Doty

pheromone as described in nonhuman mammals. To some While much has been made of odor-related dissortative
extent, the chemical profile of breast secretions overlaps mating preferences in mice related to genes of the major
with that of amniotic fluid. Therefore, early postnatal histocompatability complex (MHC) (Beauchamp et al.,
attraction to odors associated with the nipple/areola may 1985; Yamazaki and Boyse, 1985; Yamazaki et al., 1976,
reflect prenatal exposure and familarization.” Some foods 1998), genes at other loci are also involved in establishing
ingested by the mother markedly influence the smell of the cues employed in individual identity, and there is strong
amniotic fluid and an infant’s attraction to it (Mennella evidence that cross-fostering and diet override or attenuate
et al., 1995), as well as influence the flavor of the mother’s such genetic predispositions (Burger et al., 2001; Penn and
milk. Interestingly, amniotic fluid has other important Potts, 1998). For example, Yamazaki et al. (1988) demon-
properties for both the mother and offspring. In the rat, for strated in mice whose genetic differences were only within
example, the ingestion of amniotic fluid potentiates opiate- the MHC complex, that the preference for B6-H-2k
related analgesia (Kristal et al., 1986). males to mate with B6-J-2b females, and the preference for
A number of mammalian cross-fostering studies, B6-J-2b males to mate with B6-H-2k females, was reversed
including ones performed on ungulates (Müller-Schwarze when the mouse pups were cross-fostered by the opposite
and Müller-Schwarze, 1971), find that it is the odor of the H-2 haplotype. Singh et al. (1990) found that the urine of
species or subspecies of the cross-fostered parent, not that individual male rats born by Cesarian section and reared in
of the genetic parent, that largely establishes subsequent a germ-free environment were not discriminable by Lister
social and mating preferences. In some species or hooded rats using a habituation-dishabitutation test.
instances, such effects may be more marked in the female However, when these rats were moved to a non–germ-free
than in the male, as would be predicted from the female’s conventional animal house, such urine was discriminable
greater investment in proper mate selection (Doty, 1974). after recolonization with commensal flora, suggesting that
In a pioneering study, Mainardi (1963) found that estrous commensal bacteria are involved in the production of
female housemice of the Mus musculus domesticus unique individual odor of the urine of even MHC-congenic
subspecies, reared by both parents since weaning, rats. This observation is in accord with the fact that
preferred the odors of M. m. domesticus to those of M. m. individual odor-mediated identity seems to be largely
bactrianus, whereas analogous females reared only by influenced by diet in number of mammals, as described in
their mothers, in the absence of adult males, showed no more detail later in this chapter (for review, see Schellinck
differential preferences between these two subspecies. and Brown, 1999).
Quadagno and Banks (1970) found that female housemice As alluded to earlier in this section, learned responsive-
(Mus musculus) cross-fostered to pigmy mice (Baiomys ness of rodents and a number of other mammals to odor-
taylori) preferred the odor of pigmy mice to housemice in ants present during the preweaning period is not confined
adult preference tests. One source of the odor involved to so-called “natural,” “biological,” or “animal” odors, but
may be the preputial glands, since female mice reared with can extend to “artifical” odors as well (Cornwell, 1976;
mothers whose preputial glands have been excised prefer Galef and Kaner, 1980; Gregory and Bishop, 1975; Janus,
females without preputial glands in adulthood (Hayashi, 1993). Thus, any of a number of odorants (e.g., ethyl
1979). McCarty and Southwick (1977) found decreased benzoate, acetophenone, methyl salicyclate, Yardley’s Red
conspecific odor preferences in grasshopper mice Roses cologne, citral, cinnamon, cumin, and various
(Onychomys torridus) and white-footed mice (Peromyscus perfumes, such as Parma Violet perfume), placed in the
leucopus) that were cross-fostered to the other species’ rearing environment, can take on the same meaning as
dams; cross-fostered Peromyscus males actually switched natural biological stimuli and alter subsequent preferences
their species preference to Onychonmys. Both Mus muscu- for scented situations or scented conspecifics in later life
lus and Peromyscus maniculatus mice reared in the (Alleva et al., 1981; Carter, 1972; Carter et al., 1970;
presence of both species’ odors prove to be more success- Fillion and Blass, 1986a; Janus, 1989, 1993). For example,
ful in heterospecific agonistic encounters than conspecific rats reared on lemon-scented bedding from birth to
counterparts reared only with their own species’ odors weaning acquire a seemingly permanent preference for nest-
(Stark and Hazlett, 1972). Interestingly, gerbils (Meriones ing in lemon-scented surroundings (Rodriguez-Echandia
unguiculatus) reared with parents whose midventral et al., 1982). Adult rats previously reared with mothers and
sebaceous glands were surgically removed show lower littermates odorized by artificial odors prefer conspecifics
preferences for such odors in adulthood and engage in less odorized with such odors and are less responsive sexually
social behavior with opposite-sexed conspecifics than ger- to unodorized conspecifics (Fillion and Blass, 1986b; Marr
bils raised with parents having such glands (Blum et al., et al., 1965, 1969). The same is true for mice. Thus,
1975). Mainardi et al. (1965) reared male and female house mice
Mammalian Pheromones 357

(SWM/Mai strain) with perfumed or nonperfumed parents function to denote other individuals, as well as environ-
until the age of 21 days, when they were weaned and mental objects. Mice (Mus musculus), rats (Rattus norvegi-
separated into like-sex groups. When tested in estrus at 8 cus), hamsters (Mesocricetus auratus), Belding’s ground
months of age, 48.2% of the perfume-reared females spent squirrels (Spermophilus beldingi), and guinea pigs (Cavius
more than 60% of their time with a perfumed male, porcellus), species extensively tested on this point, can
compared to 21.4% of the normally reared females. Of the remember dozens, if not hundreds, of odors of individual
normally reared females, 67.8% spent more than 60% of conspecifics encountered in adulthood, apparently in some
the test time with the nonscented males, compared to cases for a lifetime, even after having only encountered
27.6% of the perfume-reared females. Such findings them on a single brief occasion (Beauchamp and
reiterate the fact that the olfactory system has been Wellington, 1984; Brown, 1988; Johnston, 1993; Mateo
designed, like the visual system, to recognize, remember, and Johnston, 2000; Mossman and Drickamer, 1996).*
and prefer variant salient features of the general odorifer- The association of odors with objects is obvious even in
ous environment. humans, where odors are nearly always identified with
their source, as is apparent from their names (e.g., rose,
C. Adult Learning candy, pizza, gasoline, fish, licorice, chocolate, leather,
mint, seashore, perfume, etc.). Porter et al. (1983) applied
Even though there appear to be early “critical periods” artificial odors (i.e., musk oil, oil of clove, lemon/lime, and
in many mammals when exposure to odorants is particu- cherry) every other day on four occasions after weaning to
larly effective in producing long-term alterations in odor four groups (n
4/group) of spiny mice (Acomys cahirin-
preferences and odor-mediated social behaviors (Galef jus) composed of two siblings and two nonsiblings. Each
et al., 1980), experience with odors in adulthood can also group received a single odor. Later preference tests con-
significantly alter later odor-mediated behavioral and sex- ducted with odorized kin and nonkin strangers found that
ual preferences, explaining many so-called releasing the mice interacted, for all practical purposes, only with
pheromone effects. For example, rats, dogs, and a number mice having the same odor to which they had been
of other mammals develop preferences, or markedly exposed as adolescents, regardless of genetic relatedness.
increase preexisting subtle preferences, for estrous Adult female mice of the SEC1Re/J strain (SEC), when
over diestrous female odor as a result of adult sexual exposed to the odor and sounds of C57BL6/J (C57) males
experience (e.g., Carr et al., 1965; Doty and Dunbar, for 7 days during the immediate postweaning period,
1974; Le Magnen, 1951; Lydell and Doty, 1972). Sexual exhibit a preference for bedding odors from C57 males
preferences, however, can also be influenced by adult over that males of their own strain when tested in estrus at
experiences with artificial odors placed on sexually recep- 120 days of age (Albonetti and D’udine, 1986). This pref-
tive females. In one series of studies, for example, male erence was independent of whether they were fostered by
rats were allowed to mate with estrous females who had a SEC or C57 dam until weaning.
almond extract applied to their neck and anogenital The influence of odors learned in adulthood extends to
region. In subsequent tests with almond-scented and non- the social communication of what foods are safe to eat. For
scented receptive females, these males ejaculated first and example, rats alter their preferences for foods eaten by
more frequently with the almond-scented females (Kippin other rats with whom they socialize (for review, see Doty,
and Pfaus, 2001a). Interestingly, such males tended to 1986), a phenomenon that is accentuated in protein-
preferentially mount the almond-scented females immed- deprived rats (Beck and Galef, 1989; Galef et al., 1991). In
iately prior to ejaculation, implying that the preference
may be conditioned to the ejaculatory event, per se
(Kippin and Pfaus, 2001b; Kippin et al., 2001), although *The importance of individual recognition for chordate evolution
the presence of the female during a postejaculatory period is obvious, as without such recognition natural selection could
is important for producing the conditioning (Kippin et al., not operate. As noted by (Clark, 1982), “individual recognition
2001). The effectiveness of such conditioning appears to and its effects on behavior may be the single major basis for
structuring mammalian and avian social relations. Its importance
relate to the degree of satiety and the rat’s motivational
is easily appreciated in dominance relations, mother-offspring
state (Kippin et al., 2001). No such preference was
recognition, kin-directed behavior, and mate recognition, as
observed in males having no experience with almond odor familiar examples.” Individuals from every single terrestrial
or trained in the presence of almond odor in an unpaired mammalian that has ever been tested to date, including primates
or randomly paired manner. and humans, are able to recognize conspecific individuals by
Mammals generally have the ability to rapidly acquire odor (Brown and MacDonald, 1985; Carr et al., 1976; Doty,
and maintain memories for many types of odors, and odors 1986; Goyens et al., 1975).
358 Doty

one study it was found that rats that encounter conspecifics ences. For example, female mice from social groups with
that have eaten banana-flavored food pellets are more neighboring social groups show a stronger relative prefer-
likely to enter a T-maze arm known to lead to such pellets ence for the scent marks of dominant males from their
(Galef et al., 1997). In another study, it was shown that a own groups than do females from groups with no neigh-
rat will exhibit an enhanced preference for a cinnamon- or bors (Heise and Hurst, 1994).
cocoa-flavored food recently eaten by a healthy rat placed In summary, it is obvious that many behaviors said by
in their cage for a brief period, a preference that does not some to be mediated by pheromones are likely learned, and
generalize to similarly scented nest materials or nest boxes that distinctions between inherent and learned responses
(Galef et al., 1994). Moreover, rats made sick after eating are very difficult to make, particularly since postpartum
a series of novel foods in succession are less likely to responses can be tempered by prior intrauterine learning.
exhibit a conditioned aversion to those foods whose odors While it is a truism that learning is a distinguishing feature
were previously experienced on the breath of healthy of many organisms, it is accentuated in mammals.
conspecifics to which they were briefly exposed (Galef, Chemically mediated behaviors or responses would seem
1986). This is not simply a function of greater familiarity not to be more primitive or less influenced by learning than
with the novel food odor. Rats that have learned an aver- behaviors or responses mediated by nonchemical stimuli. Is
sion to a flavored fluid and are allowed to briefly interact their any advantage in terming such chemically influenced
with healthy rats that have drunk that fluid without adverse social behaviors as being mediated by pheromones?
consequence increase their intake of the averted fluid
relative to controls that have had no such social interaction
VI. COGNITIVE PROCESSES: MICE, RATS, AND
(Galef et al., 1997).
OTHER MAMMALS ARE NOT INSECTS
Even after the meaning or health consequence of an
odor becomes manifest to a rodent, however, its behavior
As pointed out by Bronson, mammals have much more
does not need to be reflexive or invariant towards that spe-
complex nervous systems than insects, and a number
cific stimulus. For example, a male mouse typically
of their physiological responses to external stimuli—
responds to the scent marks of another male by depositing
including overt behaviors and internal hormonal changes—
urine on or around such marks. However, if this mouse
often are tempered, guided, and in some cases determined
ends up on the losing end of a fight with the other mouse,
by such intervening factors as stress or memory of past
it will become subordinate to it and will cease such scent
experiences.* As noted in Sec. V, most mammals have early
marking even when it is housed in close proximity to the
developmental periods during which odorants significantly
marked area, its circulating testosterone is maintained at a
impact upon later social behavior and endocrine responses,
high level by a silastic implant, and the other male is
possibly in a manner analogous to visual and auditory
removed from the situation (Maruniak et al., 1977).
imprinting in birds. Moreover, in many forms ideation
Hence, in this case the mouse’s scent marking is not an
divorced to some degree from concurrent external stimuli
invariant response to a pheromone or even to endogenous
can likely alter hormone levels and overt behaviors. The
levels of testosterone (which are usually highly correlated
anatomical associations of the olfactory system provide a
with scent marking behavior and glandular secretions), as
rich substrate for the cognitive mediation of complex odor-
learning has intervened (see Sec. VII. A). Social and con-
related phenomena. As noted by Slotnick (2002):
textual factors also influence responses to purported
pheromones. For example, dodecyl propionate, a putative Recent discoveries have revealed that the olfactory
maternal pheromone isolated from rat preputial glands, is system is less simple and less primitive than is
said to attract adult rats and to play a key role in regulat- generally assumed: olfactory impulses have fairly
ing postpartum maternal licking of the anogenital area of direct inputs to brain regions implicated in complex
pups, a behavior critical for initiation of defecation functions, including limbic structures and the
(Brouette-Lahlou et al., 1992). When placed on rice, how- prefrontal cortex . . . . The first link between olfaction
ever, dodecyl propionate deters licking and ingestive and cognition was the finding that cells in the olfac-
responses (Arnould et al., 1994), again pointing out that tory cortex project to the segment of the thalamic
context can establish the nature of responses to odorants.*
Social factors also influence female mouse odor prefer-
*The differentiation between mammals and invertebrates is one
of degree, not of kind. As alluded to earlier, many invertebrates
*The human analogy may be the presence of rose oil on break- learn and remember and exhibit behaviors that are influenced by
fast cereal in the morning. early experiences with odors and foodstuffs.
Mammalian Pheromones 359

mediodorsal nucleus that connects to the orbital the agents involved or have even sought to determine
frontal cortex. Subsequent reports confirmed the whether the stimulus effects are species specific.
existence of an ‘olfactory thalamocortical circuit,’ Importantly, in perhaps most of the cases where such iden-
and delineated olfactory connections to the amyg- tification has been made, the biological activity of the
dala, entorhinal cortex and hypothalamus. isolated stimulus does not faithfully mimic that noted for
the parent agents. The task of identifying sets of chemicals
Thus, it would seem that an inherent problem with the
likely inducing the activity is theoretically daunting, as in
traditionally conceived pheromone concept is that, by
many cases there are hundreds or even thousands of poten-
implication and definition, it obfuscates or even excludes
tial combinations of key components to be assessed—
from consideration the possibility that cognitive processes
components that themselves are beholden to specific
can be involved in the mediation of behavioral and
dietary and bacterial factors. One is reminded of visual
endocrine responses. This is in spite of the fact that few
research with pigeons, where complex visual stimuli used
persons who own pets or have dealt with nonhuman
in discrimination tasks are fractionated or dissociated in an
mammals deny that they appear to have ideation or
effort to find key components that maintain the discrimin-
thoughts, such as occur during episodes of dreaming
ation. While elements are found that seem to serve this
(Rasmussen et al., 1993). While there is evidence that
function, they are often idiosyncratic, suggesting that there
secretion of lutenizing hormone and testosterone can occur
need be no universal “essence” to the complex visual stim-
in rats in anticipation of sexual activity (Graham and
uli that maintain the discriminative behavior (Herrnstein,
Desjardins, 1980), cognitive/endocrine processes are more
1984; Reynolds, 1961).
easily demonstrated in humans. For example, testosterone
In this section I present five examples of behaviors said
titer increases in winners and decreases in losers of tennis,
to be caused by “releasing” pheromones and two examples
wrestling, debates, chess matches, and various games, a
of endocrine responses attributed to “priming”
number of which are largely intellectual enterprises
pheromones. In these cases it would seem that the
(Booth et al., 1989; Cavaggioni and Mucignat-Caretta,
pheromone concept is found wanting in explaining their
2000; Elias, 1981; Gonzalez-Bono et al., 1999; Mazur
quintessence. Thus, many putative “releasing” or “signal-
et al., 1992; McCaul et al., 1992; Rejeski et al., 1989; Suay
ing” pheromones seem to be dependent, largely if not
et al., 1999). Erotic dreams, expectation of sexual encoun-
entirely, upon learning during at least one stage of develop-
ters, and vicarious identification of male sports fans with
ment. Most “primer” pheromones, for which the
a winning team can increase testosterone levels
pheromone model seems at first glance to be a better
(Anonymous, 1970; Carani et al., 1990; Hellhammer et al.,
fit, seem to be much more complex than appears on the
1985). Identification with a losing team, on the other
surface and often fail to meet key criteria inherent in most
hand, can decrease testosterone levels (Bernhardt et al.,
definitions of pheromones (Doty, 2003).
1998). Chronic stress and mental concerns or conflicts are
known to alter a number of human and animal endocrine
functions (Beaumont, 1982; Fenster et al., 1999), and it is A. Releasing Pheromones
well established that stress, induced by a number of differ-
ent procedures, activates dopaminergic, noradrenergic, 1. The Maternal Pheromone of the Rat
GABAergic, and endorphinergic systems (D’Amato and
Infant rats of the Wistar and Sprague-Dawley strains have
Cabib, 1987, 1990; Nakagawa et al., 1981; Yoneda et al.,
been reported, in a series of innovative studies, to be
1983). Hence, in some cases, the influences of odors on
attracted to a maternal odor, labeled the “maternal
hormones may well be mediated via the elicitation of a
pheromone,” during the second through the fourth week of
mental image or memory, and the degree of effect may
age (Holinka and Carlson, 1976; Leidahl and Moltz, 1975;
depend upon the prior experience or lack of experience of
Leon and Moltz, 1971, 1972; Leon et al., 1972; Nyakas
the organism with the involved stimulus. Obviously, more
and Endröczi, 1970). A similar phenomenon has been
research is needed on this very fascinating topic.
reported for housemice (Breen and Leshner, 1977). The
main source of the stimulus appears to be the cecotrophe
VII. CASE STUDIES OF PURPORTED portion of the maternal anal excreta. Like many sources of
MAMMALIAN PHEROMONES odor from biological secretions, the attractive agent has
been shown to be dependent upon diet and cecal bacterial
Literally thousands of studies attribute the chemically populations, as well as the production of bile and prolactin
mediated behaviors or endocrine events they describe to (Leon, 1974, 1975, 1978; Moltz and Leidahl, 1977). Thus,
pheromones, yet few have attempted to chemically identify antibiotics that eliminate the bacterial flora eliminate the
360 Doty

attractiveness of the excretia. Pups raised with mothers on pup was placed in the apparatus the mother became
a particular diet are attracted to the odor of mothers eating extremely agitated, possibly as a result of ultrasonic
that specific diet (Leon, 1975). calling by the pup (e.g., Allin and Banks, 1972;
Although responses of rat pups to the “maternal Smith, 1975) and in moving about emitted auditory
pheromone” have been observed in several laboratories, cues. This was in sharp contrast to the non-lactating
the salience of the effect is often difficult to demonstrate or females, which tended to settle down in the goalbox
in some cases is likely even nonexistent, depending upon and go to sleep. Any maternal attraction mediated by
the rat strain evaluated, leading one to question its gener- olfactory cues was thus confounded by auditory cues
ality (Clegg and Williams, 1983; Clegg et al., 1983; Galef, and maternal retrieving. Mobile live stimuli can
1981; Kendrick, 1975). For example, Galef (1981) reports therefore not properly be used to assess maternal
that, in his laboratory, the attractiveness of a dam’s feces pheromone phenomena.
was relatively weak and that the attraction differed
In the second study of the series, Clegg and Williams
between animals fed two slightly different formulations of
tested excrement collected from lactating vs. virgin
Purina Laboratory Chow No. 5001. According to the manu-
females. The amount of the fecal material was equated
facturer, the two formulae differed subtly in only a few
across groups (3 g), since lactating females typically pro-
constituents. Whether such attraction to cecotrophe should
duce more excrement than nonlactating females, thereby
be viewed as “pheromonal” seems questionable, as even
potentially producing a confounding factor of stimulus
the scientists who first described this phenomenon readily
quantity.* No evidence for a meaningful preference for the
acknowledge that the responses are learned (Leon, 1975)
excrement of the lactating females was observed in either
and that the effectiveness of the material in attracting pups
the Wistar or Sprague-Dawley rats, although this was not
depends upon subtle dietary factors (Coopersmith et al.,
the case for the PVG/C strain rats. Of 151 PVG/C rats, 79
1984; Leon, 1975). Coopersmith et al. (1984) state that
chose the goalbox containing the maternal excrement, 47
“since there is no single maternal odor, the pups must
chose the side containing the virgin female excrement, and
become attracted to the odor that they will approach
25 remained in the arena, not moving into either compart-
through postnatal experience.” Such learning is likely
ment. While the number of PVG/C animals choosing the
facilitated by tactile stimulation, such as the licking that a
side of the maternal excrement was significantly higher
mother rat gives to her pups (Pedersen et al., 1982).
than the number who chose the side of the virgin female
In a well-controlled series of studies, Clegg et al. (1983)
excrement, the large number of pups who chose the
were unable to demonstrate a robust “maternal pheromone”
virginal excrement, in combination with those who did not
effect in several strains of rats. These investigators
make any choice at all, led these authors to the conclusion
approached their work under the assumption that a
that the effect, at best, is weak.
pheromone is an entity unique from that of simply an odor.
The remaining studies examined such factors as air flow
In the first of eight experiments, the age of the pups (18 days
rates in the testing apparatus, maternal deprivation times,
vs. 24 days), duration of maternal deprivation (3 hours vs. 18
olfactory function in the pups, different types of laboratory
hours), and type of test stimuli (e.g., own mother vs. a virgin
food, and the effects of presenting cecal contents
female; own mother vs. an adult male) were assessed.
obtained from sacificed dams in an effort to observe a
No evidence of a statistically meaningful preference was
robust phenomena. The obtained data suggested that
noted for the pup’s own lactating mother over the other
(1) pups learn to identify olfactory cues associated with the
stimulus animals. Unlike earlier studies, these authors even-
diet and (2) cecal contents produced an approach behavior
tually employed anesthetized rats as stimulus objects for the
relative to an empty goal box. In relation to the latter
following reason:
observation, the authors write, “so once more there is a
clear suggestion of olfactory cues influencing behavior,
This procedure [i.e., use of unanesthetized female
but again without the control over that behavior that would
stimulus animals] was initially followed in the
be expected on the maternal pheromone hypothesis.”
present study. Eighteen PVG/C hooded rats aged 18
Clegg and Williams (1983) reiterate the importance of
days were given the choice of their own mother as
Galef’s observations and note the following about
opposed to a virgin female of the same age and
pheromones in their discussion:
strain. Pups underwent 3 hr of pretest maternal
deprivation. All of the pups entered the goal
compartment containing their own mother. But it *The authors make the assumption that gross stimulus quantity is
was apparent that factors other than a pheromonal the involved factor, even though stimulus quality may not be
agent had helped to bring this about. As soon as a correlated with quantity.
Mammalian Pheromones 361

The implications of Galef’s discovery are indeed far control of the food.* Nonetheless, bulbectomized males
reaching, and it highlights the problem of deciding are much less prone to initiate agonistic encounters than
what does or does not constitute a pheromone. The normal mice (Ropartz, 1968). Strain differences in such
pheromone concept is employed to go beyond the behavior have also been noted (Kessler et al., 1975).†
notion that animals use olfactory cues. It carries The fact that the putative aggression-eliciting sub-
the implication that the stimulus controls behavior stance is found in both the preputial glands and in the
by eliciting a stereotyped and reliable response urine, as well as in the urine from female mice injected
within any particular species (Karlson and Lüscher, with testosterone early in life, suggests either that the
1959). If certain strains within a species fail to same “pheromone” is present in multiple biological fluids
demonstrate the appropriate response, or if members or that there are several such pheromones. Assuming the
of the species do so under highly specific conditions, former for the purpose of exposition, if the chemical
then can the effect be called “pheromonal”? signal is truly a “pheromone,” one would expect that
responses to it would be relatively invariant, perhaps
being expressed in proportion to the amount of substance
2. The Aggression-Eliciting Pheromone of the Male
being excreted by these various sources. However, a
Mouse
number of observations suggest that this is not the case.
It has been suggested that male mice produce a First, exposing a male mouse for one hour daily for
pheromone that elicits aggressive behavior from other 10 days to soiled woodshavings from the cage of a strange
males (Heyser et al., 1992; Mugford and Nowell, male mouse eliminates, for the most part, agonistic
1970,1971; Mugford et al., 1972). The laboratory behaviors directed towards that male in subsequent
demonstration of the aggression-eliciting pheromone has encounters. Such exposure also mitigates, to a much less-
typically employed paired encounters where males— er extent, such behaviors directed towards other mature
often trained fighters—engage in agonistic behavior male conspecifics (implying that familiarity with any con-
towards a castrate stimulus male or female to which specific male odor in this context decreases agonistic ten-
urine or preputial secretions from the target male animal dencies) (Kimelman and Lubow, 1974).† If such behaviors
are applied (e.g., Lee and Brake, 1971). The aggression- were being mediated by a pheromone, then one would
eliciting property of the stimulus is testosterone-depen- have to postulate that each mouse has a different aggres-
dent, since (1) male urine or preputial sebum from sion-eliciting pheromone and that familiarity to one can,
castrated males leads to fewer or briefer attacks when to some degree, influence the mouse’s behavior towards
placed on the stimulus animals than those elicited by another. A seemingly more parsimonious explanation
stimuli from non-castrate males (Heyser et al., 1992; would be that the behavior is dependent upon the relative
Mugford et al., 1971), (2) a dose-response relation exists degree of olfactory strangeness or novelty, a concept with
between the amount of testosterone injected into urine empirical support (Alberts and Galef, 1973; Mackintosh
donor animals and the attack behavior elicited (Mugford and Grant, 1966). Second, mice reared alone or isolated
et al., 1972), and (3) the efficacy of male urine is from other mice, when placed in subsequent encounters,
decreased by the administration of the anti-androgen,
cyproterone acetate (Jones and Nowell, 1974; Nowell
and Wouters, 1973). In general, female mice receive no
or few attacks in this situation, and deodorizing males *One must be cautious in interpreting the effects of olfactory
reduces aggressive responses directed towards them (Lee bulbectomy, since this operation has influences beyond simply
et al., 1971). Interestingly, female mice administered altering the perception of environmental chemicals. Thus, it also
testosterone early in life also elicit agonistic responses, effects vomeronasal organ input and, depending upon the species,
implying that this hormone somehow alters the female can enhance, depress, or have no influence on tonic gonadotropin
physiology such that the purported aggression-eliciting secretion (see Pieper and Newman, 1999, for a review).
agent is produced (Lee and Griffo, 1973). Olfaction is †Jemiolo et al. (1991) have chemically assessed the constituents
of preputal gland secretion, and have found that two sesquiter-
implicated in mediating the response, although other
penic compounds, E, E,--farnesene and E--farnesene, elicit
factors seem to be involved as well. For example, Rowe
increased investigatory behavior relative to water or bladder urine
and Edwards (1971) reported that male mice whose in sexually naïve and experienced mice.
olfactory bulbs had been removed failed to attack cas- †Conceivably the strangeness of the odor contributes to the
trates in a paired encounter, but when food deprived for elicitation of aggressive behaviors. Dixon (1982) found that
48 hours and paired with such mice in the presence of a increased aggression directed to mice injected with diazepam
food pellet, 70% of the bulbectomized mice fought for likely reflected changes in the odor of their urine.
362 Doty

are more likely to initiate social/investigatory behaviors Evans (1979) claimed that there is no logical way of
and are more prone to attack a conspecific male than mice arguing for the existence of a “pheromone” whose
reared in groups (Levine et al., 1965).* Both the duration function is to release aggressive behaviour from other
and the nature or timing of the exposure can determine the mice. This, to all intents and purposes, would involve
magnitude of this phenomenon and in some cases may a signal meaning “attack me please” and would result
even result in its reversal (i.e., less aggression from in a loss of fitness on the part of the signalling
isolated mice) (Cairns and Nakelski, 1971; King, 1957; individual. The results of the present studies support
King and Gurney, 1954). For example, mice who are this position and suggest that the odours really act as
housed alone for 24 days in the presence of the bedding of personal labels signalling “I am a threat to you”. For
another mouse and then tested in agonistic encounters example, the odour of preputial sebum of dihydrotes-
with either that mouse or a mouse whose odor is unfamil- terone-treated castrates probably identifies the donor
iar to them, exhibit more aggression towards the mouse as “an unfamiliar, mature, sexually active, territorial
whose odor is familiar to them than to the mouse whose male” to non-habituated conspecifics who respond
odor is novel, demonstrating the lability and complexity accordingly. In this way there is no loss of fitness.
of such behavior (Corridi et al., 1993; Telle, 1966). Third,
male mice cohabitating with females are more aggressive Novotny et al. (1985) have isolated two testosterone-
towards strange males than male mice who are not living dependent volatiles, which they view as pheromones, from
with females. Male mice who in the past had cohabited the urine of male mice [2-(sec-butyl)thiazoline and 2, 3-
with a female for a short period of time exhibit intermedi- dehydro-exo-brevicomin] that increase aggressive responses
ate levels of aggressive behavior (Goyens and Noirot, among males in standard behavioral bioassays. These
1975). Presumably this reflects the testosterone titer of the agents do not work when added to water, but are synergis-
aggressor, although psychological factors may also be tic when added to castrated male urine. Thus, outside of
involved. Fourth, commonly a strange mouse, or a mouse the mileau of a urine background, they seem ineffective.
introduced into a group after having been removed from Even if one accepts them as key components of a secretion
the group, is first investigated by the group members, and that signifies strangeness or elicits agonistic activity in
in most cases the attack behavior of the group depends conspecific male mice, can they be viewed as pheromones
upon the behavior of the intruder (Cairns and Nakelski, in light of the aforementioned issues? As noted by Alberts
1970). Often the degree of investigation directed towards et al. (1973) in rats, familiarity is a key component in the
the stranger depends upon group size. Thus, residents of initiation of aggressive responses: “the response of wild
small groups of rats (up to ~20 members) attack strangers Norway rats to conspecifics is determined by a multitude
and drive them away, whereas residents of large groups of stimuli perceived via several sensory modalities.
(80–100 members) do not, implying that when a local Response to a conspecific as such (amicable and sexual
group of rats becomes large enough for individual mem- behavior) can occur in the absence of olfactory inputs. On
bers to be anonymous, it may no longer be closed (Carr the other hand, the initiation of aggression would appear
et al., 1976; Telle, 1966). Fifth, in naturalistic settings to be dependent on olfactory stimuli arising from an unfa-
(e.g., demes), aggression is minimal among group mem- miliar individual. Both the duration and direction of
bers, even though at least one male member of the group aggressive behavior is further modified by the behavior
—the dominant male—undoubtedly has high testosterone [e.g., movement] of target animals.”
titers and, hence, would be expected to produce the A human analogy may be of value in describing the
aggression-eliciting pheromone. This lack of aggression complexities of mammalian agonistic encounters. If the
presumably depends on, in part, the development of a reader was to discover a large male stranger inside his or
social dominance heirarchy and the scent-marking behav- her living room in the morning upon awakening (assuming
iors of the dominant individual(s) that familiarize the an all-night party had not been going on), it is likely that
other deme members with their odor. Sixth, the existence aggressive responses, whether verbal or physical, would be
of aggression-eliciting pheromones makes little sense directed towards the intruder. However, the nature of such
from an evolutionary perspective. Brain et al. (1987) notes responses would likely depend upon the stranger’s size,
the following: age, and apparent intentions. If, for example, the intruder
was holding a shotgun, the homeowner’s agonistic
response would likely be more calculated. On the other
*Considerable species differences exist regarding the effects of hand, if the stranger was an old man in tattered rags and the
isolation on agonistic behaviors. In guinea pigs, for example, long- weather outside was freezing, sympathy or the desire to
term isolation results in more docility, not less (Sachser, 1986). help may overshadow any agonistic behavior on the part of
Mammalian Pheromones 363

the resident. In this situation, one does not postulate the monkeys was called into question by Goldfoot et al.
presence of an aggression-promoting visuomone or (1976). These authors carefully examined Michael et al.’s
audiomone, only a situation that requires a response or the original data, noting that the male’s responsiveness varied
mitigation of various response alternatives once an under- considerably from male to male and in some cases tended
standing of the situation has been established. to depend upon particular female, irrespective of an odor
cue. They pointed out that in Michael et al.’s seminal work
(Michael and Keverne, 1970b), a baseline consisting of as
3. Copulin—The Rhesus Monkey Vaginal Pheromone
many as 60 pretests over 80 days was based on only two
In a series of highly publicized studies, Richard Michael male subjects, and suggested that the application of the
and coworkers reported that vaginal secretions from female odorants after extinction of mounting could be explained
rhesus monkeys contain pheromonal substances that elicit on the basis of disinhibition, resulting in a resumption of
copulatory behavior from males. Specifically, they found mounting. Particularly damning to the vaginal pheromone
that males would perform a bar-press behavior to gain concept was evidence, some from Michael et al.’s own
access to and to copulate with estrogen-treated females, studies, that removal of olfactory bulbs has no influence on
presumably on the basis of olfactory cues (Michael and male rhesus monkey mating behavior, implying that
Keverne, 1968, 1970b). In other studies, often using the pheromones—at least ones whose effects are mediated via
same subjects who had been trained in the aforementioned the olfactory pathways—are neither necessary nor suffi-
bar press paradigm, they applied ether or water extracts of cient for such behavior (e.g., Goldfoot et al., 1978;
vaginal secretions from estrogen-treated ovariectomized Michael and Keverne, 1968). Further problems for
monkeys to the sexual skin of untreated ovariectomized Michael et al.’s claims were Goldfoot et al.’s extensive
females. During subsequent tests, application of the behavioral and analytical studies (Goldfoot et al., 1976). In
extracts resulted in an “immediate and marked stimulation contrast to Michael et al.’s work, relatively large numbers
of the sexual activity” of the male subjects (Keverne et al., of subjects were employed. For example, in the behavioral
1971). These authors concluded “male sex-attractant studies, a total of 19 adult male and 27 adult spayed female
pheromones, with powerful behavioral effects, are present rhesus monkeys were used. Unlike in Michael et al.’s
in ether extracts of estrogen-stimulated vaginal secretions.” work, the donor females had not been recently paired with
Chemical analysis of the estrous vaginal secretions males in most of the test situations, and, thus, their vaginal
resulted in the isolation of a series of volatile short-chain secretions were not contaminated by male ejaculate.
aliphatic acids—acetic, proprionic, isobutyric, n-butryic, Despite careful quantitative assessment of a range of
and isovaleric— that these authors claimed to be the active sexual behaviors (approach, genital inspection, contact,
pheromonal substance (Bonsall and Michael, 1971; mounts, intromissions, ejaculations) under a variety of
Michael and Keverne, 1970 Michael et al., 1971). A estrogen regimens and behavioral test conditions, no statis-
mixture of these agents in specific proportions, which they tically significant differences between vaginal lavage and
termed Copulin, was prepared, found to be active, and control treatments could be found, although a slight ten-
patented in several countries for employment in human per- dency for ejaculate-contaminated secretions to increase
fumes, suggesting that Michael et al. assumed the some behaviors was noted. Importantly, the amount and
pheromone was not species specific, thereby violating one relative proportions of aliphatic acids found in the vaginal
of the most common criteria for an agent being a secretions differed markedly from those found by Michael
pheromone. The involvement of aliphatic acids in primate and associates. For example, in contrast to Copulin,
sexual attraction is counterintuitive, however, since vaginal Goldfoot et al. did not detect any isovaleric acid in the
aliphatic acids primarily appear during the luteal phase of secretions even after 29 days of estrogen treatment.
the cycle, rather than during the time of optimal fertility Goldfoot et al. concluded that “comparison of our results
(Goldfoot et al., 1976; Michael et al., 1972). In humans, to those from other laboratories [i.e., Michael et al.’s] sug-
aliphatic acids are largely dependent upon bacterial fer- gests that the mechanism involved in positive effects may
mentation of glycogen, which is highest during the luteal depend upon associative learning or upon extinction or dis-
phase (Gregoire et al., 1973). In any event, Copulin’s effec- inhibition of sexual interest.”
tiveness in altering human sexual behavior was subsequent-
ly found to be nil (Cowley and Brooksbank, 1991; Morris
4. Dimethyl Disulfide—The Hamster Vaginal Attraction
and Udry, 1978), which, ironically, would be expected if
Pheromone
Copulin was truly a species-specific pheromone.
Aside from the issue of species specificity, the validity It has been reported that both sexually experienced and
or generalizability of these findings even among rhesus sexually inexperienced male hamsters are attracted to
364 Doty

conspecific vaginal secretions of females. Singer et al. 5. The Erection-Eliciting Pheromone of the Rat
(1976), employing gas chromatography–mass spectrometry,
Some mature male rats exhibit penile erections and sexual
identified dimethyl disulfide (DMDS) from volatile frac-
behaviors in the presence of inaccessible estrous females
tions of hamster vaginal secretions as the pheromone
(Sachs et al., 1994). This phenomenon appears not to
involved. In the bioassay, sexually experienced hamsters
require adult sexual experience on the part of the male and
were employed, and the latency, duration, and number of
can be induced by airborne volatiles from estrous females,
animals “approaching, sniffing and digging” near a section
even ones who are anesthetized (Sachs, 1997). Such
of the cage under which either DMDS, the volatile frac-
volatiles, however, appear to be effervescent, as bedding
tion, or the vaginal secretion were located was compared.
soiled by estrous females does not produce this effect
The number of animals exhibiting “positive responses”
(Sachs et al., 1994). This phenomenon is not dependent
towards the whole vaginal secretion or the volatile fraction
upon visual or auditory cues (Kondo et al., 1999; Sachs,
was 12 of 12. The number exhibiting such responses to
1997) and is eliminated by lesions of the olfactory, but not
DMDS varied with the DMDS concentrations chosen by
vomeronasal, nerve (Kondo et al., 1999). This is in spite of
the authors. At 128 and 22 ng, 8 of the 12 males responded;
the fact that it is eliminated by lesions within the medial
at 56 and 2 ng of material, 5 of the 12 males responded.
amygdala, a structure that serves as a major relay of the
In other words, at the best DMDS concentration
accessory olfactory system (Kondo et al., 1997). Sachs
concocted, only two thirds of the males responded to the
(1997) notes, “receptive female rats apparently broadcast a
“pheromone,” unlike the parent secretion or the volatile
volatile pheromone that promotes erection. Pheromones
fraction derived from the secretion, where 100% of the ani-
are well known to attract mates and to act in concert with
mals responded. At the other concentrations, less than half
other stimuli to promote mating. However, this is the first
of the animals responded to the agent. Of the responding
mammalian evidence for a volatile pheromone acting
subjects, the mean (  SEM) duration of “approaching,
alone to evoke a sexual fixed-action pattern and, in that
sniffing and digging” near the region of the cage under
sense, acting as an airborne aphrodisiac.”
which the stimulus was located was 58(1) seconds for the
Since no tests of species specificity of this phenomenon
whole vaginal secretion, 37(9) seconds for the volatile
have been made and no attempts have been made to iden-
fraction, and 65(16), 23(10), 51(16), and 26(11) seconds
tify the chemicals involved, its status as a pheromone—at
for dimethyl disulfide at 128, 56, 22, and 2 ng concentra-
least according to the primary criteria inherent to most def-
tions, respectively. Responses to control stimuli (water)
initions of pheromones—is unclear. It would be of interest,
were nominal (i.e., 10 seconds or less, on average). It is
for example, to determine whether lactating female rats that
interesting that the duration of behavior directed towards
exude a number of hormone-influenced odors also induce
the dimethyl disulfide in the responding animals was
penile erection. It should be noted that this effect does not
greater, in some cases, than that directed towards the orig-
occur in a large majority of rats. Thus, in one study only
inal secretion.
55% (11/20) of sexually inexperienced males displayed this
Petrulis and Johnston (1995) reasoned that if dimethyl
effect; in another study within the same series, the same
disulfide is truly a sex attractant pheromone, then (1) male
proportion (11/20) exhibited these erections, even though 5
hamsters should spend more time than female hamsters
of the 11 were sexually experienced (Sachs, 1997). One of
investigating it and (2) the attraction to the substance by
the 11 rats was upwind from the odor source, suggesting
males should be testosterone-related. In the first of two
the likelihood of a spontaneous erection.
experiments, these authors found that males investigated
female vaginal secretions more than did females, but this
was not the case with dimethyl disulfide, where both sexes B. Priming Pheromones
equally investigated the agent. In the second experiment,
castrated males given testosterone investigated the vaginal Since the 1960s, the effects of urine or body odors on con-
secretions more than castrated males not given testo- specific reproductive endocrine responses have been attrib-
sterone. However, neither castration nor testosterone reple- uted to priming pheromones. Most such agents have been
tion influenced attraction towards dimethyl disulfide or a described in mice and a few other rodent forms. Whether,
control odor, leading these authors to conclude that or to what degree, such phenomena exist in a wide range
“DMDS does not elicit sex differences in attraction and of other species or whether, in fact, they even exist outside
that in males the attraction to DMDS is not dependent on of the laboratory is a matter of controversy (see Bronson,
gonadal hormones. These results suggest that DMDS is not 1979; Bronson and Coquelin, 1979; Labov, 1981).
a sex attractant by itself nor is it a major component of an According to the classical mammalian pheromone para-
attractant mixture.” digm, a pheromone exists in the urine of group-housed
Mammalian Pheromones 365

female mice isolated from conspecific males that pro- (Norris and Adams, 1979). The female appears to be most
duces, in other females, a lengthening of the diestrous vulnerable to pregnancy block within 48 hours of coitus, with
component of their estrous cycles or, in some cases, exposure during the first 12 hours being sufficient in the
pseudopregnancy. Another pheromone is purportedly pre- majority of cases (Bruce, 1961). The effectiveness of the
sent in the urine of noncastrate mature male mice, which, pregnancy block is reportedly not augmented by increasing
when presented to isolated females or females who have the number of males to which a female is exposed (Bronson
become acyclic as a result of being housed with other et al., 1963; Bruce, 1963; Chipman and Fox, 1966a),
females in the absence of a male, intiates estrous cycling although Chipman and Fox (1966b) reported six strange
and induces shorter and more regular cycles. This or a sim- males blocked a greater percentage of pregnancies than a sin-
ilar pheromone is also said to accelerate puberty in young gle male (85% vs. 42%), conceivably reflecting the intensity
females, and a pheromone in the urine of females is reported or complexity of the stimulus.
to delay puberty in young females. The blockage of ovum The Bruce effect has been attributed to a pheromone,
implantation in a recently inseminated female by a strange, since (1) it occurs even when the strange male is separated
nonstud, male or his odor is also believed to depend upon from the female by a wire partition (Bruce, 1959), (2) it is
a pheromone, although in this case the odor of the stud eliminated by olfactory bulbectomy or damage to the
male is said to be remembered by the female in order for female’s accessory olfactory system (Bellringer et al.,
the strange male pheromone to be effective in inducing 1980; Bruce and Parrott, 1960; Lloyd-Thomas and
the blockage. In this section two so-called priming Keverne, 1982; Rajendren and Dominic, 1985; Reynolds
pheromones are examined—one associated with the block- and Keverne, 1979), and (3) urine or previously soiled
age of pregnancy and the other with the acceleration of bedding from a strange male is as effective, in many
puberty. strains, as the strange or alien male himself in producing
the effect (Dominic, 1964, 1965, 1966b). Memory of the
stud male is believed to be important, since the effect is
1. The Strange Male Pregnancy Blocking Pheromone
eliminated by infusion of phentoamine or other memory-
(Bruce Effect)
blocking agents into the female’s accessory olfactory bulb
In 1959, the endocrinologist Hilda Bruce reported that after initial mating (Kaba and Keverne, 1988; Kaba et al.,
only 29% of a group of recently inseminated albino mice 1989). Nonetheless, at least limited physical contact with
became pregnant when paired immediately after insemin- the male seems to be needed in some strains or test situa-
ation with a nonstud male mouse of the wild-type strain, tions to block pregnancy (deCatanzaro et al., 1995b).
compared to 100% who were similarly paired with the stud While in one study exposure of recently mated Parks albi-
male (Bruce, 1959). Housing with a nonstud albino male no females to urine-soiled bedding from strange males in
mouse decreased the pregnancy rate to 72%. The latter boxes was ineffectual (Bruce, 1960b), the pregnancy block
decrement was present regardless of whether the male albino could be produced by housing the mice in tall, poorly ven-
mouse was intact or castrated (Bruce, 1960a), a finding tilated, glass jars containing bedding to which strange male
now believed to be aberrant, given numerous subsequent odor urine had been added. Another study using the same
reports—including ones from Bruce’s own laboratory— type of albino mice found that once-daily renewal of the
that castrates are ineffectual in blocking pregnancies (e.g., soiled bedding is markedly inferior to twice-daily renewal,
Bruce, 1965; Spironello-Vella and deCatanzaro, 2001). suggesting that “. . . the operative substances are evanes-
A number of studies have verified the “Bruce effect,” cent, highly volatile or highly labile, probably both”
finding in some cases that same-strain (“strange”) male mice (Parkes et al., 1962). Three 15-minute exposures over a 4-
block pregnancies in 25–30% of the females, whereas differ- day test period have been found sufficient to produce the
ent-strain (“alien”) males do so in up to 80% of the females phenomenon in outbred Swiss strain females exposed to
(Parkes and Bruce, 1962). This general phenomenon has wild-type males (Chipman et al., 1966).
been described in numerous Mus strains (e.g., C3H, BALB/c, The physiological basis for the Bruce effect is not
CBA), as well as in Peromyscus maniculatus, Microtus entirely understood. It has been generally assumed to
agrestis, Microtus ochrogaster, and Microtus pennsylvanicus depend upon the leuteotrophic function of prolactin from
(Bronson and Eleftheriou, 1963; Bronson et al., 1969; Bruce, the adenohypophysis,* since it is (1) blocked by injecting
1959, 1960a; Clulow and Clarke, 1968; Clulow and
Langford, 1964; Clulow et al., 1968; Stehn and Richmond,
1975; Terman, 1969; Watson et al., 1983). It apparently does *Crowding can influence postimplantation intrauterine mortality
not occur in some highly inbred Mus strains (Bruce, 1968; as well in a number of mammals, including housemice and
Kakihana et al., 1974) or in gerbils (Meriones unguiculatus) deermice (Helmreich, 1960).
366 Doty

recently inseminated females with the primary leuteogenic blocks. In mice, the pregnancy of recently inseminated
agent in the mouse, prolactin, or with progesterone during females can be blocked by (1) human handling (an affect
the strange male odor exposure period (Bruce and Parkes, eliminated by the injection of progesterone) (Chipman
1960; Dominic, 1966b; Rajendren and Dominic, 1987),† et al., 1966b; Runner, 1959; Weir and DeFries, 1963),
(2) blocked by the implantation of a functional ectopic (2) enforced swimming for 3 minutes and/or exposure to
pituitary graft known to produce prolactin (Bronson et al., loud tones and open areas (Weir et al., 1963), (3) exposure
1969; Dominic, 1966a, 1966b, 1967), (3) absent in post- to male rats (with or without tactile contact) (deCatanzaro,
partum pregnant females whose prolactin production is 1988), (4) exposure to predators (deCatanzaro, 1988),
induced by suckling (Bruce and Parkes, 1961), and (5) exposure to male or female rat urine (deCatanzaro,
(4) blocked by the injection of reserpine (Dominic, 1966b, 1988), and (6) the induction of nutritional or restraint
1966c), an agent that depletes stores of catecholamines and stress (Euker and Riegle, 1973; McClure, 1959), the latter
serotonin in the brain and suppresses the inhibitory hypo- of which is reduced by the administration of estrogen anti-
thalamic center that controls release of prolactin (Dominic, bodies during the preimplantation and early implantation
1966b). Further evidence of a role of prolactin comes from stages (deCatanzaro et al., 1994). McClure et al. (1987)
the observation that exposure of recently inseminated found that by starving and feeding mice in a series of alter-
females to strange male urine when prolactin surges nating 2-day periods, the appearance of litters could be
induces the block, but not at other times (Rosser et al., completely prevented. Moreover, the effectiveness of the
1989). Bromocriptine, a dopamine agonist, was just as pregnancy block is related to the aggressiveness and sexual
effective as the exposure to strange male odor, suggesting behavior of the strange mouse, with intromissions being
that dopaminergic reduction of prolactin may be the basis particularly important (deCatanzaro and Storey, 1989;
of the effect. It has also been suggested that LH release Storey and Snow, 1990). In the deermouse, Peromyscus
may be the critical or at least an initial determinant of the maniculatus, the nature of the postinsemination environ-
pregnancy blockage (Chapman et al., 1970). Estrogen may ment influences implantation (Eleftheriou et al., 1962),
be involved, since (1) the presence of males appears to such that postinsemination housing in different-sized
enhance the synthesis and release of follicle-stimulating cages decreases pregnancy success by 30–60%, depending
hormone (FSH) in gonadectomized female mice (Bronson upon the difference in size of the new environment relative
and Desjardins, 1969), (2) the administration of estrogen, to that of the old. In rats, exposure to noxious sounds 4–6
including minute amounts applied to the nose, eliminates days after mating reduces the number of pregnancies, as
successful implantation in nonlactating female mice does chronic restraint (Euker et al., 1973; Zondek and
(Bloch, 1971; deCatanzaro et al., 2001), and (3) antibodies Tamari, 1967). The adrenal gland is likely involved in
to 17-estradiol prevent the Bruce effect in recently some cases, since the Bruce effect is not present in some
inseminated females (deCatanzaro et al., 1995a). Exo- inbred mouse strains that exhibit attenuated adrenal
genous administration of androstenedione and dehy- responses (Marsden and Bronson, 1965; Snyder and
droepiandrosterone, agents that can be converted to estro- Taggart, 1967) and, in some strains, is not present or is mit-
gens, is capable of blocking a recently-inseminated igated in adrenalectomized females (Snyder et al., 1967;
female’s pregnancy; the major stress-related adrenal hor- however, see Sahu and Dominic, 1981). Exogenous
mone corticosterone, which is not convertible into estro- adrenocorticotropic hormone (ACTH) is capable of
gens, fails to do so (de Catanzaro et al., 1991). Epinephrine inhibiting luteinization in adrenalectomized mice, but is
is ineffectual, even though it does disrupt female mating more effective in mice with adrenal glands (Christian et al.,
behavior (deCatanzaro and Graham, 1992). 1965). Interestingly, endogenous estrogen rises in response
Whatever the hormonal factors involved, it seems quite to acute stress during early pregnancy as well as during
conceivable that they reflect responses to stress. As noted nonpregnant states (MacNiven et al., 1992b; Shors et al.,
in several recent reviews (e.g., deCatanzaro and 1999), and injections of ACTH can produce increases in
MacNiven, 1992; Marchlewska-Koj, 1997), blockage of estrogen levels in some species (Arai et al., 1972; Strott et
implantation is not uncommon in rodents and can be al., 1975), adding further credence to the notion that estro-
induced by a variety of stressors, suggesting that the Bruce gen may be a primary factor in stress-induced blockage of
effect is one of a number of stressor-induced pregnancy implantation.*

†It is noteworthy that progesterone is as effective as prolactin in

this regard if presented early enough (Bruce et al., 1960; *Estrogen also rises in response to chronic stress (MacNiven
Dominic, 1966b; Rajendren et al., 1987). et al., 1992a)
Mammalian Pheromones 367

Evidence that stress-reducing manipulations can protect exception of those targeted on odor memory formation of
against the Bruce effect include the observations that the stud or strange male (e.g., phentolamine, anisomycin),
(1) the presence of the stud along with a strange male miti- have anxiolytic or antidepressant properties, including
gates the strange male’s blocking ability to some degree amitriptyline, chloropromazine, haloperidol, pimozide,
(Parkes et al., 1961; Terman, 1969), conceivably decreas- progesterone, prolactin, propranolol, and reserpine (Bloch
ing the strangeness of the situation or counteracting the and Wyss, 1973; Dominic, 1966b, 1966c; Rajendren and
strange male odor, (2) replacement of a strange male with Dutta, 1988; Sahu and Dominic, 1980; Saletu et al., 1975;
an original stud who was present during the period of the Torner et al., 2001). Antidepressants such as fluoxetine,
previous pregnancy prevents the pregnancy block, whereas amitriptyline, desipramine, and buspirone have been
replacement with a stud who had limited contact with the demonstrated to enhance habitutation to novel stimuli in
female does not (Bloch, 1974),† (3) familiarization of the olfactory bulbectomized rats—rats that exhibit many
female with a male before she is mated with another male characteristics of stress. This raises the possibility that
mitigates the familiarized male’s ability to block the preg- such agents may depress the degree of novelty of a strange
nancy, regardless of the strain of the female, the strain of male or his odors on implantation (Mar et al., 2000). Of
the stud male, or the strain of the familiarized male course, such actions need not be independent of the influ-
(Furudate and Nakano, 1981; Parkes et al., 1961),‡ and (4) ences of these agents on the hypothalamic-pitutary-
the presence of other females or their urine—even urine gonadal axis.
from spayed or androgenized females—during the critical If one accepts the proposition that the Bruce effect is
postcoital period can effectively prevent the pregnancy induced by stress, then it would seem that testosterone, or
block in some strains, with the prophylactic effect being some other testicular-influenced agent, is involved in the
directly related to the number of females present (at least production of the stress-inducing odor, since (1) urine
up to the largest number tested, i.e., 6) (Bruce, 1963; from male castrates is typically ineffectual (Bruce, 1965;
Dominic, 1965). Whether this reflects the calming effect of Spironello-Vella et al., 2001), (2) males are capable of
such odors, familiarity with such odors, prior conditioning blocking pregnancy only after the age of puberty (Bruce,
induced in the suckling setting, or some type of chemical 1965), (3) urine from males housed alone is less effective
cross-adaptation to the strange male odor is not clear. In an than urine from males housed near females (whose testos-
atypically reactive housemouse strain, in which the thresh- terone titer would be expected to be comparatively higher)
old for pregnancy blockage is low, repeated handling of the (deCatanzaro et al., 1999), (4) the antiandrogen cypro-
rat pups in infancy decreases their susceptibility to the terone acetate mitigates the pregnancy blocking ability of
Bruce effect in later life (Bruce et al., 1968), presumably the urine (Bloch et al., 1973), and (5) male mice who have
reflecting mitigation of later stress responses (Denenberg achieved sexual satiety are less effective than those who
et al., 1977; King, 1959; Levine and Broadhurst, 1963). It have not in producing the pregnancy blockage (Spironello
is noteworthy that hyperprolactinemia is associated with and deCatanzaro, 1999), presumably reflecting decreased
suppression of stressor-induced elevations in plasma corti- testosterone titer (Batty, 1978). The accessory sexual
costerone levels (Drago et al., 1986; Endroczi and Nyakas, glands (e.g., the vesicular and coagulating glands) or the
1974) and that reserpine reverses the density-dependent preputial glands are unlikely important, since pregnancy
adrenal hypertrophy and reproductive decrements blocking efficacy remains in mice lacking these organs
observed in male housemice (Christian, 1956). It is of par- (Hoppe, 1975; Zacharias et al., 2000). Administration of
ticular interest that most, if not all, drugs and hormones epiandrosterone, androstenedione, androsterone, or testos-
that have been shown to block the Bruce effect, with the terone to SJL female mice results in blocked pregnancies;
administration of progesterone or dehydroepiandrosterone
does not (Hoppe, 1975). Alternatively, it is possible that
testosterone simply produces a strong salient smell that,
†One study reports that duration of exposure required to make a without familiarization or adaptation on the part of the
stud male familiar to the female is 3–4.5 hours (Rosser and female, (1) is clearly discernible to the female, (2) allows
Keverne, 1985).
‡Some data do not support the familiarization hypothesis. For for accurate differentiation between different males, and
(3) elicits enhanced sensorineural activity. Urine from
example, Lott and Hopwood (1972) reported that if the stud is
removed from the female within 3 hours of mating, pregnancy noncastrate male rodents is much more intense, even to
block from a strange male is less likely to occur than if he humans, than that from castrates or females.
remains with the female for 24 hours or longer, suggesting that Whatever the role of testosterone, it seems clear that the
exposure to the stud “sensitized” the female to subsequent preg- female rapidly learns the odor of the stud male and by fur-
nancy block. ther exposure accommodates herself to or adapts to his
368 Doty

odor over time, making it either weak or familiar, and substance accelerates sexual maturation rather than simply
relatively nonstressful.* Presumably the degree of strange- serving as a trigger for puberty (Colby and Vandenbergh,
ness of the strange male is a function of the degree of qual- 1974). Urine from males whose preputial glands are
itative difference between its odor and that of the stud—a removed is as effective as urine from intact males.
difference likely learned and somehow discerned by, or Although conceivably strain differences are present, it
dependent upon, the accessory olfactory system (Brennan is noteworth that (1) urine from dominant males produces
and Keverne, 1997). If castrate males have relatively little greater acceleration than urine from subordinant males,
smell, then they or their odors would probably not seem (2) the male urine must be present for at least 2–3 hours
particularly strange or stressful to the female. per day, or the male must be present for 1 hour per day, to
Unfortunately, no control studies have been performed that produce the acceleration, (3) urine from the same male
have attempted to equate artificial odors on intensity or presented each day produces the same degree of accelera-
quality with noncastrate male urine odors, or to differen- tion as urine from different males presented each day,
tially odorize the stud and strange males to see if artificial (4) urine from the father or a full brother exerts the same
odors can produce the Bruce effect. degree of acceleration as urine from unrelated males, and
Inherently, the Bruce effect would seem not to be well (5) excreted or bladder urine from adrenalectomized males
described as being due to a pheromone, since the is as effective as urine from an intact male (for review see
pheromone would have to vary from male mouse to male Drikamer, 1986b). However, the presence of an adrenalec-
mouse as long as the mouse differs from the stud male in tomized male does not produce the same degree of accel-
its identifying chemical composition or odor. Hence, there eration as the presence of an intact male, possibly because
could be as many pheromones as there are male mice that adrenalectomized males pursue young females less and
can be identified as individuals, at least as individuals who attempt fewer mounts (Drickamer, 1983). To date, the
have testosterone-related odors. Moreover, this phenom- active “pheromone(s)” have not been identified, although
enon has a major learning component associated with it they are said to be absent in food-deprived males and in
and seems to be a subclass of a group of phenomena that males maintained under short photoperiods when testos-
rely upon the fragile nature of the implantation process. terone would be expected to be low.
Analogous to the situation with the Bruce effect, how-
2. The Puberty-Accelerating Pheromone of Male Mice ever, pubertal acceleration is not uniquely determined by
male urine, and it is well established that a number of stim-
Prepubertal female mice housed or otherwise exposed to
uli and stressors can accelerate the puberty of female mice
noncastrate males or their urine attain puberty sooner than
and other rodents.* Thus, similar acceleration can be
females not so exposed, as measured, for example, by the
obtained by exposing prepubertal female mice to urine
time of vaginal opening (Andervont, 1944; Castro, 1967;
from (1) pregnant or lacting female mice (Drickamer,
Cowley and Wise, 1972; Vandenbergh, 1967,1969). This
1984) (2) estrous female mice (Drikamer, 1986a), and (3)
acceleration of puberty, which has been attributed to a
male rats (Colby and Vandenbergh, 1974). The latter
pheromone, is stronger when the male is present, conceiv-
observation implies that if specific urinary agents are
ably reflecting the addition of contact stimulation
involved in this phenomenon, they may not be species-
(Bronson and Maruniak, 1975; Drickamer, 1974, 1975).
specific. Evidence that stressful situations can advance
Application of male urine directly to the oral-nasal grooves
puberty in a number of mammals comes from several quar-
of postweanling female mice accelerates the time of onset
ters. First, female rat pups that are handled and placed
of first estrus by 4–6 days relative to contols exposed to tap
individually in separate containers for 3 minutes each day
water. Exposure for 3 days between the ages of 21 and 29
days is sufficient to advance puberty. Exposure to male
urine tends to shorten the interval between the day of vagi- *Not all stressful stimuli need to activate adrenal responses. For
nal opening and the day of first estrus, implying that the example, social isolation, long considered to be a “stressor,” can
influence a number of behaviors (e.g., agonistic behavior, scent
marking, emotionality) independent of clear changes in plasma
*A more common explanation of this phenomenon is that the corticosterone levels or adrenal gland weight (Spencer et al.,
inseminated female requires time to learn the characteristics of 1973). Yoshimura (1980) found-prolonged isolation increased
the stud male (Brennan et al., 1990; Bronson, 1976). scent marking of male gerbils, a behavior known to be largely
Circumstantial support for the hypothesis that it reflects habitu- androgen dependent. The cholinergic antagonist scopoalmine,
ation comes from rats, where continuous exposure of a female to however, suppressed such marking behavior independent of any
the same male produces fewer estrus cycles than successive expo- measurable changes in central acetylcholinesterase or choline
sures to different males (Cooper et al. 1972). acetyltransferase activity.
Mammalian Pheromones 369

from birth to 24 days of age display first estrus, on aver- influenced by, prenatal or postnatal experience? If so, are
age, 10 days before unhandled controls (Morton et al., they best described as being mediated by pheromones?
1963). Second, handling female rats from birth to 30 days While the examples described earlier in this chapter
of age and rehousing them in small cages advances the demonstrate the complexity of chemically mediated
time of vaginal opening. In contrast, housing handled behaviors in mammals, some responses to isolated agents
female rats in groups of 10 in a large cage within an have been reported that, at first glance, would seem to be
enriched environment delays the age of pubertal opening somewhat invariant and independent of obvious postnatal
(Swanson et al., 1983). Third, applying intense visual, learning [e.g., the “aphrodisiac pheromone” protein in
auditory, and/or electrical stimuli shortly before the time of female hamster discharge (Singer et al., 1984, 1986,
normal physiological puberty accelerates the time of 1987)]. However, in most such cases the possibility of pre-
puberty in rats, although if such stimuli are applied much natal learning has not been addressed, and species speci-
in advance of this time, puberty can be delayed (Árvay, ficity has not been established. Moreover, the isolated
1967). Fourth, repeated exposure of young rats to cold materials are often not as effective in inducing the behav-
stress brings about vaginal opening 3–4 days earlier than in ior as the parent compound or may require urine or some
the controls (Mandl and Zuckerman, 1952). Fifth, stress other biological matrix to be effective (implying that other
seems to accelerate the age of menarchy in humans. For factors are involved, such as a combination of agents or
example, girls from divorced families and families with symbiosis with releasing strata). Thus, evoking the
greater interparental conflict tend to have an earlier menar- pheromone concept in such cases would seem premature.
che than girls from intact families (Wierson et al., 1993). There are, however, rare instances where a number of basic
These and other findings suggest that whatever factors operational criteria found in a number of definitions of
influence male-induced pubertal acceleration, they are pheromones (e.g., species specificity, stereotypical
likely complex. Like many other situations, the rearing response) have been tested and, for the most part, have
condition of female mice can influence the degree to which been met. One of these exceptional cases is described
stimuli from males accelerate the time of puberty. For below. Even in this case, however, there are questions as to
example, Mucignat-Caretta et al. (1995) reared females in whether the pheromone concept best describes the behav-
three conditions: with both parents, with two females, or ior, as postnatal learning may well intervene at some point
with two females and the presence of urine from adult and one might question whether the influences of
males. Nine days after weaning, the females were exposed intrauterine learning have been completely ruled out.
to either adult male urine or to prepubertal male urine. The Rabbit pups are said to be dependent upon a “short-
adult male urine resulted in larger uteri and more cornified range pheromone” located on the doe’s belly to release and
vaginal smears than the prepubertal male urine in the two guide a steretotyped search behavior for locating the nip-
groups reared with male odors (i.e., the one with both par- ple—guidance that occurs even in preterm-delivered pups
ents present and the one with two females and male odor). (Hudson and Distel, 1983). While at the time of the first
In the group reared with females only, no significant suckling episode the nipple may be moistened with amni-
changes in uterus weight or vaginal smears were noted. otic fluid, saliva is apparently not attractive (Hudson et al.,
The authors concluded that “the data support the notion 1983) and self-grooming does not seem to transfer an
that early experience of pheromonal cues may influence attractive agent to body areas distal to the nipple
the response to pheromones in a later period, even if the (Coureaud et al., 2001). Research suggests that two
preweaning exposure to males had no direct influences on sources of attractive material may be present, one distrib-
early signs of puberty onset.” uted over the nipple epidermis and one released within the
nipple, perhaps via sebaceous structures (Moncomble
et al., 2002). Apparently one or both of these stimuli forms
VIII. NONLEARNED PHEROMONAL a gradient that helps in guiding the pup towards the nipple
RESPONSES? (Coureaud et al., 2001). Such help is important, as a pup
must find and attach to a nipple quickly to survive, since
Aside from the induction of stress-related changes in the mother is available for nursing only about 3 minutes
endocrine function, or possibly the influences of hormones each day (Coureaud et al., 2000).* Olfactory bulbectomy
or hormone-like agents from urine or other secretions on
the reproductive processes of some species, are there *Interestingly, during the nursing episode rabbit pups switch
examples where one or a few chemicals exert species-spe- nipples periodically (average 2.6 times per minute), a behavior
cific stereotypic or reflexive behaviors or endocrine not dependent upon the amount of milk available from any one
responses that are not dependent upon, or significantly nipple (Hudson et al., 1983).
370 Doty

eliminates the nipple search behavior and, hence, suckling meets the following criteria for a pheromone: (1) chemical
(Distel and Hudson, 1985). In nonbreeding females, the simplicity; (2) unambiguous, morphologically invariant,
emission of the substance is influenced by day length, and functionally obvious behavioral response of the
peaking in the early summer, and is depressed by ovariec- receiver; (3) high selectivity of stimulus-response cou-
tomy and restored by estrogen administration (Distel and pling; (4) species specificity of reception; (5) species
Hudson, 1984). Its potency seems greatest during the specificity of emission; and (6) unconditioned stimulus-
immediate postpartum period, and it elicits the greatest response coupling. Among the studies performed on this
interest on the part of pups just prior to the time of compound, whose name has not yet been made available
regularly scheduled nursing periods (Coureaud et al., for proprietary reasons, were ones showing that (1) the
2001). Both progesterone and prolactin, probably in agent, when placed on a glass rod, elicited stereotypic
concert, seem to increase its emission in estrogen-primed searching motions and attempts to orally grasp the source
does (Gonzalez-Mariscal et al., 1994), although oxytocin of the stimulus, (2) detectable impurities co-occuring with
may also be involved (Fuchs et al., 1984). The nipple the commercially available agent do not elicit these
search response is not dependent upon an intact behaviors, (3) a decrease in these responses parallels a
vomeronasal organ (Hudson and Distel, 1986). decrease in the amount of this agent as the milk ages, and
According to Hudson et al. (1983), the “nipple search (4) replenishment of the aged milk with the substance
releasing odour of rabbit does may be considered as a true reactivates the behavioral activity directed towards the
pheromone as the behaviour elicited is reliable and highly milk. The compound’s effectiveness was, however, found
stereotyped.” Nevertheless, it should be noted that female to be concentration dependent, with optimal elicitation of
rabbit pups also display attraction towards abdominal the behavior occurring at concentrations 108 to 1010
odors of adult male rabbits, nonlacting female rabbits, and g/mL. At concentrations above 10 4 g/mL, the “behavior
nonlactating nonpregnant female rabbits, although the efficiency vanished steeply.” None of a wide range of other
odor of lactating females is more preferred (Coureaud and compounds, tested across a range of concentrations, were
Schaal, 2000). Moreover, learning cannot be ruled out found to produce this behavior, including (1) other
completely as a possible modifier of this response. Thus, if volatiles found in rabbit milk, (2) volatiles not found in
the rabbit mother is perfumed before nursing, the pups rabbit milk but found in other rabbit secretions, and
learn to respond to the novel odor with the characteristic (3) volatiles not found in rabbit milk but reportedly active
nipple-search behavior in a single 3 to 4 minute nursing in eliciting such behaviors in other neonatal mammals.
episode (Kindermann et al., 1991, 1994). In an insightful When same-strain females were fed during gestation and
review stressing the importance of learning in the process lactation with two isocaloric diets composed of exclusive
of odor perception, Hudson (1999) notes that, despite the constituents bearing unique aromatic compounds, their
fact that rabbit pups delivered by caesarean section exhibit pups responded to the same degree to the target substance,
normal search and suckling behavior when placed by a suggesting that its activity is not “dependent upon the indi-
lactating doe, “this does not exclude the possibility that the vidual bouquet of volatiles passed into the aminon or
response is dependent on prenatal experience of chemical milk.” These authors state, “The high behavior efficiency
characteristics of the uterine environment. In fact, this of pure “agent x” in any of these contrasted chemo-
might even be considered likely given the steep rise in ecological contexts proves the generality of its releasing
pheromone emission in late pregnancy (Distel et al., 1984) properties within the breed under study, and mitigates the
and reports that rabbit pups are able to learn prenatally possibility that responses to “agent x” may be generated by
odor cues associated with their mother’s diet (Bilko et al., experience in the individual-specific odor environment
1994; Hudson and Altbäcker, 1982; Semke et al., 1995; before and after birth.”
Coureaud, et al., 1997).” This compound was effective in a number of different
That being said, Schaal and associates have recently breeds of rabbits, implying that it is “an efficient releaser
reported that a single rather specific compound found in in newborns of O. cuniculatus regardless of their geno-
doe milk elicits the aforementioned nipple search behavior type.” The authors infer species specificity, since this agent
—a compound whose meaning, they argue, is learned nei- did not elicit the searching and nipple grasping responses
ther in utero nor post-natally (Schaal et al., 2003). In this in newborns from another Lagomorph species, Lepus
study, active peaks were initially identified using a split europaeus, or from several rodent species or cats (Rattus
stream gas chromatograph to establish biological reactions rattus, Mus musculus, Felis cattus). They further noted that
to various components of doe’s milk in a New freshly obtained colostrum and/or milk from rats, sheep,
Zealand–California cross-breed. These investigators per- cattle, horses, and humans did not elicit the searching-
formed a series of tests on this agent which they claim grasping response in the rabbit pups, and that upon
Mammalian Pheromones 371

chemical analysis bovine milk did not contain agent X. information about the environment, stimuli sampled by
That the response is not learned postnatally was suggested each sense are often fused into complex higher-order
from studies that found normal levels of attraction to this mental or cognitive constructs (Gibson, 1966; Marks,
agent in (1) vaginally delivered pups immediately isolated 1978). Thus, many apparent sensory redundancies are
from their mother or her secretions and (2) pups delivered interactive and nonorthogonal, and caution is warranted in
by Cesarean section a day before gestational term. Lack of assuming that any one sense provides a totally unique
influences of intrauterine exposure to the agent was contribution to the organism’s Umwelt (Doty, 1986).
inferred from the following observations. First, when Are there cases where, in fact, employing the term
placed on glass rods, neither blood nor amniotic fluid pro- pheromone aids in understanding or explaning chemically
duced the behavioral responses in 1 to 3-day-old pups. mediated mammalian behaviors or endocrine responses? It
Second, “the dosage of X in the headspace developing over would seem to the present author that if one cannot practic-
these substrates resulted in negative results.” The authors ally define or test whether a “pheromonal” substance
concluded, therefore, “that X may not directly contact the differs from a “nonpheromonal” substance, or if a large
developing nasal chemoreceptors through either blood or number of differing definitions are available for making
amniotic pathways. Accordingly, the development of pup such a distinction, the pheromone concept adds little to the
responsiveness to X apparently does not depend upon pre- scientific understanding of chemically mediated biological
natal induction through stimulus exposure.” processes. If one accepts the common elements of most
While one might argue that it unusual for a extant definitions of pheromones (e.g., species specificity,
pheromone’s activity to be narrowly tuned to a range of minimal influences of learning), one is hard-pressed to
concentrations, or that postnatal smelling of blood or find verified examples of pheromones in mammals.
amniotic fluid is not a conclusive test for the possibility Moreover, the general and uncritical use of the term is
that a preference was not learned in some way in utero, the widespread and evokes a number of unwarranted, or at best
authors of this study deliberately made a concerted effort untested, assumptions about the nature of the communica-
to determine whether the agent they identified met a strin- tive process under consideration. Thus, even its name
gent set of operational criteria for a pheromone. Such an conjures up the idea that the social organization of animals
effort reflects their awareness of the need for an oper- is akin to the endocrine organization of an organism, with
ational definition of this term. Only time will tell whether, disparate parts being influenced by chemicals that circu-
in fact, others will agree that this specific agent is truly a late within the social milieu. For some nonvertebrates this
pheromone. may be true, given a relatively high degree of stereo-
typic behavior and evidence for comparatively simple
stimuli that induce behavioral or endocrinological
IX. CONCLUSIONS changes. However, for many vertebrates, particularly
mammals, such a perspective would seem to be, with rare
The pheromone concept has attracted the imagination of exception, an oversimplification of the underlying biolog-
scientists and laypersons alike and on the surface appears ical processes. While auditory and visual stimuli can alter
to provide a straightforward explanation for a number of hormone levels in birds and a number of mammals, includ-
mammalian behaviors. However, it is apparent from the ing human beings, no scientist has found it necessary to
material presented in this chapter that there is mixed agree- evoke the terms “audiomones” or “visuomones” to
ment on what, in fact, constitutes a pheromone and how a describe such phenomena. Why, then, should the term
pheromone is to be recognized. Many investigators assume pheromone be employed to describe chemically mediated
that nearly every type of chemically mediated behavioral behaviors? Even though there may be instances where hor-
or endocrine response is mediated by a pheromone, even in mones or hormone-like chemicals are detected by chemical
the absence of any specific chemical stimulants and an receptors or are ingested or taken into the circulation of
agreed-upon set of criteria for distinguishing pheromones mammals via the lungs or via the highly vascularized nasal
from nonpheromones. While a range of chemicals can sig- cavity, these instances appear to be the exception rather
nificantly alter the behavioral and endocrine responses of than the rule. Making the assumption that most chemically
mammals, the mechanisms underlying such influences are mediated social or endocrinological responses of
poorly understood and seeming complex. Moreover, by far mammals are due to pheromones would seem, therefore,
the majority of phenomena attributed to pheromones can to oversimplify complex phenomena, providing a term
be evoked by other sensory stimuli, reflecting the redun- rather than an explanation for the observed responses.
dancy and complexity of the communicative process. Thus, it would appear that the current less-than-judicial
Since a primary function of the senses is to provide employment of the insect-derived pheromone concept in
372 Doty

describing chemically mediated behaviors and endocrine Alberts, J. R., and Galef, B. G. (1973). Olfactory cues and move-
responses of mammals is questionable. ment: stimuli mediating intraspecific aggression in the wild
That being said, there is no doubt that biologically Norway rat. J. Comp. Physiol. Psychol. 85:233–242.
derived chemicals have profound influences on mam- Alberts, J. R., and May, B. (1984). Nonnutritive, thermotactile
induction of filial huddling in rat pups. Dev. Psychobiol.
malian behaviors and endocrine responses, particularly
17:161–181.
among forms that live in burrows and are largely nocturnal
Albone, E. S. (1984). Mammalian Semiochemistry. New York:
or crepuscular. Moreover, one must not lose sight of the John Wiley & Sons.
fact that chemical communication has some inherent Albonetti, M. E., and D’Udine, B. (1986). Social experience
advantages over communication using most other types of occurring during adult life: its effects on socio-sexual olfactory
sensory stimuli. First, odorants can provide unique infor- preferences in inbred mice, Mus musculus. Anim. Behav.
mation about space occupancy or territory, being easily 34:1844–1847.
distributed in both space and time. “Time-coded” mes- Alleva, E., D’udine, B., and Oliverio, A. (1981). Effet d’une
sages can be sent, such as the length of time since a given experience olfactive precoce sur les preferences sexualles de
area has been visited or occupied or specific information deux souches de souris consanguines. Biol. Behav. 6:73–78.
about reproductive state. In the Norway rat, for example, Allin, J. T., and Banks, E. M. (1972). Functional aspects of ultra-
sound reproduction by infant Albino rats (Rattus norvegicus).
estrous urine loses its attractiveness within 24 hours
Anim. Behav. 20:175–185.
(Lydell and Doty, 1972). Second, chemical communication
Andervont, H. B. (1944). Influence of environment on mammary
is useful in situations where the receiver and the sender are cancer in mice. J. Natl. Cancer Inst. 4:579–581.
not present at the same point in time. Thus, chemical stim- Anonymous. (1970). Effects of sexual activity on beard growth in
uli can remain in the environment for rather long periods man. Nature 226:869–870.
of time even in the absent of the sender. If an animal emit- Arai, K., Kuwabara, Y., and Okinaga, S. (1972). The effect of
ted a continuous noise or visual signal in a manner analo- adrenocorticotropic hormone and dexamethasone, adminis-
gous to leaving a long-lasting odor, an inordinate amount tered to the fetus in utero, upon maternal and fetal estrogens.
of energy would be expended and predators could easily Am. J. Obstet. Gynecol. 113:316–322.
locate the sender. This long-lasting property of odors Arden, M. A., and Dye, L. (1998). The assessment of menstrual
allows a dominant male, for example, to make his odor synchrony: Comment on Weller and Weller (1997). J. Comp.
Psychol. 112:323–324.
nearly continuously present in the social environment.
Arnould, C., Rousmans, S., and Vernet-Maury, E. (1994).
Third, chemical communication is efficient, making use of
Influence of dodecyl propionagte, attractive pheromone from
excretory and secretory products that often have other rat pups, on rats’ food intake. Adv. Biosci. 93:377–382.
functions. Finally, scent marking makes it possible for the Aron, C. (1979). Mechanisms of control of the reproductive func-
sender and receiver to communicate even outside of the tion by olfactory stimuli in female mammals. Physiol. Rev.
range of hearing and sight, minimizing the physical harm, 59:229–284.
expenditure of energy, and exposure to predation that can Árvay, A. (1967). Effects of exteroceptive stimuli on fertility and
result from direct physical encounters or inappropriately their role in the genesis of malformations. In: Effects of
directed mating advances. External Stimuli on Reproduction, Ciba Foundation Study
Group No. 26. Boston: Little, Brown and Company, pp. 20–28.
Batty, J. (1978). Acute changes in plasma testosterone levels and
ACKNOWLEDGMENTS their relation to measures of sexual behaviour in the male
house mouse (Mus musculus). Anim. Behav. 26:349–357.
This chapter was supported, in part, by the following grants Beach, F. A. (1955). The descent of instinct. Psychol. Rev.
from the National Institutes of Health: PO1 DC 00161, 62:401–410.
RO1 DC 04278, RO1 DC 02974, and RO1 AG 17496. I Beach, F. A., and Jaynes, J. (1954). Effects of early experience
thank Lee Drickamer, Jack King, Igor Kratskin, Matthais upon the behavior of animals. Psychol. Bull. 51:239–263.
Laska, Joel Maruniak, Michael Meredith, Vincent Sava, Bear, M., Connors, B. W., and Paradiso, M. A. (1996). Neuroscience:
Benoist Schaal and John Stiller for their constructive com- Exploring the Brain. Baltimore: Williams and Wilkins.
Beauchamp, G. K., and Wellington, J. L. (1984). Habituation to
ments on a previous version of this manuscript.
individual odors occurs following brief, widely-spaced pre-
sentations. Physiol. Behav. 32:511–514.
REFERENCES Beauchamp, G. K., Doty, R. L., Moulton, D. G., and Mugford, R. A.
(1976). The pheromone concept in mammals: A critique. In
A Dictionary of Science, 4th ed. Oxford: Oxford University Mammalian Olfaction, Reproductive Processes, & Behavior,
Press, 1999. Doty, R. L. (Ed.). New York: Academic Press, pp. 143–160.
Alberts, J. R., and Brunjes, P. C. (1978). Ontogeny of thermal and Beauchamp, G. K., Yamazaki, K., and Boyse, E. A. (1985). The
olfactory determinants of huddling in the rat. J. Comp. chemosensory recognition of genetic individuality. Sci. Am.
Physiol. Psychol. 92:897–906. 253:86–92.
Mammalian Pheromones 373

Beaumont, P. J. V. (1982). Handbook of Psychiatry and Bronson, F. H. (1971). Rodent pheromones. Biol. Reprod.
Endocrinology. Amsterdam: Elsevier. 4:344–357.
Beck, M., and Galef, B. G., Jr. (1989). Social influences on the Bronson, F. H. (1976). Urine marking in mice: Causes and
selection of a protein-sufficient diet by Norway rats (Rattus effects. In Mammalian Olfaction, Reproductive Processes and
norvegicus). J. Comp. Psychol. 103:132–139. Behavior, Doty, R. L. (Ed.). New York: Academic Press,
Bellringer, J. F., Pratt, H. P., and Keverne, E. B. (1980). pp. 119–143.
Involvement of the vomeronasal organ and prolactin in Bronson, F. H. (1979). The reproductive ecology of the house
pheromonal induction of delayed implantation in mice. mouse. Q. Rev. Biol. 54:265–299.
J. Reprod. Fertil. 59:223–228. Bronson, F. H., and Coquelin, A. (1979). The modulation of repro-
Belluscio, L., Koentges, G., Axel, R., and Dulac, C. (1999). A duction by priming pheromones in housemice: speculations on
map of pheromone receptor activation in the mammalian adaptive function. In Chemical Signals: Vertebrates and
brain. Cell 97:209–220. Aquatic Invertebrates. New York: Plenum Press, pp. 243–265.
Bernhardt, P. C., Dabbs, J. M. J., Fielden, J. A., and Lutter, C. D. Bronson, F. H., and Desjardins, C. (1969). Release of
(1998). Testosterone changes during vicarious experiences of gonadotrophin in ovariectomized mice after exposure to
winning and losing among fans at sporting events. Physiol. males. J. Endocrinol. 44:293–297.
Behav. 65:59–62. Bronson, F. H., and Eleftheriou, B. E. (1963). Influence of
Bethe, A. (1932). Vernachlässigte Hormone. Naturwissens- strange males on implantation in the deer mouse. Gen. Comp.
chaften 11:177–181. Endocrinol. 3:515–518.
Bilko, A., Altbacker, V., and Hudson, R. (1994). Transmission of Bronson, F. H., and Maruniak, J. A. (1975). Male-induced puberty
food preference in the rabbit: The means of information trans- in female mice: evidence for a synergistic action of social
fer. Physiol. Behav. 56:907–912. cues. Biol. Reprod. 13:94–98.
Birch, M. C. (1974). Pheromones. New York: American Elsevier. Bronson, F. H., Eleftheriou, B. E., and Dezell, H. E. (1969).
Bitterman, M. E. (1965). The evolution of intelligence. Sci. Am. Strange male pregnancy block in deermice: prolactin and
212:92–100. adrenocortical hormones. Biol. Reprod. 1:302–306.
Bloch, S. (1971). Enhancement of on-time nidations in suckling Brouette-Lahlou, I., Vernet-Maury, E., and Vigouroux, M.
pregnant mice by the proximity of strange males. (1992). Role of pups’ ultrasonic calls in a particular maternal
J. Endocrinol. 49:431–436. behavior in Wistar rat: pups’ anogenital licking. Behav. Brain
Bloch, S. (1974). Observations on the ability of the stud male to Res. 50:147–154.
block pregnancy in the mouse. J. Reprod. Fertil. 38:469–471. Brown, R. E. (1979). Mammalian social odors: A critical review.
Bloch, S., and Wyss, H. I. (1973). An anti-androgen (cyproterone In Advances in the Study of Behavior. New York: Academic
acetate) inhibits the pregnancy block in mice caused by the Press, pp. 103–162.
presence of strange males (Bruce-effect). J. Endocrinol. Brown, R. E. (1988). Individual odors of rats are discriminable
59:365–366. independently of changes in gonadal hormone levels. Physiol.
Blum, S. L., Balsiger, D., Ricci, J. S., and Spiegel, D. K. (1975). Behav. 43:359–363.
Effects of early exposure to ventral gland odor on physical Brown, R. E., and MacDonald, D. W. (1985). Social Odours in
and behavioral development and adult social beha- Mammals. Oxford: Clarendon Press.
vior in Mongolian gerbils. J. Comp. Physiol. Psychol. Brownlee, R. G., Silverstein, R. M., Müller-Schwarze, D., and
89:1210–1219. Singer, A. G. (1969). Isolation, identification, and function of
Bonsall, R. W., and Michael, R. P. (1971). Volatile constituents of the chief component of the male tarsal scent in black-tailed
primate vaginal secretions. J. Reprod. Fertil. 27:478–479. deer. Nature 221:284–285.
Booth, A., Shelley, G., Mazur, A., Tharp, G., and Kittok, R. Bruce, H. M. (1959). An exteroceptive block to pregnancy in the
(1989). Testosterone, and winning and losing in human com- mouse. Nature 184:105.
petition. Horm. Behav. 23:556–571. Bruce, H. M. (1960a). A block to pregnancy in the mouse caused
Brain, P. F., Homady, M. H., Castano, D., and Parmigiani, S. by the proximity of strange males. J. Reprod. Fertil.
(1987). Pheromones and behaviour of rodents and primates. 1:96–103.
Boll. Zool. 4:279–288. Bruce, H. M. (1960b). Further observations on pregnancy block
Breen, M. F., and Leshner, A. I. (1977). Maternal pheromone: in mice caused by the proximity of strange males. J. Reprod.
A demonstration of its existence in the mouse (Mus musculus). Fertil. 1:311–312.
Physiol. Behav. 18:527–529. Bruce, H. M. (1961). Time relations in the pregnancy-block
Brennan, P. A., and Keverne, E. B. (1997). Neural mechanisms of induced in mice by strange males. J. Reprod. Fertil.
mammalian olfactory learning. ProgNeurobiol. 51:457–481. 2:138–142.
Brennan, P., Kaba, H., and Keverne, E. B. (1990). Olfac- Bruce, H. M. (1963). Olfactory block to pregnancy among
tory recognition: A simple memory system. Science grouped mice. J. Reprod. Fertil. 6:451.
250:1223–1226. Bruce, H. M. (1965). Effect of castration on the reproductive
Bronson, F. H. (1968). Pheromonal influences on mammalian pheromones of male mice. J. Reprod. Fertil. 10:141–143.
reproduction. In Pheromonal Influences on Mammalian Bruce, H. M. (1968). Absence of pregnancy-block in mice when
Reproduction, Diamond, M. (Ed.). Bloomington: Indiana stud and test males belong to an inbred strain. J. Reprod.
University Press, pp. 341–361. Fertil. 17:407–408.
374 Doty

Bruce, H. M. (1970). Pheromones. Br. Med. Bull. 26:10–13. Chipman, R. K., and Fox, K. A. (1966a). Factors in pregnancy
Bruce, H. M., and Parkes, A. S. (1960). Hormonal factors in blocking: Age and reproductive background of females:
exteroreceptive block to pregnancy in mice. J. Endocrinol. Numbers of strange males. J. Reprod. Fertil. 12:399–403.
20:29–30. Chipman, R. K., and Fox, K. A. (1966b). Oestrous synchroniza-
Bruce, H. M., and Parkes, A. S. (1961). The effect of concurrent tion and pregnancy blocking in wild house mice (Mus
lactation on the olfactory block to pregnancy in the mouse. musculus). J. Reprod. Fertil. 12:233–236.
J. Reprod. Fertil. 6:vi–vii. Chipman, R. K., Holt, J. A., and Fox, K. A. (1966). Pregnancy
Bruce, H. M., and Parrott, D. M. V. (1960). Role of olfactory failure in laboratory mice after multiple short-term exposure
sense in pregnancy block by strange males. Science 131:1526. to strange males. Nature 210:653.
Bruce, H. M., Land, R. B., and Falconer, D. S. (1968). Inhibition Christian, J. J. (1956). Reserpine suppression of density-depend-
of pregnancy-block in mice by handling. J. Reprod. Fertil. ent adrenal hypertrophy and reproductive hypoendocrinism in
15:289–294. populations of male mice. Am. J. Physiol. 187:356.
Buck, L. B. (2000). The molecular architecture of odor and Christian, J. J., Lloyd, J. A., and Davis, D. E. (1965). The role of
pheromone sensing in mammals. Cell 100:611–618. endocrines in the self-regulation of mammalian populations.
Burger, B. V., Smit, D., Spies, H. S., Schmidt, C., Schmidt, U., Recent Prog. Horm. Res. 21:501.
and Telitsina, A. Y. (2001). Mammalian exocrine secretions Clark, A. B. (1982). Scent marks as social signals in Galago
XVI. Constitutents of secretion of supplementary sacculi of crassicaudatus II. Discrimination between individuals by
dwarf hamster, Phodopus sungorus sungorus. J. Chem. Ecol. scent. J. Chem. Ecol. 8:1153–1165.
27:1277–1288. Clegg, F., and Williams, D. (1983). Maternal pheromone in
Cairns, R. B., and Nakelski, J. S. (1970). On fighting in mice: Rattus norvegicus. Behav. Neur. Biol. 37:223–236.
situational determinants of intragroup dyadic stimulation. Clulow, F. V., and Clarke, J. E. (1968). Pregnancy-block in
Psychon. Sci. 18:16–17. Microtus agrestis, an induced ovulator. Nature 219:511.
Cairns, R. B., and Nakelski, J. S. (1971). On fighting in mice: Clulow, F. V., and Langford, P. E. (1964). Pregnancy-block in the
ontogenetic and experiential determinants. J. Comp. Physiol. meadow vole Microtus pennsylvanicus. J. Reprod. Fertil.
Psychol. 74:354–364. 24:275–277.
Campbell, N. A. (1996). Biology, 4th ed. Menlo Park, CA: Colby, D. R., and Vandenbergh, J. G. (1974). Regulatory effects
Benjamin/Cummings Publishing Co., Inc. of urinary pheromones on puberty in the mouse. Biol. Reprod.
Carani, C., Bancroft, J., Del, R. G., Granata, A. R., Facchinetti, F., 11:268–279.
and Marrama, P. (1990). The endocrine effects of visual erot- Cooper, K. J., Purvis, K., and Haynes, N. B. (1972). Further obser-
ic stimuli in normal men. Psychoneuroendocrinology vations on the ability of the male to influence the oestrous
15:207–216. cycle of the underfed rat. J. Reprod. Fertil. 28: 473–475.
Caroom, D., and Bronson, F. H. (1971). Responsiveness of Coopersmith, R., and Leon, M. (1984). Enhanced neural
female mice to preputial attractant: effects of sexual experi- response to familiar olfactory cues. Science 225:849–851.
ence and ovarian hormones. Physiol. Behav. 7:659–662. Coopersmith, R., and Leon, M. (1986). Enhanced neural
Carr, W. J., Loeb, L. S., and Dissinger, M. E. (1965). Responses response by adult rats to odors experienced early in life. Brain
of rats to sex odors. J. Comp. Physiol. Psychol. 59:370–377. Res. 371:400–403.
Carr, W. J., Loeb, L. S., and Wylie, N. R. (1966). Responses to Cornwell, C. A. (1976). Selective olfactory exposure alters social
feminine odors in normal and castrated male rats. J. Comp. and plant odor preferences of immature hamsters. Behav. Biol.
Physiol. Psychol. 62:336–338. 17:131–137.
Carr, W. J., Yee, L., Gable, D. and Marasco, E. (1976) Olfactory Corridi, P., Chiarotti, F., Bigi, S., and Alleva, E. (1993).
recognition of conspecifics by domestic Norway rats. Familiarity with conspecific odor and isolation-induced
J. Comp. Physiol. Psychol. 90:821–828. aggressive behavior in male mice (Mus domesticus). J. Comp.
Carter, C. S. (1972). Effects of olfactory experience on the behaviour Psychol. 107:328–335.
of the guinea-pig (Cavia porcellus). Anim. Behav. 20: 54–60. Coureaud, G., and Schaal, B. (2000). Attraction of newborn
Carter, C. S., and Marr, J. N. (1970). Olfactory imprinting and rabbits to abdominal odors of adult conspecifics differing in
age variables in the guinea-pig, Cavia porcellus. Anim. Behav. sex and physiological state. Dev. Psychobiol. 36:271–281.
18:238–244. Coureaud, G., Schaal, B., Orgeur, P., Hudson, R., Lebas, F.,
Castro, B. M. (1967). Age of puberty in female mice: and Coudert, P. (1997). Perinatal odour disruption impairs
Relationship to population density and the presence of adult neonatal milk intake in the rabbit. Adv. Ethol. 32:134.
males. Acad. Brasil. Cienc. 39:289–291. Coureaud, G., Schaal, B., Coudert, P., Rideaud, P., Fortun-
Cavaggioni, A., and Mucignat-Caretta, C. (2000). Major urinary Lamothe, L., Hudson, R., and Orgeur, P. (2000). Immediate
proteins, alpha(2U)-globulins and aphrodisin. Biochim. postnatal sucking in the rabbit: its influence on pup survival
Biophys. Acta 1482:218–228. and growth. Reprod. Nutr. Dev. 40:19–32.
Chapman, V. M., Desjardins, C., and Whitten, W. K. Coureaud, G., Schaal, B., Langlois, D., and Perrier, G. (2001).
(1970). Pregnancy block in mice: changes in pituitary Orientation response of newborn rabbits to odours of lactating
LH and LTH and plasma progestin levels. J. Reprod. females: relative effectiveness of surface and milk cues.
Fertil. 21:333–337. Anim. Behav. 61:153–162.
Mammalian Pheromones 375

Cowley, J. J., and Brooksbank, B. W. (1991). Human exposure Dember, W. N., and Jenkins, J. J. (1970). General Psychology.
to putative pheromones and changes in aspects of social Englewood Cliffs, NJ: Prentice-Hall, Inc.
behaviour. J. Ster. Biochem. Mol. Biol. 39:647–659. Denenberg, V. H., Desantis, D., Waite, S., and Thoman, E. B.
Cowley, J. J., and Wise, D. R. (1972). Some effects of mouse (1977). The effects of handling in infancy on behavioral states
urine on neonatal growth and reproduction. Anim. Behav. in the rabbit. Physiol. Behav. 18:553–557.
20:499–506. Distel, H., and Hudson, R. (1984). Nipple-search pheromone in
Cross, B. A., and Harris, G. W. (1952). The role of the neurohy- rabbits: dependence on season and reproductive state.
pophysis in the milk ejection reflex. J. Endocrinol. 8:148–161. J. Comp. Physiol. 155:13–17.
Curtis, R. F., Ballantine, J. A., Keverene, E. B., Bonsall, R. W., Distel, H., and Hudson, R. (1985). The contribution of the olfac-
and Michael, R. P. (1971). Identification of primate sexual tory and tactile modalities to the nipple-search behaviour of
pheromones and the properties of synthetic attractants. Nature newborn rabbits. J. Comp. Physiol. 157:599–605.
232:396–398. Dixon, A. K. (1982). A possible olfactory component in the
D’Amato, F. R., and Cabib, S. (1987). Chronic exposure to a effects of diazepam on social behavior of mice.
novel odor increases pups’ vocalizations, maternal care, and Psychopharmacologia 77:246–252.
alters dopaminergic functioning in developing mice. Behav. Dominguez, H. D., Lopez, M. F., and Molina, J. C. (1999).
Neur. Biol. 48:197–205. Interactions between perinatal and neonatal associative learn-
D’Amato, F. R., and Cabib, S. (1990). Behavioral effects of ing defined by contiguous olfactory and tactile stimulation.
manipulations of the olfactory environment in developing Neurbiol. Learn. 71:272–288.
mice: Involvement of the dopaminergic system. In Dominic, C. J. (1964). Source of the male odour causing preg-
Psychobiology of Stress, Puglisi-Allegra, S., and Oliverio, A. nancy block in mice. J. Reprod. Fertil. 8:266–267.
(Eds.). Amsterdam: Kluwer Academic Publishers, pp. 59–71. Dominic, C. J. (1965). The origin of the pheromones causing
deCatanzaro, D. (1988). Effect of predator exposure upon early pregnancy block in mice. J. Reprod. Fertil. 10:469–472.
pregnancy in mice. Physiol. Behav. 43:691–696. Dominic, C. J. (1966a). Effects of single ectopic pituitary grafts
deCatanzaro, D., and Graham, C. (1992). Influences of exo- on the oestrous cycle of the intact mouse. J. Reprod. Fertil.
genous epinephrine on two reproductive parameters in female 12:533–538.
mice: disruption of receptivity but not implantation. Horm. Dominic, C. J. (1966b). Observations on the reproductive
Behav. 26:330–338. pheromones of mice. II. Neuro-endocrine mechanisms involved
deCatanzaro, D., and MacNiven, E. (1992). Psychogenic preg- in the olfactory block to pregnancy. J. Reprod. Fertil. 11:415.
nancy disruptions in mammals. Neurosci. Biobehav. Rev. Dominic, C. J. (1966c). Reserpine: inhibition of olfactory block-
16:43–53. age of pregnancy in mice. Science 152:1764–1765.
deCatanzaro, D., and Storey, A. E. (1989). Partial medication of Dominic, C. J. (1967). Effect of ectopic pituitary grafts on the
strange-male-induced pregnancy blocks by sexual activity in olfactory block to pregnancy in mice. Nature 213:1242.
mice (Mus musculus). J. Comp. Psychol. 103:381–388. Dorland’s Illustrated Medical Dictionary, 26th ed. (1981).
deCatanzaro, D., MacNiven, E., and Ricciuti, F. (1991). Philadelphia: W. B. Saunders Company.
Comparison of the adverse effects of adrenal and Doty, R. L. (1974). A cry for the liberation of the female rodent:
ovarian steroids on early pregnancy in mice. Psychoneuro- courtship and copulation in Rodentia. Psychol. Bull.
endocrinology 16:525–536. 81:159–172.
deCatanzaro, D., MacNiven, E., Goodison, T., and Richardson, Doty, R. L. (1986). Odor-guided behavior in mammals.
D. (1994). Estrogen antibodies reduce vulnerability to stress- Experientia 42:257–271.
induced failure of intrauterine implantation in inseminated Doty, R. L. (2003). Mammalian pheromones, audiomones,
mice. Physiol. Behav. 55:35–38. visuomones and snarks.
deCatanzaro, D., Muir, C., O’Brien, J., and Williams, S. (1995a). Doty, R. L., and Dunbar, I. (1974). Attraction of beagles to con-
Strange-male-induced pregnancy disruption in mice: reduc- specific urine, vaginal and anal sac secretion odors. Physiol.
tion of variability by 17-estradiol antibodies. Physiol. Behav. 12:825–833.
Behav. 58:401–404. Drago, F., Amir, S., Continella, G., Alloro, M. C., and
deCatanzaro, D., Wyngaarden, P., Griffiths, J., Ham, M., Hancox, Scapagnini, U. (1986). Effects of endogenous hyperpro-
J., and Brain, D. (1995b). Interactions of contact, odor cues, and lactinemia on adaptive responses to stress. In Prolactin—
androgens in strange-male-induced early pregnancy disruptions Basic and Clinical Correlates, MacLeod, R. M., Thorner, M.,
in mice (Mus musculus). J. Comp. Psychol. 109: 115–122. and Scapagnini, U. (Eds.). Padova: Liviana Press,
deCatanzaro, D., Muir, C., Sullivan, C., and Boissy, A. (1999). pp. 609–614.
Pheromones and novel male-induced pregnancy disruptions in Drickamer, L. C. (1974). Contact stimulation, androgenized
mice: exposure to conspecifics is necessary for urine alone to females and accelerated sexual maturation in female mice.
induce an effect. Physiol. Behav. 66:153–157. Behav. Biol. 12:101–110.
deCatanzaro, D., Baptista, M. A., and Spironello-Vella, E. (2001). Drickamer, L. C. (1975). Contact stimulation and accelerated
Administration of minute quantities of 17beta-estradiol on the sexual maturation of female mice. Behav. Biol. 15:113–115.
nasal area terminates early pregnancy in inseminated female Drickamer, L. C. (1983). Male acceleration of puberty in female
mice. Pharmacol. Biochem. Behav. 69:503–509. mice (Mus musculus). J. Comp. Psychol. 97:191–200.
376 Doty

Drickamer, L. C. (1984). Acceleration of puberty in female mice Galef, B. G. J., and Kaner, H. C. (1980). Establishment
by a urinary chemosignal from pregnant or lactating females: and maintenance of preference for natural and artificial
timing and duration of stimulation. Dev. Psychobiol. olfactory stimuli in juvenile rats. J. Comp. Physiol.
17:451–455. Psychol. 94:588–595.
Drickamer, L. C. (1986a). Effects of urine from females in oestrus Galef, B. G. J., and Sherry, D. F. (1973). Mother’s milk: A medium
on puberty in female mice. J. Reprod. Fertil. 77:613–622. for transmission of cues reflecting the flavor of mother’s diet.
Drickamer, L. C. (1986b). Puberty-accelerating chemosignals in J. Comp. Physiol. Psychol. 83:374–378.
mice; ecological and evolution considerations. In Chemical Galef, B. G. J., and Henderson, P. W. (1972). Mother’s milk: a
Signals in Vertebrates IV, Duvall, D., Muller-Schwarze, D. determinant of the feeding preferences of weaning rat pups.
and Silvestein, R.M. (Eds.) NY: Plenum Press, pp. 441–455. J. Comp. Physiol. Psychol. 78:213–219.
Drickamer, L. C. (1992). Chemosignals and reproduction in adult Galef, B. G., Beck, M., and Whiskin, E. E. (1991). Protein defi-
female house mice. In Chemical Signals in Vertebrates VI, ciency magnifies social influence on the food choices of
Doty, R. L. Muller-Schwarze, D. (Eds). New York: Plenum Norway rats (Rattus norvegicus). J. Comp. Psychol.
Press, pp. 245–251. 105:55–59.
Dulac, C., and Axel, R. (1995). A novel family of genes encoding Galef, B. G. J., Iliffe, C. P., and Whiskin, E. E. (1994). Social
putative pheromone receptors in mammals. Cell 83:195–206. influences on rats’ (Rattus norvegicus) preferences for fla-
Dulac, C., and Axel, R. (1998). Expression of candidate vored foods, scented nest materials, and odors associated with
pheromone receptor genes in vomeronasal neurons. Chem. harborage sites: are flavored food special? J. Comp. Psychol.
Senses 23:467–475. 108:266–273.
Eibl-Eibesfeldt, I. (1970). Ethology: The Biology of Behavior. Galef, B. G. J., Whiskin, E. E., and Bielavska, E. (1997).
New York: Holt, Rinehart and Winston. Interaction with demonstrator rats changes observer rats’
Eleftheriou, B. E., Bronson, F. H., and Zarrow, M. X. (1962). affective responses to flavors. J. Comp. Psychol. 111:
Interaction of olfactory and other environmental stimuli on 393–398.
implantation in the deer mouse. Science 137:764. Gibson, J. J. (1966). The Senses Considered as Perceptual
Elias, M. (1981). Serum cortisol, testosterone, and testosterone- Systems. Boston: Houghton Mifflin.
binding globulin responses to competitive fighting in human Goldfoot, D. A., Kravetz, M. A., Goy, R. W., and Freeman, S. K.
males. Aggress. Behav. 7:215–224. (1976). Lack of effect of vaginal lavages and aliphatic acids
Endroczi, E., and Nyakas, C. S. (1974). Pituitary-adrenal func- on ejaculatory responses in rhesus monkeys: behavioral and
tion in lactating rats. Endokrinologie (Budapest) 63:1–5. chemical analyses. Horm. Behav. 7:1–27.
Euker, J. S., and Riegle, G. D. (1973). Effects of stress on preg- Goldfoot, D. A., Essock-Vitale, S. M., Asa, C. S., Thornton, J. E.,
nancy in the rat. J. Reprod. Fertil. 34:343–346. and Leshner, A. I. (1978). Anosmia in male rhesus monkeys
Evans, C. M. (1979). Studies on relationships between gonadal does not alter copulatory activity with cycling females.
hormones and intra-specific aggression behaviour in rodents. Science 199:1095–1096.
Ph.D. thesis, University of Wales. Gonzalez-Bono, E., Salvador, A., Serrano, M. A., and Ricarte, J.
Fenster, L., Waller, K., Chen, J., Hubbard, A. E., Windham, G. C., (1999). Testosterone, cortisol, and mood in a sports team
Elkin, E., and Swan, S. (1999). Psychological stress in the competition. Horm. Behav. 35:55–62.
workplace and menstrual function [published erratum appears Gonzalez-Mariscal, G., Chirino, R., and Hudson, R. (1994).
in Am J Epidemiol. Am. J. Epidemiol. 149:127–134. Prolactin stimulates emission of nipple pheromone in ovariec-
Fillion, T. J., and Blass, E. M. (1986a). Infantile behavioural reac- tomized New Zealand white rabbits. Biol. Reprod.
tivity to oestrous chemostimuli in Norway rats. Anim. Behav. 50:373–376.
34:123–133. Goyens, J., and Noirot, E. (1975). Effects of cohabitation with
Fillion, T. J., and Blass, E. M. (1986b). Infantile experience with females on aggressive behavior between male mice. Dev.
suckling odors determines adult sexual behavior in male rats. Psychobiol. 8:79–84.
Science 231:729–731. Graham, J. M., and Desjardins, C. (1980). Classical conditioning:
Fuchs, A. R., Cubile, L., Dawood, M. Y., and Jorgensen, F. N. Induction of luteinizing hormone and testosterone secretion in
(1984). Release of oxytocin and prolactin by suckling rabbits anticipation of sexual activity. Science 210:1039–1041.
throughout lactation. Endocrinology 114:462–469. Gregoire, A. T., Ledger, W. D., and Moran, M. J. (1973). The
Furudate, S., and Nakano, T. (1981). [Studies on the pheromonal glycogen content of the human female genital tract in cycling,
pregnancy block in the mouse. II. Discriminatory and/or menopausal and women with endometrial and cervical carci-
memorial mechanisms in female mice responding to the stud noma. Fertil. Steril. 24:198–201.
male]. [Japanese]. Jikken Dobutsu. Exp. Animals 30:1–5. Gregory, E. H., and Bishop, A. (1975). Development of olfactory-
Galef, B. G., Jr. (1981). Preference for natural odors in rat pups: guided behavior in the golden hamster. Physiol. Behav.
implications of a failure to replicate. Physiol. Behav. 15:373–376.
26:783–786. Groves, P., and Schlesinger, K. (1979). Introduction to Biological
Galef, B. G., Jr. (1986). Social identification of toxic diets by Psychology. Dubuque, IA: William C. Brown Company.
Norway rats (Rattus norvegicus). J. Comp. Psychol. 100: Halpern, M. (1987). The organization and function of the
331–334. vomeronasal system. Ann. Rev. Neurosci. 10:325–362.
Mammalian Pheromones 377

Hatanaka, T. (1992). Is the mouse vomeronasal organ a sex pheromones from oestrous cows increase synchronization of
pheromone receptor? In Chemical Signals in Vertebrates VI, oestrus in dairy heifers after PGF-2 alpha injection. J. Reprod.
Doty, R. L., and Müller-Schwartze (Eds.). New York: Plenum Fertil. 66:189–196.
Press, pp. 27–30. Jacobson, L. (1811). Description anatomique d’un organ observé
Hayashi, S. (1979). A role of female preputial glands in social dans les Mammiféres. Ann. Musée Hist. Natl. Paris
behavior of mice. Physiol. Behav. 23:967–969. 18:412–424.
Heise, S., and Hurst, J. L. (1994). Territorial experience causes a Janus, C. (1989). The development of olfactory preferences for
shift in the responsiveness of female house mice to odours artificial odors briefly experienced by the precocial spiny
from dominant males. Adv. Biosci. 93:291–296. mouse young. Behav. Neur. Biol. 52:430–436.
Hellhammer, D. H., Hubert, W., and Schurmeyer, T. (1985). Janus, C. (1993). Stability of preference for odors after short-term
Changes in saliva testosterone after psychological stimulation exposure in young spiny mice. Dev. Psychobiol. 26:65–79.
in men. Psychoneuroendocrinology 10:77–81. Jemiolo, B., Xie, T. M., and Novotny, M. (1991). Socio-sexual
Helmreich, R. L. (1960). Regulation of reproductive rate by intra- olfactory preference in female mice: attractiveness of synthetic
uterine mortality in the deer mouse. Science 132:417–418. chemosignals. Physiol. Behav. 50:1119–1122.
Hepper, P. G. (1987). The amniotic fluid: An important priming Jennings, J. W., and Keffer, L. H. (1969). Olfactory learning set
role in kin recognition. Anim. Behav. 35:1343–1346. in two varieties of domestic rat. Psychol. Rep. 24:3–15.
Hepper, P. G. (1988). Adaptive fetal learning: prenatal exposure Johnston, R. E. (1993). Memory for individual scent in hamsters
to garlic affects postnatal preferences. Anim. Behav. (Mesocricetus auratus) as assessed by habituation methods.
36:935–936. J. Comp. Psychol. 107:201–207.
Herrnstein, R. J. (1984). Objects, categories, and discriminative Johnston, R. E. (2000). Chemical communication and
stimuli. In Animal Cognition, Roitblat, H., Bever, T., and pheromones: The types of chemical signals and the role of the
Terrace, H. (Eds.). Hillsdale, NJ: Erlbaum, pp. 233–261. vomeronasal system. In The Neurobiology of Taste and Smell,
Heyser, C. J., Spear, N. E., and Spear, L. P. (1992). Effects of Finger, T. E., Silver, W. L., and Restrepo, D. (Eds.). New
prenatal exposure to cocaine on conditional discrimination York: Wiley-Liss, pp. 101–127.
learning in adult rats. Behav. Neurosci. 106:837–845. Jones, R. B., and Nowell, N. W. (1974). Effects of cyproterone
Hildebrand, J. G., and Shepherd, G. M. (1997). Mechanisms of acetate upon urinary aversive cues and accessory sex glands in
olfactory discrimination: converging evidence for common male albino mice. J. Endocrinol. 62:167–168.
principles across phyla. Ann. Rev. Neurosci. 20:595–631. Kaba, H., and Keverne, E. B. (1988). The effect of microinfu-
sions of drugs into the accessory olfactory bulb on the olfac-
Hines, P. J. (1997). Unconscious odors. Science 278:79.
tory block to pregnancy. Neuroscience 25:1007–1011.
Hodos, W., and Campbell, C. B. G. (1969). Scala naturae: Why
Kaba, H., Rosser, A., and Keverne, B. (1989). Neural basis of
there is no theory in comparative psychology. Psychol. Bull.
olfactory memory in the context of pregnancy block.
76:337–350.
Neuroscience 32:657–662.
Holinka, C. F., and Carlson, A. D. (1976). Pup attraction to lac-
Kakihana, R., Ellis, L. B., Gerling, S. A., Blum, S. L., and
tating Sprague-Dawley rats. Behav. Biol. 16:489–505.
Kessler, S. (1974). Bruce effect competence in yellow-lethal
Hoppe, P. C. (1975). Genetic and endocrine studies of the preg-
heterozygous mice. J. Reprod. Fertil. 40:483–486.
nancy-blocking pheromone of mice. J. Reprod. Fertil.
Karlson, P., and Lüscher, M. (1959). “Pheromones”: A new term
45:109–115.
for a class of biologically active substances. Nature 183:55–56.
Hudson, R. (1999). From molecule to mind: The role of experi-
Kendrick, K. (1975). Maternal pheromone: Discrimination by
ence in shaping olfactory function. J. Comp. Physiol. pre-weanling rats. Unpublished dissertation, University of
185:297–304. Durham, Durham, England.
Hudson, R., and Altbäcker, V. (1982). Development of feeding Kendrick, K. K., Haupt, M. A., Hinton, M. R., Broad, K. D., and
and food preference in the European rabbit: environmental Skinner, J. D. (2001). Sex differences in the influences on
and maturational determinants. In Behavioral Aspects of mothers on the socio sexual preferences on their offspring.
Feeding: Basic and Applied Research in Mammals, Galef, B. G., Horm. Behav. 40:322–338.
Mainardi, M., and Valsecchu, P. (Eds.). Newark: Harood Kessler, S., Harmatz, P., and Gerling, S. A. (1975). The genetics
Academic Publishers, pp. 125–145. of pheromonally mediated aggression in mice. I. Strain differ-
Hudson, R., and Distel, H. (1983). Nipple location by newborn ence in the capacity of male urinary odors to elicit aggression.
rabbits: Behavioural evidence for pheromonal guidance. Behav. Genetics 5:233–238.
Behaviour 85:260–275. Keverne, E. B. (1983). Pheromonal influences on the endocrine
Hudson, R., and Distel, H. (1986). Pheromonal release of suck- regulation of reproduction. Trends Neurosci. 6:381–384.
ling in rabbits does not depend on the vomeronasal organ. Keverne, E. B., and Michael, R. P. (1971). Sex-attractant proper-
Physiol. Behav. 37:123–128. ties of ether extracts of vaginal secretions from rhesus mon-
Hudson, R., and Distel, H. (1999). Induced peripheral sensitivity keys. J. Endocrinol. 51:313–322.
in the developing vertebrate olfactory system. Ann. NY Acad. Kimelman, B. R., and Lubow, R. E. (1974). The inhibitory effect
Sci. 856:109–115. of preexposed olfactory cues on intermale aggression in mice.
Izard, M. K., and Vandenbergh, J. G. (1982). Priming Physiol. Behav. 12:919–922.
378 Doty

Kindermann, U., Gervais, R., and Hudson, R. (1991). Rapid odor Leidahl, L. C., and Moltz, H. (1975). Emission of the maternal
conditioning in newborn rabbits: amnesic effect of hypother- pheromone in the nulliparous female and failure of emission
mia. Physiol. Behav. 50:457–460. in the adult male. Physiol. Behav. 14:421–424.
Kindermann, U., Hudson, R., and Distel, H. (1994). Learning of LeMagnen, J. (1951). Etude des phénomènes olfacto-sexuels
suckling odors by newborn rabbits declines with age and chez le rat blanc. C. R. Acad. Sci. Paris 145:1636–1641.
suckling experience. Dev. Psychobiol. 27:111–122. Leon, M. (1974). Maternal pheromone. Physiol. Behav.
King, J. A. (1957). Relationships between early social experience 13:441–453.
and adult aggressive behavior in inbred mice. J. Genetic Leon, M. (1975). Dietary control of maternal pheromone in the
Psychol. 90:151–166. lactating rat. Physiol. Behav. 14:311–319.
King, J. A. (1959). Effects of early handling upon adult behavior Leon, M. (1978). Emission of maternal pheromone. Science
in two subspecies of deermice, Permyscus maniculatus. 201:938–939.
J. Comp. Physiol. Psychol. 52:82–88. Leon, M., and Moltz, H. (1971). Maternal pheromone: discrimin-
King, J. A., and Gurney, N. L. (1954). Effect of early social expe- ation by pre-weanling albino rats. Physiol. Behav. 7: 265–267.
rience on adult aggressive behavior in inbred mice. J. Comp. Leon, M., and Moltz, H. (1972). The development of the
Physiol. Psychol. 47:326–330. pheromonal bond in the albino rat. Physiol. Behav.
Kippin, T. E., and Pfaus, J. G. (2001a). Nature of the conditioned 8: 683–686.
response mediating olfactory conditioned ejaculatory prefer- Levine, L., Barsel, G. E., and Diakow, C. A. (1965). Interaction
ence in the male rat. Behav. Brain Res. 122:11–24. of aggressive and sexual behavior in male mice. Behaviour
Kippin, T. E., and Pfaus, J. G. (2001b). The development of olfactory 25:272–280.
conditioned ejaculatory preferences in the male rat. I. Nature of Levine, S., and Broadhurst, P. L. (1963). Genetic and ontogenetic
the unconditioned stimulus. Physiol. Behav. 73:457–469. determinants of adult behavior in the rat. J. Comp. Physiol.
Kippin, T. E., Samaha, A.-N., Sotiropoulos, V., and Pfaus, J. G. Psychol. 56:423–428.
(2001). The development of olfactory conditioned ejaculatory Liman, E. R. (1996). Pheromone transduction in the vomeronasal
preferences in the male rat. II. Parametric manipulation of organ. Curr. Opin. Neurobiol. 6:487–493.
conditioning session number and duration. Physiol. Behav. Lloyd-Thomas, A., and Keverne, E. B. (1982). Role of the brain
73:471–485. and accessory olfactory system in the block to pregnancy in
Kirk-Smith, M. D., and Booth, D. A. (1980). Effect of mice. Neuroscience 7:907–913.
androstenone on choice of location in other’s presence. In Lorenz, K. (1950). The comparative method of studying innate
Olfaction and Taste VII, van der Starre, H. (Ed.). London: IRL behavior patterns. In Physiological Mechanisms of Animal
Press, pp. 397–400. Behavior, Danielli, J. F., and Brown, R. (Eds.). New York:
Kondo, Y., Sachs, B. D., and Sakuma, Y. (1997). Importance of Academic Press, pp. 221–268.
the medial amygdale in rat penile erection evoked by Lott, D. F., and Hopwood, J. H. (1972). Olfactory pregnancy-
remote stimuli from estrous females. Behav. Brain Res. block in mice (Mus musculus): An unusual response acquisi-
88:153–160. tion paradigm. Anim. Behav. 20:263–267.
Kondo, Y., Tomihara, K., and Sakuma, Y. (1999). Sensory Lydell, K., and Doty, R. L. (1972). Male rat of odor preferences
requirements for noncontact penile erection in the rat. Behav. for female urine as a function of sexual experience, urine age,
Neurosci. 113:1062–1070. and urine source. Horm. Behav. 3:205–212.
Kristal, M. B., Thompson, A. C., and Abbott, P. (1986). Ingestion Mackintosh, J. H., and Grant, E. C. (1966). The effect of olfac-
of amniotic fluid enhances opiate analgesia in rats. Physiol. tory stimuli on the agonistic behaviour of laboratory mice.
Behav. 38:809–815. Z. Tierpsychol. 23:584–587.
Labov, J. B. (1981). Pregnancy blocking in rodents: adaptive MacNiven, E., deCatanzaro, D., and Younglai, E. V. (1992a).
advantages for females. Am. Naturalist 118:361–371. Chronic stress increases estrogen and other steroids around
Lee, C. T., and Brake, S. C. (1971). Reactions of male fighters to intrauterine implantation in inseminated rats. Physiol. Behav.
male and female mice, untreated or deodorized. Psychon. Sci. 52:159–162.
24:209–211. MacNiven, E., deCatanzaro, D., and Younglai, E. V. (1992b).
Lee, C. T., and Griffo, W. (1973). Early androgenization and Chronic stress increases estrogen and other steroids in insem-
aggression pheromone in inbred mice. Horm. Behav. inated rats. Physiol. Behav. 52:159–162.
4:181–189. Macrides, F., Clancy, A. N., Singer, A. G., and Agosta, W. C. (1984).
Lee, C. T., and Griffo, W. (1974). Progesterone antagonism of Male hamster investigatory and copulatory responses to vaginal
androgen-dependent aggression-promoting pheromone in discharge: An attempt to impart sexual significance to an arbi-
inbred mice (Mus musculus). J. Comp. Physiol. Psychol. trary chemosensory stimulus. Physiol. Behav. 33:627–632.
87:150–155. Mainardi, D. (1963). Elimanazione della barriera etologica
Lee, S. V. d., and Boot, L. M. (1955). Spontaneous pseudopreg- all’Isolamento riproduttivo tra Mus musculus domesticus e
nancy in mice. Acta Physiol. Pharmacol. Neerlandica M.m. bactrianus mediante azione sull’apprendimento infan-
4:442–443. tile. Inst. Lombardo Accad. Sci. Lett. 97:291–299.
Lehrman, D. S. (1953). A critique of Konrad Lorenz’s theory of Mainardi, D., Marsan, M., and Pasquali, A. (1965). Causation of
instinctive behavior. Q. Rev. Biol. 28:337–363. sexual preferences of the house mouse. The behaviour of mice
Mammalian Pheromones 379

reared by parents whose odour was artificially altered. Mennella, J. A., and Beauchamp, G. K. (1991b). The transfer of
Soc Ital Nat Milano 104:325–338. alcohol to human milk. Effects on flavor and the infant’s
Mandl, A. M., and Zuckerman, S. (1952). Factors influencing behavior. N. Engl. J. Med. 325:981–985.
the onset of puberty in albino rats. J. Endocrinol. Mennella, J. A., and Beauchamp, G. K. (1996). The human
8:357–364. infants’ response to vanilla flavors in mother’s milk and for-
Mar, A., Spreekmeester, E., and Rochford, J. (2000). Anti- mula. Infant Behav. Dev. 19:13–19.
depressants preferentially enhance habituation to novelty in Mennella, J. A., Johnson, A., and Beauchamp, G. K. (1995).
the olfactory bulbectomized rat. Psychopharmacology Garlic ingestion by pregnant women alters the odor of amni-
150:52–60. otic fluid. Chem. Senses 20:207–209.
Marchlewska-Koj, A. (1997). Sociogenic stress and rodent Meredith, M. (1982). Stimulus accesss and other processes
reproduction. Neurosci. Biobehav. Rev. 21:699–703. involved in nasal chemosensory function.: Potential substrates
Marks, L. E. (1978). The Unity of the Senses. New York: for neuronal and hormonal influence. In Olfaction and
Academic Press. Endocrine Regulation, W. Breiphol (Ed.). London: IRL Press
Marr, J. N., and Gardner, L. E., Jr. (1965). Early olfactory expe- Ltd., pp. 223–236.
rience and later social behavior in the rat: Preference, sexual
Meredith, M. (1991). Sensory processing in the main and acces-
responsiveness, and care of young. J. Genet. Psychol. 107:
sory olfactory systems: comparisons and contrasts. J. Ster.
167–174.
Biochem. Mol. Biol. 39:601–614.
Marr, J. N., and Lilliston, L. G. (1969). Social attachment in rats
Meredith, M. (1998). Vomeronasal, olfactory, hormonal conver-
by odor and age. Behaviour XXXIII:277–282.
gence in the brain. Cooperation or coincidence? Ann. NY
Marsden, H. M., and Bronson, F. H. (1965). Strange male block
to pregnancy: its absence in inbred mouse strains. Nature Acad. Sci. 855:349–361.
207:878. Michael, R. P., and Keverne, E. B. (1968). Pheromones in the
Marshall, D. A., Doty, R. L., Lucero, D. P., and Slotnick, B. M. communication of sexual status in primates. Nature
(1981). Odor detection thresholds in the rat for the vapors of 218:746–749.
three related perfluorocarbons and ethylene glycol dinitrate. Michael, R. P., and Keverne, E. B. (1970a). A male sex-attractant
Chem. Senses 6:421–433. pheromone in rhesus monkey vaginal secretions.
Martin, I. G. (1980). “Homeochemic,” intraspecific chemical J. Endocrinol. 46:xx–xxi.
signal. J. Chem. Ecol. 6:517–519. Michael, R. P., and Keverne, E. B. (1970b). Primate sex
Maruniak, J. A., Desjardins, C., and Bronson, F. H. (1977). pheromones of vaginal origin. Nature 225:84–85.
Dominant-subordinate relationships in castrated male mice Michael, R. P., Keverne, E. B., and Bonsall, R. W. (1971).
bearing testosterone implants. Amer. J. Physiol. 233: Pheromones: isolation of male sex attractants from a female
E495–E499. primate. Science 172:964–966.
Mateo, J. M., and Johnston, R. E. (2000). Retention of social Michael, R. P., Zumpe, D., Keverne, E. B., and Bonsall, R. W.
recognition after hibernation in Belding’s ground squirrels. (1972). Neuroendocrine factors in the control of primate
Anim. Behav. 59:491–499. behavior. Recent Prog. Horm. Res. 28:665–706.
Matsunami, H., and Buck, L. B. (1997). A multigene family Moltz, H., and Leidahl, L. C. (1977). Bile, prolactin, and the
encoding a diverse array of putative pheromone receptors in maternal pheromone. Science 196:81–83.
mammals. Cell 90:775–784. Mombaerts, P., Wang, F., Dulac, C., Chao, S. K., Nemes, A.,
Mazur, A., Booth, A., and Dabbs, J. M. Jr. (1992). Testosterone Mendelsohn, M., Edmondson, J., and Axel, R. (1996).
and chess competition. Soc. Psychol. Q. 55:70–77. Visualizing an olfactory sensory map. Cell 87:675–686.
McCarty, R., and Southwick, C. H. (1977). Cross-species fos- Moncomble, A. S., Schaal, B., Quenedey, B., Quenedey, A.,
tering: Effects on the olfactory preference of Onychomys Coureaud, G., Brossut, R., Langlois, D., and Perrier, G. (2002).
torridus and Peromyscus leucopus. Behav. Biol. 19: In search of the exocrine origin(s) of the rabbit mammary
255–260.
pheromone. Presentation at the European Chemoreception
McCaul, K. D., Gladue, B. A., and Joppa, M. (1992). Winning,
Research Organization, Erlangen, Germany, 2002 (abstr).
losing, mood, and testosterone. Horm. Behav. 26:488–504.
Morris, N. M., and Udry, J. R. (1978). Pheromonal influences on
McClure, D., Barker, G., Barker, B., and Feil, P. (1987). Oral
human sexual behavior: an experimental search. J. Biosoc.
management of the cancer patient, part II: Oral complications
of radiation therapy. Compendium 8:88–2. Sci. 10:147–157.
McClure, T. J. (1959). Temporary nutritional stress and infertility Morton, J. R. C., Denenberg, V. H., and Zarrow, M. X. (1963).
in mice. Nature 181:1132. Modification of sexual development through stimulation in
Melrose, D. R., Reed, H. C., and Patterson, R. L. (1971). infancy. Endocrinology 72:439–442.
Androgen steroids associated with boar odour as an aid to the Mossman, C. A., and Drickamer, L. C. (1996). Odor preferences
detection of oestrus in pig artificial insemination. Brit. Vet. J. of female house mice (Mus domesticus) in seminatural enclo-
127:497–502. sures. J. Comp. Psychol. 110:131–138.
Mennella, J. A., and Beauchamp, G. K. (1991a). Maternal diet Mucignat-Caretta, C. Caretta, A., and Cavaggioni, A. (1995).
alters the sensory qualities of human milk and the nursling’s Pheromonally accelerated puberty is enhanced by previous
behavior. Pediatrics 88:737–744. experience of the same stimulus. Physiol. Behav. 57:901–903.
380 Doty

Mugford, R. A., and Nowell, N. W. (1970). Pheromones and their Pedersen, P. E., Stewart, W. B., Greer, C. A., and Shepherd, G. M.
effect on aggression in mice. Nature 226:967–968. (1983). Evidence for olfactory function in utero. Science
Mugford, R. A., and Nowell, N. W. (1971). Shock-induced 221:478–480.
release of the preputial gland secretions that eleicit fighting in Penn, D., and Potts, W. (1998). MHC-disassortative mating pref-
mice. J. Endocrinol. 51:xvi–xvii. erences reversed by cross-fostering. Proc. Roy. Soc. Lond. B:
Mugford, R. A., and Nowell, N. W. (1972). The dose-response to Biol. Sci. 265:1299–1306.
testosterone propionate of preputial glands, pheromones and Petrulis, A., and Johnston, R. E. (1995). A reevaluation of
aggression in mice. Horm. Behav. 3:39–46. dimethyl disulfide as a sex attractant in golden hamsters.
Müller-Schwarze, D. (1971). Pheromones in black-tailed deer Physiol. Behav. 57:779–784.
(Odocoileus heminonus columbianus). Anim. Behav. Pieper, D. R., and Newman, S. W. (1999). Neural pathway from
19:141–152. the olfactory bulbs regulating tonic gonadotropin secretion.
Müller-Schwarze, D. (1977). Complex mammalian behavior and Neurosci. Biobehav. Rev. 23:555–562.
pheromone bioassay in the field. In Chemical Signals in Pihet, S., Mellier, D., Bullinger, A., and Schaal, B. (1997).
Vertebrates, Müller-Schwarze, D., and Mozell, M. M. (Eds.). Réponses comportementales aux odeurs chez le nouveau-né
New York: Plenum Press, pp. 413–433. prématuré: étude préliminaire. Enfance 1:33–46.
Müller-Schwarze, D., and Muller-Schwarze, C. (1971). Porter, R. H., and Schaal, B. (2000). Olfaction and the develop-
Olfactory imprinting in a precocial mammal. Nature ment of social behavior in neonatal mammals. In Handbook of
229:55–56. Olfaction and Gustation, Doty, R. L. (Ed.). New York: Marcel
Müller-Schwarze, R., Müller-Schwarze, D., Singer, A. G., and Dekker.
Silverstein, R. M. (1974). Mammalian pheromone: Porter, R. H., and Winberg, J. (1999). Unique salience of mater-
Identification of active component in the subauricular scent of nal breast odors for newborn infants. Neurosci. Biobehav. Rev.
the male pronghorn. Science 183:860–862. 23:439–449.
Nakagawa, R., Tanaka, M., Kohno, Y., Noda, Y., and Nagasaki, N. Porter, R. H., Matochik, J. A., and Makin, J. W. (1983). Evidence
(1981). Regional responses of rat brain noradrenergic neurons for phenotype matching in spiny mice (Acomys cahirinus).
to acute intense stress. Pharmacol. Biochem. Behav. Anim. Behav. 31:978–984.
14:729–732. Quadagno, D., and Banks, E. M. (1970). The effects of recipro-
Nigrosh, B. J., Slotnick, B. M., and Nevin, J. A. (1975). Olfactory cal cross fostering on the behaviour of two species of rodents,
discrimination, reversal learning, and stimulus control in rats. Mus musculus and Baiomys taylori ater. Anim. Behav. 18:
J. Comp. Physiol. Psychol. 89:285–294. 379–390.
Norris, M. L., and Adams, C. E. (1979). Exteroceptive factors
Rajendren, G., and Dominic, C. J. (1985). Effect of transection of
and pregnancy block in the Mongolian gerbil, Meriones
the vomeronasal nerve on the male-induced implantation fail-
unguiculatus. J. Reprod. Fertil. 57:401–404.
ure (the Bruce effect) in mice. Indian J. Exp. Biol.
Novotny, M., Harvey, S., Jemiolo, B., and Alberts, J. (1985).
23:635–637.
Synthetic pheromones that promote intermale aggression in
Rajendren, G., and Dominic, C. J. (1987). The male-induced
mice. Proc. Natl. Acad. Sci. USA 82:2059–2061.
pregnancy block (the Bruce effect) in mice: re-evaluation of
Novotny, M., Jemiolo, B., and Harvey, S. (1990). Chemistry of
the ability of exogenous progesterone in preventing implanta-
rodent pheromones: molecular insights into chemical signal-
tion failure. Exp. Clin. Endocrinol. 89:188–196.
ing in mammals. In Chemical Signals in Vertebrates. Vol. 5,
Rajendren, G., and Dutta, A. K. (1988). Effect of haloperidol on
MacDonald, D. W., Muller-Schwarze, D., and Natynczuk, S. E.
(Eds.). Oxford: Oxford University Press, pp. 1–22. male-induced implantation failure (Bruce effect) in mice.
Nowell, N. W., and Wouters, A. (1973). The effect of cyproterone Indian J. Exp. Biol. 26:909–910.
acetate upon aggressive behavior in the laboratory mouse. J. Rasmussen, J. L., Rajecki, D. W., and Craft, H. D. (1993).
Endocrinol. 57:xxxvi–xxxvii. Humans’ perceptions of animal mentality: Ascriptions of
Nyakas, C., and Endröczi, E. (1970). Olfaction guided approach- thinking. J. Comp. Psychol. 107:283–290.
ing behaviour of infantile rats to the mother in maze box. Acta Reid, I. C., and Morris, R. G. (1993). The enigma of olfactory
Physiol. Acad. Sci. Hungar. 38:59–65. learning. Trends Neurosci. 16:17–20.
Parkes, A. S., and Bruce, H. M. (1961). Olfactory stimuli in Rejeski, W. J., Gagne, M., Parker, P. E., and Koritnik, D. R.
mammalian reproduction. Science 134:1049–1054. (1989). Acute stress reactivity from contested dominance in
Parkes, A. S., and Bruce, H. M. (1962). Pregnancy block in dominant and submissive males. Behav. Med. 15:118–124.
female mice placed in boxes soiled by males. J. Reprod. Reynolds, G. S. (1961). Attention in the pigeon. J. Exp. Anal.
Fertil. 4:303–308. Behav. 4:203–208.
Pause, B. M., Rogalski, K. P., Sojka, B., and Ferstl, R. (1999). Reynolds, J., and Keverne, E. B. (1979). The accessory olfactory
Sensitivity to androstenone in female subjects is associated system and its role in the pheromonally mediated suppression
with an altered brain response to male body odor. Physiol. of oestrus in grouped mice. J. Reprod. Fertil. 57:31–35.
Behav. 68:129–137. Rodriguez-Echandia, E. L., Foscolo, M., and Broitman, S. T.
Pedersen, P. E., and Blass, E. M. (1982). Prenatal and postnatal (1982). Preferential nesting in lemon-scented environment in
determinants of the 1st suckling episode in albino rats. Dev. rats reared on lemon-scented bedding from birth to weaning.
Psychobiol. 15:349–355. Physiol. Behav. 29:47–49.
Mammalian Pheromones 381

Rodriguez, I., Greer, C. A., Mok, M. Y., and Mombaerts, P. Schaal, B., Coureaud, G., Langlois, D., Giniès, C. Sémon, E., and
(2000). A putative pheromone receptor gene expressed in Perrier, G. (2003). An odorant for survival? The mammary
human olfactory mucosa. Nature Genetics 26:18–19. pheromone of the rabbit. Unpublished manuscript.
Ropartz, P. (1968). The relation between olfactory stimulation Schank, J. C. (2000). Menstrual-cycle variability and measure-
and aggressive behaviour in mice. Anim. Behav. 16:97–100. ment: further cause for doubt. Psychoneuroendocrinology
Rosser, A. E., and Keverne, E. B. (1985). The importance of cen- 25:837–847.
tral noradrenergic neurones in the formation of an olfactory Schank, J. C. (2001). Menstrual-cycle synchrony: problems and
memory in the prevention of pregnancy block. Neuroscience new directions for research. J. Comp. Psychol. 115:3–15.
15:1141–1147. Schellinck, H. M., and Brown, R. E. (1999). Searching for the
Rosser, A. E., Remfry, C. J., and Keverne, E. B. (1989). source of urinary odors of individuality in rodents. In
Restricted exposure of mice to primer pheromones coincident Advances in Chemical Signals in Vertebrates, Johnston, R. E.,
with prolactin surges blocks pregnancy by changing hypo- Muller-Schwarze, D, and Sorensen, P. W. (Eds.). New York:
thalamic dopamine release. J. Reprod. Fert. 87:553–559. Plenum, pp. 267–280.
Rowe, F. A., and Edwards, D. A. (1971). Olfactory bulb removal: Semke, E., Distel, H., and Hudson, R. (1995). Specific enhancement
Influences on the aggressive behaviors of male mice. Physiol. of olfactory receptor sensitivity associated with foetal learning of
Behav. 7:889–891. food odors in the rabbit. Naturwissenschaften 82:148–149.
Runner, M. N. (1959). Embryocidal effect of handling pregnant Shorey, H. H. (1976). Animal Communication by Pheromones.
mice and its prevention with progesterone. Anat. Rec. New York: Academic Press.
133:330–331. Shors, T. J., Pickett, J., Wood, G., and Paczynski, M. (1999).
Rutowski, R. L. (1981). The function of pheromones. J. Chem. Acute stress persistently enhances estrogen levels in the
Ecol. 7:481–484. female rat. Stress 3:163–171.
Sachs, B. D. (1997). Erection evoked in male rats by airborne Signoret, G. P. (1976). Chemical communication and reproduc-
scent from estrous females. Physiol. Behav. 62:921–924. tion in domestic mammals. In Mammalian Olfaction,
Sachs, B. D., Akasofu, K., Citron, J. H., Daniels, S. B., and Reproductive Processes, & Behavior., Doty, R.L. (Ed.). New
Natoli, J. H. (1994). Noncontact stimulation from estrous York: Academic Press, pp. 243–254.
females evokes penile erection in rats. Physiol. Behav. Singer, A. G. (1991). A chemistry of mammalian pheromones.
55:1073–1079.
J. Ster. Biochem. Mol. Biol. 39:627–632.
Sachser, N. (1986). The effects of long-term isolation on physiology
Singer, A. G., Agosta, W. C., O’Connell, R. J., Pfaffmann, C.,
and behavior in male guinea pigs. Physiol. Behav. 38:31–39.
Bowen, D. V., and Field, F. H. (1976). Dimethyl disulfide: an
Sahu, S. C., and Dominic, C. J. (1980). Chlorpromazine inhibi-
attractant pheromone in hamster vaginal secretion. Science
tion of the pheromonal block to pregnancy (the Bruce effect)
191:948–950.
in mice. Indian J. Exp. Biol. 18:1025–1027.
Singer, A. G., Clancy, A. N., Macrides, F., and Agosta, W. C.
Sahu, S. C., and Dominic, C. J. (1981). Failure of adrenalectomy
(1984). Chemical studies of hamster vaginal discharge: Male
to prevent the male-induced pregnancy block (Bruce effect) in
behavioral responses to a high molecular weight fraction
mice. Endokrinologie 78:156–160.
require physical contact. Physiol. Behav. 33:645–651.
Sam, M., Vora, S., Malnic, B., Ma, W., Novotny, M. V., and Buck,
Singer, A. G., Macrides, F., Clancy, A. N., and Agosta, W. C.
L. B. (2001). Neuropharmacology. Odorants may arouse
instinctive behaviours. Nature 412:142. (1986). Purification and analysis of a proteinaceous aphro-
Sarnat, H. B. (1978). Olfactory reflexes in the newborn infant. disiac pheromone from hamster vaginal discharge. J. Biol.
J. Pediatrics 92:624–626. Chem. 261:13323–13326.
Savic, I., Berglund, H., Gulyas, B., and Roland, P. (2001). Singer, A. G., Agosta, W. C., Clancy, A. N., and Macrides, F.
Smelling of odorous sex hormone-like compounds causes (1987). The chemistry of vomeronasally detected
sex-differentiated hypothalamic activations in humans. pheromones: characterization of an aphrodisiac protein. Ann.
Neuron 31:661–668. NY Acad. Sci. 519:287–298.
Schaal, B., Orgeur, P., and Marler, M. (1994). Amniotic fluid Singh, P. B., Herbert, J., Roser, B., Arnott, L., Tucker, D. K., and
odor in neonatal adaptation: A summary of recent research in Brown, R. E. (1990). Rearing rats in a germ-free environment
mammals. Adv. Biosci. 93:239–245. eliminates their odors of individuality. J. Chem. Ecol. 16:
Schaal, B., Orgeur, P., and Arnould, C. (1995). Olfactory prefer- 1667–1682.
ences in newborn lambs: Possible influences of prenatal expe- Sipos, M. L., Kerchner, M., and Nyby, J. G. (1992). An ephemeral
rience. Behaviour 132:351–365. sex pheromone in the urine of female house mice (Mus
Schaal, B., Marlier, L., and Soussignan, R. (1998). Olfactory domesticus). Behav. Neur. Biol. 58:138–143.
function in the human fetus: evidence from selective neonatal Slotnick, B. M. (1994). The enigma of olfactory learning revisited.
responsiveness to the odor of amniotic fluid. Behav. Neurosci. Neuroscience 58:1–12.
112:1438–1449. Slotnick, B. M. (2000). Can rats acquire an olfactory learning
Schaal, B., Marlier, L., and Soussignan, R. (2000). Human foe- set? Behav. Neurosci. 114:814–829.
tuses learn odours from their pregnant mother’s diet. Chem. Slotnick, B. (2002). Animal cognition and the rat olfactory sys-
Senses 25:729–737. tem. Trends Cogn. Sci. 5:216–222.
382 Doty

Slotnick, B. M., Kufera, A., and Silberberg, A. M. (1991). Suay, F., Salvador, A., Gonzalez-Bono, E., Sanchis, C., Martinez,
Olfactory learning and odor memory in the rat. Physiol. M., Martinez-Sanchis, S., Simon, V. M., and Montoro, J. B.
Behav. 50:555–561. (1999). Effects of competition and its outcome on serum
Smith, J. C. (1975). Sound communication in rodents. Symp. testosterone, cortisol and prolactin. Psychoneuro-
Zool. Soc. Lond. 37:317–330. endocrinology 24:551–566.
Smotherman, W. P. (1982). Odor aversion learning by the rat Swanson, H. H., and van de Poll, N. E. (1983). Effects of an iso-
fetus. Physiol. Behav. 29:769–771. lated or enriched environment after handling on sexual matu-
Smotherman, W. P., and Robinson, S. R. (1987). Prenatal expres- ration and behaviour in male and female rats. J. Reprod.
sion of species-typical action patterns in the rat fetus (Rattus Fertil. 69:165–171.
norvegicus). J. Comp. Psychol. 101:190–196. Takigami, S., Osada, T., Yoshida-Matsuoka, J., Matsuoka, M., Mori,
Smotherman, W. P., and Robinson, S. R. (1990). Rat fetuses respond Y., and Ichikawa, M. (1999). The expressed localization of rat
to chemical stimuli in gas phase. Physiol. Behav. 47: 863–868. putative pheromone receptors. Neurosci. Lett. 272:115–118.
Snyder, R. L., and Taggart, N. E. (1967). Effects of adrenalectomy Teicher, M. H., and Blass, E. M. (1976). Suckling in newborn
on male induced pregnancy block in mice. J. Reprod. Fertil. rats: Eliminated by nipple lavage, reinstated by pup saliva.
14:451. Science 193:422–425.
Sorensen, P. W. (1996). Biological responsiveness to pheromones Teicher, M. H., and Blass, E. M. (1977). First suckling response
provides fundamental and unique insight into olfactory func- of the newborn albino rat: The roles of olfaction and amniot-
tion. Chem. Senses 21:245–256. ic fluid. Science 198:635–636.
Sorensen, P. W., and Stacey, N. E. (1999). Evolution and special- Telle, H. J. (1966). Beitrag zur Kenntnis der Verhaltensweise von
ization of fish hormonal pheromones. In Advances in Ratten, vergleichend dargestellt bei Rattus norvegicus und
Chemical Signals in Vertebrates, Johnston, R. E., Müller- Rattus rattus. Z. Ang. Zool. 53:129–196.
Schwarze, D., and Sorensen, P. W. (Eds.). New York: Kluwer Terman, C. R. (1969). Pregnancy failure in female prairie deer-
Academic, pp. 15–47. mice related to parity and social environment. Anim. Behav.
Spencer, J., Gray, J., and Dalhouse, A. (1973). Social isolation in 17:104–108.
the gerbil: Its effect on exploratory or agonistic behavior and The Random House College Dictionary, revised ed. (1975). New
adrenocortical activity. Physiol. Behav. 10:231–237. York: Random House, Inc.
Spironello, E., and deCatanzaro, D. (1999). Sexual satiety dimin- Thiessen, D. D., Regnier, F. E., Rice, M., Goodwin, M., Isaacks, N.,
ishes the capacity of novel males to disrupt early pregnancy in and Lawson, N. (1974). Identification of a ventral scent mark-
inseminated female mice (Mus musculus). J. Comp. Psychol. ing pheromone in the male Mongolian gerbil (Meriones
113:218–222. unguiculatus). Science 184:83–85.
Spironello-Vella, E., and deCatanzaro, D. (2001). Novel male Tirindelli, R., Mucignat-Caretta, C., and Ryba, N. J. (1998).
mice show gradual decline in the capacity to disrupt early Molecular aspects of pheromonal communication via the
pregnancy and in urinary excretion of testosterond and 17- vomeronasal organ of mammals. Trends Neurosci.
estradiol during the weeks immediately after castration. 21:482–486.
Horm. Metab. Res. 33:681–686. Torner, L., Toschi, N., Pohlinger, A., Landgraf, R., and Neumann,
Stark, B., and Hazlett, B. A. (1972). Effects of olfactory experi- I. D. (2001). Anxiolytic and antistress effects of brain pro-
ence on aggression in Mus musculus and Peromyscus manic- lactin: Improved efficacy of antisense targeting of the pro-
ulatus. Behav. Biol. 7:265–269. lactin receptor by molecular modeling. J. Neurosci.
Stedman’s Medical Dictionary, 26th ed. (1999). Baltimore: 21:3207–3214.
Williams and Wilkins. Vandenbergh, J. G. (1967). Effect of the presence of a male on
Stehn, R. A., and Richmond, M. E. (1975). Male-induced preg- the sexual maturation of female mice. Endocrinology
nancy termination in the prairie vole, Microtus ochrogaster. 81:345–349.
Science 1876:1211–1213. Vandenbergh, J. G. (1969). Male odor accelerates female sexual
Stern, K., and McClintock, M. K. (1998). Regulation of ovulation maturation in mice. Endocrinology 84:658–660.
by human pheromones. Nature 392:177–179. Vierling, J. S., and Rock, J. (1967). Variations in olfactory sensi-
Stickrod, G., Kimble, D. P., and Smotherman, W. P. (1982). In tivity to exaltolide during the menstrual cycle. J. Appl.
utero taste/odor aversion conditioning in the rat. Physiol. Physiol. 22:311–315.
Behav. 28:5–7. Wang, H. W., Wysocki, C. J., and Gold, G. H. (1993). Induction
Storey, A. E., and Snow, D. T. (1990). Postimplantation pregnancy of olfactory receptor sensitivity in mice. Science
disruptions in meadow voles: relationship to variation in male 260:998–1000.
sexual and aggressive behavior. Physiol. Behav. 47:19–25. Watson, M., Clulow, F. V., and Mariotti, F. (1983). Influence of
Strassmann, B. I. (1999). Menstrual synchrony pheromones: olfactory stimuli on pregnancy of the meadow vole, Microtus
cause for doubt. Hum. Reprod. 14:579–580. pennsylvanicus, in the laboratory. J. Mammal. 64:706–708.
Strott, C. A., Sundel, H., and Stahlman, M. T. (1975). Maternal Webster’s New Collegiate Dictionary. (1999). Springfield, MA:
and fetal plasma progesterone, cortisol, testosterone and G. & C. Merriam Company.
17-estradiol in preparturient sheep: response to fetal ACTH Weir, M. W., and DeFries, J. C. (1963). Blocking of pregnancy in
infusion. Endocrinology 95:1327–1322. mice as a function of stress. Psychol. Rep. 13:365–366.
Mammalian Pheromones 383

Wierson, M., Long, P. J., and Forehand, R. L. (1993). Toward a Yamazaki, K., Boyse, E. A., Mike, V., Thaler, H. T., Mathieson,
new understanding of early menarche: the role of environ- B. J., Abbott, J., Boyse, Zayas, Z. A., and Thomas, L.
mental stress in pubertal timing. Adolescence 28: 913–924. (1976). Control of mating preferences in mice by genes in
Whitten, W. K. (1956). Modification of the oestrous cycle of the the major histocompatibility complex. J. Exp. Med.
mouse by external stimuli associated with the male. 144:1324–1335.
J. Endocrinol. 13:399–404. Yamazaki, K., Beauchamp, G. K., Kupniewski, D., Bard, J.,
Whitten, W. K. (1966). Pheromones and mammalian reproduc- Thomas, L., and Boyse, E. A. (1988). Familial imprinting
tion. In Advances in Reproductive Physiology, McLaren, A. determines H-2 selective mating preferences. Science
(Ed.). New York: Academic Press, pp. 159–177. 240:1331–1332.
Whitten, W. (1999). Pheromones and regulation of ovulation. Yamazaki, K., Singer, A., and Beauchamp, G. K. (1998). Origin,
Nature 401:232–233. functions and chemistry of H-2 regulated odorants. Genetica
Whitten, W. K., and Champlin, A. K. (1972). Bibliography (with 104:235–240.
review) on pheromones and olfaction in mammalian repro- Yoneda, Y., Kanmori, K., Ida, S., and Kuriyama, K. (1983).
duction. Bibliography Reprod. 19:149–156, 297–303. Stress-induced alteration in metabolism of aminobutyric acid
Whitten, W. K., and Champlin, A. K. (1973). The role of olfac- in rat brain. J. Neurochem. 40:350–356.
tion in mammalian reproduction. In Handbook of Physiology. Yoshimura, H. (1980). Cholinergic mechanisms in scent marking
Section 7: Endocrinology. Washington, DC: American behavior by Mongolian gerbils (Meriones unguiculatus).
Physiological Society, pp. 109–123. Pharmacol. Biochem. Behav. 13:519–523.
Wilson, E. O. (1963). Pheromones. Sci. Am. 208:100–114. Youngentob, S. L., and Kent, P. F. (1995). Enhancement of odor-
Wilson, E. O., and Bossert, W. H. (1963). Chemical communica- ant-induced mucosal activity patterns in rats trained on an
tion among animals. Rec. Prog. Horm. Res. 19:673–710. odorant identification task. Brain Res. 670:82–88.
Wilson, H. C. (1992). A critical review of menstrual synchrony Zacharias, R., de Catanzaro, D., and Muir, C. (2000). Novel
research. Psychoneuroendocrinology 17:565–591. male mice disrupt pregnancy despite removal of vesicular-
Wöhrmannj-Repenning, A. (2000). The human vomeronasal coagulating and preputial glands. Physiol. Behav.
organ and pheromonal communication: facts and fantasy. 68:285–290.
Z Säugetierkunde 65:380–382. Zondek, B., and Tamari, I. (1967). Effects of auditory stimuli on
Wysocki, C. J., and Meredith, M. (1987). The vomeronasal sys- reproduction. Ciba Foundation Study Group 26:4–19.
tem. In The Neurobiology of Taste and Smell, Finger, T. E.,
and Silver, W. L. (Eds.). New York: Wiley, pp. 125–150.
18

Psychophysical Evaluation of Olfaction in Nonhuman Mammals

Lloyd Hastings
University of Pennsylvania, Philadelphia, Pennsylvania, U.S.A.

1. INTRODUCTION mechanisms underlying olfactory function. With the devel-

opment of psychophysics, attention turned to investigating
the relationship between changes in physical stimuli and
Experimental study of sense organs must be made on the resulting psychological sensations in a quantitative
man because animals can not directly account to us manner.
the sensations which they experience. The initial objectives of psychophysical research
Claude Bernard, 1865 were to determine the minimum detectable energy levels
of stimulation (absolute threshold) and the minimum
Results of studies employing animals in sensory research detectable difference between values on a continuum,
over the last 50 years have shown the above statement to be e.g., difference thresholds. Numerous psychophysical
categorically wrong. In fairness, Bernard was partially right techniques have since been developed to assess sensitiv-
in that animals cannot directly convey their sensory experi- ity, including ones that challenge the concept of a
ences. However, he erred when he assumed that the only threshold as a fixed point, e.g., signal detection theory.
valid method for investigating sensory experience was by Other techniques have been developed to investigate
verbal report. His summary dismissal of animal testing can how people respond to odor stimuli that fall in the
be better understood when it is realized that the field of ani- suprathreshold range, i.e., odors that are clearly percep-
mal experimental psychology, pioneered by Thorndike tible. The latter includes procedures to examine (1) odor
(1911) and Watson (1914), had not yet been developed. quality discrimination/generalization, (2) odor quality
This chapter will show that, contrary to Bernard’s supposi- recognition/identification, (3) odor intensity, and (4)
tion, behavioral testing of animals can provide much impor- odor memory.
tant information about sensory systems, including In this chapter I describe various methods for assess-
olfaction, when the appropriate methodology is employed. ing olfactory function in animals. The focus is on dogs,
Moreover, animal testing is the only way in which both rats, and mice, given their widespread use in laboratory
behavioral testing and experimental physiological proce- settings, although the basic procedures that are reviewed
dures, e.g., lesions, histological examination, genetic are applicable to other forms. These methods allow for
manipulation, etc., can be undertaken in the same organism. determining the effects of numerous experimental
Early attempts to understand the olfactory system manipulations on the ability to smell, including influ-
focused on developing classification systems based upon ences of drugs, brain lesions, and genetic manipulations.
various “primary” odor groups. Since these attempts were The reader is referred to Hübener and Laska (2001) for
usually subjective, there was little agreement and, con- a recent review of nonhuman primate odor discrimination
sequently, little progress made in understanding the basic paradigms.

385
386 Hastings

II. HISTORICAL BACKGROUND the odors of other animals. His subjects also responded,
however, to other odors such as camphor, carbon disulfide,
A. Early Canine Studies and vanillin, which did not have any apparent biological
significance. Using the same technique, Seffrin (1915)
The olfactory powers of a number of animals have attempted to determine some “minima perceptible,” i.e.,
attracted considerable attention through the years, but the the absolute threshold, for some of the pure substances
interest was usually casual in nature. Early observations from Zwaardemaker’s classes, as well as a number of ani-
were made concerning the ability of animals, like deer mal products such as urine.
and kangaroos, to smell human scent from great distances, While the attempts to control the concentrations of the
and of the ability of dogs and other animals to track game olfactory stimuli in these early studies can, at best, be con-
and to find escaped convicts. It was not until the late nine- sidered crude, and the accuracy of the actual data collected
teenth and early twentieth centuries, however, that system- as being suspect, these studies, along with the others still
atic application of the scientific method was employed to to be described in this chapter, nevertheless paved the way
better understand such abilities. In 1910, Schmidt pub- for the more rigorously controlled experiments of the last
lished one of the first books on canine tracking, half of the twentieth century.
Verbrecherspur und Polizeihund, which placed great In 1916, Henning conducted studies of discrimination
emphasis on the necessity for the experimental controls behavior by training dogs to select a faintly perfumed
and testing of hypotheses, in contrast to most contempo- handkerchief from a number of unadulterated ones.
rary approaches that were generally preoccupied with Conversely, he also trained dogs to ignore strong animal
results and gave little detail on test conditions or proce- odors to which they normally responded vigorously. He
dures. Nearly two decades earlier, Passy (1892) conducted concluded that methods based on motor responses revealed
one of the first empirical studies of odor preferences in nothing about the olfactory acuity or hedonics of an odor,
dogs (see Chapter 19 for a discussion of odor preference since absence of response did not necessarily indicate the
paradigms). Paper was dipped into alcoholic solutions of inability to perceive the odor. This early observation pre-
odorous substances and allowed to dry. When presented to saged the necessity of well-developed instrumental and
a dog, a preference was scored if the dog smelled them operant conditioning paradigms that clearly define the
attentively and tried to seize and eat the paper. If the dog nature of the response to an odorant stimulus as a prerequi-
turned its head away, a negative odor preference was site for studying animal behavior and olfactory psy-
scored. In this study, preference was operationally defined chophysics.
by the dog’s movement; motivational status of the subject According to McCartney (1968), Neuhaus conducted
was not considered. Thus, this early investigator was able the first well-controlled studies of canine olfaction in the
to ascertain some quantitative information from animals early 1950s. Neuhaus developed not only an apparatus that
about the perceived hedonics of olfactory cues. attempted to regulate the olfactory cues presented to dogs,
In 1907, Hamilton devised a operant box from which a but also a protocol that incorporated a forced-choice para-
young dog was rewarded by being allowed to escape if the digm. He determined the absolute thresholds for a variety
correct of four pedals was pushed. Visual, olfactory, and of pure compounds as well as mixtures of compounds. He
other cues were associated with each pedal and random- also performed one of the earliest animal studies to inves-
ized across trials. Unfortunately, the olfactory component tigate the increase in stimulus concentration that was nec-
of the study was attenuated when shock was employed to essary to be judged different from a second one, i.e., a just
punish wrong responding, leading the subject to freeze noticeable difference (JND). In addition to his behavioral
thereafter when placed in the test situation. More extensive studies, Neuhaus performed histological examination of
studies on the olfactory sensitivity or discrimination ability the canine olfactory region and counted the number of
of dogs appeared in Germany around the same time, such cells in different breeds of dogs. Finally, he included in his
as when Kalischer (1909) trained dogs to take food only studies an investigation of the influences on olfactory
when it was associated with a certain odor. He then pro- thresholds of orally administered agents.
ceeded to studies involving odor mixtures, where he mixed About the same time of Neuhaus’s studies, researchers
the original odor with as many as three or four additional in the United States and England initiated studies in canine
odoriferous compounds before presenting the stimuli to olfaction. Using a technique that insured the dog was
the dog. Subsequently, Heitzenrohder (1913) recorded the responding only to olfactory cues and employing a large
respiratory movements of a dog using kymographical number of trials, Ashton et al. (1957) determined the
recordings, in an attempt to be more quantitative. He absolute threshold for eight straight-chain fatty acids from
found, not surprisingly, that dogs responded forcefully to formic to caprylic acid. When the threshold values obtained
Psychophysical Evaluation of Olfaction in Nonhuman Mammals 387

in this study were compared with those obtained by olfaction is not generally perceived as being as vital to
Neuhaus, they were found to differ by a factor of 107. These humans as sight or hearing. In addition, another major
findings point out a monumental problem in olfactory impediment has been the inability to distinguish, a priori,
research that does not presently exist for other sensory sys- between receptors that respond to different odorants. With
tems—namely, the marked difficulty in accurately measur- the discovery of the large multigene family for olfactory
ing the stimulus reaching the nose of the subject. This is receptors by Buck and Axel (1991) and greater knowledge
especially true in threshold studies, where stimulus concen- of the molecular events in the transduction process, tremen-
trations may be well below the detection limits of available dous progress is now being made in delineating the basic
instruments. The similar inability to easily and accurately mechanisms of olfactory function. Research on olfactory
measure qualitative differences in olfactory stimuli also function that relies on behavioral measures, however, has
contributes to the uncertainty that accompanies many stud- not experienced a corresponding resurgence in growth.
ies involving olfactory function and makes comparisons Those researchers interested primarily in the behavioral
between studies difficult. expression of olfaction, especially in animals, have always
represented only a small minority in the field of olfactory
B. Ascension of Rodents in Olfactory Research research. This state of affairs is again due to a variety of fac-
tors, but with the development of sophisticated operant and
While most psychophysical studies of olfactory function instrumental conditioning paradigms, the presumed inabil-
during the first half of the twentieth century used dogs as ity of animals to convey information about the sensory sys-
subjects, the laboratory rat was making its ascent to promi- tems, as asserted by Bernard (1865), is not a primary
nence in psychological research. Liggett (1928) and reason. More likely the major impediment has been techni-
Swann (1933) investigated olfactory function in rats using cal in nature, such as difficulties in accurately generating,
the buried food test—a test still widely used today. In the controlling, and measuring olfactory stimuli. Not until the
1940s, Stone (1941) and Lashley and Sperry (1943) devel- development of precision olfactometers, which allowed
oped procedures for investigating olfactory discrimination some control over the olfactory stimuli, was it possible to
behavior in rats using simple choice situation, which, conduct meaningful studies on such psychophysical mea-
unfortunately, allowed only crude control of relevant sures as absolute thresholds.
stimulus parameters. Among the first attempts to use oper- Unlike the production of visual or auditory stimuli,
ant conditioning techniques, as well as a functional olfac- generating and controlling olfactory stimuli are more
tometer to control the presentation of stimuli, were studies arduous tasks. First, an olfactory stimulus must be
of Pfaffmann et al. (1958) and Goff (1961). These investi- volatile; if the compound is a liquid or solid, it must be
gators used flow-dilution olfactometers to accurately pre- converted to a vapor phase. To achieve the desired con-
sent the odorant stimuli and changes in the performance of centrations of a stimulus, a known quantity of the vapor
various schedules of reinforcement to determine absolute phase is mixed with varying quantities of background air
thresholds. Once animal behavior could be reliably deci- by the process of flow dilution. Also, in contrast to visual
phered through the use of operant conditioning techniques or auditory stimuli—which can be easily generated and
and olfactory stimuli could be controlled (if not measured), turned on and off with great precision—olfactory stimuli
the laboratory rat became a most useful subject in the linger until dispersed by diffusion or scavenged from the
investigation of olfactory function. Soon, more precise surrounding environs by a vacuum source. Furthermore,
olfactometers were developed, as were more sophisticated the presentation and removal of olfactory stimuli must be
operant conditioning protocols, and the investigation of accomplished without producing extraneous cues, such as
olfactory function using animal behavioral assessment auditory cues or changes in temperature, flow, or pressure.
techniques began in earnest (Braun and Marcus, 1969; Construction of an olfactometer that is capable of produc-
Davis, 1973; Pierson, 1974; Nigrosh et al., 1975). ing olfactory stimuli without such confounding artifacts
requires considerable time and effort, as well as money
(for review, see Prah et al., 1995).
III. METHODOLOGICAL ISSUES
Although olfactometers have become more sophisticated
A. Stimulus Generation and Control and more precise, due mainly to the incorporation of com-
puters and mass flow controllers in their design, one contin-
Until recently, research investigating the mechanisms uing deficiency in many applications is the failure to
underlying olfactory function has lagged behind similar measure and verify the olfactory stimuli produced. Once
research in vision and audition. This is due to a variety of again, unlike in auditory and visual research, there is no
reasons, but probably first and foremost is the fact that widely available instrument that can easily and routinely be
388 Hastings

used to accurately measure olfactory stimuli. Although great Numerous studies have shown that (1) the threshold for
care may be taken in generating the stimuli, problems with activating the trigeminal system for most compounds is
leaks in the system, loss due to adsorption on the instru- much higher than the olfactory threshold (Tucker, 1971),
ments walls, and/or contamination of the stock odorant can and (2) the trigeminal system can be eliminated and the
substantially alter the actual stimulus that reaches the sub- olfactory system can still function in a normal manner
ject. The failure to verify stimulus concentration, especially (Silver et al., 1985). What has not been addressed, how-
in threshold measurements, undoubtedly contributes to the ever, is whether the trigeminal system, when the olfactory
large variability often found in published reports (Cain, system has been compromised or destroyed, can be used
1977; Stevens et al., 1988). When stimuli are measured, gas by animals to detect chemosensory stimuli to guide subse-
chromotography is usually employed, but the technology is quent behavior. The possibility has been suggested that in
complicated and costly. A major challenge facing the field humans, albeit in a minor way, the olfactory and trigemi-
of olfactory research is finding new and more sensitive nal systems interact and that the sensitivity of one may
methods of measuring olfactory stimuli. Potential method- change when the other is altered (Bouvet et al., 1987;
ologies include “electronic noses” (see Chapter 14), but so Livermore et al., 1992). To what degree this is true in non-
far there has been very little use of these instruments in human mammals is not known.
olfactory research. The vomeronasal organ (VNO), which shares many
morphological and embryonic similarities with the olfac-
B. Involvement of Other Systems in Olfaction tory system, is located in the anterior portion of the nasal
cavity. The sensory fibers of the VNO terminate in the
Besides the difficulties inherent in dealing with the genera- accessory olfactory bulb instead of the main olfactory
tion and control of olfactory stimuli, care must also be bulb, unlike those of the olfactory neuroepithelium. The
taken to assure that the animal is responding only to olfac- VNO appears to mediate sensory information important
tory cues, and not to cues from some other sensory system. for reproductive physiology and behavior, although it does
While Cranial Nerve I (CN I) is the major neuronal system respond to some volatiles that are detected by CN I
involved in olfactory function, there are other neuronal sys- (Johnston, 1998). While olfactory function appears normal
tems within the nasal cavity that may contribute to or at in the absence of the VNO (Brouette-Lahlou et al., 1994),
least subtly modulate the sense of smell. These include the and CN I and VNO neuronal pathways are relatively inde-
trigeminal system (CN V), the vomeronasal organ pendent of one another, there is a remote possibility the
(Jacobson’s organ), the septal organ of Masara, and the VNO can mediate olfactory cues in the absence of a func-
nervus terminalis. If understanding how the olfactory sys- tional olfactory system. Such information is noteworthy,
tem functions is the primary goal, then the degree, if any, to since damage to the VNO is often minimized during expo-
which these other neuronal systems contribute to the sense sure to airborne toxic agents, compared to the olfactory
of olfaction should be determined. Inherent in this state- epithelium (Gaafar et al., 1992; Youngentob et al., 1997).
ment is the implication that the function or purpose that The third neuronal system in the nasal cavity, the septal
these other systems perform or subserve is known; unfortu- organ (SO) of Masara, is a small patch of sensory epithe-
nately, this is not entirely the case. One perplexing issue lium located bilaterally on the septal wall in close proximity
germane to this topic is the fact that when most of the olfac- to the VNO. The structure of the neuroepithelium of the SO
tory epithelium in rodents is destroyed, olfactory sensitivity is very similar to that of the main olfactory epithelium, and
does not appear to be greatly affected (Hastings et al., 1991; its afferent fibers terminate in the caudal part of the olfac-
Youngentob et al., 1997). Two theories exist to explain this tory bulb on glomeruli known as “septal glomeruli.” It was
phenomenon: (1) only a small percentage of the olfactory originally hypothesized that, due to its proximity to the
epithelium is actually needed to subserve this element of entrance of the nasal cavity, it may have an alerting func-
olfactory function, and/or (2) the animal is responding to tion (Rodolfo-Masera, 1943). That is, by continuously sam-
cues generated by one of these other neuronal systems in pling the incoming olfactory stimuli during periods of rest
the absence of a functional olfactory system. Some (sleep), it performed an alerting function, which in turn
attempts have been made to establish the relative contribu- might modify the function of the main olfactory system. An
tions of the different systems to olfactory function, but empirical test of this hypothesis (Giannetti et al., 1995)
definitive information about the potential interaction of showed this not to be the case. No other role for the SO has
these other systems with olfaction is still lacking. been proposed, and its true functions remain unknown.
Of the four, the trigeminal system (CN V) has been the The final neuronal system to innervate the olfactory
most extensively studied. Its primary function is to detect epithelium is the nervus terminalis (NT), a plexiform,
and respond to airborne irritants (see Chapter 47). ganglionated nerve originating in the epithelium of the
Psychophysical Evaluation of Olfaction in Nonhuman Mammals 389

VNO (Schwanzel-Fukuda and Pfaf, 1995). Branches of the A. Assessment of Perceived Intensity
NT intermingle with afferent fibers of both the VNO and
the trigeminal nerve as they course through the septal Determination of olfactory thresholds has long been con-
mucosa. The fibers eventually terminate in the accessory sidered the sine qua non parameter for best describing the
olfactory bulb as well as in the olfactory tubercle, the sep- overall function of the olfactory system. The absolute
tal, the precommissural, and the preoptic areas of the brain. threshold is the measure of the minimal odorant concen-
The function of the NT is poorly understood. It is possible tration that can be detected from clean air, while the recog-
that this nerve may be part of a luteinizing hormone–releas- nition threshold is the minimal amount that can not only be
ing hormone (LHRH) system involved in regulating the detected, but also identified. One variation of the absolute
VNO “pump” (Wirsig-Wiechmann and Lepri, 1991). threshold measure introduced in recent years is the odor
Another possible function includes mediation of some rather mixture threshold (Doty et al., 1999; Lu and Slotnick,
specific chemosensory responses (Kyle et al., 1987). 1998; Xu and Slotnick, 1999). In this procedure, the con-
Although the function of the olfactory nerve and the centration of the test odorant is varied within the context of
trigeminal system is apparent, and at least certain attributes a second odorant, whose concentration remains constant
of the VNO understood, the close proximity and intermin- and which also serves as the “blank.” The assumption,
gling of the fibers of the SO and NT make it very difficult which may or may not be valid, is that since there is more
to experimentally determine their actual function, as well “interference” in the system, the task should be more diffi-
as any role they may play in olfaction. If one is only inter- cult to perform and, thus, more sensitive to any perturba-
ested in determining some psychophysical measure such tion in the system.
as an absolute threshold in an intact, functioning animal,
then there is little need to know the relative contributions
1. Establishing an Absolute Threshold
that each of these subsytems may be providing to the
via Operant Procedures
process of olfaction. On the other hand, if one is interested
in discerning some specific process of the olfactory system While the goal is the same, there are a variety of ways to
such as recovery of olfactory function after toxic insult or determine the absolute threshold, including stimulus pre-
genetic manipulation of a gene for a specific odorant sentation procedures, such as the method of limits, stair-
receptor, it then becomes very important to know the ori- cases with reversals, and the method of constant stimuli
gin of the sensory information to which the animal is (Figs. 1, 2). In all such procedures, the animal is taught to
responding. Too often in the past, this information has not make some conditioned operant response, usually a bar
been known or even considered. press, nose poke, or lick, when an olfactory cue is detected
and to withhold that response when only clean air is pre-
sented. This is termed a go/no-go paradigm. On rare occa-
IV. PSYCHOPHYSICAL EVALUATION sions, a go/go differential response paradigm is used, e.g.,
OF OLFACTION go-right if odor is present/go-left if odor is absent. It is
used less frequently, however, since more training trials are
Stebbins (1970) defined animal psychophysics as an area of usually necessary to acquire the task. Studies that have
research in which the primary concern is the behavioral used some type of operant conditioning paradigm to assess
analysis of sensory function. The basic data consist of condi- absolute detection thresholds are presented in Table 1. In
tioned responses obtained from awake, intact animals in most cases the absolute threshold measure was used to
response to sensory stimulation. Function of the sensory sys- evaluate the functional status of the olfactory system after
tem is then inferred from observation of the overt behavioral some form of insult, such as an anatomical lesion (olfac-
response. A change in the perception of a measureable para- tory bulb or CNS pathway), hormonal manipulation, or
meter of an olfactory stimulus by the animal is reflected in a chemical damage of the olfactory neuroepithelium. Most
corresponding change in its conditioned behavior. As indi- studies have used either mice or rats, although the absolute
cated earlier, psychophysical evaluation of olfaction in ani- threshold has been examined in dogs for a few specific
mals did not really begin until progress was made in stimulus compounds (Fig. 2).
control, i.e., olfactometry, and in behavioral analysis in ani- In the method of limits, olfactory stimuli consisting of
mals, i.e., operant conditioning techniques. Earlier reviews of a series of either ascending or descending concentrations
this topic include those by Passe and Walker (1985), Slotnick of the odorant are presented to a subject. When used with
(1990), and Walker and Jennings (1991). In this section, pro- animals, usually a descending series is used in order to
tocols for assessing the perceived intensity and quality of maintain stimulus control. Threshold is reached when the
olfactory stimuli in animals are described. subject’s ability to correctly detect the stimulus reaches
390 Hastings

values, usually in a randomized sequence, is presented and

the threshold calculated from the generated response prob-
abilities for the various stimuli. The method of constant
stimuli has rarely been used in animal olfactory research,
largely because of the large number of trials that are
required and the propensity for adaptation.

2. Establishing an Absolute Threshold Using

Classical Conditioning or Related Conditioned
Responses
While the majority of studies investigating olfactory func-
tion have used tasks based upon operant conditioning tech-
niques, there are some that employ classical conditioning
paradigms. These include studies based upon both
conditioned avoidance (or approach) and conditioned sup-
pression paradigms. In the conditioned avoidance para-
digm, an odor is paired with negative reinforcement
(positive reinforcement for conditioned approach) and the
animal learns to perform some form of avoidance (or
approach) response whenever it detects the odor. By suc-
cessively lowering the concentration and noting when the
conditioned behavior stops, i.e., when the animal can no
longer detect the odor, a threshold measure is obtained.
The conditioned suppression paradigm is very similar. The
animal learns to suppress an ongoing operant response,
e.g., licking, whenever it detects a specific odorant. As
before, a threshold is determined by successively lowering
the concentration and identifying the point at which the
animal no longer suppresses its behavior. Both the method
of limits and staircase stimulus presentation procedures
Figure 1 (top) Schematic diagram of the air dilution olfac- can be used with either technique.
tometer used to assess sensitivity to ethyl acetate vapor. D1, D2, Table 2 lists a number of studies that have made absolute
D3, and D4 are successive dilution stages. The airflows in chan- threshold determinations using classical conditioning/con-
nels A (odor) and B (clean air) were set at 0.1 L/min. The flow in ditioned response procedures. Included in this listing are
channel C (carrier flow) was set at 1.9 L/min. (bottom) Schematic three recent studies which have used an olfactory cue as the
airflows in channels A and B were set at 0.05 and 1.95 L/min
conditioned stimulus in a conditioned reflex paradigm
respectively. (From Bodyak and Slotnick, 1999.)
(Hunt et al., 1997; Nsegbe et al., 1998; Richardson et al.,
1999). While these studies did not actually measure olfac-
tory detection thresholds, they could be easily modified to
do so. As with the other examples, one only has to lower the
chance or some other prescribed criteria. In the staircase concentration after the behavior has been established and
method, the same procedure is used as the method of lim- look for alterations in the behavior. The benefits of using a
its, but the threshold region is traversed back and forth. conditioned reflex, like changes in heart rate, ventilation, or
The stimulus concentration continues to increase until a startle response, are that little training is required—com-
prescribed number of correct responses are made, and then pared to operant techniques—and that the techniques can
the direction of stimulus change reverses until poor per- be adapted to test very young animals.
formance occurs. This procedure is repeated for a set num-
ber of reversals and then the reversal values are averaged. 3. Assessment of Suprathreshold Stimuli
Thresholds obtained with this procedure are usually more
reliable than those obtained with the standard method of In humans, suprathreshold odor intensity is usually studied
limits, since more data are used in the determination. In by either quantifying the growth in the magnitude of
the method of constant stimuli, a fixed set of stimulus odor sensation as the odor concentration increases
Psychophysical Evaluation of Olfaction in Nonhuman Mammals 391

control of the olfactory stimulus, along with the ability to

verify the actual concentrations presented to the subject, are
required for the collection of reliable data.
While the use of olfactory intensity-difference thresholds
appears to have several advantages over the measurement of
olfactory absolute thresholds, only a few such studies have
been reported in the animal literature. Using this technique,
Slotnick and Ptak (1974) compared olfactory intensity-dif-
ference thresholds in rats and humans. For rats, a go/no-go
discrete trials, successive stimulus presentation was
employed. Rats were exposed to either a reference concen-
tration or to one of a series of concentrations, which varied
according to a modified ascending method-of-limits proce-
dure. The olfactory-difference threshold (Weber fraction)
obtained was approximately 0.03 (compared to ~0.3 for
humans). In two later studies (Slotnick and Schoonover,
1984; Slotnick et al., 1997), a Weber fraction approximately
10 times as great (or the same as humans in the previous
study) was obtained. This difference was attributed to vari-
ations in the training protocol, reinforcing the principle that
Figure 2 Apparatus for testing dogs in an odor-detection task.
if behavioral measures are to be used to obtain sensory
The test chamber is housed in a controlled environment room
occupying part of a laboratory. For the purposes of illustration,
information in animals, such measures must be well
many details have been simplified or omitted. (For example, a gas defined. When used as a diagnostic test to evaluate whether
chromatograph and water reservoir bottles are normally housed olfactory function had been impaired by an experimental
on the roof of the chamber, and an air conditioning unit and procedure, no effects were observed on the intensity-differ-
purification stages lie on the roof of the room.) The olfactometer ence threshold in rats after lesions of the anterior amygdala
is shown in semi-schematic form. (From Moulton, 1977.) (Slotnick, 1985). However, transection of the lateral olfac-
tory tract (Slotnick and Schoonover, 1993) and application
(suprathreshold scaling) or by measuring the ability to of intranasal zinc (Slotnick and Gutman, 1977) did have a
detect small changes in odor concentration (differential greater effect on intensity-difference thresholds than on
threshold or JNDs) (Walker and Jennings, 1992). In human absolute detection thresholds. This suggests that the inten-
suprathreshold scaling studies, the increase in perceived sity-difference threshold test might be a more sensitive mea-
intensity is quantified either by verbal report or by some sure for assessing damage to the olfactory system than the
cross-modal manipulation, e.g., adjusting the length of a more frequently measured absolute threshold. Table 3 lists
line or arrow. No comparable paradigm is available for use studies that have measured olfactory intensity-difference
with animals, so tests of differential sensitivity have been thresholds in rats.
used with animals to investigate perception of suprathresh-
old stimulus intensity. B. Assessment of Perceived Quality
When animals are asked to discriminate between two dif-
ferent odorants, it is assumed they can do so based on qual- Besides odor intensity, the other most frequently examined
itative differences. However, since it is difficult to match the parameter is odor quality. It is the qualitative differences of
intensities of two qualitatively different odors, the animals olfactory stimuli that allow us to perceive, discriminate,
could be responding to differences in intensity rather than and enjoy the myriad of smells in our environment.
quality. Knowing how sensitive animals are to differences in Probably the single most important issue in the field of
intensity is necessary to rule out any confounding effects in olfactory research is the question of how the molecular
studies focusing on discrimination of qualitative differences. properties of an odorant determine odor quality. Crucial to
On a more practical note, generation of a suprathreshold understanding this relationship was the discovery of the
stimulus is more reliable than a threshold stimulus, and large multigene family of olfactory receptors (Buck and
instrumentation is available to measure these higher levels. Axel, 1991). The question of whether a certain type of
This is usually not the case when dealing with olfactory receptor responds to an individual odor or is broadly tuned
threshold testing, where the required concentrations are usu- to respond to many odors is being addressed through the
ally below the detection limits of most instruments. Precise tools of molecular biology, specifically the development of
392 Hastings

Table 1 Threshold Testing: Operant Conditioning

Species Experimental manipulation Paradigm Ref.
Mice Technique Go/no-go Walker and O’Connell, 1986
Mutant preference Deiss and Baudoin, 1997
Transgenic Go/no-go Youngentob and Margolis, 1999
Lesion Go/no-go Bodyak and Slotnick, 1999
Technique Go/no-go Mihalick et al., 2000
Rats Technique Go/no-go Slotnick and Nigrosh, 1974
Baseline/compound Go/no-go Marshall et al., 1981b
Lesion Go/no-go Slotnick B. M. and Schoonover, 1984
Drug Go/no-go Doty and Ferguson-Segall, 1987
Drug Go/no-go Doty et al., 1988
Lesion Go/no-go Doty and Ferguson-Segall, 1989
Drug Go/no-go Doty and Risser, 1989
Lesion Go/no-go Slotnick et al., 1987
Lesion Go/no-go Slotnick and Pazos, 1990
Drug Go/no-go Doty et al., 1990
Lesion Go/no-go Hunt and Slotnick, 1991
Baseline/compound Go/no-go Youngentob et al., 1991a
Lesion Go/no-go Doty et al., 1991
Lesion Go/no-go Slotnick and Schoonover, 1992
Drug Go/no-go Brosvic et al., 1996
Lesion Go/no-go Youngentob et al., 1997
Lesion Go/no-go Apfelbach et al., 1998
Lesion Go/no-go Lu and Slotnick, 1998
Drug Go/no-go Doty et al., 1998
Drug Go/no-go Dhong et al., 1999
Drug Go/no-go Doty et al., 1999
Lesion Go/no-go Slotnick et al., 2000
Dogs Baseline/compound Go/no-go Moulton and Marshall, 1976
Baseline/compound Go/no-go Marshall et al., 1981a
Baseline/compound Go/no-go Marshall and Moulton, 1981
Baseline/compound Operant/handler Arner et al., 1986
Baseline/compound Operant/handler Kurz et al., 1994
Baseline/compound Operant/handler Kurz et al., 1996

Table 2 Threshold Testing: Classical Conditioning/Conditioned Response

Species Experimental manipulation Paradigm Ref.
Rats Baseline/compound CA Eayrs and Moulton, 1960
Rats Baseline/compound CA Moulton and Eayrs, 1960
Rats Baseline/compound CA Moulton, 1960
Rats,dog Baseline/compound CS Davis, 1973
Rats Baseline/compound CS Pierson, 1974
Dogs Baseline/compound CS Krestel et al., 1984
Rat pups Drug CA Welson et al., 1991
Mice Chemical lesion CA Kimura et al., 1991
Rats Chemical lesion CA Peele et al., 1991
Rats Technique CA Darling and Slotnick, 1994
Rats Chemical lesion CS Sun et al., 1996
Rats Technique/chemical lesion CA Owens et al., 1996
Rat pups Conditioned reflex (change in heart rate) CR Hunt et al., 1997
Rats Conditioned reflex (ventilation) CR Nsegbe et al., 1998
Rats Conditioned odor potentiation of startle CR Richardson et al., 1999
CA
conditioned avoidance/approach; CS
conditioned suppression; CR
conditioned reflex.
Psychophysical Evaluation of Olfaction in Nonhuman Mammals 393

Table 3 Intensity-Difference Threshold tional capacity of the system. This is also an important
Experimental issue in recovery of function after the system has experi-
manipulation Paradigm Ref. enced extensive damage.

Technique/baseline Go/no-go Slotnick and Ptak, 1977

Lesion Go/no-go Slotnick and Gutman, 1977 1. Olfactory Discrimination Behavior
Lesion Go/no-go Slotnick and Berman, 1980
As with the absolute threshold task, the predominant para-
Lesion Go/no-go Slotnick and Schoonover, 1984
Lesion Go/no-go Slotnick, 1985 digm employed in most studies of olfactory discrimination
Lesion Go/no-go Slotnick and Schoonover, 1993 behavior has been the go/no-go, successive trials para-
Lesion Go/no-go Slotnick et al., 1997 digm, which incorporates a learned operant response such
as a nose poke or bar press. Since most problems presented
in an olfactory discrimination task are binary in nature,
e.g., differentiate between odor A and clean air or between
genetically modified mice, e.g., transgenic and “knockout” odor A and odor B, the go/no-go task is the most efficient
mice, which have had genes added or deleted (Youngentob way to answer the question. While the same testing para-
and Margolis, 1999). However, olfaction is a perceptual digm was used in almost all the studies listed in Table 4,
process, and even if there were specific receptors for each the rationale behind the testing did differ. Sometimes dis-
odorant, there would still be additional processing at the crimination testing was used to assess the functional status
level of the olfactory bulb and other higher cortical centers. of the system after local insult to the olfactory epithelium.
Only by examining an awake, responding organism is the In other studies, discrimination testing was used to assess
fully integrated process of olfaction available for study. the effects of lesions to the olfactory bulb or other CNS
The issue of perceived quality can be examined in pathways in an effort to understand the neuronal circuitry
numerous ways. These differing approaches, which in and structures mediating olfaction. Other times, when it
themselves are not mutually exclusive, help define the was apparent that the organism could perform simple
nature of the questions addressed as well as the techniques olfactory discrimination behavior, more complex tasks
employed to provide the answers. Thus, perceived quality were used or parameters such as varying delays in
can be examined in terms of discrimination behavior, as an responding, changes in intertrial intervals, and reversal
issue of similarity/dissimilarity, and in terms of perceptual learning procedures were incorporated in the testing
constancy. Discrimination involves the process of differen- design to examine the role learning and memory played in
tiating the elements of a group into two or more subgroups, the processing of olfactory cues. An attempt was made to
based upon some attribute. This is a fundamental property identify the predominant question being asked in the
of any sensory system; exactly how the olfactory system studies listed in Table 4. However, the distinctions made
accomplishes this and the nature of the basic mechanisms between the various study rationales are not absolute, and
or processes are not known. On a more practical level, learning is obviously inherent in all tasks.
assessing the ability of an organism to discriminate among
odorants is frequently used as an index of the functional
2. Stimulus Generalization
status of the system. While the discrimination process
seeks to separate elements into different groups, usually in Studies on discrimination behavior have been used in a
a binary (e.g., same, different) fashion, studies on stimulus diagnostic manner to investigate which neuronal struc-
generalization seek to investigate in greater detail the rela- tures or pathways are necessary for olfaction and, to a
tionships that exist among the various elements. The test- lesser degree, how the system actually processes qualita-
ing situation is thus structured so that the degree of tive differences/similarities. In the discrimination process,
similarity (or dissimilarity) among stimuli is reflected in a the general goal is to elucidate the differences
performance gradient. It is hoped that by examining these between/among stimuli in order to differentiate them. In
relationships, a clear understanding of how the system studies on stimulus generalization (Table 5), the goal is to
processes olfactory stimuli can be gained. Tests of similar- determine how closely different stimuli are perceived as
ity / dissimilarity can also be used in a diagnostic mode, being similar in quality. By focusing on the degree of per-
e.g., intensity-difference thresholds. The final issue, per- ceptual similarity among a group of stimuli, it is hoped
ceptual constancy, examines how the system maintains its that a continuum of similarity can be found among either
functional integrity over time. The primary olfactory some physiochemical property of the odorants, a com-
receptor cells, which also are the first order afferents, are monality among members of the olfactory receptor
replaced over time without any adverse effects on the func- gene family, or some other variable that will aid in under-
394 Hastings

Table 4 Discrimination Tasks was, in essence, a go/no-go paradigm, the response metric
Experimental —fixed-ratio responding—allowed for a gradation of
manipulation Paradigm Ref. responding, as opposed to the usual single nose poke or bar
press response. While some of the steric properties of the
Technique Go/no-go Nigrosh et al., 1975 odorant molecules could be shown to correlate with the
Baseline/
gradation in responding, there were too few data to sub-
learning-seta Go/no-go Slotnick and Katz, 1974
stantiate any hypothesis.
Lesionc Go/no-go Slotnick and Kaneko, 1981
Lesionb Go/no-go Slotnick, 1985 Almost 20 years elapsed before another example of a
Lesionb Go/no-go Slotnick et al., 1987 study that examined stimulus generalization in rats
Metabolic activityb Go/no-go Slotnick et al., 1989 appeared in the literature (Duncan et al., 1992). It was
Odor masking in similar in design to that of Braun and Marcus (1969), but
mixturesb Go/no-go Laing et al., 1989 instead of using fixed-ratio responding as the operant, a
Lesionc Go/no-go Lu an Slotnick, 1990 touch response was employed in the go/no-go paradigm.
Lesionc Go/no-go Slotnick and Risser, 1990 Odorants were assigned as S and S and the rats
Baselinec Go/no-go Slotnick et al., 1991 trained to a certain level of proficiency. On test days, a
Chemical lesiona Buried food Hastings et al., 1991 percentage of the trials was replaced with unreinforced
Lesionb Go/no-go Slotnick and Schoonover,
probe trials, in which new test odorants were presented.
1992
Performance on the probe trials was graded; the more the
Baseline/match
to samplec Go/no-go Lu et al., 1993 test odorant was like the S, the better the performance.
Lesionb Go/no-go Lu and Slotnick, 1994 One caveat offered by the authors applies to most gener-
Chemical Lesiona Go/no-go Evans et al., 1995 alization studies of similar design: “generalization
Odor memoryc Runway Lovelace and Slotnick, responses must be compared to levels of S responding;
1995 a low level of response does not mean that the test stim-
Lesionc Go/no-go Thanos and Slotnick, 1997 ulus is similar to S.”
Lesionb Go/no-go McBride and Slotnick, A procedure that has considerable potential as a tool
1997 for investigating the degree of similarity/dissimilarity
Chemical lesiona Go/no-go Setzer and Slotnick, 1998 among different odorants in animals is an animal analog
Lesionc Go/no-go Zhang et al., 1998
of the confusion matrix task used to evaluate human
Lesionb Go/no-go Lu and Slotnick, 1998
olfactory function (Wright, 1987) as developed by
Transgenic Buried food Smith et al., 1998
Lesion Buried food Berger-Sweeney et al., Youngentob and coworkers (Kent et al., 1995;
1998 Youngentob et al., 1990, 1991b, 1995) (Fig. 3). In this
Chemical lesion Go/no-go Xu and Slotnick, 1999 procedure, rats are initially trained to sample a test odor-
Diet Go/no-go Greiner et al., 1999 ant at a central location and then to traverse a specially
Drug Go/go Winters et al., 2000 marked alley to obtain reinforcement. Similar training is
Chemical lesion Go/no-go Slotnick et al., 2000 given for all test odorants. During this training period, the
a
To assess functional status of neuroepithelium.
rat is not allowed access to any alley except the one asso-
b
To assess effect of manipulation on olfactory function (nonneuroepithe- ciated with the test odorant. In the testing phase, all
lium effects).
c
To assess learning/memory function.
Table 5 Stimulus Generalization Studies
Paradigm Ref.
standing how the system actually processes qualitative Fixed ratio responding during
differences of odorants. S,S with probe trial Braun and Marcus, 1969
The initial publication of a technique to study stimulus Go/no-go: masking by similar/
generalization among odorants in rats occurred over 30 dissimilar odorants Laing et al., 1989
years ago (Braun and Marcus, 1969). Braun and Marcus Confusion matrix Youngentob et al., 1990
taught rats to respond on a fixed-rate schedule in the pres- Confusion matrix Youngentob et al., 1991b
ence of one odorant and to refrain from responding in the Go/no-go with probe trials Duncan et al., 1992
presence of a second odorant. On a nonrewarded probe Approach/avodiance Heth et al., 1992
Confusion maxtrix Youngentob et al., 1995
trial, a third, novel odor was presented and the degree to
Confusion maxtrix Kent et al., 1995
which the rat responded was recorded as a measure of the
Buried odor mixtures Linster and Smith, 1999
similarity to the original S odorant. Although this task
Psychophysical Evaluation of Olfaction in Nonhuman Mammals 395

reinforcement. By manipulating the composition of the

odorant cue in the scented cup they were able to demon-
strate that rats can generalize between a component and a
binary mixture that contains that component. While this
task requires much less baseline training than some of the
aforementioned tasks, its overall utility is limited by the
lack of rigorous control over stimulus generation and pre-
sentation. Even with this limitation, the modified buried
food task can still provide useful information concerning
odor identification and processing,

3. Perceptual Constancy
As noted earlier, olfactory receptor neurons (ORNs) are
replaced throughout life on a continuous basis. Whenever
the neuroepithelium is damaged after exposure to toxic
compounds or diseases, this process is accelerated. The
fact that perceptual quality remains more or less constant
over long periods suggests that during normal replace-
ment, new ORNs either reestablish the same bulbar con-
nections as their predecessors or are reprogrammed once
they successfully connect to the bulb. Furthermore, under
normal conditions, only a small percentage of the total
number of neurons is being replaced, leaving the bulk of
the system intact to function in a normal manner.
The situation is quite different when the olfactory neuro-
epithelium is extensively damaged by toxic agents, disease,
Figure 3 Five tunnel differential response chamber for the ani- or experimental nerve transaction. Under these conditions,
mal odorant confusion matrix: (A) top view; (B) side view. (From almost all normal connections are lost and the system must
Youngentob et al., 1990).
reconstitute itself, de novo. While a number of studies have
examined the functional status of the olfactory system after
response alleys become available and the test odorant is extensive insult, usually by employing threshold or dis-
varied from trial to trial. By examining the pattern of crimination tasks, only a few have attempted to establish
errors made during a session, the similarity/dissimilarity whether perceptual quality remains constant during and/or
of all odor combinations can be investigated. The major after insult. Anecdotal reports suggests that perceptual
drawbacks of this procedure are the long and complicated qualities can be severely disrupted after pervasive damage
training and testing periods required to perform the task, to the olfactory system.
as well as the need for a highly sophisticated testing To address the question of perceptual constancy, there
apparatus. must be a way to assess perceptual quality with enough
Not all procedures for examining stimulus generaliza- sensitivity to detect even small perturbations to the sys-
tion are based upon operant conditioning paradigms con- tem. Few studies have attempted to address this challeng-
ducted in operant chambers. Heth et al. (1992) used ing issue. Yee and Costanzo (1995) manually trained
positioning of the nest and food store and the preferred hamsters to perform a go/no-go task that involved
location of the tested animal (mole rats) as indicators for responding initially to an odorized container while ignor-
preference/aversion to various pairs of enantiomeric ing a nonodorized container, followed by analogous dis-
compounds. Linster and Smith (1999) employed one of crimination training between two odorants. After
the oldest tests available for assessing olfactory func- successfully learning to discriminate the two odorants,
tion—the buried food test—to investigate generalization the olfactory nerve fibers were severed. Return of olfac-
between binary odor mixtures in rats. Instead of simply tory-mediated behavior was first seen 19 days after
allowing the rats to dig in sand to find food reinforce- surgery, with complete recovery evident by day 40. While
ment, they trained the rats to choose between a scented these results confirmed that replacement ORNs regained
cup with reinforcement and an unscented cup without full functionality, it left unanswered whether the system
396 Hastings

displayed perceptual constancy during the recovery new era in olfactory research. Much progress has been
process. This was because testing after surgery used rein- made in delineating not only the large multigene family of
forced trials. Thus, the complete recovery of olfactory olfactory receptors, but also the physical location of these
function could have been due to recovery of the system receptors within the nasal cavity. The next crucial step is to
and performance of the originally task, or the animals link specific receptors and / or families of receptors with
could have learned to discriminate between the two odors the detection of individual odorants. Some progress has
based upon the associated reinforcement contingencies, been made using cell cultures (Zhao et al., 1998) or elec-
even if the odors were being perceived as two new, dis- trophysiological measures such as the electro-olfactogram
tinct odors. To address this issue, Yee and Costanzo (Buiakove et al., 1996) and patch-clamping (Ma et al.,
(1998) repeated the study, but did not test animals until 1999). While these studies provide very important infor-
40 days after surgery, to allow time for the new neurons mation, they cannot address the issues of perception that
to fully mature. Sham-operated animals were able to per- occur in an intact, responding organism. The development
form the original discrimination task after 40 days of rest, of genetically modified mice in which the ORN genes have
but the nerve-transected animals could not, when tested been manipulated, or mice that mimic a disease known to
without reinforcement. When reinforcement was re- be associated with olfactory dysfunction, e.g., Alzheimer’s
instated, rats could perform the discrimination. This sug- disease (Doty, 1991), along with suitable techniques for
gested that the nerve transection and subsequent regrowth assessing olfactory function offers promise in this regards.
had changed the odor perceptual qualities of the stimuli An approach is outlined below, which incorporates many
and that the animals had to relearn the task based upon of the techniques already discussed, that can be used to
the reinforcement contingencies of the new stimuli. assess olfactory function in genetically modified mice.
The only other study to assess perceptual constancy of
odorant quality during recovery of function after insult to A. Adults
ORNs was conducted by Youngentob and Schwob
(1997). Rats were trained on a five-odorant identification Although mice can be evaluated using the lengthy psy-
confusion matrix task and then their nasal neuroepithe- chophysical paradigms described earlier, there is a need for
lium was almost completely destroyed by exposure to the rapid tests that can be used to assess large numbers of
olfactotoxin methyl bromide. Like the preceding study, mice. Three screening tests—the buried food test, a condi-
rats were not tested until the olfactory neuroepithelium tioned odor aversion task, and a odor habituation task—
had been reconstituted (2 months for methyl bromide). have been found to be useful for this purpose. In the
Under these conditions, control rats retained the identifi- simplest form of the buried food test, food reinforcement
cation task at a 75% level, in contrast to methyl bro- is buried in bedding or sand and the mouse must locate it
mide–treated rats, who performed at a 30% level. It also and dig it up within a specified time frame (Liggett, 1928;
took the lesioned rats almost three times as long as the Hastings et al,. 1991). The test can be made more sophis-
control rats to once again achieve criterion performance. ticated by requiring the mouse to discriminate between
The authors interpreted this poorer performance by the two odors or odors of different concentrations (Berger-
lesioned rats as an indication that odorant quality percep-
Sweeney et al., 1998; Linster and Smith, 1999; Smith
tion had been altered. This is in contrast to the findings of
et al., 1998). In conditioned taste aversion, an aversion to
their earlier study, where they reported no change in
a flavored solution is induced by following ingestion with
absolute thresholds for similarly lesioned rats
toxicosis (Garcia et al., 1966). However, development of
(Youngentob et al., 1997). This would suggest that the
an aversion to an odor alone is much more difficult
absolute threshold test might not be the most sensitive
(Hankins et al., 1973). Holder and Garcia (1987) have
test to use to evaluate the functional status of the olfac-
shown, though, that if a taste is paired with an odorant, a
tory system after insult. However, additional studies are
strong aversion to the odorant alone can be obtained. Only
necessary to establish whether this is true.
one trial conditioning is needed, and the learned associ-
ation is quite persistent. Tests involving both detection
V. BRIEF SCREENING TESTS FOR thresholds and discrimination tasks can be developed using
ASSESSING OLFACTION IN conditioned odor aversion (Darling and Slotnick, 1994;
GENETICALLY MODIFIED MICE Kimura et al., 1991). In an odor habituation paradigm, one
odor is presented repeatedly in short trials until investiga-
Development of methodologies in molecular biology that tory interests in it wanes. A novel odor is then presented,
allow direct manipulation of genes has opened an entirely producing a dramatic increase in sniffing, thereby demon-
Psychophysical Evaluation of Olfaction in Nonhuman Mammals 397

strating detection. If differences are found in the olfactory ACKNOWLEDGMENTS

function of genetically modified mice using these screen-
ing tasks, more sophisticated psychophysical tasks can Supported, in part, by Grants RO1 DC 02974 (R. L. Doty),
then be used to evaluate subtle changes in olfactory func- RO1 DC 04083 (I. Kratskin) and R43 DC 04024 (L.
tion. Hastings). The comments of Dr. R. L. Doty during the
preparation of this manuscript were especially helpful.
B. Neonates

Since the senses of hearing and sight are underdeveloped REFERENCES

at birth, mouse pups must rely almost entirely on their
sense of smell for bonding and nutrition. In fact, any Amiri, L., Dark, T., Noce, K. M., and Kirstein, C. L. (1998). Odor
preferences in neonatal and weanling rats. Dev. Psychobiol.
manipulation that results in anosmia at birth would most
33:157–162.
likely be lethal to the pup. Nonetheless, some genetic alter- Apfelbach, R., Weiler, E., Asselbergs, W. F., Polak, E. H., and
ations may influence olfactory function in neonates. The Slotnick, B. (1998). Selective and reversible reduction of odor
use of neonates could also be cost effective, since they do sensitivity in the rat by concanavalin A. Physiol. Behav. 65:
not have to be housed until they reach maturity. One task 513–516.
that has been used to evaluate olfaction in mouse pups Arner, L. D., Johnson, G. R., and Slotnick, B. (1986). Delineating
involves a preference test for soiled “home-cage” bedding toxic areas by canine olfaction. EPA project summary.
or the dam (Montella and Reddy, 1991; Wong et al., 2000). Ashton, E. H., Eayrs, J. T., and Moulton, D. G. (1957). Olfactory
A novel task that is used extensively with neonatal rats and acuity in the dog. Nature 179:1069–1070.
could be easily adapted to mice is associative conditioning Berger-Sweeney, J., Libbey, M., Arters, J., Junagadhwalla, M.,
and Hohmann, C. F. (1998). Neonatal monoaminergic deple-
using odorants. Pups are exposed to an odor paired with
tion in mice (Mus musculus) improves performance of a novel
tactile stimulation (stroking the skin with a paint brush), odor discrimination task. Behav. Neurosci. 112:1318–1326.
with preference measured as time spent near the odor ver- Bernard, C. (1865). An Introduction to the Study of Experimental
sus a nonodorized space (Weldon et al. 1991). Another Medicine, H. C. Greene, (Transl., 1949). Henry Schuman, Inc.,
approach to allow greater quantification of responses is to New York
establish conditioned reflexes, e.g., heart rate, to odorant Bodyak, N., and Slotnick, B. (1999). Performance of mice in an
cues (Hunt et al. 1997; Nsegbe et al., 1998; Richardson automated olfactometer: odor detection, discrimination and
et al., 1999). It should be noted, however, that, given the odor memory. Chem. Senses 24:637–645.
limited response repertoire of neonates, it can be difficult Bouvet, J. F., Delaleu, J. C., and Holley, A. (1987). Olfactory
to detect subtle deficits with these tasks. receptor cell function is affected by trigeminal nerve activity.
Neurosci. Lett. 77:181–186.
Braun, J. J., and Marcus, J. (1969). Stimulus generalization
among odorants by rats. Physiol. Behav. 4:245–248.
VI. SUMMARY AND CONCLUSION
Brosvic, G. M., Risser, J. M., Mackay-Sim, A., and Doty, R. L.
(1996). Odor detection performance in hypothyroid and
Contrary to Bernard’s supposition that humans are the euthyroid rats. Physiol. Behav. 59:117–121.
only animals for whom the study of sensation and percep- Brouette-Lahlou, I., Soulier, N., and Vernet-Maury, E. (1994).
tion is possible, nonhuman animals have been used to Does the rat vomernonasal organ play a role in olfactory
study the sense of olfaction with much success. learning? Adv. Biosci. 93:457–464.
Techniques have been developed that allowed behavioral Buck, L., and Axel, R. (1991). A novel multigene family may
responses of animals to be interpreted and quantified as encode odorant receptors: a molecular basis for odor recogni-
well as the precise generation and presentation of olfac- tion. Cell 65:175–187.
tory stimuli. Currently, animals are routinely used to study Buiakove, O. I., Baker, H., Scott, J. W., Farbman, A., Kream, R.,
the olfactory process, including the coding of odor inten- Grillo, M., Franzen, L., Richman, M., Davis, L. M.,
Abbondanzo, S., Stewart, C. L., and Margolis, F. L. (1996).
sity and quality. The effects of a wide variety of experi-
Olfactory marker protein (OMP) gene deletion causes altered
mental manipulations, including lesions, drugs, and physiological activity of olfactory sensory neurons. Proc.
various learning paradigms, on the ability to perceive Natl. Acad. Sci. USA 93:9858–9863.
odors have been employed. Future application of these Cain, W. S. (1977). Differential sensitivity for smell: “noise” at
techniques to the study of olfactory development and the the nose. Science 195:796–798
genetic basis of olfactory function in mice are now being Darling, F. M., and Slotnick, B. M. (1994). Odor-cued taste
investigated. avoidance: a simple and efficient method for assessing olfac-
398 Hastings

tory detection, discrimination and memory in the rat. Physiol. Giannetti, N., Saucier, D., and Astic, L. (1995). Analysis of the
Behav. 55:817–822. possible altering function of the septal organ in rats: a lesional
Darwin, C. (1873). Inherited instinct. Nature 7:281. and behavioral study. Physiol. Behav. 58:837–845.
Davis, R. G. (1973). Olfactory psychophysical parameters in man, Goff, W. R. (1961). Measurement of absolute olfactory sensitiv-
rat, dog, and pigeon. J. Comp. Physiol. Psychol. 85: 221–232. ity in rats. Am. J. Psychol. 74:384–393.
Deiss, V., and Baudoin, C. (1997). Hyposmia for butanol and Greiner, R. S., Moriguchi, T., Hutton, A., Slotnick, B. M., and
vanillin in mutant staggerer male mice. Physiol. Behav. Salem, N., Jr. (1999). Rats with low levels of brain docosa-
61:209–213. hexaenoic acid show impaired performance in olfactory-based
Dhong, H. J., Chung, S. K., and Doty, R. L. (1999). Estrogen pro- and spatial learning tasks. Lipids 34 (supp):S239–S243.
tects against 3-methylindole-induced olfactory loss. Brain Hamilton, G. van T. (1907). An experimental study of an unusual
Res. 824:312–315. type of reaction in a dog. J. Comp. Neurol. 17:329–341.
Doty, R. L. (1991). Olfactory capacities in aging and Alzheimer’s Hankins, W. G., Garcia, J., and Rusiniak, K. W. (1973).
disease. Psychophysical and anatomic considerations. Ann. Dissociation of odor and tase in baitshyness. Behav. Biol.
NY Acad. Sci. 640:20–27. 8:407–419.
Doty, R. L., and Ferguson-Segall, M. (1987). Odor detection per- Hastings, L., Miller, M. L., Minnema, D. J., Evans, J. and Radike,
formance of rats following d-amphetamine treatment: a signal M. (1991). Effects of methyl bromide on the rat olfactory sys-
detection analysis. Psychopharmacology 93:87–93. tem. Chem. Senses 16:43–55.
Doty, R. L., and Ferguson-Segall, M. (1989). Influence of adult Heitzenrohder, C. (1913). Über das Verhalten des Hundes gegen
castration on the olfactory sensitivity of the male rat: a signal einige Riechstoffe. Z. Biol. 62:491–507.
detection analysis. Behav. Neurosci. 103:691–694. Henning, H. (1916). Der Geruch. Barth, Leipzig.
Doty, R. L., and Risser, J. M. (1989). Influence of the D-2 Heth, G., Nevo, E., Ikan, R., Weinstein, V., Ravid, U., and
dopamine receptor agonist quinpirole on the odor detection Duncan, H. (1992). Differential olfactory perception of enan-
performance of rats before and after spiperone administration. tiomeric compounds by blind subterranean mole rats (Spalax
Psychopharmacology 98:310–315. ehrenbergi). Experientia 48:897–902.
Doty, R. L., Ferguson-Segall, M., Lucki, I., and Kreider, M. Holder, M. D., and Garcia, J. (1987). Role of temporal order and
(1988). Effects of intrabulbar injections of 6-hydroxy- odor intensity in taste-potentiated odor aversions. Behav.
dopamine on ethyl acetate odor detection in castrate and non- Neurosci. 101:158–163.
castrate male rats. Brain Res. 444:95–103. Hübener, F., and Laska, M. (2001). A two-choice discrimination
Doty, R. L., Li, C., and Risser, J. M. (1990). Fluprazine method to assess olfactory percformance in pigtailed
hydrochloride: no influence on the odor detection perfor- macaques, Macaca nemestrina. Physiol. Behav. 72:511–519.
mance of male rats. Pharmacol. Biochem. Behav. 35: Hunt, N. L., and Slotnick, B. M. (1991). Functional capacity of
699–703. the rat olfactory bulb after neonatal naris occlusion. Chem.
Doty, R. L., Risser, J. M., and Brosvic, G. M. (1991). Influence Senses 16:131–142.
of adrenalectomy on the odor detection performance of rats. Hunt, P. S., Richardson, R., Hess, M. F., and Campbell, B. A.
Physiol. Behav. 49:1273–1277. (1997). Emergence of conditioned cardiac responses to an
Doty, R. L., Li, C., Bagla, R., Huang, W., Pfeiffer, C., Brosvic, G. olfactory CS paired with an acoustic startle UCS during
M., and Risser, J. M. (1998). SKF 38393 enhances odor detec- development: form and autonomic origins. Dev. Psychobiol.
tion performance. Psychopharmacology 136:75–82. 30:151–163.
Doty, R. L., Bagla, R., and Kim, N. (1999). Physostigmine Johnston, R. E. (1998). Pheromones, the vomeronasal system,
enhances performance on an odor mixture discrimination test. and communication. From hormonal responses to individual
Physiol. Behav. 65:801–804. recognition. Ann. NY Acad. Sci. 855:333–348.
Duncan, H. J., Beauchamp, G. K., and Yamazaki, K. (1992). Kailscher, O. (1909). Weitere Mitteilung über die Ergebnisse der
Assessing odor generalization in the rat: a sensitive technique. Dressur als physiologische Untersuchungsmethode auf dem
Physiol. Behav. 52:617–620. Gebiete des Gehör-;Geruchs-und Farbensinnes. Arch. Physiol.
Eayrs, J. T., and Moulton, D. G. (1960). Studies in olfactory acu- 303–322.
ity: I. Measurement of olfactory thresholds in the rat. Q. J. Kendal-Reed, M., Walker, J. C., Morgan, W. T., LaMacchio, M. and
Exp. Psychol. 12:90–98. Lutz, R. W. (1998). Human responses to propionic acid. I.
Evans, J. E., Miller, M. L., Andringa, A., and Hastings, L. (1995). Quantification of within- and between-participant variation in per-
Behavioral, histological, and neurochemical effects of nickel ception by normosmics and anosmics. Chem. Senses 23: 71–82.
(II) on the rat olfactory system. Toxicol. Appl. Pharmacol. Kent, P. F., Youngentob, S. L., and Sheehe, P. R. (1995). Odorant-
130:209–220. specific spatial patterns in mucosal activity predict perceptual
Gaafar, H., Girgis, R., Hussein, M., and el-Nemr, F. (1992). The differences among odorants. J. Neurophysiol. 74:1777–1781.
effect of ammonia on the respiratory nasal mucosa of mice. Kimura, Y., Miwa, T., Furukawa, M., and Umeda, R. (1991).
A histological and histochemical study. Acta Otolaryngol. Effects of topical application of steroids on olfactory disturb-
112:339–342. ance in mice. Chem. Senses 16:297–302.
Garcia, J., Ervin, F. r., and Koelling, R. A. (1966). Learning with Krestel, D., Passe, D., Smith, J. C., and Jonsson, L. (1984).
prolonged delay of reinforcement. Psychonom. Sci. Behavioral determination of olfactory thresholds to amyl
5:121–122. acetate in dogs. Neurosci. Biobehav. Rev. 8:169–174.
Psychophysical Evaluation of Olfaction in Nonhuman Mammals 399

Kurz, M. E., Billard, M., Rettig, M., Augustiniak, J., Lange, J., asymmetrical lesion paradigm in rats. Behav. Neurosci.
Larsen, M., Warrick, R., Mohns, T., Bora, R., and Broadus, K. 111:1273–1284.
(1994). Evaluation of canines for accelerant detection at fire McCartney, W. (1968). Olfaction and Odours. Springer-Verlag,
scenes. J. Forensic Sci. 39:1528–1536. New York, pp. 28–31.
Kurz, M. E., Schultz, S., Griffith, J., Broadus, K., Sparks, J., Mihalick, S. M., Langlois, J. C., Krienke, J. D., and Dube, W. V.
Dabdoub, G., and Brock, J. (1996). Effect of background (2000). An olfactory discrimination procedure for mice.
interference on accelerant detection by canines. J. Forensic J. Exp. Anal. Behav. 73:305–318.
Sci. 41:868–873. Montella, P. G., and Reddy, C. S. (1991). Neurotoxic effects of
Kyle, A. L., Sorenson, P. W., Stacey, N. E., and Dulka, J. G. secalonic acid D in mice during subchronic postnatal expo-
(1987). Medial olfactory tract pathways controlling sexual sure. Pharmacol. Biochem. Behav. 40:241–247.
reflexes and behavior in teleosts. Ann. NY Acad. Sci. Moulton, D. G. (1960). Studies in olfactory acuity. III. Relative
519:107. detectability of n-aliphatic acetates by the rat. Q. J. Exp.
Laing, D. G., Panhuber, H., and Slotnick, B. M. (1989). Odor Psychol. 12:203–213.
masking in the rat. Physiol. Behav. 45:689–694. Moulton, D. G. (1977). Minimum odorant concentration
Lashley, K. S., and Sperry, R. W. (1943). Olfactory discrimina- detectable by the dog and their implications for olfactory
tion after destruction of the anterior thalamic nuclei. Am. receptor sensitivity. In Chemical Signals in Vertebrates,
J. Physiol. 139:450. Muller-Schware, D., and Mozell, M. M., (Eds.). Plenum
Liggett, J. R. (1928). An experimental study of the olfactory sen- Press, New York, pp. 455–464.
sitivity of the white rat. Genet. Psychol. Monogr. 3:3–64. Moulton, D. G., and Eayrs, J. T. (1960). Studies in olfactory acu-
Linster, C., and Smith, B. H. (1999). Generalization between ity. II. Relative detectability of n-aliphatic alcohols by the rat.
binary odor mixtures and their components in the rat. Physiol. Q. J. Exp. Psychol. 12:99–109.
Behav. 66:701–707. Moulton, D. G. and Marshall, D. A. (1976). The performance of
Livermore, A., Hummel, T., and Kobal, G. (1992). dogs in detecting -ionone in the vapor phase. J. Comp.
Physiol. 110:287–306.
Chemosensory event-related potentials in the investigation of
Nigrosh, B. J., Slotnick, B. M., and Nevin, J. A. (1975). Olfactory
interactions between the olfactory and the somatosensory
discrimination, reversal learning, and stimulus control in rats.
(trigeminal) systems. Electroencephalog. Clin. Neurophysiol.
J. Comp. Physiol. Psychol. 89:285–294.
83:201–210.
Nsegbe, E., Vardon, G., Dauger, S., Perruchet, P., Gaultier, C.,
Lovelace, C. T., and Slotnick, B. M. (1995). Memory for brief,
and Gallego, J. (1998). Classical conditioning to hypoxia
widely spaced odor presentations in the rat. Chem. Senses
using odors as conditioned stimuli in rats. Behav. Neurosci.
20:183–190.
112:1393–1401.
Lu, X. C., and Slotnick, B. M. (1990). Acquisition of an olfactory
Owens, J. G., James, R. A., Moss, O. R., Morgan, K. T.,
learning-set in rats with lesions of the mediodorsal thalamic
Bowman, J. R., Struve, M. F., and Dorman, D. C. (1996).
nucleus. Chem. Senses 15:713–724. Design and evaluation of an olfactometer for the assessment
Lu, X. C., and Slotnick, B. M. (1994). Recognition of propionic of 3-methylindole-induced hyposmia. Fundam. Appl.
acid vapor after removal of the olfactory bulb area associated Toxicol. 33:60–70.
with high 2-DG uptake. Brain Res. 639:26–32. Passe, D. H., and Walker, J. C. (1985). Odor psychophysics in
Lu, X. C., and Slotnick, B. M. (1998). Olfaction in rats with vertebrates. Neurosci. Biobehav. Rev. 9:431–467.
extensive lesions of the olfactory bulbs: implications for odor Passy, J. (1892). Sur les minimums perceptibles de quelques
coding. Neuroscience 84:849–866. odeurs. C. R. Acad. Sci. (Paris) 114:306–308.
Lu, X. C., Slotnick, B. M., and Silberberg, A. M. (1993). Odor Peele, D. B., Allison, S. D., Bolon, B., Prah, J. D., Jensen, K. F.,
matching and odor memory in the rat. Physiol. Behav. 53: and Morgan, K. T. (1991). Functional deficits produced by
795–804. 3-methylindole-induced olfactory mucosal damage revealed
Ma, M., Chen, W. R., and Shepherd, G. M. (1999). Electro- by a simple olfactory learning task. Toxicol. Appl.
physiological characterization of rat and mouse olfactory Pharmacol. 107:191–202.
receptor neurons from an intact epithelial preparation. Pfaffmann, C., Goff, W. R., and Bare, J. K. (1958). An olfac-
J. Neurosci. Methods 92:31–40. tometer for the rat. Science 128:1007–1008.
Marshall, D. A., and Moulton, D. G. (1981). Olfactory sensitivity Pierson, S. C. (1974). Conditioned suppression to odorous stim-
to alpha-ionone in humans and dogs. Chem. Senses 6:53–61. uli in the rat. J. Comp. Physiol. Psychol. 86:708–717.
Marshall, D. A., Blumer, L., and Moulton, D. G. (1981a). Odor Prah, J. D., Sears, S. B., and Walker, J. C. (1995). Modern
detection curves for n-pentanoic acid in dogs and humans. approaches to air dilution olfactometry. In Handbook of
Chem. Senses 6:445–453. Olfaction and Gustation, Doty, R. L. (Ed.). Marcel Dekker,
Marshell, D. A., Doty, R. L., Lucero, D. P., and Slotnick, B. M. Inc., New York, pp. 227–256.
(1981b). Odor detection thresholds in the rat for the vapors of Richardson, R., Vishney, A., and Lee, J. (1999). Conditioned odor
three related perfluorocarbons and ethylene glycol dinitrate. potentiation of startle in rats. Behav. Neurosci. 113:787–794.
Chem. Senses 6:421–433. Rodolfo-Masera, T. (1943). Su l’esistenza di un particolare
McBride, S. A., and Slotnick, B. (1997). The olfactory thalamo- organo olfattivo nel setto nasale della cavia e di altri roditori.
cortical system and odor reversal learning examined using an Arch. Ital. Anat. Embryol. 48:157–212.
400 Hastings

Schmidt, F. (1910). Verbrecherspur und Polizeihund. Selbstverlag Slotnick, B. M., Panhuber, H., Bell, G. A., and Laing, D. G.
des Vereins für deutsche Schäferhunde, Augsburg. (1989). Odor-induced metabolic activity in the olfactory bulb
Schwanzel-Fukuda, M., and Pfaf, D. W. (1995). Structure and of rats trained to detect propionic acid vapor. Brain Res.
function of the nervus terminalis. In Handbook of Olfaction 500:161–168.
and Gustation, Doty, R. L. (Ed.). Marcel Dekker, Inc., New Slotnick, B. M., Kufera, A., and Silberberg, A. M. (1991).
York, pp. 835–864. Olfactory learning and odor memory in the rat. Physiol.
Seffrin, L. (1915). über die kleinsten noch wahrnehmbar Behav. 50:555–561.
Geruchsmengen einiger Riechstoffe beim Hund. Z. Biol. Slotnick, B. M., Bell, G. A., Panhuber, H., and Laing, D. G.
47:493–513. (1997). Detection and discrimination of propionic acid
Setzer, A. K., and Slotnick, B. (1998). Odor detection in rats with after removal of its 2-DG identified major focus in the
3-methylindole-induced reduction of sensory input. Physiol. olfactory bulb: a psychophysical analysis. Brain Res. 762:
Behav. 65:489–496. 89–96.
Silver, W. L., Mason, J. R., Marshall, D. A., and Maruniak, J. A. Slotnick, B., Glover, P., and Bodyak, N. (2000). Does intranasal
(1985). Rat trigeminal, olfactory and taste responses after application of zinc sulfate produce anosmia in the rat? Behav.
capsaicin desensitization. Brain Res. 333:45–54. Neurosci. 114:814–829.
Slotnick, B. M. (1985). Olfactory discrimination in rats with Smith, D. R., Striplin, C. D., Geller, A. M., Mailman, R. B.,
anterior amygdala lesions. Behav. Neurosci. 99:956–963. Drago, J., Lawler, C. P., and Gallagher, M. (1998). Behavioral
Slotnick, B. M. (1990). Olfactory Perception. In Comparative assessment of mice lacking D1A dopamine receptors.
Perception: Basic mechanisms, Berkely, M. A., and Stebbins, Neuroscience 86:135–146.
W. C. (Eds.). John Wiley and Sons, New York. Stanton, M. E. (1991). Neonatal exposure to triethyltin disrupts
Slotnick, B. M., and Berman, E. J. (1980). Transection of the lat- olfactory discrimination learning in preweanling rats.
eral olfactory tract does not produce anosmia. Brain Res. Bull. Neurotox. Teratol. 13:515–524.
5:141–145. Stebbins, W. C. (1970). Principles of animal psychophysics. In
Slotnick, B. M., and Gutman, L. A. (1977). Evaluation of Animal Psychophysics: The Design and Conduct of Sensory
intranasal zinc sulfate treatment on olfactory discrimination in Experiments, Stebbins, W. C. (Ed.). Appleton-Century-Crofts,
rats. J. Comp. Physiol. Psychol. 91:942–950. New York, pp. 1–20.
Slotnick, B. M., and Kaneko, N. (1981). Role of mediodorsal Stevens, J. C., William, C. S., and Burke, R. J. (1988). Variability
thalamic nucleus in olfactory discrimination learning in rats. of olfactory thresholds. Chem. Senses 13:643–653.
Science 214:91–92. Stone, C. P. (1941). A simple method for developing olfactory
Slotnick, B. M., and Katz, H. M. (1974). Olfactory learning-set discrimination habits in rats. J. Genet. Psychol. 58:419–421.
formation in rats. Science 185:796–798. Sun, T. J., Miller, M. L., and Hastings, L. (1996). Effects of
Slotnick, B. M., and Nigrosh, B. J. (1974). Olfactory stimulus inhalation of cadmium on the rat olfactory system: behavior
control evaluated in a small animal olfactometer. Percept. and morphology. Neurotox. Teratol. 18:89–98.
Motor Skills 39:583–597. Swann, H. G. (1933). The function of the brain in olfaction: l.
Slotnick, B. M., and Pazos, A. J. (1990). Rats with one olfactory Olfactory discrimination and an apparatus for its test.
bulb removed and the contralateral naris closed can detect J. Comp. Physiol. 15:229–241.
odors. Physiol. Behav. 48:37–40. Thanos, P. K., and Slotnick, B. M. (1997). Short-term odor mem-
Slotnick, B. M., and Ptak, J. E. (1977). Olfactory intensity-dif- ory: effects of posterior transection of the lateral olfactory
ference thresholds in rats and humans. Physiol. Behav. tract in the rat. Physiol. Behav. 61:903–906.
19:795–802. Thorndike, E. L. (1911). Animal Intelligence: Experimental
Slotnick, B. M., and Risser, J. M. (1990). Odor memory and odor Studies. Macmillan, New York.
learning in rats with lesions of the lateral olfactory tract and Tucker, D. (1971). Nonolfactory responses from the nasal cavity:
mediodorsal thalamic nucleus. Brain Res. 529:23–29. Jacobson’s organ and the trigeminal system. In Handbook of
Slotnick, B. M., and Schoonover, F. W. (1984). Olfactory thresh- Sensory Physiology, Beidler, L. M. (Ed.). Springer-Verlag,
olds in unilaterally bulbectomized rats. Chem. Senses New York.
9:325–340. Walker, J. C., and Jennings, R. A. (1992). Comparison of odor
Slotnick, B. M., and Schoonover, F. W. (1992). Olfactory path- perception in humans and animals. In The Human Sense of
ways and the sense of smell. Neurosci. Biobehav. Rev. Smell, Laing, D. G., Doty, R. L., and Breipohl, W. (Eds.).
16:453–472. Springer-Verlag, New York, pp. 261–280.
Slotnick, B. M., and Schoonover, F. W. (1993). Olfactory sensi- Walker, J. C., and O’Connell, R. J. (1986). Computerized odor
tivity of rats with transection of the lateral olfactory tract. psychophysical testing in mice. Chem. Senses 11:439–453.
Brain Res. 616:132–137. Watson, J. B. (1914). Behavior: An Introduction to Comparative
Slotnick, B. M., Graham, S., Laing, D. G., and Bell, G. A. (1987). Psychology. Holt, New York.
Detection of propionic acid vapor by rats with lesions of Weldon, D. A., Travis, M. L., and Kennedy, D. A. (1991).
olfactory bulb areas associated with high 2-DG uptake. Brain Posttraining D1 receptor blockade impairs odor conditioning
Res. 417:343–346. in neonatal rats. Behav. Neurosci. 105:450–458.
Psychophysical Evaluation of Olfaction in Nonhuman Mammals 401

Winters, B., Matheson, W. R., McGregor, I. S., and Brown, R. E. Youngentob, S. L., and Schwob, J. E. (1997). Changes in odorant
(2000). An automated two-choice test of olfactory working quality perception following methyl bromide induced lesions
memory in the rat: effect of scopolamine. Psychobiology 28: of the olfactory epithelium. Chem. Senses 22: 830–831.
21–31. Youngentob, S. L., Markert, L. M., Mozell, M. M., and Hornung,
Wirsig-Wiechmann, C. R., and Lepri, J. J. (1991). LHRH- D. E. (1990). A method for establishing a five odorant identi-
immunoreactive neurons in the pterygopalatine ganglia of fication confusion matrix task in rats. Physiol. Behav.
voles: a component of the nervus terminalis? Brain Res. 47:1053–1059.
568:289–293. Youngentob, S. L., Hornung, D. E., and Mozell, M. M. (1991a).
Wong, S. T., Trinh, K., Hacker, B., Chan, G. C., Lowe, G., Determination of carbon dioxide detection thresholds in
Gaggar, A., Xia, Z., Gold, G. H., and Storm, D. R. (2000). trained rats. Physiol. Behav. 49:21–26.
Disruption of the type III adenylyl cyclase gene leads to Youngentob, S. L., Markert, L. M., Hill, T. W., Matyas, E. P., and
Mozell, M. M. (1991b) Odorant identification in rats: an
peripheral and behavioral anosmia in transgenic mice. Neuron
update [comment]. Physiol. Behav. 49:1293–1296.
27:487–497.
Youngentob, S. L. Kent, P. F., Sheehe, P. R., Schwob, J. E., and
Wright, H. N. (1987). Characterization of olfactory dysfunction.
Tzoumaka, E. (1995). Mucosal inherent activity patterns in
Arch. Otolaryngol. Head Neck Surg. 113:163–168.
the rat: evidence from voltage-sensitive dyes. J. Neurophysiol.
Xu, W., and Slotnick, B. (1999). Olfaction and peripheral olfac-
73:387–398.
tory connections in methimazole-treated rats. Behav. Brain Youngentob, S. L., Schwob, J. E., Sheehe, P. R., and Youngentob,
Res. 102:41–50. L. M. (1997). Odorant threshold following methyl bromide-
Yee, K. K., and Costanzo, R. M. (1995). Restoration of olfactory induced lesions of the olfactory epithelium. Physiol. Behav.
mediated behavior after olfactory bulb deafferentation. 62:1241–1252.
Physiol. Behav. 58:959–968. Zhang, Y., Burk, J. A., Glode, B. M., and Mair, R. G. (1998). Effects
Yee, K. K., and Costanzo, R. M. (1998). Changes in odor quality of thalamic and olfactory cortical lesions on continuous olfac-
discrimination following rec–overy from olfactory nerve tran- tory delayed nonmatching-to-sample and olfactory discrimina-
section. Chem. Senses 23:513–519. tion in rats (Rattus norvegicus). Behav. Neurosci. 112:39:53.
Youngentob, S. L., and Margolis, F. L. (1999). OMP gene dele- Zhao, H., Ivic, L., Otaki, J. M., Hashimoto, M., Mikoshiba, K.,
tion causes an elevation in behavioral threshold sensitivity. and Firestein, S. (1998). Functional expression of a mam-
Neuroreport 10:15–19. malian odorant receptor. Science 279:237–242.
19

Methods for Determining Odor Preferences

in Nonhuman Mammals

Richard L. Doty
University of Pennsylvania, Philadelphia, Pennsylvania, U.S.A.

. . . the more we study animals, the more that we per- also McIndoo, 1926). Quantitative assessment of odor
ceive that smell is the chief agent which provokes preferences and aversions in nonhuman mammals, how-
attractions and aversions between animals them- ever, appeared a number of decades later and, with rare
selves and between animals and men. exception (e.g., Swann, 1933), largely followed the devel-
opment of formal taste preference testing in the 1930s
—Gilbert Coleridge, 1920
(e.g., the two-bottle preference test) (see Richter, 1936,
1939; Richter and Campbell, 1940). Nonetheless, anecdo-
tal reports regarding mammalian odor preferences
I. INTRODUCTION appeared at earlier dates (e.g., Binet and Passy, 1895,
1896; Hamilton, 1907), and attempts to understand the
Operationally, one stimulus is assumed to be preferred tracking abilities and preferences of dogs have a long his-
to another if an animal behaviorally increases the tory (e.g., Romanes, 1887; von Uexküll and Sarris, 1931)
probability of being stimulated by that stimulus over the (see Chapter 18). In 1949, Beach and Gilmore employed
probability of being stimulated by the other. In general, boxes containing sawdust over which the urine was sprin-
behavioral inferences about preferences or aversions are kled to quantitatively demonstrate that male dogs spend
established on a relative basis, although marked prefer- more time sniffing urine odor from an estrous than from
ences or aversions are easily identified. Increased or an anestrous female and that sexual experience or compe-
decreased stimulus access on the part of an animal can be tence may be associated with this behavior. Le Magnen
achieved in a number of ways, such as by spending more (1952), using a Y-maze, showed that adult male rats
or less time in the physical proximity to a stimulus or by (Rattus norvegicus) prefer the odor of receptive females to
increasing or decreasing sniffing behavior in its presence. nonreceptive ones, whereas prepubertal or castrated rats
Despite the fact that preference paradigms have been do not (unless they have been injected with testosterone).
used to infer discrimination, it should be emphasized that Four years later, Godfrey (1958) found, using a similar
a lack of preference does not necessarily mean a lack of maze, that estrous female bank voles (Clethrionomys) pre-
discrimination. fer homospecific male odors to heterospecific male odors
Numerous invertebrate odor preference studies and that hybrids were discriminated against. Subsequent
appeared in the late nineteenth and early twentieth cen- workers like Carr et al. (1965, 1970) devised even simpler
turies, such as those examining the behavior of paramecia preference paradigms that could be placed within the sub-
(Jennings, 1906) and the pomice fly (Barrows, 1907; see ject’s home cage, demonstrating, among other things, the

403
404 Doty

important role of sexual experience in establishing hetero- 1984; Pfaff and Pfaffman, 1969; Smith, 1965; Stern, 1970;
sexual odor preferences. and Thiessen et al., 1971.*)
In this chapter, basic methodologies are described for The approach paradigm has a number of advantages.
determining odor preferences and aversions in mammals, First, it is easy to set up and can be placed within the
with an emphasis on rodents. Five types of paradigms are subject’s home cage or home range in some instances,
conceptualized and critiqued: approach, forced approach- making it amenable to both laboratory and field
avoidance, bar-press stimulus presentation, odor trail, and situations. Second, several odorants can be tested
sniff-rate analysis. Although these paradigms are not neces- simultaneously. Third, given its reliance upon simple
sarily mutually exclusive, they provide a useful taxonomy for exploratory or foraging behaviors, there is no need for
the classification and comparison of most odor preference elaborate training or shaping procedures (of either the
test situations. Tests designed to measure detection sensitiv- subject or the technician) to produce the desired behav-
ity, discrimination, and learning, including conditioned aver- ior. Unfortunately, this paradigm also has a number of
sion paradigms, are not reviewed in this chapter, and the potential disadvantages. First, depending upon the
reader is referred elsewhere for information on such pro- nature of the test situation, the possibility of odor diffu-
cedures (e.g., Slotnick, 1990; Stevens, 1975) (see also sion and mixing cannot be ruled out, particularly where
Chapters 18, 20, 40). Results of specific studies are discussed exhaust systems are absent or poorly planned, adding
only if they add to the description of a particular procedure. noise to the measurements. Second, this paradigm
assumes that a positively preferred odor elicits marked
approach behavior in the subject. This may not always
II. APPROACH PARADIGM be the case, particularly if the animal can stand and
simply sniff an odor stream without approaching its
The most popular odor preference test paradigm is one in source. Third, position biases or preferences, particu-
which two or more odorants are placed at different locations larly in males, are common in such test situations
within a home cage or test chamber or room. In some cases, (see Barnett and Spencer, 1953; Doty, 1971). Such
anesthetized conspecifics to whom an odor has been added preferences potentially limit repeated testing of the
are employed. The duration and/or frequency of investigation same subjects, require careful counterbalancing of the
of the odorant or odorized stimulus object usually serves positions of the stimulus presentations, and may
as the dependent measure, although other measures have necessitate the use of a relatively large number of
been employed as well (e.g., the amount of food or water subjects to decrease error variance. Fourth, test situa-
consumed near the stimulus) (see Barnett and Spencer, 1953; tions which use the amount of food or water consumed
Drickamer, 1972; Millman, 1968). Some investigators have as their dependent measure may confound olfactory and
required subjects to dig through sand, sawdust, or wood-chip gustatory factors (e.g., some of the “odor” stimuli may
bedding towards an odor source, using, for example, the become ingested and tasted), limiting the degree of gen-
time required to locate the stimulus as the dependent eralization of their results to other situations. Such con-
measure (e.g., Heth et al., 1992; Howard et al., 1969; Swann, founding may vary with the animal’s deprivation level.
1933). Studies employing discrete one-trial choices between For example, the odor of a novel flower may elicit more
portions of a test apparatus containing experimental odors investigation than that of food pellets in a satiated rat,
(e.g., Y-or T-maze arms) commonly assess the proportion of whereas such a flower odor associated with food may be
subjects choosing the appropriate section (e.g., Godfrey, avoided by a hungry rat in favor of food pellets not so
1958; Huck and Banks, 1984; Leone and Moltz, 1971; odorized or food pellets associated with a different odor.
Thiessen et al., 1971). When duration of investigation is the Imagine the odor of roses in your breakfast cereal at
dependent measure, such times are either recorded on stop- 7:00 a.m.! Fifth, this procedure does not allow accurate
watches or via photoelectric cells or mechanical tripping measurement of negative odor preferences or aversions.
devices connected to electrical recording systems or comput- For example, both a neutral and a highly aversive odor
ers (Additional examples of early studies to employ this may receive an initial exploratory sniffing bout by an
paradigm include Baran and Glickman, 1970; Beach and animal and then receive no more attention during a test
Gilmore, 1949; Beauchamp, 1973; Carr et al., 1965, 1970;
Doty, 1971, 1972; Doty and Anisko, 1973; Doty and Dunbar,
1974; Doty et al., 1971; Godfrey, 1958; LeMagnen, 1952; * These examples reflect studies from the first few decades of the
Lydell and Doty, 1972; Mainardi et al., 1965; Marr and widespread employment of this methodology and should not be
Gardner, 1965; Moore, 1965; O’Connell and Meredith, viewed as inclusive or even representative of the hundreds of
studies that have since used this paradigm.
Determining Odor Preferences in Nonhuman Mammals 405

session. A literal interpretation of the data would odors throughout the entire period. Third, it is conceivable
suggest, erroneously, that both odorants are equally that a preference measure inferred from this paradigm is
attractive. A sixth limitation of this paradigm is related not the same as a preference measure obtained from
to the potential interaction between a subject’s activity paradigms in which more direct comparisons of two odor
level and the preference measure. During periods of stimuli are possible (Doty, 1973). Although few data are
high circadian running activity, for example, a prefer- available that directly test this point, results from a study
ence observed during other periods may wash out, using a bar-press stimulus presentation paradigm suggest
depending upon the type of test apparatus (Doty, 1971). this possibility. Tapp and Long (1971) found that a
hierarchy of odor preferences obtained by giving a rat a
choice between a lever producing a puff of air and one
III. FORCED APPROACH-AVOIDANCE
producing a puff of odorized air correlated only 0.02 with
PARADIGM
a hierarchy obtained from a paired-comparison two-lever
situation. Research is needed to determine if similar
Under the assumption that an animal is motivated to
preferences are obtained from the approach and forced
approach attractive odorants and avoid aversive ones, a
approach-avoidance paradigms. Fourth, this paradigm
simple test employing a relatively small two-chamber
requires the testing of subjects in a test chamber indepen-
apparatus can be employed. Air or some other presumed
dent of the home cage, where competing cues are not
neutral stimulus is typically present or contingent upon the
present. Finally, a number of the shortcomings of the
subject’s presence in one chamber, whereas the experi-
approach paradigm are also present in the forced
mental odorant is present or contingent upon the subject’s
approach-avoidance situation. For example, circadian
presence in the other chamber.* Attraction of the subject to
activity and position preferences within the test apparatus
the odorant is operationally defined as his or her spending
conceivably influence the dependent measure.
more than 50% of a test period on the odor side, and avoid-
ance is defined as spending less than 50% of the test period
on that side. A preference measure between two odorants
IV. BAR-PRESS STIMULUS PRESENTATION
can be inferred indirectly by comparing the relative time
PARADIGM
spent with each odor when they are individually paired
against a control alternative in separate tests. Experimental
A number of studies from a single laboratory have used a
test situations that can be categorized as forced approach-
test paradigm in which a puff of odorized or nonodorized
avoidance include those of Brown and Willner (1983),
air is made contingent upon a bar press in a two-lever
Carter (1972), Doty (1973), Rottman and Snowdown
modified Skinner box (e.g., Long and Stein, 1969; Long
(1972), and Cornwell-Jones (1981).
and Tapp, 1967, 1970; Tapp and Long, 1968, 1971). Odors
The major advantage of the forced approach-avoidance
are subsequently exhausted from the test apparatus by
paradigm is that both aversive and attractive qualities of
means of a flow-through exhaust system. This test situa-
odorants can be determined. However, this procedure has
tion differs from the approach paradigm primarily in its
many shortcomings. First, developing preference measures
brief presentation of the odorant (usually 1-second puff/bar
for a number of odorants becomes quite time consuming.
press) and in its use of an extremely objective and measur-
Only one odorant can be tested at a time, and large num-
able dependent variable.
bers of subjects must be used to decrease error variance
This paradigm has a number of positive features. First,
resulting from large individual differences in the prefer-
various parameters of the stimulus can be independently
ence measure. Second, since the animal must be on one
manipulated with maximal precision and minimal difficulty
side of the apparatus or the other, his or her position is con-
(e.g., duration of odor puff, concentration of odorant, sched-
stantly being recorded during the test session, even though
ule of reinforcement, magnitude of energy expenditure
it is unlikely that he is attending to the environmental
required to produce an odor puff). Second, the use of bar
presses as a dependent variable allows the reinforcing prop-
erties of various odorants to be determined using standard
*Some investigators employ two odors, rather than an odor and a
operant conditioning equipment. Third, response rates
blank, in this paradigm (e.g., Brown and Willner, 1983). This is
usually done so that all pairs of a set of odorants can be tested elicited in food deprived rats for odors of powdered food
against one another. Although this does allow for the develop- (see, e.g., Long and Tapp, 1967) are quite high in
ment of a preference hierarchy, it does not allow for a delineation comparison to response rates elicited in nondeprived ones,
of the degree to which a given finding is due to attractive or aver- suggesting the sensitivity of this measure to biological need
sive qualities of the odorants involved. states of the organism. Fourth, since the subject must
406 Doty

perform a discrete behavior in order to smell an odorant, an increase in running speed is an even more difficult task
bypassing of the recording system is not possible, as can unless one can assume that running speed is related to the
occur in the approach paradigm. Fifth, a number of odorants amount of odorant reaching the olfactory mucosa, an
can be presented simultaneously in this test situation by untested and somewhat counterintuitive notion in the con-
simply adding additional levers. Sixth, a measure of odor text of a run down a short T-maze runway. Common sense
aversion could be determined, at least theoretically, by would suggest that an odor (or lack of odor) that does not
requiring a subject to push a lever to turn off an odorant. signify “frustration” or “impending illness” would be pre-
Major drawbacks of this paradigm include: (1) its reliance ferred to an odor that does. If we assume that running speed
upon complex mechanical and electrical components; (2) has no significant influence on olfactory function, then the
potential difficulties involved in adequately cleaning odor- finding that rats actually slow down in the presence of “non-
ants from the valves and tubes after their use; (3) potential reward” rather than “reward” odors would make speed
influences of position preferences and activity levels upon inversely related to preference. Given, however, the fact that
the subjects’ responses; and (4) the necessity of shaping such measures are contaminated by freezing or other emo-
animals to press the bars in some test situations. tion-related responses, it is difficult to see how this specific
paradigm would be generally useful in establishing odor
preferences.
V. ODOR TRAIL PARADIGM
VI. SNIFF-RATE ANALYSIS PARADIGM
Many mammals are capable of detecting subtle differences in
odor trails laid down by conspecifics and preferentially
Teichner (1966) and Teichner et al. (1967) developed a
follow one trail over another (e.g., Eibl-Eibesfeldt, 1970;
sophisticated electronic system to monitor and record the
Ewer, 1968). According to our earlier definition of a prefer-
number, duration, and intensity of sniffs by unrestrained rats
ence, the trail that is followed is in effect preferred to the
in an odor-controlled environment. The “repellency” and
nonfollowed one. This paradigm differs from the ones men-
“attractiveness” of various concentrations of a number of dif-
tioned earlier in terms of the unique presentation of the stim-
ferent odorants were determined by comparing the amount of
uli and the types of odorants typically tested. Douglas (1966)
sniffing during a control period to that during a period when
systematically examined various cues used by rats in pro-
an odorant was present. Decreased sniffing was the measure
ducing “spontaneous alternation” in the T-maze and found
of aversiveness and increased sniffing the measure of attrac-
that the rat’s tendency to avoid its own recently laid odor trail
tiveness. It was found that the less restriction to movement
was a major determinant of the alternation. Klein and Brown
imposed upon their subjects, the more they sniffed.
(1969) found that rats made either anosmic or both anosmic
The sniff-rate analysis paradigm is one of the more inter-
and blind did not alternate at rates significantly above chance
esting and underutilized preference test situations reported in
levels, whereas control and osmic blinded rats did alternate
the literature and, when used prudently, should provide
at such levels, supporting Douglas’s findings. These studies
important information about relative preferences to odorants.
imply that the rats preferred earlier laid trails or clean
However, one must be careful in interpreting what changes in
flooring to recently laid ones in this test situation.
sniffing rates signify. Welker (1964), for example, reports
Studies using straight alley runways indicate rats
that “mildly novel” auditory, olfactory, tactile or visual stim-
deposit a distinctive odor trail when their expectation of
uli are capable of arousing a characteristic sniffing pattern in
food reward has been recently thwarted or when they have
the rat and that removal of the olfactory bulbs has no marked
encountered a cue related to impending illness (e.g.,
effect on the average rate of sniffing or upon the integration
Batsell et al., 1990; Ludvigson and Sytsma, 1967;
of various behaviors related to sniffing. Furthermore, strong
Morrison and Ludvigson, 1970; Seago et al., 1970). The
stimulation of any of the aforementioned modalities tended
deposited odor trails can be used as discriminative stimuli
to inhibit sniffing in a number of subjects. Keeping these
in learning situations and produce decrements in the
potential issues in mind, the sniff-rate analysis paradigm still
running speed of animals passing over them.
appears to have many merits and potential applications in the
Although the tracking of one odor trail rather than
development of odor preference hierarchies.
another clearly meets our general definition of preference,
the real preference decision was made at a choice point and
the time spent on the trail can be indicative of a number of VII. CONCLUSIONS
factors, including the length or integrity of the trail. Hence,
calculations of preferences based upon time on trails is There are many reasons for wanting to know an animal’s
problematic. Attempting to define a preference in terms of preference for one odorant over another. Aside from better
Determining Odor Preferences in Nonhuman Mammals 407

understanding animal behavior and establishing factors original sex partners in rats. J. Comp. Physiol. Psychol.
responsible for mediating odor communication among ani- 71:216–222.
mals, knowledge gained from preference testing is critical Carter, C. S. (1972). Effects of olfactory experience on the behav-
in applied settings. For example, developing lures as well ior of the guinea-pig (Cavia porcellus). Animal Behav.
20:54–60.
as repellants is critical to modern wildlife-management
Coleridge, G. (1920). Animal attractions and repulsions.
programs, and preference tests are essential in such
Contemp. Rev. 117:539–545.
development. Animal preference paradigms are key com- Cornwell-Jones, C. A. (1981). Conspecific odor preferences of
ponents of research within the pet food industry and, in male albino rats are reversed by intracerebral 6-hydroxy-
some situations, even have applications in the development dopamine. Brain Res. 213:379–385.
of model systems for human food preferences. Although Doty, R. L. (1971). Homospecific and heterospecific odor prefer-
the present review is not a comprehensive assessment of all ences in sexually-naive Peromyscus maniculatus bairdi and
olfactory preference studies reported in the literature, it Peromyscus leucopus noveboracensis. Unpublished doctoral
does provide a framework that includes most odor prefer- dissertation, Michigan State University.
ence test situations devised to date. Doty, R. L. (1972). Odor preferences of female Peromyscus man-
iculatus bairdi for male mouse odors of P. m. bairdi and
P. leucopus noveboracensis as a function of estrous state.
ACKNOWLEDGMENTS J. Comp. Physiol. Psychol. 81: 191–197.
Doty, R. L. (1973). Reactions of deer mice (Peromyscus manicu-
latus) and white-footed mice (Peromyscus leucopus) to homo-
The preparation of this chapter, an update of an earlier one,
specific and heterospecific urine odors. J. Comp. Physiol.
Doty (1975), was supported, in part, by Grants PO1 DC
Psychol. 84:296–303.
00161, RO1 DC 04278, RO1 DC 02974, and RO1 AG Doty, R. L., and Anisko, J. J. (1973). Procaine hydrochloride
27496 from the National Institutes of Health, Bethesda, olfactory block eliminates mating behavior in the male
MD (R. L. Doty, principal investigator). golden hamster. Physiol. Behav. 10:395–397.
Doty, R. L., and Dunbar, I. A. (1974). Attraction of Beagles to
conspecific urine, vaginal, and anal sac secretion odors.
REFERENCES Physiol. Behav. 35:729–731.
Doty, R. L., Carter, C. S., and Clemens, L. G. (1971). Olfactory
Baran, D., and Glickman, S. E. (1970). “Territorial marking” in control of sexual behavior in the male and early-androgenized
the Mongolian gerbil: A study of sensory control and func- female hamster. Horm. Behav. 2:325–335.
tion. J. Comp. Physiol. Psychol. 71:237–245. Douglas, R. J. (1966). Cues for spontaneous alternation. J. Comp.
Barnett, S. A., and Spencer, M. M. (1953). Responses of wild rats Physiol. Psychol. 62:171–183.
to offensive smells and tastes. Br. J. Animal Behav. I:32–37. Drinkamer, L. C. (1972). Experience and selection behavior in
Barrows, W. M. (1907). The reactions of the pomice fly, the food habits of Peromyscus: use of olfaction. Behaviour
Drosophila ampelophila Loew, to odorous substances. J. Exp. 41:269–287.
Zool. 4:515–537. Eibl-Eibesfeldt, I. (1970). Ethology: The Biology of Behavior.
Batsell, W. R. J., Ludvisgon, H. W., and Kunko, P. M. (1990). Holt, Rinehart and Winston, New York.
Odor from rats tasting a signal of illness. J. Exp. Psychol. Ewer, R. F. (1968). Ethology of Mammals. Plenum, New York.
Anim. Behav. Proc. 16:193–199. Gleason, K. K., and Reynierse, J. H. (1969). The behavioral sig-
Beach, F. A., and Gilmore, R. W. (1949). Response of male dogs nificance of pheromones in vertebrates. Psychol. Bull.
to urine from females in heat. J. Mammal. 30:391–392. 71:58–73.
Beauchamp, G. K. (1973). Attraction of male guinea pigs to con- Godfrey, J. (1958). The origin of sexual isolation between bank
specific urine. Physiol. Behav. 10:589–594. voles. Proc. R. Phys. Soc. Edinburgh 27:47–55.
Binet, A., and Passey, J. (1895/1896). Contribution à l’étude de Goff, W. R. (1961). Measurement of absolute olfactory sensitiv-
l’olfaciton chez le chien. C. R. 24ms Session, Assoc. Frac. ity in rats. Am. J. Psychol. 74:384–393.
Avance. Sci. ii, pp. 659–661. Hamilton, G. van T. (1907). An experimental study of an unusual
Bronson, F. H., and Caroom, D. (197.1). Preputial gland of the type of reaction in a dog. J. Comp. Neurol. 17:329–341.
male mouse: Attractant function. J. Reprod. Fertil. Heath, G., Nevo, E., Ikan, R., Weinstein, V., Ravid, U., and
25:279–282. Duncan, H. (1992). Differential olfactory perception of enan-
Brown, R. E., and Willner, J. A. (1983). Establishing an “affec- tiomeric compounds by blind subterranean mole rats (Spalax
tive scale” for odor preferences of infant rats. Behav. Neur. ehrenbergi). Experientia 48:897–902.
Biol. 38:251–260. Howard, W. E., Palmateer, S. D., and Marsh, R. E. (1969). A body
Carr, W. J., Loeb, L. S., and Dissinger, M. E.(1965). Responses capacitor-olfactometer for squirrels and rats. J. Mammal.
of rats to sex odors. J. Comp. Physiol. Psychol. 55:415–417. 50:771–776.
Carr, W. J., Krames, L., and Costanzo, D. J. (1970). Previous Huck, U. W., and Banks, E. M. (1984). Social olfaction in male
sexual experience and olfactory preference for novel versus brown lemmings (Lemmus sibiricus
trimucronatus) and
408 Doty

collared lemings (Dicrostonyx groenlandicus): I. Discrimination by the main and accessory olfactory systems. Behav.
of species, sex, and estrous condition. J. Comp. Psychol. Neurosci. 98:1083–1093.
98:54–59. Pfaff, D, and Pfaffmann, C. (1969). Behavioral and electrophysi-
Jennings, H. S. (1906). Behaviour of the Lower Organisms. ological response of male rats to female rat urine odors. In
Columbia University Press, New York. Olfaction and Taste: Proceedings of the Third International
Klein, D., and Brown, T. S. (1969). Exploratory behavior and Symposium, C. Pfaffman (Ed.). Rockefeller University Press,
spontaneous alternation in blind and anosmic rats. J. Comp. New York, pp. 258–267.
Physiol. Psychol. 68:107–111. Ralls, K. (1971). Mammalian scent marking. Science 171:
LeMagnen, J. (1952). Les phenomenenes olfacto-sexuels chez le 443–449.
rat blanc. Arch. Sci. Physiol. 6:295–332. Richter, C. P. (1936). Increased salt appetite in adrenalectomized
Leone, M., and Moltz, H. (1971). Maternal pheromone: discrim- rats. Am. J. Physiol. 115:115–161.
ination by preweaning albino rats. Physiol. Behav. Richter, C. P. (1939). Salt taste thresholds of normal and adrena-
7: 265–267. lectomized rats. Endocrinology 24:367–371.
Long, C. J., and Stein, G. W. (1969). An analysis of the reinforc- Richter, C. P., and Campbell, K. H. (1940). Taste thresholds and taste
ing properties of food odor. Can. J. Psychol. 23:212–218. preferences of rats for five common sugars. J. Nutr. 20: 31–46.
Long, C. J., and Tapp, J. T. (1967). Reinforcing properties of Romanes, G. J. (1887). Experiments on the sense of smell in
odors for the albino rat. Psychonom. Sci. 7:17–18. dogs. Nature 36:273–274.
Long, C. J., and Tapp, J. T. (1968). An apparatus for the assess- Rottman, S. J., and Snowdon, C. T. (1972). Demonstration and
ment of the reinforcing properties of odors in small animals. analysis of an alarm pheromone in mice. J. Comp. Physiol.
J. Exp. Anal. Behav. 11:49–51. Psychol. 81:483–490.
Long, C. J., and Tapp, J. T. (1970). Significance of olfactory Seago, J. D., Ludvigson, H. W., and Remley, N. R. (1970).
tracts in mediating response to odors in the rat. J. Comp. Effects of anosmia on apparent double alternation in the rat.
Physiol. Psychol. 72:435–443. J. Comp. Physiol. Psychol. 71:435–442.
Ludvigson, H. W., and Sytsma, D. (1967). The sweet smell of Slotnick, B. M. (1990). Olfactory perception. In Comparative
success: apparent double alternation in the rat. Psychonom. Perception. Vol. 1, Basic Mechanisms, M. A. Berkley, W. C.
Sci. 9:283–284. Stebbins (Eds.). John Wiley & Sons, New York, pp. 155–214.
Lydell, K., and Doty, R. L. (1972). Male rat odor preferences for Smith, M. H. (1965). Behavioral discrimination shown by
female urine as a function of sexual experience, urine age, and allopatric and sympatric males of Peromyscus eremicus and
urine source. Horm. Behav. 3:205–212. Peromyscus californicus between females of the same two
Ludvigson, H. W., and Sytsma, D. (1967). The sweet smell of species. Evolution 19:430–435.
success: apparent double alternation in the rat. Psychonom. Stern, J. J. (1970). Responses of male rats to sex odors. Physiol.
Sci. 9:283–284. Behav. 5:519–524.
Mainardi, D., Marsan, M., and Pasquali, A. (1965). Causation of Stevens, D. A. (1975). Laboratory methods for obtaining olfac-
sexual preferences of the house mouse. The behavior of mice tory discrimination in rodents. In Methods in Olfactory
reared by parents whose odor was artificially altered. Atti Soc. Research, D. G. Moulton, A. Turk and J. W. Johnston, Jr.
Ital. Sci. Naturali e del Museo Civico di Storia Naturae di (Eds.). Academic Press, London, pp. 375–394.
Milano 104:325–338. Swann, H. G. (1933). The function of the brain in olfaction. I.
Marr, J. N., and Gardner, L. E. (1965). Early olfactory experience Olfactory discrimination and an apparatus for its test.
and later social behavior in the rat: preference, sexual respon- J. Comp. Physiol. 15:229–241.
siveness, and care of young. J. Genet. Psychol. 107: 167–174. Tapp, J. T., and Long, C. J. (1968). A comparison of the
McIndoo, N. E. (1926). An insect olfactometer (to measure reinforcing properties of stimulus onset for several sense
attraction to non-flowering plants). J. Econ. Entomol. modalities. Can. J. Psychol. 22:449–455.
19:545–571. Tapp, J. T., and Long, C. J. (1971). Olfactory preferences in rats.
Millman, B. (1968). Odor preferences and aversions in the rat. Unpublished manuscript, Vanderbilt University.
Unpublished master’s thesis, University of Calgary, Teichner, W. H. (1966). A method for studying olfaction in the
Canada. unrestrained rat. J. Psychol. 63:291–297.
Moore, R. E. (1965). Olfactory discrimination as an isolating Teichner, W. H., Price, L. M., and Nalwalk, T. (1967).
mechanism between Peromyscus maniculatus and Suprathreshold olfactory responses of the rat measured by
Peromyscus polionotus. Am. Midland Naturalist 73: 85–100. sniffing. J. Psychol. 66:63–65.
Morrison, R. R., and Ludvigson, H. W. (1970). Discrimination by Thiessen, D. D., Lindzey, G., Blum, S. L., and Wallace, P.
rats of conspecific odors of reward and nonreward. Science (1971). Social interaction and scent marking in the
167:904–905. Mongolian gerbil (Meriones unguiculatus). Anim. Behav.
Moulton, D. G. and Eayrs, J. T. (1960). Studies in olfactory acu- 19:505–513.
ity. II. Relative detectability of n-aliphatic alcohols by the rat. Uexküll, J. von, and Sarris, E. G. (1931). Das Duftfeld des
Q. J. Exp. Psychol. 12:99–109. Hundes (Hund und Eckstein). Z. Hundeforsch. 1 (3/4):55–68.
O’Connell, R. J., and Meredith, M. (1984). Effects of volatile and Welker, W. I. (1964). Analysis of sniffing of the albino rat.
nonvolatile chemical signals on male sex behaviors mediated Behaviour 22:223–244.
20

Olfactory Memory

Aras Petrulis
Georgia State University, Atlanta, Georgia, U.S.A.

Howard Eichenbaum
Boston University, Boston, Massachusetts, USA

I. INTRODUCTION olfactory nucleus, piriform cortex, olfactory tubercle, cor-

tical amygdala, and the entorhinal cortex (ENT). The piri-
A wealth of data exists on olfactory memory and its neural form cortex (PIR), the largest area and the one most
substrates in experimental animals and, increasingly, in extensively innervated by the OB, is heavily intercon-
humans. This review is an attempt to comprehensively sur- nected with other olfactory structures and provides the
vey this research within the domain of studies on learning majority of olfactory input to the orbitofrontal cortex
and memory mediated by the main olfactory system in (OFC) both directly and indirectly via the mediodorsal
adult mammals. Even within these limitations, there is a thalamus (MDthal). PIR also has extensive bidirectional
voluminous literature from many different behavioral para- connections with ENT, which, in turn, provides the major-
digms. To organize this review, we have subdivided the lit- ity of cortical connections with the hippocampus. Not sur-
erature to highlight important differences in mnemonic and prisingly, these interconnected components are all
cognitive processes inherent in various memory tasks. We involved in olfactory memory, but each area appears to
deal first with olfactory discrimination learning, a process have different and unique functional properties in both ani-
in which odors are learned by association with positive and mals and humans (Eichenbaum, 1997; Savic et al., 2000).
negative reinforcers. Next, we consider several paradigms
that assess the recognition of odors. Third, we examine the
II. DISCRIMINATION LEARNING
existing literature on more complex forms of olfactory
memory that involve the formation of associations between A. Appetitive Conditioning of Odor Cues
odors and other stimuli (including other odors). Lastly, we
1. Behavior
include evidence showing how olfactory learning is impor-
tant for the social lives of animals. Some of the earliest attempts to train animals to discrimi-
Our aim is to understand how information flows within nate between odors have involved appetitive conditioning;
the main olfactory system and its projection targets. The that is, rewarding animals with water or food for
anatomy of this system is well known and has been performing some action (nose-poke, bar-press) after
reviewed extensively (e.g., Haberly, 1985; Shipley and detecting particular odors but not after detecting other
Ennis, 1996) (see also Chapters 1–9). Briefly, olfactory odorants (Slotnick, 1990) (see Chapter 18). This basic
receptor neurons project to the apical dendrites of technique has persisted and has undergone refinements
mitral/tufted cells in the olfactory bulbs (OB). These pro- over time that have introduced better temporal and spatial
jection neurons make reciprocal synapses with inhibitory control over odor stimulation (Nigrosh et al., 1975;
granule cells, via their basal dendrites, and project mainly Slotnick and Katz, 1974; Slotnick and Nigrosh, 1974). The
through the lateral olfactory tract (LOT) to the anterior use of computer-controlled olfactometers, in particular, led

409
410 Petrulis and Eichenbaum

to substantial progress in understanding the behavioral and 2. Lesion Studies

neural processes underlying olfactory memory and percep-
Before the mid-1970s, some studies failed to find deficits
tion in nonhumans (Slotnick, 1990). Moreover, increased
in olfactory discrimination tasks following substantial
interest in studying olfactory discrimination and memory
lesions to cortical and subcortical components or projec-
followed the realization that olfactory cues are highly
tions of the main olfactory system (Brown, 1963; Kimble
salient stimuli for laboratory rodents and can be used to
and Zack, 1967; Lashley and Sperry, 1943; Schuckman et
probe the cognitive and neural architecture of stimulus rep-
al., 1969; Swann, 1934, 1935), although other reports
resentation, learning, and memory (Davis and
indicated deficits following lesions of temporal lobe in
Eichenbaum, 1991; Slotnick, 1990).
monkeys (Brown et al., 1963; Santibanez and Hamuy,
Rodents learn to discriminate odors more rapidly than
1957). More recent lesion studies, outlined below, have
auditory or visual cues (Nigrosh et al., 1975; Slotnick,
shed light on the differential roles of distinct components
1984), even when extremely similar odors such as those of
of this system.
individual hetero- and conspecifics are used (Bowers and
Alexander, 1967; Gheusi et al., 1997; Schellinck et al.,
1991; Yamazaki et al., 1990). Furthermore, rats rapidly a. Olfactory Bulb and Cortex. The PIR is widely
develop a learning set for odor discriminations, reflecting considered the olfactory system structure most likely to be
their acquisition of abstract rules or procedures, such as critical for olfactory memory (Ambros-Ingerson et al.,
“win-stay, loose-shift” (Jennings and Keefer, 1969; Nigrosh 1990; Haberly and Bower, 1989; Hasselmo et al., 1990;
et al., 1975; Slotnick, 1984; Slotnick and Katz, 1974). see also Davis and Eichenbaum 1991). PIR damage
Although this interpretation has been questioned (Reid and severely impairs discrimination of odor mixtures that share
Morris, 1992, 1993), subsequent research revealed that components and prevents development of an odor learning
learning set formation is robust (Slotnick, 1994). Rats also set (Staubli et al., 1987b). However, pretraining and use of
develop reversal learning sets for odor discrimination simple odorants can alleviate these deficits (Staubli et al.,
(Nigrosh et al., 1975; O’Grady and Jennings, 1974). 1987b; Zhang et al., 1998). Transection of the lateral olfac-
The strength and nature of odor discrimination learning tory tract sparing OB and anterior PIR impairs odor dis-
has been explored using measures of positive savings in crimination learning with brief, but not long, intertrial
relearning discriminations and of negative savings in learn- intervals (Slotnick, 1985; Slotnick and Berman, 1980;
ing the reversal of a previously acquired discriminations. Slotnick and Risser, 1990; Slotnick and Schoonover, 1992;
These assessments show that animals can remember many Thanos and Slotnick, 1997). In addition, olfactory infor-
different pairs of odors for at least several weeks, indicat- mation needed for discrimination performance can also
ing that olfactory memory is both quite large and highly reach the PIR via fibers running within the anterior com-
resistant to interference and degradation [rats (Slotnick missure as well as the LOT (Bennett, 1968; Slotnick and
et al. 1991; Staubli et al., 1987a), squirrel monkeys (Laska Schoonover, 1992). This pattern of results suggests that the
and Hudson, 1993; Laska et al., 1996)]. Rats encode both posterior PIR and ENT may be necessary for longer-term
the positively and the negatively reinforced odors within a odor memories and that short-term memories can be main-
discrimination, rather than simply ignoring one of the tained in anterior PIR. Furthermore, PIR may be necessary
cues, and they appear to encode, like humans (see Chapter for discriminations between highly similar odorants, but is
10), multicomponent odors as unitary, “gestalt” stimuli not critical for easy discriminations, which may be sup-
and not as a collection of independent components. Also, ported by other anterior olfactory structures such as the
rats quickly learn olfactory discriminations with intertrial AON (Hamrick et al., 1993; Slotnick and Schoonover,
intervals ranging from several seconds to 30 minutes, 1992). In addition, posttraining inactivation of the OB
demonstrating that they form odor-reward associations impairs retention of discriminations of “electric odors,”
even with long delays between each odor presentations suggesting a role for the OB in memory consolidation
(Lovelace and Slotnick, 1995). (Mouly et al., 1993).
Olfactory discrimination has also been explored by
training rats to differentiate electrical stimulation of dis- b. Hippocampus. Early studies reported no deficit in
tinct regions of the olfactory bulbs or lateral olfactory tract. olfactory discrimination following hippocampal lesions
Rats retain discriminations between “electrical odors” over (Allen, 1940; Swann, 1934; Kimble and Zack, 1967). In
long periods, form learning sets for them, and can dis- addition, several early studies reported that damage to sep-
criminate stimulation of extremely close regions of the tal nuclei and its connections to the hippocampus via the
bulb (Mouly and Holley, 1986; Mouly et al., 1985; Roman fornix, as well as damage to the hippocampus proper, can
et al., 1987). facilitate acquisition of odor discriminations (Carlson and
Olfactory Memory 411

Vallante, 1974; Schmajuk and Isaacson, 1984; Vom Saal The deficits following MDthal ablations are less severe
et al., 1975). By contrast, lesions of the lateral entorhinal than those after OFC damage, although animals with
cortex (ENT) severely impair the acquisition of simultane- MDthal damage are more impaired in discriminations of
ous olfactory discriminations acquired with long intertrial qualitatively similar and novel odors (Eichenbaum et al.,
intervals and impaired retention of a discrimination 1980; but see Lu and Slotnick, 1990). In addition, using
learned one hour before (Staubli et al., 1984). However, simultaneous presentations of odors in an olfactory maze,
ENT lesions do not affect preoperatively acquired odor Staubli et al. (1987b) demonstrated that rats with MDthal
discriminations, suggesting this area is not the critical site lesions were severely impaired in initial learning, but par-
of memory storage (Staubli et al., 1986). tially recovered with extensive overtraining. Also, MDthal
Spared, impaired, or facilitated performance on olfac- lesioned rats are severely impaired in odor reversal learn-
tory discrimination following hippocampal system dam- ing (Slotnick and Kaneko, 1981) and do not form learning
age depends on whether odors are presented in a way sets during olfactory discrimination training (Lu and
that facilitates or hinders learning relationships between Slotnick, 1990). However, MDthal is not critical for reten-
them. Rats with damage to the hippocampal system, via tion of preoperatively learned odor discriminations
fornix transection (FNX) or ENT lesions, acquire odor (Slotnick and Risser, 1990). This combination of results
discriminations and learning set for the task as quickly suggests that the MDthal projections to OFC allow animals
as normal animals if the odors are presented successively to rapidly organize odor-reward associations (McBride and
in a go/no-go task (hold nose in odor port for rewarded Slotnick, 1997).
odors, withdraw for unrewarded odors) (Eichenbaum
et al., 1986, 1988; Otto et al., 1991b). By contrast, ani- d. Amygdala. Generally, lesions or disconnection of
mals with FNX and ENT damage perform poorly when the amygdala do not impair olfactory discrimination learn-
two odors are presented simultaneously and the animal is ing in rats or rhesus monkeys (Eichenbaum et al., 1986;
rewarded for nose-poking into the source of the positive Schuckman et al., 1969; Slotnick, 1985). It is likely that the
odor (Eichenbaum et al., 1988; Otto and Garruto, 1997), amygdala is not critically involved in memory formation
although they can learn some discriminations by treating that requires incremental learning involving small rewards.
the odors as a compound stimulus (Eichenbaum et al., Instead, it may be primarily engaged when learning about
1989). It has been argued that this task requires not only odors with strong affective associations (see below).
that animals remember individual odor cues and their
valences but also encourages the animal to learn the rela-
3. Cellular Correlates
tionship between simultaneously presented odors. Short-
term maintenance of memory for individual odors does a. Olfactory Bulb and Cortex. Some of the earliest
not depend on processing by either the hippocampus or evidence for physiological correlates of olfactory learning
ENT (Eichenbaum et al., 1994). However, maintenance came from EEG recordings in the OB and PIR of rabbits.
of odor memories for longer periods does seem to Stable changes in the amplitude and location of OB EEG
require the posterior PIR and/or the ENT (Thanos and bursts were observed following presentation of condi-
Slotnick, 1997). tioned odors, suggesting the EEG primarily reflects the
reinforcement history of odorants rather than their sensory
c. Orbitofrontal Cortex and Mediodorsal Thalamic qualities (Di Prisco and Freeman, 1985; Freeman, 1991;
System. Lesions of the OFC and the MDthal result in a Freeman and Schneider, 1982; Grajski and Freeman,
pattern of deficits different from that observed following 1989). Also, large-scale changes in the synaptic dynamics
hippocampal damage. Like hippocampal or ENT lesions, between OB and PIR have been observed during sniffing
damage to the OFC and MDthal does not impair basic to conditioned odors, but not to unreinforced odors
olfactory perception, in that detection ability and odor (Bressler, 1988). These findings suggest the OB plays
thresholds are similar to sham animals. In contrast, rats a dynamic role in the formation of odor memories.
with either OFC or MDthal lesions display severe deficits Other studies have examined changes in PIR synaptic
in go/no-go successive odor discrimination (Eichenbaum physiology following odor discrimination learning. Roman
et al., 1980). Furthermore the impairment could be attrib- et al. (1987) demonstrated that “electrical odor” discrimi-
uted to increased perseveration of responses, a hallmark nation learning leads to long-term potentiation (LTP) in
of prefrontal damage (Fuster, 1989; Kolb, 1984). In this PIR. Furthermore, population responses of PIR neurons to
case the deficit is selective to the olfactory domain, LOT stimulation increased with the number of successful
as OFC rats are not impaired in spatial alternation discrimination trials but not if the animals were presented
(Eichenbaum et al., 1983a). with familiar but unconditioned odors (Roman et al.,
412 Petrulis and Eichenbaum

1993a). Increased excitability of PIR neurons, as well as rewarded odors, irrespective of odor identity. In addition,
synaptic facilitation between PIR neurons, was also hippocampal neurons responded to highly specific combi-
observed following odor discrimination training (Saar nations of odor and location and in association with particu-
et al., 1998, 1999). lar odor pairs or sequences, supporting the idea that the
Given the importance of the piriform cortex in olfac- hippocampus is primarily involved in processing the rela-
tory perception and memory, it is surprising that only a tionship between various stimuli (Eichenbaum et al., 1994).
handful of studies have evaluated PIR neuronal firing pat- During the acquisition of odor discriminations, the hip-
terns during odor discrimination learning. McCollum pocampal theta rhythm synchronizes with sniffing
et al. (1991) reported that most PIR neurons rapidly habit- (Macrides et al., 1982). Also, hippocampal neurons tend to
uate to discriminated odors. Moreover, very few cells fire in high-frequency bursts in phase with the ongoing
responded to more than two odors presented to the rats, theta rhythm, suggesting a link between conditions optimal
indicating a sparse coding in the PIR (see also for synaptic plasticity and periods of odor stimulus evalu-
Schoenbaum and Eichenbaum, 1995a). Most interest- ation and learning (Otto et al., 1991a). For example, dis-
ingly, like the OB (Kay and Laurent, 1999), PIR neurons crimination of “electrical odors” results in the enhanced
respond to a variety of discrimination task parameters that excitability of ENT inputs to the dentate gyrus (DG)
are not obviously olfactory in nature, including trial initi- (Chaillan et al., 1996). Enhancement of the DG field
ation cues and reward (Schoenbaum and Eichenbaum, potential emerges early in training and the degree of poten-
1995a). Some PIR neurons also fire associated with the tiation is correlated with discrimination performance
acquired reward-valence of odors, with predictive rela- (Chaillan et al., 1999). These findings suggest that plastic-
tionships between odor cues when one odor occurred reli- ity at the first steps of hippocampal processing occurs
ably prior to another, and with the expectation of reward. when the animal is learning the significance of stimuli.
Thus, PIR neurons encode associations between nonolfac-
tory and olfactory cues that underlie odor discrimination d. Interactions Between Systems. Using expression
performance, suggesting PIR might be a locus of long- of the c-fos gene as a marker for neuronal activation, Hess
term odor memories. et al. (1995a,b, 1997) have shown that different areas of the
brain are activated during distinct stages of olfactory dis-
b. Orbitofrontal Cortex. OFC neurons fire in associ- crimination learning. During exploration of the olfactory
ation with many events throughout all periods of an odor maze prior to training and during the learning of the oper-
discrimination trial. Many OFC neurons are responsive to ant task (nose-poking), all divisions of the hippocampus, as
particular odors, but these responses are usually not spe- well as PIR, the granule cell layers of the OB, and the ante-
cific to a single odor (Tanabe et al., 1975a), and most OFC rior medial amygdala rats were activated. The basolateral
cells fire more associated with the valence of particular amygdala was also activated during discrimination learn-
odors and the consumption of water reward that with odor ing, suggesting that this area is engaged, even if its role is
quality (Alvarez et al., 1999; Schoenbaum and not necessary to odor discrimination (Eichenbaum et al.,
Eichenbaum, 1995a). Like PIR, some OFC neurons also 1986; Slotnick, 1985). During initial learning, CA3 is more
encode predictive relationships between odors and the activated than CA1, and the reverse was observed later,
expectation or anticipation of reward. However, ensemble indicating a shift of processing within the hippocampus.
activity in the OFC is predominantly correlated with Finally, with overtraining, PIR and the DG activity became
“essential” trial information, such as odor identity and correlated, suggesting a facilitation of information transfer
valence, and not with incidental information such as pre- between cortex and the hippocampus (Hess et al., 1995b).
dictive odor-odor relationships (Schoenbaum and
Eichenbaum, 1995b). During initial acquisition, OFC neu-
ronal responses to odor identity and valence increase in
4. Biochemical Substrates
selectivity with improved performance and continue to
change, even in animals that have learned the task well a. Glutamatergic System. Antagonists of gluta-
(Alvarez et al., 1999). matergic N-methyl-D-aspartate (NMDA) receptor impair
acquisition of discriminations between low-intensity odors
c. Hippocampus. During successive and simultane- presented at long intertrial intervals, suggesting that
ous odor discrimination, learning neurons in the CA1 NMDA receptors are involved in persistent synaptic
region encode the full array of task-relevant parameters changes (Griesbach et al., 1998; Staubli et al., 1989). Also,
(Eichenbaum et al., 1987; Wiener et al., 1989). In particu- drugs that increase fast (AMPA-mediated) glutamatergic
lar, many cells respond preferentially to presentation of conductance facilitate olfactory discrimination learning
Olfactory Memory 413

(Larson et al., 1995; Staubli et al., 1994b; see also Morris administered during acquisition (Mouly et al., 1990).
and Davis, 1994). Manipulation of glutamate receptors Injections of NE directly into the OB during discrimination
after learning are ineffective, suggesting a selective role in training stimulates the EEG changes normally seen follow-
encoding odor memories (Staubli et al., 1994a, 1996; see ing exposure to novel odors and delayed EEG habituation
also Miserendino et al., 1990; Morris et al., 1986). In addi- to repeated presentations of unreinforced odors (Gray et al.,
tion, reduction of synaptic plasticity via inhibition of cel- 1986). Moreover, NE is released in the OB of mice per-
lular protease also impairs acquisition of olfactory forming an odor discrimination task (Brennan et al., 1998).
discriminations (Staubli et al., 1985). These findings support the idea that NE is involved in odor
learning (Gray et al., 1986) and may be important for odor
b. Cholinergic System. Acetylcholine (ACh) has memory consolidation (Sara et al., 1999).
been shown to modulate PIR synaptic physiology by
increasing the excitability of pyramidal cells and suppress- B. Aversive Conditioning of Odor Cues
ing intrinsic, but not afferent, connections of pyramidal
cells (Hasselmo and Bower, 1993). Formal models suggest Learning about aversive situations must, at some level,
that increased ACh release puts PIR into an acquisition require different neural circuits than those that underlie
mode, whereas decreased ACh tone facilitates retrieval of appetitive learning. These differences may be expected in
patterns stored within pyramidal cell ensembles (Hasselmo pathways for reinforcement and behavioral output, as well
and Bower, 1993). This modulation may be critical for as in the autonomic nervous system. At the same time, one
allowing new odor representations to be incorporated into might expect similarities of responses in the olfactory sys-
the network without massive interference between the new tem itself. The early literature does suggest similarities in
and old representations (Hasselmo, 1995, 1999). In sup- the neural systems underlying olfactory discrimination
port of this model, injections of ACh receptor antagonists learning using aversive and appetitive reinforcers. For
preferentially impair a rat’s ability to discriminate odor example, in studies of conditioned paw-lifting to odors fol-
pairs when one odor in the pair had a different reward his- lowing odor-shock pairing (Allen, 1937), lesions of the
tory (De Rosa and Hasselmo, 2000). Conversely, increas- piriform cortex, amygdala, fornix, hippocampus, and neo-
ing ACh activity facilitates discriminations between cortex did not impair odor conditioning, although both pir-
compound odors and their elements (Doty et al., 1999). iform and prefrontal cortex damage resulted in a lack of
The major source of ACh to the olfactory system is discrimination between odors (Allen, 1938, 1940, 1941).
from the horizontal limb of the diagonal band (HLDB) Rats with similar lesions could also acquire shock-moti-
(Linster and Hasselmo, 2000, Zaborszky et al., 1986). vated olfactory discriminations (Brown and Ghiselli,
HLDB stimulation replicates the effects of application of 1938). Moreover, only damage to various hypothalamic
ACh agonists on the dynamics of piriform cortex physiol- structures and the piriform cortex impaired retention of a
ogy, indicating that increased activity of HLDB is critical shock-motivated odor discrimination in rats, whereas
for putting the PIR into an acquisition mode (Linster et al., lesions to various neocortical structures, mediodorsal thal-
1999). Furthermore, rats with HLBD lesions are impaired amus, hippocampus, amygdala, septum, and cerebellum
in acquiring an odor discrimination with long but not short had little effect (Thompson, 1980a,b,c). These findings,
intertrial intervals, and these animals rapidly forget odor indicating that only damage to primary olfactory structures
discriminations (Roman et al., 1993b). consistently impairs simple olfactory discriminations
using aversive stimuli, are consistent with the results on
c. Monoaminergic System. Noradrenergic (NE) olfactory learning using appetitive reinforcers. (Long and
release in the olfactory system has profound modulatory Tapp, 1970; Swann, 1935; Thanos and Slotnick, 1997).
effects on both the OB and PIR. NE decreases intrinsic More recent studies have demonstrated that odors can
activity within the OB and PIR without greatly affecting serve as conditioned stimuli in Pavlovian fear-conditioning
afferent input (Hasselmo et al., 1997; Jiang et al., 1996). paradigms (Otto et al., 1997; Richardson et al., 1999) and
This may facilitate processing of weak odor stimuli by the that the neural substrates underlying performance on this
OB (Jiang et al., 1996) and may prime the PIR for acquisi- task are somewhat different than those underlying
tion of olfactory information (Hasselmo et al., 1997). performance on appetitive learning tasks. Otto et al. (1997)
Perfusion of the OB with ß-adrenergic antagonists prevents found that rats retain conditioned freezing to an odor for
the normal change in EEG patterns following exposure to more than 2 weeks after repeated pairing with foot shock.
conditioned odors during olfactory discrimination training Similarly, odor-shock pairings result in an increased startle
(Gray et al., 1986). In addition, intrabulbar injections of NE response to a loud tone in the presence of the conditioned
antagonist can impair long-term retention of odors when odor but not other odors (Richardson et al., 1999).
414 Petrulis and Eichenbaum

As observed in visual or auditory fear-conditioning A. Delayed Match and Nonmatch to Sample Tasks
tasks (LeDoux, 1995), lesions of the basolateral amygdala
1. Behavior
(BLA) impairs freezing both to the conditioned olfactory
cues and to shock, implicating the amygdala as a critical Several simple recognition memory tests have been used
node in the expression of fear responses (Cousens and to develop animal models of amnesia. Of these, the most
Otto, 1998). Rats with lesions of the anterior perirhinal successful and most widely adopted is the delayed non-
cortex (PRC), a structure with substantial and reciprocal matching to sample test (DNMTS) (Eichenbaum et al.,
connections to the BLA and olfactory areas, also are 2000; Gaffan, 1974; Mishkin, 1978). As first applied to
impaired in olfactory fear conditioning (Herzog and Otto, monkeys, this test consists of a sample phase during
1997, 1998). By contrast, lesions of the BLA or the PRC which an animal is rewarded for displacing a novel com-
do not impair odor discrimination using water reward plex object, then a variable delay period, then a test phase
(Eichenbaum et al., 1986; Otto et al., 1991b; Slotnick, where the animal is rewarded for selecting a novel (non-
1985), suggesting differences in the neural substrates matching) object over the familiar one (Eichenbaum et al.,
underlying appetitive and aversive olfactory memory. 2000). Other studies have used a continuous recognition
In a more direct comparison, rats with BLA lesions variant in which a series of stimuli are presented, and the
normally prefer an odor previously paired with a sucrose animal must respond differentially to each stimulus
reward but do not avoid odors previously paired with foot- depending on whether it is a match or nonmatch with the
shock (Cahill and McGaugh, 1990), indicating a selective previous stimulus.
deficit in aversive odor learning. However, BLA lesions do Rats are able to rapidly acquire and perform odor-
not impair odor aversion induced by pairing with quinine, guided DNMTS and match to sample (DMTS) tasks at
a mildly aversive stimulus, indicating that the BLA is levels of accuracy observed in monkeys on visually guided
involved in olfactory memory only when learning involves versions of the task, and performance is sensitive to simi-
highly arousing stimuli. Consequently, the observed speci- lar parametric manipulations (Lu et al., 1993; Otto and
ficity of the amygdala in aversive olfactory conditioning Eichenbaum, 1992a). For example, both species show
may simply reflect the fact that few appetitive stimuli pro- more forgetting if the delay between the sample and choice
duce high levels of arousal, whereas aversive stimuli often phases is increased (Koger and Mair, 1994; Otto and
do. Nevertheless, the differential involvement of PRC in Eichenbaum, 1992a). Also, in both species performance
aversive versus appetitive olfactory learning suggests that decreases substantially when the same stimuli appear more
real differences may exist in how positive and negative frequently (Koger and Mair, 1994; Otto and Eichenbaum,
valence is attributed to odor cues. 1992a). In both species, performance is sensitive to aging
Conditioned affective responses to odors may also (Zyzak et al., 1995).
involve a part of the caudate-putamen, because damage to
the ventrolateral aspect of this area impairs acquisition of
2. Lesion Studies
conditioned suppression of water licking when exposed to
an odor previously paired with foot shock (Viaud and White, a. Hippocampus. Selective lesions of the PRC and
1989). Processing of this conditioned emotional response ENT of rats and monkeys dramatically impair object-cued
involves dopaminergic activity in the ventrolateral neostria- DNMTS performance, whereas damage to the hippocam-
tum as posttraining injections of amphetamine or dopamine pus or to FNX produces either no deficit or less severe and
D2 receptor agonist into this area facilitate retention of the transient deficits (Mumby and Pinel, 1994; Mumby et al.,
conditional response (Viaud and White, 1989; White and 1992; Murray and Mishkin, 1998; Zola-Morgan et al.,
Viaud, 1991). 1989). Similarly, hippocampal removal or FNX transection
does not impair olfactory-guided DNMTS performance
with memory delays of less than 15 minutes (Mair et al.,
III. RECOGNITION MEMORY 1998; Otto and Eichenbaum, 1992a; Sutherland et al.,
1989) but may impair performance over longer delays
A critical feature of adaptive behavior in animals and (30–60 min) (Dudchenko et al., 2000). In contrast, rats
humans is the ability to remember previously encountered with damage to the PRC and ENT are impaired at delays
stimuli over relatively long time intervals and to judge the of 30 and 60 seconds (Otto and Eichenbaum, 1992a).
relative familiarity or novelty of current percepts. The abil- These animals acquire the task normally and perform well
ity to recognize prior occurrence of individual stimuli has at short delays (3 sec), indicating the PRC and ENT are
been well characterized at the behavioral and neural levels selectively involved with maintaining the memory of the
(Eichenbaum et al., 2000). odor for extended periods.
Olfactory Memory 415

b. Orbitofrontal Cortex. Lesions of the OFC result showed the preponderance of odor-selective firing during
in a pattern of deficits on DNMTS different from those fol- the choice phase. In addition, the odor-specificity of
lowing hippocampal system lesions. Unlike hippocampal match/nonmatch neural responses decreased from the DG
or PRC-ENT lesions, OFC lesions severely impair to CA3 to CA1, suggesting an increased level of abstrac-
DNMTS acquisition (Koger and Mair, 1994; Otto and tion through the trisynaptic circuit. Finally, although
Eichenbaum, 1992a). However, once the task is learned, increased or decreased neuronal activity was observed
OFC lesions do not affect memory, except in conditions of across delays in hippocampal neurons, none of these
high stimulus interference and longer delays. MDthal responses was specific to any particular odor, suggesting
lesions have minimal effects, indicating that the task-rele- that representations of specific odors are not maintained in
vant information used by the OFC is not transmitted the hippocampus.
through its connections with the MDthal (Zhang et al., Neurons in the parahippocampal region (PRC and
1998). ENT) and OFC differ from hippocampal neurons in their
responses associated with DNMTS events (Ramus and
Eichenbaum, 2000; Young et al., 1997). In contrast to the
c. Olfactory Cortex. Lesions of PIR impair retention hippocampal neurons, many cells in PRC-ENT and OFC
and reacquisition on DNMTS irrespective of the memory are odor-selective during odor sampling and maintain this
delay and level of interference (Zhang et al., 1998). This odor-selectivity during the delay period. In addition, unlike
suggests that short-term recognition of odors requires PIR hippocampal neurons, PRC-ENT and OFC neurons fire in
processing. In addition, both systemic and intraolfactory association with match/nonmatch status of specific odors.
bulb injections of ACh receptor antagonists impair DMTS Differences are also apparent between PRC-ENT and OFC
performance at 30-second but not 4-second memory delays neurons. More OFC cells fire associated with the reward
(Ravel et al., 1992, 1994). These results, along with other and to match/nonmatch features of the task than PRC-ENT
evidence (see below), indicate an important role for ACh in cells. By contrast, more PRC-ENT cells show stimulus-
memory early in the central processing of olfactory cues. selective activity firing over delay periods (Ramus and
Eichenbaum, 2000; Young et al., 1997). In general, OFC
cellular responses appear to correlate with important task
3. Cellular Correlates
parameters, whereas PRC-ENT neurons maintain the rep-
Several studies have reported task-related neuronal activity resentations of particular odor stimuli over delay periods.
in the hippocampus, PIR, ENT, and OFC of rats perform-
ing DNMTS tasks. Although neurons in all of these areas
respond to virtually all aspects of task performance, differ- B. Juvenile Recognition
ences exist in the proportion of cells encoding various 1. Behavior
aspects of the task, and some of these differ between ver-
sions of the task. For example, in a DNMTS task in which Recognition memory can also be demonstrated using etho-
animals nose-poked into an odor port for reward, CA1 neu- logically relevant tasks that do not require extrinsic
rons in the hippocampus responded indiscriminately to all rewards. Thor and Holloway (1982) described a test of
match/nonmatch decisions rather than to the odor identity social memory that takes advantage of a rat’s inherent
on any given match/nonmatch episode (Otto and curiosity in novel conspecifics. In the initial study, adult
Eichenbaum, 1992b). In contrast, when rats performed the male rats investigated a juvenile for several minutes
formally identical task in a rich spatial environment where whereupon, after a variable delay, the adult was presented
digging in odorized sand for food was the operant with either the same juvenile or a novel juvenile. After
response, hippocampal cells encoded not only the abstract delay intervals of 30 minutes or less, male rats investigated
match/match rule and positional information, but also odor the familiar juvenile less than they had on the first trial,
identity (Wood et al., 1999). whereas males presented with a novel juvenile maintained
Wiebe and Staubli (1999) reported that CA1/CA3 hip- a high rate of investigation. This phenomenon does not
pocampal neurons also fire differentially to odors during appear to be due to differences in behavior of novel and
the sample choice phases of the DNMTS task. familiar juveniles, as juveniles are not able to remember
Interestingly, neurons in different subfields of the hip- the adult animal over delays exceeding 10 minutes (Thor
pocampus had distinctive firing patterns depending on the and Holloway, 1982) and almost all social encounters are
particular phase of their version of the task. Most strik- initiated by the adults (Gheusi et al., 1994). The ability to
ingly, CA1 cells showed the greatest odor-selective activ- recognize a familiar juvenile over long delays is facilitated
ity during the sample phase, whereas dentate gyrus cells by repeated exposure to the familiar juvenile (Sekiguchi et
416 Petrulis and Eichenbaum

al., 1991b) and can be impaired by repeated exposure to a dence exists that septo-hippocampal AVP may be more
novel juvenile during the delay period (Thor and involved in juvenile recognition than in other hippocam-
Holloway, 1982), suggesting the memory is subject to pal-dependent memory tasks (Engelmann et al., 1992,
interference. Female rats and mice are able to remember 1996). In addition, AVP may have effects on early stages
juveniles over longer delays (up to 2 hours) and, unlike of olfactory processing; injections of AVP directly into the
male rats, do not appear to require the vasopressinergic olfactory bulbs (OB) improve memory, and this effect
system for modulation of this memory (Bluthe and appears to be due to AVP modulation of NE activity, as
Dantzer, 1990; Bluthe et al., 1993). depletion of bulbar NE eliminates this facilitatory effect
Several lines of evidence indicate that juvenile recogni- (Dluzen et al., 1998, 2000).
tion is mediated by olfactory cues. First, adults decrease Other peptides, such as oxytocin (OXY) and cholecys-
their investigation of juveniles if their odors were pre- tokinin (CCK), also modulate juvenile recognition. For
sented on the initial trial (Sawyer et al., 1984). Second, example, OXY release facilitates recognition, although
adult investigation of juveniles is primarily centered on the this effect is more important for female than male rats,
ano-genital area (Gheusi et al., 1994), and removal of the and appears to have effects on different neural structures
preputial glands, a prominent source of chemical signals in than does AVP (Engelmann et al., 1998; Popik and van
rats, impairs recognition of juveniles (Popik et al., 1991a). Ree, 1991; van Wimersma Greidanus and Maigret, 1996).
Third, removal of the olfactory bulbs eliminates the reduc- CCK can either facilitate juvenile recognition by stimu-
tion in investigation of familiar juveniles (Dantzer et al., lating CCK A receptors in the periphery (via the vagus
1990). Although juvenile recognition has been widely nerve) or impair memory for juveniles via CCK B recep-
attributed to functions of the vomeronasal organ (VNO), tors in the CNS (Lemaire et al., 1992, 1994a,b), suggest-
removal of the VNO results in only transient impairments ing multiple mechanisms for peptide action on mnemonic
(Bluthe and Dantzer, 1993), whereas relatively selective processes.
peripheral olfactory lesions impair recognition (Popik
et al., 1991a), indicating that the main olfactory system b. Neurotransmitters. Modulation of more tradi-
may be more important. tional neurotransmitter systems also can affect recogni-
tion of juveniles. Acute depletion of norepinephrine (NE)
in the central nervous system (CNS) impairs juvenile
2. Biochemical Substrates
recognition at 30-minute delays, whereas increasing NE
a. Neuropeptides. The juvenile recognition para- release facilitates memory in the face of retroactive inter-
digm has been widely adopted by behavioral pharmacolo- ference (Griffin and Taylor, 1995). However, increased
gists as a fast and inexpensive assay for the mnemonic NE release postinvestigation does not extend recognition
effects of various drugs and endogenous neurochemicals. after a long delay, suggesting that NE does not directly aid
Most studies have focused on the role of the neuropeptide consolidation but may, instead, reduce interference from
arginine-vasopressin (AVP). Several lines of evidence sup- intervening stimuli (Griffin and Taylor, 1995). While the
port the idea that increased vasopressin release in the site of NE action on juvenile recognition is not well char-
rodent septo-hippocampal system, after investigation of a acterized, the amnestic effect of CNS-wide NE depletion
juvenile, improves memory. Stimulating release of AVP or cannot be attributed to effects on the OB, as selective
giving injections of AVP into the septum result in recogni- depletion of NE in the OB does not impair recognition
tion of the familiar juvenile over long delays (Dantzer et (Dluzen et al., 1998).
al., 1988; Engelmann and Landgraf, 1994, 1995; Recognition can also be facilitated by dopaminergic
Engelmann et al., 1994; LeMoal et al., 1987; Popik et al., activity in the nucleus accumbens (Ploeger et al., 1991).
1991b; Sekiguchi et al., 1991a), whereas blockade of AVP Similarly, increasing cholinergic (ACh) activity following
action in the septum/hippocampus impairs recognition investigation of juveniles or ovariectomized females facili-
over short delay intervals (Dantzer et al., 1987; Landgraf et tated memory over long delays, whereas blockade of mus-
al., 1995; van Wimersma Greidanus and Maigret, 1996). It carinic ACh receptors impaired memory over short delays
is unclear what processes are being affected by AVP that if given directly after the encounter (Perio et al., 1989;
ultimately lead to improved memory. For example, many Soffie and Lamberty, 1988; Winslow and Camacho, 1995).
of the manipulations of the AVP system that modulate
juvenile recognition also modulate fear and anxiety-like c. Genetic Manipulations. Unlike rats, group-
behaviors in nonsocial tests as well as aggressive behavior housed mice, but not isolated mice, are able to recognize
(Everts and Koolhaas, 1999; Koolhaas et al., 1990, 1998; juveniles after delays of 1–7 days. This long-term memory
Landgraf et al., 1995; Liebsch et al., 1996). However, evi- appears to involve protein synthesis as injections with
Olfactory Memory 417

protein synthesis inhibitors impair recognition over long (Johnston and Robinson, 1993; Schellinck et al., 1995) and
delays (Kogan et al., 2000). One particular protein, cAMP- between artificial odors, although the magnitude of olfac-
responsive element-binding protein (CREB), previously tory investigation is reduced (Hunter and Murray, 1989).
implicated in the cellular cascade surrounding hippocam-
pal-dependent, nonsocial memory (Silva et al., 1998), 2. Lesions
appears to be involved in juvenile recognition as mutant
The physiological underpinnings of social odor recogni-
mice with a specific knock-out of the gene that produces
tion have not been well characterized. However, initial evi-
CREB are unable to show long-term memory for individ-
dence indicates that social odor recognition shares neural
ual juveniles (Kogan et al., 2000).
substrates with other forms of recognition, such as those
critical to DNMTS. Recognition of individual odors in
3. Lesions
female hamsters depends on the integrity of the olfactory,
Although it is clear that the septum is involved in juvenile rather than vomeronasal, system (Petrulis et al., 1999a),
recognition, the role of the hippocampus has been more and requires processing by the posterior parts of the main
difficult to evaluate. Indirect manipulations that damage olfactory bulb projection zone (Petrulis et al., 1999b).
the hippocampus such as ischemia and perforant path Hamsters with lesions of OFC or medial amygdala
transections impair recognition of familiar juveniles at (Petrulis and Johnston, 1999) showed no deficits in either
30-minute delays (Andersen and Sams-Dodd, 1997; habituation or discrimination (Petrulis et al., 1998),
Lemaire et al., 1994a). Moreover, transection of FNX, whereas lesions of the parahippocampal region selectively
but not lesions of the BLA, resulted in overall reduced impaired recognition of a novel individual’s odor with lit-
investigation of juveniles and an apparent deficit in tle effect on recognition of the familiar odor (Petrulis et al.,
recognition, although damage to other limbic system 2000). FNX transaction (Petrulis et al., 2000) and selective
pathways makes interpretation of hippocampal involve- lesions of the hippocampus (Petrulis and Eichenbaurn,
ment problematic (Maaswinkel et al., 1996). Selective 2000) do not eliminate individual odor recognition in ham-
lesions of the hippocampus in mice impair recognition sters. Similar results have been observed in rats: lesions of
after a 30-minute delay, but not if recognition was tested the hippocampus or septum do not impair recognition of
immediately after the first exposure, implicating the hip- odors, whereas animals with parahippocampal region dam-
pocampus in this type of long-term recognition memory age appear to have deficits in recognizing novel urine odors
(Kogan et al., 2000). (Hunter and Murray, 1989; A. Petrulis and A. Armenakis,
unpublished observations). Surprisingly, lesions of ENT in
rats actually facilitate the memory for familiar, artificial
C. Habituation/Discrimination
odors, in that lesions allow rats to recognize familiar odors
1. Behavior over longer delays than normal rats (Wirth et al., 1998).
This result is difficult to reconcile with either the impair-
Habituation paradigms have also been used to investigate
ments on the DNMTS task using artificial odors after
the kinds of information available in social odors and how
parahippocampal lesions (Otto and Eichenbaum, 1992a) or
these individualized odors are generated in several species
the deficits in recognition of social odors.
(e.g., Halpin, 1986; Johnston et al., 1993, 1994; Schellinck
et al., 1995). Although there are several variants on the task
3. Biochemical Substrates
(Gregg and Thiessen, 1981; Johnston, 1993; Sundberg
et al., 1982), in each animals are repeatedly presented with A limited set of pharmacological manipulations suggests
a particular odor from one individual and decrements in that the neurochemical substrates underlying habituation
sniffing directed at the odor occur over repeated presenta- and discrimination of odors are similar to those involved in
tions. On a subsequent test trial, the same odor from a other tests of olfactory-guided recognition memory. For
novel individual is presented either alone or opposed to the example, administration of scopolamine to rats, prior to
now-familiar odor, and increased or preferred sniffing of testing, impairs habituation to artificial odors as well as pre-
the novel odor is taken as a reflection of recognition of the venting increased investigation of novel odors (Hunter and
familiar one. Hamsters and guinea pigs demonstrate recog- Murray, 1989). Scopolamine is ineffective in blocking
nition of social odors over delays of several seconds to recognition if given after repeated exposures to the familiar
several weeks (Beauchamp and Wellington, 1984; odor but before presentation of the novel odor, suggesting
Johnston, 1993). This procedure taps into a general olfac- that the cholinergic system is necessary for encoding the
tory recognition process because animals are able to dis- odor rather than involved in retention or retrieval of the
criminate between individual odors of other species memory. Similarly, scopolamine injections in male mice
418 Petrulis and Eichenbaum

impaired habituation to repeated presentations of cues, suggesting that spatial cues are sufficient. However,
ovariectomized females, whereas blockade of acetyl- providing either visual or olfactory cues significantly
cholinesterase facilitated habituation (Winslow and increased correct food site identification. Olfactory cues
Camacho, 1995). These manipulations of the ACh system were more salient than visual markers when these two cues
may selectively affect the OB or PIR-ENT, as lesions of the were dissociated, suggesting that olfaction is the dominant
horizontal diagonal band, the major source of ACh in olfac- modality for local detection of objects. Moreover, rats can
tory areas, eliminate habituation to artificial odors (Paolini use both environmental and self-generated odor cues to
and McKenzie, 1993), whereas lesions of the medial sep- orient themselves in space if no illumination is available
tum, the main source of ACh to the hippocampus, have lit- (Lavenex and Schenk, 1996, 1998). Rats are also able to
tle effect (Hunter and Murray, 1989). associate odors sampled at a fixed location with rewards
As in other olfactory memory paradigms, the NE sys- delivered at places differentiated by visual, tactile and
tem may be involved in social odor recognition. Depletion positional properties (Youngentob et al., 1990, 1991). OFC
of NE in OB of rats did not greatly impair habituation to may be one area where these cross-modal representations
social odors, but after the first presentation of a novel are formed. Using a task in which rats were trained to
social odor, animals failed to show reinvestigation of sub- detect a unique odor at each of four locations, OFC neu-
sequent, novel urine odors (Guan et al., 1993). The lack of rons responded not only to specific odors or places, but
interest was not observed if animals were used as the also demonstrated odor-specific firing during arrival at the
familiar and novel stimuli, suggesting that, lacking a func- location where the odor was presented previously (Lipton
tioning olfactory system, recognition of individual animals et al., 1999).
may be achieved using nonolfactory cues.
B. Odor-Tactile Associations
4. Cellular Correlates
The few studies that have investigated physiological corre- Tomie and Whishaw developed a task in which rats were
lates of olfactory habituation have recorded from anes- required to pull strings to obtain attached food rewards
thetized animals and often use extremely long odor (Tomie and Whishaw, 1990). By using strings of different
exposures (Buonviso et al., 1998; Buonviso and Chaput, thickness and texture and painted with different odorants,
2000), and so have limited direct comparisons with behav- they showed that rats could be trained to use particular
ioral recognition. Nevertheless, recent studies by Wilson configurations of string texture and odor to access food
demonstrate that neurons in PIR show decrements in rewards (Tomie and Whishaw, 1990). Acquisition of these
response to repeated presentations of the same odor and configural discriminations, but not simple discriminations,
that this decrement occurs even though the OB is highly was impaired in animals with large hippocampal lesions
active (Wilson, 1998a). Intracellular recordings indicate (Whishaw and Tomie, 1991), although eventually other
that some of this reduction is due to reduced efficacy of brain systems could support the association. In particular,
synaptic potentials generated by OB afferent input to PIR large lesions of OFC, but not MDthal (Tomie and
(Wilson, 1998b). The habituation of PIR neurons appears Whishaw, 1996), permanently impaired acquisition and
specific to particular odors as presentation of highly similar retention of this task (Whishaw et al., 1992), suggesting an
or overlapping mixtures of odorants leads to cellular disha- alternative pathway.
bituation (Wilson, 2000). This pattern of results strongly
suggests that even highly similar odors are processed and
stored as separate representations in piriform cortex. C. Odor-Odor Associations
1. Formal Learning Tasks
IV. STIMULUS-STIMULUS ASSOCIATIONS
Several odor-odor association paradigms have been devel-
A. Odor-Place Associations oped to investigate the role of the hippocampus in “declar-
ative” memory in animals (Eichenbaum, 1997;
Several paradigms have been used to assess how animals Eichenbaum and Cohen, 2000). This kind of memory can
associate odors with places in the environment. Lavenex be studied in nonlinguistic species by characterizing it as
and Schenk (1995, 1998) trained rats where to find food memory for relationships between memories that can be
hidden at various open field locations flagged either by expressed “flexibly,” that is, in situations different than
odors or by visual cues or left unmarked. Latency to find repetition of the learning event (Cohen and Eichenbaum,
the food was unaffected by eliminating olfactory or visual 1993; Eichenbaum, 1997).
Olfactory Memory 419

In humans, paired-associate learning depends on hip- learn the palatability of different substances through social
pocampal function (Cohen and Eichenbaum, 1993). This interactions with conspecifics. The social transmission of
task involves presenting a list of two-word (object, etc.) diet preference has been studied experimentally in both
pairs, such as “hamster-dance, army-table,” and then, after domestic and wild rats and has resulted in fascinating
some delay, presenting them with one word from each pair insights into the information dynamics of animal social
and requiring recall or recognition of the second, associ- groups (Galef and Allen, 1995; Galef and White, 1997),
ated word. Using an olfactory variant of this task, Bunsey adaptive foraging (Galef, 1993), and how odor-odor asso-
and Eichenbaum (1993) demonstrated that rats can learn to ciations are formed naturally in animals (Galef and
discriminate assigned pairings of odors from “mispairs” of Wigmore, 1983). Three studies, published nearly simulta-
the same odors and that lesions of the PRC-ENT block this neously, reported that naïve rats (observers) show a sub-
capacity. In contrast, rats with selective lesions of the hip- stantial preference for the food that the demonstrator had
pocampus are able to form odor-odor associations (Bunsey eaten after an interaction with the demonstrator (Galef and
and Eichenbaum, 1996; Li et al., 1999). However, these Wigmore, 1983; Posadas-Andrews and Roper, 1983;
associations are highly inflexible and are bound to the cir- Strupp and Levitsky, 1984). Exposure to the diet alone
cumstances in which they are learned. Hippocampal rats does not lead to an increased preference (Galef et al.,
are unable to infer a transitive relationship between over- 1985), and the critical cues appear to be olfactory: (1) pref-
lapping odor pairs (e.g., A is associated with B and B is erences are not formed by anosmic animals or by animals
associated with C, therefore A is associated with C). And separated by a Plexiglas screen (Galef and Wigmore,
they do not demonstrate symmetry in their responses to 1983); (2) rats can show a preference after investigating
odor pairs (A is associated with B, therefore B is associ- the snout of anesthetized demonstrators (Galef and Stein,
ated with A). Taken together, these data suggest that the 1985); and (3) diet preferences can be induced by pairing
PRC-ENT, but not the hippocampus proper, is critical to constituents of rat breath (e.g., carbon disulfide) with a
form stimulus-stimulus associations and that the hip- particular diet odor (Galef et al., 1988). The memory can
pocampus is needed for a representation that allows flexi- last for 4 weeks, depending on the number of interactions
ble expression of these associations. with demonstrators (Galef, 1989; Galef and Whiskin,
Disconnection of the hippocampus, either by fornix tran- 1998; Galef and Wigmore, 1983), does not suffer from
section or PRC-ENT lesions, also results in severe impair- interference between successive demonstrators (Galef,
ment in flexible memory expression in another odor-guided 1983), is insensitive to the level of food deprivation, social
transitive inference task (Davis, 1992). In this study, rats familiarity, age, sex, and health status of observers and/or
were trained on a series of overlapping two-odor discrimi- demonstrators (Galef et al., 1984, 1991; Galef and Smith,
nations that could be organized to form a hierarchy (AB, 1994; Galef and Whiskin, 1998), and can reverse toxicosis-
BC, CD, DE, where “”
is selected over). If ani- induced diet aversions (Galef, 1985, 1986b) and prevent
mals form the hierarchical representation, they should be formation of new food aversions (Galef, 1986a, 1987).
able to judge between any two odors (especially BD), Socially acquired diet preferences can be formed even to
even though both of these odors had been equally rewarded foods that are relatively unpalatable, protein-deficient, and
during training. Whereas normal rats demonstrate this tran- require long handling times (Galef, 1986b; Galef and
sitive inference, rats with hippocampal system damage do Whiskin, 1995).
not, even though they learned the premise pairs and could
discriminate the odor that was always rewarded from the b. Lesions. This natural learning of associations
one that was never rewarded (AE) (Dusek and between odors has proved to be an attractive paradigm for
Eichenbaum, 1997). Lesions of the hippocampal system the study of the neural mechanisms underlying relational
also impair acquisition of an olfactory version of transverse memory. As with other forms of relational memory
patterning, a protocol that involves a circular organization of (Eichenbaum, 1997), rats with lesions of the hippocam-
odors (AB, BC, CA) (Dusek and Eichenbaum, 1998). pal formation or the parahippocampal region, but not
MDthal, were able to learn the association but were not
able to retain this information over 1–2 days (Alvarez
2. Social Transmission of Food Preferences
et al., 2001; Bunsey and Eichenbaum, 1995; Winocur,
a. Behavior. One of the most critical decisions that 1990). One recent study did not replicate this observation
animals have to make is which foods to eat and which but had to use four times the normal odor concentrations
foods to avoid. Although animals have ingestive mecha- to obtain learning by normal rats, suggesting that a dif-
nisms that reduce the risks of consuming toxic substances ferent kind of learning may have guided the observed
(Palmerino et al., 1980), social living animals can also preferences (Burton et al., 2000). Rats with hippocampal
420 Petrulis and Eichenbaum

damage also showed amnesia for odor preferences distress. Thereafter, rats avoid drinking water in the pres-
learned a few days prior to surgery, indicating that the ence of the taste or the odor alone. If presented with only
hippocampus is involved in consolidation of this associa- an odor and then poisoned after delays of 15 minutes or
tion (Winocur, 1990). longer, rats do not show aversion to that odor, whereas a rat
presented with only a tastant will show robust avoidance of
c. Biochemical Substrates. New molecular techniques that taste even at delays of several hours (Durlach and
have provided further support for the critical role of the Rescorla, 1980; Palmerino et al., 1980). Rats can show
hippocampus in this form of olfactory memory. Selective odor-toxicosis learning but only if the interval between
deletion of a subunit of the glutamatergic NMDA receptor, consumption of the compound stimulus and toxicosis is
a key component of synaptic plasticity (Morris and Davis, sufficiently short (Durlach and Rescorla, 1980; Pain and
1994), in the CA1 region of the mouse hippocampus pro- Booth, 1968; Palmerino et al., 1980). Consequently, the
duced substantial deficits in long-term retention of potentiation of odor aversion by taste refers to the ability
socially-induced diet preference (Rampon et al., 2000). of taste in compound with odor to extend the duration of
Manipulations of other biochemical systems involved in the sensory trace until it can be associated with toxicosis.
synaptic plasticity and learning have also lead to deficits in The literature surrounding the psychological nature of this
retention of diet preferences. Mutant mice lacking several association has been contentious and filled with contradic-
forms of CREB, a substance previously implicated in tory reports (e.g., Bouton and Whiting, 1982; Droungas
synaptic plasticity and other forms of memory (Silva et al., and LoLordo, 1991; Durlach and Rescorla, 1980; Holder
1998), demonstrated impairments in retaining the diet et al., 1987; Miller et al., 1986a; Palmerino et al., 1980).
preference over 24 hours, but memory was normal on an However, it now seems clear that taste does not have privi-
immediate test (Kogan et al., 1997). leged access to learning about illness as previously postu-
In addition, several experiments have demonstrated a lated, but that odor, taste, or texture can support learning
role for AVP in both recall and retention of socially over long delays (Martin and Lawrence, 1979). In particu-
acquired diet preferences. Injections of AVP facilitated lar, strong odors in drinking solution that, by themselves,
recall when rats were tested at delays that result in no diet have no taste, potentiate aversion to tastes that are ineffec-
preference but showed impaired recall if tested on delays tive when conditioned alone (Darling and Slotnick, 1994;
that animals will normally show preferences over (Strupp Slotnick et al., 1997). In spite of the controversy, much is
et al., 1990). Thus, AVP modulates memory retrieval known about neural mechanisms underlying odor-taste
depending on how well animals remember the stimulus associations by using TPOA and related tasks.
associations and may play a role in consolidation (Popik
and Van Ree, 1993). 2. Amygdala System
Several lines of evidence point to the basolateral amygdala
D. Odor-Taste and Odor-Toxicosis Association (BLA) as a critical convergence site between olfactory and
gustatory information in TPOA. The BLA is one of the
1. Behavior
first regions that receive both olfactory and gustatory cues
Strictly speaking, any form of discrimination learning (Alheid et al., 1995). Temporary inactivation of the amyg-
in which odors are paired with water and food reinforcers dala (Bermudez-Rattoni et al., 1983; Ferry et al., 1995),
includes, in part, odor-taste associations. However, since selective lesions of the BLA (Bermudez-Rattoni et al.,
water and food reinforcers contain other types of stimuli, such 1986; Ferry et al., 1995; Hatfield et al., 1992), and
as texture and physiological reactions to ingestion, and not catecholaminergic depletions of the amygdala (Fernandez-
just gustatory stimuli, the precise association formed during Ruiz et al., 1993) all impair TPOA acquisition. Also, rats
odor discrimination learning is not clear. Consequently, in this with injections of NMDA antagonists into the BLA are
section we will be primarily concerned with the formation of impaired in TPOA acquisition but are able to express pre-
associations between odors and well-defined taste cues. viously learned TPOA (Hatfield and Gallagher, 1995). In
The vast majority of information about taste-odor asso- all cases, rats demonstrated aversion to the taste stimulus
ciations comes from studies of taste-potentiated odor aver- alone, indicating that these manipulations interfere with
sions (TPOA) and from variations on this paradigm. In this the association between odor and taste, rather than
task, rats are allowed to drink a solution adulterated with impairing associations with toxicosis. Conversely, stimula-
both an odorant and a taste stimulus or with an odor pre- tion of BLA after ingestion facilitates odor-toxicosis learn-
sented in direct proximity to the taste solution. After a vari- ing, allowing for longer delays between odor and
able delay, rats are administered LiCl to induce gastric poisoning, suggesting that the BLA is involved in
Olfactory Memory 421

prolonging the olfactory memory until it can be paired the odors were changed (Rolls et al., 1996a; Schoenbaum
with gastric distress (Ferry and DiScala, 1997). et al., 1999). Surprisingly, correlated firing between cells
Lastly, in a task in which rats need to discriminate in the OFC of rats increased after reversal, suggesting that
between one odor paired with an attractive taste stimulus the original odor-taste associations are still maintained
and another paired with an aversive taste, neurons in the within the network of OFC neurons (Schoenbaum et al.,
BLA anticipate the appearance of the positive or negative 2000). This pattern of results, when compared to the firing
taste following odor sampling, and this selectivity emerges characteristics of BLA neurons, has been interpreted as
prior to the animal’s discriminative behavior (Schoenbaum showing that OFC is responsible for accessing information
et al., 1998). In addition, when BLA neurons fire during about the significance of an odor from BLA and then link-
odor sampling, they appear to be primarily encoding the ing this representation with the appropriate behavioral out-
valence of the odor stimuli and not odor identity per se, as put (Schoenbaum et al., 1999).
these neurons quickly reverse their selectivity to an odor
when its reward value changes (Schoenbaum et al., 1999). 4. Hippocampal System
Collectively, these results strongly suggest that the BLA is
The hippocampal system also plays a role in odor-taste
important for making odor-taste and odor-toxicosis associ-
association learning. Lesions of the hippocampus prevent
ations and, by extension, is a region where odor valence,
the acquisition of both CTA and TPOA; this impairment is
based on taste or visceral information, is generated. The
not attributable to deficient neophobia observed after these
mechanism of this association is not yet known, but it is
lesions (Miller et al., 1986b). Blockade of ACh receptors
likely that it involves activation of NMDA receptors, a type
prior to aversion training enhanced the strength of TPOA,
of glutamate receptor previously implicated in other forms
whereas increasing ACh activity in the hippocampus
of associative memory (Morris and Davis, 1994).
impaired the acquisition of TPOA and odor-toxicosis asso-
ciations (Bermudez-Rattoni et al., 1987). Although hip-
3. Orbitofrontal Cortex System pocampal damage impairs TPOA, lesions of the ENT,
which is the source of cortical input to the hippocampus,
The OFC and insular cortex compose another prominent
facilitate odor-toxicosis learning in that ENT-lesioned ani-
region of convergence between odor and taste information
mals showed aversions even after long delays between
(Rolls, 1997). Early studies suggested that the OFC-insu-
odor and poisoning (Ferry et al., 1996). This facilitatory
lar cortex is involved in odor-taste association (Tanabe
effect requires interactions of ENT with the BLA because
et al., 1975a,b) and, more generally, in learning contin-
inactivation of BLA in rats with ENT lesions prevents
gencies between odor and reward (Eichenbaum et al.,
them from displaying odor-toxicosis learning with long
1980, 1983b). Using both conditioned taste aversion
delays between odor and poisoning (Ferry et al., 1999).
(CTA) and TPOA paradigms, Lasiter et al. (1985) defined
the anterior insular cortex as being the most likely site of
5. Olfactory Bulb and Cortex
odor-taste association as lesions within this area selec-
tively impaired TPOA. Lastly, using a task in which odors are paired in solution
More recently, evidence for the integration of odor and with either attractive or aversive tastants, Kay and Laurent
taste in the OFC-insular cortex has been provided by the (1999) demonstrated that mitral cell activity in the OB
observation that removal of one OB and the contralateral reflects the reward (taste) contingencies of odors. While
ventrolateral OFC impairs the ability of rats to distinguish modulation of OB activity by nonolfactory factors, such as
an olfactory-taste compound from its component features hunger, satiety, or malaise, have been observed previously
(Schul et al., 1996). Physiological evidence also suggests (Chaput and Holley, 1976; Pager et al., 1972; Pager and
an intimate relationship between odors and taste in the Royet, 1976), this report showed that the strongest predic-
OFC-insular cortex of rhesus monkeys and rats. During tor of mitral cell firing was not odor identity but odor
performance of discrimination tasks that demand the for- valence and task-relevant behaviors, such as licking for
mation of associations between odors and attractive or water reward. For example, apparent odor-selective neu-
aversive tastes, different populations of OFC neurons rons changed their firing patterns when the taste with
coded odor identity and odor valence (taste) in both rats which it is paired was changed and some mitral cells even
and monkeys (Critchley and Rolls, 1996; Rolls et al., fired in anticipation of reinforcement (Kay and Laurent,
1996b; Schoenbaum et al., 1998, 1999). OFC neurons 1999), much like BLA and OFC neurons (Schoenbaum
showed selectivity in their firing during odor presentation et al., 1998, 1999). These results suggest that even the ear-
only after the animal had learned the task, and few cells liest stages of olfactory processing can be strongly influ-
reversed this selectivity after the reward contingencies for enced by prior experience and that odor memory may be
422 Petrulis and Eichenbaum

highly distributed between primary olfactory areas such as cues as animals preferred anesthetized novel animals or
the OB and PIR. This interpretation is in agreement with their odors over those of a prior mate (Carr et al., 1970,
current theoretical and empirical work in other neural sys- 1979, 1980; Huck et al., 1984; Johnston and Rasmussen,
tems showing that sensory and motor memory can be 1984). Furthermore, destruction of the olfactory epithe-
encoded by the same neurons that initially process this lium, but not vomeronasal organ removal, eliminates pref-
information (Fuster, 1995). erences of male hamsters for novel females (Johnston and
Rasmussen, 1984). Although the neural circuitry
underlying this form of olfactory recognition remains to be
V. SOCIAL BEHAVIOR AND ODOR LEARNING identified, it may involve increased dopamine release in
the nucleus accumbens (Fiorino et al., 1997) as well as
Recognition of individual identity can alter interactions processing by PRC-ENT, but does not require an intact
between animals in many contexts and is mediated, to a hippocampus (Petrulis and Eichenbaum, 2000).
great extent, by odors in several macrosmatic taxa (Brown Olfactory-based, long-term recognition of mates has
and Macdonald, 1985) (see Section III. C. and Chapters 15 been observed in pair-bonding species such as prairie voles
and 17). (Newman and Halpin, 1988) and Djungarian hamsters
(Vasilieva and Sokolov, 1994), as well as solitary species
A. Aggressive Behavior such as collared lemmings (Huck and Banks, 1979). Much
work in prairie voles has shown that pair-bonding develops
Familiarity can either increase (Corridi et al., 1993) or either by long-term cohabitation or, more usually, follow-
decrease agonistic behavior (Daly, 1977; Halpin, 1976; ing mating and depends on increased activity of the vaso-
Kimelman and Lubow, 1974). Conversely, aggressive pressin and oxytocin systems in the brain (Carter et al.,
interactions can alter attraction and/or preference for odors 1995; Young et al., 1998). Recently, a role for dopamine in
of familiar animals. For example, following agonistic pair formation has been described following the observa-
interactions, male guinea pigs spend more time near the tion that dopamine antagonists given before or immedi-
odors of animals that they had defeated than near the odors ately after mating impair pair-bonding (Wang et al., 1999).
of the male that had dominated them (Martin and This effect of dopamine (DA) appears to be related to con-
Beauchamp, 1982; see also Nyby et al., 1970). Similarly, solidation and not sensorimotor processing, as antagonist
rats that win agonistic encounters increase scent marking given 24 hours after mating did not impair mate recogni-
and olfactory investigation of odors from defeated males, tion. Dopamine may be acting through the nucleus accum-
more so than to unfamiliar animals (Brown, 1992). bens (NAcc) as injections of dopamine D-2 antagonists
Moreover, the odors of familiar dominant rats potentiate into this region impaired pair-bonding whereas D-2 ago-
freezing responses of defeated animals during presentation nists facilitated pair formation (Gingrich et al., 2000).
of shock, indicating that social odors can be readily condi- Evidence from both lesion and c-fos activation experi-
tioned by the aversive experience of defeat (Williams and ments have implicated other neural structures, such as the
Scott, 1989; Williams et al., 1990). Although little is cortico-medial amygdala and BLA, in the pair-bonding
known about neural mechanisms underlying learning process (Demas et al., 1997; Kirkpatrick et al., 1994a;
about odors within agonistic circumstances, the medial Wang et al., 1997). Although formation of pair-bonds also
amygdala (MeA) may play a critical role in this process. requires a functioning vomeronasal/olfactory system
For example, MeA lesions in subordinate rats attenuated (Kirkpatrick et al., 1994b; Williams et al., 1992), there is
their defeat-induced behaviors, such as the reduction of little direct evidence to link changes in oxytocin, vaso-
olfactory investigation of dominant animals (Bolhuis et al., pressin, or other neurochemical systems to the olfactory
1984; Luiten et al., 1985). memory underlying long-term attachment.

B. Mate Recognition and Pair-Bonding C. Sexual Behavior

Olfactory memory is also important for mate recognition Sexual behavior in animals is often characterized as a
in both polygamous and monogamous pair-bonding stereotyped and rigid process that is impermeable to learn-
species. In species that do not form long-term social ing. However, the available evidence suggests that adaptive
attachments, such as rats and hamsters, both males and expression of courtship and copulatory behavior may
females increase preference for novel mates shortly fol- require the formation of associations between the odors of
lowing copulation with another conspecific. This conspecifics and copulatory stimuli. For example, male
“Coolidge effect” appears to be dependent on olfactory mice produce ultrasonic vocalizations (USV) as part of
Olfactory Memory 423

their precopulatory behavior and when they encounter uri- olfactory bulbs (OB). For example, birth or VGS leads to a
nary cues produced by females (Nyby and Whitney, 1980). release of centrifugal pathway neurotransmitters such as
If males are not allowed to interact with females, the NE and ACh, as well as stimulating release of intrinsic
number of USVs in response to female urine declines over neurotransmitters, such as GABA and glutamate
time; this decrease is reversed by brief exposures to (Kendrick et al., 1988a,b).
females (Dizinno et al., 1978; Nyby and Whitney, 1980). The behavioral plasticity that underlies recognition of
The maintenance of male USV depends on the formation offspring is correlated with changes in the responsiveness
of associations between odor cues emitted by the female of OB mitral cells to lamb odors. Mitral cells normally do
and some aspect of their interaction with them. Males not respond to lamb odors prior to parturition or after the
interacting with artificially perfumed females will later close of the sensitive period for bonding, but they do show
produce USVs in response to the perfume when presented increased activity to lamb odors during the time when
alone (Kerchner et al., 1986; Nyby et al., 1978). Similarly, bonding would normally occur (Kendrick et al., 1992).
the hormonal response of male rats to sexually receptive More specifically, parturition results in changes in the den-
females can be conditioned to neutral odors by pairing drodendritic reciprocal synapses between mitral cells and
these odors with female presentation (Graham and the inhibitory granule cells in the OB. Levels of both glu-
Desjardins, 1980). Moreover, pairing neutral odors with tamate (from mitral cells) and GABA (from granule cells)
access to receptive females also results in increased num- are increased by exposure to lamb odors, and proportion-
ber of ejaculations with females scented with this odor ally more GABA is released in response to glutamate post-
(Kippin et al., 1998). The nucleus accumbens, a region partum than before birth, suggesting that these synapses
long implicated in reward learning (Robbins and Everitt, are altered by VGS associated with the birthing process
1996), may be important for the association of odor with (Kendrick et al., 1992). Nitric oxide (NO), a retrograde
the rewarding stimuli of sociosexual interaction. In anes- messenger molecule implicated in synaptic plasticity, is
thetized rats, the presentation of odors previously paired also produced in the OB during birth and appears to be
with copulation elicited greater neuronal activity in this required for olfactory memory formation (Kendrick et al.,
area than unpaired odors or odors paired with unreceptive 1997). Inhibiting NO release in the OB during bonding
females (West et al., 1992). blocks the formation of a selective bond but does not
impair recall. NO is formed in the OB by glutamate acting
D. Maternal Behavior via the NMDA receptor and appears responsible for the
increased glutamate release from the mitral cells during
Olfactory learning is also important during maternal exposure to lamb odor since blocking NO impairs this
behavior (Fleming and Korsmit, 1997; Lee et al., 1999). increase (Kendrick et al., 1997).
One particularly good example of this is the formation of The release of extrinsic NE and ACh into the OB asso-
a selective bond between a recently parturient sheep and ciated with the period of selective bonding also appears
their offspring. After birth, the ewe recognizes her mobile to be a critical event for the acquisition of the olfactory
offspring and selectively feeds it while repulsing other memory. Lesioning the source of NE innervation of the
lambs. The formation of the bond is rapid and limited to a OB prevents ewes from forming selective bonds with her
short time after birth, such that an alien lamb can be fos- own lambs (Pissonnier et al., 1985). Similarly, blockade
tered successfully during this period and the ewe’s own of -adrenergic receptors in the OB during imprinting
lamb will be rejected if the ewe has not interacted with it significantly impaired formation of the olfactory memory
during this time (Poindron and Le Neindre, 1980). Altering (Levy et al., 1990). Manipulating the ACh system also
the olfactory cues originating from the lamb or removing has profound consequences for memory formation.
main olfactory input in the ewe impairs bond formation, Injections of scopolamine around parturition as well as
implicating the olfactory system in this imprinting phe- immediately after bond formation impaired the ability of
nomenon (Levy et al., 1994; Poindron and Le Neindre, ewes to bond with their own lamb, but did not impair
1980). The critical stimulus that precipitates maternal recall of a previously learned odor memory (Ferreira
behavior and bonding is the stimulation of the cervix and et al., 1999; Levy et al., 1997a). However, it is likely that
vaginal tract during delivery, as nonpregnant ewes can be NE and ACh release in the OB is also involved in recall-
induced to show selective bonding with lambs by artificial ing or processing lamb odors, as ACh and NE are also
vaginocervical stimulation (VGS) (Kendrick et al., 1991; released in response to lamb odor after bonding has
Keverne et al., 1983). The effect of VGS on olfactory already occurred (Kendrick et al., 1992). The precise role
imprinting appears to be due, in part, to its influence on of ACh and NE in OB dynamics during imprinting
both intrinsic and neuromodulatory systems of the main remains obscure but may function to increase signal effi-
424 Petrulis and Eichenbaum

cacy and/or to facilitate the storage of nonoverlapping 1978; Mair et al., 1980; Potter and Butters, 1980).
memory representations (Jiang et al., 1996; Linster and However, their deficit appears less selective than those
Hasselmo, 1997). observed in OFC patients, as several reports indicate
Although much is known about the pharmacology of altered olfactory thresholds, attentional deficits, and prob-
olfactory memory formation in ewes, comparatively less is lems with nonolfactory tasks in Korsakoff’s patients (Jones
known about what brain systems, other than the OB, are et al., 1978; Mair et al., 1986; Potter and Butters, 1980).
involved in selective bonding. Based on immediate early Deficits in odor identification and recognition memory
gene expression, several olfactory structures are activated are also observed after unilateral temporal lobe damage,
following parturition and exposure to lambs including PIR, with the greatest deficit seen after right side damage
ENT, OFC, and the dentate gyrus (DaCosta et al., 1997). (Jones-Gotman and Zatorre, 1988; Rausch and Seraf-
Clearly more research is needed to identify the critical etinides, 1975; Rausch et al., 1977). Severe deficits in odor
site(s) of plasticity responsible for the olfactory imprinting recognition were observed in patient H. M., who had bilat-
phenomenon in sheep. eral resection of the medial temporal lobe. H. M. per-
formed normally on odor detection or in odor intensity
discrimination. However, he was completely unable to
VI. HUMAN OLFACTORY MEMORY make same-different judgments using odor quality, match
previously sampled odors to current odors, or identify
Can the extensive body of knowledge on olfactory common objects by smell (Eichenbaum et al., 1983b). At
memory in experimental animals provide insight into the present, it is not clear what structures in the temporal lobe
neural basis of olfactory memory in humans? Before con- are critical for odor discrimination, but recent evidence
sidering this question, one must confront several prob- points to damage in the piriform cortex located along the
lems. First, the complex and unavoidable relationship anterior medial temporal lobe (Jones-Gotman et al., 1997).
between linguistic and olfactory processes may result in Results from functional imaging of human brain activ-
different patterns of activation between animals and ity during olfactory stimulation have been largely congru-
humans, making meaningful comparisons difficult (see ent with the neuropsychological literature by showing that
Serby and Chobor, 1992). Second, the sensory world of the OFC, ENT, insular and piriform cortex are all activated
humans, like many other Catarrhine primates, is domi- in response to odor stimulation (Kettenmann et al., 1997;
nated primarily by visual and auditory cues, and, not sur- Levy et al., 1997b; Savic et al., 2000; Sobel et al., 1998a,
prisingly, we devote much less brain space to olfaction 2000) (see Chapter 12). Odor exposure also activates struc-
than do most other mammals, including rodents. This tures more remotely connected to the olfactory system,
raises the possibility that the mechanisms and areas that such as the cerebellum and the cingulate cortex, with at
we study in animals are selective adaptations to macros- least some of this activation related to the control of sniff-
matic life and, thus, may differ in complexity and function ing behavior (Levy et al., 1997b; Savic et al., 2000; Sobel
from odor processing in microsmatic humans. Lastly, et al., 1998b).
observations on human olfactory memory are contradic- Several recent studies have explicitly assessed the brain
tory on even the most basic of issues and therefore lack areas involved in olfactory memory using functional imag-
the consensus needed for reasonable comparisons with ing. Royet et al. (1999) observed increased activation of
animal research (for reviews, see Herz and Engen, 1996; the right OFC, as well as in other frontal and cingulate
Richardson and Zucco, 1989; White, 1998). regions, during judgments of odor familiarity (Royet et al.,
Nonetheless, there are parallels between the neural 1999). Savic et al. (2000) observed that discrimination of
mechanisms underlying human and animal olfactory odor quality specifically activates the hippocampal forma-
memory. Neuropsychological evaluations of patients have tion, as well as the caudate, but these areas are not acti-
shown that damage to olfactory structures results in vated during recognition of an odor learned one hour
impairments on tests of odor discrimination and memory. previously. In contrast, during recognition large areas of
For example, damage to right OFC impairs odor-quality association cortex, including OFC and PIR, showed
discrimination, common odor identification, and odor increased activity. This widespread activation of temporal
recognition memory (Jones-Gotman and Zatorre, 1988, and parietal cortex may reflect the reactivation of repre-
1993; Potter and Butters, 1980). In most cases OFC lesions sentations of nonolfactory stimuli that invariably sur-
do not greatly affect odor detection thresholds or discrimi- rounded the initial encoding of odor stimuli. Finally, the
nation of nonolfactory stimuli. Patients with Korsakoff’s amygdala is activated primarily to unpleasant and
syndrome have severe damage to the MDthal and perform presumably highly arousing odors (Zald and Pardo, 1997).
quite poorly on odor discrimination tasks (Jones et al., Taken together, the observations from neuropsychological
Olfactory Memory 425

and brain imaging studies on humans is largely consistent Petrulis and Eichenbaum, 2000; Petrulis et al., 2000).
with the descriptions of the neuroanatomical substrates Neither of these structures appears to be the final site of
that underlie olfactory memory in animals. Across species, memory storage, a view consistent with the literature on
widespread areas of the old “rhinencephalon” participate the role of the hippocampal system in nonolfactory
in odor processing in characteristic patterns associated memory (Cohen and Eichenbaum, 1993).
with particular types of olfactory memory.
2. Orbitofrontal Cortex System
VII. CONCLUSIONS The OFC and the associated mediodorsal thalamus are
involved in learning the “rules” that must be applied in dif-
Although a clear and comprehensive synthesis of this field ferent learning tasks. This conceptualization incorporates
is not yet possible, several themes emerge from this the observation that animals with OFC or MDthal damage
review. First, olfactory memory is not a unitary phenome- have difficulty acquiring odor discrimination and discrim-
non. Instead, it a highly distributed process that involves ination reversal learning sets and learning the match/non-
the activation of a distinct set of pathways when particu- match rule in DNMTS (Eichenbaum et al., 1980, 1983a;
lar cognitive demands are placed on the animal. In corre- McBride and Slotnick, 1997). Also, OFC is involved in
spondence with this view, damage to different parts of representing associations between odors and other stimuli,
olfactory system and its projections selectively disrupts a particularly those with intrinsic valence, such as taste,
particular kind of odor memory, or shifts the strategies nociception, and visceral sensations, and the ability to alter
used in learning the task (Eichenbaum and Cohen, 2000). behavior associated with changes in stimulus-reinforcer
Second, although it has been known for some time that associations (Rolls, 2000). The OFC does not, however,
structures within the olfactory system are highly intercon- seem to be a “higher” olfactory center in the sense of sup-
nected, recent evidence has provided insights into the porting more complex discriminations independent of
functional consequences of reentrant connections within reward (Petrulis et al., 1998).
the olfactory system. Lastly, modeling and empirical stud-
ies of ACh and NE effects on the olfactory system are 3. Amygdala System
beginning to provide an understanding of neuromodula-
The amygdala is also involved in some aspects of olfac-
tory systems that regulate the dynamics of memory for-
tory memory formation. Neurons in the basolateral amyg-
mation. In particular, research on the neurochemistry
dala are primarily active during the initial stages of
underlying juvenile odor recognition has provided hith-
odor-reward learning (Hess et al., 1997; Schoenbaum
erto unsuspected insights into the complex nature of con-
et al., 1998, 1999, 2000). Paradoxically, lesions of this
solidation of olfactory information. We consider each of
area have no effect on the acquisition of odor-reward asso-
these topics in more detail below.
ciations (Eichenbaum et al., 1986; Slotnick, 1985). This
discrepancy is probably due to differences in methodol-
A. Multiple Memory Systems ogy and on the strength and nature of the reinforcers used
1. Hippocampal System between studies and remains to be fully elucidated.
However, studies of odor-guided fear conditioning sug-
The observations from both lesion and electrophysiologi- gest requisite amygdala involvement only when there is
cal studies indicate that different aspects of olfactory high motivational content (Otto et al., 2000) or a highly
memory require or engage distinct brain pathways depend- arousing context (Cahill and McGaugh, 1990).
ing on the required cognitive operations. The hippocampus
is involved in situations that require animals to rapidly
4. Olfactory Bulb and Cortex
learn the relationships between odor stimuli, or between
odors and other cues, and then express this knowledge in Several models of the piriform cortex and olfactory bulb
situations different from those of acquisition (Eichenbaum, suggest that the connectional architecture of these areas
1997; Eichenbaum et al., 1994, 1999). In contrast, the hip- is sufficient to store self-organized odor memories
pocampus appears not to be critical for incremental learn- (Ambros-Ingerson et al., 1990; Haberly and Bower,
ing of odor valences or for recognition of particular odors 1989; Hasselmo et al., 1990). Neural plasticity associated
as familiar or unfamiliar (Eichenbaum et al., 1986; Petrulis with learning or learning-like contingencies has been
and Eichenbaum, 2000; Petrulis et al., 2000). Instead, the demonstrated in PIR (Kanter and Haberly, 1990; Roman
ability to retain memories of individual odors requires the et al., 1993a; Saar et al., 1999). Electrophysiological data
parahippocampal region (Otto and Eichenbaum, 1992a; are consistent with the idea that olfactory memories are
426 Petrulis and Eichenbaum

stored in a sparse and distributed manner within the piri- C. Neuromodulation

form cortex and the olfactory bulbs (Freeman, 1991;
McCollum et al., 1991; Schoenbaum and Eichenbaum, ACh and/or NE are critical for olfactory memory formation
1995a; Wilson, 2000). Unfortunately, this type of in a variety of contexts, including social recognition (Perio
arrangement makes it difficult to interpret the effects of et al., 1989; Winslow and Camacho, 1995), delayed non-
PIR lesions on olfactory memory. Nevertheless, lesions match-to-sample (Ravel et al., 1994), and olfactory imprint-
of PIR can selectively impair odor memory formation ing (Levy et al., 1990). Detailed analysis of ACh and NE
without rendering the animal anosmic (Slotnick and action on the microcircuitry of the OB and the piriform cor-
Schoonover, 1992; Zhang et al., 1998). Ironically, we tex have revealed that these modulators operate by reducing
know much less about the role of PIR in olfactory behav- intrinsic activity and allowing new afferent input to
ior and memory than about other regions receiving sec- mitral/pyramidal cells (Hasselmo and Bower, 1993). This
ondary and tertiary olfactory projections. modulation appears necessary for reducing interference
Surprisingly, in some situations memory-related plas- between the stored odor representations and the acquisition
ticity can be observed in olfactory receptor neurons. For of new olfactory information when the neuronal ensembles
example, in mice with congenitally low sensitivity to par- encoding these odors overlap (Hasselmo, 1995; Linster and
ticular odorants, repeated exposure can increase respon- Hasselmo, 1997). Consequently, preventing ACh function
siveness of the main olfactory epithelium to these odors eliminates the ability of the network, and thence the animal,
(Wang et al., 1993). More dramatically, exposure to artifi- to discriminate between the odors. This kind of marriage
cial odors later used for homing increases sensitivity in the between modeling, micro-circuit analysis, and behavior is
olfactory receptor neurons of salmon (Dittman et al., 1997; critical for further understanding of olfactory memory and
Hasler et al, 1978; Nevitt et al., 1994), and this odor mem- could profitably be applied to understanding the role of
ory may be triggered by hormonal changes when salmon other neuromodulators, such as vasopressin, that have pro-
are learning about their natal stream (Hasler et al., 1978; found effects on consolidation of social memories.
Morin et al., 1989). This kind of receptor-based olfactory
learning is likely limited to very specific and time-delim-
REFERENCES
ited contexts in which errors in learning are not susceptible
to correction and have extreme fitness consequences.
Alheid, G. F., de Olmos, J. S., and Beltramino, C. A. (1995).
Nevertheless, these results suggest that behaviorally rele-
Amygdala and extended amygdala. In The Rat Nervous
vant plasticity may be widely distributed in olfactory cir- System, G. Paxinos (Ed.). Academic Press, San Diego,
cuits and even found at the very earliest stages of olfactory pp. 495–578.
processing. Allen, W. F. (1937). Olfactory and trigeminal conditioned
reflexes in dogs. Am. J. Physiol. 118:532–539.
Allen, W. F. (1938). Relationship of the conditioned olfactory-
B. Reentrant Loops in the Olfactory System fore-leg response to the motor centers of the brain. Am.
J.Physiol. 121:657–668.
Although the massive interconnections between olfactory Allen, W. F. (1940). Effect of ablating the frontal lobes, hip-
structures (Haberly and Price, 1978a,b), systemwide elec- pocampi, and occipito-parieto-temporal (excepting pyriform
trophysiological oscillations (Freeman, 1991), and the areas) lobes on positive and negative olfactory conditioned
modulation of olfactory structures by motivational state reflexes. Am. J. Physiol. 128:754–771.
(Pager, 1974, 1983) have been known for some time, it is Allen, W. F. (1941). Effect of ablating the pyriform-amygdaloid
only recently that we have come to appreciate the func- areas and hippocampi on positive and negative olfactory con-
tional consequences of this interconnectivity. This feed- ditioned reflexes and on conditioned olfactory differentiation.
back from higher structures is most apparent in OB and Am. J. Physiol. 132:81–92.
PIR neuronal activity in animals performing olfactory dis- Alvarez, P., Schoenbaum, G., and Eichenbaum, H. (1999).
Cortical plasticity continues after learning: dynamic represen-
crimination tasks. The major finding is that most neurons
tations of odors in the orbitofrontal cortex of the rat. Soc.
in these “primary sensory” areas are driven more by nonol-
Neurosci. Abs. 25:876.
factory task demands, such as responding to water reward, Alvarez, P., Lipton, P. A., Melrose, R., and Eichenbaum, H.
trial onset cues, etc., than by the sensory qualities of the (2001). Differential effects of damage within the hippocampal
odors (Di Prisco and Freeman, 1985; Kay and Laurent, region on memory for a natural, non-spatial odor-odor associ-
1999; Schoenbaum and Eichenbaum, 1995a). It seems ation. Learn. Memory 8:79–86.
likely that the sculpting of neural activity in these areas, in Ambros-Ingerson, J., Granger, R., and Lynch, G. (1990).
compliance with critical features of the task, is due to feed- Simulation of paleocortex performs hierarchical clustering.
back from central structures. Science 247:1344–1348.
Olfactory Memory 427

Andersen, M. B., SamsDodd, F. (1997). Transient cerebral Brown, T. S., Rosvold, H. E., and Mishkin, M. (1963). Olfactory
ischemia inhibits juvenile recognition in the mongolian gerbil. discrimination after temporal lobe lesions in monkeys.
Pharmacol. Biochem. Behav. 56:719–725. J. Comp. Physiol. Psychol. 56:190–195.
Beauchamp, G. K., and Wellington, J. L. (1984). Habituation to Bunsey, M., and Eichenbaum, H. (1993). Critical role of the
individual odors occurs following brief, widely-spaced pre- parahippocampal region for paired-associate learning in rats.
sentations. Physiol. Behav. 32:511–514. Behav. Neurosci. 107:740–747.
Bennett, M. H. (1968). The role of the anterior limb of the ante- Bunsey, M., and Eichenbaum, H. (1995). Selective damage to the
rior commissure in olfaction. Physiol. Behav. 3:507–515. hippocampal region blocks long-term retention of a natural
Bermudez-Rattoni, F., Rusiniak, K. W., and Garcia, J. (1983). and nonspatial stimulus-stimulus association. Hippocampus
Flavor-illness aversions: potentiation of odor by taste is dis- 5:546–556.
rupted by application of novocaine into amygdala. Behav. Bunsey, M., and Eichenbaum, H. (1996). Conservation of hip-
Neural Biol. 37:61–75. pocampal memory function in rats and humans. Nature
Bermudez-Rattoni, F., Grijalva, C. V., Kiefer, S. W., and Garcia, 379:255–257.
J. (1986). Flavor-illness aversions: the role of the amygdala in Buonviso, N., and Chaput, M. (2000). Olfactory experience
the acquisition of taste-potentiated odor aversions. Physiol. decreases responsiveness of the olfactory bulb in the adult rat.
Behav. 38:503–508. Neuroscience 95:325–332.
Bermudez-Rattoni, F., Coburn, K. L., Fernandez, J., Chavez, A. Buonviso, N., Gervais, R., Chalansonnet, M., and Chaput, M.
F., and Garcia, J. (1987). Potentiation of odor by taste and (1998). Short-lasting exposure to one odour decreases general
odor aversions in rats are regulated by cholinergic activity of reactivity in the olfactory bulb of adult rats. Eur. J. Neurosci.
dorsal hippocampus. Pharmacol. Biochem. Behav. 26: 10:2472–2475.
553–559. Burton, S., Murphy, D., Qureshi, U., Sutton, P., and O’Keefe,
Bluthe, R. M., and Dantzer, R. (1990). Social recognition does J. (2000). Combined lesions of the hippocampus and subicu-
not involve vasopressinergic neurotransmission in female rats. lum do not produce deficits in a nonspatial social olfactory
Brain Res. 535:301–304. memory task. J. Neurosci. 20:5468–5475.
Bluthe, R. M., and Dantzer, R. (1993). Role of the vomeronasal Cahill, L., and McGaugh, J. L. (1990). Amygdaloid complex
system in vasopressinergic modulation of social recognition lesions differentially affect retention of tasks using appeti-
in rats. Brain Res. 604:205–210. tive and aversive reinforcement. Behav. Neurosci. 104:
Bluthe, R. M., Gheusi, G., and Dantzer, R. (1993). Gonadal 532–543.
steroids influence the involvement of arginine vasopressin in Carlson, N. R., and Vallante, M. A. (1974). Enhanced cue func-
social recognition in mice. Psychoneuroendocrinology tion of olfactory stimulation in mice with septal lesions.
18:323–335. J. Comp. Physiol. Psychol. 87:237–248.
Bolhuis, J. J., Fitzgerald, R. E., Dijk, D. J., and Koolhaas, J. M. Carr, W. J., Krames, L., and Costanzo, D. J. (1970). Previous sex-
(1984). The corticomedial amygdala and learning in an ago- ual experience and olfactory preference for novel versus
nistic situation in the rat. Physiol. Behav. 32:575–579. original sex partners in rats. J. Comp. Physiol. Psychol.
Bouton, M. E., and Whiting, M. R. (1982). Simultaneous odor- 71:216–222.
taste and taste-taste compounds in poison-avoidance learning. Carr, W. J., DeMesquita-Wander, M., Sachs, S. R., and Maconi,
Learn. Motiv. 13:472–494. P. (1979). Responses of female rats to odors from familiar vs.
Bowers, J. M., and Alexander, B. K. (1967). Mice: individual novel males. Bull. Psychon. Soc. 14:118–120.
recognition by olfactory cues. Science 158:1208–1210. Carr, W. J., Hirsch, J. T., and Balazs, J. M. (1980). Responses of
Brennan, P. A., Schellinck, H. M., de la Riva, C., Kendrick, K. male rats to odors from familiar vs novel females. Behav.
M., and Keverne, E. B. (1998). Changes in neurotransmitter Neural Biol. 29:331–337.
release in the main olfactory bulb following an olfactory Carter, C. S., DeVries, A. C., and Getz, L. L. (1995).
conditioning procedure in mice. Neuroscience 87: 583–590. Physiological substrates of mammalian monogamy: the
Bressler, S. L. (1988). Changes in electrical activity of rabbit prairie vole model. Neurosci. Biobehav. Rev. 19:303–314.
olfactory bulb and cortex to conditioned odor stimulation. Chaillan, F. A., Roman, F. S., and Soumireu-Mourat, B. (1996).
Behav. Neurosci. 102:740–747. Modulation of synaptic plasticity in the hippocampus and pir-
Brown, C. W., and Ghiselli, E. E. (1938). Subcortical mecha- iform cortex by physiologically meaningful olfactory cues in
nisms in learning. IV, Olfactory discrimination. J. Comp. an olfactory association task. J. Physiol. Paris 90:343–347.
Psychol. 26:109–120. Chaillan, F. A., Truchet, B., Roman, F. S., and Soumireu-Mourat,
Brown, R. E. (1992). Responses of dominant and subordinate B. (1999). Early polysynaptic potentiation recorded in the
male rats to the odors of male and female conspecifics. dentate gyrus during an associative learning task.
Aggress. Behav. 18:129–138. Neuroscience 94:443–451.
Brown, R. E., and Macdonald, D. W. (Eds.). (1985). Social Odors Chaput, M., and Holley, A. (1976). Olfactory bulb responsiveness
in Mammals. Clarendon Press, Oxford. to food odour during stomach distension in the rat. Chem.
Brown, T. S. (1963). Olfactory and visual discrimination in the Senses Flav. 2:189–201.
monkey after selective lesions of the temporal lobe. J. Comp. Cohen, N. J., and Eichenbaum, H. (1993). Memory, Amnesia, and
Physiol. Psychol. 56:764–768. the Hippocampal System. MIT Press, Cambridge, MA.
428 Petrulis and Eichenbaum

Corridi, P., Chiarotti, F., Bigi, S., and Alleva, E. (1993). tocin preservation of social recognition responses in rats.
Familiarity with conspecific odor and isolation-induced Neurosci. Lett. 254:161–164.
aggressive behavior in male mice (Mus domesticus). J. Comp. Dluzen, D. E., Muraoka, S., and Landgraf, R. (2000). Olfactory
Psychol. 107:328–335. bulb norepinephrine depletion abolishes vasopressin and oxy-
Cousens, G., and Otto, T. (1998). Both pre- and posttraining exci- tocin preservation of social recognition responses in rats. Eur.
totoxic lesions of the basolateral amygdala abolish the expres- J. Neurosci. 12:760–766.
sion of olfactory and contextual fear conditioning. Behav. Doty, R. L., Bagla, R., and Kim, N. (1999). Physostigmine
Neurosci. 112:1092–1103. enhances performance on an odor mixture discrimination test.
Critchley, H. D., and Rolls, E. T. (1996). Olfactory neuronal Physiol. Behav. 65:801–804.
responses in the primate orbitofrontal cortex: analysis in Droungas, A., and LoLordo, V. M. (1991). Taste-mediated poten-
an olfactory discrimination task. J. Neurophysiol. 75: tiation of odor aversion induced by lithium chloride: effects of
1659–1672. preconditioning exposure to the conditioned stimulus and
DaCosta, A. P. C., Broad, K. D., and Kendrick, K. M. (1997). postconditioning extinction of the taste aversion. Learn.
Olfactory memory and maternal behaviour-induced changes Motiv. 22:291–310.
in c-fos and zif/268 mRNA expression in the sheep brain. Dudchenko, P. A., Wood, E. R., and Eichenbaum, H. (2000).
Mol. Brain Res. 46:63–76. Neurotoxic hippocampal lesions have No effect on odor span
Daly, M. (1977). Some experimental tests of the functional sig- and little effect on odor recognition memory but produce sig-
nificance of scent-marking by gerbils (Meriones unguicula- nificant impairments on spatial span, recognition, and alterna-
tus). J. Comp. Physiol. Psychol. 91:1082–1094. tion. J. Neurosci. 20:2964–2977.
Dantzer, R., Bluthe, R. M., Koob, G. F., and Le Moal, M. (1987). Durlach, P. J., and Rescorla, R. A. (1980). Potentiation rather
Modulation of social memory in male rats by neurohypophy- than overshadowing in flavor-aversion learning: an analysis in
seal peptides. Psychopharmacology 91:363–368. terms of within-compound associations. J. Exp. Psychol.
Dantzer, R., Koob, G. F., Bluthe, R. M., and Le Moal, M. (1988). Anim. Behav. Proc. 6:175–187.
Septal vasopressin modulates social memory in male rats. Dusek, J. A., and Eichenbaum, H. (1997). The hippocampus and
Brain Res. 457:143–147. memory for orderly stimulus relations. Proc. Natl. Acad. Sci.
Dantzer, R., Tazi, A., and Bluthe, R. M. (1990). Cerebral lateral- USA 94:7109–7114.
ization of olfactory-mediated affective processes in rats. Dusek, J. A., and Eichenbaum, H. (1998). The hippocampus and
Behav. Brain Res. 40:53–60. transverse patterning guided by olfactory cues. Behav.
Darling, F. M., and Slotnick, B. M. (1994). Odor-cued taste Neurosci. 112:762–771.
avoidance: a simple and efficient method for assessing olfac- Eichenbaum, H. (1997). Declarative memory: Insights from cog-
tory detection, discrimination and memory in the rat. Physiol. nitive neurobiology. Ann. Rev. Psychol. 48:547–572.
Behav. 55:817–822. Eichenbaum, H., and Cohen, N. J. (2000). From Conditioning to
Davis, H. (1992). Transitive inference in rats (Rattus norvegicus). Conscious Recollection. Oxford Press, Oxford.
J. Comp. Psychol. 106:342–349. Eichenbaum, H., Shedlack, K. J., and Eckmann, K. W. (1980).
Davis, J. L., and Eichenbaum, H. (Eds.). (1991). Olfaction: A Thalamocortical mechanisms in odor-guided behavior. I.
Model System for Computational Neuroscience. MIT Press, Effects of lesions of the mediodorsal thalamic nucleus and
Cambridge, MA. frontal cortex on olfactory discrimination in the rat. Brain
De Rosa, E., and Hasselmo, M. E. (2000). Muscarinic choliner- Behav. Evol. 17:255–275.
gic neuromodulation reduces proactive interference between Eichenbaum, H., Clegg, R. A., and Feeley, A. (1983a).
stored odor memories during associative learning in rats. Reexamination of functional subdivisions of the rodent pre-
Behav. Neurosci. 114:32–41. frontal cortex. Exp. Neurol. 79:434–451.
Demas, G. E., Williams, J. M., and Nelson, R. J. (1997). Eichenbaum, H., Morton, T. H., Potter, H., and Corkin, S.
Amygdala but not hippocampal lesions impair olfactory (1983b). Selective olfactory deficits in case H. M. Brain 106:
memory for mate in prairie voles (Microtus ochrogaster). Am. 459–472.
J. Physiol. 273:R1683–R1689. Eichenbaum, H., Fagan, A., and Cohen, N. J. (1986). Normal
Di Prisco, G. V., and Freeman, W. J. (1985). Odor-related bulbar olfactory discrimination learning set and facilitation of rever-
EEG spatial pattern analysis during appetitive conditioning in sal learning after medial-temporal damage in rats: implica-
rabbits. Behav. Neurosci. 99:964–978. tions for an account of preserved learning abilities in amnesia.
Dittman, A. H., Quinn, T. P., Nevitt, G. A., Hacker, B., and Storm, J. Neurosci. 6:1876–1884.
D. R. (1997). Sensitization of olfactory guanylyl cyclase to a Eichenbaum, H., Kuperstein, M., Fagan, A., and Nagode,
specific imprinted odorant in coho salmon. Neuron 19: 381–389. J. (1987). Cue-sampling and goal-approach correlates of hip-
Dizinno, G., Whitney, G., and Nyby, J. (1978). Ultrasonic vocal- pocampal unit activity in rats performing an odor-discrimina-
izations by male mice (Mus musculus) to female sex tion task. J. Neurosci. 7:716–732.
pheromone: experiential determinants. Behav. Biol. Eichenbaum, H., Fagan, A., Mathews, P., and Cohen, N. J.
22:104–113. (1988). Hippocampal system dysfunction and odor discrimi-
Dluzen, D. E., Muraoka, S., and Landgraf, R. (1998). Olfactory nation learning in rats: impairment or facilitation depending
bulb norepinephrine depletion abolishes vasopressin and oxy- on representational demands. Behav. Neurosci. 102: 331–339.
Olfactory Memory 429

Eichenbaum, H., Mathews, P., and Cohen, N. J. (1989). Further Ferry, B., Wirth, S., and Di Scala, G. (1999). Functional interac-
studies of hippocampal representation during odor discrimi- tion between entorhinal cortex and basolateral amygdala dur-
nation learning. Behav. Neurosci. 103:1207–1216. ing trace conditioning of odor aversion in the rat. Behav.
Eichenbaum, H., Otto, T., and Cohen, N. J. (1994). Two func- Neurosci. 113:118–125.
tional components of the hippocampal memory system. Fiorino, D. F., Coury, A., and Phillips, A. G. (1997). Dynamic
Behav. Brain Sci. 17:449–518. changes in nucleus accumbens dopamine efflux during the
Eichenbaum, H., Dudchenko, P., Wood, E., Shapiro, M., and Coolidge effect in male rats. J. Neurosci. 17:4849–4855.
Tanila, H. (1999). The hippocampus, memory, and place cells: Fleming, A. S., and Korsmit, M. (1997). Plasticity in the mater-
is it spatial memory or a memory space? Neuron 23:209–226. nal circuit: effects of maternal experience on Fos-Lir in hypo-
Eichenbaum, H., Alvarez, P., and Ramus, S. J. (2000). Animal thalamic, limbic, and cortical structures in the postpartum rat.
models of amnesia. In Handbook of Neuropsychology: Horm. Behav. 31:145–158.
Memory Disorders, L. Cermak, (Ed.). Elsevier, Amsterdam, Freeman, W. J. (1991). Nonlinear dynamics in olfactory informa-
pp. 175–198. tion processing. In Olfaction: A Model System for
Engelmann, M., and Landgraf, R. (1994). Microdialysis adminis- Computational Neuroscience, J. L. Davis and H. Eichenbaum
tration of vasopressin into the septum improves social recog- (Eds.). MIT Press, Cambridge, MA, pp. 225–249.
nition in Brattleboro rats. Physiol. Behav. 55:145–149. Freeman, W. J., and Schneider, W. (1982). Changes in spatial pat-
Engelmann, M., and Landgraf, R. (1995). Microdialysis adminis- terns of rabbit olfactory EEG with conditioning to odors.
tration of vasopressin into the septum improves social recog- Psychophysiology 19:44–56.
nition in Brattleboro rats. J. Neuroendocrinol. 7:243–253. Fuster, J. M. (1989). The Prefrontal Cortex. Raven Press, New
Engelmann, M., Bures, J., and Landgraf, R. (1992). Vasopressin York.
administration via microdialysis into the septum interferes Fuster, J. M. (1995). Memory in the Cerebral Cortex. MIT Press,
with the acquisition of spatial memory in rats. Neurosci. Lett. Cambridge, MA.
142:69–72. Gaffan, D. (1974). Recognition impaired and association intact in
Engelmann, M., Ludwig, M., and Landgraf, R. (1994). the memory of monkeys after transection of the fornix.
Simultaneous monitoring of intracerebral release and behav- J. Comp. Physiol. Psychol. 86:1100–1109.
ior: endogenous vasopressin improves social recognition. Galef, B. G. (1983). Utilization by Norway rats (R. norvegicus)
J. Neuroendocrinol. 6:391–395. of multiple messages concerning distant foods. J. Comp.
Engelmann, M., Wotjak, C. T., Neumann, I., Ludwig, M., and Psychol. 97:364–371.
Landgraf, R. (1996). Behavioral consequences of intra- Galef, B. G. (1985). Socially induced diet preference can par-
cerebral vasopressin and oxytocin: Focus on learning and tially reverse a LiCl-induced diet aversion. Anim. Learn.
memory. Neurosci. Biobehav. Rev. 20:341–358. Behav. 13:415–418.
Engelmann, M., Ebner, K., Wotjak, C. T., and Landgraf, R. Galef, B. G., Jr. (1986a). Social identification of toxic diets
(1998). Endogenous oxytocin is involved in short-term olfac- by Norway rats (Rattus norvegicus). J. Comp. Psychol.
tory memory in female rats. Behav. Brain Res. 90:89–94. 100:331–334.
Everts, H. G., and Koolhaas, J. M. (1999). Differential modula- Galef, B. G., Jr. (1986b). Social interaction modifies learned
tion of lateral septal vasopressin receptor blockade in spatial aversions, sodium appetite, and both palatability and han-
learning, social recognition, and anxiety-related behaviors in dling-time induced dietary preference in rats (Rattus norvegi-
rats. Behav. Brain Res. 99:7–16. cus). J. Comp. Psychol. 100:432–439.
Fernandez-Ruiz, J., Miranda, M. L., Bermudez-Rattoni, F., and Galef, B. G. (1987). Social influences on the identification of
Drucker-Colin, R. (1993). Effects of catecholaminergic deple- toxic foods by Norway rats. Anim. Learn. Behav. 15:327–332.
tion of the amygdala and insular cortex on the potentiation of Galef, B. G., Jr. (1989). Enduring social enhancement of rats’
odor by taste aversions. Behav. Neural Biol. 60:189–191. preferences for the palatable and the piquant. Appetite
Ferreira, G., Gervais, R., Durkin, T. P., and Levy, F. (1999). 13:81–92.
Postacquisition scopolamine treatments reveal the time course Galef, B. G. J. (1993). Functions of social learning about food: a
for the formation of lamb odor recognition memory in par- causal analysis of effects of diet novelty on preference trans-
turient ewes. Behav. Neurosci. 113:136–142. mission. Anim. Behav. 46:257–265.
Ferry, B., and DiScala, G. (1997). Bicuculline administration into Galef, B. G., Jr., and Allen, C. (1995). A new model system for
basolateral amygdala facilitates trace conditioning of odor studying behavioural traditions in animals. Anim. Behav.
aversion in the rat. Neurobiol. Learn. Mem. 67:80–83. 50:705–717.
Ferry, B., Sandner, G., and Di Scala, G. (1995). Neuroanatomical Galef, B. G., Jr., and Smith, M. A. (1994). Susceptibility of arti-
and functional specificity of the basolateral amygdaloid ficially reared rat pups to social influences on food choice.
nucleus in taste-potentiated odor aversion. Neurobiol. Learn. Dev. Psychobiol. 27:85–92.
Mem. 64:169–180. Galef, B. G., and Stein, M. (1985). Demonstrator influence
Ferry, B., Oberling, P., Jarrard, L. E., and Di Scala, G. (1996). on observer diet preference: analysis of critical social
Facilitation of conditioned odor aversion by entorhinal cortex interactions and olfactory signals. Anim. Learn. Behav.
lesions in the rat. Behav. Neurosci. 110:443–450. 13:31–38.
430 Petrulis and Eichenbaum

Galef, B. G., Jr., and Whiskin, E. E. (1995). Are socially acquired Guan, X., Blank, J., and Dluzen, D. (1993). Depletion of olfactory
behaviours irreversible? Behav. Process. 34:279–284. bulb norepinephrine by 6-OHDA disrupts chemical cue but not
Galef, B. G., Jr., and Whiskin, E. E. (1998). Determinants of the social recognition responses in male rats. Brain Res. 622: 51–57.
longevity of socially learned food preferences of Norway rats. Haberly, L. B. (1985). Neuronal circuitry in olfactory cortex:
Anim. Behav. 55:967–975. anatomy and functional implications. Chem. Senses
Galef, B. G., Jr., and White, D. J. (1997). Socially acquired infor- 10:219–238.
mation reduces Norway rats’ latencies to find food. Anim. Haberly, L. B., and Bower, J. M. (1989). Olfactory cortex: model
Behav. 54:705–714. circuit for study of associative memory? Trends Neurosci.
Galef, B. G., and Wigmore, S. W. (1983). Transfer of information 12:258–264.
concerning distant foods: a laboratory investigation of the Haberly, L. B., and Price, J. L. (1978a). Association and com-
‘information-centre’ hypothesis. Anim. Behav. 31:748–758. missural fiber systems of the olfactory cortex of the rat I.
Galef, B. G., Kennett, D. J., and Wigmore, S. W. (1984). Transfer Systems origination in the piriform cortex and adjacent areas.
of information concerning distant food in rats: a robust phe- J. Comp. Neurol. 178:711–740.
nomenon. Anim. Learn. Behav. 12:292–296. Haberly, L. B., and Price, J. L. (1978b). Association and com-
Galef, B. G., Kennett, D. J., and Stein, M. (1985). Demonstrator missural fiber systems of the olfactory cortex of the rat. II.
influence on observer diet preference: effects of simple expo- Systems originating in the olfactory peduncle. J. Comp.
sure and the presence of a demonstrator. Anim. Learn. Behav. Neurol. 181:781–807.
13:25–30. Halpin, Z. T. (1976). The role of individual recognition by odors
Galef, B. G., Jr., Mason, J. R., Preti, G., and Bean, N. J. (1988). in the social interactions of the mongolian gerbil (Meriones
Carbon disulfide: a semiochemical mediating socially- unguiculatus). Behaviour 58:117–130.
induced diet choice in rats. Physiol. Behav. 42:119–124. Halpin, Z. T. (1986). Individual odors among mammals: origins
Galef, B. G., Beck, M., and Whiskin, E. E. (1991). Protein defi- and functions. Adv. Study Behav. 16:39–70.
ciency magnifies social influence on the food choices Hamrick, W. D., Wilson, D. A., and Sullivan, R. M. (1993).
of Norway rats (Rattus norvegicus). J. Comp. Psychol. Neural correlates of memory for odor detection conditioning
105:55–59. in adult rats. Neurosci. Lett. 163:36–40.
Gheusi, G., Bluthe, R. M., Goodall, G., and Dantzer, R. (1994). Hasler, A. D., Scholz, A. T., and Horrall, R. M. (1978). Olfactory
Ethological study of the effects of tetrahydroaminoacridine imprinting and homing in salmon. Am. Scientist 66:347–355.
(THA) on social recognition in rats. Psychopharmacology Hasselmo, M. E. (1995). Neuromodulation and cortical function:
(Berl) 114:644–650. modeling the physiological basis of behavior. Behav. Brain
Gheusi, G., Goodall, G., and Dantzer, R. (1997). Individually dis- Res. 67:1–27.
tinctive odours represent individual conspecifics in rats. Anim. Hasselmo, M. E. (1999). Neuromodulation: acetylcholine and
Behav. 53:935–944. memory consolidation. Trends Cog. Sci. 3:351–359.
Gingrich, B., Liu, Y., Cascio, C., Wang, Z., and Insel, T. R. Hasselmo, M. E., and Bower, J. M. (1993). Acetylcholine and
(2000). Dopamine D2 receptors in the nucleus accumbens are memory. Trends Neurosci. 16:218–222.
important for social attachment in female prairie voles Hasselmo, M. E., Wilson, M. A., Anderson, B. P., and Bower, J.
(Microtus ochrogaster). Behav. Neurosci. 114:173–183. M. (1990). Associative memory function in piriform (olfac-
Graham, J. M., and Desjardins, C. (1980). Classical conditioning: tory) cortex: computational modeling and neuropharmacol-
induction of luteinizing hormone and testosterone secretion in ogy. Cold Spring Harb. Symp. Quant. Biol. 55:599–610.
anticipation of sexual activity. Science 210:1039–1041. Hasselmo, M. E., Linster, C., Patil, M., Ma, D., and Cekic, M.
Grajski, K. A., and Freeman, W. J. (1989). Spatial EEG correlates (1997). Noradrenergic suppression of synaptic transmission
of nonassociative and associative olfactory learning in rabbits. may influence cortical signal-to-noise ratio. J. Neurophysiol.
Behav. Neurosci. 103:790–804. 77:3326–3339.
Gray, C. M., Freeman, W. J., and Skinner, J. E. (1986). Chemical Hatfield, T., and Gallagher, M. (1995). Taste-potentiated odor
dependencies of learning in the rabbit olfactory bulb: acquisi- conditioning: impairment produced by infusion of an N-
tion of the transient spatial pattern change depends on norepi- methyl-D-aspartate antagonist into basolateral amygdala.
nephrine. Behav. Neurosci. 100:585–596. Behav. Neurosci. 109:663–668.
Gregg, B., and Thiessen, D. D. (1981). A simple method of olfac- Hatfield, T., Graham, P. W., and Gallagher, M. (1992). Taste-
tory discrimination of urines for the Mongolian gerbil, potentiated odor aversion learning: role of the amygdaloid
Meriones unguiculatus. Physiol. Behav. 26:1133–1136. basolateral complex and central nucleus. Behav. Neurosci.
Griesbach, G. S., Hu, D., and Amsel, A. (1998). Effects of MK- 106:286–293.
801 on vicarious trial-and-error and reversal of olfactory dis- Herz, R. S., and Engen, T. (1996). Odor memory: review and
crimination learning in weanling rats. Behav. Brain Res. analysis. Psychonom. Bull. Rev. 3:300–313.
97:29–38. Herzog, C., and Otto, T. (1997). Odor-guided fear conditioning in
Griffin, M. G., and Taylor, G. T. (1995). Norepinephrine modula- rats: 2. Lesions of the anterior perirhinal cortex disrupt fear
tion of social memory: evidence for a time-dependent func- conditioned to the explicit conditioned stimulus but not to the
tional recovery of behavior. Behav. Neurosci. 109:466–473. training context. Behav. Neurosci. 111:1265–1272.
Olfactory Memory 431

Herzog, C., and Otto, T. (1998). Contributions of anterior perirhi- Jones-Gotman, M., and Zatorre, R. J. (1988). Olfactory identifi-
nal cortex to olfactory and contextual fear conditioning. cation deficits in patients with focal cerebral excision.
Neuroreport 9:1855–1859. Neuropsychologia 26:387–400.
Hess, U. S., Lynch, G., and Gall, C. M. (1995a). Changes in c-fos Jones-Gotman, M., and Zatorre, R. J. (1993). Odor recognition
mRNA expression in rat brain during odor discrimination memory in humans: role of right temporal and orbitofrontal
learning: differential involvement of hippocampal subfields regions. Brain Cog. 22:182–198.
CA1 and CA3. J. Neurosci. 15:4786–4795. Jones-Gotman, M., Zatorre, R. J., Cendes, F., Olivier, A.,
Hess, U. S., Lynch, G., and Gall, C. M. (1995b). Regional Andermann, F., McMackin, D., Staunton, H., Siegel, A. M.,
patterns of c-fos mRNA expression in rat hippocampus fol- and Wieser, H. G. (1997). Contribution of medial versus lat-
lowing exploration of a novel environment versus perform- eral temporal-lobe structures to human odour identification.
ance of a well-learned discrimination. J. Neurosci. 15: Brain 120:1845–1856.
7796–7809. Kanter, E. D., and Haberly, L. B. (1990). NMDA-dependent
Hess, U. S., Gall, C. M., Granger, R., and Lynch, G. (1997). induction of long-term potentiation in afferent and association
Differential patterns of c-fos mRNA expression in amygdala fiber systems of piriform cortex in vitro. Brain Res. 525:
during successive stages of odor discrimination learning. 175–179.
Learn Mem. 4:262–283. Kay, L. M., and Laurent, G. (1999). Odor- and context-dependent
Holder, M. D., Leon, M., Yirmiya, R., and Garcia, J. (1987). modulation of mitral cell activity in behaving rats. Nat.
Effect of taste preexposure on taste and odor aversions. Anim. Neurosci. 2:1003–1009.
Learn. Behav. 15:55–61. Kendrick, K. M., Keverne, E. B., Chapman, C., and Baldwin,
Huck, U. W., and Banks, E. M. (1979). Behavioral components of B. A. (1988a). Intracranial dialysis measurement of oxy-
individual recognition in the collared lemming (Dicrostonyx tocin, monoamine and uric acid release from the olfactory
groenlandicus). Behav. Ecol. Sociobiol. 6:85–90. bulb and substantia nigra of sheep during parturition, suck-
Huck, U. W., Banks, E. M., and Coopersmith, C. B. (1984). ling, separation from lambs and eating. Brain Res.
Social olfaction in male brown lemmings (Lemmus sibiricus 439:1–10.

trimucronatus) and collared lemmings (Dicrostonyx groen- Kendrick, K. M., Keverne, E. B., Chapman, C., and Baldwin, B.
landicus): II. Discrimination of mated and unmated females. A. (1988b). Microdialysis measurement of oxytocin, aspar-
J. Comp. Psychol. 98:60–65. tate, gamma-aminobutyric acid and glutamate release from
Hunter, A. J., and Murray, T. K. (1989). Cholinergic mechanisms the olfactory bulb of the sheep during vaginocervical stimula-
in a simple test of olfactory learning in the rat. tion. Brain Res. 442:171–174.
Psychopharmacology (Berl) 99:270–275. Kendrick, K. M., Levy, F., and Keverne, E. B. (1991). Importance
Jennings, J. W., and Keefer, L. H. (1969). Olfactory learning set of vaginocervical stimulation for the formation of maternal
in two varieties of domestic rat. Psychol. Rep. 24:3–15. bonding in primiparous and multiparous parturient owes.
Jiang, M. R., Griff, E. R., Ennis, M., Zimmer, L. A., and Shipley, Physiol. Behav. 50:595–600.
M. T. (1996). Activation of locus coeruleus enhances the Kendrick, K. M., Levy, F., and Keverne, E. B. (1992). Changes in
responses of olfactory bulb mitral cells to weak olfactory the sensory processing of olfactory signals induced by birth in
nerve input. J. Neurosci. 16:6319–6329. sleep. Science 256:833–836.
Johnston, R. E. (1993). Memory for individual scent in hamsters Kendrick, K. M., Guevara-Guzman, R., Zorrilla, J., Hinton, M.
(Mesocricetus auratus) as assessed by habituation methods. R., Broad, K. D., Mimmack, M., and Ohkura, S. (1997). for-
J. Comp. Psychol. 107:201–207. mation of olfactory memories mediated by nitric oxide.
Johnston, R. E., and Rasmussen, K. (1984). Individual recogni- Nature 388:670–674.
tion of female hamsters by males: role of chemical cues and Kerchner, M., Vatza, E. J., and Nyby, J. (1986). Ultrasonic vocal-
of the olfactory and vomeronasal systems. Physiol. Behav. izations by male house mice (Mus musculus) to novel odors:
33:95–104. roles of infant and adult experience. J. Comp. Psychol. 100:
Johnston, R. E., and Robinson, T. A. (1993). Cross-species dis- 253–261.
crimination of individual odors by hamsters (Muridae: Kettenmann, B., Hummel, C., Stefan, H., and Kobal, G. (1997).
Mesocricetus auratus, Phodopus campbelli), Ethology Multiple olfactory activity in the human neocortex identified
94:317–325. by magnetic source imaging. Chem. Senses 22:493–502.
Johnston, R. E., Derzie, A., Chiang, G., Jernigan, P., and Lee, Keverne, E. B., Levy, F., Poindron, P., and Lindsay, D. R. (1983).
H.-C. (1993). Individual scent signatures in golden hamsters: Vaginal stimulation: an important determinant of maternal
evidence for specialization of function. Anim. Behav. bonding in sheep. Science 219:81–83.
45:1061–1070. Kimble, D. P., and Zack, S. (1967). Olfactory discrimination in
Johnston, R. E., Chiang, G., and Tung, C. (1994). The informa- rats with hippocampal lesions. Psychon. Sci. 8:211–212.
tion in scent over-marks of golden hamsters. Anim. Behav. Kimelman, B. R., and Lubow, R. E. (1974). The inhibitory effect
48:323–330. of preexposed olfactory cues on intermale aggression in mice.
Jones, B. P., Butters, N., Moskowitz, H. R., and Montgomery, K. Physiol. Behav. 12:919–922.
(1978). Olfactory and gustatory capacities of alcoholic Kippin, T. E., Talianakis, S., Schattmann, L., Bartholomew, S.,
Korsakoff patients. Neuropsychologia 16:323–337. and Pfaus, J. G. (1998). Olfactory conditioning of sexual
432 Petrulis and Eichenbaum

behavior in the male rat (Rattus norvegicus). J. Comp. Lavenex, P., and Schenk, F. (1996). Integration of olfactory infor-
Psychol. 112:389–399. mation in a spatial representation enabling accurate arm
Kirkpatrick, B., Carter, C. S., Newman, S. W., and Insel, T. R. choice in the radial arm maze. Learn. Mem. 2:299–319.
(1994a). Axon-sparing lesions of the medial nucleus of the Lavenex, P., and Schenk, F. (1998). Olfactory traces and spatial
amygdala decrease affiliative behaviors in the prairie vole learning in rats. Anim. Behav. 56:1129–1136.
(Microtus ochrogaster): behavioral and anatomical speci- LeDoux, J. E. (1995). Emotion: clues from the brain. Ann. Rev.
ficity. Behav. Neurosci. 108:501–513. Psychol. 46:209–235.
Kirkpatrick, B., Williams, J. R., Slotnick, B. M., and Carter, C. Lee, A., Li, M., Watchus, J., and Fleming, A. S. (1999).
S. (1994b). Olfactory bulbectomy decreases social behavior Neuroanatomical basis of maternal memory in postpartum
in male prairie voles (M. ochrogaster). Physiol. Behav. 55: rats: selective role for the nucleus accumbens. Behav.
885–889. Neurosci. 113:523–538.
Kogan, J. H., Frankland, P. W., Blendy, J. A., Coblentz, J., Lemaire, M., Piot, O., Roques, B. P., Bohme, G. A., and
Marowitz, Z., Schutz, G., and Silva, A. J. (1997). Spaced Blanchard, J. C. (1992). Evidence for an endogenous chole-
training induces normal long-term memory in CREB mutant cystokininergic balance in social memory. Neuroreport
mice. Curr. Biol. 7:1–11. 3:929–932.
Kogan, J. H., Frankland, P. W., and Silva, A. J. (2000). Long-term Lemaire, M., Barneoud, P., Bohme, G. A., Piot, O., Haun, F.,
memory underlying hippocampus-dependent social recogni- Roques, B. P., and Blanchard, J. C. (1994a). CCK-A and
tion in mice. Hippocampus 10:47–56. CCK-B receptors enhance olfactory recognition via distinct
Koger, S. M., and Mair, R. G. (1994). Comparison of the effects neuronal pathways. Learn. Mem. 1:153–164.
of frontal cortical and thalamic lesions on measures of olfac- Lemaire, M., Bohme, G. A., Piot, O., Roques, B. P., and
tory learning and memory in the rat. Behav. Neurosci. Blanchard, J. C. (1994b). CCK-A and CCK-B selective recep-
108:1088–1100. tor agonists and antagonists modulate olfactory recognition in
Kolb, B. (1984). Functions of the frontal cortex of the rat: a com- male rats. Psychopharmacology (Berl) 115:435–440.
parative review. Brain Res. Rev 8:65–98. LeMoal, M., Dantzer, R., Michaud, B., and Koob, G. F. (1987).
Koolhaas, J. M., van den Brink, T. H. C., Roozendaal, B., and Centrally injected arginine vasopressin (AVP) facilitates
Boorsma, F. (1990). Medial amygdala and aggressive behav- social memory in rats. Neurosci. Lett. 77:353–359.
ior: interaction between testosterone and vasopressin. Levy, F., Gervais, R., Kindermann, U., Orgeur, P., and Piketty, V.
Aggress. Behav. 16:223–229. (1990). Importance of beta-noradrenergic receptors in the
Koolhaas, J. M., Everts, H., de Ruiter, A. J., de Boer, S. F., and olfactory bulb of sheep for recognition of lambs. Behav.
Bohus, B. (1998). Coping with stress in rats and mice: differ- Neurosci. 104:464–469.
ential peptidergic modulation of the amygdala-lateral septum Levy, F., Locatelli, A., Piketty, V., Tillet, Y., and Poindron, P.
complex. Prog. Brain Res. 119:437–448. (1994). Involvement of the main but not the accessory olfac-
Landgraf, R., Gertsberger, R., Montkowski, A., Probst, J. C., tory system in the maternal behavior of primiparous and mul-
Wotjak, C. T., Holsboer, F., and Engelmann, M. (1995). V1 tiparous ewes. Physiol. Behav. 57:97–104.
vasopressin receptor antisense oligodeoxynucleotide into sep- Levy, F., Richard, P., Meurisse, M., and Ravel, N. (1997a).
tum reduces vasopressin binding, social discrimination abili- Scopolamine impairs the ability of parturient ewes to learn to
ties, and anxiety-related behavior in rats. J. Neurosci. recognize their lambs. Psychopharmacology 129:85–90.
15:4250–4258. Levy, L. M., Henkin, R. I., Hutter, A., Lin, C. S., Martins, D., and
Larson, J., Lieu, T., Petchpradub, V., LeDuc, B., Ngo, H., Schellinger, D. (1997b). Functional MRI of human olfaction.
Rogers, G. A., and Lynch, G. (1995). Facilitation of olfactory J. Comput. Assist. Tomogr. 21:849–856.
learning by a modulator of AMPA receptors. J. Neurosci. Li, H., Matsumoto, K., and Watanabe, H. (1999). Different
15:8023–8030. effects of unilateral and bilateral hippocampal lesions in rats
Lashley, K. S., and Sperry, R. W. (1943). Olfactory discrimina- on the performance of radial maze and odor-paired associate
tion after destruction of the anterior thalamic nuclei. Am. tasks. Brain Res. Bull. 48:113–119.
J. Physiol. 139:446–450. Liebsch, G., Wotjak, C. T., Landgraf, R., and Engelmann, M.
Lasiter, P. S., Deems, D. A., and Garcia, J. (1985). Involvement (1996). Septal vasopressin modulates anxiety-related behav-
of the anterior insular gustatory neocortex in taste-potentiated iour in rats. Neurosci. Lett. 217:101–104.
odor aversion learning. Physiol. Behav. 34:71–77. Linster, C., and Hasselmo, M. (1997). Modulation of inhibition
Laska, M., and Hudson, R. (1993). Assessing olfactory perfor- in a model of olfactory bulb reduces overlap in the neural
mance in a New World primate, Saimiri sciureus. Physiol. representation of olfactory stimuli. Behav. Brain Res.
Behav. 53:89–95. 84:117–127.
Laska, M., Alicke, T., and Hudson, R. (1996). A study of long- Linster, C., and Hasselmo, M. E. (2000). Neural activity in the
term odor memory in squirrel monkeys (Saimiri sciureus). horizontal limb of the diagonal band of broca can be modu-
J. Comp. Psychol. 110:125–130. lated by electrical stimulation of the olfactory bulb and cortex
Lavenex, P., and Schenk, F. (1995). Influence of local environ- in rats. Neurosci. Lett. 282:157–160.
mental olfactory cues on place learning in rats. Physiol. Linster, C., Wyble, B. P., and Hasselmo, M. E. (1999). Electrical
Behav. 58:1059–1066. stimulation of the horizontal limb of the diagonal band of
Olfactory Memory 433

broca modulates population EPSPs in piriform cortex. tioned odor aversion, and conditioned taste aversion in rats
J. Neurophysiol. 81:2737–2742. with hippocampal lesions. Behav. Neural Biol. 45:240–253.
Lipton, P. A., Alvarez, P., and Eichenbaum, H. (1999). Miserendino, M. J. D., Sanannes, C. B., Melia, K. R., and Davis,
Crossmodal associative memory representations in rodent M. (1990). Blocking of acquisition but not expression of con-
orbitofrontal cortex. Neuron 22:349–359. ditioned fear-potentiated startle by NMDA antagonists in the
Long, C. J., and Tapp, J. T. (1970). Significance of olfactory amygdala. Nature 345:716–718.
tracts in mediating response to odors in the rat. J. Comp. Mishkin, M. (1978). Memory in monkeys severely impaired by
Physiol. Psychol. 72:435–443. combined but not by separate removal of amygdala and hip-
Lovelace, C. T., and Slotnick, B. M. (1995). Memory for brief, pocampus. Nature 273:297–298.
widely spaced odor presentations in the rat. Chem. Senses Morin, P.-P., Dodson, J. J., and Dore, F. Y. (1989). Thyroid activ-
20:183–190. ity concomitant with olfactory learning and heart rate changes
Lu, X. C., and Slotnick, B. M. (1990). Acquisition of an olfactory in Atlantic salmon, Salmo salar, during smoltification. Can.
learning-set in rats with lesions of the mediodorsal thalamic J. Fish. Aquat. Sci. 46:131–136.
nucleus. Chem. Senses 15:713–724. Morris, R. G. M., Anderson, E., Lynch, G., and Baudry, M.
Lu, K. C., Slotnick, B. M., and Silberberg, A. M. (1993). Odor (1986). Selective impairment of learning and blockade of
matching and odor memory in the rat. Physiol. Behav. long-term potentiation by an N-methyl-D-aspartate receptor
53:795–804. antagonist, AP5. Nature 319:774–776.
Luiten, P. G., Koolhaas, J. M., de Boer, S., and Koopmans, S. J. Morris, R. G. M., and Davis, M. (1994). The role of NMDA
(1985). The cortico-medial amygdala in the central nervous receptors in learning and memory. In The NMDA Receptor, J.
system organization of agonistic behavior. Bran Res. C. Watkins and G. L. Collingridge (Eds.). Oxford Press,
332:283–297. Oxford, pp. 52–76.
Maaswinkel, H., Baars, A. M., Gispen, W. H., and Spruiji, B. M. Mouly, A. M., and Holley, A. (1986). Perceptive properties of the
(1996). Roles of the basolateral amygdala and hippocampus multi-site electrical microstimulation of the olfactory bulb in
in social recognition in rats. Physiol. Behav. 60:55–63. the rat. Behav. Brain Res. 21:1–12.
Macrides, F., Eichenbaum, H. B., and Forbes, W. B. (1982). Mouly, A. M., Vigouroux, M., and Holley, A. (1985). On the abil-
Temporal relationship between sniffing and the limbic theta ity of rats to discriminate between microstimulations of the
rhythm during odor discrimination reversal learning. olfactory bulb in different locations. Behav. Brain Res.
J. Neurosci. 2:1705–1717. 17:45–58.
Mair, R. G., Capra, C., McEntee, W. J., and Engen, T. (1980). Mouly, A. M., Gervais, R., and Holley, A. (1990). Evidence for
Odor discrimination and memory in Korsakoffs psychosis. the involvement of rat olfactory bulb in processes supporting
J. Exp. Psychol. Hum. Percept. Perform. 6: 445–458. long-term olfactory memory. Eur. J. Neurosci. 2:978–984.
Mair, R. G., Doty, R. L., Kelly, K. M., Wilson, C. S., Langlais, P. Mouly, A. M., Kindermann, U., Gervais, R., and Holley, A.
J., McEntee, W. J., and Vollmecke, T. A. (1986). Multimodal (1993). Involvement of the olfactory bulb in consolidation
sensory discrimination deficits in Korsakoffs psychosis. processes associated with long-term memory in rats. Behav.
Neuropsychologia 24:831–839. Neurosci. 107:451–457.
Mair, R. G., Burk, J. A., and Porter, M. C. (1998). Lesions of the Mumby, D. G., and Pinel, J. P. J. (1994). Rhinal cortex lesions
frontal cortex, hippocampus, and intralaminar thalamic nuclei and object recognition in rats. Behav. Neurosci. 108:11–18.
have distinct effects on remembering in rats. Behav. Neurosci. Mumby, D. G., Wood, E., and Pinel, J. P. J. (1992). Object-recog-
112:772–792. nition memory is only mildly impaired in rats with lesions of
Martin, I. G., and Beauchamp, G. K. (1982). Olfactory recogni- the hippocampus and amygdala. Psychobiology 20:18–27.
tion of individuals by male cavies (Cavia aperea). J. Chem. Murray, E. A., and Mishkin, M. (1998). Object recognition and
Ecol. 8:1241–1249. location memory in monkeys with excitotoxic lesions of the
Martin, L. T., and Lawrence, C. D. (1979). The importance of amygdala and hippocampus. J. Neurosci. 18:6568–6582.
odor and texture cues in food aversion learning. Behav. Nevitt, G. A., Dittman, A. H., Quinn, T. P., and Moody, W. J., Jr.
Neural Biol. 27:503–515. (1994). Evidence for a peripheral olfactory memory in
McBride, S. A., and Slotnick, B. (1997). The olfactory thalamo- imprinted salmon. Proc. Natl. Acad. Sci. USA 91:4288–4292.
cortical system and odor reversal learning examined using an Newman, K. S., and Halpin, Z. T. (1988). Individual odours and
asymmetrical lesion paradigm in rats. Behav. Neurosci. mate recognition in the prairie vole, Microtus ochrogaster.
111:1273–1284. Anim. Behav. 36:1779–1787.
McCollum, J., Larson, J., Otto, T., Schottler, F., Granger, R., and Nigrosh, B. J., Slotnick, B. M., and Nevin, J. A. (1975). Olfactory
Lynch, G. (1991). Short-latency single unit processing in discrimination, reversal learning, and stimulus control in rats.
olfactory cortex. J. Cog. Neurosci. 3:293–299. J. Comp. Physiol. Psychol. 89:285–294.
Miller, J. S., McCoy, D. F., Kelly, K. S., and Bardo, M. T. Nyby, J., and Whitney, G. (1980). Experience affects behav-
(1986a). A within-event analysis of taste-potentiated odor and ioral responses to sex odors. In Chemical Signals in
contextual aversions. Anim. Learn. Behav. 14:15–21. Vertebrates and Aquatic Invertebrates, D. Muller-Schwarze
Miller, J. S., Nonneman, A. J., Kelly, K. S., Neisewander, J. L., and R. M. Silverstein (Eds.). Plenum Press, New York,
and Isaac, W. L. (1986b). Disruption of neophobia, condi- pp. 173–192.
434 Petrulis and Eichenbaum

Nyby, J., Thiessen, D. D., and Wallace, P. (1970). Social inhibi- Perio, A., Terranova, J. P., Worms, P., Bluthe, R. M., Dantzer, R.,
tion of territorial marking in the mongolian gerbil. Psychon. and Biziere, K. (1989). Specific modulation of social memory
Sci. 21:310–312. in rats by cholinomimetic and nootropic drugs, by benzodi-
Nyby, J., Whitney, G., Schmitz, S., and Dizinno, G. (1978). azepine inverse agonists, but not by psychostimulants.
Postpubertal experience establishes signal value of mam- Psychopharmacology 97:262–268.
malian sex odor. Behav. Biol. 22:545–552. Petrulis, A., and Eichenbaum, H. (2000). The hippocampal sys-
O’Grady, R. S., and Jennings, J. W. (1974). Reversal learning set tem, individual odor discrimination and the Coolidge effect in
in successive discrimination reversal learning and transfer golden hamsters (Mesocricetus auratus). Soc. Neurosci. Abs.
with rats using visual and olfactory cues. Psychol. Rec. 26:468.
24:243–251. Petrulis, A., and Johnston, R. E. (1999). Lesions centered on the
Otto, T., and Eichenbaum, H. (1992a). Complementary roles of medial amygdala impair scent-marking and sex- odor recog-
the orbital prefrontal cortex and the perirhinal-entorhinal cor- nition but spare discrimination of individual odors in female
tices in an odor-guided delayed-nonmatching-to-sample task. golden hamsters. Behav. Neurosci. 113:345–357.
Behav. Neurosci. 106:762–775. Petrulis, A., DeSouza, I., Schiller, M., and Johnston, R. E. (1998).
Otto, T., and Eichenbaum, H. (1992b). Neuronal activity in the Role of frontal cortex in social odor discrimination and scent-
hippocampus during delayed non-match to sample perfor- marking in female golden hamsters (Mesocricetus auratus).
mance in rats: evidence for hippocampal processing in recog- Behav. Neurosci. 112:199–212.
nition memory. Hippocampus 2:323–334. Petrulis, A., Peng, M., and Johnston, R. E. (1999a). Effects of
Otto, T., and Garruto, D. (1997). Rhinal cortex lesions impair vomeronasal organ removal on individual odor discrimina-
simultaneous olfactory discrimination learning in rats. Behav. tion, sex-odor preference, and scent marking by female ham-
Neurosci. 111:1146–1150. sters. Physiol. Behav. 66:73–83.
Otto, T., Eichenbaum, H., Wiener, S. I., and Wible, C. G. (1991a). Petrulis, A., Peng, M., and Johnston, R. E. (1999b). Lateral olfac-
Learning-related patterns of CA1 spike trains parallel stimu- tory tract transections impair discrimination of individual
lation parameters optimal for inducing hippocampal long- odors, sex-odor preferences, and scent marking by female
term potentiation. Hippocampus 1:181–192. hamsters (Mesocricetus auratus). In Advances in Chemical
Otto, T., Schottler, F., Staubli, U., Eichenbaum, H., and Lynch, G. Communication in Vertebrates, R. E. Johnston, D. Muller-
(1991b). Hippocampus and olfactory discrimination learning: Schwarze, and P. Sorensen (Eds.). Plenum Press, New York,
effects of entorhinal cortex lesions on olfactory learning and pp. 549–562.
memory in a successive-cue, go-no-go task. Behav. Neurosci. Petrulis, A., Peng, M., and Johnston, R. E. (2000). The role of the
105:111–119. hippocampal system in social odor discrimination and scent-
Otto, T., Cousens, G., and Rajewski, K. (1997). Odor-guided fear marking in female golden hamsters (Mesocricetus auratus).
conditioning in rats: 1. Acquisition, retention, and latent inhi- Behav. Neurosci. 114:184–195.
bition. Behav. Neurosci. 111:1257–1264. Pissonnier, D., Thiery, J. C., Fabre-Nys, C., Poindron, P., and
Otto, T., Cousens, G., and Herzog, C. (2000). Behavioral and Keverne, E. B. (1985). The importance of olfactory bulb nora-
neuropsychological foundations of olfactory fear condition- drenalin for maternal recognition in sheep. Physiol. Behav.
ing. Behav. Brain Res. 110:119–128. 35:361–363.
Pager, J. (1974). A selective modulation of the olfactory bulb Ploeger, G. E., Willemen, A. P., and Cools, A. R. (1991). Role of
electrical activity in relation to the learning of palatability in the nucleus accumbens in social memory in rats. Brain Res.
hungry and satiated rats. Physiol. Behav. 12:189–195. Bull. 26:23–27.
Pager, J. (1983). Unit responses changing with behavioral out- Poindron, P., and Le Neindre, P. (1980). Endocrine and sensory
come in the olfactory bulb of unrestrained rats. Brain Res. regulation of maternal behavior in the ewe. Adv. Study Behav.
289:87–98. 11:75–119.
Pager, J., and Royet, J. P. (1976). Some effects of conditioned Popik, P., and Van Ree, J. M. (1991). Oxytocin but not vaso-
aversion on food intake and olfactory bulb electrical pressin facilitates social recognition following injection into
responses in the rat. J. Comp. Physiol. Psychol. 90:67–77. the medial preoptic area of the rat brain. Eur.
Pager, J., Giachetti, I., Holley, A., and Le Magnen, J. (1972). A Neuropsychopharmacol. 1:555–560.
selective control of olfactory bulb electrical activity in relation Popik, P., and Van Ree, J. M. (1993). Social transmission of fla-
to food deprivation and satiety in rats. Physiol. Behav. vored tea preferences: facilitation by a vasopressin analog and
9:573–579. oxytocin. Behav. Neural Biol. 59:63–68.
Pain, J. F., and Booth, D. A. (1968). Toxiphobia for odors. Popik, P., Vetulani, J., Bisaga, A., and van Ree, J. M. (1991a).
Psychon. Sci. 10:363–364. Recognition cue in the rat’s social memory paradigm. J. Basic
Palmerino, C. C., Rusiniak, K. W., and Garcia, J. (1980). Flavor- Clin. Physiol. Pharmacol. 2:315–327.
illness aversions: the peculiar roles of odor and taste in Popik, P., Wolterink, G., De Brabander, H., and Van Ree, J. M.
memory for poison. Science 208:753–755. (1991b). Neuropeptides related to [Arg 8] vasopressin facili-
Paolini, A. G., and McKenzie, J. S. (1993). Effects of lesions in tates social recognition in rats. Physiol. Behav. 49: 1031–1035.
the horizontal diagonal band nucleus on olfactory habituation Posadas-Andrews, A., and Roper, T. J. (1983). Social transmission
in the rat. Neuroscience 57:717–724. of food-preferences in adult rats. Anim. Behav. 31: 265–271.
Olfactory Memory 435

Potter, H., and Butters, N. (1980). An assessment of olfactory Royet, J. P., Koenig, O., Gregoire, M. C., Cinotti, L., Lavenne, F.,
deficits in patients with damage to prefrontal cortex. Le Bars, D., Costes, N., Vigouroux, M., Farget, V., Sicard, G.,
Neuropsychologia 18:621–628. Holley, A., Mauguiere, F., Comar, D., and Froment, J. C.
Rampon, C., Tang, Y. P., Goodhouse, J., Shimizu, E., Kyin, M., (1999). Functional anatomy of perceptual and semantic pro-
and Tsien, J. Z. (2000). Enrichment induces structural cessing for odors. J Cogn. Neurosci. 11:94–109.
changes and recovery from nonspatial memory deficits in Saar, D., Grossman, Y., and Barkai, E. (1998). Reduced after-
CA1 NMDAR1-knockout mice. Nat. Neurosci. 3:238–244. hyperpolarization in rat piriform cortex pyramidal neurons is
Ramus, S. J., and Eichenbaum, H. (2000). Neural correlates of associated with increased learning capability during operant
olfactory recognition memory in the rat orbitofrontal cortex. conditioning. Eur. J. Neurosci. 10:1518–1523.
J. Neurosci. 20:8199–8208. Saar, D., Grossman, Y., and Barkai, E. (1999). Reduced synaptic
Rausch, R., and Serafetinides, E. A. (1975). Specific alteration of facilitation between pyramidal neurons in the piriform cortex
olfaction function in humans with temporal lobe lesions. after odor learning. J. Neurosci. 19:8616–8622.
Nature 225:557–558. Santibanez, G., and Hamuy, T. P. (1957). Olfactory discrimina-
Rausch, R., Serafetinides, E. A., and Crandall, P. H. (1977). tion deficits in monkeys with temporal lobe ablations.
Olfactory memory in patients with anterior temporal lobec- J. Comp. Physiol. Psychol. 50:472–473.
tomy. Cortex 13:445–452. Sara, S. J., Roullet, P., and Przybyslawski, J. (1999).
Ravel, N., Vigouroux, M., Elaagouby, A., and Gervais, R. Consolidation of memory for odor-reward association: beta-
(1992). Scopolamine impairs delayed matching in an adrenergic receptor involvement in the late phase. Learn.
olfactory task in rats. Psychopharmacology (Berl) Mem. 6:88–96.
109:439–443. Savic, I., Gulyas, B., Larsson, M., and Roland, P. (2000).
Ravel, N., Elaagouby, A., and Gervais, R. (1994). Scopolamine Olfactory functions are mediated by parallel and hierarchical
injection into the olfactory bulb impairs short-term olfactory processing. Neuron 26:735–745.
memory in rats. Behav. Neurosci. 108:317–324. Sawyer, T. F., Hengehold, A. K., and Perez, W. A. (1984).
Reid, I. C., and Morris, R. G. (1992). Smells are no surer: rapid Chemosensory and hormonal mediation of social memory in
improvement in olfactory discrimination is not due to the male rats. Behav. Neurosci. 98:908–913.
acquisition of a learning set. Proc. R. Soc. Lond. B Biol. Sci. Schellinck, H. M., Brown, R., and Slotnick, B. M. (1991).
247:137–143. Training rats to discriminate between the odors of individual
Reid, I. C., and Morris, R. G. (1993). The enigma of olfactory conspecifics. Anim. Learn. Behav. 19:223–233.
learning. Trends Neurosci. 16:17–20. Schellinck, H. M., Rooney, E., and Brown, R. E. (1995). Odors
Richardson, J. T. E., and Zucco, G. M. (1989). Cognition and of individuality of germfree mice are not discriminated by rats
olfaction: a review. Psychol. Bull. 105:352–360. in a habituation-dishabituation procedure. Physiol. Behav.
Richardson, R., Vishney, A. and Lee, J. (1999). Conditioned odor 57:1005–1008.
potentiation of startle in rats. Behav. Neurosci. 113:787–794. Schmajuk, N. A., and Isaacson, R. L. (1984). Classical contin-
Robbins, T. W., and Everitt, B. J. (1996). Neurobehavioural gencies in rats with hippocampal lesions. Physiol. Behav.
mechanisms of reward and motivation. Curr. Op. in 33:889–893.
Neurobiol. 6:228–236. Schoenbaum, G., and Eichenbaum, H. (1995a). Information cod-
Rolls, E. T. (1997). Taste and olfactory processing in the brain ing in the rodent prefrontal cortex. I. Single-neuron activity in
and its relation to the control of eating. Crit. Rev. Neurobiol. orbitofrontal cortex compared with that in pyriform cortex.
11:263–287. J. Neurophysiol. 74:733–750.
Rolls, E. T. (2000). The orbitofrontal cortex and reward. Cereb. Schoenbaum, G., and Eichenbaum, H. (1995b). Information cod-
Cortex 10:284–294. ing in the rodent prefrontal cortex. II. Ensemble activity in
Rolls, E. T., Critchley, H. D., Mason, R., and Wakeman, E. A. orbitofrontal cortex. J. Neurophysiol. 74:751–762.
(1996a). Orbitofrontal cortex neurons: role in olfactory and Schoenbaum, G., Chiba, A. A., and Gallagher, M. (1998).
visual association learning. J. Neurophysiol. 75:1970–1981. Orbitofrontal cortex and basolateral amygdala encode expected
Rolls, E. T., Critchley, H. D., and Treves, A. (1996b). outcomes during learning. Nat. Neurosci. 1: 155–159.
Representation of olfactory information in the primate Schoenbaum, G., Chiba, A. A., and Gallagher, M. (1999). Neural
orbitofrontal cortex. J. Neurophysiol. 75:1982–1996. encoding in orbitofrontal cortex and basolateral amygdala during
Roman, F., Staubli, U., and Lynch, G. (1987). Evidence for olfactory discrimination learning. J. Neurosci. 19: 1876–1884.
synaptic potentiation in a cortical network during learning. Schoenbaum, G., Chiba, A. A., and Gallagher, M. (2000).
Brain Res. 418:221–226. Changes in functional connectivity in orbitofrontal cortex and
Roman, F. S., Chaillan, F. A., and Soumireu-Mourat, B. (1993a). basolateral amygdala during learning and reversal training.
Long-term potentiation in rat piriform cortex following dis- J. Neurosci. 20:5179–5189.
crimination learning. Brain Res. 601:265–272. Schuckman, H., Kling, A., and Orbach, J. (1969). Olfactory dis-
Roman, F. S., Simonetto, I., and Soumireu-Mourat, B. (1993b). crimination in monkeys with lesions in the amygdala.
Learning and memory of odor-reward association: selective J. Comp. Physiol. Psychol. 67:212–215.
impairment following horizontal diagonal band lesions. Schul, R., Slotnick, B. M., and Dudai, Y. (1996). Flavor and the
Behav. Neurosci. 107:72–81. frontal cortex. Behav. Neurosci. 110:760–765.
436 Petrulis and Eichenbaum

Sekiguchi, R., Wolterink, G., and van Ree, J. M. (1991a). Staubli, U., Ivy, G., and Lynch, G. (1984). Hippocampal dener-
Analysis of the influence of vasopressin neuropeptides on vation causes rapid forgetting of olfactory information in rats.
social recognition of rats. Eur. Neuropsychopharmacol. Proc. Natl. Acad. Sci. USA 81:5885–5887.
1:123–126. Staubli, U., Baudry, M., and Lynch, G. (1985). Olfactory dis-
Sekiguchi, R., Wolterink, G., and van Ree, J. M. (1991b). Short crimination learning is blocked by leupeptin, a thiol protease
duration of retroactive facilitation of social recognition in rats. inhibitor. Brain Res. 337:333–336.
Physiol. Behav. 50:1253–1256. Staubli, U., Fraser, D., Kessler, M., and Lynch, G. (1986). Studies
Serby, M., and Chobor, K. (Eds.). (1992). The Science of on retrograde and anterograde amnesia of olfactory memory
Olfaction. Springer-Verlag, New York. after denervation of the hippocampus by entorhinal cortex
Shipley, M. T., and Ennis, M. (1996). Functional organization of lesions. Behav. Neural Biol. 46:432–444.
olfactory system. J. Neurobiol. 30:123–176. Staubli, U., Fraser, D., Faraday, R., and Lynch, G. (1987a).
Silva, A. J., Kogan, J. H., Frankland, P. W., and Kida, S. (1998). Olfaction and the “data” memory system in rats. Behav.
CREB and memory. Ann. Rev. Neurosci. 21:127–148. Neurosci. 101:757–765.
Slotnick, B. M. (1984). Olfactory stimulus control in the rat. Staubli, U., Schottler, F., and Nejat-Bina, D. (1987b). Role of
Chem. Senses 9:157–165. dorsomedial thalamic nucleus and piriform cortex in process-
Slotnick, B. M. (1985). Olfactory discrimination in rats with ing olfactory information. Behav. Brain Res. 25:117–129.
anterior amygdala lesions. Behav. Neurosci. 99:956–963. Staubli, U., Thibault, O., DiLorenzo, M., and Lynch, G. (1989).
Slotnick, B. M. (1994). The enigma of olfactory learning revis- Antagonism of NMDA receptors impairs acquisition but not
ited. Neuroscience 58:1–12. retention of olfactory memory. Behav. Neurosci. 103:54–60.
Slotnick, B. M., and Berman, E. J. (1980). Transection of the lat- Staubli, U., Perez, Y., Xu, F. B., Rogers, G., Ingvar, M., Stone-
eral olfactory tract does not produce anosmia. Brain Res. Bull. Elander, S., and Lynch, G. (1994a). Centrally active modula-
5:141–145. tors of glutamate receptors facilitate the induction of
Slotnick, B. M., and Kaneko, N. (1981). Role of mediodorsal long-term potentiation in vivo. Proc. Natl. Acad. Sci. USA
thalamic nucleus in olfactory discrimination learning in rats.
91:11158–11162.
Science 214:91–92.
Staubli, U., Rogers, G., and Lynch, G. (1994b). Facilitation of
Slotnick, B. M., and Katz, H. M. (1974). Olfactory learning-set
glutamate receptors enhances memory. Proc. Natl. Acad. Sci.
formation in rats. Science 185:796–798.
USA 91:777–781.
Slotnick, B. M., and Nigrosh, B. J. (1974). Olfactory stimulus
Staubli, U., Izrael, Z., and Xu, F. B. (1996). Remembrance of
control evaluated in a small animal olfactometer. Percept.
odors past: enhancement by central facilitation of AMPA
Motor Skills 39:583–597.
receptors. Behav. Neurosci. 110:1067–1073.
Slotnick, B. M., and Risser, J. M. (1990). Odor memory and odor
Strupp, B. J., and Levitsky, D. A. (1984). Social transmission of
learning in rats with lesions of the lateral olfactory tract and
food preference in adult hooded rats (Rattus norvegicus).
mediodorsal thalamic nucleus. Brain Res. 529:23–29.
Slotnick, B. M. and Schoonover, F. W. (1992). Olfactory path- J. Comp. Psychol. 98:257–266.
ways and the sense of smell. Neurosci. Biobehav. Rev. Strupp, B. J., Bunsey, M., Bertsche, B., Levitsky, D. A., and
16:453–472. Kesler, M. (1990). Enhancement and impairment of memory
Slotnick, B. M., Kufera, A., and Silberberg, A. M. (1991). retrieval by a vasopressin metabolite: an interaction with
Olfactory learning and odor memory in the rat. Physiol. accessibility of the memory. Behav. Neurosci. 104: 268–276.
Behav. 50:555–561. Sundberg, H., Doving, K., Novikov, S., and Ursin, H. (1982). A
Slotnick, B. M., Westbrook, F., and Darling, F. M. C. (1997). method for studying responses and habituation to odors in
What the rat’s nose tells the rat’s mouth: long delay aversion rats. Behav. Neural Biol. 34:113–119.
conditioning with aqueous odors and potentiation of taste by Sutherland, R. J., McDonald, R. J., Hill, C. R., and Rudy, J. W.
odors. Anim. Learn. Behav. 25:357–369. (1989). Damage to the hippocampal-formation in rats selec-
Sobel, N., Prabhakaran, V., Desmond, J. E., Glover, G. H., tively impairs the ability to learn cue relationships. Behav.
Goode, R. L., Sullivan, E. V., and Gabrieli, J. D. E. (1998a). Neural Biol. 52:331–356.
Sniffing and smelling: separate subsystems in the human Swann, H. G. (1934). The function of the brain in olfaction. II. The
olfactory cortex. Nature 392:282–286. results of destruction of olfactory and other nervous structures
Sobel, N., Prabhakaran, V., Hartley, C. A., Desmond, J. E., Zhao, upon the discrimination of odors. J. Comp. Neurol. 59:175–201.
Z., Glover, G. H., Gabrieli, J. D., and Sullivan, E. V. (1998b). Swann, H. G. (1935). The function of the brain in olfaction. I.
Odorant-induced and sniff-induced activation in the cerebel- The effects of large cortical lesions on olfactory discrimina-
lum of the human. J. Neurosci. 18:8990–9001. tion. Am. J. Physiol. 111:257–262.
Sobel, N., Prabhakaran, V., Zhao, Z., Desmond, J. E., Glover, G. Tanabe, T., Iino, M., and Takagi, S. F. (1975a). Discrimination of
H., Sullivan, E. V., and Gabrieli, J. D. (2000). Time course of odors in olfactory bulb, pyriform-amygdaloid areas, and
odorant-induced activation in the human primary olfactory orbitofrontal cortex of the monkey. J. Neurophysiol.
cortex. J. Neurophysiol. 83:537–551. 38:1284–1296.
Soffie, M., and Lamberty, Y. (1988). Scopolamine effects on Tanabe, T., Yarita, H., Iino, M., Ooshima, Y., and Takagi, S. F.
juvenile conspecific recognition in rats: possible interaction (1975b). An olfactory projection area in orbitofrontal cortex
with olfactory sensitivity. Behav. Proc. 17:181–190. of the monkey. J. Neurophysiol. 38:1269–1283.
Olfactory Memory 437

Thanos, P. K., and Slotnick, B. M. (1997). Short-term odor Whishaw, I. Q., Tomie, J. A., and Kolb, B. (1992). Ventrolateral
memory: Effects of posterior transection of the lateral olfac- prefrontal cortex lesions in rats impair the acquisition and
tory tract in the rat. Physiol. Behav. 61:903–906. retention of a tactile-olfactory configural task. Behav.
Thompson, R. (1980a). Odor discrimination performance as Neurosci. 106:597–603.
affected by cortical, thalamic and cerebellar lesions in rats. White, N. M., and Viaud, M. (1991). Localized intracaudate
Physiol. Behav. 24:797–800. dopamine D2 receptor activation during the post-training
Thompson, R. (1980b). Some subcortical regions critical for period improves memory for visual or olfactory conditioned
retention of an odor discrimination in albino rats. Physiol. emotional responses in rats. Behav. Neural Biol. 55:
Behav. 24:915–921. 255–269.
Thompson, R. (1980c). Sparing of an olfactory discrimination White, T. L. (1998). Olfactory memory: the long and short of it.
habit following extensive neocortical removals in rats. Chem. Senses 23:433–441.
Physiol. Behav. 25:405–408. Wiebe, S. P., and Staubli, U. V. (1999). Dynamic filtering of
Thor, D. H., and Holloway, W. R. (1982). Social memory of recognition memory codes in the hippocampus. J. Neurosci.
the male laboratory rat. J. Comp. Physiol. Psychol. 19:10562–10574.
96:1000–1006. Wiener, S. I., Paul, C. A., and Eichenbaum, H. (1989). Spatial and
Tomie, J. A., and Whishaw, I. Q. (1990). New paradigms for tac- behavioral correlates of hippocampal neuronal activity.
tile discrimination studies with the rat: methods for simple, J. Neurosci. 9:2737–2763.
conditional, and configural discriminations. Physiol. Behav. Williams, J. L., and Scott, D. K. (1989). Influence of conspecific
48:225–231. and predatory stressors and their associated odors on defen-
Tomie, J. A., and Whishaw, I. Q. (1996). Configural learning, sive burying and freezing responses. Anim. Learn. Behav.
spatial navigation, and amphetamine-induced activity spared 17:383–393.
with ibotenic acid but not quinolinic acid lesions of the dor- Williams, J. L., Rogers, A. G., and Adler, A. P. (1990). Prolonged
exposure to conspecific and predator odors reduces fear reac-
somedial thalamus in rats. Psychobiology 24:99–109.
tions to those odors during subsequent prod-shock tests.
van Wimersma Greidanus, T. B., and Maigret, C. (1996). The role
Anim. Learn. Behav. 18:453–461.
of limbic vasopressin and oxytocin in social recognition.
Williams, J. R., Slotnick, B. M., Kirkpatrick, B. W., and Carter,
Brain Res. 713:153–159.
C. S. (1992). Olfactory bulb removal affects partner prefer-
Vasilieva, N. Y., and Sokolov, V. E. (1994). The role of midven-
ence development and estrus induction in female prairie voles.
tral gland secretion in individual discrimination by
Physiol. Behav. 52:635–639.
Djungarian hamster (Phodopus campbelli Thomas, 1905)
Wilson, D. A. (1998a). Habituation of odor responses in the rat
females. Ethology 98:192–200.
anterior piriform cortex. J. Neurophysiol. 79:1425–1440.
Viaud, M. D., and White, N. M. (1989). Dissociation of visual
Wilson, D. A. (1998b). Synaptic correlates of odor habituation
and olfactory conditioning in the neostriatum of rats. Behav. in the rat anterior piriform cortex. J. Neurophysiol.
Brain Res. 32:31–42. 80:998–1001.
Vom Saal, F. S., Hamilton, L. W., and Gandelman, R. J. (1975). Wilson, D. A. (2000). Odor specificity of habituation in the rat
Faster acquisition of an olfactory discrimination following anterior piriform cortex. J. Neurophysiol. 83:139–145.
septal lesions in male albino rats. Physiol. Behav. Winocur, G. (1990). Anterograde and retrograde amnesia in rats
14:697–703. with dorsal hippocampal or dorsomedial thalamic lesions.
Wang, H. W., Wysocki, C. J., and Gold, G. H. (1993). Induction Behav. Brain Res. 38:145–154.
of olfactory receptor sensitivity in mice. Science Winslow, J. T., and Camacho, F. (1995). Cholinergic modulation
260:998–1000. of a decrement in social investigation following repeated
Wang, Z., Hulihan, T. J., and Insel, T. R. (1997). Sexual and contacts between mice. Psychopharmacology (Berl)
social experience is associated with different patterns of 121:164–172.
behavior and neural activation in male prairie voles. Brain Wirth, S., Ferry, B., and Di Scala, G. (1998). Facilitation of olfa-
Res. 767:321–332. tory recognition by lateral entorhinal cortex lesion in rats.
Wang, Z., Yu, G., Cascio, C., Liu, Y., Gingrich, B., and Insel, T. Behav. Brain Res. 91:49–59.
R. (1999). Dopamine D2 receptor-mediated regulation of Wood, E. R., Dudchenko, P. A., and Eichenbaum, H. (1999). The
partner preferences in female prairie voles (Microtus ochro- global record of memory in hippocampal neuronal activity.
gaster): a mechanism for pair bonding? Behav. Neurosci. Nature 397:613–616.
113:602–611. Yamazaki, K., Beauchamp, G. K., Imai, Y., Bard, J., Phelan, S. P.,
West, C. H., Clancy, A. N., and Michael, R. P. (1992). Enhanced Thomas, L., and Boyse, E. A. (1990). Odor types determined
responses of nucleus accumbens neurons in male rats to novel by the major histocompatibility complex in germfree mice.
odors associated with sexually receptive females. Brain Res. Proc. Natl. Acad. Sci. USA 87:8413–8416.
585:49–55. Young, B. J., Otto, T., Fox, G. D., and Eichenbaum, H. (1997).
Whishaw, I. Q., and Tomie, J. A. (1991). Acquisition and reten- Memory representation within the parahippocampal region.
tion by hippocampal rats of simple, conditional, and config- J. Neurosci. 17:5183–5195.
ural tasks using tactile and olfactory cues: implications for Young, L. J., Wang, Z., and Insel, T. R. (1998). Neuroendocrine
hippocampal function. Behav. Neurosci. 105:787–797. bases of monogamy. Trends Neurosci. 21:71–75.
438 Petrulis and Eichenbaum

Youngentob, S. L., Markert, L. M., Mozell, M. M., and Hornung, olfactory stimulation. Proc. Natl. Acad. Sci. USA 94:
D. E. (1990). A method for establishing a five odorant identi- 4119–4124.
fication confusion matrix task in rats. Physiol. Behav. 47: Zhang, Y., Burk, J. A., Glode, B. M., and Mair, R. G. (1998).
1053–1059. Effects of thalamic and olfactory cortical lesions on continuous
Youngentob, S. L., Markert, L. M., Hill, T. W., Matyas, E. P., and olfactory delayed nonmatching-to-sample and olfactory dis-
Mozell, M. M. (1991). Odorant identification in rats: an crimination in rats (Rattus norvegicus). Behav. Neurosci.
update. Physiol. Behav. 49:1293–1296. 112:39–53.
Zaborszky, L., Carlsen, J., Brashear, H. R., and Heimer, L. Zola-Morgan, S., Squire, L. R., and Amaral, D. G. (1989). Lesions
(1986). Cholinergic and GABAergic afferents to the olfactory of the hippocampal formation but not lesions of the fornix or the
bulb in the rat with special emphasis on the projection neurons mammillary nuclei produce long-lasting memory impairment in
in the nucleus of the horizontal limb of the diagonal band. monkeys. J. Neurosci. 9:898–913.
J. Comp. Neurol. 243:488–509. Zyzak, D. R., Otto, T., Eichenbaum, H., and Gallagher, M.
Zald, D. H., and Pardo, J. V. (1997). Emotion, olfaction, and the (1995). Cognitive decline associated with normal aging in
human amygdala: amygdala activation during aversive rats: a neuropsychological approach. Learn. Mem. 2:1–16.
21

Nasal Patency and the Aerodynamics of Nasal Airflow:

Measurement by Rhinomanometry and Acoustic Rhinometry,
and the Influence of Pharmacological Agents

Richard E. Frye
Children’s Hospital, Boston, Massachusetts, U.S.A.

I. INTRODUCTION damaged nasal function. Tonndorf (1939) demonstrated

turbulent flow in nasal models, throwing into question
A. Recognition of the Nose in Respiratory
Kayser’s laminar flow assumption. The rhinomanometer
Function: A Brief History
was further improved upon when Spoor (1965) incorporated
electronic pressure transducers, thereby eliminating the U-
Ebers’ Papyrus, the only complete Egyptian papyrus, men-
tube water manometer—a device with limited frequency
tions the nose as a respiratory organ: “As to the breath
response (Nakano, 1967; Randall, 1962).
which enters into the nose; it enters the heart and lungs;
these give to the whole belly” (Ebbell, 1937). The
Egyptians knew that the nose secretes mucus, contains B. Functional Physiology of the Nasal Airway:
arteries and veins, and is responsible for olfaction. A Brief Review
Operations to repair nasal bone fractures and remove
polyps were recorded (Pahor, 1992; Pahor and Kimura, The nasal airway optimizes gas exchanged by conditioning
1991). Galen was the only Greek physician or philosopher and filtering inspired air; incoming air is heated, humidi-
on record to recognize the importance of the nose in respi- fied, and filtered of airborne pathogens and environmental
ration (Kimmelman, 1989). pollutants. Thermoregulation and systemic water balance is
In 1844, Piorry espoused the importance of the nose in maintained by recovering heat and water vapor from
respiratory function. He divided the etiology of nasal steno- expired air and is independent of nasal airway resistance
sis into those produced by septal deviation or alternating (Keck et al., 2000). Thomson and Dudley-Buxton (1923)
vasomotor stenosis (Williams et al., 1970). Half a century highlighted the functional significance of nasal morphol-
later, Franke (1894) measured nasopharyngeal pressure ogy by demonstrating that the anthropological cephalomet-
changes in patients with nasal abnormalities. Soon there- ric nasal index varies across race in relation to indigenous
after, Kayser (1895) created the first anterior rhino- climate. Indeed, the major change in nasal morphology
manometer by simultaneously measuring nasal volume flow from Australopithecus sp. to Homo sapiens supposedly
and nasopharyngeal pressure. Based on measurements from provided a selective advantage by conserving moisture in
these devices, Franke and Kayser performed operations to arid environments (Franciscus and Trinkaus, 1988).
enlarge the nasal lumen. However, in many cases the Nasal airway patency is functionally coordinated
operations failed to improve the underlying condition and with pulmonary function. By matching lower airway

439
440 Frye

impedance, the nose assists in the control of breathing pressure across a conduit (i.e., the nasal cavity) to the vol-
frequency and expiration length and provides positive end- ume airflow rate through the conduit:
expiratory pressure. Turbinate swelling is increased by
pulmonary stretch, lower airway irritation, normocapnic 8   lq
p

progressive hypoxia, and respiratory drive (Lung and  r4o
Wang, 1991; Maltais et al., 1991; Nishihira and In this equation p is pressure, q is volume flow, l is
McCaffrey, 1987; Series et al., 1989). Although stimula- length of the conduit,  is dynamic viscosity, and ro is the
tion of neither the nose nor the larynx influences elastic or radius of the tube. This equation is easily modified to
resistive lung properties (Jacobs and Dickson, 1986), the express Poiseuille resistance, R, an analog to the electrical
nasal airway can influence distal airway function. For ohmic-type resistance:
example, removal of nasal obstruction improves sleep
apnea, and nasal breathing improves exercise tolerance in p 8  l

R

patients with chronic obstructive pulmonary disease and q  ro4
decreases pulmonary work during exercise in the nonath- Poiseuille resistance is used to describe nasal airway
lete (Lamblin et al., 2000; Lavie, 1987; Morrison et al. resistance (NAR), as well as upper and lower airway resist-
1989; Petruson and Bjuro, 1990, Tanaka et al., 1988). ance. This equation assumes that the airway is circular,
straight, rigid, and uniform and that air is homogeneous and
C. Influence of the Nose on Olfactory Function noncompressible. Although the latter assumption is rarely
violated to any significant extent, the former assumptions
The nose contains the olfactory neuroepithelium, a small are most certainly violated by the complicated shape of the
(~2 cm2) region of the nasal mucosa located on the cribri- nasal cavity and the compliant nature of the nasal valve.
form plate, upper nasal septum, dorsal superior turbinate, Since the resistance of a conduit is inversely proportional to
and regions of the middle turbinate. The irregular contour the fourth power of the radius (Leyton, 1975), small
of the nasal cavity, combined with high velocity airflow, changes in the diameter of the nasal cavity can greatly
produces nonlinear aerodynamics that promote odorant influence Poiseuille resistance.
mixing while producing a complicated odorant distribu- In straight, rigid, uniform circular conduits, laminar
tion. Only 15% of the incoming airstream passes near the flow predominates at low flow rates, whereas turbulent
olfactory epithelium. Subtle alternations in nasal geometry flow predominates at high flow rates (Rohrer, 1915).
can deflect the airstream away from the olfactory epithe- Reynolds (1883) developed an index to measure the rela-
lium; in many cases nasal function and patency perception tive turbulence of fluid flow:
are unaffected. Only 15–20% of patients referred to clini-
2ro   q
cal smell and taste centers have diagnosable obstructive Re


airflow abnormalities (Deems et al., 1991; Mott and
Leopold, 1991). The Reynolds number depends on fluid density () and
molecular viscosity (). A Reynolds number below 1500
usually indicates laminar flow, while a Reynolds number
II. AERODYNAMICS OF THE NASAL CAVITY between 1500 and 2000 is indicative of mixed turbulent
and laminar flow. Turbulence usually develops when the
Nasal aerodynamics are influenced by nasal anatomy and Reynolds number exceeds 2000, but this is dependent on
physiology. Functional nasal cavity aerodynamics is eval- the shape and roughness of the conduit. For example, a
uated in vivo by rhinomanometry—a technique that meas- bifurcating conduit will develop turbulence at a Reynolds
ures differential pressure across, and volume flow number of 900 (Jones et al., 1969).
through, the nasal cavity. Nasal pressure and flow are non- Although the average velocity through a conduit can be
linearly related in normal and disordered airways due to calculated by dividing the volume flow rate by the cross-
turbulent flow and the dynamic narrowing of the nasal sectional area of the conduit, the velocity at any particular
valve. point is a function of distance from the conduit’s wall. Fluid
near the wall moves slowly owing to shear stress, while fluid
A. Turbulent Flow in the center moves most rapidly. The velocity profile gradient
depends on the flow type. Laminar flow is organized in
Nasal cavity aerodynamics is customarily described by the concentric layers, with the outermost layer traveling at almost
equations of fluid mechanics, particularly the Hagen- zero velocity and the centermost layer traveling at maximum
Poiseuille equation. This equation relates the driving- velocity. This arrangement results in a parabolic flow profile.
Nasal Patency and Nasal Airflow Aerodynamics 441

Turbulent flow develops at high flow rates or when the lower critical pressure (Adamson, 1987; Goode, 1985;
geometry of a conduit contains irregular walls, bifurca- Kasperbauer and Kern, 1987) or a static nasal valve
tions, bends, or abrupt changes in cross-sectional area (Guillette and Perry, 1990). Obstruction in this region is
(CSA). Flow becomes disordered and local eddies develop more critical than any in other region of the nasal chamber
within the jetstream. Once turbulence develops, the force (Berkinshaw et al., 1987).
that previously promoted forward fluid motion moves the
fluid in directions perpendicular to the forward motion. C. Measuring Nasal Airway Properties
Efficiency is greatly decreased and a substantial increase
in pressure is needed to produce a small increase in flow Many approaches are used to measure nasal cavity proper-
rate. The turbulent flow profile is flat, with a quick veloc- ties. Changes in nasal peak flow and pressures were the
ity drop near the wall of the conduit. first parameters recognized. Simultaneous pressure and
flow measurements allow nasal airway resistance (NAR)
B. The Collapsible Nasal Valve calculations. Various preset points for pressure or flow val-
ues have been selected to standardize NAR. However, no
The pressure drop across a tube due to airflow depends on standardized measurement is universally accepted. Despite
the smallest CSA, also know as the minimal cross-sectional the development of advanced curve fitting algorithms,
area (MCA) (Williams et al., 1970). In the normal nasal mathematical model coefficients are not familiar to many
airway, this is the nasal valve, which, under normal condi- clinicians and documentation of clinical correlation is
tions, accounts for 90% of the pressure drop (Jones et al., lacking. NAR is best measured by anterior rhinomanome-
1988). Although Uddstromer recognized the importance of try, although this method is the most technically difficult,
the nasal valve in 1939, its function was not quantitatively NAR varies significantly among subjects, although within-
described until 1970 (Bridger and Proctor, 1970). subject variation is quite small and physiological
The nasal valve is located approximately 2 cm posterior alterations in NAR result from the nasal cycle, body
to the entrance of the naris; its CSA is normally about position, and nasal decongestion.
1.5 cm2. It is composed of compliant tissue, allowing it to Acoustic rhinometry is a relatively new technique for
partly collapse when the differential pressure reaches a quantitatively measuring nasal cavity dimensions. Sound
critical value; thus, a collapsible tube model can be pulse trains are introduced into the nose and the sound
applied. The differential pressure across a collapsible tube reflections are analyzed. Nasal CSA is graphed as a func-
depends on the fluid flow rate, which, in turn, depends on tion of distance from the naris. Rigid nasal endoscopy has
resistance. Since resistance depends on the MCA, and the correlated anatomical landmarks with sequential CSA
MCA depends on the differential pressure, pressure and minima: the first corresponds to the nasal valve, the second
flow indirectly regulate each other. Collapsible tubes are corresponds to the anterior end of the inferior turbinate,
known as flow regulators, which prevent flow from and the third corresponds to the anterior end of the middle
exceeding an upper limit (Conrad, 1969; Holt, 1969). turbinate (Corey et al., 1999). Primary reportable measure-
Critical pressure depends on both upstream and ments include unilateral or total nose MCA as well as nasal
downstream resistance; thus, the dynamics of the nasal cavity volume (NCV).
valve depend on the nasal airflow direction (i.e., expiration Acoustic rhinometry was introduced to the otorhino-
or inspiration). laryngology community only a decade ago—a relatively
Nasal alar muscle activity reduces nasal valve short time for a new technology to grow in popularity.
elasticity, thereby increasing critical pressure (Cole et al., Since its introduction the reliability and accuracy has been
1985). Indeed, nasal resistance on the paralyzed side of studied repeatedly, and normative values have been devel-
patients with unilateral facial palsy is four times that of the oped for a wide variety of ages and ethnic groups. This
nonparalyzed side owing to the inactivity of the nasal alar technique requires minimal cooperation from the subject
muscles (Van Dishoeck, 1964). Many factors influence the and does not require breathing effort, making it particular
onset and magnitude of alar muscle activity, including helpful in evaluating children with nasal obstruction. The
breathing rate, maximum flow rate, acceleration of airflow, utility of acoustic rhinometry is clearly demonstrated in its
sleep state, CO2 concentration, resistive loading, and nega- ability to measure parameters in the nasal airway of infants —
tive airway pressure (Mezzanotte et al., 1992; Strohl et al., a feat rather difficult to perform with rhinomanometry.
1980, 1982). The advantages and drawbacks of acoustic rhinometry
Sagging or depression of the upper lateral nasal carti- and rhinomanometry must be considered individually, as
lage, anterior nasal septum buckling, or inferior turbinate each method measures a different entity. Both methods are
inflammation can narrow the nasal valve area, leading to a reliable over several weeks, if performed by an experienced
442 Frye

operator under controlled circumstances (Silkoff et al., air takes the path of least resistance, it is easily redistribu-
1999). Anatomical nasal cavity parameters, as measured by ted if one path is blocked. Thus, airstream distribution can
acoustic rhinometry, provide important information con- be profoundly influenced without a substantial change in
cerning the size and location of the maximum flow veloc- nasal resistance. Indeed, nasal resistance is typically
ity. However, only anterior nasal cavity measurements elevated only by severe nasal abnormalities, yet particle
correlate with clinical abnormalities. Indeed, the accuracy distribution can be altered by rather minor abnormalities.
of acoustic rhinometric measurements is greatly reduced Clinical studies have associated certain nasal abnormalities
beyond the MCA. with olfactory dysfunction; however, the nasal abnormal-
Special caution is required when interpreting posterior ities that are described are rarely localized to a specific
CSA and NCV measurements, since the sound reflections nasal area. Several attempts to correlate olfactory function
depend on nasal cavity shape and the size and location of with specific anatomic nasal areas have been made
the MCA. For example, a differential change in the (Hornung and Leopold, 1999; Leopold, 1988).
expansion of the first and second minima, after pharma-
ceutical, operative, or prosthetic manipulation of the A. Nasal Anatomy and Patency
nose, may produce apparent movement of the MCA; if
1. Simulated Nasal Abnormalities
the MCA has moved, the change in the MCA value will
be based on the CSA of a different anatomical landmark, Several investigators have introduced artificial obstruc-
making the measure’s validity questionable (Tomkinson tions in the human airway. For example, Cole et al. (1988)
and Eccles, 1998). found that fiberfoam protruding 3–5 mm into the nonde-
A small MCA, as seen during mucosal inflammation or congested airway at the upper lateral cartilage area (i.e.,
severe septal deviation, underestimates posterior NCV by nasal valve area), but not into other areas of the nasal cav-
limiting the transmission and reflection of sound distal to the ity, significantly influenced NAR. When the mucosa was
MCA. For example, NCV measured by coronal high-reso- decongested, 4- and 5-mm obstructions in the upper lat-
lution computed tomography (CT) correlates well with eral cartilage area were required to meaningfully affect
acoustic rhinometry measurements in the anterior, but NAR. In the turbinate region of the nasal chamber larger-
not the posterior, nasal cavity (Dastidar et al., 1999a,b). sized obstructions blocking an extensive portion of the
Following decongestion or other procedures that expand the nondecongested airway were required to significantly
MCA, the posterior NCV will be less underestimated, if at alter NAR (Chaban et al., 1988). A simple nasal valve
all. Thus, NCV expansion will be falsely elevated, albeit not stent was inserted in the noses of patients with a static
reliably. Indeed, the application of external nasal valve dila- nasal valve and normal subjects with artificially created
tor strips causes not only a significant increase in nasal valve midseptal nasal obstructions (Guillette and Perry, 1990).
CSA but also an increase in NCV (Ng et al., 1998). In addi- Artificially created obstructions and nasal valve
tion, turbulent resistance as well as dynamics of the nasal abnormalities both resulted in equivalent NAR. However,
valve cannot be measured with a “snapshot” approach. a nasal stent only significantly improved the patients with
Despite the potential problems with acoustic rhino- nasal valve disorders. Haight et al. (1985) found that
manometry, it may provide the needed information to selective decongestion of the nasal valve region, but not
explain the between-subject variability in rhinomanomet- the nasal choanal region, markedly influenced NAR.
ric measurements and fill in some of the structural para- Using a clear acrylic model of the nasal passageway,
meters of the aerodynamic equations. Alternatively, the Levine et al. (1986) showed that moderate and severe ante-
information from both rhinomanometry and acoustic rior septal deviations significantly influenced Rohrer’s K1
rhinometry can be combined into a unique index. This and K2 coefficients—values that measure laminar and tur-
approach was taken, for example, by Kesavanathan et al. bulent airflow, respectively. Moderate and severe posterior
(1995), who determined the characteristics of the nasal septal deviations changed these coefficients to a lesser
pressure-volume ratio relationship in the anterior turbinate degree. Septal deviations placed in the turbinate region of
and nasal valve regions and their relation to nasal resist- the nasal chamber, even when severe, did not influence
ance for the normal and decongested nasal mucosa. these coefficients. Simulated mild and moderate enlarge-
ment of all turbinates altered the K1 coefficient.
These studies indicate that anterior airway obstructions,
III. NASAL CAVITY ANATOMY particularly in the region of the nasal valve, have a much
larger impact on NAR than posterior obstructions or flow
Nasal cavity anatomy influences nasal resistance, aerody- limitations. Only severe obstructions influence NAR
namics, particle deposition, and olfactory function. Since within the turbinate region.
Nasal Patency and Nasal Airflow Aerodynamics 443

2. Clinical Abnormalities NAR, anterior CSA, and symptoms (Kamami, 1997,

Kamami et al., 2000). Increasing the anterior nasal cavity
Grymer et al. (1997) studied 230 randomly selected adults,
size to a critical level may mitigate nasal valve dysfunc-
14% of whom had a subjective feeling of nasal obstruction.
tion. For example, lateral rhinotomy with medial maxillec-
Patients with a small anterior CSA or symptoms of sinu-
tomy improves NAR and nasal valve CSA despite
sitis or rhinitis, mostly as a result of anterior septal devia-
interrupting nasal valve support. This is presumably due to
tions or severe mucosa swelling, were most likely to have
increased CSA in the anterior nasal chamber by concomi-
subjective nasal obstruction.
tant resection of the anterior inferior turbinate (Leug et al.,
The importance of the anterior nasal chamber and the
1998). Nasal valve suspension in patients with nasal
nasal valve in producing NAR has been repeatedly demon-
obstruction due to nasal valve collapse improves subject
strated. Roithmann et al. (1994) showed a significant, nega-
nasal patency and NAR without a significant increase in
tive, nonlinear relationship between NAR and MCA area
MCA (Paniello, 1996).
in 78 patients suffering from nasal obstruction—small
Septoplasty with or without turbinoplasty, inferior
intrusions into the nasal lumen produced unusually large
turbinate cauterization, rhinoplasty, or uvulopalatopharyn-
increases in NAR when located in the valve region.
goplasty improves objective nasal patency and acoustic
Patients with postrhinoplasty nasal obstruction had a sig-
rhinometric measurements (Reber et al., 1998; Shemen and
nificantly smaller nasal valve CSA. In a later study these
Hamburg, 1997). However, acoustic rhinometric measure-
same authors found that symptoms, as well as CSA anom-
ments vary widely and do not correlate well with subjective
alies, were mitigated by external nasal dilation (Roithmann
nasal patency. Inferior turbinoplasty and endoscopic sinus
et al., 1997a).
surgery improved symptoms and acoustic rhinometric
Acoustic rhinometry and anterior rhinomanometry are
measurements for patients with chronic nasal obstruction
most sensitive in revealing severe deviations in the ante-
not caused by septal deviation. Radical trimming of the
rior nasal cavity, but are less sensitive in demonstrating
inferior turbinate reduced total NAR (Wight et al., 1988a).
middle and posterior deviations (Szucs and Clement,
Acoustic rhinometric patency measures and subjective
1998). Dinis et al. (1997) examined 45 consecutive adult
symptoms improved after bilateral inferior turbinoplasty,
subjects with complaints of nasal obstruction; CT scans
although the degree of subjective and objective improve-
delineated the specific etiology. Rhinomanometry uncov-
ment did not correlate well (Grymer et al., 1996). Although
ered abnormalities in patients with anterior septal devia-
several procedures for improving inferior turbinate hyper-
tion, but not posterior septal deviation or sinusitis.
trophy improve nasal function, Passali et al. (1999) found
Although disorders in the anterior nasal cavity dispro-
that submucosal resection without lateral displacement is
portionately influence NAR, general nasal airway obstruc-
superior to electrocautery, cryotherapy, laser cautery, or
tion also causes airflow and symptomatic changes.
turbinectomy. Illum (1997) showed that compensatory
Adenoid hypertrophy decreases nasopharyngeal CSA
anterior inferior turbinectomy with septoplasty does not
(Cho et al., 1999; Mostafa, 1997). Nasal provocation with
improve symptoms beyond septoplasty alone. Functional
allergins results in a dose-dependent change in MCA and
endoscopic sinus surgery supposedly improves chronic
NAR (Austin and Foreman, 1994; Roithmann et al.,
sinusitis by reducing mucosal edema without changing
1997b; Zweiman et al., 1997).
structural anatomy (Keles et al., 1998).
3. Nasal Surgery
4. Pharmaceutical Treatments
The influence of corrective surgery on subjective and
objective nasal patency varies from report to report. This is Pharmaceuticals used in the treatment of rhinnitis, specific-
probably due to the lumping of heterogeneous disorders ally topical decongestants, antihistamines, and steroids,
together and the failure to employ standardized outcome have been evaluated by objective nasal patency measure-
measures. However, studies that use well-validated out- ments. Rhinomanometry is commonly used to monitor
come measures and study populations consisting of rela- NAR and airflow, whereas acoustic rhinometry can meas-
tive specific disorders or operations do show some positive ure mucosal swelling. Caveats mentioned above concerning
effects of surgery. The studies confirm that the anterior changes in NCV with large changes in mucosal swelling
nose, in the region of the vestibule and nasal valve, is par- should be carefully considered when examining the results
ticularly important for improving nasal patency. of these studies.
Enlarging the anterior nasal cavity improves objective A number of recent studies have quantitatively exam-
and subjective nasal patency. Laser-assisted outpatient sep- ined the effects of the topical decongestant oxymetazoline
toplasty for moderate anterior septal deviation improves on airway patency. Most of these studies have employed
444 Frye

placebo and double-blind controls. Oxymetazoline is effi- allergy. Hilberg (1995) demonstrated that pretreatment
cacious when evaluated by NCV, MCA, NAR, or rhino- with beclomethasone is better than terfenadine in prevent-
stereometry (Bickford et al., 1999; Graf et al., 1999) and ing the allergen-induced increase in NAR. Intranasal
appears to be the most effective imidazoline derivative steroids also reduce the histamine-induced provocation
(Hochban et al., 1999). Many studies do not show dose- response in perennial rhinitis (Mygind et al., 1997). The
response effect beyond an effective dose. For example, efficacy of the thromboxane A2 receptor antagonist rama-
although 0.25 and 0.50 mg/mL significantly increased troban was demonstrated after allergen challenge with
NCV, the effect was not different between the two doses house dust; treatment prevented significant change in NCV
(Hummel et al., 1998). Twenty-five and 50 g, but not 6.25 and MCA from baseline (Terada et al., 1998).
or 12.5 of oxymetazoline, significantly reduced NAR.
Although total NCV increased in a dose-dependent man-
ner, MCA did significantly increase beyond the lowest
5. Nasal Valve Dilation
dose (Taverner et al., 1999). Although rebound swelling is
not seen in several studies (Graf et al., 1999; Hochban et The nasal valve and anterior vestibule are important con-
al., 1999), other studies have demonstrated its dynamics tributors to NAR. As noted earlier in this review, the nasal
(Morris et al., 1997). However, it may be related to dosage. valve is dynamic and collapsible, allowing it to change its
Few other nasal decongestants have been objectively resistance in proportion to airflow rate. A smaller resting
evaluated. One dose of pseudoephedrine (60 mg) signifi- nasal valve CSA results in a higher baseline valvular resis-
cantly improved symptoms and increased total MCA and tance and a relatively lower maximum airflow rate.
NCV, but did not change NAR, in healthy patients with Anterior septal deviations may deform the nasal valve,
moderate to severe acquired NAR as a complication making its baseline resistance higher, thereby causing
of the common cold (Taverner et al., 1999). Phenylpro- nasal obstruction.
panolamine (100 mg twice) a day had effects similar to top- External nasal dilator strips, such as the The Breath-
ical oxymetazoline, as measured by rhinostereometry and RightTM strip (CNS. Inc., Minneapolis, MN), can effec-
MCA (Graf et al., 1999). Inhaled furosemide significantly tively relieve most anterior nasal abnormalities.
decreased NAR in 12 patients with perennial nonallergic Application of such strips increased nasal valve CSA in the
rhinitis (Masieri et al., 1997). pre- and postdecongested nose (Ng et al., 1998). Strip
Intranasal steroid efficacy has also been studied with application improved MCA and NAR in normal subjects,
quantitative nasal patency measurements. Fluticasone and patients with septal deviation in the nasal valve area, and
beclomethasone aqueous nasal spray increased NCV in 32 patients with mucosal congestion, although the most
patients with severe polyposis as compared to placebo, marked change occurred in the septal deviation group
although fluticasone but not beclomethasone significantly (Roithmann et al., 1998). NAR and CSA are improved in
improved morning peak inspiratory flow rate after the first normal Caucasian subjects after the application of nasal
week of treatment (Lund et al., 1998). Beclomethasone strips (Gosepath et al., 1997). However, similar improve-
and budesonide improved nasal airflow in patients with ment was not reliably seen in African American subjects
nasal polyposis, although beclomethasone but not flu- (Portugal et al., 1997). Application of other nasal dilators
nisolide improved NAR in nasal polyposis patients follow- (Airplus, Prevancure AB, Vastra Frolunda, Sweden;
ing surgery. The efficacy of intranasal steroids for Improved Mechanical Therapeutic Nasal Dilator, Breath
perennial rhinitis in children and adults and in seasonal EEZ Corp., Brooklyn, New York) increases MCA and
allergic rhinitis has been confirmed using rhinomanomet- decreases NAR (Chaudhry et al., 1996; Nielsen et al.,
ric measurements (see Mygind et al., 1997). 1997). In a comparative study, Lorino et al. (1998) showed
Nasal provocation is used to determine the efficacy of that the internal nasal mechanical dilator Nozovent
specific pharmaceutical treatments for allergic rhinitis. (Prevancure AB, Vastra Frolunda, Sweden), but not the
Spaeth et al. (1996) documented the efficacy of azelastine external nasal strip device Respir (Kentia Diffusion,
using rhinomanometry, acoustic rhinometry, rhinoscopy, Boulogne, France) decreased NAR.
and symptom scores after histamine and allergen provoca- Nasal strips have physiological effects. Breathe-Right
tion. Astemizole improves NAR and symptom scores after nasal strips improved the respiratory disturbance index for
long-term provocation in subjects allergic to grass pollen patients with obstructive sleep apnea and snoring if the
(Horak et al., 1993). Nielsen et al. (1996) confirmed the patient also suffered from hyperplasia or hypertrophy of
utility of steroids for treating seasonal allergic rhinitis by the lower turbinates, septal deviation or allergic rhinitis,
demonstrating a difference in NCV and symptom scores only minor pharyngeal obstruction, or was less than
after metacholine challenge in patient’s with grass pollen 55 years in age (Gosepath et al., 1999). Nasal strips also
Nasal Patency and Nasal Airflow Aerodynamics 445

significantly decrease heart rate, ventilation, and VO2 in septum presumably redistribute the airstream within the
athletes during submaximal exercise (Griffin et al., 1997) nasal cavity. Airflow obstruction is only one mechanism
of olfactory dysfunction, and changes in the olfactory epi-
B. Nasal Anatomy and Olfactory Ability thelium due to, for example, inflammatory processes may
also have a significant effect (for review, see Doty and
1. Clinical Observations
Mishra, 2001).
Schneider and Wolf (1960) pioneered the study of nasal
anatomy and olfactory function. Unfortunately, the study
2. Nasal Cavity Volume and Olfactory Function
has a number of pitfalls, including the fact that the clin-
ician that made the nasal cavity measurements was not To identify the relationship between nasal anatomy and
blind to the threshold data. Most modern studies have olfactory dysfunction, Leopold (1988) obtained nasal cav-
not properly measured olfaction, evaluated sufficient ity CT scans of 34 patients with conductive or idiopathic
sample sizes, or used adequate methodological proce- hyposmia. The nasal cavity superior to the middle turbinate
dures (for reviews, see Doty and Frye, 1989; Leopold, was divided into nine sections by placing coronal borders
1986; Mott and Leopold, 1991). In general, loss of olfac- anterior and posterior to the cribriform plate and horizontal
tory function is related to allergic rhinitis, polyposis, borders at 5 and 10 mm below the cribriform plate. The
nasal sinus disease, or adenoid hypertrophy. For exam- volumes of these regions were measured and entered into a
ple, both allergic and nonallergic rhinitis patients have stepwise regression analysis using olfactory test score (as
hyposmia, but nonallergic rhinitis patients have a higher measured by an odor confusion matrix) (see Chapter 9) as
olfactory threshold than allergic rhinitis patients (Simola the dependent variable. Interestingly, a larger volume in the
and Malmberg, 1998). region 10–15 mm below the cribriform plate and a smaller
Hyposmia associated with nasal disease may be due to volume 1–5 mm inferior and anterior to the cribriform
airflow obstruction or olfactory mucosa inflammation. For plate was associated with higher olfactory scores. A larger
example, patients with chronic sinusitis who have inflam- volume in the region 10–15 mm below and posterior to the
matory cell influx into the olfactory mucosa are much cribriform plates potentiates both effects.
more likely to have an olfactory deficit (Kern, 2000). Hornung and Leopold (1999) measured unilateral, as
A reduction in the N1 component of olfactory and opposed to bilateral, olfactory function in 19 patients
trigeminal chemosensory event–related potentials are with static conductive hyposmia due to polyposis or
correlated with acute rhinitis symptoms. Only the mucosal edema. Like their previous study, the volume
trigeminal N1 component normalizes when mucus between the cribriform plate and middle meatus was
secretion subsides; olfactory chemosensory event–related divided up into nine sections. In addition, the nasal
potentials returned towards normal over a one-month chamber and vestibule was divided into 12 additional
period (Hummel et al., 1998a). This protracted hyposmia volumetric areas (Fig. 1). Interestingly, a larger volume
lends evidence to a nonobstructive etiology. 5–10 mm below and anterior to the cribriform plate and a
In some cases, olfactory function can be improved by smaller nasal vestibule volume correlated with better
endoscopic operative procedures (Seiden and Smith, 1988) olfactory function. A larger volume along the septum in
or adenoidectomy (Ghorbanian et al., 1983). Septoplasty the region between the inferior and middle turbinates
can improve olfaction and nasal resistance (Stevens and markedly increased the former effect.
Stevens, 1985), and the degree of septal curving before These studies suggest that critical areas of the nasal
surgery may be correlated with the degree of olfactory cavity, both near and remote from the olfactory receptors,
improvement (Kittle and Waller, 1973; Shevrygin, 1973). influence olfactory function, presumably by directing air
Although total removal of the inferior turbinate is related towards the olfactory cleft. A smaller nasal vestibule may
to improvement of olfactory function (Ophir et al., 1986), direct air to the olfactory cleft from the anterior aspect, and
such improvement varies with the technique used (Elwany a larger volume anterior and inferior to the cribriform plate
and Harrison, 1990). may increase airflow to the olfactory cleft. A larger
Even though studies suggest that surgical or medical volume along the septum just below the middle turbinate
intervention can improve olfaction, success is limited. may direct airflow to the olfactory cleft from a medial
Indeed, many patients do not experience restoration of aspect. In should be recognized that by excluding patients
olfactory function following intervention. Nasal sinus dis- and noses with anosmia or normosmia, as has been done in
ease associated with swelling of the superior nasal cavity the aforementioned studies, the variation in the olfactory
likely obstructs airflow to the olfactory region, while dis- data, as well as the degree to which these data can be gen-
orders associated with the inferior turbinate or nasal eralized to normal subjects, is reduced. In addition, even
446 Frye

Figure 1 Cutaway view of a right nasal cavity showing the subdivided nasal regions whose volume was correlated with olfactory func-
tion. (From Hornung and Leopold, 1999).

patients with conductive hyposmia may have some com- A. Nasal Airway Models
ponent of inflammation at the olfactory mucosa.
Nasal cavity models do not stimulate elastic or dynamic
properties of the mucosa or the nasal valve. Since most
models are based on a single nasal cavity, finding general-
IV. NASAL AIRFLOW PATTERNS
ization may be limited. In addition, suspended particles
may not simulate molecular dynamics accurately, and
The complicated shape of the nasal cavity divides the
induction of Pitot tubes or anemometers into a model may
incoming air into jet streams that flow between the nasal
alter aerodynamics. Nevertheless, these studies provide
turbinates and along the nasal septum. Turbulent flow,
information on general principles of nasal airflow and
vortices, and regions of negative pressure are created.
insight into the particle transport and distribution.
Vortices promote mixing, whereas high-speed flows asso-
ciated with regions of negative pressure may produce
1. Water Flow Models
nasal abnormalities. For example, Ogawa (1986) reports
that polyps occur more often on the concave side of uni- Aluminum particle deposition in the olfactory region of a
lateral septal deviated noses, possibly because of higher plastic nasal model increases with water velocity (Stuiver,
flow rates. 1958). Masing (1967) injected ink into the naris and
To localize specific nasal cavity areas that influence nasopharynx during simulated inspiration and expiration,
the direction and velocity of the airstream, investigators respectively, at a velocity of 0.75 L/min. Injection into the
have turned to nasal models. Water or air, along with an lateral, dorsal, or ventral naris caused ink flow along the
indicator substance such as dye or smoke, respectively, inferior nasal cavity, while injection into the medial or cen-
is moved through a model in order to identify flow tral naris resulted in diversion of ink to the superior and
stream dynamics. Most nasal models are qualitative, but middle nasal regions. Injection into the dorsal nasopharynx
some studies have directly measured velocity. diverted ink to superior nasal regions, while ink injection
Mathematical models of the nasal cavity, which simulate into the medial, central, and ventral nasopharynx was
nasal cavity aerodynamics and particle disposition, have diverted into the middle and inferior nasal regions. Ink
also been used. release in the medial nasopharynx caused the stream to
Nasal Patency and Nasal Airflow Aerodynamics 447

travel in the inferior nasal region after forming postturbinal clear bioplastic model at a flow rate of 10 L/min. A total of
vortices. 100–350 measurements were made at various horizontal
Swift and Proctor (1977) used both water and air in and vertical points at five cross sections: the nasal valve,
clear polyester resin cadaver casts to observe streamlines preturbinal, midturbinal, postturbinal, and nasopharynx
and measure airflow rates. With an inspiratory volume rate regions. During inspiration, the nasal valve directed air
of 12.5 L/min and a nostril CSA of 0.9 cm2, the air veloc- along the floor and lateral aspects of the nasal cavity at a
ity was 138 m/min and laminar. The nasal valve, which had rate of 63 m/min and along the upper nasal cavity entrance
a CSA of 0.32 cm2, had a computed average velocity of at lower velocities. In the preturbinal region, flow velocity
390 m/min and a measured maximum velocity of 1116 was greatest in the lower portion and, like the nasal valve
m/min. The abrupt increase in nasal chamber CSA poste- region, showed a very irregular velocity profile. Airflow in
rior to the nasal valve resulted in turbulence and a marked the midturbinal area was more regular, with the highest
decrease in airflow velocity. Turbulence continued flow rates along the medial floor and middle nasal cavity
throughout the nasal cavity. Most of the flow passed areas. At the postturbinal and nasopharyngeal regions, the
between the middle meatus and the nasal septum above the flow profiles remained uniform but demonstrated low
inferior turbinate at a rate of 120–180 m/min. A lesser velocities owing to the larger total CSA. Expiratory veloc-
amount of air followed the nasal septum and then changed ity flow profiles appeared less uniform and, in general, had
direction, following the floor of the nasal chamber. A small greater maximum velocities than inspiratory flow profiles.
fraction of the main stream broke off and formed a stand- While the majority of the inspiratory flow was shunted
ing eddy in the olfactory region. through the inferior and middle meatuses, most of the
Morgan et al. (1991) used a water-dye siphon system expiratory flow was shunted along the septum. Although
to study nasal airflow in the F344 rat and rhesus monkey. Reynolds numbers were indicative of laminar flow, the
Distribution of the dye differed between species for a flow profiles indicated that turbulent flow predominated
given release point, indicating that a common distribution throughout the nasal cavity. Air velocity in the superior
mechanism does not exist across species. However, the posterior nasal cavity was low during inspiration and high
data suggest that the anterior portion of the nose is impor- during expiration.
tant for directing the incoming airstream, regardless of Hornung et al. (1987) used a vacuum pump to draw
the species. Xenon 133 gas though a plastic nose at 2.5, 7.0, and 20
L/min. A catheter released radioactive gas in four locations
within the nostril and at the initial opening of the nasal
2. Gas Flow Models
chamber. Positioning the catheter in the center of the naris
Proetz (1951) pumped smoke through models of both nor- distributed radioactivity evenly throughout the nose, with
mal and abnormal nasal airways. Inspired air was directed the exception of the superior region, whereas positioning
toward the septum until it reached the nasal vestibule, at the catheter in the ventral medial portion of the nostril
which point it fanned out. The streams reconverged as they increased the radioactivity in the dorsal middle and inferior
approached the choana. The middle turbinate divided the meatuses. The greatest superior region radioactivity was
expiratory airstream in half, with half following the inspi- observed when the catheter was placed in the initial por-
ratory pathway and the other half forming eddies and vor- tion of the nasal chamber. Increasing the flow rate
tices. Nasal turbinates were not important for directing increased radioactivity in the anterior olfactory region.
inspiratory flow, but were essential for directing expiratory Simmen et al. (1999) pumped aerosolized water parti-
flow. Nasal valve region abnormalities changed the direc- cles through an anatomical human model of the choana
tion of the smoke flow. Swelling of the upper two thirds of with a translucent replica of the original nasal septum.
the nose directed smoke flow in a straight path toward the Physiological pressures were simulated. Turbulence was
nasopharynx and prevented the smoke flow from fanning present throughout all flow velocities, with turbulence
out inside the nasal cavity. Septal spurs and middle meatus being most prominent during airflow acceleration and
polyps did not alter normal smoke flow, while superior deceleration and less prominent at near-steady flow. Under
meatus polyps deflected the air currents downward and normal conditions the majority of the main flow stream
away from the olfactory meatus. External deformities passed through the middle meatus at all rates and the olfac-
resulting from the narrowing or collapse of the naris aug- tory region was aerated toward the end of inspiration and
mented the pressure drop across the nose but did not during the entire expiration phase. Simulated hypertrophic
change the pattern of smoke flow. mucosal membranes and turbinates increased the propor-
Girardin et al. (1983) used a laser Doppler device to tion of air passing through the middle meatus, and
measure the velocity of a water aerosol introduced into a turbinate decongestion resulted in more even flow
448 Frye

distribution. Turbinectomy redirected airflow along the

floor of the nose. Although this model demonstrates some
interesting results, the lack of a simulated nasal valve may
have decreased the turbulence and dynamics of the
airstream.
Using a large-scale mock human nasal cavity and a hot
wire anemometer, Scherer et al. (1989) measured nasal air
currents at inspiratory flow rates between 15 and 120 L/min.
At least 50% of the airflow passed through the inferior and
middle meatuses, while 15% passed through the olfactory
region. Air velocity was relatively high through the inferior
and middle meatuses and lower olfactory slit. Simulated
nasal hairs increased turbulence. Although the air velocity
profiles were typical of turbulent flow through the inspira-
tory airflow range, Reynolds numbers were only 400 at
lower flow rates. When this model was studied with steady
physiological flow rates [1,100 (66), 560 (33.6), and 180
(10.8) mL/sec (L/min)], laminar flow predominated,
although moderate turbulence was seen.

3. Computer Simulated Flow

Using a mathematical simulation of incompressible, steady,
laminar flow through three-dimensional nasal cavity repre-
sented by a trapezoid outline and two curved plates as the
inferior and middle turbinates, Elad et al. (1993) showed
that the majority of air flowed along the nasal cavity floor, Figure 2 Nasal model used to calculate concentration field and
while the turbinate structures directed flow in an anterior- odorant mass flux in order to investigate inspired odorant mole-
posterior direction. These researchers suggest that the cule transport and uptake. (Top) Three-dimensional finite ele-
turbinates and the nasal cavity shape are responsible for ment mesh of the right nasal cavity with view of septum and nasal
forcing airflow towards the olfactory region. Keyhani et al. floor. (Left) Slice through the three-dimensional mesh at plane 8.
(1995) solved the steady-state Navier-Stokes and continuity (Right) Midsagittal outline of the model nasal cavity and loca-
equations for an anatomically correct finite element mesh tions of nine coronal planes. Three coronal planes are demon-
designed from a CT scan of a healthy adult nose in order to strated with the olfactory region highlighted with hatch marks.
(From Keyhani et al., 1997.)
determine the laminar airflow patterns in the nasal cavity at
quiet breathing flow rates (Fig. 2). The highest inspiratory
velocities occurred along the nasal floor and between the
inferior and middle turbinates with laminar flow predomi- olfactory region and the amount of particle deposition in
nating as velocity varied between resting breathing rates of the nose, the number of particles reaching the olfactory
125 (7.5)–200 (12) mL/sec (L/min); about 10% of the air- region can be calculated. However, since particle deposition
flow passed into the olfactory cleft. These results were is a function of such variables as airstream velocity, subject
found to be consistent with the results of the large-scale age, and particle size, such calculations are complex.
mock human nasal cavity developed by Scherer et al. (1989)
and Hahn et al. (1993). The lack of representation of the 1. Theoretical Studies
nasal valve in these simulations calls into question the valid- Scott et al. (1978) developed a theoretic geometric model
ity of determination of turbulent flow within the nasal cavity. of the nasal cavity from dimensions reported by Proctor
and Swift (1971) and cadaver specimens. Particle deposi-
B. Particle Deposition and Uptake tion was calculated for five nasal regions: the nostril, the
nasal valve, the expansion area, the nasal cavity area (fur-
Determining the filtering efficiency of the nose may be of ther divided into four regions), and the entrance to the
help in understanding odorant transport. For example, by nasopharynx. Deposition of particles within the first 0.5
knowing the proportion of inspired air shunted to the mm of the nostril was due to nasal hairs, whereas particle
Nasal Patency and Nasal Airflow Aerodynamics 449

deposition in the nasal valve region was found only for Odorant flux decreased along the olfactory slit from ante-
particles that crossed the main flow stream into the bound- rior to posterior and from inferior to superior, with this gra-
ary streamline created by convergence of the lateral wall dient being dependent on odorant mucus solubility. Thus,
toward the nasal septum. Particle deposition in the expan- different odorants generated discernibly different flux pat-
sion region, which represented the highest region of depo- terns across the olfactory mucosa. The nasal valve was
sition in the nose, was due to turbulence and vortices. included in this model, and this model assumed a steady
Deposition within the nasal cavity and nasopharynx flow, thereby reducing the contribution of turbulent flow.
entrance was caused by bending and sudden constrictions The significance of this is not known; however, turbulent
in CSA. A combination of a 15% decrease in the airway currents and vortices in the olfactory silt will probable
CSA and a 15% increase in nostril hair decreased particle influence the odorant flux pattern.
deposition. Changes in particle deposition were attributed
to changes in air velocity. The olfactory fissure was not
3. In Vivo Studies
modeled in this study, but since its anatomical characteris-
tics are similar to those of the expansion region, a substan- In vivo particle deposition measurement indicates the
tial amount of particle deposition would be expected. number of particles available for transport to the olfactory
Although the validity of this model was assessed only region, not the specific deposition location, and is calcu-
by comparing total deposition values to empirical data, lated in one of two ways. Either the ratio of the particle
identification of the expansion area as the major site of concentration inhaled through the nose to the concentra-
major particle deposition is in accord with other reports tion exhaled through the nose is subtracted from a similar
(Fry and Black, 1972). Indeed, areas of large smoke parti- ratio measured at the mouth or an aerosol is drawn into the
cle deposition have been identified just posterior to the nose and out of the mouth during a breath-hold.
nasal valve and in the olfactory fissure (Proetz, 1951). This Becquemin et al. (1991) measured deposition of 1-,
has been attributed to the “impingement effect,” which 2.05-, and 2.8-m particles in the noses of adults, older
causes particle deposition just distal to an abrupt bend or children (12–15 years), and younger children (5.5–11.5
expansion of a tube. years) during rest and exercise. Average inspiratory flow
rates were 27.5, 18.5, and 17.0 L/min during rest and 63.4,
31.3, and 28.0 L/min during exercise for adults, older chil-
2. Computer Simulation
dren, and younger children, respectively. Particle deposi-
Hahn et al. (1994) developed an odorant transport model, tion was significantly lower in younger children than
which included odorant molecule aerodynamic transport adults for two particle sizes during rest and for all sizes
through bulk and lateral flow mechanisms and local odor- during exercise. Particle deposition was significantly
ant molecule movement from the mucosa surface to olfac- lower in older children for the smallest particle during
tory receptor by sorption and diffusion, as well as olfactory exercise. Particle deposition increased with flow rate,
receptor interaction. The model predicted that increasing implicating an inertial impaction rather than a gravita-
the flow rate would increase the perceived odor intensity tional sedimentation deposition mechanisum. This obser-
for highly soluble odorants but decrease odor intensity for vation confirms the conclusion of earlier, less
insoluble odorants. However, if only a limited sorption sur- sophisticated, studies (Landahl and Black, 1947; Landahl
face area is available, perceived odor intensity should and Tracewell, 1949). Since average volume flow rate
decrease for all odorants regardless of the solubility. decreased with age, the filtering efficiency of the child’s
In a more detailed study, Keyhani et al. (1997) investi- nose may be lower due to airflow rate or anatomy, or both.
gated inspired odorant molecule transport and uptake Although deposition may decrease with air velocity, a
using the simulation designed by Keyhani et al. (1995). lower volume airflow rate does not necessarily indicate a
The concentration field and odorant mass flux at the nasal lower velocity across subjects since velocity is dependent
walls was calculated by uncoupled steady convective-dif- on nasal CSA.
fusion equations. Total odorant flux, a measure highly cor- Particle deposition is related to particle density and size
related with perceived odor intensity, was a function of through the formula d 2Q ( is the particle density, d is the
several transport parameters including the odorant solubil- particle diameter, and Q is the volume airflow rate). This
ity and diffusivity of the mucosal lining and the thickness value is equivalent to the Stokes equation and correlates
of the mucus layer. Odorant flux increased with inlet con- well with nasal deposition when it is above 337 g m sec
centration in a nonlinear fashion and depended on odorant for inspiration or 215 g m sec for expiration, suggesting
solubility. For example, relative flux decreased for poorly that diffusional effects are more important than inertial
soluble odorants and increased for highly soluble odorants. effects for submicrometer particles (see Yu et al., 1981).
450 Frye

V. THE NASAL CYCLE proportion of subjects exhibiting a “classic” nasal cycle

decreases with age (Mirza et al., 1997).
A. The Nasal Cycle as an Ultradian Rhythm Nasal turbinate swelling, which causes changing in
NCV and NAR, is controlled by autonomic vascular
A periodic simultaneous change in the volume of each tonicity of the nasal mucosa. Although the medulla may
nasal cavity has been reported in up to 80% of the adult produce the autonomic changes through an N-methyl-
population (Principato and Ozenberger, 1970). Although D-aspartate–mediated system, the production of the under-
this change is classically believed to be opposing across lying rhythm probably originates in the hypothalamus
the two sides of the nose, several variations of this cycle (Eccles, 1978; Galioto et al., 1991; Haxhiu et al., 1987).
exist, and in some patients the direction of volume Sympathetic tonus decreases both turbinate blood flow and
change may be in the same direction on both sides of the size (Malm, 1977) by vasoconstriction of deep capacitance
nose or may only involve one nasal cavity. Alteratively, vessels without altering superficial blood vessel flow
the degree to which the volume and resistance of each (Kurita et al., 1988). Parasympathetic tonus causes oppo-
nasal cavity changes may not be equal, and a dominant site effects, although to a lesser extent (Anggard, 1977;
nasal cavity may result. This ultradian rhythm, called the Haight and Cole, 1986).
nasal cycle, is observed in a number of animals. When The cyclical changes in asymmetrical autonomic tonic-
present, its periodicity reportedly ranges from 40 minutes ity are related to the basic rest-activity cycle (BRAC) of
to 4 hours (Bojsen-Muller and Fahrenkrug, 1971; Eccles, Kleitman (1967). Such diverse phenomena as asymmetrical
1978). hemispheric electroencephalographic (EEG) activity
The criterion for defining the nasal cycle is complex (Wertz et al., 1983), adrenal gland secretion of cate-
and depends on the detection method. For example, cholamines (Kennedy et al., 1986), performance on visual/-
autocorrelation analysis assumes a regular periodic sig- spatial psychological tasks (Klein et al., 1986), and sleep
nal that is consistent in amplitude and frequency. As dis- stage (Frye and Doty, 1992) are associated with the BRAC.
cussed above, nasal cavity volume and turbinate The “rest” phase of the BRAC is associated with propor-
swelling is influenced by many physiological and envi- tionately more right hemispheric integrated EEG activity, a
ronmental factors, and, as described below, the nasal spatial cognitive mode, a parasympathetic tonicity in
cycle is highly influenced by physiological and environ- unpaired organs, sympathetic tonicity on the left side of the
mental stimuli. Thus, the idea of an idealized nasal cycle body, parasympathetic tonicity on the right side of the body,
with equal and consistent right and left nasal cavity a reduction in spontaneous vigilance, and an increase in
volume changes with a consistent frequency is rather napping behavior. Relatively greater airflow through the
naive. It is not surprising that investigators who use strict right nasal chamber is associated with the “activated” phase
criteria (i.e., Gilbert and Rosenwasser, 1987, Mirza et of the BRAC, during which effects opposite to those noted
al., 1997) do not find a nasal cycle in a substantial num- above are observed.
ber of subjects evaluated. The influence of the nasal cycle on NAR is modulated
Total NAR remains relatively constant, while the differ- by many factors. Although emotional state, humidity, tem-
ence in NAR between the two sides of the nose may be as perature, exercise, hyperventilation, anoxia, hypercapnia,
large as 12 cm H2O/L/sec for the normal airway (Cole and environmental pollution, and psychological factors symmet-
Haight, 1986). Although controversy exists regarding the rically alter total NAR, unilateral sensory stimulation asym-
operational definition of the nasal cycle and its prevalence, metrically alters NAR (Cole, 1982). For example, unilateral
much of the discord probably stems from differences in pressure to the axilla (Davies and Eccles, 1985) or lateral
measurement technique (i.e., Flanagan and Eccles, 1997). recumbency (Haight and Cole, 1986) will decrease NAR in
Recent studies using acoustic rhinometry demonstrate that the opposite or superior nasal cavity, respectively. Afferent
the cycle is present, in some form, in a majority of adults pressure receptors in the pelvic and pectoral girdles and the
and in children as young as 3 years. It persists after cessa- deep subcutaneous tissue of the thorax are responsible for the
tion of nasal airflow and occurs independent of structural latter response. Extended lateral recumbency can result in
abnormalities such as septal deviation (Gungor et al., sustained inhibition of the nasal cycle (Haight and Cole,
1999; Lund, 1996; Sung et al., 2000). The prevalence of 1984). Abnormal responsiveness of the nasal turbinates, as
the nasal cycle appears to change with age. For example, seen in nasal sinus disease (e.g., allergic rhinitis, sinusitis)
Fisher et al. (1995) found a “classical” reciprocal alternat- (Hilberg et al., 1995, Ophir et al., 1988), vasomotor rhinitis
ing pattern in 80%, an “in concert” pattern in 7%, and an (Kuening, 1968), and multiple chemical sensitivities (Doty et
“irregular” pattern in 13% of 15 healthy children ranging al., 1988), may influence the nasal cycle. In addition,
in age from 3 to 10 years. There is some evidence that the mucocilliary clearance and reactivity to nasal allergen provo-
Nasal Patency and Nasal Airflow Aerodynamics 451

cation is enhanced in the congested nasal chamber (Brooks centration was determined by a unilateral olfactory
et al., 1991; Littlejohn et al., 1992). Furthermore, the threshold test preceding the study; the contralateral nos-
extremes of the nasal cycle can adversely affect patients with tril was blocked during odor sampling. The nasal cycle
obstructed airways by further increasing NAR. and unilateral nasal resistance were also measured. The
left dominant phase of the nasal cycle was associated
B. The Nasal Cycle and Olfactory Function with an increased sensitivity and decreased subject
response bias for odors presented to the right and left
From a theoretical perspective, the nasal cycle may influ- sides of the nose, respectively. Since asymmetrical
ence olfaction, on at least one side of the nose, in four changes in integrated hemispheric EEG activity are
ways. First, a change in turbinate size alters airstream correlated with the nasal cycle and the majority of the
velocity and distribution, thereby altering odorant access olfactory bulb’s afferents project unilaterally, interhemi-
to the olfactory epithelium. Second, changes in autonomic spheric alteration in decision processing associated with
tonicity of the mucosa can alter penetration or concentra- olfactory recognition may explain this finding.
tion of the odorant molecules reaching the olfactory
receptors by changing the quantity and consistency of
VI. NASAL AIRFLOW PERCEPTION
nasal secretions. Third, fluctuations in central arousal
mechanism may alter olfactory function, especially since
A. Perception of Changes in Nasal Patency
the locus coeruleus has direct connections to the anterior
olfactory nucleus and olfactory bulb. Finally, if it is
Self-reported perceived changes in and measurements of
assumed that olfactory processing is localized, at least to
nasal patency correlate following large modifications in
some degree, to the right hemisphere, variations in relative
nasal lumen size. For example, septoplasty (Broms et al.,
hemispheric EEG activity correlated with the nasal cycle
1982; Larsen and Kristensen, 1990), inferior turbinate
could alter odor information processing and perception. submucosal diathermy (Jones et al., 1985a), and aerosol
To determine if a relationship exists between the nasal steroid treatment of seasonal allergic rhinnitis and severe
cycle and olfactory sensitivity, Frye and Doty (1992) mea- polyposis (Lancer et al., 1987, Lund et al., 1998)
sured unilateral 2-butanone olfactory thresholds and NAR improve both perceived and measured nasal patency.
for 33 men and 44 women in two sessions separated by 4 Changes in perceived nasal patency induced by
hours. In approximately half of the subjects the nostril not histamine or nasal allergen challenge are reliably corre-
sampling the odorant was occluded. If the nasal cycle had lated with NAR and rhinostereometry, but not MCA
not spontaneously changed its phase at the beginning of measurements (Graf, 1996; Lane et al., 1996; Zweiman
the second session, an attempt was made to change the et al., 1997). However, the perceived change in nasal
phase by applying pressure under the armpit, in the palm patency after decongestion with 1% phenylephrine did
of the hand, or by auditory stimulation on the side of the not correlate with changes in MCA or NAR (Kim et al.,
more patent airway. In subjects whose contralateral naris 1998). Although symptom score does not correlate with
was blocked, low right NAR was associated with NAR (Hardcastle et al., 1988a,b), perceived nasal
decreased olfactory thresholds on both sides of the nose, patency does correlate with preoperative, but not postop-
whereas low left NAR was associated with comparatively erative, NAR (Gordon et al., 1989). The inspiratory
increased olfactory thresholds on both sides of the nose. turbulent parameter of Rohrer’s equation was found to
Thus, augmentation of olfactory sensitivity was associated be related to the severity of nasal obstructive symptoms
with increased arousal and greater left hemispheric inte- in 75 patients (Naito et al., 1995).
grated EEG activity, and olfactory sensitivity attenuation The poor relationship between perceived and measured
was associated with decreased arousal and greater right nasal patency in normal subjects may be due to large inter-
hemispheric integrated EEG activity. These data suggest subject variability. For example, individual correlations
that NAR, per se, was not responsible of alterations in between daily nasal inspiratory peak flow rate and perceived
olfactory sensitivity and that cerebral EEG or central nasal patency were strong despite significant differences in
arousal mechanisms likely are involved. individual regression lines (Fairley et al., 1993). Although
To better define this relationship, NAR was measured an overall relationship between perceived nasal patency and
in eight right-handed male subjects every 15 minutes MCA or NCV was not found in subjects with nasal obstruc-
over 6 hours (R. E. Frye, A. Valle, and R. L. Doty, tion, these variables were correlated when 10 new subjects
unpublished data). Olfactory sensitivity to phenyl ethyl were studied on an individual level (Tai et al., 1998).
alcohol and subject response bias was measured using a Unilateral measurements and higher baseline NAR
signal detection paradigm. The perithreshold odor con- improve the strength of the relationship between perceived
452 Frye

and measured nasal patency. For example, the correlation needles, and methyl salicylate) but not vanilla produced
between NAR and perceived nasal patency in normal sub- significantly higher perceived nasal patency ratings.
jects was improved when unilateral, rather than total nasal L-Menthol lozenges also increased airflow sensation
airflow, was evaluated (Sipila et al., 1995) or when only (Eccles et al., 1990; Naito et al., 1991). However, this
the side of the nose with the greater NAR was studied effect was not induced by D-menthol, D-isomethol, or
(Hirschberg and Rezek, 1998). A correlation between uni- D-neomenthol vapors (Eccles et al., 1988). Simulation of
lateral perceived nasal patency and pre- and postdeconges- trigeminal nasal flow receptors or interference with
tion MCA and NAR was found in symptomatic patients trigeminal thermoreceptor calcium conductance may cause
(Roithmann et al., 1994). these findings since L-menthol is a trigeminal stimulant.
Small variations in measurable nasal patency changes are
not reliably perceived in normal subjects. NAR changes
induced by aspirin or allowed to occur spontaneously over a VII. CONCLUSIONS: AIRFLOW DYSNAMICS
long or short observation period (i.e., single observation to IN RELATION TO OLFACTION
6 weeks) do not correlate with perceived patency changes
(Jones et al., 1985b; Jones et al., 1989a). Gungor et al. The contribution of airflow within the olfactory cleft to
(1999) found that nasal cycle related fluctuations in CSA olfaction has been considered by a large number of inves-
and NCV did not correlate with perceived nasal patency. tigators. Unfortunately, many studies have significant
design limitations and provide only limited insight into the
B. Nasal Flow Receptors mechanisms involved. Nonetheless, a window of elucida-
tion for some basic principles in this area is provided.
Nasal flow receptors mediate the sensation of nasal airflow. Incoming air currents are directed by the nasal
These receptors are located in both the nasal vestibule and vestibule. Turbulent flow is important for both mixing of
the nasal chamber, although the receptors at each location odorant molecules and proper humidification and filter-
convey a different sensation. Local anesthesia of the nasal ing of the incoming air. Although Reynolds numbers are
vestibule receptors produces a sensation of nasal obstruction not indicative of turbulent flow, such flow develops in the
(Jones et al., 1987), whereas local anesthesia of the nasal expansion region during inspiration and remains through-
chamber receptors results in a sensation of increased nasal out the nose. Disorders within this region may alter vor-
patency (Jones et al., 1986). Nasal chamber flow receptors tices and turbulent flow by redirection of airstreams and
are likely localized in the middle and posterior inferior augmenting nasal resistance. However, turbulence air-
turbinate mucosa (Wight et al., 1988b). Nasal patency recep- flow prior to reaching the olfactory cleft may not be par-
tors are probably predominant since local anesthesia of both ticularly important since secondary air currents and
the nasal vestibule and chamber produce a sensation of nasal vortices are probably produced in the olfactory cleft dur-
obstruction (Jones et al., 1989b). Indeed, the sensitivity of the ing inspiration, thereby creating a standing eddy and pro-
nose to an air jet pulse and to temperature changes is great- moting odorant mixing and inertial deposition. Although
est at the nasal vestibule (Clarke and Jones, 1992, 1994). the particular portion of the naris responsable for direct-
Trigeminal nerve afferents have been implicated in the ing particles to the olfactory region is controversial, it is
cat (Davis and Eccles, 1987) and rat (Tsubone, 1989). In known that the expiratory airstream exhibits higher
the human, the palatine nerve (a branch of V2) innervates velocities and a more direct path to the olfactory regions
the turbinates and nasal septum, while the ethmoid nerve than the inspiratory airstream (Girardin et al., 1983).
(a branch of V1) innervates the vestibule. It is not known Inspiratory airflow velocity and volume flow corre-
whether the different receptors causes different sensations; lates with particle deposition and the proportion of parti-
however, thermoreceptors may mediate airflow sensation cles transported to the olfactory region. Humans will
since mucosal temperature, as measured by noncontact adjust duration of a sniff to optimize odorant perception
infrared thermometry, is correlated with subjective nasal in a low-flow nostril (Sobel et al., 2000). Indeed, in
patency (Willatt, 1993; Willatt and Jones, 1996). humans, airflow rate is positively correlated with odor
intensity (Rehn, 1978), the number of odorants identified
in a confusion matrix (Schwartz et al., 1987), and the
C. Influence of Odors on Airflow Sensation magnitude of the olfactory evoked potential (Kobal and
Hummel, 1991).
Eccles et al. (1987) demonstrated that particular odors In the clinical setting, a variety of mechanisms related
influence nasal airflow perception without changing NAR. to diseases of the nasal cavity can alter olfaction.
An aromatic combination (menthol, camphor, oil of pine Conductive airflow abnormalities definitely alter airflow
Nasal Patency and Nasal Airflow Aerodynamics 453

patterns. In some cases, endoscopic operative procedures, REFERENCES

septoplasty, turbinoplasty, or adenoidectomy can improve
olfactory function. However, many patients do not experi- Adamson, J. E. (1987). Constriction of the internal nasal valve in
ence restoration of olfactory function following interven- rhinoplasty: treatment and precention. Ann. Plast. Surg.
tion. Nasal sinus disease associated with swelling of the 18:114–121.
superior nasal cavity likely obstructs airflow to the olfac- Anggard, A. (1977). Parasympathetic influence on the nasal
tory region, while disorders associated with the inferior mucosa. Acta Orolaryngol. 83:22–24.
turbinate or nasal septum presumably redistribute the Austin, C. E., and Foreman, J. C. (1994). Acoustic rhinometry
airstream within the nasal cavity. Local changes in the vol- compared with posterior rhinomanometry in the measurement
ume of the olfactory meatus more than 1 cm inferior to the of histamine- and bradykinin-induced changes in nasal airway
cribriform plate influence olfactory function, and polyps patency. Br. J. Clin. Pharm. 37:33–37.
located in the superior meatus can deflect incoming air Becquemin, M. H., Swift, D. L., Bouchikhi, A., Roy, M., and
currents away from the olfactory meatus. Since only a Teillac, A. (1991). Particle deposition and resistance in the
small percentage of air flows through this region, abnor- noses of adults and children. Eur. Respir. J. 4: 694–702.
malities in the olfactory region most likely have little, if Berkinshaw, E. R. Spalding, P. M., and Vig, P. S. (1987). The
any, influence on nasal resistance, while profoundly affect- effect of methodology in the determination of nasal reistance.
ing the proper delivery of odorant to the olfactory cleft. Am. J. Orthod. Dentofac. Orthop. 92:329–335.
Airflow obstruction only accounts for a subset of alter- Bickford, L., Shakib, S., and Taverner, D. (1999). The nasal air-
ations in olfactory perception. Many other nonobstructive ways response in normal subjects to oxymetazoline spray:
mechanisms may result in anosmia or dysosmia. For randomized double-blind placebo-controlled trial. Br. J. Clin.
example, infiltration of inflammatory cells into the olfac-
Pharmacol. 48:53–56.
tory mucosa may alter function of the olfactory receptors.
Bojsen-Muller, F., and Fahrenkrug, J. (1971). Nasal swell-bodies
Although direct destruction of olfactory receptors by
and cyclic changes in the air passages of the rat and rabbit
inflammatory cells has not been shown, patients with
nose. J. Anat. 110:25–27.
inflammatory changes in the olfactory mucosa are at risk
Bridger, G. P., and Proctor, D. F. (1970). Maximal nasal inspira-
for olfactory deficits. In addition, changes in mucosa thick-
tory flow and nasal resistance. Ann. Otol. 79:481–488
ness and mucus consistency as a result of inflammation
Broms, P., Jonson, B., and Malm, L. (1982). A pre- and post-
can alter odorant flux.
operative evaluation in functional septoplasty. Acta
Changes in olfactory mucosa topography may change
Otolaryngol. 94:523–529.
olfactory perception as well as intensity. For example, a
Brooks, C. D., Karl, K. J., and Francom, S. F. (1991).
limited odorant sorption surface area will reduce, rather
Unilaterality of obstruction after acute nasal allergen provo-
than increase, odor intensity as airflow increases (Hahn
cation. Relation of allergic dose, nasal reactivity and the nasal
et al., 1994). Thus, decreased available olfactory mucosa
will paradoxically decrease olfactory sensation as a cycle. Clin. Exp. Allergy 21:583–587.
patient attempts to use adaptive mechanisms of odor Chaban, R., Cole, P., and Naito, K. (1988). Simulated septal devi-
exploration. Odorant flux along the olfactory slit in an ations. Arch. Otolaryngol. Head Neck Surg. 114:413–415
anterior to posterior and inferior to superior direction is Chaudhry, M. R.,. Akhtar, S., and Dwalsaint, F. (1996).
unique to odorants of different solubility (Keyhani et al., Rhinomanometric evaluation of the improved mechanical
1997). These flux patterns produce a unique odor signa- therapeutic nasal dilator in patients with anterior nasal
ture and contribute to odor identification. Destruction of obstruction. Rhinology 34:32–34.
portions of the olfactory mucosa can change this pattern Cho, J. H., Lee, D. H., Lee, N. S., Won, Y. S., Yoon, H. R., and
identification in several ways. For example, a punched- Suh, B. D. (1999). Size assessment of adenoid and nasopha-
out pattern of olfactory receptor dysfunction randomly ryngeal airway by acoustic rhinometry in children.
disrupts identification of particular portions of the pat- J. Laryngol. Otol. 113:899–905
tern, whereas olfactory receptor dysfunction in the ante- Clarke, R. W., and Jones, A. S. (1992). Nasal airflow receptors:
rior inferior portion of the olfactory slit decreases odor the relative importance of temperature and tactile stimulation.
perception at the point of maximum flux. Random dis- Clin. Otolaryngol. 17:388–392.
ruption of olfactory receptor function could result in per- Clarke, R. W., and Jones, A. S. (1994). The distribution of nasal
ception of a nonunique pattern resulting in a dysosmia, airflow sensitivity in normal subjects. J. Laryngol. Otol.
whereas disruption of olfactory receptor function at the 108:1045–1047.
point of maximum flux may completely prevent odorant Cole, P. (1982). Upper respiratory airflow. In The Nose: Upper
perception. Airway Physiology and the Atmospheric Environment, D. F.
454 Frye

Proctor and I. B. Andersen (Eds.). Elsevier Biomedical Press, Ebbell, B. (1937). The Papyrus Ebers. Oxford University Press,
Amsterdam, pp. 163–189. London.
Cole, P., and Haight, J. S. (1986). Posture and the nasal cycle. Eccles, R. (1978). The central rhythm of the nasal cycle. Acta
Ann. Otol. Rhinol. Laryngol. 95:233–237. Otolaryngol. 86:464–468.
Cole, P. Haight, J. S. J., Love, L., and Oprysk, D. (1985). Eccles, R., Lancashire, B., and Tolley, N. S. (1987). Experimental
Dynamic components of nasal resistance. Am. Rev. Respir. studies on nasal sensation of airflow. Acta Otolaryngol.
Dis. 132:1229–1232. 103:303–306.
Cole, P., Chaban, R., Naito, K., and Oprysk, D. (1988). The Eccles, R., Griffiths, D. H., Newton, C. G., and Tolley, N. S.
obstructive nasal septum: effect of simulated septal deviations (1988). The effect of menthol isomers on nasal sensation to
on nasal airflow resistance. Arch. Otolaryngol. Head Neck airflow. Clin. Otolaryngol. 13:25–29.
Surg. 114:410–412.
Eccles, R., Jawad, M. S. M., and Morris, S. (1989). Olfactory and
Conrad, W. A. (1969). Pressure-flow relations in collapsible
trigeminal thresholds and nasal resistance to airflow. Acta
tubes. IEEE Trans. Bio. Med. Eng. 16:284–295.
Otolaryngol. 108:268–273.
Corey, J. P., Nalbone, V. P., and Ng, B. A. (1999). Anatomic
Eccles, R., Jawad, M. S., and Morris, S. (1990). The effects of
correlates of acoustic rhinometry as measured by rigid
oral administration of (-)-menthol on nasal resistance to air-
nasal endoscopy. Otolaryngol. Head Neck Surg. 121:
flow and nasal sensation of airflow in subjects suffering from
572–576.
Dastidar, P., Heinonen, T., Numminen, J., Rautiainen, M., and nasal congestion associated with the common cold. J. Pharm.
Laasonen, E. (1999a). Semi-automatic segmentation of com- Pharmacol. 42:652–654.
puted tomographic images in volumetric estimation of nasal Elad, D., Liebenthal, R., Wenig, B. L., and Einav, S. (1993).
airway. Eur. Arch. Otorhinolaryngol. 256:192–198. Analysis of airflow patterns in the human nose. Med. Biol.
Dastidar, P., Numminen, J., Heinonen, T., Ryymin, P., Eng. Comput. 31:585–592
Rautiainen, M., and Laasonen, E. (1999b). Nasal airway vol- Elwany, S., and Harrison, R. (1990). Inferior turbinectomy: com-
umetric measurement using segmented HRCT images and parison of four techniques. J. Laryngol. Otol. 104:206–209.
acoustic rhinometry. Am. J. Rhinol. 13:97–103 Fairley, J. W., Durham, L. H., and Ell, S. R. (1993). Correlation
Davies, A. M., and Eccles, R. (1985). Reciprocal changes in nasal of subjective sensation of nasal patency with nasal inspiratory
resistance to airflow caused by pressure applied to the axilla. peak flow rate. Clin. Otolaryngol. 18:19–22.
Acta Otolaryngol. 99:154–159. Fisher, E. W., Palmer, C. R., and Lund, V. J. (1995). Monitoring
Davies, A. M., and Eccles, R. (1987). Electromyographic fluctuations in nasal patency in children: acoustic rhinome-
response of a nasal muscle to stimulation of the nasal try versus rhinohygrometry. J. Laryngol. Otol.
vestibule in the cat. J. Physiol. 291:25–38. 109:503–508.
Deems, D. A., Doty, R. L., Settle, R. G., Moore-Gillon, V., Flanagan, P. and Eccles, R. (1997). Spontaneous changes of uni-
Shaman, P., Mester, A. F., Kimmelman, C. P., Brightman, V. lateral nasal airflow in man. A re-examination of the ‘nasal
J., and Snow, J. J. Jr. (1991). Smell and taste disorders, a study cycle’. Acta Otolaryngol. 117:590–5.
of 750 patient from the University of Pennsylvania Smell and Franciscus, R. G., and Trinkaus, E. (1988). Nasal morphology
Taste Center. Arch. Otolaryngol. Head Neck Surg. 117:
and the emergence of Homo erectus. Am. J. Phys. Anthropol.
519–528.
75:517–527.
Dinis, P. B., Haider, H., and Gomes, A. (1997). Rhinomanometry,
Franke, G. (1894). Experimentelle Untersuchungen über
sinus CT-scan and allergy testing in the diagnostic assessment
Luftbewegung, und Luftwechsel in der Nase und ihren
of chronic nasal obstruction. Rhinology 35:158–160.
Nebenhöhlen. Arch. Laryngol. 1:230–249.
Doty, R. L., and Frye, R. E. (1989). Influence of nasal obstruc-
Fry, F. A., and Black, A. (1973). Regional deposition and clear-
tion on smell function. Otolaryngol. Clin. North Am.
22:397–411. ance of particles in human nose. J. Aerosol Sci. 4:113–124.
Doty, R. L., Deems, D. A., Frye, R. E., Pelberg, R., and Shapiro, Frye, R. E., and Doty, R. L. (1992). The influence of ultradian
A. (1988). Olfactory sensitivity, nasal resistance, and auto- autonomic rhythms as indexed by the nasal cycle on unilateral
nomic function in patients with multiple chemical sensitivi- olfactory thresholds. In Chemical Signals in Vertebrates VI, R.
ties. Arch. Otolaryngol. Head Neck Surg. 114:1422–1427. L. Doty and D. Muller-Schwarze (Eds.) Plenum Press, New
Doty, R. L., and Mishra, A. (2001). Influence of nasal obstruc- York, pp. 595–598.
tion, rhinitis, and rhinosinusitis on the ability to smell. Galioto, G., Mevio, E., Galioto, P., Fornasari, G., Cisternino, M.,
Laryngoscope 111:409–423. and Fraietta, L. (1991). Modification of the nasal cycle in
Frye, R. E. (1989). Influence of nasal obstruction on smell func- patients with hypothalamic disorders: Kallmann’s syndrome.
tion. Otolaryngol. Clin. North Am. 22:397–411. Ann. Otol. Rhinol. Laryngol. 100:559–562.
Nasal Patency and Nasal Airflow Aerodynamics 455

Ghorbanian, S. N., Paradise, J. L., and Doty, R. L. (1983). Odor Hahn, I., Scherer, P. W., and Mozell, M. M. (1993). Velocity pro-
perception in children in relation to nasal obstruction. files measured for airflow through a large-scale model of the
Pediatrics 72:510–516. human nasal cavity. J. Appl. Physiol. 75:2273–2287.
Gilbert, A. N. and Rosenwasser, A. M. (1987). Biological rhyth- Hahn, I., Scherer, P. W., and Mozell, M. M. (1994). A mass trans-
micity of nasal patency: a re-examination of the nasal cycle. port model of olfaction. J. Theor. Biol. 167:115–128.
Acta Otolaryngological 104:180–186. Haight, J. S. J., and Cole, P. (1984). Reciprocating nasal airflow
Girardin, M., Bilgen, E., and Arbour, P. (1983). Experimental resistance. Acta Orolaryngol. 97:93–98.
study of velocity fields in a human nasal fossa by laser Haight, J. S. J., and Cole, P. (1986). Unilateral nasal resistance
anemometry. Ann. Otol. Rhinol. Laryngol. 92:231–236. and asymmetrical body pressure. J. Otolaryngol. S16:1–31.
Goode, R. L. (1985). Surgery of the incompetent nasal valve. Haight, J. S. J., Cole, P., Cooper, P. W., and Kassel, R. R. (1985).
Laryngoscope 95:546–555. The choana and nasal obstruction. J. Otolaryngol. 14: 99–102.
Gordon, A. S. D., McCaffrey, T. V., Kern, E. B., and Pallanch, Hardcastle, P. F., White, A., and Prescott, R. J. (1988a). Clinical
J. F. (1989). Rhinomanometry for preoperative and postoper- and rhinometric assessment of the nasal airway–do they mea-
ative assessment of nasal obstruction. Otolaryngol. Head sure the same entity? Clin. Otolaryngol. 13:185–191.
Neck Surg. 101:20–26. Hardcastle, P. F., White, A., and Prescott, R. J. (1988b). Clinical
Gosepath, J., Mann, W. J., and Amedee, R. G. (1997). Effects of or rhinomanometric assessment of the nasal airway—which is
the Breathe Right nasal strips on nasal ventilation. Am. better? Clin. Otolaryngol. 13:381–385.
J. Rhinol. 11:399–402. Haxhiu, M. A., Strohl, K. P., Norcia, M. P., Van Lunteren, E.,
Gosepath, J., Amedee, R. G., Romantschuck, S., and Mann, W. J. Deal, R. J., Jr., and Cherniack, N. S. (1987). A role for the
(1999). Breathe Right nasal strips and the respiratory distur- ventral surface of the medulla in regulation of nasal resis-
bance index in sleep related breathing disorders. Am. tance. Am. J. Physiol. 253:R494–R500.
J. Rhinol. 13:385–389. Hilberg, O. (1995). Effect of terfenadine and budesonide on nasal
Graf, P., and Hallen, H. (1996). Clinical and rhinostereometric symptoms, olfaction and nasal airway patency following
assessment of nasal mucosal swelling during histamine chal- allergen challenge. Allergy 50:683–688.
lenge. Clin. Otolaryngol. All. Sci. 21:72–75. Hilberg, O., Grymer, L. F., and Pedersen, O. F. (1995).
Graf, P., and Juto, J. E. (1994). Correlation between objective Spontaneous variations in congestion of the nasal mucosa.
nasal mucosal swelling and estimated stuffiness during long- Ann. Allergy Asthma Immunol. 74:516–521.
term use of vasoconstrictors. J. Oto-Rhino-Laryngol. Rel. Hirschberg, A., and Rezek, O. (1998). Correlation between
Spec. 56:334–339. objective and subjective assessments of nasal patency.
Graf, P., Enerdal, J., and Hallen, H. (1999a). Ten days’ use of J. Otorhinolaryngol. Relat. Spec. 60:206–211.
oxymetazoline nasal spray with or without benzalkonium Hochban, W., Althoff, H., and Ziegler, A. (1999). Nasal decon-
chloride in patients with vasomotor rhinitis. Arch. gestion with imidazoline derivatives: acoustic rhinometry
Otolaryngol. Head Neck Surg. 125:1128–1132. measurements. Eur. J. Clin. Pharmacol. 55:7–12.
Graf, P., Toll, K., Palm, J., and Hallen, H. (1999b). Effects of sus- Holt, J. P. (1969). Flow through collapsible tubes and through in
tained-release oral phenylpropanolamine on the nasal mucosa situ veins. IEEE Trans. Bio. Med. Eng. 16:274–283.
of healthy subjects. Acta Otolaryngol. 119:837–842. Horak, F., Toth, J., Jager, S., and Berger, U. (1993). Effects of
Griffin, J. W., Hunter, G., Ferguson, D., and Sillers, M. J. (1997). H1-receptor antagonists on nasal obstruction in atopic
Physiologic effects of an external nasal dilator. Laryngoscope patients. Allergy 48:226–229.
107:1235–1238. Hornung, D. E., and Leopold, D. A. (1999). Relationship
Grymer, L. F., Illum, P., and Hilberg, O. (1996). Bilateral inferior between uninasal anatomy and uninasal olfactory ability.
turbinoplasty in chronic nasal obstruction. Rhinology Arch. Otolaryngol. Head Neck Surg. 125:53–58.
34:50–53. Hornung, D. E., Leopold, D. A. Youngentob, S. L., Sheebe, P. R.,
Grymer, L. F., Hilberg, O., and Pedersen, O. F. (1997). Prediction Gagne, G. M., Thomas, D., and Mozell, M. M. (1987).
of nasal obstruction based on clinical examination and Airflow patterns in a human nasal model. Arch. Otolaryngol.
acoustic rhinometry. Rhinology 35:53–57. Head Neck Surg. 113:169–172.
Guillette, B. J., and Perry, C. J. (1990). Use of nasal valve stent Hummel, T., Rothbauer, C., Barz, S., Grosser, K., Pauli, E., and
with anterior rhinomanometry to quantitate nasal valve Kobal, G. (1998a). Olfactory function in acute rhinitis. Ann.
obstruction. Ann. Oto. Rhino. Laryngol. 99:175–178. NY Acad. Sci. 855:616–624.
Gungor, A., Moinuddin, R., Nelson, R. H., and Corey, J. P. Hummel, T., Rothbauer, C., Pauli, E., and Kobal, G. (1998b).
(1999). Detection of the nasal cycle with acoustic rhinometry: Effects of the nasal decongestant oxymetazoline on human
techniques and applications. Otolaryngol. Head Neck Surg. olfactory and intranasal trigeminal function in acute rhinitis.
120:238–247. Eur. J. Clin. Pharmacol. 54:521–528.
456 Frye

Illum, P. (1997). Septoplasty and compensatory inferior turbinate Keyhani, K., Scherer, P. W., and Mozell, M. M. (1995).
hypertrophy: long-term results after randomized turbinoplasty. Numerical simulation of airflow in the human nasal cavity.
Eur. Arch. Otorhinolaryngol. Suppl. 1: S89–92. J. Biomech. Eng. 117:429–441.
Jacobs, J. R., and Dickson, C. B. (1986). Effects of nasal and Keyhani, K., Scherer, P. W., and Mozell, M. M. (1997). A numer-
laryngeal stimulation upon peripheral lung function. ical model of nasal odorant transport for the analysis of
human olfaction. J. Theor. Biol. 186:279–301.
Otolaryngol. Head Neck Surg. 95:298–302.
Kim, C. S., Moon, B. K., Jung, D. H., and Min, Y. G. (1998).
Jones, A. S., Lancer, J. M., Moir, A. A., and Stevens, J. C.
Correlation between nasal obstruction symptoms and objec-
(1985a). The effect of submucosal diathermy to the inferior tive parameters of acoustic rhinometry and rhinomanometry.
turbinate on nasal resistance to airflow in allergic and vaso- Auris Nasus Larynx 25:45–48.
motor rhinitis. Clin. Otolaryngol. 10:249–252. Kimmelman, C. P. (1989). The problem of nasal obstruction.
Jones, A. S., Lancer, J. M., Moir, A. A., and Stevens, J. C. Otolaryngol. Clin. North Am. 22:253–264.
(1985b). Effect of aspiring on nasal resistance to airflow. Br. Kittle, G., and Waller, G. (1973). Small improving effect of
Med. J. 290:1171–1173. Cottle’s septum operation. Z. Laryngol. Rhinol. Otol.
Jones, A. S., Lancer, J. M., Shone, G. R., and Stevens, J. C. Grenzgeg. 52:280–284.
(1986). The effect of lignocaine on nasal resistance and nasal Klein, R., Pilon, D., Prosser, S., and Shannahoff-Khalsa, D.
sensation to airflow. Acta Otolaryngol. 101:328–330. (1986). Nasal airflow asymmetries and human performance.
Biol. Psych. 23:127–137.
Jones, A. S., Crosher, R., Wight, R. G., Lancer, J. M., and
Kleitman, N. (1967). Basic rest activity cycle in relationship to
Beckingham, E. (1987). The effect of local anesthesia of the
sleep and wakefullness. In Sleep: Physiology and Pathology,
nasal vestibule on nasal sensation to airflow and nasal resis- A. Kales (Ed.). Lippincott, Philadelphia, pp. 33–38.
tance. Clin. Otolaryngol. 12:461–464. Kobal, G., and Hummel, T. (1991). Human electro-olfactograms
Jones, A. S., Wight, R. G., Stevens, J. C., Phil, D., and and brain responses to olfactory stimulation. In The Human
Beckingham, E. (1988). The nasal valve: a physiological and Sense of Smell, D. G. Laing, R. L. Doty, and W. Breipohl
clinical study. J. Laryngol. Otol. 102:1089–1094. (Eds.). Springer-Verlag, Berlin, pp. 135–150.
Jones, A. S., Willatt, D. J., and Durham, L. M. (1989a). Nasal air- Kuening, J. (1968). On the nasal cycle. Int. Rhinol. 6:99–136.
flow: resistance and sensation. J. Laryngol. Oto. 103:909–911. Kunkel, M., and Hochban, W. (1994). Acoustic rhinometry: a
Jones, A. S., Wight, R. G., Crosher, R., and Durham, L. H. new diagnostic procedure—experimental and clinical experi-
(1989b). Nasal sensation of airflow following blockade of the ence. Int. J. Oral Maxillofacial Surg. 23:409–412.
nasal trigeminal afferents. Clin. Otolaryngol. 14:285–289. Kurita, N. Hasegawa, M., Ohki, M., and Watabane, I. (1988).
Jones, J. G., Clarke, S. W., and Oliver, D. R. (1969). Two-phase Nasal resistance and nasal blood flow in postural changes.
gas-liquid flow in airways. Br. J. Anaesth. 41:192–193. Acta Otolaryngol. 106:448–452.
Kamami, Y. V. (1997). Laser-assisted outpatient septoplasty Lamblin, C., Brichet, A., Perez, T., Darras, J., Tonnel, A. B., and
results on 120 patients. J. Clin. Laser Med. Surg. 15:123–129. Wallaert, B. (2000). Long-term follow-up of pulmonary func-
Kamami, Y. V., Pandraud, L., and Bougara, A. (2000). Laser- tion in patients with nasal polyposis. Am. J. Respir. Crit. Care
assisted outpatient septoplasty: results in 703 patients. Med. 161:406–413.
Otolaryngol. Head Neck Surg. 122:445–449. Lancer, J. M., Jones, A. S., Stevens, J. C., and Beckingham, E.
Kasperbauer, J. L., and Kern, E. B. (1987). Nasal valve physiol- (1987). A comparison by rhinomanometry of beclomethasone
ogy. Otolaryngol. Clin. North Am. 20:699–719. and tergenadine in the treatment of seasonal rhinitis.
Kayser, R. (1895). Die exacte Messung der Luftdurchgängigkeit J. Laryngol. Otol. 101:350–354.
der Nase. Arch. Laryngol. 3:101–120. Landahl, H. D., and Black, S. (1947). Penetration of airborne particu-
Keck, T., Leiacker, R., Riechelmann, H., and Rettinger, G. late through the human nose. J. Ind. Hyg. Toxicol. 29: 269–277.
(2000). Temperature profile in the nasal cavity. Laryngoscope Landahl, H. D., and Tracewell, T. (1949). Penetration of airborne
110:651–654. particulate through the human nose II. J. Ind. Hyg. Toxicol.
Keles, N., Ilicali, O. C., and Deger, K. (1998). Objective and sub- 31:55–59.
jective assessment of nasal obstruction in patients undergoing Lane, A. P., Zweiman, B., Lanza, D. C., Swift, D., Doty, R.,
endoscopic sinus surgery. Am. J. Rhino. 12:307–309. Dhong, H. J., and Kennedy, D. W. (1996). Acoustic rhinome-
Kennedy, B., Ziegler, M. G., and Shannahoff-Khalsa, D. S. try in the study of the acute nasal allergic response. Ann. Oto.
(1986). Alternating lateralization of plasma catecholamines Rhino. Laryngol. 105:811–888.
and nasal patency in humans. Life Sci. 38:1203–1214. Larsen, K., and Kristensen, S. (1990). Peak flow nasal patency
Kern, R. C. (2000). Chronic sinusitis and anosmia: pathologic indices and self-assessment in septoplasty. Clin. Otolaryngol.
changes in the olfactory mucosa. Laryngoscope 110:
All. Sci. 15:327–334.
1071–1077.
Kesavanathan, J., Swift, D. L., and Bascom, R. (1995). Nasal Lavie, P. (1987). Rediscovering the importance of nasal breathing
pressure-volume relationships determined with acoustic rhi- in sleep, or shut your mouth and save your sleep. J. Laryngol.
nometry. J. Appl. Physiol. 79:547–553. Otol. 101:558–563.
Nasal Patency and Nasal Airflow Aerodynamics 457

Leopold, D. A. (1986). The relation between nasal anatomy and monary disease patients exercising with noseclip.mouthpiece.
function. In Clinical Measurement of Taste and Smell, H. L. Am. J. Cardiol. 64:1180–1184.
Meisekman and R. S. Rivlin (Eds.). Macmillan, New York, Mostafa, B. E. (1997). Detection of adenoidal hypertrophy using
pp. 529–549. acoustic rhinomanometry. Eur. Arch. Otorhinolaryngol.
Leopold, D. A. (1988). The relationship between nasal anatomy Suppl 1:S27–29.
and human olfaction. Laryngoscope 98:1232–1238. Mott, A. E., and Leopold, D. A. (1991). Disorders in taste and
Levine, S. C., Levine, H., Jacobs, G., and Kasick, J. (1986). smell. Med. Clin. North Am. 75:1321–1353.
A technique to model the nasal airway for aerodynamic study. Mygind, N., Dahl, R., Nielsen, L. P., and Hilberg, O., and Bjerke,
Otolaryngol. Head Neck Surg. 95:442–449. T. (1997). Effect of corticosteroids on nasal blockage in rhini-
Leyton, L. (1975). Fluid Behavior in Biological Systems. tis measured by objective methods. Allergy 52(40 Suppl):
Clarendon Press, Oxford. 39–44.
Littlejohn, M. C., Stiernburg, C. M., Hokanson, J. A., Quinn, F. Naito, K., Ohoka, E., Kato, R., Kondo, Y., and Iwata, S. (1991).
J., Jr., and Bailey, B. J. (1992). The relationship between the The effect of L-menthol stimulation of the major palatine
nasal cycle and mucociliary clearance. Laryngoscope nerve on nasal patency. Auris Nasus Larynx 18:221–226.
102:117–120. Naito, K., Kondo, Y. Ohoka, E., Komori, M., Takeuchi, M., and
Lorino, A. M., Lofaso, F., Drogou, I., Abi-Nader, F., Dahan, E., Iwata, S. (1995). New aerodynamic aspects of nasal patency.
Coste, A., and Lorino, H. (1998). Effects of different mechan- Rhinology 33:26–29.
ical treatments on nasal resistance assessed by rhinometry. Nakano, T. (1967). Influence of nozzles on pressure and flow
Chest 114:166–170. measurement studies by means of the artificial nose and con-
Lund, V. J. (1996). Nasal physiology: neurochemical receptors, ductivity meter. Rhinol. Int. 5:183–196.
nasal cycle, and ciliary action. Allergy Asthma Proc. Ng, B. A., Mamikoglu, B., Ahmed, M. S., and Corey, J. P. (1998).
17:179–184. The effect of external nasal dilators as measured by acoustic
Lund, V. J., Flood, J., Sykes, A. P., and Richards, D. H. (1998).
rhinometry. Ear Nose Throat J. 77:840–844.
Effect of fluticasone in severe polyposis. Arch. Otolaryngol.
Nielsen, H., Orntoft, S., and Nielsen, T. G. (1997). Change in
Head Neck Surg. 124:513–518.
minimal cross sectional area using the new nasal dilator,
Lung, M. A., and Wang, J. C. (1991). Mechanical stimulation of
“Airplus.” Acta Oto. Laryngol. Supp. 529:130–132.
canine respiratory tract and nasal vascular and airway resis-
Nielsen, L. P., Bjerke, T., Christensen, M. B., Pedersen, B.,
tance. Respir. Med. 85 (Suppl A):67–68.
Rasmussen, T. R., and Dahl, R. (1996). Assessment of the
Malm, L. (1977). Sympathetic influence on the nasal mucosa.
allergic reaction in seasonal rhinitis: acoustic rhinometry is a
Acta Otolaryngol. 83:20–21.
sensitive and objective method. Clin. Exp. Allergy
Maltais, F., Dinh, L. Cormier, Y., and Series, F. (1991). Changes
26:1268–1275.
in upper airway resistance during progressive normocapnic
Nishihira, S., and McCaffrey, T. V. (1987). Reflex control of nasal
hypoxia in normal men. J. Appl. Physiol. 70:548–553.
blood vessels. Otolaryngol. Head Neck Surg. 96:273–277.
Masieri, S., Cavaliere, F., and Filiaci, F. (1997). Nasal obstruction
improvement induced by topical furosemide in subjects Ogawa, H. (1986). A possible role of aerodynamic factors in nasal
affected by perennial nonallergic rhinitis. Am. J. Rhino. polyp formation. Acta Orolaryngol. (Stockh.) S430:18–20.
11:443–447. Ophir, D. Gross-Isseroff, R., Lancet, D., and Marshak, G. (1986).
Mezzanotte, W. S., Tangel, D. J., and White, D. P. (1992). Changes in olfactory acuity induced by total inferior turbinec-
Mechanisms of control of alae muscle activity. J. Appl. tomy. Arch. Otolaryngol. Head Neck Surg. 112: 195–197.
Physiol. 72:925–933. Ophir, D. Elad, Y. Dolev, Z., and Geller-Bernstein, C. (1988).
Mirza, N., Kroger, H. and Doty, R. L. (1997). Influence of age on Effects of inhaled humidified warm air on nasal patency and
the ‘nasal cycle’. Laryngoscope 107:62–66. nasal symptoms in allergic rhinitis. Ann. Allergy 60:239–242.
Morgan, K. T., Kimbell, J. S., Monticello, T. M., Patra, A. L., and Pahor, A. L. (1992). Eat, nose and throat in ancient Eqypt. J.
Fleshman, A. (1991). Studies of inspiratory airflow patterns in Laryngol. Otol. 106:773–779.
the nasal passages of the F344 rat and rhesus monkey using Pahor, A. L., and Kimura, A. (1991). History of removal of nasal
nasal molds: relevance to formaldehyde toxicity. Toxicol. polyps. A. Folha Medica (Brazil) 102:183–186.
Appl. Pharmacol. 110:223–240. Paniello, R. C. (1996). Nasal valve suspension. An effective treat-
Morris, S., Eccles, R., Martez, S. J., Riker, D. K., and Witek, T. ment for nasal valve collapse. Arch. Otolaryngol. Head Neck
J. (1997). An evaluation of nasal response following different Surg. 122:1342–1346.
treatment regimes of oxymetazoline with reference to Passali, D., Lauriello, M., Anselmi, M., and Bellussi, L. (1999).
rebound congestion. Am. J. Rhino. 11:109–115. Treatment of hypertrophy of the inferior turbinate: long-term
Morrison, D. A., Collins, M., Stovall, J. R., and Friefeld, G. results in 382 patients randomly assigned to therapy. Ann.
(1989). Reduced exercise capacity of chronic obstructive pul- Otol. Rhinol. Laryngol. 108:569–575.
458 Frye

Petruson, B., and Bjuro, T. (1990). The importance of nose- Scott, W. R., Taulbee, D. B., and Yu, C. P. (1978). Theoretical
breathing for the systolic blood pressure rise during exercise. study of nasal deposition. Bull. Math. Biol. 40:581–603.
Acta Otolaryngol. (Stockh.) 109:461–466. Seiden, A. M., and Smith, D. V. (1988). Endoscopic intranasal
Portugal, L. G., Mehta, R. H., Smith, B. E., Sabnani, J. B., and surgery on olfaction. Chem. Senses 13:736.
Matava, M. J. (1997). Objective assessment of the breathe- Series, F., Cormier, Y., Desmeules, M., and LaForge, J. (1989).
right device during exercise in adult males. Am. J. Rhinol. Influence of respiratory drive on upper airway resistance in
11:393–397. normal men. J. Appl. Physiol. 66:1242–1249.
Principato, J. J., and Ozenberger, J. M. (1970). Cyclical changes Shemen, L., and Hamburg, R. (1997). Preoperative and postoper-
in nasal resistance. Acta Otolaryngol. 91:71–77. ative nasal septal surgery assessment with acoustic rhinome-
Proctor, D. F., and Swift, D. L. (1971). The nose—a defense against try. Otolaryngol. Head Neck Surg. 117:338–342.
the atmospheric environment. In Inhaled Particles III, Vol I., W. Shevrygin, B. V. (1973). Surgical intervention on the nasal sep-
H. Walton (Ed.). Greshman Press, Surrey, Unwin, pp. 55–69. tum for the purpose of improving and preserving olfaction.
Proetz, A. W. (1951). Air currents in the upper respiratory tract Zh. Ushn. Nos. Gor. Bolezn. 31:247–249.
and their clinical importance. Ann. Otol. Rhinol. Laryngol. Silkoff, P. E., Chakravorty, S., Chapnik, J., Cole, P., and Zamel,
60:439–467. N. (1999). Reproducibility of acoustic rhinometry and rhino-
Randall, J. E. (1962). Elements of Biophysics, 2nd Year Book manometry in normal subjects. Am. J. Rhinol. 13:131–135.
Publishers, Chicago. Simmen, D., Scherrer, J. L., Moe, K., and Heinz, B. (1999). A
Reber, M., Rahm, F., and Monnier, P. (1998). The role of acoustic dynamic and direct visualization model for the study of nasal
rhinometry in the pre-and postoperative evaluation of surgery airflow. Arch. Otolaryngol. Head Neck Surg. 125:1015–1021.
for nasal obstruction. Rhinology 36:184–187. Simola, M., and Malmberg, H. (1998). Sense of smell in allergic
Rehn, T. (1978). Perceived odor intensity as a function of airflow and nonallergic rhinitis. Allergy 53:190–194.
through the nose. Sen. Proc. 2:198–205. Sipila, J., Suonpaa, J., Silvoniemi, P., and Laippala, P. (1995).
Reynolds, O. (1883). An experimental investigation of the cir- Correlations between subjective sensation of nasal patency
cumstances which determine whether the motion of water and rhinomanometry in both unilateral and total nasal assess-
shall be direct or sinuous, and the law of resistance in parallel ment. J. Otorhinolaryngol. Relat. Spec. 57:260–263.
channels. Trans. Roy. Soc. Lond. W4:935–982. Sobel, N., Khan, R. M., Hartley, C. A., Sullivan, E. V., and
Rohrer, F. (1915). Der Stromungswiderstand in den men- Gabrieli, J. D. (2000). Sniffing longer rather than stronger to
schlichen Atemwegen und der Einfluss der unregelmassigen maintain olfactory detection threshold. Chem. Senses 25:1–8.
Verzweigung des Bronchialysystems auf den Atmungs- Spaeth, J., Schultze, V., Klimek, L., Lengersdorf, A., and Mosges,
verlaugin verschiedenen Lungenbezirken. Pfluger’s Arch. R. (1996). Azelastine reduces histamine-induced swelling of
Ges. Physiol. 162:225–299. nasal mucosa. J. Otorhinolaryngol. Relat. Spec. 58:157–163.
Roithmann, R., Cole, P., Chapnik, J., Barreto, S. M., Szalai, J. P., Spoor, A. (1965). A new method for measuring nasal
and Zamel, N. (1994). Acoustic rhinometry, rhinomanometry, conductivity. Rhinol. Int. 3:27–35.
and the sensation of nasal patency: a correlative study. Stevens, C. N., and Stevens, M. H. (1985). Quantitative effects of
J. Otolaryngol. 23:454–458. nasal surgery on olfaction. Am. J. Otolaryngol. 6:264–267.
Roithmann, R., Chapnik, J., Zamel, N., Barreto, S. M., and Cole, Strohl, K. P., Hensley, M. J., Hallett, M., Saunders, N. A., and
P. (1997a). Acoustic rhinometric assessment of the nasal Ingram, R. H. (1980). Activation of upper airway muscles
valve. Am. J. Rhinol. 11:379–385. before onset of inspiration in normal humans. J. Appl.
Roithmann, R., Shpirer, I., Cole, P., Chapnik, J., Szalai, J. P., Physiol. 49:638–642.
and Zamel, N. (1997b). The role of acoustic rhinometry in Strohl, K. P., O’Cain, C. F., and Slutsky, A. S. (1982). Alae nasi
nasal provocation testing. Ear Nose Throat J. 76:747–750, activation and nasal resistance in healthy subjects. J Appl.
752. Physiol. 52:1432–1437.
Roithmann, R., Chapnik, J., Cole, P., Szalai, J., and Zamel, N. Stuiver, M. (1958). Biophysics of the sense of smell. Master’s
(1998). Role of the external nasal dilator in the management thesis, Groningen, The Netherlands.
of nasal obstruction. Laryngoscope 108:712–715. Sung, Y. W., Lee, M. H., Kim, I. J., Lim, D. W., Rha, K. S.,
Schaeffer, J. P. (1920). The Nose and Paranasal Sinuses (Nasal and Park, C. I. (2000). Nasal cycle in patients with septal devi-
Index). Blakiston Press, Philadephia. ation: evaluation by acoustic rhinometry. Am. J. Rhinol. 14:
Scherer, P. W., Hahn, I. I., and Mozell, M. M. (1989). The bio- 171–174.
physics of nasal airflow. Otolaryngol. Clin. Am. 22:265–278. Swift, D. L., and Proctor, D. F. (1977). Access of air to the respi-
Schneider, R. A., and Wolf, S. (1960). Relationship of olfactory acu- ratory tract. In Respiratory Defense Mechanisms, J. D. Brian,
ity to nasal membrane function. J. Appl. Physiol. 15: 914–920. D. F. Proctor, and L. M. Reid (Eds.). Marcel Dekker, New
Schwartz, D. N. Mozell, M. M., Youngentob, S. L., Leopold, D. York, pp. 63–93.
L., and Sheebe, P. R. (1987). Improvement of olfaction in Szucs, E., and Clement, P. A. (1998). Acoustic rhinometry and
laryngectomized patients with the larynx bypass. Laryn- rhinomanometry in the evaluation of nasal patency of patients
goscope 97:1280–1286. with nasal septal deviation. Am. J. Rhinol. 12:345–352.
Nasal Patency and Nasal Airflow Aerodynamics 459

Tai, C. F., Ho, K. Y., and Hasegawa, M. (1998). Evaluating the Van Dishoeck, H. (1964). The part of the valve and the turbinate
sensation of nasal obstruction with acoustic rhinometry and in total nasal resistance. Int. Rhinol. 2:19–26.
rhinomanometry. Kao Hsiung I Hsueh Ko Hsueh Tsa Chih 14: Werntz, D. A., Bickford, R. G., Bloom, F. E., and Shannahoff-
548–553. Khalsa, D. S. (1983). Alternating cerebral hemispheric activity
Tanaka, Y., Morikawa, T., and Honda, Y. (1988). An assessment and the lateralization of autonomic nervous function. Hum.
of nasal function in control of breathing. J. Appl. Physiol. Neurobiol. 2:39–43
65:1520–1524. Wight, R. G., Jones, A. S., and Beckingham, E. (1988a). Radical
Taverner, D., Bickford, L., Shakib, S., and Tonkin, A. (1999a). trimming of the inferior turbinate and its effect on nasal resis-
Evaluation of the dose-response relationship for intra-nasal tance to airflow. J. Laryngol. Otol. 102:694–696.
oxymetazoline hydrochloride in normal adults. Eur. J. Clin. Wight, R. G., Jones, A. S., and Clegg, R. T. (1988b). Comparison
Pharmacol. 55:509–513. of anterior and radical trimming of the inferior turbinates and
Taverner, D., Danz, C., and Economos, D. (1999b). The effects of the effects on nasal resistance to airflow. Clin. Otolaryngol.
oral pseudoephedrine on nasal patency in the common cold: a 13:223–226.
double-blind single-dose placebo-controlled trial. Clin. Willatt, D. J. (1993). Continuous infrared thermometry of the
Otolaryngol. 24:47–51. nasal mucosa. Rhinology 31:63–67.
Terada, N., Yamakoshi, T., Hasegawa, M., Tanikawa, H., Willatt, D. J., and Jones, A. S. (1996). The role of the tempera-
Maesako, K., Ishikawa, K., and Konno, A. (1998). The ture of the nasal lining in the sensation of nasal patency. Clin.
effect of ramatroban (BAY u 3405), a thromboxane A2 Otolaryngol. 21:519–523.
receptor antagonist, on nasal cavity volume and minimum Williams, H. L., Banvoetz, J. D., Brewer, D. W., Hinderer, K. H.,
cross-sectional area and nasal mucosal hemodynamics McLaurin, J. W., Murtagh, J. A., Ogura, J. H., Sauders, S. H.,
after nasal mucosal allergen challenge in patients with Hyatt, R. E., Semenov, H., and Tonndorf, J. (1970). Report of
perennial allergic rhinitis. Acta Otolaryngol. Suppl. 537: Committee on Standardization of Definitions, Terms, Symbols
32–37. in Rhinometry of the American Academy of Ophthalmology
Tonndorf, J. (1939). Der Weg der Atemluft in der menschlichen and Otolaryngology: A Handbook and Glossary.
Nase. Arch. Ohren. Nasen Kehlkopfhail 146:41–63. American Academy of Ophthalmology and Otolaryngology,
Tomkinson, A., and Eccles, R. (1998). Acoustic rhinometry: an Rochester, MN.
explanation of some common artefacts associated with nasal Yu, C. P., Diu, C. K., and Soong, T. K. (1981). Statistical analy-
decongestion. Clin. Otolaryngol. 23:20–26. sis of aerosol deposition in nose and mouth. Am. Ind. Hyg.
Tsubone, H. (1989). Nasal “flow” receptors of the rat. Respir. Assoc. J. 42:726–733.
Physiol. 75:51–64. Zweiman, B., Getsy, J., Kalenian, M., Lane, A., Schwartz, L. B.,
Uddstromer, M. (1939). Nasal respiration: a critical survey of Doty, R., and Lanza, D. (1997). Nasal airway changes
some current physiological and clinical aspects on the respi- assessed by acoustic rhinometry and mediator release during
ratory mechanism with a description of a new method of diag- immediate and late reactions to allergen challenge. J. Allerg.
nosis. Acta Otolaryngol. (Stockh.) S42:1–146. Clin. Immunol. 100:624–631.
22

Clinical Disorders of Olfaction

Claire Murphy
San Diego State University and University of California, San Diego, School of Medicine
San Diego, California, U.S.A.

Richard L. Doty
University of Pennsylvania, Philadelphia, Pennsylvania, U.S.A.

Heather J. Duncan
University of Cincinnati College of Medicine, Cincinnati, Ohio, U.S.A.

I. INTRODUCTION these topics are not specifically addressed here. The reader
is referred to Chapter 44 for disorders of taste function.
Olfactory dysfunction can arise from a variety of reasons
and can profoundly influence a patient’s quality of life.
II. CLASSIFICATION OF OLFACTORY
Such problems are not uncommon, being present in at least
DISORDERS
1% of the population under the age of 65 years, and in well
over 50% of the population older than 65 years (Doty et
Olfactory disorders can be reliably classified as follows:
al., 1984, 1986; Hoffman et al., 1998; Murphy et al., 2001;
Schiffman, 1983). We now know that decrements in olfac- (a) Anosmia: inability to detect qualitative olfactory
tory function are among the first clinical signs of sensations (i.e., absence of smell function)
Alzheimer’s disease and idiopathic Parkinson’s disease, (b) Partial anosmia: ability to perceive some, but not
and are commonly present in epilepsy, multiple sclerosis, all, odorants
and schizophrenia (see Chapter 23). Although some (c) Hyposmia or microsmia: decreased sensitivity to
patients initially present with a frank complaint of a smell odorants
disturbance, others are unaware of their dysfunction, point- (d) Hyperosmia: abnormally acute smell function
ing out the need for routine quantitative olfactory assess- (e) Dysosmia (sometimes termed cacosmia or paros-
ment, which is now easily performed in the office (see mia): distorted or perverted smell percep-
Chapter 10). tion to odorant stimulation
In this chapter, we describe the major olfactory disorders, (f) Phantosmia: a dysosmic sensation perceived in
how they are classified, and how they are evaluated and the absence of an odor stimulus (a.k.a. olfac-
treated. Since other chapters focus, in detail, on olfactory tory hallucination)
dysfunction observed as a result of head trauma (Chapter (g) Olfactory agnosia: inability to recognize an odor
30), epilepsy (Chapter 23), toxic chemical exposure sensation, even though olfactory processing,
(Chapter 27), nutritional disturbances (Chapter 42), and language, and general intellectual functions are
neurodegenerative diseases and schizophrenia (Chapter 23), essentially intact, as in some stroke patients.

461
462 Murphy et al.

Presbyosmia is sometimes used to describe smell loss TestTM (PST)(Doty et al., 1995), the 12-odor Brief-Smell
due to aging, but this term is less specific than those Identification TestTM (B-SIT; also known as the Cross-
noted above (e.g., it does not distinguish between anos- Cultural Smell Identification TestTM) (Doty et al., 1996),
mia and hyposmia) and is laden, by definition, with the the 12-item Odor Memory TestTM (OMT) (Bromley and
notion that it is age, per se, that is causing the age-related Doty, 1995; Doty et al., 1995), the 40-odor University of
deficit. Pennsylvania Smell Identification Test (UPSIT; known
When possible, it is useful to classify olfactory impair- commercially as the Smell Identification TestTM or SIT)
ments into three general classes: (1) conductive or (Doty, 1995), an Odor Confusion Matrix Test (Wright,
transport impairments from obstruction of the nasal pas- 1987), the Scandinavian Odor Identification Test (SOIT)
sages (e.g., by chronic nasal inflammation, polyposis, (Nordin et al., 1999), the ‘Sniffin’ Sticks’ test (Hummel et
etc.); (2) sensorineural impairment from damage to the al., 1997), the Viennese Olfactory Test Battery (WOTB)
olfactory neuroepithelium (e.g., by viruses, airborne tox- (Lehrner and Deecke, 1999), the Jet Stream Olfactometer
ins, etc.); and (3) central olfactory neural impairment Test (Ikeda et al., 1999), and an 8-odor identification test
from central nervous system (CNS) damage (e.g., tumors, (Simmen et al., 1999). Electrophysiological tests
masses impacting on olfactory tract, neurodegenerative (reviewed in detail in Chapter 11) are available in some
disease, etc.). However, definitive classification of a given specialized medical centers and can aid in the detection of
patient’s disorder into a given class is often not feasible, malingering. Normative odor event-related potential
and these categories are not mutually exclusive. For (OERP) data have recently been published (Murphy et al.,
example, both damage and blockage of airflow to the 2000b). Like psychophysical tests, such measures are sen-
receptors can occur from chronic rhinosinusitis, and some sitive to aging, gender, and a number of diseases. Unlike
viruses that damage the olfactory neuroepithelium also their visual and auditory counterparts, however, OERPs
are transported into the CNS via the olfactory nerves, sub- are presently unable to discern where in the olfactory
sequently damaging central elements of the system as pathway an anomaly exists. Although recent studies
well (see Chapter 26). employing source localization analyses can roughly locate
the source of the generator potentials involved, such local-
ization is dependent upon the assumptions made in the
III. CLINICAL EVALUATION OF OLFACTORY underlying models and does not necessarily provide local-
DISORDERS ization of pathology.

A. Quantitative Olfactory Testing B. Medical History

A common error made on the part of clinicians is to accept The etiology of most cases of olfactory dysfunction can be
a patient’s report of sensory dysfunction and not to objec- ascertained from carefully questioning the patient about
tively verify the presence or magnitude of the problem. the nature, timing, onset, duration, and pattern of their
Many persons, particularly the elderly and those with symptoms, as well as a historical determination of
dementia, are unaware of their dysfunction or are inaccu- antecedent events (e.g., head trauma, upper respiratory
rate in assessing its magnitude (Doty et al., 1987; Nordin infections, toxic exposures, nasal surgeries). Fluctuations
et al., 1995). Standardized quantitative olfactory testing in function usually reflect obstructive, rather than neural,
allows, in most instances, for (1) the characterization of the factors. Subtle symptoms of central tumors, dementia,
nature and degree of the chemosensory problem, (2) estab- tremor, and seizure activity (e.g., automatisms, occurrence
lishing the validity of the patient’s complaint, including the of blackouts, auras, and déjà vu) should be sought, given
detection of malingering, (3) monitoring changes in the frequent association between smell dysfunction and not
function over time, and (4) providing objective data for only brain tumors, but such disorders as epilepsy, idio-
establishing disability compensation. Fortunately, easy-to- pathic Parkinson’s disease, Alzheimer’s disease, and mul-
administer tests of smell function have been developed, a tiple sclerosis (see Chapter 23). Delayed puberty in
number of which are commercially available. Such tests, association with anosmia, with or without midline cranio-
which vary considerably in terms of reliability (see facial abnormalities, deafness, and renal anomalies, sug-
Chapter 10), include the T&T olfactometer test (Takagi, gests the possibility of Kallmann’s syndrome or one of its
1989), the San Diego Odor Identification Test (Anderson variants. Medications being used prior to or at the time
et al., 1992; Murphy et al., 1994), the Smell Threshold of the symptom onset should be determined, as some
TestTM (Doty, 2000), the Alcohol Sniff Test (AST) can profoundly influence olfaction [e.g., antifungal
(Davidson and Murphy, 1997), the 3-odor Pocket Smell agents, angiotensin-converting enzyme (ACE) inhibitors].
Clinical Disorders of Olfaction 463

Medical conditions potentially associated with smell ease, and most can be expected to reflect significant dam-
impairment should also be identified (e.g., liver disease, age to the olfactory neuroepithelium (Deems et al., 1991;
hypothyroidism, or diabetes). A history of epistaxis, dis- Goodspeed et al., 1987; Mott and Leopold, 1991). Listed
charge (clear, purulent, or bloody), nasal obstruction, and below are major causes of olfactory dysfunction not dis-
somatic symptoms, including headache or irritation, as cussed in detail elsewhere in the volume, along with an
well as a report as to whether the problem seems to be overview of findings for each of the disorders involved.
more prevalent on one side of the nose or the other, may be
of localizing value. Idiopathic cases that present during A. Upper Respiratory Infections
winter months (which is more common than not) suggest
the possibility of a viral origin, even if other elements of an Although rarely appreciated, the most frequent cause of
upper respiratory infection were not present or recognized. smell loss in the adult is an upper respiratory infection
It is critical for the clinician to be aware that while (URI), such as is associated with the common cold,
patients often present with the complaint of taste loss, influenza, pneumonia, or human immunodeficiency virus
quantitative testing usually reveals only an olfactory prob- (HIV) (Akerlund et al., 1995; Deems et al., 1991; Hummel
lem, reflecting decreased retronasal stimulation of the et al., 1998; Murphy et al., 2000a). Often the respiratory
olfactory receptors during deglutition (Burdach and Doty, illness is described as being more severe than usual, and in
1987; Murphy et al., 1977) (see Chapter 44). Importantly, many cases a dysosmia or phantosmia is present. The lat-
the clinician should be cognizant of the fact that combina- ter phenomena typically subside over time, leaving the
tions of causal factors may be present that need to be con- patient with a noticeable olfactory deficit. Exactly what
sidered. For example, persons with allergies or older predisposes someone to viral- or bacterial-induced smell
persons may be more susceptible than others to smell loss dysfunction or the mechanisms underlying it remains
from viral and other causes because of prior cumulative unclear, although most such losses become manifest in
damage to the olfactory epithelium. middle or older age, suggesting the potential importance of
Today, nasal endoscopy is the method of choice in cumulative insult and the challenge of regeneration of the
assessing the health of the nasal cavity, and visualization neuroepithelium in advancing age when proliferation of
of the olfactory meatal region is now possible through the basal cells and immature neurons is significantly reduced
use of both flexible and rigid rhinoscopes (Davidson et al., (Loo et al., 1996). Direct insult to the olfactory neuro-
1995). Modern imaging techniques can detect inflamma- epithelium is presumably the primary basis of the problem
tory processes within the nose and sinuses, as well as brain in URIs, as biopsy studies of olfactory epithelia from
lesions and the integrity of the olfactory bulbs, tracts, and patients with post-URI anosmia evidence extensive cica-
cortical parenchyma (see Chapter 28). For example, trization, decreases in receptor cell number, absent or
patients complaining of never having a sense of smell typi- decreased numbers of cilia on remaining receptor cells,
cally lack normal olfactory bulbs or tracts upon appropri- and replacement of sensory epithelium with respiratory
ate magnetic resonance imaging (MRI). Some laboratory epithelium (Douek et al., 1975; Jafek et al., 1990b;
tests (e.g., blood serum tests) are helpful in detecting Yamagishi et al., 1988, 1994). However, as noted in
underlying medical conditions suggested by history and Chapter 26, many viruses invade the CNS via the olfactory
physical examinations, such as infection, nutritional defi- neuroepithelium, and the possibility exists that some
ciencies (e.g., vitamins B6, B12), allergy, diabetes mellitus, viruses may influence central structures in addition to, or
and thyroid, liver, and kidney disease. Visual acuity, visual independent of, peripheral damage. Even though sponta-
field, and optic disc examinations can aid in the detection neous recovery in some of these patients is theoretically
of possible intracranial mass lesions that, in addition to possible, meaningful recovery is rare when marked loss
producing visual deficits, impinge upon the olfactory tract. has been present for a period of time. As with the case of
cessation of smoking (Frye et al., 1990), however, there
may be some moderate return of function over relatively
IV. CAUSES OF OLFACTORY DYSFUNCTION long periods of time, depending upon the magnitude of the
insult to the underlying basal cell membrane. For example,
As can be seen in Table 1, there are many reported etiol- one study followed up 21 patients with URI-related smell
ogies for olfactory disturbance. Approximately two thirds loss longitudinally (average duration
3 years), noting
of cases of chronic anosmia or hyposmia (i.e., those which that 19 evidenced significantly higher UPSIT scores on
are presumably permanent) that present to a clinic retest and that 13 reported subjective improvement
are likely due to prior upper respiratory infections, head (Duncan and Seiden, 1995). However, it should be noted
trauma (Chapter 30), and nasal and paranasal sinus dis- that, according to standardized norms (Doty, 1995), the
464 Murphy et al.

Table 1 Reported Agents, Diseases, Drugs, Interventions, and Other Etiological Categories Associated in the Medical or Toxicological
Literature with Olfactory Dysfunctiona

Drugs Endocrine/Metabolic
Adrenal steroids (chronic use) Addison’s disease
Amino acids (excess) Congenital adrenal hyperplasia
Cysteine Cushing’s syndrome
Histidine Diabetes mellitus
Analgesics Froelich’s syndrome
Antipyrine Gigantism
Anesthetics, local Hypergonadotropic hypogonadism
Cocaine HCl Hypothyroidism
Procaine HCl Kallmann’s syndrome
Tetracaine HCl Pregnancy
Anticancer agents (e.g., methotrexate) Panhypopituitarism
Antihistamines (e.g., chlorpheniramine malate) Pseudohypoparathyroidism
Antimicrobials Sjögren’s syndrome
Griseofulvin Turner’s syndrome
Lincomycin Industrial Dusts, Metals, Volatiles
Macrolides Acetone
Neomycin Acids (e.g., sulfuric)
Pencillins Ashes
Streptomycin Benzene
Tetracyclines Benzol
Tyrothricin Butyl acetate
Antirheumatics Cadmium
Mercury/gold salts Carbon disulfide
D-Penicillamine Cement
Antithyroids Chalk
Methimazole Chlorine
Propylthiouracil Chromium
Thiouracil Coke/coal
Antivirals Cotton
Cardiovascular/hypertensives Cresol
Gastric medications Ethyl acetate
Cimetidine Ethyl and methyl acrylate
Hyperlipoproteinemia medications Flour
Artovastatin calcium (Lipitor) Formaldehyde
Cholestyramine Grain
Clofibrate Hydrazine
Intranasal saline solutions with: Hydrogen selenide
Acetylcholine Hydrogen sulfide
Acetyl, -methylcholine Iron carboxyl
Menthol Lead
Strychnine Mercury
Zinc sulfate Nickel
Local vasoconstrictors Nitrous gases
Opiates Paint solvents
Codeine Paper
Hydromophone HCl Pepper
Morphine Peppermint oil
Psychopharmaceuticals (e.g., LSD, psilocybin) Phosphorus oxychloride
Sympathomimetics Potash
Amphetamine sulfate Silicone dioxide
Fenbutrazate HCI Spices
Phenmetrazine theoclate Trichloroethylene
a Categories are not mutually exclusive.
Clinical Disorders of Olfaction 465

Table 1 (continued)

Infections — Viral/Bacterial Temporal lobe tumors

Acquired immunodeficiency syndrome (AIDS) Neoplasms—Intranasal
Acute viral rhinitis Neuro-olfactory tumors
Bacterial rhinosinusitis Esthesioepithelioma
Bronchiectasis Esthesioneuroblastoma
Fungal Esthesioneurocytoma
Influenza Esthesioneuroepithelioma
Rickettsial Other benign or malignant nasal tumors
Microfilarial Adenocarcinoma
Lesions of the nose/Airway blockage Leukemic infiltration
Adenoid hypertrophy Nasopharyngeal tumors with extension
Allergic rhinitis Neurofibroma
Perennial Paranasal tumors with extension
Seasonal Schwannoma
Atrophic rhinitis Neoplasms—Extranasal and Extracranial
Chronic inflammatory rhinitis Breast
Hypertrophic rhinitis Gastrointestinal tract
Nasal polyposis Laryngeal
Rhinitis medicamentosa Lung
Structural abnormality Ovary
Deviated septum Testicular
Weakness of alae nasi Neurological
Vasomotor rhinitis Amyotrophic lateral sclerosis
Medical Interventions Alzheimer’s disease
Adrenalectomy Cerebral abscess (esp. frontal or ethmoidal regions)
Anesthesia Down syndrome
Anterior craniotomy Familial dysautonomia
Arteriography Guam ALS/PD/dementia
Chemotherapy Head trauma
Frontal lobe resection Huntington’s disease
Gastrectomy Hydrocephalus
Hemodialysis Korsakoff’s psychosis
Hypophysectomy Migraine
Influenza vaccination Meningitis
Laryngectomy Multiple sclerosis
Oophorectomy Myesthenia gravis
Paranasal sinus exenteration Paget’s disease
Radiation therapy Parkinson’s disease
Rhinoplasty Refsum’s syndrome
Temporal lobe resection Restless leg syndrome
Thyroidectomy Syphilis
Neoplasms—Intracranial Syringomyelia
Frontal lobe gliomas and other tumors Temporal lobe epilepsy
Midline cranial tumors Hamartomas
Parasagital meningiomas Mesial temporal sclerosis
Tumors of the corpus callosum Scars/previous infarcts
Olfactory groove/cribriform plate meningiomas Vascular insufficiency/anoxia
Osteomas Small multiple cerebrovascular accidents
Paraoptic chiasma tumors Subclavian steal syndrome
Aneurysms Transient ischemic attacks
Craniopharyngioma Nutritional/metabolic
Pituitary tumors (esp. adenomas) Abetalipoproteinemia
Suprasellar cholesteatoma Chronic alcoholism
Suprasellar meningioma Chronic renal failure

(continued)
466 Murphy et al.

Table 1 (continued)

Cirrhosis of liver Psychiatric

Gout Anorexia nervosa (severe stage)
Protein-calorie malnutition Attention deficit disorder
Total parenteral nutrition w/o adequate replacement Depressive disorders
Trace metal deficiencies Hysteria
Copper Malingering
Zinc Olfactory reference syndrome
Whipple’s disease Schizophrenia
Vitamin deficiency Schizotypy
Vitamin A Seasonal affective disorder
Vitamin B6 Pulmonary
Vitamin B12 Chronic obstructive pulmonary disease

average patient of this study was not anosmic on the first of whom exhibit a generally more pathological epithelium
test occasion [mean (SEM) UPSIT score
21.2 (1.7)] and (e.g., disordered arrangement of cells, more islands of
that the magnitude of improvement was modest, with the respiratory-like epithelium) (Lee et al., 2000). This
mean score still indicative of borderline severe/moderate hypothesis is further supported by findings of weak or no
microsmia [mean (SEM)
26.2 (1.5)]. Nonetheless, a associations between olfactory test scores and measures of
positive correlation (r
0.55; p  0.001) was found nasal airway patency (save severe blockage), whether
between the change in UPSIT scores and the time between measured by rhinoscopy, rhinomanometry, or acoustic rhi-
the two test administrations, implying that the longer the nometry (Apter et al., 1999; Cowart et al., 1993; Nordin
post-URI interval, the greater the recovery. More studies et al., 1998; Scott et al., 1988).
are needed to establish whether such recovery continues
over longer time periods and whether an asymptote in per-
formance occurs at some point.

B. Nasal and Sinus Disease

Olfactory impairment that accompanies nasal or sinus dis-

ease has been traditionally viewed as being solely conduc-
tive (Fig. 1). Although marked airflow blockage
undoubtedly alters olfactory sensitivity in some patients,
empirical data suggest that surgical (e.g., excision of
polyps) or medical (e.g., administration of topical or sys-
temic steroids) treatment rarely returns function to normal,
implying that blockage alone cannot completely explain
the olfactory loss (for review, see Doty and Mishra, 2001).
While, in general, olfactory dysfunction is related to the
severity of rhinosinusitis [e.g., in one study, the mean Figure 1 Schematic drawing (coronal) of the nasal cavity and
UPSIT scores were 35, 31, 26, and 23 for Kennedy Stages paranasal sinuses. The left side is typical of chronic rhinosinusi-
I to IV of the disease, respectively (Downey et al., 1996)], tis, with pansinus obstruction. Airflow to the nasal vault (and
the defining factor may, in fact, be the severity of olfactory epithelium) is obstructed by congestion of the
histopathological changes within the olfactory mucosa ostiomeatal complex (i, infundibulum; up, uncinate process; eb,
ethmoidal bulla; fr, frontal recess; mt, middle turbinate). The
(Jafek et al., 1987). Thus, Kern (2000) has reported that
right side shows the situation following endoscopic surgery,
such severity is related to olfactory test scores in patients which opens up the middle meatus, allowing freer air flow to the
with chronic rhinosinusitis. Furthermore, biopsies from the olfactory mucosa. The superimposed rectangle indicates the gen-
neuroepithelial region of patients with nasal disease are eral location of the olfactory epithelium, which is actually poste-
less likely to yield olfactory-related tissue than biopsies rior to this plane of section, on the upper lateral wall of the nasal
from controls (Feron et al., 1998). The same is true for septum, on the cribriform plate, and on the medial wall of the
anosmic vs. nonanosmic rhinosinusitis patients, the former superior turbinate. (From Smith and Duncan, 1992.)
Clinical Disorders of Olfaction 467

1. Hypertrophied Adenoids et al., 1996; Kondo et al., 1998; Mott et al., 1997;
Rydzewski et al., 2000; Simola and Malmberg, 1998).
Hypertrophied adenoid tissue can significantly block the
Overall, there appears to be general correspondence
nasal airflow of children whose airways are otherwise
between a patient’s self-report of olfactory loss and objec-
patent. In general, nasal resistance decreases by ~20–40%
tive test measures. Golding-Wood et al. (1996), for exam-
following adenoidectomy (Crysdale et al., 1985; Fielder,
ple, evaluated odor identification ability before and after 6
1985). Olfactory thresholds to phenyl ethyl alcohol, deter-
weeks of betamethasone treatment in 25 well-documented
mined in children with varying degrees of nasal obstruction,
perennial rhinitis patients. The patient group was initially
are related directly to clinical ratings of nasal obstruction
divided into two groups: those who affirmatively answered
(Fig. 2) (Delank, 1992; Ghorbanian et al., 1983).
the question “Is your sense of smell impaired?” (n
15)
2. Acute Viral-Related Rhinitis or Rhinosinusitis and those who did not (n
10). The UPSIT scores of each
of the 15 members of the former group were higher after
Two studies have quantitatively assessed olfaction the betamethasone treatment [respective group means (SD)
longitudinally following the onset of the common cold.
18.93 (9.4) and 33.4 (4.01)]. This was not the case for
Akerlund et al. (1995) measured 1-butanol odor-detection those who initially felt that they had no problems smelling
thresholds in a group of student volunteers before and 4 [respective pre-/posttreatment means (SD)
33.40 (4.01)
days after nasal inoculation with the coronavirus 229E. and 32.8 (4.94)]. Moderate correlations between the
The nine individuals who developed a cold had impaired UPSIT scores and the self-ratings of olfactory function
olfactory thresholds on the postinoculation test relative to were found both before (r
0.52) and after (r
0.58)
the controls—impairment that correlated with nasal con- treatment. As has been noted by others, however, the aver-
gestion, but not nasal discharge. Hummel et al. (1998) age posttreatment UPSIT score was still indicative of a
evaluated smell function and airway patency in 18 women mild hyposmic condition. The UPSIT scores retained,
and 18 men at the time of onset of a natural cold, as well among the patients, a similar rank order before and after
as 2, 4, 6, and 35 days later. The cold produced a decrease treatment (Spearman r
0.75).
in the volume of the anterior nasal cavity, an increase in Several studies have examined olfactory function
mucus secretion, an increase in olfactory thresholds, a before and after allergen challenges (e.g., Hilberg, 1995;
decrease in intensity ratings, and a decrease in odor- Hinriksdottir et al., 1997; Klimek and Eggers, 1997; Lane
evoked potential amplitudes (N1) to both olfactory and et al., 1996; Moll et al., 1998). In all cases, smell function
trigeminal stimuli. Even when the airway was patent and decreased as a result of the challenges, although in the few
mucus secretion was minimal, evoked potential amplitudes cases that examined measures of airway patency, no asso-
to olfactory stimuli were still depressed, indicating to the ciation was noted between the degree of smell dysfunction
authors that URIs may influence olfactory function inde- and the patency measure. Hilberg (1995) assessed the
pendent of nasal congestion. comparative effects of the oral antihistamine terfenadine
(an H1-blocker) and the topical steroid, budesonide, on an
allergen challenge in subjects with nasal allergy uncompli-
3. Acute (e.g., Allergic) and Chronic Rhinitis and
cated by polyposis on olfactory dysfunction and various
Rhinosinusitis
other hay fever symptoms. Although both drugs had an
The first large-scale empirical study of olfaction in allergic effect on the nonolfactory hay fever symptoms during the
rhinitis was that by Cowart et al. (1993). Phenyl ethyl alco- nasal pollen challenge, only the budesonide improved the
hol detection threshold measures were obtained from 91 challenge-related decrement in olfactory sensitivity. This
patients with symptoms of allergic rhinitis and from 80 steroid also was more effective in increasing nasal volume.
nonatopic controls. The allergy patients exhibited greater Unfortunately, the improvement in olfactory function
dysfunction than the controls, with 23.1% having a thresh- occurred in less than half of the patients (7/17; 41%).
old at or above the 2.5 percentile of the controls. Clinical
or radiographic evidence of rhinosinusitis or nasal polyps C. Nasal Surgery
or both was associated with hyposmia in the allergy
patients: 14.3% with no associated rhinosinusitis exhibited Gross-Isseroff et al. (1989) obtained detection threshold
hyposmia, whereas 42.9% with associated rhinosinusitis and UPSIT measures in children with choanal atresia
did so. before and after surgical repair at relatively advanced ages
These general observations have been noted by others (8–31 years). The three patients who had suffered from
for rhinitis or rhinosinusitis patients employing a variety of bilateral atresia had permanent olfactory deficits, whereas
olfactory test measures (Apter et al., 1999; Golding-Wood the one patient who had suffered from unilateral atresia
468 Murphy et al.

Figure 2 (A) Nasal obstruction ratings, based upon assessment of mouth breathing and hyponasality, in 28 children before and after
adenoidectomy. (B) Phenyl ethyl alcohol odor-detection thresholds before and after adenoidectomy in the same study population. Each
line joins preoperative and postoperative values for an individual subject. (From Ghorbanian et al., 1983).

appeared to have normal function. These findings sug- The degree of improvement is well exemplified by
gested to the authors the question as to whether early sen- examining the median pre- and postoperative UPSIT
sory exposure is needed for normal development of scores calculated across six studies in which five or more
olfactory function (see Chapter 29). patients were tested (Eichel, 1994; el Naggar et al., 1995;
The limited empirical data available suggest that sep- Friedman et al., 1999; Kimmelman, 1994; Lund and
toplasty and rhinoplasty have no adverse effects on Scadding, 1994; Seiden and Smith, 1988). These values,
olfactory function and may possibly improve function 17.0 and 25.5, respectively, are indicative of an average
very slightly if the airway is significantly constricted change in function from total anosmia to a borderline
(Kimmelman, 1994; Stevens and Stevens, 1985). moderate/severe microsmia.
Olfactory function has been evaluated before and after
common operative procedures for chronic rhinosinusitis
and/or polyposis unresponsive to more conservative D. Tumors
treatments (e.g., allergen avoidance, nasal corticos-
teroids). Among such operations are middle turbinate Olfactory dysfunction can result from a variety of
medialization, polypectomy, uncinectomy, anterior eth- intranasal and intracranial tumors. McCormack and Harris
moidectomy, posterior ethmoidectomy, and sphenoidec- (1955) reported on five cases in which neurogenic tumors
tomy, alone or in combination. With rare exception (e.g., arising in the lateral nasal wall were accompanied by anos-
Klimek et al., 1997), olfactory function improves con- mia and noted that over 100 such cases had been cited in
siderably following such surgeries (Delank and Stoll, the literature up to that time. Around 20% of the tumors of
1994, 1998; Downey et al., 1996; Hoseman et al., 2000; the temporal lobe or lesions of the uncinate convolution are
Kimmelman, 1994; Leonard et al., 1988; Seiden and said to produce some form of olfactory disturbance, most
Smith, 1988), even though the proportion of patients typically the hallucination of a bad smell (Furstenberg
regaining normal olfactory function is typically less than et al., 1943).
40% (e.g., Delank and Stoll, 1994, 1998; Leonard et al., The olfactory bulbs and tracts are very sensitive to pres-
1988; Min et al., 1995) and many cases likely regresses sure from meningiomas from the dura of the cribriform
within a year to preoperative levels (Klimek et al., 1997). plate and surrounding regions (e.g., olfactory groove
Clinical Disorders of Olfaction 469

meningiomas, suprasellar ridge meningiomas), pituitary sphere). When generalized intracranial pressure was
growths that extend above the diaphram of the sella tur- observed, decreased recognition threshold sensitivity was
cica, and tumors inside or on the floor of the third ventri- sometimes present (Elsberg, 1935b). Although modern
cle. Thus, both unilateral and bilateral hyposmia and imaging techniques may make such testing, in the general
anosmia have been reported in cases of frontal lobe sense, seem rather archaic, it is conceivable that olfactory
glioma, suprasellar meningioma, and sphenoidal ridge testing could still be of value today in the early detection of
meningioma, as well as in cases involving nonneoplastic some brain tumors.
space–filling lesions, such as large internal carotid Recently, Daniels et al. (2001) examined odor discrim-
aneurysms extending over the pituitary fossa, aneurysms ination performance and OERPs in 20 subjects with uni-
of the anterior communicating bifurcation, and hydro- lateral frontal or temporal lobe brain tumors. Patients with
cephalus that pushes the floor of the third ventricle down- right-side lesions exhibited deficits in odor discrimination
ward (Graff-Radford et al., 1997). Although signs other in both the left and right nares. Patients with left-side
than olfactory ones are typically present in such cases, lesions only exhibited attenuated function when the odor-
olfactory dysfunction can, in fact, be the sole sign ant was presented to the left side. The amplitude of the
(Barraquer-Berre and Fargus, 1950; McCormack and OERPs was decreased after left-side stimulation, but not
Harris, 1955; FitzSimon et al., 1997). According to Finelli after right-side stimulation. Interestingly, a correlation
and Mair (1991), the single most egregious error of neu- was present between the olfactory and acoustic
rologists is failure to recognize the symptom of anosmia event–related potentials in patients with right-side lesions
as the principal or sole feature of an olfactory groove after right-side stimulation.
neoplasm.
Unfortunately, the classical surgical approach to anterior E. Congenital Anosmia
skull base tumors — bifrontal craniotomy — is almost
A number of patients report never having experienced smell
always associated with postoperative anosmia and other
function, with the phenotype ranging from being unable to
complications. For this reason, a number of surgical
detect only one or a few compounds to total anosmia.* In
approaches that better spare the olfactory system have been
devised (e.g., unilateral frontal craniotomy with orbital
osteotomy; the transglabellar/subcranial approach)(Babu * Most persons are said to be constitutionally unable to detect a
et al., 1995; Jung et al., 1997). few specific odorants, a phenomenon termed specific anosmia.
It is of interest that olfactory tests were employed over First described by Blakeslee for the fragrance of verbena flowers
60 years ago to help diagnose cribriform plate menin- (Blakeslee, 1918), dozens of specific anosmias have been
giomas and to localize other tumors impinging upon the reported in the literature, including ones for musks, trimethy-
olfactory nerve (Elsberg, 1935a,b; Elsberg and Levy, 1935). lamine, hydrogen cyanide,n-butyl mercaptan, and isovaleric acid
Despite the limitations of the olfactory test procedure used (for review, see Takagi, 1989). However, like the blind spot of the
[i.e., the so-called blast injection technique (see Chapter retina, specific anosmias are not typically recognized by a patient
and are rarely reasons for seeking clinical help. Moreover, some
10)], the olfactory recognition threshold was generally ele-
specific anosmics can, in fact, detect high concentrations of odor-
vated on the side where a neoplasm exerted pressure on one
ants to which they are said to be anosmic, implying hyposmia
olfactory nerve. When both nerves were involved, bilateral rather than anosmia, and not all specific anosmias are really that
threshold elevation was noted, with greater elevation on the specific. Amoore (1971), for example, tested the thresholds of 10
most affected side. Such changes were noted for expanding subjects who exhibited specific anosmia to isobutyric acid to 17
lesions on the ventral surfaces of the frontal cerebral lobes, other compounds, finding heightened thresholds for straight
as well as for suprasellar meningiomas and aneurysms of chain fatty acids with four to seven carbon atoms. Importantly,
the internal carotid artery or the anterior part of the circle of exposure to, or repeated testing with, some odorants, including
Willis. Threshold values were not influenced by pituitary ones associated with specific anosmias, can result in an increase
adenomas confined to the region of the sella turcica, but in their sensitivity in both humans and rats (Doty and Ferguson-
were heightened if growth occurred above the sella. Segall, 1989; Doty et al., 1981). For example, Wysocki et al.
(1989) found that the ability to perceive androstenone could be
Although most intracerebral tumors were not associated
induced in 10 of 20 initially insensitive subjects by systematically
with altered recognition thresholds, they were associated
exposing them to this odorant, an observation confirmed by oth-
with prolonged duration of olfactory adaptation or fatigue. ers (Moller et al., 1999). This finding led Wysocki et al. to sug-
Thus, tumors in or near the midline of the cranial cavity gest that three categories of humans subjects may exist in regards
produced long-lasting adaptation (e.g., parasagital menin- to responsiveness to this substance; namely, the truly anosmic,
giomas, tumors of the corpus callosum, and infiltrating the inducible, and those who are either constitutionally sensitive
growths extending to the medial surface of a cerebral hemi- or have experienced incidental exposure to the agent.
470 Murphy et al.

the vast majority of cases of total anosmia, appropriate MRI aforementioned tastants, as well as for pure-tone auditory
evaluation of the cribriform and gyrus rectus regions thresholds, and for filtered and nonfiltered speech discrim-
reveals agenesis or dysgenesis of the olfactory bulbs and ination tests performed at 40 and 60 dB above threshold
stalks (Yousem et al., 1996). While some of these cases (Henkin and Daly, 1968; Henkin et al., 1967).
may represent long-term degeneration from viral or trau- Attempts to replicate or extend these findings have
matic insults to the olfactory epithelium or fila (reflecting apparently not been forthcoming. Such efforts are sorely
the withdrawal of trophic influences on the bulb from intact needed, however, since such marked hypersensitivity is
olfactory neurons), most are presently assumed to be unprecedented in the chemical senses literature, and
hereditary. Lygonis (1969) reported total anosmia (without numerous tests of auditory, gustatory, and olfactory func-
any apparent associated disorders) in members of four gen- tion in rats before and after adrenalectomy have found no
erations of a family living in an isolated island community hypersensitivity of any sort. Indeed, the available data sug-
and suggested that the pattern of inheritance was likely gest that adrenalectomy may, in fact, produce decrements
autosomal dominant. Singh et al. (1970) described a famil- in chemosensory function (e.g., Brosvic and Rowe, 1992;
ial anosmia loosely associated with premature baldness and Conn and Mast, 1973; Doty et al., 1991; Kosten and
vascular headaches, which appeared to be inherited as Contreras, 1985; Weigel et al., 1989).
dominant with varying penetrance. Several congenital
olfactory syndromes are accompanied by endocrine dys-
2. Chromatin Negative Gonadal Dysgenesis (Turner’s
function (e.g., Kallmann’s syndrome), and are discussed in
Syndrome)
the next section on endocrine disorders.
Turner’s syndrome (TS) is a form of gonadal dysgenesis
F. Endocrine Disorders resulting from a 45, X karyotype (X-chromosomal mono-
somy). TS is characterized by a female phenotype, a
Although a number of endocrine disorders have been asso- shield-like chest, short stature, short and sometimes
ciated with smell dysfunction, many more have not even webbed neck, low-set ears, small mandible, a high-
been assessed for this problem. Of those that have been arched palate, and sexual infantilism. Other problems in
examined, the mechanisms responsible for the smell loss — some cases include congenital lymphedema, skeletal
aside from obvious anatomical alterations in the primary anomalies, abnormalities of the nails, and cardiac and
olfactory pathways—are poorly understood. Listed below renal deficits. In the sole olfactory study published on TS,
are disorders for which at least some quantitative olfactory nine patients were found to have elevated detection and
test data are available. recognition thresholds to the three odorants assessed
(pyridine, thiophene, nitrobenzene) (Henkin, 1967).
Gonadal hormone therapy reversed the chemosensory
1. Adrenocortical Insufficiency (Addison’s Disease)
deficits. Interestingly, the mothers of the patients
Henkin and Bartter (1966) reported that patients with exhibited similar olfactory abnormalities, which the
untreated adrenocortical insufficiency exhibited greater author suggests is in accord with a potential genetic basis
odor detection threshold sensitivity, relative to normal con- for the chemosensory disorder.
trols, not only for the odorants pyridine, thiophene, and
nitrobenzene, but for the vapors above aqueous solutions
3. Cushing’s Syndrome
of six tastants (NaCl, KCl, HCl, NaHCO3, sucrose, and
urea). These effects were remarkable, with none of distrib- Cushing’s syndrome (CS) is a disease characterized by
utions of scores from the patient and control groups show- chronic excessive secretion of adrenal corticosteroids
ing any overlap. In the case of the volatiles arising from the (i.e., hypercortisolism). Approximately 80% of such
tastants, the patients detected differences between the cases reflect excessive secretion of adrenocorticotropic
water and the vapors of the tastants at 1/10,000 the hormone (ACTH) (e.g., ACTH-producing pituitary
concentration of that noted for normals. The increased sen- tumors), whereas the remainder reflect ACTH-indepen-
sitivity did not return to normal following daily 20 mg dent etiologies (e.g., adrenal gland adenomas and carci-
injections of deoxycorticosterone acetate (DOCA) for nomas, cortisol-secreting tumors) (Ferrante, 1999).
periods up to 10 days. In contrast, treatment of the subjects According to Henkin (1975), patients with CS exhibit a
with 20 mg of the carbohydrate-active steroid prednisolone decrease in sensory detection acuity for olfaction, gesta-
returned sensitivity to normal within 24 hours. These phe- tion, hearing, and proprioception. Dogs given chronic
nomena, however, appear quite general, as similar alter- dexamethasone (an animal model of Cushing’s syn-
ations in taste sensitivity were noted for all of the drome) do appear to have increased detection thresholds
Clinical Disorders of Olfaction 471

for benzaldehyde and eugenol, in accord with this notion an earlier mouse study linked hypothyroidism to anosmia,
(Ezeh et al., 1992). the paradigm employed was, in fact, a two-bottle prefer-
ence paradigm, leaving open the possibility that dysosmia,
rather than anosmia, induced the behavioral changes
4. Hypothyroidism
(Beard and Mackay-Sim, 1987).
It is well established that many hypothyroid patients com-
plain of chemosensory problems (Deems et al., 1991).
5. Pseudohypoparathyroidism
Thus, Lewitt et al. (1989) reported that 11 of 16 (69%)
hypothyroid patients they evaluated complained of alter- Pseudohypoparathyroidism (PHP) is a rare endocrine dis-
ations in taste and smell function, and McConnell et al. order characterized by deficiencies in responsiveness to,
(1975) reported that 7 of 18 patients with untreated primary but not in the production of, parathyroid hormone.
hypothyroidism (39%) were cognizant of some alteration in Individuals with Type Ia PHP exhibit a generalized hor-
their sense of smell, with 3 (17%) experiencing dysosmia. mone resistance, a deficiency of the alpha chain of the
Intuitively one might expect hypothyroidism to be asso- stimulatory guanine nucleotide-binding protein (Gs) of
ciated with chemosensory deficits, since it is generally adenylyl cyclase, and an unusual constellation of skeletal
assumed that this disease produces alterations in other sen- and developmental abnormalities termed Albright heredi-
sory systems. Thus, 36–83% of hypothyroid patients are tary osteodystrophy (AHO). Among the abnormalities
said to experience somesthetic disturbances, and 25–45% expressed in AHO are obesity, short stature, brachydactyly,
auditory or visual dysfunction, with night blindness report- round faces, and subcutaneous ossifications. Type Ib PHP
edly being a common feature of the disorder (Mattes et al., patients, in contrast, exhibit a specific hormone resistance
1986). Nonetheless, this literature is somewhat limited, to parathyroid hormone, do not express a deficiency in Gs
and there is controversy as to whether objective measures protein activity, and do not have AHO.
of olfactory function are, in fact, altered in this disease. PHP was first reported to be associated with olfactory
In a pioneering study on this topic, McConnell et al. dysfunction in 1968 (Henkin, 1968). Subsequent studies
(1975) reported that odor-detection thresholds for pyri- reported that Type Ia PHP, but not Type Ib PHP, is accom-
dine and nitrobenezene, as well as various taste threshold panied by decreased olfactory function (Ikeda et al., 1988;
measures, were strikingly elevated in patients with Weinstock et al., 1986). The olfactory problem was
untreated hypothyroidism—a problem that readily believed to reflect the Gs protein deficiency of the Type Ia
resolved following thyroxine treatment. In contrast, patients, since Gs proteins are involved in the first stages of
Lewitt et al. (1989) found no detection threshold differ- olfactory transduction (Ikeda et al., 1988; Weinstock et al.,
ences between 16 patients and 17 controls for the odorant 1986). However, a more recent comprehensive study, while
phenyl ethyl alcohol and only a slight, but statistically confirming that Type Ib PHP patients exhibit altered olfac-
significant, difference on a suprathreshold test of odor tory function on three types of tests (odor identification,
identification. Various measures of suprathreshold taste detection threshold, and discrimination/memory), also
function, as well as auditory and visual evoked potentials, found Type 1b patients to have similar olfactory
did not differ between the controls and patients, and no dysfunction, making it unlikely that the olfactory dysfunc-
pre-/post-thyroxin differences were observed for any tion is related to the proposed Gs protein deficiency (Doty
measure. Similar negative findings have been reported by et al., 1997). In this same study, patients with pseudo-
others for various taste measures (Pittman and Beschi, pseudohypoparathyroidism (PPHP), a disorder found in
1967). some relatives of patients with Type Ia PHP, were found to
The reason for these discrepancies is not clear, although have relatively normal smell function. Since, like PHP Type
common test substances and procedures were not used Ia, PPHP is accompanied by a Gs protein deficiency and
across studies, and it is not apparent whether the patients by AHO, but not by a generalized end organ hormone
were comparable in terms of the nature, duration, and insensitivity, the smell loss of PHP Type Ia patients clearly
severity of their hypothyroid problems among the studies. cannot be explained on the basis of Gs. Thus, while we
It is of interest that well-controlled psychophysical studies now know that several forms of PHP exhibit olfactory loss,
employing rats have found no differences in either odor- or the mechanism for the loss is obscure.
taste-detection performance before or after the induction
of hypothyroidism (Brosvic et al., 1992, 1996) but have 6. Kallmann’s Syndrome
shown large differences in taste preference measures for
NaCl, HCl, and quinine, but not for sucrose, following While there are reports of smell loss in cases of hyper-
functional thyroidectomy (Brosvic et al., 1996). Although gonadotrophic hypogonadism (Males and Schneider, 1972),
472 Murphy et al.

cases of anosmia in association with idiopathic hypogo- chronic hallucinatory psychoses (Chapter 23), seasonal
nadotrophic hypogonadism (IHH) (Kallmann’s syndrome, affective disorder (Postolache et al., 1999, 2001), and
or KS) are much more common. KS is generally believed to severe stage anorexia nervosa (Fedoroff et al., 1995).
reflect an autosomal dominant mode of inheritance with Among the more interesting of such syndromes is “olfac-
incomplete expressivity (Quinton et al., 2001). Anomalies in tory reference syndrome,” a condition generally viewed as
addition to anosmia that are sometimes seen in KS patients distinct from schizophrenic or affective disorders (Bishop,
are cryptorchidism, midline craniofacial abnormalities, 1980; Pryse-Phillips, 1971). In this disorder, the patient
tooth agenesis, deafness, and renal anomalies. Although believes that smells emanate either from his or her own
women can have KS, it is much more prevalent in men. body (intrinsic hallucinations) or from elsewhere (extrinsic
Interestingly, family members related to individuals with hallucinations). The hallucinations can be “minimal,”
this syndrome may have both hypogonadism and anosmia; where the odor is complained about but the patient does not
others may have anosmia but normal gonadal function, take steps to remove it, or the “reasonable” reaction, in
whereas others may be entirely normal. Although MRI which the patient takes steps to eliminate the odor (e.g.,
studies generally reveal bilateral agenesis or dysgenesis of complaining to authorities or plugging up the chimney with
the olfactory bulbs and tracts (Bajaj et al., 1993; newspapers). In intrinsic hallucinatory cases, compulsive
Klingmuller et al., 1987; Yousem et al., 1993), there is one washing behavior, changing clothes, and restriction of
report of a IHH patient with aplasia of the left, but not the social activity is frequently seen. In many cases, this syn-
right, olfactory tract and bulb (Wustenberg et al., 2001). drome is closely linked with an obsessive-compulsive
The limited data available suggest that the olfactory disorder and is amenable to treatment with serotonin-uptake
epithelia of patients with KS have pathological changes inhibitors (Dominguez and Puig, 1997; O’Sullivan et al.,
analogous to those of mammals whose connections 2000; Stein et al., 1998).
between the olfactory fila and the olfactory bulb are sev-
ered; namely, morphologically immature receptor neurons
lacking cilia, a decrease in the number of olfactory receptor V. TREATMENT OF OLFACTORY DISORDERS
cells, and the formation of intraepithelial neuromas
(Schwob et al., 1993). In one nasal biopsy study, the total Meaningful treatments are available for some, but not all,
lack of an olfactory neuroepithelium in a KS subject patients whose olfactory dysfunction is conductive, i.e.,
was reported (Jafek et al., 1990a), although this observation resulting from blockage of airflow to the olfactory neuro-
must be viewed conservatively in light of sampling prob- epithelium. Obstruction may arise from several causes,
lems endemic to such biopsy studies (Paik et al., 1992). including allergic rhinitis, polyps, chronic sinusitis, or
Probably because of its association with altered olfac- rhinitis. Effective therapies for olfactory loss secondary
tory pathway anatomy, the anosmia noted in Kallmann’s to allergic rhinitis include allergy management, topical
syndrome is not reversed by either gonadotropin or gonadal cromolyn, topical and systemic corticosteroid therapies,
steroid therapy. However, because the testicular derange- and surgical procedures to reduce inflammation or
ment can be largely reversed by hormones when adminis- obstructions. A brief course of systemic steroid therapy
tered at the appropriate time, it is important that attempts be can be useful in distinguishing between conductive and
made to detect the endocrine problem as early as possible. sensorineural olfactory loss, as patients with the former
Sparkes et al. (1968) stress that anosmia serves as an impor- will often respond positively to some extent to the treat-
tant diagnostic marker in the early detection of such prob- ment (Davidson et al., 1987), although longer-term sys-
lems, and Mroueh and Kase (1968) suggest that perhaps all temic steroid therapy is not advised. Topical nasal steroids
pediatric patients presenting with anosmia should undergo are often ineffectual in returning smell function because
a specific gonadal evaluation to prevent the possible devel- the steroid fails to reach the affected regions in the upper
opment of irreversible atrophic changes. Unfortunately nasal passages. Increased efficacy presumably occurs
many individuals lacking smell ability (particularly young- when the nasal drops or spray are administered in the head-
sters who do not recognize the deficit or who are too shy to down Moffett’s position (Mott et al., 1997). In patients
mention it) do not seek help for this difficulty, and routine with obstructive inflammatory disease, swelling of the
pediatric examinations rarely test olfactory function. ostiomeatal complex can prevent drainage from the
sinuses, causing chronic sinusitis. Antibiotic therapy in
G. Psychiatric Disorders combination with control of the allergic symptoms under-
lying the inflammation is effective in many of these cases.
A number of psychiatric disorders are reportedly associated Resistant cases typically require surgery to improve
with altered smell function, including schizophrenia and drainage and clear infection.
Clinical Disorders of Olfaction 473

Patients with nasal polyps typically present with long- typically have dose-related improvement in olfactory
standing history of inflammation and congestion, and function and flavor sensation over time (Frye et al.,
many have a previous history of medical or surgical treat- 1990). Central lesions, such as CNS tumors that impinge
ment for the polyps. Though relief may be time limited in on olfactory bulbs and tracts and epileptogenic foci
many of these cases, surgery is indicated both to treat the within the medial temporal lobe that result in olfactory
obstructed breathing and to improve smell function. Even seizures, can often be resected in a manner that allows for
a moderate degree of ostiomeatal disease without accom- some restoration of olfactory function, as mentioned ear-
panying polyps can result in significant olfactory impair- lier in the chapter. Medications that induce distortions of
ment (Smith and Seiden, 1991). Septal deviation can olfaction can often be discontinued and replaced with
interfere with drainage at the osteomeatal complex, other types of medications or modes of therapy. Despite
contributing to the development of sinusitis. Septal devia- the fact there are advocates for zinc and vitamin thera-
tion can also obstruct airflow, and thus odorant flow, to the pies, there is no compelling evidence that these therapies
olfactory receptors. In patients with sinusitis secondary to work except in cases where frank zinc or vitamin defi-
inflammation of the ostiomeatal complex and septal devia- ciencies exist. Similarly, the employment of amino-
tion, endoscopic sinus surgery (ESS) combined with sep- phylline, in attempts to increase the level of cAMP within
toplasty (which also improves the approach for ESS) will the olfactory receptor cells, have no sound empirical
improve many of the symptoms related to the underlying basis for improving olfaction and have never been
disease (Davidson et al. 1995; Murphy et al., 2000). assessed relative to a placebo.
Surgical procedures that reduce nasal obstruction have Prognosis for recovery seems to be better for patients with
been demonstrated to improve olfactory function less severe hyposmia or microsmia than for those with anos-
(Ghorbanian et al., 1983; Ophir et al., 1986). The allevia- mia or severe microsmia. In some etiologies, this reflects the
tion of allergic disease in the ostiomeatal complex by less extensive damage into the basal cell layer of the epithe-
endoscopic ethmoidectomy can improve or restore olfac- lia and possibly less fibrosis around the foramina of the crib-
tory sensitivity (Hosemann et al., 2000; Seiden and Smith, riform plate through which the olfactory nerve axons pass.
1988). Thus, olfactory disorders caused by obstruction are An important component of therapy for many patients is the
often amenable to treatment (Davidson et al., 1995; Doty quantitative establishment of the true degree of olfactory
and Snow, 1987). Surgical patients may require ancillary loss. This places the patient’s problem into overall perspec-
treatment. In patients who remained anosmic after surgery tive; thus, it can be therapeutic for an older person to learn
for nasal and sinus polyps, Stevens (2001) reported that that, while his or her smell function is not what it used to be,
oral but not intranasal steriods were effective in restoring it still falls above the average of his or her peer group.
smell function to normal in most patients. Tomooka et al.
(2000) reported improved function with nasal irrigation ACKNOWLEDGMENTS
after endoscopic sinus surgery.
There is no treatment for smell loss secondary to This paper was supported, in part, by the following grants
congenital or other malformations of the olfactory bulbs or from the National Institutes of Health, Bethesda, MD: RO1
stalks. In general, the olfactory dysfunction due to sen- AG 04085, RO1 AG 27496, RO1 DC02064, RO1 DC
sorineural causes is difficult to manage, and the prognosis 02974, RO1 DC 04278, and PO1 DC 00161.
for patients suffering from long-standing total loss due to
upper respiratory illness or head trauma is poor. Most
patients who recover smell function subsequent to head REFERENCES
trauma do so within 12 weeks of injury (Costanzo et al.,
1995). Although there are no verified treatments for Akerlund, A., Bende, M., and Murphy, C. (1995). Olfactory
trauma-related olfactory loss, anti-inflammatory agents threshold and nasal mucosal changes in experimentally
may minimize posttraumatic sequelae in some cases. induced common cold. Acta Oto-Laryngol. 115:88–92.
Recent rat research suggests that application of nerve Amoore, J. E. (1971). Olfactory genetics and anosmia. In
Handbook of Sensory Physiology. Vol. IV. Chemical Senses.
growth factor onto the olfactory epithelium may alleviate
Part I, Beidler, L. M. (Ed.). Springer-Verlag, Berlin,
axotomy-induced degenerative changes in the olfactory
pp. 145–156.
receptor neurons, although it is not known whether this has Anderson, J., Maxwell, L., and Murphy, C. (1992). Odorant iden-
any functional consequence or if such a procedure in tification testing in the young child. Chem Senses 17:590.
humans would be efficacious (Kasuno et al., 2000). Apter, A. J., Gent, J. F., and Frank, M. E. (1999). Fluctuating
Tobacco smoking by itself rarely causes complete loss olfactory sensitivity and distorted odor perception in allergic
of the sense of smell, although patients who quit smoking rhinitis. Arch. Otolaryngol. Head Neck Surg. 125:1005–1010.
474 Murphy et al.

Babu, R., Barton, A., and Kasoff, S. S. (1995). Resection of olfac- ders, a study of 750 patients from the University of
tory groove meningiomas: technical note revisited. Surg. Pennsylvania Smell and Taste Center. Arch. Otolaryngol.
Neurol. 44:567–572. Head Neck Surg. 117:519–528.
Bajaj, S., Ammini, A. C., Marwaha, R., Gulati, P., Khetarpal, K., Delank, K. W. (1992). Die olfaktorische Sensitivität bei der
and Mahajan, H. (1993). Magnetic resonance imaging of the Rachenmandelhyperplasie. Laryngo-Rhino-Otologie 71:
brain in idiopathic hypogonadotropic hypogonadism. Clin. 293–297.
Radiol. 48:122–124. Delank, K. W., and Stoll, W. (1994). Die Riechfunktion vor und
Beard, M. D., and Mackay-Sim, A. (1987). Loss of sense of smell nach endonasaler Operation der Chronisch-polyposen. HNO
in adult, hypothyroid mice. Brain Res. 433:181–189. 42:619–623.
Bishop, E. R., Jr. (1980). An olfactory reference syndrome — Delank, K. W., and Stoll, W. (1998). Olfactory function after
monosymptomatic hypochondriasis. J. Clin. Psychiatry 41: functional endoscopic sinus surgery for chronic sinusitis.
57–59. Rhinology 36:15–19.
Blakeslee, A. F. (1918). Unlike reaction of different individuals to Dominguez, R. A., and Puig, A. (1997). Olfactory reference syn-
fragrance in verbena flowers. Science 48:298–299. drome responds to clomipramine but not fluoxetine: a case
Bromley, S. M., and Doty, R. L. (1995). Odor recognition mem- report. J. Clin. Psychiatry 58:497–498.
ory is better under bilateral than unilateral test conditions. Doty, R. L. (1995). The Smell Identification TestTM
Cortex 31:25–40. Administration Manual, 3rd ed. Sensonics, Inc., Haddon
Brosvic, G. M., and Rowe, M. M. (1992). Methyl xanthine, Heights, NJ.
adenosine, and human taste responsivity. Physiol. Behav. 52: Doty, R. L. (2000) Odor Threshold TestTM Administration
559–563. Manual. Sensonics, Inc., Haddon Heights, NJ.
Brosvic, G. M., Doty, R. L., Rowe, M. M., Harron, A., and Doty, R. L., and Bromley, S. M. (2002). Smell. In Diseases of the
Kolodiy, N. (1992). Influences of hypothyroidism on the taste Nervous System: Clinical Neuroscience and Therapeutic
detection performance of rats: a signal detection analysis. Principles, Asbury, A. K., McKhann, G., McDonald, W. I.,
Behav. Neurosci. 106:992–998. Goadsby, P. J., and McArthur, J. C. (Eds.). Cambridge
Brosvic, G. M., Risser, J. M., Mackay-Sim, A., and Doty, R. L. University Press, Cambridge, in press.
(1996). Odor detection performance in hypothyroid and Doty, R. L., and Mishra, A. (2001). Influences of nasal obstruc-
euthyroid rats. Physiol. Behav. 59:117–121. tion, rhinitis, and rhinosinusitis on the ability to smell.
Burdach, K. J., and Doty, R. L. (1987). The effects of mouth Laryngoscope 111:409–423.
movements, swallowing, and spitting on retronasal odor per- Doty, R. L., Snyder, P. J., Huggins, G. R., and Lowry, L. D.
ception. Physiol. Behav. 41:353–356. (1981). Endocrine, cardiovascular, and psychological corre-
Conn, F. W., and Mast, T. E. (1973). Adrenal insufficiency and lates of olfactory sensitivity changes during the human men-
electrophysiological measures of auditory sensitivity. Am. strual cycle. J. Comp. Physiol. Psychol. 95:45–60.
J. Physiol. 225:1430–1436. Doty, R. L., Shaman, P., Applebaum, S. L., Giberson, R.,
Costanzo, R. M., DiNardo, L. J., and Zasler, N. D. (1995). Head Siksorski, L., and Rosenberg, L. (1984). Smell identification
injury and olfaction. In Handbook of Olfaction and ability: changes with age. Science 226:1441–1443.
Gustation, Doty, R. L. (Ed.). Marcel Dekker, New York, Doty, R. L., Gregor, T., and Monroe, C. (1986). Quantitative
pp. 493–502. assessment of olfactory function in an industrial setting.
Cowart, B. J., Flynn-Rodden, K., McGeady, S. J., and Lowry, L. J. Occun. Med. 28:457–460.
D. (1993). Hyposmia in allergic rhinitis. J. Allergy Clin. Doty, R. L., Reyes, P. F., and Gregor, T. (1987). Presence of both
Immunol. 91:747–751. odor identification and detection deficits in Alzheimer’s dis-
Crysdale, W. S., Cole, P., and Emery, P. (1985). Cephalometric ease. Brain Res. Bull. 18:597–600.
radiographs, nasal airway resistance, and the effect of ade- Doty, R. L., and Ferguson-Segall, M. (1989). Influence of adult
noidectomy. J. Otolaryngol. 14:92. castration on the olfactory sensitivity of the male rat: a signal
Daniels, C., Gottwald, B., Pause, B. M., Sojka, B., Mehdorn, H. M., detection analysis. Behav. Neurosci. 103:691–694.
and Ferstl, R. (2001). Olfactory event-related potentials in Doty, R. L., Risser, J. M., and Brosvic, G. M. (1991). Influence
patients with brain tumors. Clin. Neurophysiol. 112: 1523–1530. of adrenalectomy on the odor detection performance of rats.
Davidson, T. M., and Murphy, C. (1997). Rapid clinical evalua- Physiol. Behav. 49:1273–1277.
tion of anosmia. The alcohol sniff test. Arch. Otolaryngol. Doty, R. L., McKeown, D. A., Lee, W. W., and Shaman, P. (1995).
Head Neck Surg. 123:591–594. A study of the test-retest reliability of ten olfactory tests.
Davidson, T. M., Jalowayski, A., Murphy, C., and Jacobs, R. D. Chem. Senses 20:645–656.
(1987). Evaluation and treatment of smell dysfunction. West. Doty, R. L., Marcus, A., and Lee, W. W. (1996). Development of
J. Med. 146:434–438. the 12-item cross-cultural smell identification test (CC-SIT).
Davidson, T. M., Murphy, C., and Jalowayski, A. A. (1995). Laryngoscope 106:353–356.
Smell impairment: Can it be reversed? Postgraduate Med. Doty, R. L., Fernandez, A. D., Levine, M. A., Moses, A., and
98:107–118. McKeown, D. A. (1997). Olfactory dysfunction in type I
Deems, D. A., Doty, R. L., Settle, R. G., Moore-Gillon, V., pseudohypoparathyroidism: dissociation from Gs alpha pro-
Shaman, P., Mester, A. F. et al. (1991). Smell and taste disor- tein deficiency. J. Clin. Endocrinol. Metabol. 82:247–250.
Clinical Disorders of Olfaction 475

Douek, E., Bannister, L. H., and Dodson, H. C. (1975). Recent Ghorbanian, S. N., Paradise, J. L., and Doty, R. L. (1983). Odor
advances in the pathology of olfaction. Proc. Roy. Soc. Med. perception in children in relation to nasal obstruction.
68:467–470. Pediatrics 72:510–516.
Downey, L. L., Jacobs, J. B., and Lebowitz, R. A. (1996). Golding-Wood, D. G., Holmstrom, M., Darby, Y., Scadding, G.
Anosmia and chronic sinus disease. Otolaryngol. Head Neck K., and Lund, V. J. (1996). The treatment of hyposmia with
Surg. 115:24–28. intranasal steroids. J. Laryngol. Otol. 110:132–135.
Duncan, H. J., and Seiden, A. M. (1995). Long-term follow-up of Goodspeed, R. B., Gent, J. F., and Catalanotto, F. A. (1987).
olfactory loss secondary to head trauma and upper respiratory Chemosensory dysfunction. Clinical evaluation results from a
tract infection. Arch. Otolaryngol. Head Neck Surg. taste and smell clinic. Postgrad. Med. 81:251–257.
121:1183–1187. Graff-Radford, N. R., Lin, S. C., Brazis, P. W., Bolling, J. P.,
Eichel, B. S. (1994). Improvement of olfaction following pansi- Liesegang, T. J., Lucas, J. A., et al. (1997). Tropicamide eye-
nus surgery. ENT J 73:248–250. drops cannot be used for reliable diagnosis of Alzheimer’s
el Naggar M., Kale, S., Aldren, C., and Martin, F. (1995). Effect disease. Mayo Clin. Proc. 72:495–504.
of Beconase nasal spray on olfactory function in post-nasal Gross-Isseroff, R., Ophir, D., Marshak, G., Ganchrow, J. R.,
polypectomy patients: a prospective controlled trial. Beizer, M., and Lancet, D. (1989). Olfactory function follow-
J. Laryngol. Otol. 109:941–944. ing late repair of choanal atresia. Laryngoscope 99:
Elsberg, C. A. (1935a). XI. The value of quantitative olfactory 1165–1166.
tests for the localization of supratentorial tumors of the brain. Henkin, R. I. (1967). Abnormalities of taste and olfaction in
A preliminary report. Bull. Neurol. Inst. NY 4: 511–522. patients with chromatin negative gonadal dysgenesis. J. Clin.
Elsberg, C. A. (1935b). XII. The localization of tumors of the Endocrinol. Metab. 27:1436–1440.
frontal lobe of the brain by quantitative olfactory tests. Bull. Henkin, R. I. (1968). Impairment of olfaction and of the tastes of
Neurol. Inst. NY 4:535–543. sour and bitter in pseudohypoparathyroidism. J. Clin.
Elsberg, C. A., and Levy, I. (1935). The sense of smell: I. A new Endocrinol. Metab. 28:624–628.
and simple method of quantitative olfactometry. Bull. Neurol. Henkin, R. I. (1975). The role of adrenal corticosteroids in sen-
Inst. NY 4:5–19. sory processes. In Handbook of Physiology, Blaschko, H.,
Ezeh, P. I., Myers, L. J., Hanrahan, L. A., Kemppainen, R. J., and Smith, A. D., and Sayers, G. (Eds.). American Physiological
Cummins, K. A. (1992). Effects of steroids on the olfactory Society, Washington, DC, pp. 209–230.
function of the dog. Physiol. Behav. 51:1183–1187. Henkin, R. I., and Bartter, F. C. (1966). Studies on olfactory
Fedoroff, I. C., Stoner, S. A., Andersen, A. E., Doty, R. L., and thresholds in normal man and in patients with adrenal cortical
Rolls, B. J. (1995). Olfactory dysfunction in anorexia and insufficiency: the role of adrenal cortical steroids and of
bulimia nervosa. Int. J. Eating Disord. 18:71–77. serum sodium concentration. J. Clin. Invest. 45:1631–1639.
Feron, F., Perry, C., McGrath, J. J., and Mackay, S. (1998). New Henkin, R. I., and Daly, R. L. (1968). Auditory detection and per-
techniques for biopsy and culture of human olfactory epithe- ception in normal man and in patients with adrenal cortical
lial neurons. Arch. Otolaryngol. Head Neck Surg. insufficiency: effect of adrenal cortical steroids. J. Clin.
124:861–866. Invest. 47:1269–1280.
Ferrante, M. A. (1999). Endogenous metabolic disorders. In Henkin, R. I., McGlone, R. E., Daly, R., and Bartter, F. C. (1967).
Textbook of Clinical Neurology, Goetz, C. G. and Pappert, E. J. Studies on auditory thresholds in normal man and in patients
(Eds.). W. B. Saunders Company, Philadelphia, pp. 731–767. with adrenal cortical insufficiency: the role of adrenal cortical
Fielder, C. P. (1985). The effect of adenoidectomy on nasal resis- steroids. J. Clin. Invest. 46:429–435.
tance to airflow. Acta Otolaryngol. 100:444–449. Hilberg, O. (1995). Effect of terfenadine and budesonide on nasal
Finelli, P. F., and Mair, R. G. (1991). Disturbances of taste and symptoms, olfction, and nasal airway patency following aller-
smell. In Neurology in Clinical Practice, Fenichel, G. M. and gen challenge. Allergy 50:683–688.
Marsden, C. D. (Eds.). Butterworth-Heinemann, Boston, pp. Hinriksdottir, I., Murphy, C., and Bende, M. (1997). Olfactory
209–216. threshold after nasal allergen challenge. J. Oto-Rhino-
Fitzsimon, J. S., Waring, S. C., Kokmen, E., McLaren, J. W., and Laryngol. Rel. Spec. 59:36–38.
Brubaker, R. F. (1997). Response of the pupil to tropicamide Hoffman, H. J., Ishii, E. K., and Macturk, R. H. (1998). Age-
is not a reliable test for Alzheimer disease. Arch. Neurol. related changes in the prevalence of smell/taste problems
54:155–159. among the United States adult population. Results of the 1994
Friedman, M., Tanyeri, H., Landsberg, R., and Caldarelli, D. disability supplement to the National Health Interview Survey
(1999). Effects of middle turbinate medialization on olfaction. (NHIS). Ann. NY Acad. Sci. 855:716–722.
Laryngoscope 109:1442–1445. Hoseman, W., Goertzen, W., Wohlleben, R., Wolf, S., and
Frye, R. E., Schwartz, B. S., and Doty, R. L. (1990). Dose-related Wigand, M. E. (2000). Olfaction after endoscopic endonasal
effects of cigarette smoking on olfactory function. JAMA ethmoidectomy. Am. J. Rhinol. 7:11–15.
263:1233–1236. Hummel, T., Sekinger, B., Wolfe, S. R., Pauli, E., and Kobal, G.
Furstenberg, A. C., Crosby, E., and Farrior, B. (1943). Neurologic (1997). ‘Sniffin’ Sticks’: olfactory performance assessed by
lesions which influence the sense of smell. Arch. Otolaryngol. the combined testing of odor identification, odor discrimina-
48:529–530. tion and olfactory threshold. Chem. Senses 22:39–52.
476 Murphy et al.

Hummel, T., Rothbauer, C., Barz, S., Grosser, K., Pauli, E., and Leonard, G., Cain, W. S., and Clavet, G. (1988). Surgical correc-
Kobal, G. (1988). Olfactory function in acute rhinitis. Ann. NY tion of olfactory disorders. Chem. Senses 13:708.
Acad. Sci. 855:616–624. Lewitt, M. S., Laing, D. G., Panhuber, H., Corbett, A., and Carter,
Ikeda, K., Sakurada, T., Sasaki, Y., Takasaka, T., and Furukawa, J. N. (1989). Sensory perception and hypothyroidism. Chem.
Y. (1988). Clinical investigation of olfactory and auditory Senses 14:537–546.
function in type I pseudohypoparathyroidism: participation Loo, A. T., Youngentob, S. L., Kent, P. F., and Schwob, J. E.
of adenylate cyclase system. J. Laryngol. Otol. 102: (1996). The aging olfactory epithelium: neurogenesis,
1111–1114. response to damage, and odorant-induced activity. Int. J. Dev.
Ikeda, K., Tabata, K., Oshima, T., Nishikawa, H., Hidaka, H., and Neurosci. 14:881–900.
Takasaka, T. (1999). Unilateral examination of olfactory Lund, V. J., and Scadding, G. K. (1994). Objective assessment of
threshold using the Jet Stream Olfactometer. Auris Nasus endoscopic sinus surgery in the management of chronic rhino-
Larynx 26:435–439. sinusitis: an update. J. Laryngol. Otol. 108:749–753.
Jafek, B. W., Moran, D. T., Eller, P. M., Rowley, J. C., and Jafek, Lygonis, C. S. (1969). Familial absence of olfaction. Heredity
T. B. (1987). Steroid-dependent anosmia. Arch. Otolaryngol. 61:413–416.
Head Neck Surg. 113:547–549. Males, J. L., and Schneider, R. A. (1972). Hypergonadotrophic
Jafek, B. W., Gordon, A. S., Moran D. T., and Eller, P. M. hypogonadism with anosmia. Acta Endocrinol. 71:7–12.
(1990a). Congenital anosmia. ENT J 69:331–337. Mattes, R. D., Heller, A. D., and Rivlin, R. S. (1986).
Jafek, B. W., Hartman, D., Eller, P. M., Johnson, E. W., Strahan, Abnormalities in suprathreshold taste function in early
R. C., and Moran, D. T. (1990b). Postviral olfactory dysfunc- hypothyroidism in humans. In Clinical Measurement of
tion. Am. J. Rhinol. 4:91–100. Taste and Smell, Meiselman, H. L., and Rivlin, R. S.
Jung, T. M., TerKonda, R. P., Haines, S. J., Stromes, S., and (Eds.). Macmillan Publishing Company, New York, pp.
Marentette, L. J. (1977). Outcome analysis of the transglabel- 467–486.
lar/subcranial approach for lesions of the anterior cranial McConnell, R. J., Menendez, C. E., Smith, F. R., Henkin, R. I.,
fossa: a comparison with the classic craniotomy approach. and Rivlin, R. S. (1975). Defects of taste and smell in patients
Otolaryngol. Head Neck Surg. 116:642–646. with hypothyroidism. Am. J. Med. 59:354–364.
Kern, R. C. (2000). Chronic sinusitis and anosmia: pathologic McCormack, L. J., and Harris, H. E. (1955). Neurogenic tumors
changes in the olfactory mucosa. Laryngoscope 110: 1071–1077. of the nasal fossa. J. Am. Med. Assoc. 157:318–321.
Kimmelman, C. P. (1994). The risk to olfaction from nasal Min, Y.-G., Yun, K.-S., Song, B. H., Cho, Y. S., and Lee, K. S.
surgery. Laryngoscope 104:981–988. (1995). Recovery of nasal physiology after functional endo-
Klimek, L., and Eggers, G. (1997). Olfactory dysfunction in scopic sinus surgery: olfaction and mucociliary transport.
allergic rhinitis is related to nasal eosinophilic inflammation. ORL 57:264–268.
J. Allergy Clin. Immunol. 100:158–164. Moll, B., Klimek, L., Eggers, G., and Mann, W. (1998).
Klimek, L., Moll, B., Amedee, R. G., and Mann, W. J. (1997). Comparison of olfactory function in patients with seasonal
Olfactory function after microscopic endonasal surgery in and perennial allergic rhinitis. Allergy 53:297–301.
patients with nasal polyps. Am. J. Rhinol. 11:251–255. Moller, R., Pause, B. M., and Ferstl, R. (1999). Inducibility of
Klingmuller, D., Dewes, W., Krahe, T., Brecht, G., and olfactory sensitivity by odor exposure of persons with specific
Schweikert, H.U. (1987). Magnetic resonance imaging of the anosmia [in German]. Z. Exp. Psychol. 46:53–59.
brain in patients with anosmia and hypothalamic hypo- Mott, A. E., and Leopold, D. A. (1991). Disorders in taste and
gonadism (Kallmann’s syndrome). J. Clin. Endocrinol. smell. Med. Clin. North Am. 75:1321–1353.
Metab. 65:581–584. Mott, A. E., Cain, W. S., Lafreniere, D., Leonard, G., Gent, J. F.,
Kondo, H., Matsuda, T., Hashibia, M., and Baba, S. (1998). A and Frank, M. E. (1997). Topical corticosteroid treatment of
study of the relationship between the T&T olfactometer and anosmia associated with nasal and sinus disease. Arch.
the University of Pennsylvania Smell Identification Test in a Otolaryngol. Head Neck Surg. 123:367–372.
Japanese population. Am. J. Rhinol. 12:353–358. Mroueh, A., and Kase, N. (1968). Olfactory-genital dysplasia.
Kosten, T., and Contreras, R. J. (1985). Adrenalectomy reduces Am. J. Obstet. Gynecol. 100:525–527.
peripheral neural responses to gustatory stimuli in the rat. Murphy, C., Anderson, J. A., and Markison, S. (1994).
Behav. Neurosci. 99:734–741. Psychophysical assessment of chemosensory disorders in
Lane, A. P., Zweiman, B., Lanza, D. C., Swift, D., Doty, R., et clinical populations. In Olfaction and Taste XI, Kurihara, K.
al. (1996). Acoustic rhinometry in the study of the acute Suzuki, N., and Ogawa, H. (Eds.). Springer-Verlag, Tokyo,
nasal allergic response. Ann. Otol. Rhinol. Laryngol. pp. 609–613.
105:811–818. Murphy, C., Davidson, T. M., Jellison, W., Austin, S., Mathews,
Lee, S. H., Lim, H. H., Lee, H. M., Park, H. J., and Choi, J. O. W. C., Ellison, D. W., and Schlotfeldt, C. (2000a). Sinonasal
(2000). Olfactory mucosal findings in patients with persistent disease and olfactory impairment in HIV disease: endoscopic
anosmia after endoscopic sinus surgery. Ann. Otol. Rhinol. sinus surgery and outcome measures. Laryngoscope
Laryngol. 109:720–725. 110:1707–1710.
Lehrner, J., and Deecke, L. (1999). Die Wiener Olfaktorische Murphy, C., Morgan, C. D., Geisler, M. W., Wetter, S.,
Testbatterie (WOTB). Akt. Neurol. 26:1–8. Covington, J. W., Madowitz, M. D., et al. (2000b). Olfactory
Clinical Disorders of Olfaction 477

event-related potentials and aging: normative data. Int. J. Scott, A. E., Cain, W. S., and Clavet, G. (1988). Topical cortico-
Psychophysiol. 36:133–145. steroids can alleviate olfactory dysfunction. Chem. Senses
Murphy, C., Schubert, C., and Cruickshanks, K. J. (2001). 13:735
Prevalence of olfactory impairment in the young old and the Seiden, A. M., and Smith, D. V. (1998). Endroscopic intranasal
oldest old: results of an epidemiological study. Chem. Senses surgery as an approach to restoring olfactory function. Chem.
26:1049. Senses 13:736.
Nordin, S., Monsch, A. U., and Murphy, C. (1995). Unawareness Simmen, D., Briner, H. R., and Hess, K. (1999). Screeningtest
of smell loss in normal aging and Alzheimer’s disease: dis- des Geruchssinness mit Riechdisketten. Laryngo-Rhino-
crepancy between self-reported and diagnosed smell sensitiv- Otologie 78:125–130.
ity. J. Gerontol. 50:187–192. Simola, M., and Malmberg, H. (1998). Sense of smell in allergic
Nordin, S., Lotsch, J., Kobal, G., and Murphy, C. (1998). Effects and nonallergic rhinitis. Allergy 53:190–194.
of nasal-airway volume and body temperature on intranasal Singh, N., Grewal, M. S., and Austin, J. H. (1970). Familial anos-
chemosensitivity. Physiol. Behav. 63:463–466. mia. Arch. Neurol. 22:40–44.
Nordin, S., Bramerson, A., Liden, E., and Bende, M. (1999). The Smith, D. V., and Seiden, A. M. (1991). Olfactory dysfunction.
Scandinavian Odor-Identification Test: development, reliabil- In The Human Sense of Smell, D. G. Laing, R. L. Doty, and
ity, validity and normative data. Acta Otolaryngol. W. Breipohl, (Eds.). Springer-Verlag, New York, pp.
118:226–234. 283–305.
Ophir, D., Gross-Isseroff, R., Lancet, D., and Marshak, G. Sparkes, R. S., Simpson, R. W., and Paulsen, C. A. (1968).
(1986). Changes in olfactory acuity induced by total inferior Familial hypogonadotropic hypogonadism with anosmia.
turbinectomy. Arch. Otolaryngol. Head Neck Surg. Arch. Intern. Med. 121:534–538.
112:195–197. Stein, D. J., Le Roux, L., Bouwer, C., and Van Heerden, B.
O’Sullivan, R. L., Mansueto, C. S., Lerner, E. A., and Miguel, E. (1998). Is olfactory reference syndrome an obsessive-com-
C. (2000). Characterization of trichotillomania. A phenome- pulse spectrum disorder?: two cases and a discussion.
nological model with clinical relevance to obsessive-compul- J. Neuropsychiat. Clin. Neurosci. 10:96–99.
sive spectrum disorders. Psychiatr. Clin. North Am. Stevens, C. N., and Stevens, M. H. (1985). Quantitative effects
23:587–604. of nasal surgery on olfaction. Am. J. Otolaryngol. 6:
Paik, S. I., Lehman, M. N., Seiden, A. M., Duncan, H. J., and 264–267.
Smith, D. V. (1992). Human olfactory biopsy. The influence Takagi, S. F. (1989). Human Olfaction. University of Tokyo
of age and receptor distribution. Arch. Otolaryngol. Head Press, Tokyo.
Neck Surg. 118:731–738. Tomooka, L. T., Murphy, C., and Davidson, T. M. (2000).
Pittman, J. A., and Beschi, R. J. (1967). Taste thresholds in hyper- Clinical study and literature review of nasal irrigation.
and hypothyroidism. J. Clin. Endocrinal. Metabol. Laryngoscope 110:1189–1193.
27:895–896. Weigel, M. T., Prazma, G., and Pillsbury, H. C. (1989). Auditory
Postolache, T. T., Doty, R. L., Wehr, T. A., Jimma, L. A., Han, L., acuity in adrenocorticoid insufficiency. Am. J. Otol. 10:
Turner, E. H., Matthews, J. R., Neumeister, A., No, C., 267–271.
Kroger, H., Bruder, G. E., and Rosenthal, N. E. (1999). Weinstock, R. S., Wright, H. N., Spiegel, A. M., Levine, M. A.,
Monorhinal odor identification and depression scores in and Moses, A. M. (1986). Olfactory dysfunction in humans
patients with seasonal affective disorder. J. Affect. Disord. with deficient guanine nucleotide-binding protein. Nature
56:27–35. 322:635–636.
Postolache, T. T., Wehr, T. A., Doty, R. L., Sher, L., Turner, E. H., Wright, H. N. (1987). Characterization of olfactory dysfunction.
Bartko, J., and Rosehthal, N. E. Patients with seasonal affec- Arch. Otolaryngol. Head Neck Surg. 113:163–168.
tive disorder have lower odor detection thresholds than con- Wustenberg, E. G., Fleischer, A., Gerbert, B., Abolmaali, N.,
trol subjects. Arch Psychiatry (in press). Huttenbrink, K. B., and Hummel, T. (2001). Normal sense of
Pryse-Phillips, W. (1971). An olfactory reference syndrome. Acta smell in Kallmann syndrome. A case report [in German].
Psychiatr. Scand. 47:484–509. Laryngo-Rhino-Otologie 80:85–89.
Quinton, R., Duke, V. M., Robertson, A., Kirk, J. M., Matfin, G., Wysocki, C. J., Dorries, K. M., and Beauchamp, G. K. (1989).
de Zoysa, P. A., et al. (2001). Idiopathic gonadotrophin defi- Ability to perceive androstenone can be acquired by ostensi-
ciency: genetic questions addressed through phenotypic char- bly anosmic people. Proc. Natl. Acad. Sci. USA 86:
acterization. Clin. Endocrinol. 55:163–174. 7976–7978.
Rydzewski, B., Pruszewicz, A., and Sulkowski, W. J. (2000). Yamagishi, M., Hasegawa, S., and Nakano, Y. (1988).
Assessment of smell and taste in patients with allergic rhini- Examination and classification of human olfactory mucosa in
tis. Acta Otolaryngol. 120:323–326. patients with clinical olfactory disturbances. Arch. Oto-Rhino-
Schiffman, S. S. (1983). Taste and smell in disease. N. Engl. J. Laryngol. 245:316–320.
Med. 308:1337–1343. Yamagishi, M., Fujiwara, M., and Nakamura, H. (1994).
Schwob, J. E., Szumowski, K. E., Leopold, D. A., and Emko, P. Olfactory mucosal findings and clinical course in patients
(1993). Histopathology of olfactory mucosa in Kallmann’s with olfactory disorders following upper respiratory viral
syndrome. Ann. Otol. Rhinol. Laryngol. 102:117–122. infection. Rhinology 32:113–118.
478 Murphy et al.

Yousem, D. M., Turner, W. J., Li, C., Snyder, P. J., and Doty, R. Yousem, D. M., Geckle, R. J., Bilker, W., McKeown, D. A., and
L. (1993). Kallmann syndrome: MR evaluation of olfactory Doty, R. L. (1996). MR evaluation of patients with congenital
system. Am. J. Neuroradiol. 14:839–843. hyposmia or anosmia. Am. J. Roentgenol. 166:439–443.
23

Odor Perception in Neurodegenerative Diseases

Richard L. Doty
University of Pennsylvania, Philadelphia, Pennsylvania, U.S.A.

I. INTRODUCTION with PD. Third, in some neurodegenerative diseases, such

as PD, scores on most olfactory tests are unrelated to dis-
Since the pioneering studies of Ansari and Johnson (1975) ease stage or progression, whereas in others, such as AD,
and Waldton (1974), it has become apparent that the abil- this appears not to be the case. And finally, it is now well
ity to smell is compromised in a number of neurodegener- established that in MS the number of plaques within
ative diseases, including Alzheimer’s disease (AD), Down olfactory-related CNS structures, but not in other brain
syndrome (DS), Huntington’s disease (HD), idiopathic regions, is strongly correlated with the degree of olfactory
Parkinson’s disease (PD), multiple sclerosis (MS), and the dysfunction.
parkinsonism-dementia complex of Guam (PDC) (for The goal of this chapter is to review the empirical data
reviews, see Doty, 1991, 2001; Ferreyra-Moyano and related to smell dysfunction in major neurological disorders,
Barragan, 1989; Hawkes et al., 1999; Mesholam et al., most of which are viewed as degenerative. Greater attention
1998; Moberg et al., 1999; Murphy, 1999; Serby, 1987). is paid to the those disorders for which the largest literature
Alterations in olfaction in such a wide range of disorders— has evolved (e.g., AD, PD, and SZ) and for which olfactory
along with the findings of olfactory dysfunction in the nor- dysfunction is relatively well characterized. Although some
mal elderly, epilepsy, multi-infarct dementia (MID), information on potential neuropathological correlates is pre-
schizophrenia (SZ), and brain surgery cases—raises the sented, the bulk of this information is provided in Chapter 24.
possibility that such olfactory anomalies simply reflect
nonspecific general disruption of central nervous system
(CNS) pathways. However, this is unlikely for several rea- II. OLFACTORY FUNCTION IN ALZHEIMER’S
sons. First, in some diseases, such as AD and PD, the DISEASE
olfactory deficit presents very early in the disease process,
long before significant brain deterioration is evident. The diagnosis of AD is a pathological diagnosis possible
Second, the degree of olfactory dysfunction differs, on only at autopsy. Probable AD, a diagnosis based upon a set
average, among most of these disorders. For example, AD, of well-defined clinical criteria (e.g., an idiopathic, slowly
PD, and PDC are accompanied by marked alterations in developing memory loss), is what is typically referred to as
the ability to smell, whereas HD, MID, and SZ are accom- AD in living persons (McKhann et al., 1984). Early diag-
panied by more moderate alterations. Progressive supra- nosis of neurodegenerative diseases such as AD is impor-
nuclear palsy (PSP) and 1-methyl-4-phenyl-1, 2, 3, tant, not only for long-term care, but for increasing the
6-tetrahydropyridine–induced parkinsonism (MPTP-P) are efficacy of medications that may thwart, to some degree,
associated with only minor changes in the ability to smell, the progression of the symptoms. Hence, markers for early
in spite of the fact that they share major clinical features AD are of significant value to the physician and patient.

479
480 Doty

odor-related activation of central structures (e.g., sub-

frontal temporal lobe), as measured by functional imaging
procedures such as positron emission tomography (PET)
(Buchsbaum et al., 1991; Kareken et al., 2001). Fourth,
the chemosensory loss is relatively specific to olfaction, as
most measures of taste function seem not to be involved
(Murphy et al., 1990) (see, however, Schiffman et al.,
1990). Fifth, many AD patients are unaware of their olfac-
tory deficit until formal testing (Doty et al., 1987; Nordin
et al., 1995a). For example, Doty et al. (1987) reported
that only 2 of 34 early-stage AD patients (6%) responded
affirmatively to the question, posed before olfactory test-
ing, “Do you suffer from smell and/or taste problems?,”
despite the fact that over 90% of the patients exhibited
lower scores on the standardized 40-item University of
Pennsylvania Smell Identification Test (UPSIT) than their
age-matched controls. Sixth, the olfactory loss of AD
appears, on average, to be equivalent to that observed in
PD and PDC, suggesting these disorders may share a
common pathological substrate (Doty et al., 1991a).
Seventh, as briefly noted earlier, despite the possibility of
such a substrate, the olfactory dysfunction of AD appears
to progress with time (Corwin and Serby, 1985; Knupfer
and Spiegel, 1986; Murphy et al., 1990; Nordin et al.,
1997; Richard and Bizzini, 1981; Serby et al., 1985),
unlike the situation with PD, where the smell loss seems
Figure 1 (A). University of Pennsylvania Smell Identification
to be more stable (Doty, 1991) (see, however, Tissingh et
Test (UPSIT) scores for patients with Alzheimer’s disease (AD)
and for age-, gender, and race-matched controls. (B). Detection al., 2001). Nonetheless, the reliability of psychophysical
threshold values for phenyl ethyl alcohol for patients with AD testing in significantly demented patients must be ques-
and for matched controls. Each dot signifies an individual sub- tioned. Eighth, olfactory testing is useful in the differen-
ject’s data point. Although some overlap appears between the AD tial diagnosis of AD from disorders commonly
and control subject data when plotted in this manner, very few of misdiagnosed as AD, such as depression. Indeed, even a
the AD subjects performed better than their matched controls. three-item odor identification test accurately discrimi-
(From Doty et al., 1987.) nates patients with AD from patients with major affective
disorder, being superior to the widely used 30-item Mini-
Mental State Examination (MMSE) in this regard
(McCaffrey et al., 2000; Solomon et al., 1998). Ninth,
A considerable amount is known about the olfactory there is evidence that smell loss is present in some inher-
dysfunction of AD. First, the loss—which is usually not ited forms of AD (e.g., Nee and Lippa, 2001) and, more-
total—is present in both nasal chambers and in the earli- over, that the loss in idiopathic AD correlates with a
est stages of the disease, including some cases of ques- family history of dementia (Schiffman et al., 1990),
tionable AD (Doty et al., 1987, 1991a; Morgan et al., implying some genetic linkage, a point discussed below.
1995; Schiffman et al., 1990) (Fig. 1). Second, the smell Interestingly, in light of evidence that postmenopausal
deficit is robust, present in a large proportion of patients estrogen may mitigate some forms of AD-related cogni-
(85–90%), and detected by a wide range of olfactory tests, tive decline (e.g., in verbal memory) (Henderson et al.,
including tests of odor identification, detection threshold 1996), there is circumstantial, but yet to be confirmed,
sensitivity, discrimination, and memory; however, evidence that estrogens may also lessen to some degree
suprathreshold intensity ratings may be less sensitive to postmenopausal olfactory loss (Deems et al., 1991;
the deficit (Table 1). In a recent meta-analysis of 11 olfac- Dhong et al., 1999).
tory/AD studies, effect sizes ranging from 0.98 to 12.15 Olfactory dysfunction—particularly in conjunction
(median
2.17) were noted (Mesholam et al., 1998). with other risk factors—is a predictor of subsequent devel-
Third, the olfactory dysfunction is reflected by decreased opment of AD in older persons. Five recent studies attest to
Odor Perception in Neurodegenerative Diseases 481

Table 1 Procedural Details of Alzheimer’s Disease Olfactory Function Studies

No. of subjects Mean age Findings
Ref. and sex (SD or range) Type of test Odorant types (p-value)
Waldton, 1974 66 F 72.2 (nr) Ability to “apprehend” 6 types D (nst)
Richard and
Bizzini, 1981 2 M, 6 F 74.3 (9.5) Detection threshold n-propanol D (nst)
Corwin and Serby,
1985 11 nr nr Odor ID 10 pairs D (0.02)
Peabody and Tinklenberg,
1985 14 M, 4 F 66 (53–79) Odor ID 5 types D (nst)a
Serby et al., 1985
(note: same data
as Corwin and
Serby, 1985) 11 nr nr Odor ID 10 pairs D (0.02)
Knupfer and Spiegel,
1986 6 M, 12 F 81.6 (5.8) Detection threshold Eucalyptol D (0.001)
Detection threshold Citral D (0.001)
Detection threshold Prunolide D (0.001)
Odor memory Multiple D (0.001)
Odor ID, naming Multiple D (0.001)
Warner et al., 1986 12 M, 5 F 66.7 (nr) UPSITb 40 types D (0.001)
Doty et al., 1987 14 M, 11 F 69.0 (8.5) UPSIT 40 types D (0.001)
6 M, 9 F 68.9 (9.3) Detection threshold Phenylethanol D (0.001)
Koss et al., 1987 10 nr nr UPSIT 40 types D (0.001)
nr Detection threshold Pyridine D (ns)
Koss et al., 1988 10 M 61.7 (8.7) UPSIT 40 types D (0.001)
Detection threshold Pyridine D (ns)
Moberg et al., 1987 42 nr 72.5 (7.8) Odor Memory Multiple D (0.001)
Rezek, 1987 18 nr 70.0 (4.8) Noncued ID 5 types D (0.002)
Cued ID 5 types D (0.01)
Detection threshold Pentanol D (ns)
Detection threshold Cinnamon oil D (0.001)
Kesslak et al., 1988 8 M, 10 F 64.2 (1.7) UPSIT 40 types D (0.05)
Odor matching 15 sets D (0.05)
Green et al., 1989 5 M, 7 F 68.4 (6.3) Odor intensity ratings Phenylethanol Normal
Odor intensity ratings Eugenol Normal
Odor matching 10 types D (0.05)
Murphy et al.,
1990 10 M, 11 F 72.8 (5.4) Detection threshold n-butanol D (0.001)
Schiffman et al.,
1990 30 (nr) 69.7 (6.9) Detection of supra-
threshold odorants 14 types D (nst)
Buchsbaum et al.,
1991) 4 M, 2 F 72.1 (3.5) Odor matching 30 trials D (0.001)
Serby et al., 1991 55 nr (60–79) UPSIT 40 types D (0.004)
Almkvist et al.,
1992 3 M, 11 F 72.6 (7.0) Detection threshold Pyridine D (0.01)
Perl et al., 1992 2 M, 18 F (67–91) Facial reaction 6 types D (0.04–0.002)c
Morgan et al.,
1995 15 M, 3 F 73.5 (9.5) UPSIT 40 types D (0.001)
SDCOIDd 8 types D (0.001)
Detection threshold n-butanol D (0.01)
Nordin et al.,
1995a 42 M, 38 F 74.0 (6.7) Detection threshold n-butanol D (0.01)
(continued)
482 Doty

Table 1 (continued)

No. of subjects Mean age Findings

Ref. and sex (SD or range) Type of test Odorant types (p-value)
Nordin and Murphy,
1996 4 M, 12 F 76.4 (10.4) Odor memory 10–15 types D (0.05)e
Detection threshold n-butanol D (0.05)
Nordin et al., 1997 4 M, 14 F 70.6 (8.7) Detection threshold pyridine D (0.007)
Bacon et al., 1998 8 nr 78.5 (8.5) Detection threshold n-butanol D (0.03)
Hawkes and Shephard,
1998 8 nr nr UPSIT 40 types D (0.001)
8 nr nr OERP 2 types D (nst)f
Solomon et al.,
1998 12 M, 8 F 74.5 (7.7) PSTg 3 types D (0.001)h
Lehrner and Deecke,
1999 20 nr 63–94 WOTBi 20 types D (0.001)
Larsson et al., 1999 11 F 69.7 (8.2) Odor identification 20 types D (0.001)
Detection threshold 1-butanol D (ns)
Bacon Moore et al.,
1999 22 M, 18 F 72.83 (7.61) Detection threshold n-butanol D (0.001)
Odor fluency 10 types D (0.05)
Niccoli-Waller et al.,
1999 15 M, 17 F 76.0 (9.1) Detection threshold n-butanol D (0.001)
McCaffrey et al.,
2000 7 M, 13 F 74.2 (7.9) PST 3 types D (0.001)j
Broggio et al.,
2001 8 M, 12 F 75 [66–80] Discrimination and
identification of orally
presented stimuli multiple D (0.001)
Gray et al., 2001 4M, 9F 75.4 [53–79] UPSIT 40 types D (0.001)
Kareken et al.,
2001 4 M, 3 F 73.1 (8.3) UPSIT 40 types D (0.001)
Detection threshold Phenylethanol D (ns)
McShane et al.,
2001 39 M, 53 F 75.6 (8.0) Ability to “perceive” 1 odorant: No differ-
lavender water ences
in “anosmia”
Royet et al., 2001 3 M, 12 F 68.4 (8.4) Identification 12 items D (0.005)
Odor intensity ratings 12 items ? (ns)
Pleasantness ratings 12 items D (0.025)k
Familiarity ratings 12 items D (0.05)
Edibility ratings 12 items D (ns)

Nearly all studies employed controls of the same age [and/or in the case of the University of Pennsylvania Smell Identification Test (UPSIT), standard-
ized test norms]; nst, no statistical test applied; nr, not reported; ns, not statistically significant; D, decreased performance in values relative to age-matched
controls (note that for detection thresholds, this means elevated threshold, not decreased thresholds, and that this is independent of whether effect is sta-
tistically significant); I, increased performance relative to controls.
aDepends upon odorant and measure employed.
bUniversity of Pennsylvania Smell Identification Test.
c8 of 18 AD patients were abnormal on test (44%) compared to 1 of 26 controls (4%).
dSan Diego Children’s Odor Identification Test.
e“Questionable AD subjects,” which requires absence of reported functional changes by significant others.
fFour of the 8 patients had normal OERPs, 4 had abnormal OERPs.
gPocket Smell Test™.
hComparison was against depressed patients who have no dysfunction.
iWiener Olfaktorischen Testbatterie.
jComparison was made against depressed normals.
kOnly for 2 of 12 odorants.
Odor Perception in Neurodegenerative Diseases 483

this fact. In the first, Bacon et al. (1998) administered an n- showed a moderate to strong sensitivity and specificity for
butanol detection threshold test to 70 normal volunteers at diagnosis of AD at follow-up. In the fourth study, Swan
risk for AD (i.e., older individuals who had memory and Carmelli (2002) administered the B-SIT to 359 non-
impairment). The sensitivity of those who subsequently demented elderly persons. Testing 41/2 yrs later found low
progressed to a diagnosis of AD (n
8) was lower than test scores to be associated with declines in verbal mem-
that of those who did not, implying that increased olfactory ory, independent of APOE-4 status, but not in executive
thresholds may be an indicator of incipient disease. control or global function. In the fifth study, Royal et al.
Importantly, of 15 patients who were diagnosed as having (2002) found that, in 173 initially nondemented indepen-
questionable AD, 9 with at least one APOE-4 allele were dent retirement community residents, decreased UPSIT
less sensitive than the remaining 6, who had no such allele, scores predicted cognitive decline and an AD-like memory
suggesting an association between this genetic risk factor impairment 3 years later.
for AD and smell dysfunction.* In the second, Graves et al. In light of such findings and the association of smell loss
(1999) administered the 12-item version of the UPSIT (the with the APOE-4 allele, it is of interest that some close rel-
Brief Smell Identification TestTM or B-SIT) to 1604 nonde- atives of AD patients may exhibit smell dysfunction. Serby
mented community-dwelling senior citizens 65 years of et al. (1996) administered the UPSIT to 28 first-degree rela-
age or older. Over a subsequent 2-year time period, the B- tives of 28 AD patients and 28 healthy controls. Despite sim-
SIT scores were a better predictor of cognitive decline than ilar global cognitive scores (MMSE), the relatives exhibited
scores on a global cognitive test. Persons who were anos- lower UPSIT scores than those of the controls ( p  0.01).
mic and possessed at least one APOE-4 allele had 4.9 Although olfactory testing was suggested by Nee and Lippa
times the risk of having cognitive decline than normosmic (2001) to not be useful in predicting the symptom onset of
persons not possessing this allele. This is in contrast to the one relatively rare genetically determined form of AD (pre-
1.23 times greater risk for cognitive decline in normosmic senilin-1 AD), the aforementioned data, along with evidence
individuals possessing at least one such allele. When the of poorer odor detection or identification performance in
data were stratified by sex, women who were anosmic and individuals having one or more APOE alleles (Bacon et al.,
possessed at least one APOE-4 allele had an odds ratio of 1998) and a delayed latency to an olfactory event–related
9.71, compared to an odds ratio of 1.90 for women who potential in individuals with the APOE-4 allele (Wetter and
were normosmic and possessed at least one allele. The Murphy, 2001), are in accord with the concept of a genetic
respective odds ratios for men were 3.18 and 0.67. The vulnerability to olfactory dysfunction in this disease.
third study evaluated, at 6-month intervals, the cognitive
and olfactory function of 90 outpatients with mild cogni-
tive impairment (Devanand et al., 2000). UPSIT scores III. OLFACTORY FUNCTION IN DOWN
were lower in patients with mild cognitive impairment SYNDROME
than in controls. Of 77 patients followed over a 2-year
period, those with mild, moderate, or severe smell loss, as Down syndrome, a trisomy 21 disorder, is the most common
well as those with low UPSIT scores who reported no clinical syndrome associated with mental retardation,
problems smelling, were more likely to develop AD. Low accounting for ~17% of the retarded population. In support of
UPSIT scores accompanied by lack of awareness of olfac- early clinical observations (e.g., Brousseau and Brainerd,
tory problems predicted the time until development of AD. 1928), empirial studies have found adult DS patients to have
Even in patients with relatively high MMSE scores, low difficulty smelling, as measured by tests of odor identifica-
UPSIT scores with lack of deficit awareness remained a tion, detection, memory, and EEG responses to odorants
significant predictor of AD. UPSIT scores of 30–35 (Hemdal et al., 1993; Murphy and Jinich, 1996; Warner et al.,
1988; Wetter and Murphy, 1999; Zucco and Negrin, 1994).
Such observations are of significance, given that the average
*Apoliopoprotein E is a widely distributed cholesterol transport
smell loss observed in DS is very close to that observed in
protein that circulates in the plasma after synthesis by the liver, AD (i.e., UPSIT scores ~20) (McKeown et al., 1996; Warner
spleen, and kidneys. In the CNS it is synthesized by
et al., 1988), and DS patients who live into early adulthood
macrophages, neurons, and glia. In the PNS it is synthesized by
macrophages, nonmyelinating Schwann cells, and ganglionic
inevitably develop the clinical and neuropathological fea-
satellite cells. Common isoforms include E2, E3, and E4, with E2 tures of AD (Oliver and Holland, 1986). Importantly, deposi-
seemingly having a neuroprotectant action and E4 being associ- tion of amyloid in the form of senile plaques or diffuse
ated with a higher risk of not only developing some neurodegen- amyloid deposits occurs in cortical brain regions associated
erative diseases, such as AD, but also having a more malignant with olfactory processing (e.g., entorhinal cortex) as early as
course of degeneration (Bedlack et al., 2000). the age of 19 years in DS (Hof et al., 1995).
484 Doty

At what age do individuals with DS begin exhibiting and discrimination (Table 2). A study examining the sensi-
olfactory loss? In the sole study designed to shed light on tivity and specificity of the UPSIT in differentiating between
this question, McKeown et al. (1996) administered the clinically diagnosed PD patients and normal controls found
UPSIT and a 16-item odor discrimination test to 20 adoles- these values to be quite high (0.91 and 0.88, respectively, in
cents with DS [mean age (SD)
13.89 years (1.98)], to 20 males 60 years of age) (Doty et al., 1995a). However, as is
non-DS retarded children matched on the basis of mental also true for AD, total anosmia is not the rule. Thus, in one
age [Peabody Picture Vocabulary Test-Revised (PPVT-R)] study only 13% of 38 patients who received an odor detec-
(Dunn, 1981), and to 20 nonretarded children also matched tion threshold test were unable to detect the highest odorant
on mental age. Although no meaningful differences in concentration presented, and only 38.3% of 81 patients had
olfactory function were found among the three study UPSIT scores suggestive of anosmia (Doty et al., 1988a).
groups, the test scores of both the DS and non-DS retarded Moreover, all but one of 41 PD patients asked if an odor was
subjects were markedly lower than those of the nonretarded present on each of 40 UPSIT items answered affirmatively to
children of their own chronological age. Moreover, the 35 or more of the items, even though the majority were
UPSIT scores were similar in magnitude to those of adult unable to identify most of the odors or felt that the perceived
DS subjects (~20). According to published norms (Doty, sensation did not correspond to the response alternatives.
1995), a 14-year-old nonretarded boy or girl would be Second, female PD patients generally have less dysfunction
expected to have an UPSIT score within the range of 34–38. than male PD patients (Stern et al., 1994), as is also the case
The respective average UPSIT scores of the DS and non-DS in AD. Third, while the loss is generally bilateral, there can
retarded subjects fell well below this range (18.65 and be slight individual differences in the degree to which the left
21.35, respectively). However, UPSIT scores of this magni- and right sides of the nose are involved. No association
tude fall within the low normal range for persons 6 years of exists, however, between the side of relatively greater
age (the average mental age of the subjects). Thus, an involvement and the side of hemiparkinsonism, as might be
UPSIT score of 19 falls at the 14th percentile of a group of expected if asymmetrical damage to striatal dopamine sys-
57 6-year-olds available from our Center’s database, tems were involved in the olfactory problem (Doty et al.,
whereas a score of 20 falls at the 17th percentile. Given the 1992b). Fourth, the smell loss is indistinguishable from that
fact, however, that the average UPSIT scores of these ado- of AD and the PDC, with UPSIT scores averaging around 20
lescents were equivalent to those of adult DS subjects, it is (Doty et al., 1991a,b). Fifth, the smell loss is unrelated to the
possible that the smell loss observed in the latter group magnitude of the motor symptoms (Doty et al., 1988a, 1989,
appears quite early in life—before significant AD-like brain 1992b), although subtle variations among so-called benign
pathology appears. Alternatively, it is conceivable that smell vs. malignant forms may be present (Stern et al., 1994).
function improves somewhat during the late teen years, only Sixth, the smell loss is unrelated to numerous neuropsycho-
to fall again as the AD-like neuropathology becomes more logical measures, such as the Randt memory test, reaction
salient. time, a finger-tapping test, and selected verbal and perfor-
mance subsets of the Wechsler Adult Intelligence Scale—
Revised (Doty et al., 1989). Seventh, the smell loss appears
IV. OLFACTORY FUNCTION IN IDIOPATHIC in both familial and sporadic forms of parkinsonism and may
PARKINSON’S DISEASE be a sign of the preclinical state of the disease (Berendse,
2001; Markopoulou et al., 1997). Eighth, unlike AD, there is
PD, since its first description by James Parkinson in 1817 no apparent longitudinal progression in olfactory dysfunc-
(Parkinson, 1817), has generally been considered to be a tion as occurs in other elements of the disease process (Doty
purely motor disease. However, within the last two decades et al., 1988a). Ninth, anti-PD medications (e.g., L-dopa,
or so it has become apparent that a number of sensory dopamine agonists, anticholinergic compounds) have
changes are present in this disorder, including subtle alter- absolutely no influence on the smell deficit, which occurs as
ations in vision and hearing (Gawel et al., 1981; Rodnitzky, severely in nonmediated or never-medicated patients as in
1998). Importantly, it is now clear that smell loss is a major medicated ones (Doty et al., 1992b; Quinn et al., 1987; Roth
component of this disease, with its prevalence (~90% of all et al., 1998). Tenth, also like AD, many PD patients are
tested cases) being greater than tremor, one of the cardinal unaware of their olfactory deficit until formal testing (Doty
signs of the disorder (Doty et al., 1988a). et al., 1988a). Eleventh, olfactory testing is useful in the dif-
Much is now known about the smell dysfunction of PD. ferential diagnosis of idiopathic PD from a number of other
First, as with the case of AD, the loss is generally bilateral neurodegenerative diseases with motor symptoms, including
and robust, being detected by a wide range of olfactory tests, disorders often misdiagnosed as PD (e.g., PSP, MPTP-
including tests of odor identification, threshold detection, induced PD, and essential tremor) (Busenbark et al., 1992;
Odor Perception in Neurodegenerative Diseases 485

Table 2 Procedural Details of Parkinson’s Disease Olfactory Function Studies

Mean age Findings
Ref. No. of subjects (SD or range) Type of test Odorant types (p-value)
Ansari and Johnson,
1975 22 M 58 (41 to 67) Detection threshold Pentyl acetate D (0.05)
Korten and Meulstee,
1980 39 M, 41 M 61.5 (nr) Questionnaire Unknown or NA D (0.001)
Ward et al., 1983 45 M, 27 F 60 (nr) Odor detection PEMECa D (0.01)
Detection threshold Pentyl acetate D (0.03)
Odor discrimination 4 types D (nst)
Corwin and Serby,
1985 5 (nr) nr Odor identification 10 pairs D (nst)
Serby et al., 1985 5 (nr) nr Odor identification 10 pairs D (0.05)b
Doty et al., 1988a 46 M, 35 F 67.4 (8.17) UPSITc 40 types D (0.0001)
Detection threshold Phenethanol D (0.0001)
Quinn et al., 1987 50 M, 28 F 61.5 (10.3) Detection threshold Pentyl acetate D (0.001)
Kesslak et al., 1988 8 M, 10 F 65.4 (3.2) UPSIT 40 types D (0.05)
Odor matching task 15 sets D (nst)
Bostantjopoulou et al.,
1991 22M, 22F 61.9 (9.4) Odor detection Pentyl acetate D (0.001)
Odor naming task Fresh coffee, D (0.01)
Murofushi et al., candy
1991 11 M, 7 F 59.6 (41–76) Detection threshold 5 types D (0.01, 0.05)d
recognition 5 types D (0.01, 0.05)e
threshold
Doty et al., 1992b 30 M, 10 F 61.9 (10.0) UPSIT 40 types D (0.001)
Hawkes and Shephard,
1993 49 M, 47 F 57 (27–81) UPSIT 40 types D (0.0001)
Stern et al., 1994 68 M, 50 F 64.3 (9.34) UPSIT 40 types D (0.001)
Doty et al., 1995a 109 M, 71 F 64 Ss  60 yr UPSIT 40 types D (0.001)
55 Ss: 61–71 yr
47 Ss  71 yr
Lehrner et al., 1995 13 (nr) 64.7 (11.4) Detection threshold n-butanol D (nr)
Odor identification 20 types D (nr)
Odor memory 20 types D (nr)
Barz et al., 1997 13 M, 18 F 66.8 (44–81) Odor discrimination 8 pairs D (ns)
Odor identification 8 types D (0.001)
OERPf vanillin, H2S D (0.05)g
Hawkes et al., 1997 49 M, 47 F 57 (27–81) UPSIT 40 types D (0.0001)
66 (sex nr) OERP H2S D (0.001)h
Ahlskog et al., 1998 7 M, 2 F 66.1 (6.4) UPSIT—Modified 20 types D (0.01)
Hawkes and Shephard, 78 M, 77 F 61 (nr) UPSIT 40 types D (0.0001)
1998 19 M, 17 F 62 (27–77) OERP vanillin, H2S D (nst)
Daum et al., 2000 28 M, 12 F 63.6 (8.7) SSi 16 types D (0.001)j
Detection threshold n-butanol D (.001)
(continued)
486 Doty

Table 2 (continued)

Mean age Findings

Ref No. of subjects (SD or range) Type of test Odorant types ( p-value)
Montgomery, Jr.
et al., 2000a 9 M, 9 F 64 (45–79) UPSIT 40 types D (0.001)
Montgomery, Jr.
et al., 2000b 32 M, 26 F, 1 U 69 (38–83) UPSIT 40 types D (0.001)
Tissingh et al., 2001 25 M, 16 F 56.4 (8.9) B-SIT 12 items D (0.001)
Odor discrimination 32 trials D (0.001)
Detection threshold Phenylethanol D (0.001)
Zucco et al., 2002 3 M, 3 F 65 (4.6) Odor Matching 12 types D (0.001)k
Identification 10 types
Müller et al., 2002 35 M, 15 F 58 (38–80) SSi 16 types D (0.0001)
Sobel et al., 2002 10 M, 10 F 67.1 (9.9) UPSIT 40 types D (0.0001)
Detection threshold Vanillin D (0.007)
Detection threshold Propionic Acid D (0.003)

Nearly all studies have employed controls of the same age (and/or, in the case of UPSIT, standardized test norms); nst, no statistical test applied; nr, not
reported; ns, not statistically significant; D, decreased performance in values relative to age-matched controls (note that for detection thresholds, this
means elevated threshold, not decreased thresholds, and that this is independent of whether effect is statistically significant); I, increased performance
relative to control; U, unknown.
aBased upon comparison of combined AD and PD groups vs. 5 other groups.
bPhenyl ethyl methyl ethyl carbinol.
cUniversity of Pennsylvania Smell Identification Test.
dSignificant effects were observed in detection for phenylethanol and methyl cyclopentenolone at the 0.01 level, and for isovaleric acid, gamma-unde-

calactone, and skatole at the 0.05 level.

eSignificant effects were observed in recognition for methyl cyclopentenolone and undecalactone at the 0.01 level. Significant effects were observed for

phenylethanol at the 0.05 level. No significant effects were noted for isovaleric acid or skatole.
fOdor event related potential.
gReflects prolongation of latencies.
hThe OERP responses were somewhat equivocal. Of 66 PD patients, 5 had unclear recordings and 1 had none. Of the remaining 60 patients, 37 had

OERPs to both CO2 and H2S. Of 10 PD patients with normal UPSIT scores, one had absent H2S responses and 3 had prolonged latencies to this agent.
All had intact CO2 responses. In some cases, OERPs obtained from right naris stimulation were significantly delayed relative to controls.
iSniffin’ Sticks test, which includes tests of odor identification, discrimination, and detection.
jAll p-values 0.001, except for discrimination on the left, which was 0.05.
kValues based on combination of both types of tests in an analysis of variance.

Doty et al., 1992a, 1993, 1995a). Finally, as in the case with clinical motor signs in PD.” Such findings reiterate the point
AD, some asymptomatic first-degree relatives of patients that olfactory testing, in conjunction with other measures, is
with either familial or sporadic forms of PD appear to exhibit likely useful in the early detection of PD.
olfactory dysfunction (Montgomery et al., 1999, 2000b; There is controversy as to whether some odorants are
Wolters et al., 2000). Recently, Berendse (2001) adminis- more useful than others in discerning PD patients from con-
tered tests of odor detection, identification, and discrimina- trols. An early study from my laboratory found no evidence
tion to 250 relatives of PD patients (~84% children, ~16% that any of the UPSIT items were more useful than others in
siblings, 1 parent). In 25 hyposmic and 23 normosmic indi- making this distinction (R. L. Doty, unpublished). However,
viduals sampled from this group, nigrostriatal dopaminergic Hawkes and Shephard (1993) administered the UPSIT to 70
function was assessed using single photon emission com- PD patients and 70 age-matched controls. The difference
puter tomography (SPECT) with [125I]-CIT as a dopamine between the percentage of subjects in each group correctly
transporter ligand. An abnormal reduction in striatal identifying each odor was computed. Significant differ-
dopamine transporter binding was present in 4 of the 25 ences were noted for the UPSIT items of pizza and winter-
(16%) hyposmic relatives, 2 of whom subsequently devel- green, leading these authors to conclude that “these
oped clinical parkinsonism, and in none of the 23 nor- observations raise the possibility that there is a congenital or
mosmic relatives. These authors noted, “The observation in acquired selective hyposmia in Parkinson’s disease, compa-
the present study that significant reductions in dopamine rable with the smell blindness to androstenone, for example,
transporter binding were found only in hyposmic relatives of which affects 20–47% otherwise healthy individuals and is
PD patients suggests that olfactory dysfunction may precede probably genetically determined.” Additional research is
Odor Perception in Neurodegenerative Diseases 487

needed to determine whether this interesting observation is quately, and could respond verbally to the questions of the
repeatable and, if so, whether it reflects differences in inten- examiner without difficulty. Neither the UPSIT nor the
sity among UPSIT items or true odorant-related factors. detection threshold scores differed between the MPTP-P
and normal groups, even though the patients with MPTP-P
had more advanced parkinsonism and contained a higher
V. OLFACTORY FUNCTION IN
proportion of cigarette smokers. As expected, the UPSIT
1-METHYL-4-PHENYL-1, 2, 3,
and PEA threshold scores of the young PD patients differed
6-TETRAHYDROPYRIDINE–INDUCED
significantly from those of the controls (Fig. 2). These data
PARKINSONISM
suggest that the functional integrity of the olfactory system
of MPTP-P patients is greater than that of PD patients and
If the hypothesis is true that the smell loss of most cases of
provided early evidence that olfactory dysfunction is not a
idiopathic PD is caused by an environmental agent (e.g., a
concomitant element of all parkinsonian syndromes.
virus) that enters the nose and directly or indirectly damages
elements of the olfactory system (the olfactory vector
hypothesis) (see Chapter 24), then parkinsonian syndromes VI. OLFACTORY FUNCTION IN THE
induced intravenously by known toxins might be expected to PARKINSONISM-DEMENTIA
be spared of olfactory dysfunction. One such syndrome, dis- COMPLEX OF GUAM
covered in the early 1980s, is that induced by intravenously
administered MPTP. Although MPTP does not readily cross Amyotrophic lateral sclerosis and parkinsonism-dementia
the blood-brain barrier, its toxic metabolite, 1-methyl-4- accounted for at least 15% of adult deaths among the
phenylpyridinium, or MPP, does so and produces a syn- Chamorro populations of Guam and Rota between 1957
drome in humans and nonhuman primates remarkably and 1965 (Reed and Brody, 1975). Epidemiological stud-
similar to that of idiopathic PD (Langston et al., 1983). ies, including case-control comparisons and extensive
In the sole study addressing olfactory function in patients pedigree analyses, have failed to identify a genetic cause of
with this rare form of parkinsonism, the UPSIT and a detec- these disorders (Kurland, 1988). However, more recent
tion threshold test for phenyl ethyl alcohol was adminis- data suggest that genetic susceptibility may be involved at
tered to six young persons suffering from this disorder some level. Thus, one study genotyped 12 patients with
(Doty et al., 1992a). Thirteen rare young PD patients and 10 PDC and 12 disease-free Chamorros for APOE alleles.
normal subjects served as comparison groups. Despite their Although no differences in APOE-4 frequencies were
major motoric deficits, the MPTP-P patients evidenced no found, the PDC group had a lower frequency of APOE-2
major decrements in cognitive function, could sniff ade- (8.3%) than did the controls (33.3%) (Waring et al., 1994).

Figure 2 University of Pennsylvania Smell Identification Test (UPSIT) and phenyl ethyl alcohol odor detection threshold test scores
for patient with MPTP-induced parkinsonism (MPTP-P), young patients with idiopathic Parkinson’s disease (PD), and matched normal
controls. (From Doty et al., 1992a.)
488 Doty

These general observations were confirmed in a bradykinesia frequently appear relatively early in its pro-
subsequent study in which an additional 17 PDC gression. The hallmark feature is vertical-gaze paresis
Chamorros were tested (Buee et al., 1996). (especially down-gaze paresis), which can be overcome by
It is now apparent that Guamanian Chamorros with the oculocephalic maneuver and is of supranuclear origin.
PDC have deficits in their ability to identify odors that are PSP is commonly misdiagnosed as PD, because it shares
similar to those of AD and PD patients. In the first of two so many motor features with PD. Unlike PD, however, the
studies on this topic, 24 Chamorros with PDC were admin- parkinsonian features are less responsive to anti-PD medi-
istered the UPSIT (Doty et al., 1991a). Half were from cations (Jackson et al., 1983), and PSP is typically charac-
Umatac and Merizo, two southern villages associated with terized by comparatively more frontal lobe dysfunction,
a high prevalence of PDC. The others were from other more neuronal degeneration within the basal ganglia and
Guamanian villages with lower PDC prevalence rates. upper brain stem, and less involvement of mesolimbic and
Even though these subjects were ambulatory and living mesocortical dopamine systems than PD (Cambier et al.,
with their families, all evidenced some degree of rigidity 1985; Jankovic, 1989).
and bradykinesia at the time of testing. For comparison, Unlike patients with PD or PDC, most patients with
UPSIT data from 24 AD and 24 PD North American PSP tend to have a relatively normal sense of smell,
patients who had similar levels of smoking behavior were although slight to moderate losses are present in some
matched to the PDC data on the basis of gender and age. individuals. In an early study on this topic, the UPSIT and
All subjects received a picture test analogous to the UPSIT a phenyl ethyl alcohol odor-detection threshold test were
to control for cultural and cognitive differences that might administered to 22 patients with PSP who scored well on
influence the UPSIT scores (termed the Picture the PIT. The test scores of these individuals were com-
Identification Test, or PIT). The UPSIT scores of the three pared to those from 22 PD patients and 22 neurologically
groups did not differ significantly from one another, normal age-, gender-, and race-matched controls. The per-
despite the fact that they were markedly lower than those formance of the PSP patients was clearly superior to that
obtained for normal persons of the same age and gender. As of the PD patients [respective UPSIT means (SD): 31.59
was done in an earlier PD study (Doty et al., 1988a) each (7.18) and 18.82 (6.94)], with approximately half scoring
participant was asked whether or not he or she suffered within the normal range. However, the PSP patients did
from any smell or taste problems prior to olfactory testing. exhibit moderate deficits relative to the controls, whose
Three of the PDC patients reported such problems (13%), mean UPSIT score was 35.60 (4.06).
as compared to two of the AD (8%) and three of the PD
(13%) patients, indicating that the level of awareness of the
VIII. OLFACTORY FUNCTION IN MULTIPLE
problem is similar, if not identical, in these three disorders.
SYSTEM ATROPHY
More recently, Ahlskog et al. (1998) administered an
abbreviated version of the UPSIT to 9 Chamorros with
Wenning et al. (1996) administered the UPSIT to 29 patients
symptoms of ALS, 9 with symptoms of pure parkinsonism,
with multiple system atrophy (MSA), 15 PSP patients, 118
11 patients with pure dementia, and 31 patients with PDC,
patients with idiopathic PD, 7 patients with corticobasal
as well as to neurologically normal Chamorro Guamanians
degeneration (CBD), and 123 healthy controls. Normal test
and 25 North American controls. The UPSIT scores were
scores were noted for the PSP and CBD patients and mild
markedly depressed in the four disease groups relative to the
impairment for the MSA patients relative to the controls.
controls and did not differentiate among the four syndromes
The authors noted that “preserved or mildly impaired olfac-
of Guamanian neurodegenerative disease. Some of the con-
tory function in a parkinsonian patient is more likely to be
trol subjects had lower scores than their North American
related to atypical parkinsonism such as MSA, PSP or CBD,
counterparts, implying the possibility of a subclinical neuro-
whereas markedly reduced olfaction is more suggestive of
generative disease process in nonsymptomatic individuals.
IPD PD.” These authors found that an UPSIT score of
25 was associated with a sensitivity of 77% and a specificity
of 85% in differentiating PD from atypical parkinsonism.
VII. OLFACTORY FUNCTION IN PROGRESSIVE
SUPRANUCLEAR PALSY
IX. OLFACTORY FUNCTION IN
Progressive supranuclear palsy (also termed the Steele, HUNTINGTON’S DISEASE
Richardson, and Olszewski syndrome) accounts for ~4%
of patients who exhibit parkinsonian symptoms. Tremor Huntington’s disease (also termed Huntington’s chorea) is
is rarely present in this disorder, although rigidity and a genetic disorder of dysfunctional movement, cognitive
Odor Perception in Neurodegenerative Diseases 489

deterioration, and altered behavior with autosomal domi- adults. Again, only the patients with clinical signs of
nant transmission that becomes phenotypically expressed HD exhibited depressed olfaction (mean UPSIT score
progressively relatively late in life. As functional capacity
27.4; SD
6.5). At what time in the disease process
worsens, chorea generally lessens and dystonia intensifies. olfactory deficits appear is not known, but if analogous to
Among its primary motor symptoms are hyperkinesias that other neurodegenerative diseases in which olfaction is
take, initially, the form of chorea, being characterized by compromised, presumably the loss occurs at a relatively
fleeting movements, which, in some cases, appear semi- early stage of clinical progression.
purposively within the context of overall heightened activ-
ity and motor restlessness (Shoulson, 1986).
X. OLFACTORY FUNCTION IN MULTIPLE
The first olfactory study of HD patients of which I am
SCLEROSIS
aware employed 38 HD patients and 38 controls, and
found HD to be associated with a deficit in the ability to
For a number of years it was assumed that little olfactory
remember odor qualities (Moberg et al., 1987). This
dysfunction was present in multiple sclerosis, since the pri-
problem was noted in early-affected patients with mini-
mary olfactory neurons are unmyelinated. Moreover, rela-
mal chorea or cognitive dysfunction and with normal ver-
tively early psychophysical studies, including one
bal and visual recognition memory performance.
employing 40 subjects (Ansari, 1976), found no olfactory
Subsequent work has shown that HD patients have decre-
deficits in this disorder. Indeed, even a comparatively
ments in the general ability to smell that could explain
recent study that administered the UPSIT and a match-to-
such deficits. For example, Nordin et al. (1995b) found
sample odor discrimination test to 14 MS patients reported
performance decrements in HD subjects on tests of odor
no deficits (Kesslak et al., 1988).
detection threshold, intensity discrimination, quality dis-
In 1984, our group found that 23% of 31 MS patients
crimination, and identification, with the greatest impair-
evaluated evidenced some degree of olfactory dysfunction
ment occurring for odor identification. These authors
on the UPSIT (Doty et al., 1984). A decade later we pre-
suggested these effects may be secondary to the odor-
sented case studies in which olfactory dysfunction was the
detection deficits. More recently, this same group com-
presenting symptom of MS (Constantinescu et al., 1994).
pared the n-butanol detection threshold sensitivity of 7
More recently we have demonstrated a strong inverse cor-
mildly affected HD patients to that of 7 age- and educa-
relation (r
 0.94) between UPSIT scores and the num-
tion-matched healthy controls (Hamilton et al., 1999).
ber of MS-related plaques within the subfrontal and
The subjects were also administered the California Odor
subtemporal lobes of patients, providing a physiological
Learning Test and the California Verbal Learning Test.
basis for the dysfunction (Fig. 3a) (Doty et al., 1997b,
Odor threshold sensitivity, but not group membership,
1998b). No such relationship was present between UPSIT
accounted for significant variance in total olfactory learn-
scores and plaques in other brain regions (Fig. 3b). These
ing. Both groups learned fewer items in the olfactory
observations have been subsequently replicated by others
modality than in the verbal modality, but retained a simi-
(Zorzon et al., 2000).
lar amount following a delay.
It would seem safe to assume that the reason why some
The question arises as to whether patients who carry
workers have not observed olfactory dysfunction in
the HD mutation and, hence, are at risk for expressing its
patients with MS is due to the dynamic nature of the dis-
symptoms have decreased smell function. Moberg and
ease. Recently, in a study of 5 MS patients tested three or
Doty (1997) administered the UPSIT and a PEA detec-
four times across an 18- to 20-month period, we demon-
tion threshold test to 25 probands with HD, 12 at-risk off-
strated that UPSIT scores wax and wane longitudinally in
spring, and 37 unrelated controls. Decreased olfaction
concert with the waxing and waning of the number of MS-
was noted only in the HD group, with a mean UPSIT
related plaques within the subtemporal and subfrontal
score of 24.8 (SD
8.7) and a PEA threshold score of
lobes (Doty et al., 1999).
4.4 log vol/vol (SD
1.4). Since HD is an autosomal
dominant genetic disorder with 100% penetrance,
approximately half of the at-risk offspring would have XI. OLFACTORY FUNCTION IN
been expected to have the mutation and to evidence smell AMYOTROPHIC LATERAL SCLEROSIS
dysfunction if it was a very early marker of HD. Bylsma
et al. (1997) extended these findings, testing 20 HD As with the case of PD, amyotrophic lateral sclerosis
patients who had the disease for a mean of 8.0 years (ALS) has been traditionally considered a motor neuron
(range: 4–14 years), 20 normal subjects with the genetic disease (MND). However, today we know that this is an
mutation that causes HD, and 20 mutation-negative oversimplification of this disorder. In 1991, Elian reported
490 Doty

observed in AD or PD, with the UPSIT scores falling

around 30.
In a subsequent study, Sajjadian et al. (1994) adminis-
tered the UPSIT bilaterally to 17 female and 20 male ALS
patients and unilaterally to 7 male and 7 female ALS
patients. Age-, gender-, smoking habit-, and race-matched
controls were also evaluated. While the UPSIT scores of
the ALS patients were significantly lower than those of the
controls, the degree of dysfunction was similar to that seen
by Elian. Thus, only 11% of the ALS patients had UPSIT
scores indicative of total or near total anosmia. Despite the
fact that nearly half of the ALS patients had UPSIT scores
that fell within the normal range, 75.7% of the patients
scored below their individually matched controls.
Although no sex differences or laterality in the ALS-
related test scores were observed, an age-related decre-
ment was present and, interestingly, significant
correlations were found between UPSIT scores and neuro-
physiological measures of peripheral nerve conductance. If
the latter observation proves to be true, it is conceivable
that a common pathophysiological process influences the
motor neuron responses and segments of the afferent olfac-
tory pathway.
Recently, Hawkes et al. (1998) administered the UPSIT
to 58 patients with MND and 135 controls. Additionally,
olfactory event–related potentials (OERPs) in response to
H2S were recorded in 15 patients, and the olfactory bulbs
of 8 MND cadavers were histologically examined.
Although UPSIT scores were slightly worse in the MND
patients, only the bulbar patients exhibited significant
decrements. OERPs were normal in 9 patients and delayed
in one. OERPs from the remaining 5 subjects were not able
to be recorded. Analysis of the olfactory bulb tissue
revealed excessive lipofuscin deposition in all 8 cases
examined, indicating subclinical neuronal damage. The
authors concluded that the olfactory dysfunction in this
disorder is relatively mild.

Figure 3 (A) Relationship, in patients with multiple sclerosis

(MS), between number of plaques in subtemporal and sub- XII. OLFACTORY FUNCTION IN
frontal lobes and scores on the University of Pennsylvania SCHIZOPHRENIA
Smell Identification Test (UPSIT). (B) No such relationship
was present between UPSIT scores and plaques in brain
Although SZ is not generally considered a neurodegenera-
regions outside of the primary olfactory cortical areas. (From
Doty et al., 1997b.) tive disease, it is included in this chapter because of evi-
dence that elements of this disease may have progressive
components and because a large literature exists regarding
that patients with MND exhibit decreased UPSIT scores olfactory function in this disorder. Interest in olfactory
bilaterally. Specifically, UPSIT scores from 9 male and 6 function in SZ has grown exponentially since Campbell
female MND patients of varying severity were compared and Gregson (1972) first attempted to test odor recognition
with those of age- and sex-matched controls and proved to memory in SZ patients. Indeed, there are now more olfac-
be significantly depressed. However, the magnitude of the tion-related studies in patients with SZ than in any other
deficit was more similar to that observed in SZ than that neurological disorder. This explosion in research reflects
Odor Perception in Neurodegenerative Diseases 491

the commercial availability of easy-to-use quantitative ers (Malaspina et al., 2002). Seventh, the olfactory loss is
olfactory tests (e.g., the UPSIT) and an acute awareness discernible to some degree by functional imaging studies
that temporal-limbic brain systems are markedly altered in (e.g., PET, SPECT), with hypometabolism or decreased
schizophrenia, making olfactory testing a unique func- activation occurring within central brain structures associ-
tional probe of these brain regions (Seidman et al., 1992) ated with smell function (e.g., subtemporal and subfrontal
(for reviews, see Harrison and Pearson, 1989; Martzke lobes), particularly on the right (Bertollo et al., 1996;
et al., 1997; Moberg et al., 1999; Pantelis and Brewer, Buchsbaum et al., 1991; Clark et al., 1991; Crespo-Facorro
1995; Serby et al., 1992). There is suggestion, for example, et al., 2001; Malaspina et al., 1998; Wu et al., 1993).
that odor-identification deficits may be indicative of right Eighth, the chemosensory decrement appears to occur early
orbitofrontal lobe dysfunction (Purdon, 1998). in the disease process and, in fact, is found in many patients
The following summarizes what is known about the who may be prone to SZ, such as those with schizotypy or
olfactory dysfunction of this disease. First, as in the case of with family histories positive for significant mental illness
AD and PD, the sensory alterations appear to be largely (Becker et al., 1993; Kopala et al., 1998b, 2001; Park and
bilateral, although recent data suggest that some left:right Schoppe, 1997). Ninth, there is some evidence that at least
differences may be present that could reflect subtypes of SZ some forms of the disorder may have a genetic basis
(Good et al., 2002). Second, the degree of dysfunction, on (Kopala et al., 1991). Tenth, UPSIT scores in SZ patients
average, is considerably less in SZ than in AD or PD are inversely and significantly correlated with disease dura-
(although this may not be the case in elderly individuals) tion, suggesting that some progression of pathology may be
(Moberg et al., 1997a). Thus, the median UPSIT score of occurring somewhere in the olfactory pathways (Moberg et
13 SZ studies reporting such data in a recent meta-analysis al., 1997b). This phenomenon appears to be independent of
was 32.5 (Moberg et al., 1999), compared to average medication history and gender. Presently, odor-identifica-
UPSIT scores in AD and PD studies of ~20. The median tion ability appears to be the only neuropsychological
effect size (i.e., the mean difference between patient and marker known to correlate with disease duration in SZ.
control groups expressed in SD units) for 17 SZ studies In the aforementioned meta-analysis (Moberg et al.,
employing the UPSIT was 0.92 (range: 0.56–3.18). 1999), studies tapping the domains of odor identification,
Analogous medians for 8 AD and 4 PD studies were 2.01 detection threshold sensitivity, discrimination, and mem-
(range: 0.98–12.15) and 1.66 (range: 1.27–2.38), respec- ory were assessed. Potential moderator variables such as
tively. The median effect size for the SZ studies was signif- age, gender, medication status, and smoking history were
icantly less than the median effect sizes of the AD and PD also examined. It is noteworthy that substantial olfactory
studies ( p
0.006 and 0.06, respectively), which did not deficits, across all sensory domains, were documented in
differ significantly from one another ( p
0.73). Such dif- patients with SZ. Indeed, no differential deficits were
ferentiation is due, in part, to a lower proportion of patients observed for odor identification, detection threshold sensi-
with SZ exhibiting smell loss. Thus, in recent study from tivity, discrimination, and memory (Table 3).* These data
our center, 23% of a group of 41 patients with SZ exhibited support the view that olfactory functions, as a whole, are
UPSIT scores indicative of microsmia or anosmia, in con- deficient in patients suffering from SZ. Outlier analysis did
trast to 94.4% of a group of 18 AD patients and 96.3% of a identify two significant outlier studies, in which both the
group of 54 PD patients (P. J. Moberg et al., unpublished). patients and controls were much older than those in the
Third, the lower olfactory test scores of SZ cannot be other studies. Calculation of effect sizes for these two stud-
explained on the basis of cigarette smoking, which is typi- ies found the composite effect size for the elderly SZ
cally higher in SZ than in non-SZ study populations patients to be over twice that seen in younger SZ patients.
(Brewer et al., 1996; Houlihan et al., 1994). Fourth, UPSIT Notably, these effect sizes reflect the differences between
scores in patients with SZ appear to be more strongly
*Although there is widespread consensus that identification abil-
related to neuropsychological measures of frontal lobe
function (e.g., problems in executive function) than to mea- ity, as measured by the UPSIT, is compromised in SZ (Table 3),
sures of medial temporal lobe function (e.g., memory) there has been some debate regarding other measures, for which
fewer studies exist. Bradley (1984) reported that psychotic
(Pantelis and Brewer, 1995; Purdon, 1998; see, however,
patients, most notably men with SZ, were hypersensitive to 5-16-
Good et al., 2002). Fifth, UPSIT scores are correlated with androsten-3-one. Other work has not confirmed this finding
negative, but not positive, symptoms of the disorder, (Gross-Isseroff et al., 1987). As can be seen in Table 3, either
reflecting associations with neuropsychological tasks of heightened thresholds (i.e., decreased sensitivity) or no signifi-
prefrontal cortical function (Brewer et al., 1996). Sixth, cant threshold deficits at all are the rule, the latter often reflecting
some subgroups of SZ patients (e.g., those with deficit syn- small sample sizes, threshold measures with low reliability, and
drome), may exhibit greater olfactory dysfunction than oth- compromised statistical power.
492 Doty

Table 3 Procedural Details of Studies of Olfactory Function in Patients with Schizophrenia

Sex and no. Mean age Odorant Direction of effect
Ref. of subjects (SD or range) Type of testing types and p-value
Campbell and Memory for odor
Gregson, 1972 20 (sex nr) nr positions 12 types D (nst)
Bradley, 1984 5 M, 6 F (20–50) Detection threshold Androstenone I (0.05)a
Sreenivasan et al., 1987 32 (sex nr) (15–45) Odor matching task nr D (nst)
Gross-Isseroff et al., 1987 22 M, 20 F 30.1 (6.4) Detection threshold Androsteone D (ns)
Detection threshold Pentyl acetate D (0.01)
Hurwitz et al., 1988 15 M, 3 F 23.9 (17–41) UPSIT 40 types D (0.02)
Dunn and Weller, 1989 13 M, 2 F 54.2 (28–71) Discrimination test 4 odor sets D (ns)
Kopala et al., 1989 26 M, 15 F UPSIT 40 types D (0.005)b
Serby et al., 1990 14 M (40–49) UPSIT 40 types D (0.001)
Detection threshold Geraniol D (0.001)
Warner et al., 1990 12 M 34 (20–42) UPSIT 40 types D (0.03)c
Geddes et al., 1991 16 M, 8 F 38.5 (20–66) Detection threshold Androstenone D (ns)d
Kopala et al., 1992 30 M, 10 F 26.25 (7.41) UPSIT detection
threshold 40 types D (0.01)
10 (nr) nr n-Butanol I (ns)
10 (nr) nr Detection threshold Phenylethanol D (ns)
Seidman et al., 1992 15 M, 1 F 36.5 (8.1) UPSIT 40 types
Wu et al., 1993 19 M, 1 F 32.1 (9.3) UPSIT 40 types D (0.01)
Odor memory 10 sets D (0.001)
Houlihan et al., 1994 23 M, 19 F 33.25 (6.8) UPSIT 40 types D (0.004)
Kopala et al., 1994 131 (nr) 34.6 (5.0) UPSIT 40 types D (0.001)
Malaspina et al., 1994 5 M, 6 F 34.6 (14.8) UPSIT 40 types D (0.05)
Kopala et al., 1995a 49 M, 16 F 28.3 (16–48) UPSIT 40 types D (0.02)
Kopala et al., 1995b 27 F 42.85 (7.04) UPSIT 40 types D (0.001)
Seidman et al., 1995 14 M, 4 F 39.8 (8.3) UPSIT 40 types D (0.006)
Brewer et al., 1996 27 M 31.8 (8.5) UPSIT 40 types D (0.001)
Moberg et al., 1997b 4 M, 12 F 77.9 (6.5) UPSIT 40 types D (0.001)
Moberg et al., 1997b 18 M, 20 F 50.6 (25.5) UPSIT 40 types D (0.001)
Seidman et al., 1997 24 M, 16 F 38.5 (6.5) UPSIT 40 types D (0.02)
Good et al., 1998 65 M nr UPSIT (unilateral) 40 types D (0.05)
Detection threshold Phenylethanol D (0.10)
(unilateral)
Kopala et al., 1998a 18 (sex nr) nr UPSIT detection 40 types D (0.001)
threshold n-Butanol D (0.05)e
Kopala et al., 1998b 12 (sex nr) 36.8 (4.9) UPSIT 40 types D (0.02)f
Malaspina et al., 1998 5 M, 1 F 34.6 (14.8) UPSIT 40 types D (0.05)
Purdon, 1998 18 M, 3 F 37.1 (9.6) UPSIT 40 types D (in micros-
mic range of
norms)
Stedman and Clair, 1998 37 M, 9 F 36.39 (8.05) UPSIT 40 types D (0.001)
Purdon and Flor-
Henry, 2000) 11 M, 6 F 27.7 (7.8) UPSIT (Unilateral) 40 types D (0.05)g
Detection threshold n-Butanol D
(unilateral)
Sirota et al., 1999 19 M 36.5 (11.9) Detection threshold Pentyl acetate D (0.0001)
Detection threshold Androstenone D (ns)
12 M 34.2 (11.9) Detection threshold Pentyl acetate I (0.04)h
Detection threshold Androstenone I (0.02)i
Purdon and Flor-
Henry, 2000 11 M, 6 F 27.65 (7.75) UPSIT 40 types D (0.05)
Detection threshold n-Butanol D (0.05)j
Brewer et al., 2001 55 M, 19 F 22.3 (3.67) UPSIT 40 types D (0.001)
Odor Perception in Neurodegenerative Diseases 493

Table 3 (continued)

Sex and no. Mean age Odorant Direction of effect

Ref. of subjects (SD or range) Type of testing types and p-value
Clark et al., 2001 20 M, 6 F 26.6 (7.9) UPSIT 40 types D (nst)k
Kohler et al., 2001 22 M, 18 F 31.8 (9.3) UPSIT 40 types D (0.008)
Detection threshold Phenylethanol D (ns)
Nearly all studies have employed controls of the same age (and/or, in the case of UPSIT, standardized test norms); nst, no statistical test applied; nr, not
reported; ns, not statistically significant; D, decreased performance in values relative to age-matched controls (note that for detection thresholds, this means
elevated threshold, not decreased thresholds, and that this is independent of whether effect is statistically significant); I, increased performance relative to
control;
aSignificance found only for males, not females.
bBased upon t-tests (two-tailed) computed by present author from Table 1 of their study. Although the authors note a sex difference in this work, both males

and females differed significantly from their sex-matched control groups.

cBased upon reanalysis of data performed by Hurowitz and Clark (1990).
dThe negative symptom group was less sensitive than the positive symptom group (p  0.01).
eThis was between a SZ group with severe polydipsia (n = 5) and a control group (n = 9). No significant difference was noted between a nonpolydipsic

SZ group (n = 5) and the controls (n
9).

fBased upon t-test (two-tailed) computed by present author from Table 1 of their study between affect twins and normal comparison group.
gThis was a unilateral test and the main group effect was diminished by a nostril by group interaction.
hBased upon t-test (one-tailed) between controls and a combined map of schizophrenic and brief psychotic disorders patients.
iBased upon t-test (one-tailed) between controls and first episode schizophrenic groups.
jBased upon t-test (two-tailed) computed by present author from Table 1 of their study.
kSeven of 26 subjects score in the norm-based microsmic range.

patients in each age group and their own age-matched con- study to the next), and a variety of drugs with diverse phar-
trols. As such, the observed differences in effect size macological actions were combined in the drug analysis,
between young and elderly SZ patients do not simply precluding an assessment of the specific pharmacological
reflect normal aging effects, but rather some specific mechanisms potentially involved. More recently, Purdon
decline in olfactory function presumably due to the disease and Flor-Henry (2000) obtained n-butanol thresholds from
process itself. These findings, in conjunction with the each side of the nose of 17 nonmedicated SZ patients and
inverse relationship observed between the duration of ill- 17 matched healthy controls. The patients exhibited an
ness and UPSIT scores, are in accord with the aforemen- asymmetrical impairment of the left naris that was not pre-
tioned hypothesis that some degenerative component is sent in the controls. In a second experiment, thresholds
present within olfaction-related brain regions of patients were measured in a new sample of 10 SZ patients before
with SZ. and after neuroleptic treatment. The asymmetry noted in
The issue of whether antipsychotic medications influ- the first study was replicated, and the relative advantage of
ence smell function in SZ has received considerable inter- the right naris shifted to a relative advantage of the left
est, although the vast majority of studies find no obvious naris over the course of the 8-week-long treatment period.
influence (e.g., Brewer et al., 2001; Hurwitz et al., 1988; These authors suggested that neuroleptic drugs may bene-
Kopala et al., 1992, 1998a; Malaspina et al., 1994; Moberg fit left hemisphere functions at a cost to right hemisphere
et al., 1997b; Wu et al., 1993). If drug effects are present, functions.
they are subtle and present for only some types of psy- There is some precedent from animal studies for the
chophysical measures. Sirota et al. (1999) reported that idea that drugs that alter dopaminergic pathways can influ-
threshold sensitivity to two odorants, isoamyl acetate and ence the ability to smell. For example, d-amphetamine
androstenone, was decreased in 19 male SZ patients dur- [a drug that releases norepinephrine (N E) and dopamine
ing a drug-free period and even more after initiation of (DA) and blocks their reuptake and degradation at nerve
antipsychotic drug treatment. Twelve male SZ patients terminals] enhances the odor-detection performance of rats
naïve to antipsychotic drugs reportedly had marginally for ethyl acetate (EA) at low doses and depresses such per-
increased threshold sensitivity, leading these authors to formance at somewhat higher, yet relatively low, doses
conclude that SZ-related olfactory deficits are due to (Doty and Ferguson-Segall, 1987). Since (1) marked atten-
antipsychotic medications. Considerable variation was uation or elimination of olfactory bulb NE (but not DA) by
present, however, in the threshold measures (e.g., in the intrabulbar injection of 6-hydroxydopamine (6-OHDA)
controls a 4-log2-step difference was present from one has no significant influence on such detection performance
494 Doty

(Doty et al., 1988b) and (2) depletion of cortical and olfac- (1993) similarly found, in 16 patients with left foci and 14
tory bulb NE by injection of 6-OHDA into the dorsal nor- patients with right foci, no detection threshold deficits on
adrenergic bundle influences neither the acquisition or either side of the nose, relative to controls, to the odorant
maintenance of a two-choice odor discrimination problem phenyl ethyl alcohol.
(Mair and Harrision, 1990), amphetamine’s effects prob- In contrast to detection threshold studies, a number of
ably depend upon dopaminergic, rather than noradrener- workers have reported suprathreshold deficits in epileptic
gic, pathways. patients. For example, Abraham and Mathai (1983) noted
Direct support for potential dopaminergic regulation of decreased performance on an odor-matching task in 28
olfactory sensitivity comes from a study using various epileptic patients with right-sided foci. Interestingly,
doses of the dopamine D2 receptor agonist quinpirole patients with left-sided foci did not exhibit the same diffi-
(Doty and Risser, 1989). A dose-dependent decrease in the culty in performing the bilaterally administered olfactory
odor-detection performance for ethyl acetate was found. matching task. Eskenazi et al. (1986) reported a small
Notably, pretreatment with the D2 receptor antagonist bilateral odor identification deficit in 18 epileptic patients
spiperone, which also is a potent 5HT2A antagonist, elimi- relative to 17 normal controls, which was apparently inde-
nated these effects. In subsequent work (Doty et al., pendent of the side of the seizure focus. More recently,
1998a), the influence of a D1-selective partial agonist SKF Carroll et al. (1993) found, relative to controls, a decrease
38393 on the odor-detection performance of rats was simi- in immediate odor memory to common odorants (e.g.,
larly evaluated. In a series of four experiments, it was vinegar, coconut, coffee, nail varnish, and garlic) in 10
found that, in contrast to quinpirole, SKF 38393 improved patients with right-sided temporal lobe epilepsy. Ten
odor-detection performance in rats and that this phenom- patients with left-sided epilepsy, as well as 10 epileptics
enon could be attenuated by the D1 receptor blocker SCH with a nontemporal lobe epileptic focus, did not show this
23390. This effect was found to be neither sex-specific nor deficit. This difference was seen only for nonnameable,
confined to the odor EA. These experiments suggest that uncommon odors.
D1 and D2 agonists can have differential influences on Three other studies have appeared on this general
odor-detection abilities in rats. topic. In the first, Martinez et al. (1993) found deficits in
Despite the aforementioned animal studies, however, odor discrimination, short- and long-term odor memory,
the influence of dopaminergic drugs on olfactory perfor- and odor naming in 21 epileptics relative to controls. No
mance in humans has received little attention, particularly differences between the left and right sides of the nose
in SZ. It would be of interest, for example, to know were present in these patients, who subsequently under-
whether Risperidone improves olfactory function. This went temporal lobe resection. In the second, West et al.
drug is a novel benzisoxazole antipsychotic agent (1993) also found, in 20 patients with left foci and 20
with potent combined serotonin 5HT2A and dopamine D2 with right foci, decrements on each side of the nose on
antagonistic properties. tests of odor identification and memory relative to
matched controls. In the third, prolonged OERP laten-
cies were noted in 10 patients with right-side foci
XIII. OLFACTION IN EPILEPSY after right-side odorant stimulation, and in 12 patients
with left-side foci after left-side odorant stimulation
Although early threshold studies reported heightened over- (Hummel et al., 1995).
all bilateral sensitivity in patients with epilepsy, particu- Recently, Kohler et al. (2001) administered the UPSIT
larly prior to an ictal event (e.g., Campanella et al., 1978; and a detection threshold test to 14 patients with right-
De et al., 1976; Dimov, 1973; Santorelli and Marotta, temporal lobe epilepsy, 18 patients with left-temporal lobe
1964), this phenomenon has not been subsequently epilepsy, 40 patients with SZ, and 25 healthy controls.
observed. Whether this reflects mitigating influences of While the detection threshold measure did not differ
modern antiseizure medications is not clear, but possible, among the four groups of subjects, the UPSIT scores of
since the vast majority of epileptic patients tested in more the right temporal lobe epilepsy patients and of the SZ
recent years have been on medications. patients were similar to one another and were signifi-
In the first study to test patients with respect to side of cantly lower than the UPSIT scores of the other two
epileptogenesis, Eskenazi et al. (1986) reported normal groups. These findings are in accord with the hypothesis
bilaterally determined n-butanol odor detection thresholds that the right hemisphere may play a special role in the
in 18 epileptic patients relative to 17 controls, an observa- processing of odors, as suggested from some neuroimag-
tion confirmed for this odorant for each side of the nose by ing studies (see Chapter 12). The reader is referred else-
Martinez et al. (1993) in 21 epileptic patients. West et al. where to the influences of seizure-controlling operations,
Odor Perception in Neurodegenerative Diseases 495

including commissurectomy, on measures of smell func- central than the olfactory neuroepithelium), the basis upon
tion (Doty et al., 1997a). which they arrive at this conclusion is suspect for a num-
ber of reasons. First, the vast majority of studies find
threshold deficits in AD patients, and even their own data
XIV. CONCLUSIONS AND CONSIDERATIONS point to this conclusion. Thus, three of their subjects (30%)
were either anosmic or hyposmic by their own criteria,
In this chapter, I have reviewed a number of studies and, in fact, the average threshold value for the AD patients
demonstrating olfactory dysfunction in a wide range of was higher (i.e., indicative of less sensitivity) than that of
neurological disorders. While the focus has been on those the controls, even though statistical significance at the 0.05
for which the most olfactory data are available, it should be a level was not achieved. Second, compared to other stud-
noted that olfactory dysfunction has been observed in a ies on this topic (see Table 1), their sample size was very
number of disorders not discussed in detail in this chapter. small, compromising statistical power. In general, single
Thus, smell loss is present in Pick’s disease (Richard and ascending series detection threshold tests such as adminis-
Bizzini, 1981), Kallmann’s syndrome (e.g., Yousem et al., tered by them have test-retest reliability values below 0.40,
1996), multi-infarct dementia (Knupfer and Spiegel, 1986; compared to those of the UPSIT, which are uniformly
Gray et al., 2001), restless leg syndrome (Adler et al., above 0.90 (Doty et al., 1995b). Hence, the power of such
1998), Korsakoff’s psychosis (e.g., Mair et al., 1983), alco- a threshold test to detect a deficit is much less than that of
holism (e.g., DiTraglia et al., 1991), seasonal affective dis- a 40-item odor-identification test and is particularly
order (Postolache et al., 1999), head trauma (e.g., Doty et compromised with small subject samples. Third, the gen-
al., 1997c), attention-deficit hyperractivity disorder eral assumption that threshold tests reflect peripheral dam-
(Gansler et al., 1998), stroke (e.g., Rousseaux et al., 1996), age and identification tests reflect central damage is
and human immunodeficiency or acquired immunodefi- untenable. Thus, central lesions, in fact, have been associ-
ciency syndrome (e.g., Brody et al., 1991), to name a few ated with threshold deficits by some authors (e.g.,
(for review, see Doty, 2001). It is noteworthy that a num- Rousseaux et al., 1996; Shelley and Shelley, 2000), and
ber of neurological or neurodegenerative disorders are peripheral (i.e., neuroepithelial) lesions are known to influ-
accompanied by no, or only modest, alterations in smell ence both tests of odor-detection threshold and odor iden-
function. In addition to those mentioned earlier in this tification (Deems et al., 1991).
chapter (e.g., MPTP-P), such disorders include essential
tremor (Busenbark et al., 1992), depression (Amsterdam et
ACKNOWLEDGMENTS
al., 1987), and panic disorder (Kopala and Good, 1996).
However, lack of involvement in neurological diseases
This work was supported, in part, by Grants PO1 DC
seems to be the exception rather than the rule, as most neu-
00161, RO1 DC 04278, RO1 DC 02974, RO1 AG 08148,
rodegenerative diseases appear to be associated with at
and RO1 AG 27496 from the National Institutes of Health,
least some degree of olfactory dysfunction. The reasons for
Bethesda, MD. I thank Dr. Paul Moberg for providing
the apparent predilection of the olfactory pathways for
some of the information employed is this review.
damage in such disorders are not clear (see Chapter 24).
It is important to note that several studies of AD and SZ
patients have reported lack of significant effects on odor- REFERENCES
detection threshold tests, in contrast to highly significant
effects on such tests as the UPSIT. In some such cases, the Abraham, A., and Mathai, K. V. (1983). The effect of right tem-
lack of a threshold effect has been assumed to signify that poral lobe lesions on matching of smells. Neuropsychologia
the olfactory system is, in effect, functioning normally and 21:277–281.
that the deficit on identification reflects a central olfactory Adler, C. H., Gwinn, K. A., and Newman, S. (1998). Olfactory
agnosia. A case in point for AD is work by Koss et al. function in restless legs syndrome. Movement Dis.
These investigators reported, in two publications based 13:563–565.
upon the same set of patients, significant UPSIT, but not Ahlskog, J. E., Waring, S. C., Petersen, R. C., Esteban-Santillan,
C., Craig, U. K., O’Brien, P. C., et al. (1998). Olfactory dys-
pyridine threshold, deficits in AD patients relative to con-
function in Guamanian ALS, parkinsonism, and dementia.
trols (Koss et al., 1987, 1988). They concluded that the Neurology 51:1672–1677.
olfactory deficit of AD is associated with a “central” rather Almkvist, O., Berglund, B., Nordin, S., and Wahlund, L. O.
than a “peripheral” anomaly. While, ultimately, their over- (1992). Is pyridine odor deficit related to progression of
all conclusion may be correct (i.e., that the cause of olfac- dementia in Alzheimer’s disease? :. Dept. Psychol. Stockholm
tory deficits in AD reflects damage in structures more Univ. 748:1–15.
496 Doty

Amsterdam, J. D., Settle, R. G., Doty, R. L., Abelman, E., and bolic rate during an olfactory memory task assessed by
Winokur, A. (1987). Taste and smell perception in depression. positron emission tomography in patients with dementia of
Biol. Psychiatry 22:1481–1485. the Alzheimer type and controls. Preliminary studies. Arch.
Ansari, K. A., and Johnson, A. (1975). Olfactory function in Gen. Psychiatry 48:840–847.
patients with Parkinson’s disease. J. Chronic Dis. 28:493–497. Buee, L., Perez-Tur, J., Leveugle, B., Buee-Scherrer, V., Mufson,
Ansari, K. A. (1976). Olfaction in multiple sclerosis. With a note E. J., Loerzel, A. J., et al. (1996). Apolipoprotein E in
on the discrepancy between optic and olfactory involvement. Guamanian amyotrophic lateral sclerosis/parkinsonism-
Eur. Neurol. 14:138–145. dementia complex: genotype analysis and relationships to
Bacon, A. W., Bondi, M. W., Salmon, D. P., and Murphy, C. neuropathological changes. Acta Neuropathol. 91: 247–253.
(1998). Very early changes in olfactory functioning due to Busenbark, K. L., Huber, S. I., Greer, G., Pahwa, R., and Koller,
Alzheimer’s disease and the role of apolipoprotein E in olfac- W. C. (1992). Olfactory function in essential tremor.
tion. Ann. NY Acad. Sci. 855:723–731. Neurology 42:1631–1632.
Bacon Moore, A. S., Paulsen, J. S., and Murphy, C. (1999). A test Bylsma, F. W., Moberg, P. J., Doty, R. L., and Brandt, J. (1997).
of odor fluency in patients with Alzheimer’s and Huntington’s Odor identification in Huntington’s disease patients and
disease. J. Clin. Exp. Neuropsychol. 21:341–351. asymptomatic gene carriers. J. Neuropsychiat. Clin. Neurosci.
Barz, S., Hummel, T., Pauli, E., Majer, M., Lang, C. J., and 9:598–600.
Kobal, G. (1997). Chemosensory event-related potentials in Cambier, J., Masson, M., Viader, F., et al. (1985). Le syndrome
response to trigeminal and olfactory stimulation in idiopathic frontal de la maladie de Steele-Richardson-Olszewski. Rev.
Parkinson’s disease. Neurology 49:1424–1431. Neurol. 48:1234–1239.
Becker, E., Hummel, T., Piel, E., Pauli, E., Kobal, G., and Campanella, G., Filla, A., and De, M. G. (1978). Smell and taste
Hautzinger, M. (1993). Olfactory event-related potentials in acuity in epileptic syndromes. Eur. Neurol. 17:136–141.
psychosis-prone subjects. Int. J. Psychophysiol. 15: 51–58. Campbell, I. M., and Gregson, R. A. M. (1972). Olfactory short
Bedlack, R. S., Strittmatter, W. J., and Morgenlander, J. C. term memory in normal, schizophrenic and brain-damaged
(2000). Apolipoprotein E and neuromuscular disease: a criti- cases. Aust. J. Psychol. 24:179–185.
cal review of the literature. Arch. Neurol. 57:1561–1565. Carroll, B., Richardson, J. T., and Thompson, P. (1993).
Berendse, H. W. B. (2001). Subclinical dopaminergic dysfunc- Olfactory information processing and temporal lobe epilepsy.
tion in asymptomatic Parkinson’s disease patients’ relatives Brain Cognition 22:230–243.
with a decreased sense of smell. Ann. Neurol. 50:34–41. Clark, C., Kopala, L., Hurwitz, T., and Li, D. (1991). Regional
Bertollo, D. N., Cowen, M. A., and Levy, A. V. (1996). metabolism in microsmic patients with schizophrenia. Can.
Hypometabolism in olfactory cortical projection areas of male J. Psychiatry 36:645–650.
patients with schizophrenia: an initial positron emission Clark, C., Kopala, L., Li, D. K., and Hurwitz, T. (2001). Regional
tomography study. Psychiatr. Res. 60:113–116. cerebral glucose metabolism in never-medicated patients with
Bostantjopoulou, S., Katsarou, Z., Mentenopoulos, G., and schizophrenia. Can. J. Psychiatry 46:340–345.
Logothetis, J. (1991). Olfactory disturbances in patients with Constantinescu, C. S., Raps, E. C., Cohen, J. A., West, S. E., and
Parkinson’s disease. Neurol. Psychiatr. 12:13–15. Doty, R. L. (1994) Olfactory disturbances as the initial or
Bradley, E. A. (1984) Olfactory acuity to a pheromonal substance most prominent symptom of multiple sclerosis. J. Neurol.
and psychotic illness. Biol. Psychiatry 19:899–905. Neurosurg. Psychiatry 57:1011–1012.
Brewer, W. J., Edwards, J., Anderson, V., Robinson, T., and Corwin, J., and Serby, M. (1985). Olfactory recognition deficit in
Pantelis, C. (1996). Neuropsychological, olfactory, and Alzheimer’s and Parkinsonian dementias. IRCS Med. Sci.
hygiene deficits in men with negative symptom schizophre- 13:260.
nia. Biol. Psychiatry 40:1021–1031. Crespo-Facorro, B., Paradiso, S., Andreasen, N. C., O’Leary, D.
Brewer, W. J., Pantelis, C., Anderson, V., Velakoulis, D., Singh, S., Watkins, G. L., Ponto, L. L., and Hichwa, R. D. (2001).
B., Copolov, D. L., and McGorry, P. D. (2001). Stability of Neural mechanisms of anhedonia in schizophrenia: a PET
olfactory identification deficits in neuroleptic-naive patients study of response to unpleasant and pleasant odors. JAMA
with first-episode psychosis. Am. J. Psychiatry 158:107–115. 286:427–435.
Brody, D., Serby, M., Etienne, N., and Kalkstein, D. S. (1991). Daum, R. F., Sekinger, B., Kobal, G., and Lang, C. J. (2000).
Olfactory identification deficits in HIV infection. Am. Riechprüfung mit “Sniffin’ Sticks” zur klinischen Diagnostik
J. Psychiatry 148:248–250. des Morbus Parkinson. Nervenarzt 71:643–650.
Broggio, E., Pluchon, C., Ingrand, P., and Gil, R. (2001). Taste De, M. G., Filla, A., and Campanella, G. (1976). Recent findings
impairment in Alzheimer’s disease [French]. Rev. Neuro. 157: on the olfactory acuity in epileptic patients [Italian]. Acta
409–413. Neurol. 31:250–256.
Brousseau, K., and Brainerd, H. G. (1928). Mongolism: A Study Deems, D. A., Doty, R. L., Settle, R. G., Moore-Gillon, V.,
of the Physical and Mental Characteristics of Mongolian Shaman, P., Mester, A. F., et al. (1991). Smell and taste disor-
Imbeciles. Baltimore: Williams and Wilkins. ders, a study of 750 patients from the University of
Buchsbaum, M. S., Kesslak, J. P., Lynch, G., Chui, H., Wu, J., Pennsylvania Smell and Taste Center. Arch. Otolaryngol.
Sicotte, N., et al. (1991). Temporal and hippocampal meta- Head Neck Surg. 117:519–528.
Odor Perception in Neurodegenerative Diseases 497

Devanand, D. P., Michaels-Marston, K. S., Liu, X., Pelton, G. Doty, R. L., Stern, M. B., Pfeiffer, C., Gollomp, S. M., and
H., Padilla, M., Marder, K., et al. (2000). Olfactory deficits Hurtig, H. I. (1992b). Bilateral olfactory dysfunction in early
in patients with mild cognitive impairment predict stage treated and untreated idiopathic Parkinson’s disease.
Alzheimer’s disease at follow-up. Am. J. Psychiatry 157: J. Neurol. Neurosurg. Psychiatry 55:138–142.
1399–1405. Doty, R. L., Golbe, L. I., McKeown, D. A., Stern, M. B., Lehrach,
Dhong, H. J., Chung, S. K., and Doty, R. L. (1999). Estrogen pro- C. M., and Crawford, D. (1993). Olfactory testing differenti-
tects against 3-methylindole-induced olfactory loss. Brain ates between progressive supranuclear palsy and idiopathic
Res. 824:312–315. Parkinson’s disease. Neurology 43:962–965.
Dimov, D. (1973). The condition of olfaction in epilepsy. Vestnik Doty, R. L., Bromley, S. M., and Stern, M. B. (1995a). Olfactory
Otorinolaringol. 35:22–23. testing as an aid in the diagnosis of Parkinson’s disease:
DiTraglia, G. M., Press, D. S., Butters, N., Jernigan, T. L., development of optimal discrimination criteria. Neuro-
Cermak, L. S., Velin, R. A., et al. (1991). Assessment of olfac- degeneration 4:93–97.
tory deficits in detoxified alcoholics. Alcohol 8:109–115. Doty, R. L., McKeown, D. A., Lee, W. W., and Shaman, P.
Doty, R. L. (1991). Olfactory dysfunction in neurogenerative dis- (1995b). A study of the test-retest reliability of ten olfactory
orders. In Smell and Taste in Health and Disease, Getchell, T. tests. Chem. Senses 20:645–656.
V., Doty, R. L., Bartoshuk, L. M., and Snow, J. B. Jr. (Eds.). Doty, R. L., Bromley, S. M., Moberg, P. J., and Hummel, T.
Raven Press, New York, pp. 735–751. (1997a). Laterality in human nasal chemoreception. In
Doty, R. L. (2001). Olfaction. Ann. Rev. Psychol. 52:423–452. Cerebral Asymmetries in Sensory and Perceptual Processing,
Doty, R. L., and Ferguson-Segall, M. (1987). Odor detection per- Christman, S. (Ed.). North Holland Publishing Co.,
formance of rats following d-amphetamine treatment: a signal Amsterdam, pp. 497–542.
detection analysis. Psychopharmacology 93:87–93. Doty, R. L., Li, C., Mannon, L. J., and Yousem, D. M. (1997b).
Doty, R. L., and Risser, J. M. (1989). Influence of the D-2 Olfactory dysfunction in multiple sclerosis. N. Engl. J. Med.
dopamine receptor agonist quinpirole on the odor detection 336:1918–1919.
performance of rats before and after spiperone administration. Doty, R. L., Yousem, D. M., Pham, L. T., Kreshak, A. A., Geckle,
Psychopharmacology 98:310–315. R., and Lee, W. W. (1997c). Olfactory dysfunction in patients
Doty, R. L., Shaman, P., and Dann, M. (1984). Development of with head trauma. Arch. Neurol. 54:1131–1140.
the University of Pennsylvania Smell Identification Test: a Doty, R. L., Li, C., Bagla, R., Huang, W., Pfeiffer, C., Brosvic,
standardized microencapsulated test of olfactory function. G. M., et al. (1998a). SKF 38393 enhances odor detection
Physiol. Behav. 32:489–502. performance. Psychopharmacology 136:75–82.
Doty, R. L., Reyes, P. F., and Gregor, T. (1987). Presence of both Doty, R. L., Li, C., Mannon, L. J., and Yousem, D. M. (1998b).
odor identification and detection deficits in Alzheimer’s dis- Olfactory dysfunction in multiple sclerosis: Relation to
ease. Brain Res. Bull. 18:597–600. plaque load in inferior frontal and temporal lobes. Ann. NY
Doty, R. L., Deems, D. A., and Stellar, S. (1988a). Olfactory dys- Acad. Sci. 855:781–786.
function in parkinsonism: a general deficit unrelated to neu- Doty, R. L., Li, C., Mannon, L. J., and Yousem, D. M. (1999).
rologic signs, disease stage, or disease duration. Neurology Olfactory dysfunction in multiple sclerosis: relation to longi-
38:1237–1244. tudinal changes in plaque numbers in central olfactory struc-
Doty, R. L., Ferguson-Segall, M., Lucki, I., and Kreider, M. tures. Neurology 53:880–882.
(1988b). Effects of intrabulbar injections of 6-hydroxy- Dunn, L. M. (1981). Peabody Picture Vocabulary Test-Revised
dopamine on ethyl acetate odor detection in castrate and non- Manual for Forms L and M. Circle Pines, MN: American
castrate male rats. Brain Res. 444:95–103. Guidance Service.
Doty, R. L., Riklan, M., Deems, D. A., Reynolds, C., and Stellar, Dunn, T. P., and Weller, M. P. I. (1989). Olfaction in schizophre-
S. (1989). The olfactory and cognitive deficits of Parkinson’s nia. Percept. Motor Skills 69:833–834.
disease: evidence for independence. Ann. Neurol. 25: Elian, M. (1991). Olfactory impairment in motor neuron disease:
166–171. a pilot study. J. Neurol. Neurosurg. Psychiatry 54:927–928.
Doty, R. L., Perl, D. P., Steele, J. C., Chen, K. M., Pierce, J. D., Eskenazi, B., Cain, W. S., Novelly, R. A., and Mattson, R. (1986).
Jr., Reyes, P., and Kurland, L. T. (1991a). Odor identification Odor perception in temporal lobe epilepsy patients with and
deficit of the parkinsonism-dementia complex of Guam: without temporal lobectomy. Neuropsychologia 24:553–562.
equivalence to that of Alzheimer’s and idiopathic Parkinson’s Ferreyra-Moyano, H., and Barragan, E. (1989). The olfactory
disease. Neurology 41:77–80. system and Alzheimer’s disease. Int. J. Neurosci. 49:157–197.
Doty, R. L., Perl, D. P., Steele, J. C., Chen, K. M., Pierce, J. D., Gansler, D. A., Fucetola, R., Krengel, M., Stetson, S., Zimering,
Jr., Reyes, P., and Kurland, L. T. (1991b). Olfactory dysfunc- R., and Makary, C. (1998). Are there cognitive subtypes in
tion in three neurodegenerative diseases. Geriatrics 46(Suppl adult attention deficit/hyperactivity disorder? J. Nery. Ment.
1): 47–51. Dis. 186:776–781.
Doty, R. L., Singh, A., Tetrude, J., and Langston, J. W. (1992a). Gawel, M. J., Das, P., Vincent, S., and Rose, F. C. (1981). Visual
Lack of olfactory dysfunction in MPTP-induced parkinson- and auditory evoked responses in patients with Parkinson’s
ism. Ann. Neurol. 32:97–100. disease. J. Neurol. Neurosurg. Psychiatry 44:227–232.
498 Doty

Geddes, J., Huws, R., and Pratt, P. (1991). Olfactory acuity in the Henderson, V. W., Watt, L., and Buckwalter, J. G. (1996).
positive and negative syndromes of schizophrenia. Biol. Cognitive skills associated with estrogen replacement in
Psychiatry 29:774–788. women with Alzheimer’s disease. Psychoneuroendocrinology
Good, K. P., Martzke, J. S., Honer, W. G., and Kopala, L. C. 21:421–430.
(1998). Left nostril olfactory identification impairment in a Hof, P. R., Bouras, C., Perl, D. P., Sparks, L., Mehta, N., and
subgroup of male patients with schizophrenia. Schizophr. Res. Morrison, J. H. (1995). Age-related distribution of neu-
33: 35–43. ropathologic changes in the cerebral cortex of patients with
Good, K. P., Martzke, J. S., Milliken, H. I., Horner, W. G., and Down’s syndrome. Arch. Neurol. 52:379–391.
Kopala, L. C. (2002). Unirhinal olfactory identification Houlihan, D. J., Flaum, M., Arnold, S. E., Keshavan, M., and
deficits in young male patients with schizophrenia and related Alliger, R. (1994). Further evidence for olfactory identifica-
disorders: association with impaired memory function. Schiz. tion deficits in schizophrenia. Schizophr. Res. 12:179–182.
Res. 56:211–223. Hummel, T., Pauli, E., Schuler, P., Kettenmann, B., Stefan, H.,
Graves, A. B., Bowen, J. D., Rajaram, L., McCormick, W. C., and Kobal, G. (1995). Chemosensory event-related potentials
McCurry, S. M., Schellenberg, G. D., and Larson, E. B. in patients with temporal lobe epilepsy. Epilepsia 36:79–85.
(1999). Impaired olfaction as a marker for cognitive decline: Hurwitz, T. A., and Clark, C. (1990). Response to Warner et al.,
interaction with apolipoprotein E epsilon4 status. Neurology Olfactory functioning in schizophrenia and depression. Biol.
53:1480–1487. Psychiatry 27:458.
Gray, A. J., Staples, V., Murren, K., Dhariwal, A., and Bentham, Hurwitz, T., Kopala, L., Clark, C., and Jones, B. (1988). Olfactory
P. (2001). Olfactory identification is impaired in clinic-based deficits in schizophrenia. Biol Psychiatry 23:123–128.
patients with vascular dementia and senile dementia of Jackson, J. A., Jankovic, J., and Ford, J. (1983). Progressive
Alzheimer type. Int. J. Geriatr. Psychiatry 16:513–517. supranuclear palsy: clinical features and response to treatment
Green, J. E., Songsanand, P., Peretz, S., Hsu, P., Coarkin, S., and in 16 patients. Ann. Neurol. 13:273–278.
Growdon, J. H. (1989). Dissociation between basic and high Jankovic, J. (1989). Parkinsonism-plus syndromes. Movement
order olfactory cacpities in Alzheimer’s disease. In Dis. 4:S95–S119.
Proceedings of the Fifth Meeting of the International Study Kareken, D. A., Doty, R. L., Moberg, P. J., Mosnik, D., Chen, S.
Group on the Pharmacology of Memory Disorders Associated H., Farlow, M. R., and Hutchins, G. D. (2001). Olfactory-
with Aging, Wurtman, R. J., Corkin, S. H., Growden, J. H., evoked regional cerebral blood flow in Alzheimer’s disease.
and Ritter-Walker, E. (Eds.). Center for Brain Sciences and Neuropsychology 15:18–29.
Metabolism Charitable Trust, Cambridge, MA, pp. 449–455. Kesslak, J. P., Cotman, C. W., Chui, H. C., van den Noort, S.,
Gross-Isseroff, R. G., Stoler, M., Ophir, D., Lancet, D., and Fang, H., Pfeffer, R., and Lynch, G. (1988). Olfactory tests as
Sirota, P. (1987). Olfactory sensitivity to androstenone in possible probes for detecting and monitoring Alzheimer’s dis-
schizophrenic patients. Biol. Psychiatry 22:922–925. ease. Neurobiol. Aging 9:399–403.
Hamilton, J. M., Murphy, C., and Paulsen, J. S. (1999) Odor Knupfer, L., and Spiegel, R. (1986). Differences in olfactory test
detection, learning, and memory in Huntington’s disease. performance between normal aged, Alzheimer and vascular
J. Int. Neuropsychol. Soc. 5:609–615. type dementia individuals. Int. J. Geriatr. Psychiatry 1:3–14.
Harrison, P. J., and Pearson, R. C. (1989). Olfaction and psychi- Kohler, C. G., Moberg, P. J., Gur, R. E., O’Connor, M. J.,
atry. Brit. J. Psychiatry 155:822–828. Sperling, M. R., and Doty, R. L. (2001). Olfactory dysfunc-
Hawkes, C. H., and Shephard, B. C. (1993). Selective anosmia in tion in schizophrenia and temporal lobe epilepsy.
Parkinson’s disease? Lancet 341:435–436. Neuropsychiatr. Neuropsychol. Behav. Neurol. 14:83–88.
Hawkes, C. H., and Shephard, B. C. (1998). Olfactory evoked Kopala, L. C., and Good, K. M. (1996). Odor identification ability in
responses and identification tests in neurological disease. patients with panic disorder. J. Psychiatr. Neurosci. 21: 340–342.
Ann. NY Acad. Sci. 855:608–615. Kopala, L., Clark, C., and Hurwitz, T. A. (1989). Sex differences
Hawkes, C. H., Shephard, B. C., and Daniel, S. E. (1997). in olfactory function in schizophrenia. Am. J. Psychiatry
Olfactory dysfunction in Parkinson’s disease. J. Neurol. 146:1320–1322.
Neurosurg. Psychiatry 62:436–446. Kopala, L. C., Clark, C. C., and Bassett, A. (1991). Olfactory
Hawkes, C. H., Shephard, B. C., Geddes, J. F., Body, G. D., and deficits in schizophrenia and chromosome 5 [letter]. Biol
Martin, J. E. (1998). Olfactory disorder in motor neuron dis- Psychiatry 29:732–733.
ease. Exp. Neurol. 150:248–253. Kopala, L. C., Clark, C., and Hurwitz, T. (1992). Olfactory
Hawkes, C. H., Shephard, B. C., and Daniel, S. E. (1999). Is deficits in neuroleptic naive patients with schizophrenia.
Parkinson’s disease a primary olfactory disorder? QJM Schizophr. Res. 8:245–250.
92:473–480. Kopala, L. C., Good, K. P., and Honer, W. G. (1994). Olfactory
Hemdal, P., Corwin, J., and Oster, H. (1993). Olfactory identifi- hallucinations and olfactory identification ability in patients
cation deficits in Down’s syndrome and idiopathic mental with schizophrenia and other psychiatric disorders. Schizophr.
retardation. Neuropsychologia 31:977–984. Res. 12:205–211.
Henderson, V. W. (1997). The epidemiology of estrogen replace- Kopala, L., Good, K., Martzke, J., and Hurwitz, T. (1995a).
ment therapy and Alzheimer’s disease. Neurology 48: Olfactory deficits in schizophrenia are not a function of task
S27–S35. complexity. Schizophr. Res. 17:195–199.
Odor Perception in Neurodegenerative Diseases 499

Kopala, L. C., Good, K., and Honer, W. G. (1995b). Olfactory Olfactory dysfunction in familial parkinsonism. Neurology
identification ability in pre- and postmenopausal women with 49:1262–1267.
schizophrenia. Biol. Psychiatry 38:57–63. Martinez, B. A., Cain, W. S., de Wijk, R. A., Spencer, D. D.,
Kopala, L. C., Good, K. P., Koczapski, A. B., and Honer, W. G. Novelly, R. A., and Sass, K. J. (1993). Olfactory functioning
(1998a). Olfactory deficits in patients with schizophrenia and before and after temporal lobe resection for intractable
severe polydipsia. Biol. Psychiatry 43:497–502. seizures. Neuropsychology 7:351–363.
Kopala, L. C., Good, K. P., Torrey, E. F., and Honer, W. G. Martzke, J. S., Kopala, L. C., and Good, K. P. (1997). Olfactory
(1998b). Olfactory function in monozygotic twins discordant dysfunction in neuropsychiatric disorders: review and
for schizophrenia. Am. J. Psychiatry 155:134–136. methodological considerations. Biol. Psychiatry 42:721–732.
Kopala, L. I., Good, K. P., Morrision, K., Bassett, A. S., Alda, M., McCaffrey, R. J., Duff, K., and Solomon, G. S. (2000). Olfactory
and Honer, W. G. (2001). Impaired olfactory identification in dysfunction discriminates probable Alzheimer’s dementia
relatives of patients with familial schizophrenia. Am. J. from major depression: a cross-validation and extension.
Psychiat. 158:1286–1290. J. Neuropsychiatr. Clin. Neurosci. 12:29–33.
Korten, J. J., and Meulstee, J. (1980). Olfactory disturbances in McKeown, D. A., Doty, R. L., Perl, D. P., Frye, R. E., Simms, I.,
parkinsonism. Clin. Neurol. Neurosurg. 82:113–118. and Mester, A. F. (1996). Olfactory function in young adoles-
Koss, E., Weiffenbach, J. M., Haxby, J. V., and Friedland, R. P. cents with Down’s syndrome. J. Neurol. Neurosurg.
(1987). Olfactory detection and recognition in Alzheimer’s Psychiatry 61:412–414.
disease. Lancet 1:622. McKhann, G. M., Drachman, D., Folstein, M., Katzman, R.,
Koss, E., Weiffenbach, J. M., Haxby, J. V., and Friedland, R. P. Price, D., and Stadlan, E. M. (1984). Clinical diagnosis of
(1988). Olfactory detection and identification performance Alzheimer’s disease: report of the NINCDS-ADRDA work
are dissociated in early Alzheimer’s disease. Neurology group under the auspices of Department of Health and Human
38:1228–1232. Services Task Force on Alzheimer’s disease. Neurology
Kurland, L. T. (1988). Amyotrophic lateral sclerosis and 34:939–944.
Parkinson’s disease complex on Guam linked to an environ- McShane, R. H., Nagy, Z., Esiri, M. M., King, E., Joachim, C.,
mental neurotoxin. Trends Neurosci. 11:51–54. Sullivan, N., and Smith, A. D. (2001). Anosmia in dementia is
Langston, J. W., Ballard, P. A., Tetrude, J. W., and Irwin, I. associated with Lewy bodies rather than Alzheimer’s pathol-
(1983). Chronic parkinsonism in humans due to a product of ogy. J. Neurol. Neurosurg. Psychiatry 70:739–743.
meperidine-analog synthesis. Science 219:979–980. Mesholam, R. I., Moberg, P. J., Mahr, R. N., and Doty, R. L.
Larsson, M., Semb, H., Winblad, B., Amberla, K., Wahlund, L. (1998). Olfaction in neurodegenerative disease: a meta-analy-
O., and Backman, L. (1999). Odor identification in normal sis of olfactory functioning in Alzheimer’s and Parkinson’s
aging and early Alzheimer’s disease: effects of retrieval sup- diseases. Arch. Neurol. 55:84–90.
port. Neuropsychology 13:47–53. Moberg, P. J., and Doty, R. L. (1997). Olfactory function in
Lehrner, J., and Deecke, L. (1999). Die Wiener olfaktorische Huntington’s disease patients and at-risk offspring. Int. J.
Testbatterie (WOTB). Akt. Neurol. 26:1–8. Neurosci. 89:133–139.
Lehrner, J., Brucke, T., Kryspin-Exner, I., Asenbaum, S., and Moberg, P. J., Pearlson, G. D., Speedie, L. J., Lipsey, J. R.,
Podreka, I. (1995). Impaired olfactory function in Parkinson’s Strauss, M. E., and Folstein, S. E. (1987). Olfactory recogni-
disease. Lancet 345:1054–1055. tion: differential impairments in early and late Huntington’s
Mair, R. G., and Harrison, L. M. (1990). Influences of drugs on and Alzheimer’s diseases. J. Clin. Exper. Neuropsychol.
smell function. In The Human Sense of Smell, Laing, D. G., 9:650–664.
Doty, R. L., and Breipohl, W. (Eds.). Springer, New York, Moberg, P. J., Doty, R. L., Mahr, R. N., Mesholam, R. I., Arnold,
pp. 335–359. S. E., Turetsky, B. I., and Gur, R. E. (1997a). Olfactory iden-
Mair, R. G., McEntee, W. J., and Doty, R. L. (1983). Olfactory tification in elderly schizophrenia and Alzheimer’s disease.
perception in Korsakoff’s psychosis: correlation with brain Neurobiol. Aging 18:163–167.
noradrenergic activity. Neurology 33:64–65. Moberg, P. J., Doty, R. L., Turetsky, B. I., Arnold, S. E., Mahr, R.
Malaspina, D., Wray, A. D., Friedman, J. H., Amador, X., Yale, N., Gur, R. C., et al. (1997b). Olfactory identification deficits
S., Hasan, A., et al. (1994). Odor discrimination deficits in in schizophrenia: correlation with duration of illness. Am.
schizophrenia: association with eye movement dysfunction. J. Psychiatry 154:1016–1018.
J. Neuropsychiatr. Clin. Neurosci. 6:273–278. Moberg, P. J., Agrin, R., Gur, R. E., Gur, R. C., Turetsky, B. I.,
Malaspina, D., Perera, G. M., Lignelli, A., Marshall, R. S., Esser, and Doty, R. L. (1999). Olfactory dysfunction in schizo-
P. D., Storer, S., et al. (1998). SPECT imaging of odor identi- phrenia: a qualitative and quantitative review. Neuropsycho-
fication in schizophrenia. Psychiatr. Res. 82:53–61. pharmacology 21:325–340.
Malaspina, D., Coleman, E., Goetz, R. R., Harkavy-Friedman, J., Montgomery, E. B., Jr., Baker, K. B., Lyons, K., and Koller, W.
Corcoran, C., Amador, X., Yale, S., and Gorman, J. M. (2002). C. (1999). Abnormal performance on the PD test battery by
Odor identification, eye tracking and deficit syndrome schiz- asymptomatic first-degree relatives. Neurology 52:757–762.
ophrenia. Biol. Psychiat. 51:809–815. Montgomery, E. B., Jr., Koller, W. C., LaMantia, T. J., Newman,
Markopoulou, K., Larsen, K. W., Wszolek, E. K., Denson, M. A., M. C., Swanson-Hyland, E., Kaszniak, A. W., and Lyons, K.
Lang, A. E., Pfeiffer, R. F., and Wszolek, Z. K. (1997). (2000a). Early detection of probable idiopathic Parkinson’s
500 Doty

disease: I. Development of a diagnostic test battery. Perl, E., Shay, U., Hamburger, R., and Steiner, J. E. (1992). Taste-
Movement Dis. 15:467–473. and odor-reactivity in elderly demented patients. Chem.
Montgomery, E. B., Jr., Lyons, K., and Koller, W. C. (2000b). Senses 17:779–794.
Early detection of probable idiopathic Parkinson’s disease: II. Postolache, T. T., Doty, R. L., Wehr, T. A., Jimma, L. A., Han, L.,
A prospective application of a diagnostic test battery. Turner, E. H., et al. (1999). Monorhinal odor identification
Movement Dis. 15:474–478. and depression scores in patients with seasonal affective dis-
Morgan, C. D., Nordin, S., and Murphy, C. (1995). Odor identi- order. J. Affect. Dis. 56:27–35.
fication as an early marker for Alzheimer’s disease: impact of Purdon, S. E. (1998). Olfactory identification and Stroop inter-
lexical functioning and detection sensitivity. J. Clin. Exp. ference converge in schizophrenia. J. Psychiatr. Neurosci. 23:
Neuropsychol. 17:793–803. 163–171.
Müller, A., Reichmann, H., Livermore, A., and Hummel, T. Purdon, S. E., and Flor-Henry, P. (2000). Asymmetrical olfactory
(2002). Olfactory function in idiopathic Parkinson’s disease acuity and neuroleptic treatment in schizophrenia. Schizophr.
(IPD): results from cross-sectional studies in IPD patients and Res. 44:221–232.
long-term follow-up of de-novo IPD patients. J. Neural Trans. Quinn, N. P., Rossor, M. N., and Marsden, C. D. (1987).
109:805–811. Olfactory threshold in Parkinson’s disease. J. Neurol.
Murofushi, T., Mizuno, M., Osanai, R., and Hayashida, T. (1991). Neurosurg. Psychiatry 50:88–89.
Olfactory dysfunction in Parkinson’s disease. J. Oto-Rhino- Reed, D. M., and Brody, J. A. (1975). Amyotrophic lateral scle-
Laryngol. Rel. Spec. 53:143–146. rosis and parkinsonism-dementia on Guam 1945–1972. I.
Murphy, C. (1999). Loss of olfactory function in dementing dis- Descriptive epidemiology. Am. J. Epidemiol. 101:287–301.
ease. Physiol. Behav. 66:177–182. Rezek, D. L. (1987). Olfactory deficits as a neurologic sign in
Murphy, C., and Jinich, S. (1996). Olfactory dysfunction in dementia of the Alzheimer type. Arch. Neurol. 44:1030–1032.
Down’s syndrome. Neurobiol. Aging 17:631–637. Richard, J., and Bizzini, L. (1981). Olfaction et demences. Premiers
Murphy, C., Gilmore, M. M., Seery, C. S., Salmon, D. P., and resultats d’une etude clinique et experimentale avecle n-
Lasker, B. R. (1990). Olfactory thresholds are associated with propanol. Acta Neurol. Belg. 81:833–851.
degree of dementia in Alzheimer’s disease. Neurobiol. Aging Rodnitzky, R. L. (1998). Visual dysfunction in Parkinson’s dis-
11:465–469. ease. Clin. Neurosci. 5:102–106.
Nee, L. E., and Lippa, C. F. (2001). Inherited Alzheimer’s disease Roth, J., Radil, T., Ruzicka, E., Jech, R., and Tichy, J. (1998).
PS-1 olfactory function: a 10-year follow-up study. Am. J. Alz. Apomorphine does not influence olfactory thresholds in
Dis. Other Dementias 16:83–84. Parkinson’s disease. Funct. Neurol. 13:99–103.
Niccoli-Waller, C. A., Harvey, J., Nordin, S., and Murphy, C. Rousseaux, M., Muller, P., Gahide, I., Mottin, Y., and Romon, M.
(1999). Remote odor memory in Alzheimer’s disease: deficits (1996). Disorders of smell, taste, and food intake in a patient
as measured by familiarity. J. Adult Dev. 6:131–136. with a dorsomedial thalamic infarct. Stroke 27:2328–2330.
Nordin, S., and Murphy, C. (1996). Impaired sensory and cogni- Royall, D. R., Chiodo, L. K., Polk, M. S., and Jaramillo, C. J.
tive olfactory function in questionable Alzheimer’s disease. (2002). Severe dysosmia is specifically associated with
Neuropsychology 10:113–119. Alzheimer-like memory deficits in non-demented elderly
Nordin, S., Monsch, A. U., and Murphy, C. (1995a). retirees. Neuroepidemiology 21:68–73.
Unawareness of smell loss in normal aging and Alzheimer’s Royet, J. P., Croisile, B., Williamson-Vasta, R., Hibert, O.,
disease: discrepancy between self-reported and diagnosed Serclerat, D., and Guerin, J. (2001). Rating of different olfac-
smell sensitivity. J. Gerontol. 50:187–192. tory judgements in Alzheimer’s disease. Chem. Senses
Nordin, S., Paulsen, J. S., and Murphy, C. (1995b). Sensory- and 26:409–417.
memory-mediated olfactory dysfunction in Huntington’s dis- Sajjadian, A., Doty, R. L., Gutnick, D. N., Chirurgi, R. J., Sivak,
ease. J. Int. Neuropsychol. Soc. 1:281–290. M., and Perl, D. (1994). Olfactory dysfunction in amyotrophic
Nordin, S., Almkvist, O., Berglund, B., and Wahlund, L. O. lateral sclerosis. Neurodegeneration 3:153–157.
(1997). Olfactory dysfunction for pyridine and dementia pro- Santorelli, G., and Marotta, A. (1964). La sogli olfattometrica
gression in Alzheimer disease. Arch Neurol. 54:993–998. dell’epilettico in condizioni di base e dopa crisi. Rivista Osp-
Oliver, C., and Holland, A. J. (1986). Down’s syndrome and Psichiatr. 32:185–190.
Alzheimer’s disease: a review. Psychol. Med. 16:307–322. Schiffman, S. S., Clark, C. M., and Warwick, Z. S. (1990).
Pantelis, C., and Brewer, W. J. (1995). Neuropsychological and Gustatory and olfactory dysfunction in dementia: not specific
olfactory dysfunction in schizophrenia: relationship of frontal to Alzheimer’s disease. Neurobiol. Aging 11:597–600.
syndromes to syndromes of schizophrenia. Schizophr. Res. Seidman, L. J., Talbot, N. L., Kalinowski, A. G., McCarley, R.
17:35–45. W., Faraone, S. V., Kremen, W. S., et al. (1992).
Park, S., and Schoppe, S. (1997). Olfactory identification deficit Neuropsychological probes of fronto-limbic system dysfunc-
in relation to schizotypy. Schizophr. Res. 26:191–197. tion in schizophrenia. Olfactory identification and Wisconsin
Parkinson, J. (1817). An Essay on the Shaking Palsy. Sherwood, Card Sorting performance. Schizophy. Res. 6:55–65.
Neely and Jones, London. Seidman, L. J., Oscar-Berman, M., Kalinowski, A. G., Ajilore,
Peabody, C. A., and Tinklenberg, J. R. (1985). Olfactory deficits and O., Kremen, W. S., Pepple, J. R., and Tsuang, M. T. (1995).
primary degenerative dementia. Am. J. Psychiatry 142:524–525. Experimental and clinical neuropsychological measures of
Odor Perception in Neurodegenerative Diseases 501

prefrontal dysfunction in schizophrenia. Neuropsychology Swan, G. E., Carmelli, D. (2002). Impaired olfaction
9:481–490. predicts cognitive decline in nondemented older adults.
Seidman, L. J., Goldstein, J. M., Goodman, J. M., Koren, D., Neuroepidemiology 21:58–67.
Turner, W. M., Faraone, S. V., and Tsuang, M. T. (1997). Sex Tissingh, G., Berendse, H. W., Bergmans, P., DeWaard, R.,
differences in olfactory identification and Wisconsin Card Drukarch, B., Stoof, J. C., and Wolters, E. C. (2001). Loss of
Sorting performance in schizophrenia: relationship to atten- olfaction in de novo and treated Parkinson’s disease: possible
tion and verbal ability. Biol. Psychiatry 42:104–115. implications for early diagnosis. Movement Dis. 16: 41–46.
Serby, M. (1987). Olfactory deficits in Alzheimer’s disease. Waldton S. (1974). Clinical observations of impaired cranial nerve
J. Neural Trans. 24:69–77. function in senile dementia. Acta Psychiatr. Scand. 50: 539–547.
Serby, M., Corwin, J., Conrad, P., and Rotrosen, J. (1985). Ward, C. D., Hess, W. A., and Calne, D. B. (1983). Olfactory
Olfactory dysfunction in Alzheimer’s disease and Parkinson’s impairment in Parkinson’s disease. Neurology 33:943–946.
disease. Am. J. Psychiatry 142:781–782. Waring, S. C., O’Brien, P. C., Kurland, L. T., Thibodeau, S. N.,
Serby, M., Larson, P., and Kalkstein, D. (1990). Olfactory sense Tsai, M. S., Petersen, R. C., and Esteban-Santillan, C. E. (1994).
in psychoses. Biol. Psychiatry 28:829–830. Apolipoprotein E allele in Chamorros with amyotrophic lateral
Serby, M., Larson, P., and Kalkstein, D. (1991). The nature and sclerosis/parkinsonism-dementia complex. Lancet 343:611.
course of olfactory deficits in Alzheimer’s disease. Am. Warner, M. D., Peabody, C. A., Flattery, J. J., and Tinklenberg, J.
J. Psychiatry 148:357–360. R. (1986). Olfactory deficits and Alzheimer’s disease. Biol.
Serby, M. J., Larson, P. M., and Kalkstein, D. (1992). Olfaction and Psychiatry 21:116–118.
neuropsychiatry. In Science of Olfaction, Serby, M. J., and Warner, M. D., Peabody, C. A., and Berger, P. A. (1988). Olfactory
Chobor, K. L. (Eds.). Springer-Verlag, New York, pp. 559–584. deficits in Down’s syndrome. Biol. Psychiatry 23:836–839.
Serby, M., Mohan, C., Aryan, M., Williams, L., Mohs, R. C., and Warner, M. D., Peabody, C. A., and Csernansky, J. G. (1990).
Davis, K. L. (1996). Olfactory identification deficits in rela- Olfactory functioning in schizophrenia and depression [letter;
tives of Alzheimer’s disease patients. Biol. Psychiatry comment]. Biol. Psychiatry 27:457–458.
39:375–377. Wenning, G. K., Shephard, B., Hawkes, C., Petruckevitch, A.,
Shelley, W. B., and Shelley, E. D. (2000). The smell of burnt Less, A., and Quinn, N. (1995). Olfactory function in atypical
toast: a case report. Cutis 65:225–226. parkinsonian syndromes. Acata. Neurol. Scand. 91:247–250.
Shoulson, I. (1986). Huntington’s disease. In Diseases of the West, S. E., Doty, R. L., O’Connor, M. J., and Sperling, M. A.
Nervous System, Asbury, A.K., McKhann, G.M., and (1993). Pre- and post-operative studies of olfactory function
McDonald, W.I. (Eds.). W.B. Saunders, Philadelphia, in patients with anterior temporal lobectomy. Chem. Senses
pp. 1258–1267. 18:649.
Sirota, P., Davidson, B., Mosheva, T., Benhatov, R., Zohar, J., and Wetter, S., and Murphy, C. (1999). Individuals with Down’s syn-
Gross-Isseroff, R. (1999). Increased olfactory sensitivity in drome demonstrate abnormal olfactory event-related poten-
first episode psychosis and the effect of neuroleptic treatment tials. Clin. Neurophysiol. 110:1563–1569.
on olfactory sensitivity in schizophrenia. Psychiatr. Res. Wetter, S. and Murphy, C. (2001). Apoloprotein 4 positive indi-
86:143–153. viduals demonstrate delayed olfactory event-related poten-
Sobel, N., Thomason, M. E., Stappen, I., Tanner, C. M., Tetrud, tials. Neurobiol. Aging 22:439–447.
J. W., Bower, J. M., Sullivan, E. V., and Gabrieli, J. D. E. Wolters, E. C., Francot, C., Bergmans, P., Winogrodzka, A.,
(2001). An impairment in sniffing contributes to the olfactory Booij, J., Berendse, H. W., and Stoof, J. C. (2000). Preclinical
impairment in Parkinson’s disease. Proc. Nat. Acad. Sci. (premotor) Parkinson’s disease. J. Neurol. 247:II/103–II/109.
98:4155–4159. Wu, J., Buchsbaum, M. S., Moy, K., Denlea, N., Kesslak, P.,
Solomon, G. S., Petrie, W. M., Hart, J. R., and Brackin, H. B., Jr. Tseng, H., et al. (1993). Olfactory memory in unmedicated
(1998). Olfactory dysfunction discriminates Alzheimer’s schizophrenics. Schizophr. Res. 9:41–47.
dementia from major depression. J. Neuropsychiatr. Clin. Yousem, D. M., Geckle, R. J., Bilker, W., McKeown, D. A., and
Neurosci. 10:64–67. Doty, R. L. (1996). MR evaluation of patients with congenital
Sreenivasan, K. V., Abraham, A., and Verghese, A. (1987). Right hyposmia or anosmia. Am. J. Roentgenol. 166:439–443.
temporal lobe functions in psychiatric disorders. Ind. J. Clin. Zorzon, M., Ukmar, M., Bragadin, L. M., Zanier, F., Antonello,
Psychol. 14:40–42. R. M., Cazzato, G., and Zivadinov, R. (2000). Olfactory dys-
Stedman, T. J., and Clair, A. L. (1998). Neuropsychological, neu- function and extent of white matter abnormalities in multiple
rological and symptom correlates of impaired olfactory iden- sclerosis: a clinical and MR study. Multiple Sclerosis
tification in schizophrenia. Schizophr. Res. 32:23–30. 6:386–390.
Stern, M. B., Doty, R. L., Dotti, M., Corcoran, P., Crawford, D., Zucco, G. M., and Negrin, N. S. (1994). Olfactory deficits in
McKeown, D. A., et al. (1994). Olfactory function in Down subjects: a link with Alzheimer disease. Percept. Motor
Parkinson’s disease subtypes. Neurology 44:266–268. Skills 78:627–631.
Sturniolo, G. C., D’Inca, R., Parisi, G., Giacomazzi, F., Montino, Zucco, G., Zeni, M. T., Perrone, A., and Piccolo, I. (2001).
M. C., D’Odorico, A., et al. (1992). Taste alterations in liver Olfactory sensitivity in early-stage Parkinson patients affected
cirrhosis: Are they related to zinc deficiency? J. Trace by more marked unilateral disorder. Percept. Motor Skills
Element Electrol. Health Dis. 6:15–19. 92:894–898.
24

Olfactory System Neuropathology in Alzheimer’s Disease,

Parkinson’s Disease, and Schizophrenia

Gregory S. Smutzer, Richard L. Doty, Steven E. Arnold, and John Q. Trojanowski

University of Pennsylvania, Philadelphia, Pennsylvania, U.S.A.

I. INTRODUCTION receptor cells susceptible to damage from environmental

pathogens and toxins, as well as a potential conduit for
As noted in the previous chapter, the ability to smell is viruses and other exogenous agents from the environment
decreased in many, but not all, neurodegenerative diseases. into the brain (see Chapters 26, 27, and Introduction). This
Moreover, the proportion of affected individuals typically pathway is rather direct, in that only one synapse occurs
varies from disease to disease, as does the severity of the between an olfactory receptor neuron and olfactory infor-
affliction. What are the reasons for such selective differ- mation-processing centers such as the piriform, perirhinal,
ences? While in the case of multiple sclerosis (MS), active and entorhinal cortices. Only two synapses occur between
plaques in olfaction-related central nervous system regions the receptor cell and secondary olfactory centers such as
are known to correlate highly with olfactory deficits (Doty the hippocampus. It is noteworthy that these brain regions
et al., 1997), such clear-cut causal associations have not often exhibit the most abnormalities in AD and schizo-
been made in other neurodegenerative disorders. phrenia, albeit with very different neuropathologies
Furthermore, the problem is compounded by the observa- between the diseases. Whether the olfactory pathways are,
tion that smell loss occurs early in the progression of in fact, somehow involved in the etiology of neurodegen-
several neurological disorders, long before rampant neu- erative diseases is a source of current controversy (see
ronal death or other measurable events of later-stage Chapters 26 and 27).
disease appear.
In this chapter we examine the neuropathology of the
olfactory circuitry of three neurological disorders for II. ALZHEIMER’S DISEASE
which the most histopathological data are available; name-
ly, Alzheimer’s disease (AD), Parkinson’s disease (PD), Alzheimer’s disease is the most common neurodegenera-
and schizophrenia. Our goal is to explore the possible neu- tive disorder in humans, being the major cause of demen-
robiological causes of the olfactory losses associated with tia in the elderly. As noted in the previous chapter, smell
these disorders. Environmental factors, as well as genetic loss is prevalent (~90%) and marked (mean UPSIT scores
factors, conceivably play a role in the neurogenesis of such ~20 out of 40) in this disorder and is present even in its ear-
sensory decrements. Thus, the ciliated dendritic endings of liest “questionable” diagnostic stages. AD’s classical
olfactory receptor neurons — cells whose axons pass from defining histopathology includes neurofibrillary tangles
the nasal cavity into the brain without an intervening (comprised of abnormal filaments containing tau, neurofil-
synapse — are rather directly exposed to the external ament (NF) proteins, and ubiquitin) and neuritic plaques
milieu. This exposure to the outside world makes the (comprised of degenerating neurites interlaced with ab-
503
504 Smutzer et al.

normal filaments and extracellular -amyloid fibers). To within the recesses of the superior nasal cavity. At any
what degree such pathology is causally associated with point in time, a mixture of neuronal cells at different stages
olfactory dysfunction is not currently known. It should be of development is present within this epithelium (see
stressed, as was done in the previous chapter, that the fre- Chapters 2, 5, and 6). Among such cells are neuronal
quency and severity of olfactory dysfunction in AD is very progenitor cells, immature sensory neurons, and mature
similar to that seen in Parkinson’s disease. This observa- sensory neurons. The marker proteins expressed by the
tion, and the finding of marked olfactory dysfunction early three major cell types of the sensory epithelium are illus-
in the disease process, makes it difficult to ascribe the trated in Figure 1.
olfactory dysfunction of AD solely to large numbers of In adults, the OE consists of clusters of sensory neurons
such classical neuropathological elements. interspersed among intervening expanses of metaplastic
respiratory epithelium. These patches of nonsensory respi-
A. AD-Related Pathology Within the Olfactory ratory epithelia likely result from the replacement of sen-
Epithelium sory neurons by respiratory cells during normal aging and
by atrophy of the OE through environmental insult or
As noted elsewhere in this volume, the pseudostratified inflammation (Talamo et al., 1994). In addition, the OE of
columnar olfactory epithelium (OE) of humans lies deep elderly human subjects is characterized by convolutions of

Figure 1 Schematic illustration showing the molecular phenotype for each of the three important cell types of the adult human olfac-
tory epithelium: (A) olfactory receptor neuron cell body, (B) axon of olfactory receptor neuron, (C) sustentacular cells, (D) basal cells,
(E) peripheral nervous system axons, and (F) dystrophic neurites. Abbreviations: GAP-43, growth-associated protein-43; KER-8, keratin
8; MAP 2 and 5, microtubule-associated protein 2 and 5 respectively; N-CAM, neural-cell adhesion molecule; NF-H, M, L, high, mid-
dle, and low-molecular weight (respectively) neurofilament; NGFR, p75 nerve growth factor receptor, PERIPH, peripherin; SYN, synap-
tophysin; VF, vimentin.
Olfactory Circuitry in Neurological Disorders 505

the sensory epithelium, conceivably reflecting influences (i.e., the presence of dystrophic neurites). However, neurit-
of aging and cumulative environmental insults (Getchell et ic plaques and neurofibrillary tangles were not observed in
al., 1995; Rama Krishna et al., 1995; Smutzer et al., 1998). the epithelia, a finding confirmed by others (Lee et al.,
In general, the number of sensory neurons decreases across 1993). In contrast to control tissue, the epithelia from AD
the human life span, with the most notable decrease occur- patients had increased reactivity to antibodies raised
ring after the age of 65 years (Doty, 2001). against NF proteins as well as abnormal neural structures.
The OE has been of considerable interest to students of Thus, olfactory nerve axon bundles in postmortem AD tis-
AD (Brouillard et al., 1994). The major reasons are that sue reacted with antibodies to human neuron-specific eno-
smell dysfunction is an early clinical sign of this disorder lase (NSE) and to the dephosphorylated form of the mid-
and that the OE can be readily biopsied. Thus, a relatively sized NF subunit, NFM (P-), phosphate-independent forms
large number of studies have searched for molecular mark- of NFM, and to the low molecular weight subunit of neu-
ers within the OE that may detect AD pathology (Table 1). rofilament protein (NFL).
If AD-related pathology is present in this epithelium, then The control subjects of this study, however, were not
a potential means is available for establishing a histologi- age-matched to AD patients, being significantly younger
cal diagnosis of AD during the life of the patient, perhaps (respective medians
64 and 84 years), with over half
even before the onset of clinical systems. Biopsy samples falling below the age of the youngest AD patient. Hence,
are excellent mechanisms for collecting biological tissue the reported differences could reflect age, rather than AD,
for the potential development of stable cell lines that could per se. Nevertheless, this work received support from a
be used to examine the etiology of diseases such as AD. study by Tabaton et al. (1991) that examined the
Thus, an olfactory biopsy, if reliable, could theoretically immunoreactivity of biopsied olfactory mucosa from 8 AD
replace the neuropathological examination in establishing patients and 6 age-matched controls to tau, a microtubule-
the presence or absence of the disease (Kaakkola et al., associated protein that is an integral component of paired
1994). Early detection of AD is critical for subsequent helical filaments that make up neurofibrillary tangles, and
administration of medications that may mitigate, delay, or to ubiquitin, a proteolytic stress protein that is essential for
even reverse the disease process. cell viability. Both of these proteins may contribute to
Although the accessibility of the OE for biopsy makes aggregation events associated with the pathogensis of AD.
this tissue a promising system for clinical analysis, no Tau-reactive and partially ubiquitin-reactive dystrophic
presently identified marker proteins or messenger RNAs neurites were detected in the lamina propria of all AD
expressed in the OE of AD patients are specific to AD patients, but in no control subjects. No plaques or tangles
pathology (see Table 1). In addition, studies have not cor- were observed within the olfactory mucosa of AD patients.
related the olfactory function of AD patients with the pres- Subsequent, and more extensive, studies have been
ence of these markers for AD. It is conceivable that dam- more equivocal on the uniqueness of such pathology with-
age to the OE alone, for whatever reasons, is the factor in the olfactory epithelia of AD patients. Trojanowski et al.
involved in inducing such pathology, rather than the dis- (1991) found dystrophic neurites in the olfactory mucosa
ease process under consideration. In a recent review, of all 11 AD cases examined (where they were not partic-
Arnold et al. (1998) described the numerous non–AD- ularly abundant). However, these neurites were also found
related proteins found within the OE such as keratin 8 and in 6 of 8 neurologically normal adult controls, 4 of 4
microtubule-associated proteins (MAPs). Therefore, a pri- parkinsonism cases associated with AD, 2 of 2 cases of
mary focus of this chapter is on the detection of protein progressive supranuclear palsy (PSP), a case of Shy-
markers (and the expression levels of these markers) Drager syndrome without dementia, a case of diffuse cor-
believed to be associated in some manner with neurode- tical Lewy body disease with dementia, a case of schizo-
generative diseases and schizophrenia. phrenia, and a case of idiopathic degeneration of the sub-
stantia nigra with parkinsonism. Interestingly, only 1 of 4
elderly individuals with Down syndrome (DS) and AD
1. Dystrophic Neurites and Neurofibrillary Tangles
exhibited such neurites. In contrast, dystrophic neurites
Interest in employing olfactory system biopsies in the were lacking in all 9 fetal and neonatal cases examined.
diagnosis of AD came to the forefront in 1989, when These dystrophic neurites expressed growth-associated
Talamo et al. (1989) reported that autopsied olfactory neu- protein 43 (GAP-43), neural cell adhesion molecules (N-
roepithelia from 8 patients with AD not only had a CAMs), MAP 1B, MAP 2, tau, peripherin, and synapto-
decreased number of olfactory receptor neurons relative to physin, as well as the high (NF-H), middle (NF-M), and
controls, but exhibited selective neuropathology similar to low (NF-L) molecular weight NF triplet proteins, and were
that seen in other regions of the brains of AD patients most dense at the interface of the OE and lamina propria,
Table 1 Details of Studies Examining the Olfactory Mucosa from Alzheimer’s Disease (AD) Patients, Some Other Neurodegenerative Diseases, and Controls (C)

Number of Mean or median Major conclusions

patients and age (range, relative to the olfactory
Year Authors controls SD or SE) M:F sex ratio Autopsy or biopsy Antibodies used mucosa of AD and C

1989 Talamo et al. 9 AD 83.3 (7.0) 3:6 A RM024, DP1TA51, AD mucosa shows unique
14 C 58.7 (21.0) 7:7 HO14, RM0254, NFL, pathological changes in
ALZ50, Anti-NF, OMP, morphology, distribution, and
NSE, GFAP, P57 immunoreactivity of neuronal
Vimentin, SE2 structures relative to C mucosa.
1991 Tabaton et al. 8 AD 68 (6.91) 4:4 B NSE, ubiquitin -and ubiquitin-immunopositive
6 AMC 63 (4.00) NR dystrophic neurites in lamina
propria of all AD, but no AMC,
mucosal specimens.
1991 Trojanowski 11 AD 67.0 (6.67) 5:6 A LCB, GAP 43 MAP 2, Dystrophic neurites seen in 11
et al. 8C 60.3 (23.4) 6:2 MAP 5, NF-H, NF-M, AD, 11 of 14 cases of other
4 AD/PD 67.8 (6.8) 2:2 NF-L, Peripherin, neurodegenerative diseases,
4 AD/DS 57.8 (3.3) 2:2 PKC,  ChA, SYN, and 6 of 8 neurologically normal
6 Fetuses 5.2 M (1.0) 4:2 GFAP, Desmin, Keratin controls, but zero of 9 fetal and
2 Term 5 D, 3 M 0:2 8, N-CAM, NGFR, neonatal cases.
2 PSP 78, 82 1:1 Vimentin
1 DLDB 58 1:0
1 Schiz. 75 0:1
1 ND 84 0:1
1 S-D 55 1:0
1993 Lee et al. 12 AD 73.9 (8.55) 5:7 A The following anti- Neuritic pathology seen in 10 of 12
8C 60.3 (23.4) 6:2  antibodies: AD, 4 of 4 AD/PD, 3 of
4 AD/PD 67.8 (6.8) 2:2 ALZ50, 133, 134, 135, 4 AD/DS, 6 of 6 non-AD
4 AD/DS 57.8 (3.3) 2:2 E1, Z7, 189, 304, T1, degenerative cases, and in 4 of
6 Fetuses 5.2 M (1.0) 3:3 T14, T3p, T46, PHF1 8 controls. All  epitopes
2 Term 5 D, 3 M 0:2 examined were expressed in
2 PSP 78, 82 1:1 dystrophic neurites within the
1 DLDB 58 1:0 OE of AD, but certain ones
1 Schiz. 75 0:1 seemed preferentially expressed.
1 ND 84 0:1
1 S-D 55 1:0
1994 Kaakkola 11 AD 58 (52–76) 6:5 B NF 68, NF 150, No differential expression of
et al. 46 (17-28) 2:6 NF 200, NF 150 (p), NF proteins in AD; NF neur-
NF 200 (p), MAP 2, ons coincided with distribution
of SYN, a ubiquitous neuronal
marker; no MAP 2 or  found
in AD mucosa.
1990 Kishikawa 61 “of various “Of various ages” NR A -amyloid antibody, Positive reaction to anti--antibody
et al. diseases” -antibody in 65.5% of non-demented
subjects, including young
persons; no reaction to
anti--amyloid antibody, but
pseudo-like straining present
mainly after age of 60 years.
1994a Yamagishi 6 AD including 73.5 1:5 B -antibody, A senile plaque-like extracellular
et al. 1 familial AD 66.4 6:1 amyloid- protein, mass found that reacted strongly
7C ubiquitin to an anti-Tau antibody, and
weakly to an anti-amyloid-beta
protein antibody. Ubiquitin
immunoreactivity was also
observed in dendrites of sensory
neurons.
1995 Crino et al. 12 AD NR; most same as NR A Up107, 4G8, The majority of each class of
Lee et al., 1993 LN39, TFS patients evaluated showed
10 C immunoreactivity to these
3 AD/PD antibodies. 20–30% of controls
4 AD/DS had such immunoreactivity.
6 Fetuses
2 PSP
1 PD
1 S-D
1995 Getchell et al. 3 AD 72, 74, 85 2:1 A HSP-70 HSP-70 expressed in olfactory,
3 AMC 76, 78, 82 1:2 supporting, and acinar cells of
2 Fetuses 16–26 wks 1:1 Bowman’s glands at all ages,
1 Infant 10 wks 1:0 being greatest in prenatal
13 Others 55.61 (17.8) 7:6 specimens; expression decreased
with age, and was ~4 less in
AD than in AMC ORCs.
1996 Kulkarni-Narla 5 AD 76.8 (15.9) 4:1 A Mn-SOD, Pronounced increase in Mn- and
et al. 4 NDC 83.5 (10.0) 2:2 Cu, Zn-SOD Cu, Zn-SOD immunoreactivity in
4 C† 72.0 (7.20) 2:2 AD relative to C.
6C 37.8 (14.5) 3:3
(continued)
Table 1 (continued)

Number of Mean or median Major conclusions

patients and age (range, relative to the olfactory
Year Authors controls SD or SE) M:F sex ratio Autopsy or biopsy Antibodies used mucosa of AD and C
1996 Yamagishi 9 AD 75.0 (3.9) 7:2 A Calbindin-D28 ON from all patients contained
et al. 6 C* 80.4 (4.0) (Spot 35), NSE, OMP OMP, Spot 35, and NSE. Number
8C 34.5 (16 wks–90 y) of Spot 35 immunoreactive neurons
decreased in AD relative to C;
Mean number of OMP- and
NSE-reactive neurons were
no different.
1998 Hock et al. For biopsy, AD biopsy ages ` 5:0 B  antibody AT8, Cytoskeletal changes and 
2 mild and 3 46–82: The 7 A cases -amyloid -A4, NSE pathology within the OE
moderate AD served as positive are not primary (or specific)
For autopsy, AD autopsy 0:2 and negative features of AD, and may
2 severe AD ages: 86, 88 controls occur predominantly in late
and 5 C stages of the disease.
1998 Yamagishi 9 AD 76.2 (11.4) NR A ApoE, PGP 9.5 No AD: C differences in olfactory
et al. 10 C receptor neuron density in
contiguous OE strips; density of
apoE staining in PGP-staining
ONs 3.5  greater in AD than C.
1999 Duda et al. 7 AD 80.9 (8.2) 3:4 A -, -, -synucleins Although antibodies to - and
5 PD 65.4 (24.6) 5:0 -synucleins detected abnormal
2 MSA 55, 60 0:2 dystrophic neurites in OE of
1 AD/PD 71 1:0 patients with neurodegenerative
1 PD/DLB 63 0:1 disorders, similar pathology also
1 DLB 58 1:0 was observed in the OE of con
1 ND 86 0:1 trols.
11 C 73.6 (15.0) 5:6

A
autopsy; AD/PD: AD with parkinsonism; AD/DS
AD with Down syndrome; AMC
age-matched controls; apoE
apolipoprotein E; B
biopsy; D
days; DLBD
diffuse Lewy body dis-
ease; M
months; MSA
multiple system atrophy; ND
idiopathic nigral degeneration; NDC
nondemented controls; NR
not reported; OE
olfactory epithelium; ON
olfactory receptor neu-
rons; ORC
nonneuronal olfactory cell; PD
Parkinson’s disease; PSP
progressive supranuclear palsy; S-D
Shy-Drager syndrome; and Schiz.
schizophrenia.
Antibodies designated as follows: HSP-70 = heat shock protein 70; MAP
microtubule-associated protein; LN39
an antibody to -amyloid precursor protein; NF
neurofilament; NFM (P-)
anti-
body to dephosphorylated from of mid-sized neurofilament subunit; NSE
neuron-specific enolase; OMP
olfactory marker protein; p
phosphorylated epitope; PGP
protein gene product 9.5; SYN

synaptophysin; TFS
thioflavin-S; Up107
an antibody to β-amyloid peptide.
*Six of these controls were age-matched to the AD patients; the remaining 8 were not and included specimens ranging from a 16-week-old fetus to a 90 year-old man with hypertension. †These controls
were not age-matched, but were confirmed to have, as were the other controls, brains with no significant AD pathology.
Olfactory Circuitry in Neurological Disorders 509

as well as the lower portion of the sensory epithelium. of amyloid precursor protein (APP) by an aspartyl pro-
These data suggested that (1) the mere presence of dys- tease. Recent evidence suggests that presenilin 1 may be
trophic neurites within the olfactory mucosa is not specif- the aspartyl protease that cleaves APP to -amyloid pro-
ic to AD, (2) dystrophic neurites are a common feature of tein A 1–40 and -amyloid protein A 1–42 (Li et al.,
the olfactory neuroepithelium of neurologically normal 2000).
adults, and (3) the relevance, if any, of such neurites to Nasal mucosal biopsy tissue from AD patients has also
aging or specific disease processes was not clear at that been examined by Yamagishi et al. (1994a; 1994b). These
time. researchers detected tangle-like abnormal tau protein
Later studies by this same group further evaluated these immunoreactivity in dendrites and perikarya of olfactory
specimens or large subsets thereof. Lee et al. (1993) receptor neurons and in nerve bundles. They also detected
employed a panel of antibodies to epitopes present in sub- ubiquitin immunoreactivity in the OE. These results sug-
domains, which span much of the lengths of all six known gested that ubiquitin conjugated with other polypeptides
isoforms of human central nervous system (CNS) tau. All within dendrites, which marked these proteins for selective
six epitopes were expressed in dystrophic OE neurites of degradation. These authors concluded that in AD patients,
AD patients. While the general findings noted above were similar pathologic changes occur in both the CNS and the
observed, some antibodies (e.g., antibody 304) did not olfactory mucosa.
label dystrophic neurites in non-AD normal cases. These In a postmortem study, Crino et al. (1995) examined
results suggest that within the OE, alterations occur in tau APP and -amyloid protein within the olfactory mucosa of
protein that are specific to AD. Another important obser- patients diagnosed with AD. They utilized antibodies
vation of this work was that dystrophic neurites did not directed at -amyloid and against flanking sequences of
contain paired helical filaments characteristic of AD APP. -Amyloid immunolabeling was observed in 10 of
pathology in the CNS (Lee et al., 1991), but instead were 12 AD cases, and 2 of 3 cases with combined AD and PD.
composed of numerous bundles of 10–15 nm long fila- -Amyloid staining was detected primarily in the basal
ments. third of the OE, in axons projecting through the lamina
These anatomical studies are significant because human propria, and in metaplastic respiratory epithelium dis-
olfactory receptor neurons from normal individuals do not persed throughout the sensory OE. These authors conclud-
typically express NF proteins, the neuron-specific interme- ed that their data “suggest that deposition of A occurs in
diate filament protein peripherin, or the synaptic protein a variety of circumstances and is not restricted to patients
synaptophysin except within dystrophic neurites with AD, PD, or DS.”
(Trojanowski et al., 1991). The cause of dystrophic neurite Recently, Hock et al. (1998) used a panel of specific
formation in the OE is unknown, but could be induced by antibodies to examine whether staining for tau or -amy-
respiratory disease or response to injuries. Dystrophic neu- loid differed between (1) OE tissue obtained by biopsy
rites could also result from neuronal populations that sus- from five clinically mild to moderate AD patients and (2)
tain damage from external stimuli such as pathogens, toxic OE tissue obtained at autopsy from advanced AD cases.
chemicals, or their metabolites (Trojanowski et al., 1991). No positive staining was found in any of the mild to mod-
Additionally, some dystrophic neurite populations could erate AD cases, although tau immunoreactivity was noted
arise from aberrant basal cell development in patches of in the fine nerve fibers of the lamina propria and in a few
epithelium that have undergone replacement of damaged OE cells of advanced AD cases. The authors concluded,
sensory neurons. Since a small number of NF-positive “these results are consistent with other reports showing
olfactory neurons give rise to NF-positive dystrophic neu- that cytoskeletal changes and tau pathology in the OE are
rites, these dystrophic neurites could reflect abnormally not primary (or specific) features of AD and may occur
induced proliferation of specific neuron populations predominantly in late stages of the disease.”
(Carrell, 1988, Talamo et al., 1989). Alternatively, these
OE neurites could represent sensory neurons that fail to 3. Synuclein
form specific axonal connections with their glomerular tar-
The synucleins are a family of neuronal proteins that have
gets in the olfactory bulb (Smutzer et al., 1998).
been implicated in the pathogenesis of both AD and PD
(Kruger et al., 1998; Lippa et al., 1998). Of the synucleins,
2. Amyloid Plaques
-synuclein is a major component of cytoplasmic inclu-
A second histological marker of AD pathology is the for- sions, and this protein was originally isolated from plaques
mation of extracellular senile plaques composed of -amy- obtained from the CNS tissue of AD patients (Ueda et al.,
loid peptide. Amyloid plaques contain two major amy- 1993). -Synuclein is a soluble protein of 140 amino acids,
loidogenic peptides that are formed by enzymatic cleavage and localized primarily to presynaptic axon terminals in
510 Smutzer et al.

the nervous system (Chase, 1997). When insoluble, this the olfactory neuroepithelium to damage from a variety of
protein forms aggregates within plaques of AD tissue. - agents. An increase in hsp 70 expression has been reported
Synuclein is also the major fibrillary component of Lewy in frontal cortex white matter from postmortem AD tissue,
bodies and Lewy neurites (Spillantini et al., 1998). Lewy implying a stress response in other AD tissue (Harrison et
bodies are the primary histopathological marker for PD. al., 1993).
The -, -, and -synucleins are differentially
expressed in olfactory and respiratory epithelia, with - 6. Apolipoprotein E
synuclein being the most abundant synuclein in the OE. In
In addition to inclusions within OE cells, several genes and
addition to its expression within dystrophic neurites, -
their gene products show unusual expression patterns with-
synuclein is expressed within the perikarya and cilia of
in the human olfactory mucosa of AD patients. In a post-
olfactory receptor neurons. -, -, and -Synuclein pro-
mortem study, Yamagishi et al. (1998) examined
teins are also expressed in Bowman’s glands and in basal
apolipoprotein E (apoE) protein levels in human olfactory
cells of the OE (Duda et al., 1999). These observations are
mucosa. ApoE is a 34 kDa glycoprotein component of
consistent with a role for synucleins in the development,
most serum lipoproteins, whose function is to mobilize
regeneration, and plasticity of olfactory receptor neurons
cholesterol for membrane formation primarily during cel-
in the normal adult human OE (Duda et al., 1999). These
lular differentiation and cell repair. ApoE is encoded by a
synucleins have also been identified in the olfactory
polymorphic gene with three common alleles. A genotype
mucosa of AD, PD, dementia with Lewy body disease,
containing a fourth allele has been associated with an
multiple system atrophy, and control patients.
increased risk for familial AD. The 4 isoform of apoE is
thought to associate with tau or MAP 2 proteins in neu-
4. Superoxide Dismutaste
rofibrillary tangles and with -amyloid protein within
Another protein, superoxide dismutase (SOD), is reported- senile plaques (Rebeck et al., 1993). This association may
ly increased in the OE of AD patients relative to controls. partly explain why the 4 allele is a risk factor for familial
Manganese-SOD and Cu-Zn SOD are two ubiquitous AD.
enzymes that protect cells against destructive superoxide In the OE of elderly individuals, apoE has been detect-
anion free radicals and their reaction products during cel- ed throughout the cytoplasm of sensory neurons and
lular processes that include mitochondrial oxidative phos- Schwann cells associated with olfactory nerve bundles, in
phorylation. SODs scavenge free radicals and protect cells macrophages, and in blood vessels along the olfactory
from oxidation. Immunocytochemical analysis of human basement membrane. Its origin is presumably glial cells,
OE for Mn- and Cu-Zn SOD protein expression in post- including Schwann cells, and possibly intraepithelial
mortem tissue identified both enzymes in olfactory recep- macrophages (more prevalent in the OE of older than of
tor cells (including axons), sustentacular cells, basal cells, younger persons), since no evidence of its expression with-
Bowman’s glands, and vascular endothelium (Kulkarni- in olfactory neurons has surfaced. Indirect support for the
Narla et al., 1996). Significantly, immunoreactivity for hypothesis that olfactory receptor neurons may endocytose
both SOD enzymes heavily labeled cells near the surface apoE from surrounding cells comes from the demonstra-
of the OE and in basal region of the OE from AD patients. tion of receptors for its uptake on CNS neurons (Rebeck et
This pronounced increase in SOD enzyme immunoreactiv- al., 1993). While it does not appear to be a specific mark-
ity in the OE supports the hypothesis that oxidative stress er for AD, it is noteworthy that apoE levels are more than
in the sensory epithelium could be partially responsible for threefold greater in AD than in control olfactory receptor
olfactory deficits observed in these patients. cells (Yamagishi et al., 1998).

5. Heat Shock Protein 7. Calbindin D28k (Spot 35 protein)

Heat shock proteins are highly conserved proteins that are Intracellular levels of Ca2 are maintained at submicromo-
upregulated in cells after exposure to various stressors that lar concentrations to prevent Ca2 toxicity and to inhibit
include temperature shock, ischemia, injury, and toxic sub- the formation of insoluble phosphorylated and carboxylat-
stances (Welch, 1992). Such upregulation presumably aids ed Ca2 salts (Hubbard, 2000). Calbindin D28k is a solu-
in the protection of cells from injury. It has been demon- ble protein that binds cellular Ca2 and buffers cytosolic
strated that heat shock protein (hsp) 70 is markedly lower Ca2 concentration within cells. Calbindin D28k localizes
(~fourfold decrease in immunoreactivity) in the OE of AD to cells of the OE, olfactory bulb, and AON of the human
patients relative to controls (Getchell et al., 1995). This adult (Ohm et al., 1991; Yagamishi et al., 1996). In the OE,
downregulation may result in increased susceptibility of all known cell types express this Ca2-binding protein.
Olfactory Circuitry in Neurological Disorders 511

Intracellular Ca2 flux within olfactory tissue is critical epithelial thickness was observed. Importantly, a signifi-
for both odorant signal transduction (Kleene 1993; Zufall cantly lower number of calbindin D28k reactive receptor
et al., 2000) and perireceptor events that include secretion cells was noted in AD epithelia than in age-matched con-
(Smutzer et al., 1997). For odorant signal transduction, the trol epithelia.
cyclic nucleotide-gated channel (CNG) is a nonspecific
cation channel that regulates cation influx after odorant
8. Metallothionein
stimulation and activation with cyclic nucleotides. An
increase in divalent cations reduces the sensitivity of the Metallothioneins (MTs) are a family of proteins, present in
CNG channel to ligand (Kleene, 1999). Also, intracellular a number of mammalian tissues, that are characterized by
Ca2 activates a Ca2-dependent chloride channel during atypical cysteine abundance and high heavy metal [Cu(I),
odorant signal transduction (Kleene, 1993). Calcium is Zn(II)] content. All MT isoforms have numerous physio-
also required for type III adenylate cyclase activity, and logical functions, such as aiding in zinc and copper metab-
this cation may amplify the response of this enzyme. In olism and in protecting against stress and oxygen free rad-
addition, Ca2 may be involved in the migration of neural ical species (e.g., superoxide radicals). In mammals, four
progenitor cells from the mitotically active subventricular major isoforms have been described in numerous tissues,
zone of the forebrain to the bulb and cerebral cortex (Baker including the brain. One of these isoforms, MT-3 (also
et al., 2001; Haughey et al., 2002). termed growth inhibitory factor), may play a role in neu-
In cells, Ca2 regulates its own release from endoplas- romodulatory events and in the pathogenesis of AD
mic reticulum (ER) stores through activated IP3 receptors (Richarz and Bratter, 2002). MT expression is increased in
(Bezprozvanny et al., 1991) and/or activated ryanodine cortical glial cells of confirmed and possible AD cases
receptors (Li and Chen, 2001). The emptying of ER Ca2 (Adlard et al., 1998).
stores by either or both of these Ca2 channels could trig- Chuah and Getchell (1999) reported that immuno-
ger capacitative-like Ca2 entry into the dendrites and staining for MT in OE obtained from the septum of AD
soma of olfactory receptor neurons via Ca2- patients was increased relative to that of age-matched
release–activated currents similar to those identified in controls; indeed, four of six control specimens showed no
nonexcitable cells (Zufall et al., 2000). This intracellular MT staining at all. More intense and more frequent
Ca2 pool could modulate secretory activity, or possibly immunostaining was observed in the epithelium,
amplify the magnitude and duration of Ca2 transients Bowman’s glands, and underlying lamina propria. In a
that spread the odor-induced Ca2 wave in receptor neu- mouse study described in the same publication, such
rons (Zufall et al., 2000). staining did not differ in the OE or olfactory bulbs
It is possible that a decrease in calbindin D28k expres- between apo-E–deficient and wild-type mice. The
sion within neurons could affect the ability of these cells to authors concluded that “the induction of MT in a periph-
buffer Ca2, which in turn could adversely affect intracel- eral organ such as the olfactory mucosa is possibly
lular signaling, neuroblast migration and differentiation, or indicative of the widespread increased level of ROS reac-
perireceptor processes. Furthermore, an increase in cytoso- tive oxygen species associated with AD.”
lic Ca2 could cause these cells to be more susceptible to
damage or necrosis. For example, the expression of cal- In summary, the data from these and other studies are
bindin D28k in a pancreatic islet beta-cell line protects mixed as to whether specific neuropathology discernible
these cells against cytokine-induced apoptosis and necro- by various antibodies may be present in the OE of AD
sis (Rabinovitch, 2001). patients that would meaningfully distinguish them from
An age-related decrease in calbindin D28k protein normal elderly persons. Clearly, more research is needed
expression in olfactory receptor cells has been reported. in this area, particularly longitudinal biopsy studies to
The number of immunoreactive olfactory receptor cells examine if quantifiable changes occur over time in per-
were greatest in tissues from prenatal fetuses and notice- sons who develop AD. Such studies could determine
ably less in tissue from a 10-week-old postnatal infant, whether such changes are distinct enough to provide a
three “young” (24-, 45-, and 47-year-old) and two “old” sensitive and specific diagnosis of AD. As noted earlier,
(65- and 90-year-old) specimens (Yamagishi et al., 1996). no study has correlated the occurrence of such markers
Although the number of calbindin D28k immunoreactive with actual psychophysical measurements of olfactory
olfactory receptor neurons was similar among the postna- function, which could readily be performed in conjunc-
tal groups examined in this study, statistically fewer tion with biopsy studies. This approach would seem to
immunoreactive neurons were found in the elderly group be important if one is to identify complex AD-related
relative to the fetal group, and an age-related decrease in factors in the OE.
512 Smutzer et al.

B. Olfactory Bulb Pathology in Alzheimer’s Disease among the first olfactory structures to exhibit neurofibril-
lary tangle formation and neuronal damage in AD (Kovacs
As described in other chapters of this volume (e.g., et al., 2001). This neurodegeneration in the bulb would
Chapter 7), the olfactory bulb is the first olfactory system likely affect olfactory information processing in other
relay station in the brain. After forming synapses with regions of the CNS.
incoming olfactory receptor cell axons within the
glomeruli, the major bulbar projection neurons — the C. Olfactory Cortex Pathology of Alzheimer’s
mitral and tufted cells — send their axons via the olfacto- Patients
ry tract to cortical regions collectively termed the olfacto-
ry cortex (see next section). Included among such regions As described in detail in Chapters 8 and 9, the olfactory
is the anterior olfactory nucleus (AON), which is a rela- tract projects to central structures collectively termed the
tively well-defined structure largely comprised of large primary olfactory cortex. These structures are found large-
multipolar neurons located deep within the bulb’s granular ly within the medial temporal and basal frontal areas of the
cell layer (Esiri and Wilcock, 1984; ter Laak et al., 1994). brain and include elements of the AON, the piriform and
The AON, in turn, sends projections to a number of con- prepiriform cortices, the corticomedial amygdala, the
tralateral structures via the anterior commissure, including olfactory tubercle, and the entorhinal cortex. Most mitral
the contralateral AON and olfactory bulb. cell axons terminate within the prepiriform or piriform
Marked age-related changes occur in the olfactory bulb, cortices, which have extensive connections with CNS
and even nondemented elderly persons exhibit neurofibril- structures involved in both cognition and behavior (Reyes
lary tangles within the bulb and AON (Kishikawa et al., et al., 1987). The latter structures include insular,
1990; Meisami et al., 1998; Okamoto et al., 1990; Smith, orbitofrontal, and dorsolateral cortices, which collectively
1942; Yousem et al., 1998). The massive loss of olfactory are considered the secondary olfactory cortex. Of these
bulb glomeruli in older persons has been known for over structures, the entorhinal cortex serves as a link between
60 years (Smith, 1942). Kishikawa et al. (1990) reported the neocortex and the limbic system and contains a large
the following frequencies of intrabulbar tangles as a func- number of connections with association cortices in all four
tion of decade of life in nondemented individuals: 5th lobes of the brain. Perhaps the most important projection
decade, 17.9% (n
28); 6th decade, 25.6% (n
43); 7th of the entorhinal cortex is to the hippocampus by way of
decade, 57.1% (n
28); 8th decade, 86.7% (n
15); and the perforant path, as this is the main conduit by which
9th decade, 100% (n
2). Similar findings were reported information regarding sensory experience reaches the hip-
by Okamoto et al. (1990). Such age-related changes are pocampus (van Hoesen et al., 2000).
accelerated and exacerbated in AD, where decrements in It is well established that in AD, central nervous system
some neurotransmitter receptors (e.g., dopamine) have regions closely associated with the olfactory system exhib-
also been reported (Loopuijt and Sebens, 1990). it major pathological changes (Arnold et al., 1998).
The formation of neurofibrillary tangles and cell loss Neurofilament proteins and -amyloid plaques occur most
occurs in all layers of the bulb and within the AON early in prominently in the association areas of the cerebral cortex,
the disease process of AD (Averback, 1983; Kovacs et al., which are interconnected with elements of the olfactory
2001; Ohm and Braak, 1987; Struble and Clark, 1992; ter cortex. This finding is in contrast to the primary motor,
Laak et al., 1994). For example, Esiri and Wilcock (1984) somatosensory, auditory, and visual areas, which exhibit
reported finding 62% fewer cells in the AON of AD relatively little AD pathology (Pearson, 1996). Thus, pro-
patients relative to age-matched controls. Interestingly, nounced loss of cholinergic neurons in the basal forebrain
while neurofibrillary tangles are frequently present within nuclei, entorhinal area, and hippocampus have all been
the neurons of the AON, they are relatively rare within the reported in AD (Hyman et al., 1984; Kasa et al., 1997).
mitral, tufted, or granule cells of the bulb. However, severe Histopathological examination of the prepiriform cortex
and specific losses of olfactory bulb cells occur even in the has shown increased numbers of neurofibrillary tangles
earliest stages of AD (Davies et al., 1993; Struble and and neuritic plaques in this region. Arnold et al. (1991a)
Clark, 1992). Such losses do not require the prior forma- mapped the densities of neurofibrillary tangles and senile
tion of tangles or plaques and appear to be among the plaques in 39 cortical areas of brains of AD patients, and
major pathological changes of AD (Struble and Clark, observed a distinct distribution of NF tangles. The brain
1992). A loss of ~50% of myelinated axons, associated regions with the highest densities of neurofibrillary tangles
with a reduction in cross-sectional area of the olfactory were the entorhinal cortex, subiculum, temporal pole,
tract, has been reported in AD patients (Davies et al., perirhinal cortex, amygdala, and prepiriform cortex. Other
1993). Recent data suggest that the olfactory bulb is areas that were less affected included limbic proisocortical
Olfactory Circuitry in Neurological Disorders 513

regions and the neocortical association areas. Senile Larger decreases were observed in the severe AD group,
plaques were more evenly distributed than were neurofib- suggesting that clinical deficits occur only when signifi-
rillary tangles in Alzheimer’s brains. Recent studies further cant neural loss is present (Price et al., 2001).
showed that antibodies to presenilin proteins stained neu- Overall, the olfactory losses observed in AD patients
rofibrillary tangles in AD tissue (Murphy et al., 1996). may reflect, in whole or in part, the neuropathology occur-
These antibodies reacted with a subset of neurofibrillary ring in the olfactory bulb and primary and secondary olfac-
tangles, but not with dystrophic neurites in the entorhinal tory cortices. Clearly, increased numbers of neurofibrillary
cortex and hippocampus. tangles and senile plaques are present in regions of the
In a large series of cases from across the adult life span, olfactory system believed important for odor discrimina-
Braak and Braak (1991) staged the presence of neurofib- tion, identification, and even odor detection. While such
rillary pathology throughout the brain. They described six inclusions may correlate with the well-characterized loss-
stages of neuropathological severity, with dementia pre- es of olfactory function in AD, more studies are required to
senting in the third or fourth stage. Neurofibrillary tangles link these inclusions with quantifiable olfactory dysfunc-
were first noted in the “transentorhinal” region (compris- tion in humans. Moreover, the classical pathology of AD
ing the perirhinal and lateral portion of entorhinal cortex) may not entirely explain the olfactory loss, since patients
before being found more extensively in the entorhinal cor- with vascular dementia exhibit a similar degree of olfacto-
tex, hippocampus and other limbic regions, and ultimately ry dysfunction (Gray et al., 2001). While vascular demen-
the neocortex. Other studies have examined olfaction- tia is frequently accompanied with an admixture of AD-
related cortices from AD brains in considerable detail, and like pathology, such pathology is generally believed to be
these studies have demonstrated particularly severe pathol- more variable than in AD (Kalaria and Ballard, 1999).
ogy in the entorhinal cortex. Severe pathology has also
been reported in the corticomedial nuclei of the amygdala
(Kromer Vogt et al., 1990), prepiriform cortex (Reyes et III. PARKINSON’S DISEASE
al., 1987), temporal pole (Arnold et al., 1994), insular and
orbitofrontal cortices (Chu et al., 1997), and basal fore- Parkinson’s disease is the second most common neurode-
brain (Whitehouse et al., 1982). generative disorder in humans. Several forms of parkin-
Recently, McShane et al. (2001) reported that anosmia sonism exist, and these forms may be idiopathic, familial,
is more closely associated in demented patients with Lewy or induced by respiratory chain inhibitors such as MPP.
bodies than with typical measures of AD-related pathology. Parkinsonism also occurs in many patients diagnosed with
However, olfactory testing in this study was rudimentary AD. PD is a movement disorder that is characterized by
since patients were simply asked whether or not they per- tremor, rigidity, and bradykinesia. PD can be identified in
ceived an odor. A large number of studies have established postmortem tissue by a loss of pigmented dopaminergic
that relatively few patients with either AD or PD have anos- neurons of the substantia nigra pars compacta of the brain-
mia, making this all-or-none single-stimulus classification stem and gliosis. PD is further characterized by a distinct
questionable. Moreover, the brain regions sampled for pathology that includes the formation of Lewy bodies and
Lewy bodies were apparently not the same as those sam- Lewy neurites (Forno, 1996; Lewy, 1912; Trojanowski and
pled for AD-related pathology, making the comparisons Lee, 1998). Lewy bodies are composed of a dense core of
somewhat enigmatic. Nonetheless, this is perhaps the first filamentous and granular material surrounded by radially
study to attempt a direct correlation between pathology and oriented filaments (Roy and Wolman, 1969). A number of
function. Clearly, more refined studies are needed to deter- proteins have been identified in Lewy bodies and Lewy
mine whether this relatively startling conclusion is valid. neurites of PD patients. These proteins include NF proteins
Price et al. (2001) determined, in a postmortem study, (Schmidt et al., 1991), -synuclein (Spillantini et al.,
the volume and number of neurons in the entorhinal cortex 1998), and ubiquitin (Forno, 1996). The transformation of
and hippocampus of preclinical AD cases using a stereo- soluble synuclein proteins into insoluble aggregates (or
logical procedure (Price et al., 2001). These subjects had interactions with other neural proteins or molecules to pro-
no measurable cognitive decline, but postmortem analysis duce aggregates) may lead to Lewy body formation. After
demonstrated neuropathological evidence of AD (Price et their formation, these Lewy bodies could compromise the
al., 2001). No significant decrease in neuron number or survival of affected neurons (Trojanowski and Lee, 1998),
cell volume was found in the control group, and little or no including those of the olfactory pathway.
decrease was detected between the control and preclinical Olfactory dysfunction is one measurable feature of indi-
AD groups. In contrast, substantial decreases in neuron viduals who suffer from PD (Doty et al., 1988). Clinical
number and volume were observed in very mild AD cases. evidence indicates that olfactory deficits may occur even
514 Smutzer et al.

before the onset of movement disorders that are character- in PD (Hurtig et al., 2000). Thus, the pathology of the OE
istic of PD (Berendse, 2001). Thus, olfactory testing is one could have potential use in antemortem analysis of PD.
promising tool to identify olfactory deficits that may occur
prior to clinical symptoms (Berendse, 2001; Tissingh B. Olfactory Bulb Pathology in PD
et al., 2001; Wolters et al., 2000).
Recent evidence also suggests that disease-specific pathol-
A. Olfactory Epithelium Pathology in PD ogy may localize to the olfactory bulb and olfactory tract
of PD patients. A number of investigators (Daniel and
Relatively few studies have characterized the olfactory Hawkes, 1992; Hawkes et al., 1997; Pearce et al., 1995)
system of patients with PD at the cellular level. The few have detected Lewy body pathology by classic ubiquitin
published studies have focused on the OE and olfactory immunostaining in the AON of PD cases. Lewy bodies in
bulb. Crino et al. (1995) were the first to identify -amy- the olfactory bulb and tract were similar in morphology to
loid protein in the OE of postmortem tissue from PD those found in the cerebral cortex. Surprisingly, Lewy bod-
patients. As in AD, immunostaining was observed in the ies were not detected in control olfactory bulb tissue. In
basal third of the OE, in axons that projected through the addition, the hippocampus (which is a secondary olfactory
lamina propria, and in metaplastic respiratory epithelium center) also contained high densities of synuclein-positive
within the OE. This group also examined postmortem tis- lesions in presynaptic axon terminals. These results are
sue from PD patients with dementia due to AD. As in AD, consistent with synaptic dysfunction in the hippocampus
dystropic neurites were detected in all four cases exam- (Galvin et al., 1999).
ined. The dystropic neurites stained with antibodies Pearce et al. (1995) detected a dramatic cell loss in the
against GAP-43, and GAP-43 protein is a marker for post- AON of patients with PD, a phenomenon also observed in
mitotic immature neurons. Dystrophic neurites also the AON of patients with PD-dementia complex of Guam
expressed proteins that were not detected in normal olfac- (Doty, 1997). The cell loss correlated well with the num-
tory receptor neurons. These proteins included synapto- ber of Lewy bodies, a noteworthy observation because
physin, tau, and NF proteins. Lewy body formation in PD is almost exclusively confined
More recent studies have utilized synuclein immunore- to the brainstem, cerebral cortex, and hippocampus. As in
activity for positive identification of dystrophic neurites in AD, these results suggest a susceptibility of certain neu-
the OE of PD patients. -Synuclein has received intense ronal populations to PD pathology. If Lewy bodies localize
scrutiny because two missense mutations in the -synucle- to the olfactory bulb and tract tissue in PD but not in con-
in gene (an A53T and an A30P nucleotide substitution) trol tissue, then it is possible that PD could be diagnosed in
have been associated with a rare autosomal dominant form postmortem tissue by immunostaining this tissue for Lewy
of familial PD (Kruger et al., 1998; Papadimitriou et al., body proteins.
1999; Polymeropoulos et al., 1997). Also, -synuclein Whether dopaminergic pathology is associated with the
aggregates of wild-type protein are a major protein com- olfactory dysfunction of patients with PD is not known.
ponent of filaments of Lewy bodies and Lewy neurites in The adult human olfactory bulb is thought to contain tyro-
idiopathic PD (Spillantini et al., 1998; Trojanowski and sine hydroxylase activity in superficial tufted cells and in
Lee, 1998). As in AD, -synuclein was the most abundant small periglomerular neurons (Halasz et al., 1981;
synuclein species found in dystrophic neurites of the OE of Hoogland and Huisman, 1999; Liberini et al., 2000; Smith
PD patients. Again, no synuclein pathology was specific to et al., 1991), although the ability of these neurons to syn-
the OE of PD patients (Duda et al., 1999). thesize dopamine and the role of this neurotransmitter in
-Synuclein can associate with the cytoplasmic protein olfactory function is unclear. The olfactory dysfunction of
synphilin-1. Synphilin-1 forms intracellular inclusions that PD patients is not restored with dopamine repletion, sug-
resemble Lewy bodies in cells that are co-transfected with gesting that the deficit, if dopamine dependent, reflects a
the central domain of synuclein (Engelender et al., 1999). lack of (or dysfunctional) receptors rather than the avail-
Furthermore, synphilin-1 has been identified as a protein ability of dopamine (Doty et al., 1992). Nevertheless, ani-
component of Lewy bodies in PD (Wakabayashi et al., mal studies suggest that dopaminergic neurons may mod-
2000). This 90 kDa protein is thought to allow -synucle- ulate olfactory function to some degree. For example,
in to interact with intracellular proteins involved in vesicle dopamine receptor activation in rat olfactory bulbs causes
transport or in cytoskeletal functions (Engelender et al., a significant depression of synaptic transmission at the first
1999). These immunocytochemical studies within the OE synapse between olfactory receptor neurons and mitral
are significant because the expression of -synuclein pro- cells (Hsia et al., 1999). Systemic injection of the
tein in cortical Lewy bodies correlates well with dementia dopamine D2 receptor agonist quinpirole impairs operantly
Olfactory Circuitry in Neurological Disorders 515

determined odor detection performance of rats. In contrast, C. Olfactory Cortex Pathology of Parkinson’s
injection of the dopamine D1 receptor agonist SKF 38393 Patients
enhances such performance (Doty and Risser 1989; Doty
et al., 1998), effects that could be modulated at the level of PD is characterized by a decrease in CNS dopamine pro-
the bulb and/or in more central regions (see next section). duction. If dopamine affects CNS olfactory function in
Ligands that directly or indirectly enhance dopaminergic humans, then a loss of dopaminergic neurons could nega-
neurotransmission stimulate sniffing behavior, and this tively modulate olfactory processing in the brain. In
effect is blocked by selective dopamine D2 antagonists, rodents, recent studies have identified dopaminergic fibers
including 1-sulperide (Le Moal, 1995). in the piriform cortex (Datiche and Cattarelli, 1996). In
Dopaminergic neurons may also play a role in olfacto- addition, biochemical studies and binding studies have
ry bulb cell differentiation. In rodents and nonhuman pri- shown that the rodent piriform cortex contains high con-
mates, the constant differentiation of dopaminergic neu- centrations of dopamine and dopamine receptors
rons from precursor cells takes place in the olfactory bulb (Descarries et al., 1987). For example, the olfactory tuber-
(Kornack and Rakic, 2001), where these cells differentiate cle is the main source of innervation to the piriform cortex,
into periglomerular cells. The latter cells originate as neu- and this region has markedly decreased dopamine levels in
roblasts in the subventricular zone of the forebrain PD (Bogerts et al., 1983).
(Haughey et al., 2002). The mitotically active neuronal Histopathological studies of postmortem PD tissue have
progenitor cell population undergoes restricted chain shown that neurofibrillary tangles and Lewy bodies were
migration to the bulb and cerebral cortex via the rostral found within lamina II of the entorhinal cortex (Braak and
migratory stream. A subset of these neuroblasts then Braak, 1990). In one study, Lewy bodies were detected in
migrates to the bulb and differentiates into dopaminergic this region in 48% of the cases examined (20/42), whereas
neurons (Baker et al., 2001; Pencea et al., 2001). In the neurofibrillary tangles were positively identified in 98% of
absence of disease, these neuroblasts may provide a reser- the cases (41/42) (Mattila et al., 1999). In a related
voir for replacement of neurons lost during cell turnover in immunohistochemical study, -synuclein-immunoreactive
the olfactory bulb or after brain injury (Haughey et al., cortical Lewy bodies were identified in the amygdala (pri-
2002). In the presence of disease, however, neurogenesis mary olfactory cortex), hippocampus (secondary olfactory
may be deficient.* processing center), or cortical gyrus of 43 of 45 PD
In addition to a possible role in olfaction, bulbar patients (Mattila et al., 2000). This study further demon-
dopaminergic neurons may be important for the future strated that -synuclein–positive cortical Lewy bodies
treatment of PD. When human neural progenitor (embry- were associated with cognitive impairment in PD that was
onic) cells were transplanted into the subventricular zone independent of any AD-type pathology.
of neonatal rat brains, these embryonic cells underwent
migration to, and differentiation within, the olfactory bulb
(Englund et al., 2002). However, it is not known if the con- IV. SCHIZOPHRENIA
tinuous differentiation of dopaminergic neurons occurs in
the human olfactory bulb. If this does occur, one promising Schizophrenia, a severe psychiatric illness that affects
source of transplantable neurons in PD patients could be approximately 1% of the world’s population, is character-
adult olfactory bulb dopamine progenitor cells that origi- ized by thought disorder, hallucinations, delusions, and
nate in the anterior subventricular zone (Baker et al., impaired cognition, particularly in memory, attention, and
2001). executive functioning. Numerous neuropsychological and
clinical neurobiological studies have implicated abnormal-
ities of fronto-temporal-limbic circuitry in the disorder
* Interestingly, amyloid -peptide, a marker for AD that is also (Arnold, 1999; Moberg et al., 1999). However, in contrast
present in some patients with PD, may contribute to this process. to AD and PD, schizophrenia is not defined by distinct
This protein specifically impairs neurogenesis in the subventricu- pathological lesions (Arnold and Trojanowski, 1996;
lar zone and cerebral cortex of adult mice and in human embry-
Damadzic et al., 2002; Harrison, 1999), and no good ani-
onic cortical neural progenitor cells grown in culture (Haughey
et al., 2002). The adverse effects of amyloid -peptide on neurogen-
mal models are presently available to examine its etiology.
esis parallels a disruption of Ca2  regulation in neural progenitor While research has highlighted both neurodevelopmental
cells. If this situation occurs in vivo, then a systematic examina- and neurodeteriorative aspects of the disorder, neurodevel-
tion of the inhibitory role of amyloid -peptide on neurogenesis opmental factors seem to play the predominant determin-
could allow insights into the olfactory loss of PD and AD, and per- ing roles. This disorder is associated with such factors
haps even into the progressive elements of these disorders. as (1) maternal starvation during the first trimester of
516 Smutzer et al.

pregnancy (Susser et al., 1996), (2) influenza infection suggesting that the olfactory deficits associated with schiz-
during the second trimester (Mednick et al., 1988), (3) rhe- ophrenia could be due to abnormalities in more central
sus and ABO blood type incompatibility (Hollister et al., olfactory structures.
1996), and (4) perinatal anoxic birth injuries (Cannon et Two more recent studies, however, have found evidence
al., 2000). Monozygotic twin studies suggest that schizo- for abnormalities in the OE of patients with schizophrenia.
phrenia has both epigenetic and genetic components, since In the first study, neuronal cultures derived from biopsy tis-
concordance for schizophrenia is between 30 and 50%. sue of living patients were evaluated (Feron et al., 1999).
Antipsychotic drugs used to treat the positive symptoms of Tissue from the schizophrenics exhibited, relative to con-
schizophrenia specifically block dopamine receptors, and trol tissue, a reduced ability to attach to culture slides and
this observation has implicated aberrant dopamine trans- contained a greater proportion of cells undergoing mitosis.
mission in the pathogenesis of schizophrenia (Goldstein Interestingly, the addition of dopamine to the cultures from
and Deutch, 1992). As noted in Chapter 23, olfactory schizophrenia patients reduced the level of apoptotic cell
deficits in schizophrenia are generally less severe relative death. In a subsequent histopathologic study, the OE of 13
to those observed in AD or PD.* elderly subjects with schizophrenia and 10 controls were
removed at autopsy (Arnold et al., 2001a). Sections were
A. The Olfactory Epithelium in Schizophrenia immunolabeled with antibodies that distinguish OE neu-
rons in different stages of development, including basal
The question as to whether the OE of patients with schiz- cells (low-affinity nerve growth factor receptor,
ophrenia differs from that of controls has been addressed p75NGFR), postmitotic immature neurons (GAP-43), and
in several studies. Smutzer et al. (1998), in an early quali- mature olfactory receptor neurons (olfactory marker pro-
tative study, found that olfactory neurons and basal cell tein). These results are shown in Figure 2. The density of
populations of both schizophrenic and control groups p75NGFR basal cells in specimens from schizophrenic
equally expressed MAP 1B and N-CAM, and all receptor patients was decreased to 37% of that in controls, whereas
neurons were immunopositive for protein gene product the density of GAP-43–positive postmitotic immature neu-
9.5. Basal cells from both groups also expressed nerve rons was increased 316% relative to controls. Marked
growth factor receptor. Schizophrenic and control subjects increases were noted in (1) the ratio of GAP-43–positive
exhibited dystrophic neurites immunoreactive for either postmitotic immature neurons to p75NGFR positive cells
synaptophysin, MAP 1B, tau, or NF proteins, and evi- (665%) and (2) the ratio of olfactory marker protein–posi-
denced seemingly equivalent immunostaining patterns, tive mature neurons to p75NGFR-positive basal cells
(328%). These findings led these authors to conclude that
“abnormal densities and ratios of OE neurons at different
stages of development indicate dysregulation of OE neu-
* One of the most consistent neurobiological findings in schizo- ronal lineage in schizophrenia.” This dysregulation could
phrenia has been ventriculomegaly seen in CT and magnetic res- be due to intrinsic factors controlling differentiation or an
onance imaging (MRI) studies (Frith et al., 1995). The degree of
inability to gain trophic support from axonal targets in the
ventricular enlargement correlates with poor premorbid adjust-
olfactory bulb. While caution is necessary in extrapolating
ment and has been found at the first onset of symptoms. Notably,
many of the postmortem brain findings suggestive of aberrant developmental findings in mature OE to early brain devel-
neurodevelopment in schizophrenia have been observed in olfac- opment, similarities in molecular events suggest that such
tory cortices and closely related areas. Cytoarchitectural and neu- studies may be instructive.
ronal morphometric studies have described abnormalities in neu-
ron arrangement, size, shape, orientation, and packing density in B. The Olfactory Bulb in Schizophrenia
various limbic and prefrontal regions (Arnold and Trojanowski,
1996). Several groups have reported abnormalities in interstitial Recently it was shown that patients with schizophrenia
white matter neuron density and placement, which suggests aber- exhibit a 20% decrease in bilateral olfactory bulb volume
rant neuronal migration or subplate neuron pruning during cere- relative to controls, as measured by quantitative MRI
brogenesis (Akbarian et al., 1996; Arnold et al., 1991a). Other
(Turetsky et al., 2000). Such a decrease is of greater magni-
findings have been interpreted as subtle “miswiring” in cortical
tude than the decreases noted in other brain structures in this
connections that may be neurodevelopmental or neuroplastic in
nature (Benes, 1993). Strengthening the neurodevelopmental the- disorder. For example, bilateral reduction of hippocampus
ory has been the failure to find any evidence for postmaturational volumes in patients with schizophrenia is 4% (Gur et al.,
neural injury or neurodegeneration (Arnold, 1999; Arnold and 1987). In the bulb this phenomenon could reflect a decrease
Trojanowski, 1996; Arnold et al., 1991b, 1998; Bloom, 1993; in cell numbers, a decrease in cell volumes, or both. One
Honer et al., 1995). possible explanation for a decrease in bulb size could be
Olfactory Circuitry in Neurological Disorders 517

Figure 2 The olfactory epithelium (OE) in schizophrenia. (a) Coronal section of nasal cavities stained with toluidine blue. The OE lines
the walls of the left and right nasal cavities along turbinates (lateral walls) and septum (medial walls). Segments for analysis were from
the OE adjacent to the apices of the nasal cavities. (Bar indicates 1 mm.) Nonpsychiatric control (b) and schizophrenia case (c) OE was
double-labeled for p75 nerve growth factor receptor (p75NGFR) plus basal cells (brown), and immature olfactory receptor neurons
(ORN) expressing growth-associated protein-43 (GAP-43) plus immature neurons (black or dark gray). Nonpsychiatric control (d) and
schizophrenia case (e). OE was double-labeled for p75NFGR and basal cells (brown), and olfactory marker protein (black). BC indicates
basal cell layer; and LP, lamina propia. Bar indicates 10 mm for b to e. (See color insert.)

apoptosis of mitral cells possibly induced by noninnervation C. The Olfactory Cortex in Schizophrenia
(Naruse and Keino, 1995), incorrect innervation of regener-
ating olfactory neurons, or interruption of neuroblast migra- A number of abnormalities in cytoarchitecture, neuronal
tion via the rostral migratory stream (Chazal et al, 2000). morphology, innervation, and neuronal protein expression
Whatever its basis, this finding suggests that the olfactory have been reported in schizophrenia that suggests a distur-
bulb may be especially vulnerable to the disease processes bance in nascent neuronal migration during fetal develop-
that produce structural brain changes in this disease. ment (Arnold et al., 1998). One example is in the entorhi-
To date, only two preliminary studies of OB immuno- nal cortex, a component of the primary olfactory cortex
histopathology have been reported. Arnold et al. (2001b) (Arnold et al., 1991c; Falkai et al., 2000; Jakob and
used a panel of antibodies and quantitative microscopy to Beckmann, 1986, 1994). Several reports have described (1)
characterize olfactory bulb glomeruli in 17 elderly schizo- poorly formed neuron clusters in layer II, (2) apparent het-
phrenics and 17 matched controls. They found significant erotopic displacement of layer II-type neurons deep into
decreases in glomerular expression of the presynaptic layer III, (3) developmental anomalies in pre-alpha cell
SNARE protein SNAP-25, decreases in receptor tyrosine clusters, and (4) poor laminar differentiation and attenua-
kinase B expression (this protein binds brain-derived neu- tion of deeper layers of the cortex (Arnold et al., 1991b;
rotrophic factor), and a decrease in expression of the post- Jakob and Beckmann, 1986). Other findings include small-
synaptic dendritic protein MAP 2. In a related study, a sig- er neuron size (Arnold et al., 1995), increased density of
nificant reduction in the expression of phosphorylation- glutamatergic vertical axons (Longson et al., 1996), and
independent MAP 2 protein was observed in olfactory decreased MAP 2 protein expression (Arnold et al.,
bulbs from an elderly, highly chronic, sample of subjects 1991c).
with schizophrenia (Rioux et al., 2001). If aberrant synap- In contrast, several studies have reported no significant
tic reinnervation is an underlying feature of schizophrenia, differences between schizophrenics and controls in
then olfactory receptor cell regeneration and synapse for- entorhinal cytoarchitecture (Akil and Lewis, 1997;
mation in the bulb could be an excellent model to examine Heinsen et al., 1996; Krimer et al., 1997a). Krimer and
aberrant neurodevelopment in this disorder. colleagues (1997a) qualitatively studied all sectors of the
518 Smutzer et al.

entorhinal cortex and found no major differences between and significant advances have been made in our basic
schizophrenics and controls. They examined neuronal understanding of odorant processing and related phenom-
numbers, density, and layer volumes in the entorhinal cor- ena, including neuronal regeneration within the OE and
tex. However, their quantitative analyses were limited to a bulb. The next decade should allow scientists and clini-
small group of seven schizophrenics and eight controls. cians to utilize this and future knowledge as a tool to
They did report a trend toward lower neuronal numbers examine and dissect the pathology and etiology of neuro-
and density (by 12–18%) in almost all layers of the most logical disorders that include AD, PD, and schizophrenia.
rostral entorhinal sectors (Krimer et al., 1997b). Due to the In particular, a better understanding of the mechanisms
role of the entorhinal cortex in olfactory processing, behind the continuously regenerating neurons of the OE,
anatomical or physiological disturbances within this tissue their inherent ability to form synapses with specific
could mediate the observed aberrant behavior and/or olfac- glomeruli within the bulb, the nature of higher-order olfac-
tory dysfunction in schizophrenic patients. tory processing, and the roles of various neurotransmitters,
The parahippocampal region has been implicated in including dopamine, in determining, modulating, and
olfactory memory processes. This region consists of the coordinating olfactory information will likely provide key
presubiculum, parasubiculum, entorhinal cortex, perirhinal information. This information is necessary for elucidating
cortex, and the parahippocampal cortex proper (Arnold, the pathology responsible for the olfactory anomalies of a
2000). Global changes in the parahippocampal region of number of neurological diseases and, in some cases, the
schizophrenics have been suggested based on volumetric diseases themselves. Such an understanding will be aided
changes in the parahippocampal gyrus. These changes are not only by the painstaking examination of specific lesions
manifested as a decrease in gyral volume in the left hemi- or genetic polymorphisms in neuropsychiatric disorders,
sphere of the brain. The first report was by Bogerts et al. but by the application of advanced molecular approaches
(1985), who conducted a systematic, but nonstereological, to this problem. Such approaches will most likely include
analysis of serial brain sections of unmedicated schizo- the use of single-cell mRNA expression profiling in con-
phrenics and normal controls that underwent autopsy from junction with cDNA microchip array technology, thereby
1928 to 1953. These authors reported a 44% mean reduc- allowing for the examination of differences in expression
tion in parahippocampal gyrus volume. This decrease in profiles of key genes that may modulate olfactory function
volume was greater than that observed in any other brain in neurological disorders (Mirnics et al., 2000).
structure they studied in these autopsy cases. These other
brain structures included the amygdala, caudate, hip-
VI. ACKNOWLEDGMENTS
pocampal formation, nucleus accumbens, putamen, and
external pallidum.
The authors wish to acknowledge the support of the
National Alliance for Research on Schizophrenia and
Depression, the National Institute for Dental and
V. SUMMARY Craniofacial Research (R03-DE 13760), the National
Institute of Aging (RO1 AG 17496), the National Institute
At present, the physiological mechanisms that underlie on Deafness and Other Communication Disorders (PO1
olfactory dysfunction in neuropsychiatric disorders remain DC 00161, RO1 DC 04278, RO1 DC 02974), National
largely unknown. In both AD and PD, olfactory deficits Institute of Mental Health (ROI MH 59344), and the
occur early in the disease process, and considerable data Theodore and Vada Stanley Foundation. The authors wish
suggest that CNS neural networks involved in processing to thank John Duda and Graeme Lowe for their valuable
olfactory information may be especially vulnerable to neu- assistance.
rodegeneration (Liberini et al., 2000). Whether or to what
degree environmental agents play a role in the etiology of
the olfactory dysfunction in these disorders is not known. REFERENCES
In AD, PD, and schizophrenia, alterations in all major
Adlard, P. A., West, A. K., and Vickers, J. C. (1998). Increased
stages of the neural circuitry responsible for identifying
density of metallothionein I/II-immunopositive cortical glial
and processing olfactory information appear to be present. cells in the early stages of Alzheimer’s disease. Neurobiol.
Which of these degenerative processes or neurode- Dis. 5:349–356.
velopmental anomalies are responsible for the olfactory Akbarian, S., Kim, J. J., Potkin, S. G., Hetrick, W. P., Bunney,
dysfunction remains a mystery. W. E. Jr, and Jones, E. G. (1996). Maldistribution of interstitial
On a positive note, the olfactory system has generated neurons in prefrontal white matter of the brains of schizo-
intense interest among scientists within the last decade, phrenic patients. Arch. Gen. Psychiatry 53:425–436.
Olfactory Circuitry in Neurological Disorders 519

Akil, M, and Lewis, D. A. (1997). Cytoarchitecture of the Subclinical dopaminergic dysfunction in asymptomatic
entorhinal cortex in schizophrenia. Am. J. Psychiatry Parkinson’s disease patients’ relatives with a decreased sense
154:1010–1012. of smell. Ann. Neurol. 50:34–41.
Arnold, S. E. (1999). Neurodevelopmental abnormalities in Bezprozvanny, I., Watras, J., and Ehrlich, B. E. (1991). Bell-
schizophrenia: insights from neuropathology. Dev. shaped calcium-response curves of Ins (1, 4, 5)P3-and calci-
Psychopathol. 11:439–456. um-gated channels from endoplasmic reticulum of cerebel-
Arnold, S. E. (2000). Cellular and molecular neuropathology of lum. Nature 351:751–754.
the parahippocampal region in schizophrenia. Ann. NY Acad. Bloom, F. E. (1993). Advancing a neurodevelopmental origin for
Sci. 911:275–292. schizophrenia. Arch. Gen. Psychiatry 50:224–227.
Arnold, S. E., and Trojanowski, J. Q. (1996). Cognitive impair- Bogerts, B., Hantsch, J., and Herzer, M. (1983). A morphometric
ment in elderly schizophrenia: a dementia (still) lacking dis- study of the dopamine-containing cell groups in the mesen-
tinctive histopathology. Schizophr. Bull. 22:5–9. cephalon of normals, Parkinson patients, and schizophrenics.
Arnold, S. E., Hyman, B. T., Flory, J., Damasio, A. R., and van Biol. Psychiatry 18:951–969.
Hoesen, G. W. (1991a). The topographical and neuroanatom- Bogerts, B., Meertz, E., and Schonfeldt-Bausch, R. (1985). Basal
ical distribution of neurofibrillary tangles and neuritic plaques ganglia and limbic system pathology in schizophrenia. A mor-
in the cerebral cortex of patients with Alzheimer’s disease. phometric study of brain volume and shrinkage. Arch. Gen.
Cerebral Cortex 1:103–116. Psychiatry 42:784–791.
Arnold, S. E., Hyman, B. T., van Hoesen, G. W., and Damasio, A. Braak, H., and Braak, E. (1990). Cognitive impairment in
R. (1991b). Some cytoarchitectural abnormalities of the Parkinson’s disease: amyloid plaques, neurofibrillary tangles,
entorhinal cortex in schizophrenia. Arch. Gen. Psychiatry and neuropil threads in the cerebral cortex. J. Neural
48:625–632. Transmission Parkinson’s Dis and Dement. Sec. 2:45–57.
Arnold, S. E., Lee, V. M., Gur, R. E., and Trojanowski, J. Q. (1991c). Braak, H., and Braak, E. (1991). Neuropathological staging of
Abnormal expression of two microtubule-associated proteins Alzheimer-related changes. Acta Neuropathol. 82:
(MAP2 and MAP5) in specific subfields of the hippocampal 239–259.
formation in schizophrenia. Proc. Natl. Acad. Sci. USA Brouillard, M., Laccourreye, L., Jabbour, W., Emile, J., and
88:10850–10854. Pouplard-Barthelaix, A. (1994). (Ultrastructural and immuno-
Arnold, S. E., Hyman, B. T., and van Hoesen, G. W. (1994). histochemical study of the olfactory mucosa in Alzheimer’s
Neuropathologic changes of the temporal pole in Alzheimer’s disease) (in French). Bull. Assoc. Anat. 78:25–28.
disease and Pick’s disease. Arch. Neurol. 51:145–150. Cannon, T. D., Rosso, I. M., Hollister, J. M., Bearden, C. E.,
Arnold, S. E., Franz, B. R., Gur, R. C., Gur, R. E., Shapiro, R. M., Sanchez, L. E., and Hadley, T. (2000). A prospective cohort
Moberg, P. J., and Trojanowski, J. Q. (1995). Smaller neuron study of genetic and perinatal influences in the etiology of
size in schizophrenia in hippocampal subfields that mediate schizophrenia. Schizophr. Bull. 26:351–366.
cortical-hippocampal interactions. Am. J. Psychiatry Carrell, R. W. (1988). Alzheimer’s disease. Enter a protease
152:738–748. inhibitor. Nature 331:478–479.
Arnold, S. E., Smutzer, G. S., Trojanowski, J. Q., and Moberg, P. Chase, T. N. (1997). A gene for Parkinson disease. Arch. Neurol.
J. (1998). Cellular and molecular neuropathology of the 54:1156–1157.
olfactory epithelium and central olfactory pathways in Chazal, G., Durbec, P., Jankovski, A., Rougon, G., and Cremer,
Alzheimer’s disease and schizophrenia. Ann. NY Acad. Sci. H. (2000). Consequences of neural cell adhesion molecule
855:762–775. deficiency on cell migration in the rostral migratory stream of
Arnold, S. E., Han, L.–Y., Moberg, P. J., Turetsky, B. I., Gur, R. the mouse. J. Neurosci. 20:1446–1457.
E., Trojanowski, J. Q., and Hahn, C.–G. (2001a). Chu, C. C., Tranel, D., Damasio, A. R., and van Hoesen, G. W.
Dysregulation of olfactory receptor neuron lineage in schizo- (1997). The autonomic-related cortex: pathology in
phrenia. Arch. Gen. Psychiatry 58:829–835. Alzheimer’s disease. Cerebral Cortex 7:86–95.
Arnold, S. E., Rioux, L., Han, L-Y. (2001b). Molecular markers Chuah, M. I., and Getchell, M. L. (1999). Metallothionein in
of axon guidance and synaptogenesis in the olfactory bulb in olfactory mucosa of Alzheimer’s disease patients and apoE-
schizophrenia. Soc. Neurosci Abst. Prog. No. 454.5. deficient mice. Neuroreport 10:1919–1924.
Averback, P. (1983). Two new lesions in Alzheimer’s disease. Crino, P. B., Martin, J. A., Hill, W. D., Greenberg, B., Lee, V. M.,
Lancet 2(8360):1203. and Trojanowski, J. Q. (1995). Beta-amyloid peptide and
Baker, H., Liu, N., Chun, H. S., Saino, S., Berlin, R., Volpe, B., amyloid precursor proteins in olfactory mucosa of patients
and Son, J. H. (2001). Phenotypic differentiation during with Alzheimer’s disease, Parkinson’s disease, and Down syn-
migration of dopaminergic progenitor cells to the olfactory drome. Ann. Otol. Rhinol. Laryngol. 104:655–661.
bulb. J. Neurosci. 21:8505–8513. Daniel, S. E., and Hawkes, C. H. (1992). Preliminary diagnosis
Benes, F. M. (1993). The relationship between structural brain of Parkinson’s disease by olfactory bulb pathology. Lancet
imaging and histopathologic findings in schizophrenia 340:186.
research. Harv. Rev. Psychiatry 1:100–109. Datiche, F., and Cattarelli, M. (1996). Catecholamine innervation
Berendse, H. W., Booij, J., Francot, C. M., Bergmans, P. L., of the piriform cortex: a tracing and immunohistochemical
Hijman, R., Stoof, J. C., and Wolters, E. C. (2001). study in the rat. Brain Res. 710:69–78.
520 Smutzer et al.

Descarries, L., Lemay, B., Doucet, G., and Berger, B. (1987). quantitative evidence of a developmental abnormality. Biol.
Regional and laminar density of the dopamine innervation in Psychiatry 47:937–943.
adult rat cerebral cortex. Neuroscience 21:807–824. Feron, F., Perry, C., Hirning, M. H., McGrath, J., and Mackay-
Davies, D. C., Brooks, J. W., and Lewis, D. A. (1993). Axonal Sim, A. (1999). Altered adhesion, proliferation and death in
loss from the olfactory tracts in Alzheimer’s disease. neural cultures from adults with schizophrenia. Schizophr.
Neurobiol. Aging 14:353–357. Res. 40:211–218.
Damadzic, R., Shuangshoti, S., Giblen, G., and Herman, M. M. Forno, L. S. (1996). Neuropathology of Parkinson’s disease.
(2002). Neuritic pathology is lacking in the entorhinal cortex, J. Neuropath. Exper. Neurol. 55:259–272.
subiculum and hippocampus in middle-aged adults with Frith, C. D., Friston, K. J., Herold, S., Silbersweig, D., Fletcher,
schizophrenia, bipolar disorder or unipolar depression. Acta P., Cahill, C., Dolan, R. J., Frackowiak, R. S., and Liddle, P. F.
Neuropathol. 103:488–494. (1995). Regional brain activity in chronic schizophrenic
Doty, R. I. (1997). Studies from the University of Pennsylvania patients during the performance of a verbal fluency task. Br. J.
Smell and Taste Center. Chem. Senses 22:565–586. Psychiatry 167:343–349.
Doty, R. L. (2001). Olfaction and gustation in normal aging and Galvin, J. E., Uryu, K., Lee, V. M., and Trojanowski, J. Q. (1999).
Alzheimer’s disease. In Functional Neurobiology of Aging, Axon pathology in Parkinson’s disease and Lewy body
P. R. Hof and C. V. Mobbs (Eds.). Academic Press, San Diego, dementia hippocampus contains alpha-, beta-, and gamma-
pp. 647–658. synuclein. Proc. Natl. Acad. Sci. USA 96:13450–13455.
Doty, R. L., and Risser, J. M. (1989). Influence of the D-2 Getchell, T. V., Krishna, N. S., Dhooper, N., Sparks, D. L., and
dopamine receptor agonist quinpirole on the odor detection Getchell, M. L. (1995). Human olfactory receptor neurons
performance of rats before and after spiperone administration. express heat shock protein 70: age-related trends. Ann. Otol.
Psychopharmacology 98:310–315. Rhinol. Laryngol. 104:47–56.
Doty, R. L., Deems, D. A., and Stellar, S. (1988). Olfactory dys- Goldstein, M., and Deutch, A. Y. (1992). Dopaminergic mecha-
function in parkinsonism: a general deficit unrelated to neu- nisms in the pathogenesis of schizophrenia. FASEB J.
rologic signs, disease stage, or disease duration. Neurology 6:2413–2421.
38:1237–1244. Gray, A. J., Staples, V., Murren, K., Dhariwal, A., and Bentham,
Doty, R. L., Stern, M. B., Pfeffer, C., Gollomp, S. M., and Hurtig, P. (2001). Olfactory identification is impaired in clinic-based
H. I. (1992). Bilateral olfactory dysfunction in early stage patients with vascular dementia and senile dementia of
treated and untreated idiopathic Parkinson’s disease. J. Alzheimer type. Int. J. Geriatr. Psychiatry 16:513–517.
Neurol. Neurosurg. Psychiatry 55:138–142. Gur, R. E., Resnick, S. M., Alavi, A., Gur, R. C., Caroff, S., Dann,
Doty, R. L., Li, C., Mannon, L. J., and Yousem, D. M. (1997). R., Silver, F. L., Saykin, A. J., Chawluk, J. B., and Kushner,
Olfactory dysfunction in multiple sclerosis. N. Engl. J. Med. M. (1987). Regional brain function in schizophrenia. I. A
336:1918–1919. positron emission tomography study. Arch. Gen. Psychiatry
Doty, R. L., Li, C., Bagla, R., Huang, W., Pfeiffer, C., Brosvic, G. 44:119–125.
M., and Risser, J. M. (1998). SKF 38393 enhances odor detec- Halasz, N., Johansson, O., Hokfelt, T., Ljungdahl, A., and
tion performance. Psychopharmacology 136:75–82. Goldstein, M. (1981). Immunohistochemical identification of
Duda, J. E., Shah, U., Arnold, S. E., Lee, V. M., and Trojanowski, two types of dopamine neuron in the rat olfactory bulb as seen
J. Q. (1999). The expression of alpha-, beta-, and gamma- by serial sectioning. Neurocytology 10:251–259.
synucleins in olfactory mucosa from patients with and Harrison, P. J. (1999). The neuropathology of schizophrenia. A
without neurodegenerative diseases. Exp. Neurol. critical review of the data and their interpretation. Brain
160:515–522. 122:593–624.
Engelender, S., Kaminsky, Z., Guo, X., Sharp, A. H., Amaravi, R. Harrison, P. J., Procter, A. W., Exworthy, T., Roberts, G. W.,
K., Kleiderlein, J. J., Margolis, R. L., Troncoso, J. C., Najlerahim, A., Barton, A. J., and Pearson, R. C. (1993). Heat
Lanahan, A. A., Worley, P. F., Dawson, V. L., Dawson, T. M., shock protein (hsx70) mRNA expression in human brain:
and Ross, C. A. (1999). Synphilin-1 associates with alpha- effects of neurodegenerative disease and agonal state.
synuclein and promotes the formation of cytosolic inclusions Neuropath. Appl. Neurobiol. 19:10–21.
Nat. Genet. 22:110–114. Haughey, N. J., Liu, D., Nath, A., Borchard, A. C., and Mattson,
Englund, U., Fricker-Gates, R. A., Lundberg, C., Bjorklund, A., M. P. (2002). Disruption of neurogenesis in the subventricular
and Wictorin, K. (2002). Transplantation of human neural zone of adult mice, and in human cortical neuronal precursor
progenitor cells into the neonatal rat brain: extensive migra- cells in culture, by amyloid beta-peptide: Implications for the
tion and differentiation with long-distance axonal projections. pathogenesis of Alzheimer’s disease. Neuromol. Med.
Exp. Neurol. 173:1–21. 1:125–135.
Esiri, M. M., and Wilcock, G. K. (1984). The olfactory bulbs of Hawkes, C. H., Shephard, B. C., and Daniel, S. E. (1997).
Alzheimer’s disease. J. Neurol. Neurosurg. Psychiatry Olfactory dysfunction in Parkinson’s disease. J. Neurol.
47:56–60. Neurosurg. Psychiatry 62:436–446.
Falkai, P., Schneider-Axmann, T., and Honer, W. G. (2000). Heinsen, H., Grossman, E., Rub, U., Eisenmenger, W., Bauer, M.,
Entorhinal cortex pre-alpha cell clusters in schizophrenia: Ulmar, G., et al. (1996). Variability in the human entorhinal
Olfactory Circuitry in Neurological Disorders 521

region may confound neuropsychiatric diagnoses. Acta Kovacs, T., Cairns, N. J., and Lantos, P. I. (2001). Olfactory cen-
Anatom. 157:226–237. tres in Alzheimer’s disease: Olfactory bulb is involved in early
Hock, C., Golombowski, S., Muller-Spahn, F., Peschel, O., Braak’s stages. Neuroreport 12:285–288.
Riederer, A., Probst, A., Mandelkow, E., and Unger, J. (1998). Krimer, L. S., Herman, M. M., Saunders, R. C., Boyd, J. C.,
Histological markers in nasal mucosa of patients with Hyde, T. M., Carter, J. M., et al. (1997a). A qualitative and
Alzheimer’s disease. Eur. Neurol. 40:31–36. quantitative analysis of the entorhinal cortex in schizophrenia.
Hollister, J. M., Laing, P., and Mednick, S. A. (1996). Rhesus Cerebral Cortex 7:732–739.
incompatibility as a risk factor for schizophrenia in male Krimer, L. S., Hyde, T. M., Herman, M. M., and Saunders, R. C.
adults. Arch. Gen. Psychiatry 53:19–24. (1997b). The entorhinal cortex: an examination of cyto- and
Honer, W. G., Squires-Wheeler, E., Smith, G. N., Sharif, Z., myeloarchitectonic organization in humans. Cerebral Cortex
Chan, S., and Gewirtz, G. (1995). Developmental abnormali- 7:722–731.
ties and cortical sulcal enlargement in psychosis. Schizophr. Kromer Vogt, L. J., Hyman, B. T., van Hoesen, G. W., and
Res. 16:121–125. Damasio, A. R. (1990). Pathological alterations in the amyg-
Hoogland, P. V., and Huisman, E. (1999). Tyrosine hydroxylase dala in Alzheimer’s disease. Neuroscience 37:377–385.
immunoreactive structures in the aged human olfactory bulb Kruger, R., Kuhn, W., Muller, T., Woitalla, D., Graeber, M.,
and olfactory peduncle. J. Chem. Neuroanat. 17:153–161. Kosel, S., Przuntek, H., Epplen, J. T., Schols, L., and Riess, O.
Hsia, A. Y., Vincent, J. D., and Lledo, P. M. (1999). Dopamine (1998). Ala30Pro mutation in the gene encoding alpha-synu-
depresses synaptic inputs into the olfactory bulb. clein in Parkinson’s disease. Nature Genetics 18:106–108.
J. Neurophysiol. 82:1082–1085. Kulkarni-Narla, A., Getchell, T. V., Schmitt, F. A., and Getchell,
Hubbard, M. J. (2000). Calcium transport across the dental M. L. (1996). Manganese and copper-zinc superoxide dismu-
enamel epithelium. Cri. Rev. Oral Biol. Med. 11:437–466. tases in the human olfactory mucosa: increased immunoreac-
Hurtig, H. I., Trojanowski, J. Q., Galvin, J., Ewbank, D., tivity in Alzheimer’s disease. Exp. Neurol. 140:115–125.
Schmidt, M. L., Lee, V. M., Clark, C. M., Glosser, G., Stern, Lee, J. H., Goedert, M., Hill, W. D., Lee, V. M., and Trojanowski,
M. B., Gollomp, S. M., and Arnold, S. E. (2000). Alpha-synu- J. Q. (1993). Tau proteins are abnormally expressed in olfac-
clein cortical Lewy bodies correlate with dementia in tory epithelium of Alzheimer patients and developmentally
Parkinson’s disease. Neurology 54:1916–1921. regulated in human fetal spinal cord. Exp. Neurol. 121:93–105.
Hyman, B. T., Van Horsen, G. W., Damasio, A. R., and Barnes, Lee, V. M., Balin, B. J., Otvos, L., Jr., and Trojanowski, J. Q.
C. L. (1984). Alzheimer’s disease: cell-specific pathology iso- (1991). A68: a major subunit of paired helical filaments and
lates the hippocampal formation. Science 225:1168–1170. derivatized forms of normal Tau. Science 251:675–678.
Jakob, H., and Beckmann, H. (1986). Prenatal developmental dis- Le Moal, M. (1995). Mesocorticolimbic dopaminergic neurons:
turbances in the limbic allocortex in schizophrenics. J. Neural Functional and regulatory roles. In Psychopharmacology; The
Trans. 65:303–326. Fourth Generation of Progress, F. E. Bloom and D. Kupfer
Jakob, H., and Beckmann, H. (1994). Circumscribed malforma- (Eds.). Raven Press, New York, pp. 283–299.
tion and nerve cell alterations in the entorhinal cortex of Lewy, F. H. (1912). Paralysis agitans: I. Pathologische anatomie.
schizophrenics. Pathogenetic and clinical aspects. J. Neural In Handbüch der Neurologie III. Springer, Berlin, pp.
Trans. 98:83–106. 920–933.
Kaakkola, S., Palo, J., Malmberg, H., Sulkava, R., and Virtanen, Li, P., and Chen, S. R. (2001). Molecular basis of Ca2activation
I. (1994). Neurofilament profile in olfactory mucosa of of the mouse cardiac Ca2 release channel (ryanodine recep-
patients with a clinical diagnosis of Alzheimer’s disease. tor). J. Gen. Physiol. 118:33–44.
Virchows Arch. 424:315–319. Li, Y. M., Xu, M., Lai, M. T., Huang, Q., Castro, J. L., DiMuzio-
Kalaria, R. N., and Ballard, C. (1999). Overlap between patholo- Mower, J., Harrison, T., Lellis, C., Nadin, A., Neduvelil, J. G.,
gy of Alzheimer disease and vascular dementia. Alz. Dis. Register, R. B., Sardana, M. K., Shearman, M. S., Smith, A. L.,
Assoc. Dis. 13(suppl 3):S115–S123. Shi, X. P., Yin, K. C., Shafer, J. A., and Gardell, S. J. (2000).
Kasa, P., Rakonczay, Z., and Gulya, K. (1997). The cholinergic Photoactivated gamma-secretase inhibitors directed to the
system in Alzheimer’s disease. Prog. Neurobiol. 52:511–535. active site covalently label presenilin 1. Nature 405:689–694.
Kishikawa, M., Iseki, M., Nishimura, M., Sekine, I., and Fujii, H. Liberini, P., Parola, S., Spano, P. F., and Antonini, L. (2000).
(1990). A histopathological study on senile changes in the Olfaction in Parkinson’s disease: methods of assessment and
human olfactory bulb. Acta Pathol. Jpn. 40:255–260. clinical relevance. J. Neurol. 247:88–96.
Kleene, S. J. (1993). Origin of the chloride current in olfactory Lippa, C. F., Fujiwara, H., Mann, D. M., Giasson, B., Baba, M.,
transduction. Neuron 11:123–132. Schmidt, M. L., Nee, L. E., O’Connell, B., Pollen, D. A.,
Kleene, S. J. (1999). Both external and internal calcium reduce George-Hyslop, P., Ghetti, B., Nochlin, D., Bird, T. D.,
the sensitivity of the olfactory cyclic-nucleotide-gated chan- Cairns, N. J., Lee, V. M., Iwatsubo, T., and Trojanowski, J. Q.
nel to cAMP. J. Neurophysiol. 81:2675–2682. (1998). Lewy bodies contain altered alpha-synuclein in brains
Kornack, D. R., and Rakic, P. (2001). The generation, migration, of many familial Alzheimer’s disease patients with mutations
and differentiation of olfactory neurons in the adult primate in presenilin and amyloid precursor protein genes. Am. J.
brain. Proc. Natl. Acad. Sci. USA 98:4752–4757. Pathol. 153:1365–1370.
522 Smutzer et al.

Longson, D., Deakin, J. F., and Benes, F. M. (1996). Increased Pearce, R. K., Hawkes, C. H., and Daniel, S. E. (1995). The anter-
density of entorhinal glutamate-immunoreactive vertical ior olfactory nucleus in Parkinson’s disease. Mov. Disord.
fibers in schizophrenia. J. Neural Trans. 103:503–507. 10:283–287.
Loopuijt, L. D., Sebens, J. B. (1990). Loss of dopamine receptors Pearson, R. C. (1996). Cortical connections and the pathology of
in the olfactory bulb of patients with Alzheimer’s disease. Alzheimer’s disease. Neurodegeneration 5:429–434.
Brain Res. 529:239–244. Pencea, V., Bingaman, K. D., Freedman, L. J., and Luskin, M. B.
Mattila, P. M., Rinne, J. O., Helenius, H., and Roytta, M. (1999). (2001). Neurogenesis in the subventricular zone and rostral
Neuritic degeneration in the hippocampus and amygdala in migratory stream of the neonatal and adult primate forebrain.
Parkinson’s disease in relation to Alzheimer pathology. Acta Exp. Neurol. 172:1–16.
Neuropathol. 98:157–164. Polymeropoulos, M. H., Lavedan, C., Leroy, E., Ide, S. E.,
Mattila, P. M., Rinne, J. O., Helenius, H., Dickson, D. W., and Dehejia, A., Dutra, A., Pike, B., Root, H., Rubenstein, J.,
Roytta, M. (2000). Alpha-synuclein-immunoreactive cortical Boyer, R., Stenroos, E. S., Chandrasekharappa, S.,
Lewy bodies are associated with cognitive impairment in Athanassiadou, A., Papapetropoulos, T., Johnson, W. G.,
Parkinson’s disease. Acta Neuropathol. (Berl) 100:285–290. Lazzarini, A. M., Duvoisin, R. C., Di Iorio, G., Golbe, L. I., and
McShane, R. H., Nagy, Z., Esiri, M. M., King, E., Joachim, C., Nussbaum, R. L. (1997). Mutation in the alpha-synuclein
Sullivan, N., and Smith, A. D. (2001). Anosmia in dementia is gene identified in families with Parkinson’s disease. Science
associated with Lewy bodies rather than Alzheimer’s patholo- 276:2045–2047.
gy. J. Neurol. Neurosurg. Psychiatry 70:739–743. Price, J. L., Ko, A. I., Wade, M. J., Tsou, S. K., McKeel, D. W., and
Mednick, S. A., Machon, R. A., Huttunen, M. O., and Bonett, D. Morris, J. C. (2001). Neuron number in the entorhinal cortex
(1988). Adult schizophrenia following prenatal exposure to an and CA1 in preclinical Alzheimer disease. Arch. Neurol.
influenza epidemic. Arch. Gen. Psychiatry 45:189–192. 58:1395–1402.
Meisami, E., Mikhail, L., Baim, D., Bhatnagar, K. P. (1998). Rabinovitch, A., Suarez-Pinzon, W. L., Sooy, K., Strynadka, K.,
Human olfactory bulb: aging of glomeruli and mitral cells and and Christakos, S. (2001). Expression of calbindin-D(28k) in
a search for the accessory olfactory bulb. Ann. NY Acad. Sci. a pancreatic islet beta-cell line protects against cytokine-
855:708–715. induced apoptosis and necrosis. Endocrinology
Mirnics, K., Middleton, F. A., Marquez, A., Lewis, D. A., and 142:3649–3655.
Levitt, P. (2000). Molecular characterization of schizophrenia Rama Krishna, N. S., Getchell, M. L., Buiakova, O. I., Margolis,
viewed by microarray analysis of gene expression in pre- F. L., and Getchell, T. V. (1995). Human olfactory receptor
frontal cortex. Neuron 28:53–67. neurons contain OMP mRNA in their dendritic and axonal
Moberg, P. J., Agrin, R., Gur, R. E., Gur, R. C., Turetsky, B. I., processes. Neuroreport 6:817–821.
and Doty, R. L. (1999). Olfactory dysfunction in schizophre- Rebeck, G. W., Reiter, J. S., Strickland, D. K., and Hyman, B. T.
nia: a qualitative and quantitative review. Neuropsycho- (1993). Apolipoprotein E in sporadic Alzheimer’s disease:
pharmacology 21:325–340. allelic variation and receptor interactions. Neuron 11:575–580.
Murphy, G. M., Jr., Forno, L. S., Ellis, W. G., Nochlin, D., Levy- Reyes, P. F., Golden, G. T., Fagel, P. L., Fariello, R. G., Katz, L.,
Lahad, E., Poorkaj, P., Bird, T. D., Jiang, Z., and Cordell, B. and Carner, E. (1987). The prepiriform cortex in dementia of
(1996). Antibodies to presenilin proteins detect neurofibrillary the Alzheimer type. Arch. Neurol. 44:644–645.
tangles in Alzheimer’s disease. Am. J. Pathol. Richarz, A. N., and Bratter, P. (2002). Speciation analysis of trace
149:1839–1846. elements in the brains of individuals with Alzheimer’s disease
Naruse, I., and Keino, H. (1995). Apoptosis in the developing with special emphasis on metallothioneins. Anal. Bioanal.
CNS. Prog. Neurobiol. 47:135–155. Chem. 372:412–417.
Ohm, T. G., and Braak, H. (1987). Olfactory bulb changes in Rioux, L., Ruscheinsky, D., and Arnold, S. E. (2001). Expression
Alzheimer’s disease. Acta Neuropathol. 73:365–369. and phosphorylation of microtubule-associated protein MAP2
Ohm, T. G., Muller, H., and Braak, E. (1991). Calbindin-D-28k- in the olfactory bulb in schizophrenia. Soc. Neurosci. Abstr.
like immunoreactive structures in the olfactory bulb and anter- Prog. No. 454.5.
ior olfactory nucleus of the human adult: distribution and cell Roy, S., and Wolman, L. (1969). Ultrastructural observations in
typology — partial complementarity with parvalbumin. Parkinsonism. J. Pathol. 99:39–44.
Neuroscience 142:823–840. Schmidt, M. L., Murray, J., Lee, V. M., Hill, W. D., Wertkin, A.,
Okamoto, K., Hirai, S., Shoji, M., and Takatama, M. (1990). and Trojanowski, J. Q. (1991). Epitope map of neurofilament
Senile changes in the human olfactory bulbs. In Basic, protein domains in cortical and peripheral nervous system
Clinical, and Therapeutic Aspects of Alzheimer’s and Lewy bodies. Am. J. Pathol. 139:53–65.
Parkinson’s Diseases, T. Nagatsu (Ed.). Plenum Press, New Smith, C. G. (1942). Age incident of atrophy of olfactory nerves
York, pp. 349–352. in man. J. Comp. Neurol. 77:589–594.
Papadimitriou, A., Veletza, V., Hadjigeorgiou, G. M., Patrikiou, Smith, R. L., Baker, H., Kolstad, K., Spencer, D. D., and Greer,
A., Hirano, M., and Anastasopoulos, I. (1999). Mutated alpha- C. A. (1991). Localization of tyrosine hydroxylase and olfac-
synuclein gene in two Greek kindreds with familial PD: tory marker protein immunoreactivities in the human and
Incomplete penetrance? Neurology 52:651–654. macaque olfactory bulb. Brain Res. 548:140–148.
Olfactory Circuitry in Neurological Disorders 523

Smutzer, G., Zimmerman, J. E., Hahn, C. G., Ruscheinsky, D. D., Ueda, K., Fukushima, H., Masliah, E., Xia, Y., Iwai, A.,
Rodriguez, A., Han, L. Y., and Arnold, S. E. (1997). Inositol Yoshimoto, M., Otero, D. A., Kondo, J., Ihara, Y., and Saitoh,
1,4,5-trisphosphate receptor expression in mammalian olfac- T. (1993). Molecular cloning of cDNA encoding an unrecog-
tory tissue. Brain Res. 44:347–354. nized component of amyloid in Alzheimer disease. Proc. Natl.
Smutzer, G., Lee, V. M., Trojanowski, J. Q., and Arnold, S. E. Acad. Sci. USA 90:11282–11286.
(1998). Human olfactory mucosa in schizophrenia. Ann. Otol. van Hoesen, G. W., Augustinack, J. C., Dierking, J., Redman, S.
Rhinol. Laryngol. 107:349–355. J., and Thangavel, R. (2000). The parahippocampal gyrus in
Spillantini, M. G., Crowther, R. A., Jakes, R., Hasegawa, M., and Alzheimer’s disease. Clinical and preclinical neuroanatomical
Goedert, M. (1998). Alpha-synuclein in filamentous inclu- correlates. Ann. NY. Acad. Sci. 911:254–274.
sions of Lewy bodies from Parkinson’s disease and dementia Wakabayashi, K., Engelender, S., Yoshimoto, M., Tsuji, S., Ross,
with Lewy bodies. Proc. Natl. Acad. Sci. USA 95:6469–6473. C. A., and Takahashi, H. (2000). Synphilin-1 is present in
Struble, R. G., and Clark, H. B. (1992). Olfactory bulb lesions in Lewy bodies in Parkinson’s disease. Ann. Neurol. 47:521–523.
Alzheimer’s disease. Neurobiol. Aging 13:469–473. Welch, W. J. (1992). Mammalian stress response: cell physio-
Susser, E., Neugebauer, R., Hoek, H. W., Brown, A. S., Lin, S., logy, structure/function of stress proteins, and implications for
Labovitz, D., and Gorman, J. M. (1996). Schizophrenia after pre- medicine and disease. Physiol. Rev. 72:1063–1081.
natal famine. Further evidence. Arch. Gen. Psychiatry 53:25–31. Whitehouse, P. J., Price, D. L., Struble, R. G., Clark, A. W.,
Tabaton, M., Cammarata, S., Mancardi, G. L., Cordone, G., Coyle, J. T., and Delon, M. R. (1982). Alzheimer’s disease
Perry, G., and Loeb, C. (1991). Abnormal tau-reactive fila- and senile dementia: loss of neurons in the basal forebrain.
ments in olfactory mucosa in biopsy specimens of patients Science 215:1237–1239.
with probable Alzheimer’s disease. Neurology 41:391–394. Wolters, E. C., Francot, C., Bergmans, P., Winogrodzka, A.,
Talamo, B. R., Rudel, R., Kosik, K. S., Lee, V. M., Neff, S., Booij, J., Berendse, H. W., and Stoof, J. C. (2000). Preclinical
Adelman, L., and Kauer, J. S. (1989). Pathological changes in (premotor) Parkinson’s disease. J. Neurol. 247 (suppl
olfactory neurons in patients with Alzheimer’s disease. Nature 2):II103–II109.
337:736–739. Yamagishi, M., Ishizuka, Y., and Seki, K. (1994a). Pathology of
Talamo, B. R., Feng, W. H., and Stockmayer, M. (1994). Human olfactory mucosa in patients with Alzheimer’s disease. Ann.
olfactory epithelium: normal patterns and types of lesions Otol. Rhinol. Laryngol. 103:421–427.
found in the general population. Inhal. Toxicol. 6:249–275. Yamagishi, M., and Ishizuka, Y. (1994b). (Abnormal tau protein
ter Laak, H. J., Renkawek, K., and van Workum, F. P. (1994). The expression in biopsied human olfactory mucosa) (in
olfactory bulb in Alzheimer disease: a morphologic study of Japanese). Nippon Jibiinkoka Gakkai Kaiho J. Oto-Rhino-
neuron loss, tangles, and senile plaques in relation to olfac- Laryngol. Soc. Jpn. 97:466–472.
tion. Alz. Dis. Assoc. Dis. 8:38–48. Yamagishi, M., Takami, S., and Getchell, T. V. (1996). Ontogenetic
Tissingh, G., Berendse, H. W., Bergmans, P., DeWaard, R., expression of spot 35 protein (calbindin-D28k) in human olfac-
Drukarch, B., Stoof, J. C., and Wolters, E. C. (2001). Loss of tory receptor neurons and its decrease in Alzheimer’s disease
olfaction in de novo and treated Parkinson’s disease: Possible patients. Ann. Otol. Rhinol. Laryngol. 105:132–139.
implications for early diagnosis. Mov. Disord. 16:41–46. Yamagishi, M., Getchell, M. L., Takami, S., and Getchell, T. V.
Trojanowski, J. Q., and Lee, V. M. (1998). Aggregation of neuro- (1998). Increased density of olfactory receptor neurons
filament and alpha-synuclein proteins in Lewy bodies: impli- immunoreactive for apolipoprotein E in patients with
cations for the pathogenesis of Parkinson disease and Lewy Alzheimer’s disease. Ann. Otol. Rhinol. Laryngol. 107:
body dementia. Arch. Neurol. 55:151–152. 421–426.
Trojanowski, J. Q., Newman, P. D., Hill, W. D., and Lee, V. M. Yousem, D. M., Geckle, R. J., Bilker, W. B., and Doty, R. L.
(1991). Human olfactory epithelium in normal aging, (1998). Olfactory bulb and tract and temporal lobe volumes.
Alzheimer’s disease, and other neurodegenerative disorders. Normative data across decades. Ann. NY Acad. Sci.
J. Comp. Neurol. 310:365–376. 855:546–555.
Turetsky, B. I., Moberg, P. J., Yousem, D. M., Doty, R. L., Arnold, Zufall, F., Leinders-Zufall, T., and Greer, C. A. (2000).
S. E., and Gur, R. I. (2000). Olfactory bulb volume is reduced Amplification of odor-induced Ca2release and its role in
in patients with schizophrenia. Am. J. Psychiatry 157:828–830. olfactory signal transduction J. Neurophysiol. 83:501–512
25

Multiple Chemical Intolerance

Claudia S. Miller
University of Texas Health Science Center, San Antonio, Texas, U.S.A.

I. BACKGROUND substances these patients implicate are structurally unre-

lated, an observation unsettling for toxicologists and
From time to time, physicians encounter patients with a immunologists, who tend to think of receptors, biochemi-
chief complaint of hyperosmia accompanied by multisys- cal pathways, and target organs as being substance- or at
tem symptoms and intolerances for a wide variety of least class-specific.
chemicals, foods, and/or drugs (Ashford and Miller, 1998; Third, these patients report a baffling array of symp-
Berglund et al., 1992; Doty et al., 1998). Often these toms involving any and every organ system, and often sev-
patients say they become ill when exposed to various eral systems simultaneously. Fourth, the vast majority of
odors, often in response to cacosmia (Ryan et al., 1988) or patients attest that chemical odors and even nonodorous
dysosmia (Miller, 1996). For example, they may describe chemical exposures trigger cognitive difficulties and mood
everyday exposures to fragrances, diesel exhaust, new disturbances, symptoms physicians tend to see as psy-
plastic car interiors, household cleaners, etc. (Table 1), as chogenic. The fact that these patients’ symptoms overlap
being overpowering, “stronger than ever before,” or with those of chronic fatigue syndrome, somatoform
“extremely irritating” and triggering symptoms such as disorder, fibromyalgia, panic disorder, and posttraumatic
headaches, fatigue, memory difficulties, mental confusion, stress disorder adds to the confusion.
anxiety, irritability, depression, myalgias, arrhythmias, Despite the phenomenon’s seeming implausibility, in
dyspnea, and every sort of gastrointestinal problem (Table recent years similar patterns of multisystem symptoms
2). This clinical presentation has come to be known as and multiple chemical, food, and drug intolerances have
multiple chemical sensitivity (MCS) or multiple chemical surfaced in more than a dozen countries (nine European
intolerance. Patients presenting with this peculiar combi- nations, the United States, Canada, Japan, Australia,
nation of subjective hyperosmia, multisystem symptoms, New Zealand) among demographically diverse groups—
and multiple intolerances appear to be on the increase groups having little in common, save some initial chem-
(AOEC, 1992; NIEHS, 1997; NRC, 1992). Later in this ical exposure event (Ashford and Miller, 1998). Among
chapter, epidemiological and clinical studies of this phe- these groups are radiology workers in New Zealand
nomenon will be reviewed. exposed to x-ray developer solutions containing glu-
Several things are puzzling about these patients. First, taraldehyde (Genton, 1998), EPA employees in the
the levels of chemicals they say trigger their symptoms are agency’s Washington, D.C., headquarters exposed to air-
orders of magnitude below established safety limits, lead- borne organic chemicals arising from construction,
ing some physicians to dismiss the illness on the basis that painting, and new carpeting (Hirzy and Morison, 1989),
it violates a fundamental tenet of toxicology—evidence of families in Germany exposed to pentachlorophenol used
a dose-response relationship (Waddell, 1993). Second, the to preserve log homes (Ashford et al., 1995), sheep dip-

525
526 Miller

Table 1 Triggering Exposures Reported by 80% or More of Table 2 Symptoms Commonly Reported by Chemically
Persons with Chemical Intolerances That Developed Following Intolerant Individualsa
an Exposure to Pesticides (n
37) or Indoor Air Contaminants
(n
75) Neuromuscular Cardiac
Loss of consciousness Heart pounding
Stumbling/dragging foot Rapid heart rate
New carpeting Enclosed mall
Seizures Irregular heart rate
New automobile interior Oil-based paint
Print moving/vibrating on Chest discomfort
Poorly ventilated meeting rooms Particle board
page Affective
Perfume Gas engine exhaust
Feeling off balance Feeling tense/nervous
Detergent aisle in grocery Hotel rooms
Tingling in fingers/toes Uncontrollable crying
Newspaper/printed materials Phenolic disinfectants
Double vision Feeling irritable/edgy
Fresh asphalt/tar Dry-cleaned clothes
Muscle jerking Depressed feelings
Diesel exhaust Insecticides
Fainting Thoughts of suicide
Felt-tip markers Gasoline
Numbness in fingers/toes Nerves feel like vibrating
Nail polish/remover Potpourri
Clumsiness Sudden rage
Restroom deodorizers New tires
Problems focusing eyes Loss of motivation
Fabric stores Cigar smoke
Cold or blue nails/fingers Trembling hands
Heavy traffic Cigarette smoke
Uncontrollable sleepiness Insomnia
New plastic shower curtain Incense
Head-related Airway
Hairspray Insect repellent
Head fullness/pressure Cough
Source: Miller and Mitzel, 1995. Tender face/sinuses Bronchitis
Sinus infections Asthma or wheezing
Tightness in face/scalp Postnasal drainage
Brain feels swollen Excessive mucus production
pers in Great Britain exposed to organophosphate pesti-
Ringing in ears Shortness of breath
cides (Ashford and Miller, 1998; Monk, 1996; Stephens
Headache Eye burning/irritation
et al., 1995), hospital workers in Nova Scotia exposed to Feeling groggy Susceptible to infections
building air contaminants (Ashford and Miller, 1998), Musculoskeletal Dry eyes
casino card dealers in Lake Tahoe, California, exposed Joint pain Enlarged/tender lymph
to solvents and pesticides (Cone and Sult, 1992), breast Muscle aches nodes
and temporomandibular joint implant recipients (Miller Weak legs Hoarseness
and Prihoda, 1999a,b), and Gulf War veterans exposed to Weak arms Cognitive
solvents, smoke, fuels, pesticides, and various drugs General stiffness Memory difficulties
(Bell et al., 1998, Fiedler et al., 1996b; Miller and Cramps in toes/legs Problems with spelling
Prihoda, 1999a,b). Painful trigger points Slowed responses
Gastrointestinal Problems with arithmetic
What has attracted scientific attention to this problem
Abdominal gas Problems with handwriting
is the fact that such diverse groups—people from differ-
Foul gas Difficult concentration
ent occupations, socioeconomic classes, countries and Problems digesting food Difficulty making decisions
cultures, people who do not see the same doctors, watch Abdominal swelling/ Speech difficulty
the same television shows, or read the same books—are bloating Feelings of unreality/spacey
presenting with such similar patterns of multisystem Foul burping
Other
symptoms and new-onset intolerances preceded by an Diarrhea
Feeling tired/lethargic
initial chemical exposure event. The fact that the new- Abdominal pain/cramping
Dizziness/lightheadedness
onset intolerances these patients report include medica- Constipation
tions, foods, alcohol, and caffeine—not just chemical aCategories were derived via factor analysis of symptoms reported by
inhalants—makes these worldwide observations com- 112 individuals who said they became ill following exposure to indoor air
pelling. Some scientists regard this repeating pattern as contaminants (n
75) or cholinesterase-inhibiting pesticides (n
37).
early evidence of an emerging new paradigm or theory Source: Miller and Mitzel, 1995.
of disease, one with the potential to explain a broad
spectrum of illnesses including certain cases of asthma,
migraine headaches, and depression, as well as
chronic fatigue syndrome, fibromyalgia, and Gulf War
syndrome.
Multiple Chemical Intolerance 527

A. Historical Background B. Recent Developments

In the 1950s an allergist named Theron Randolph described Over the past decade there has been an outpouring of tech-
a cosmetic saleswoman who experienced dyspnea, asthma, nical reports, concept papers, and hypotheses concerning
fatigue, irritability, depression, and intermittent loss of con- chemical intolerance. The terms “multiple chemical sensi-
sciousness whenever she smelled “man-made combustion tivity” and “environmental illness” appear on the National
products and derivatives of gas, oil, and coal” (Randolph, Library of Medicine’s bibliographical database, Medline.
1962; Randolph and Moss, 1980). Randolph coined the term While it is generally agreed that a problem exists and that
“chemical susceptibility” to describe her condition. patients are suffering, medical opinion remains polarized
Subsequently, other physicians, seeing similar problems in as to whether the condition is a unitary one or a potpourri,
their patients, allied with Randolph to found the Society for and whether it arises from chemical exposures, psychol-
Clinical Ecology (renamed in 1984 the American Academy ogical factors, or a blend of these. There is mounting
of Environmental Medicine). These clinicians adopted concern that if low-level chemical exposures were found to
Randolph’s principal diagnostic and therapeutic approach— cause this problem, the implications for environmental
trial avoidance of common chemicals and foods and, if policy, product liability, workers’ compensation, and
patients’ symptoms cleared, re-introduction of single sub- medical treatment would be staggering.
stances one at a time to determine which, if any, triggered Several nations have examined the issue, with Canada
symptoms. Over time, some clinical ecologists adopted leading the way through its 1985 Thomson Report
unorthodox diagnostic and treatment approaches, such as (Thomson, 1985) and sponsorship of clinical studies and
the administration of “neutralizing” chemical and food scientific meetings. In the United States, the phenomenon
extracts (via injection or sublingually) and sauna “detoxifi- has been explored by New Jersey, Maryland, and California
cation” to “sweat out” chemical contaminants, practices that (Ashford and Miller, 1989; Bascom, 1989; Kreutzer et al.,
drew criticism from professional medical societies 1999), various federal environmental agencies (ATSDR,
(AAAAI, 1981, 1986, 1999; ACP, 1989; AMA, 1992). 1994; Fiedler and Kipen, 1997a), the National Academy of
Professional concerns over these and other “alterna- Sciences (NRC, 1992), and professional organizations
tive” treatments continue. Recently, there has been a soft- (ACS, 1999; AOEC, 1992). While promising research
ening of positions taken against the illness as a new group strategies have emerged from these meetings (summarized
of doctors—board-certified occupational and environ- in Ashford and Miller, 1998), few comprehensive or illumi-
mental medicine physicians in universities—have begun nating studies have been funded. At this time, underlying
to study this phenomenon (ACOEM, 1999; Ashford and mechanisms remain unknown, treatments are empirical,
Miller, 1998). The American College of Occupational and and no environmentally controlled hospital facility is avail-
Environmental Medicine now “supports scientific able in the U.S. for clinical research, diagnosis, or treat-
research into the phenomenon of MCS to help explain and ment. Patients continue to suffer, while funding for serious
better describe its pathophysiologic features and define scientific study is mired by the very medical debate such
appropriate clinical interventions” (ACOEM, 1999). studies are needed to settle. Amid the confusion of opinion
There is widespread agreement that these patients report swirling around the illness, affected individuals and their
certain distinctive features—multisystem symptoms and caregivers are in need of rational, low-risk interim inter-
multiple intolerances—whether they have seen a clinical ventions with the potential to alleviate suffering and foster
ecologist or not (Davidoff and Keyl, 1996). In 1987, Mark recovery. Of equal importance, there is a need to prevent
Cullen at Yale University edited a compendium of papers exposures (e.g., to pesticides, chemicals associated with
entitled Workers with Multiple Chemical Sensitivities: An new construction) that could disable currently healthy, but
Overview, introducing occupational/environmental medi- potentially susceptible, people.
cine practitioners to the problem (Cullen, 1987). He
defined “multiple chemical sensitivity” as “an acquired
disorder characterized by recurrent symptoms, referable II. DEFINING THE PROBLEM
to multiple organ systems, occurring in response to
demonstrable exposure to many chemically unrelated At the present time, physicians and researchers cannot agree
compounds at doses far below those established in the upon a name for this condition, much less whether it is a
general population to cause harmful effects. No single single illness or a group of related or unrelated conditions
widely accepted test of physiologic function can be shown that simply share the symptom of chemical intolerance. One
to correlate with symptoms.” thing is clear. The problems these patients report do not con-
528 Miller

stitute a syndrome: By definition, a syndrome is “a group of in this chapter. The latter presumes no particular etiol-
symptoms or signs typical of a disease” (Webster’s, 1986). ogy; instead, it nourishes fresh ideas and encourages new
The symptoms these patients report are simply too hetero- discoveries. Tolerance, as used here, is the ability to with-
geneous to be collapsed into a single syndrome—perhaps a stand an insult. Chemically intolerant individuals appear
collection of syndromes, but not a single one. to have lost their prior natural or innate tolerance for a
wide variety of chemicals, foods, and drugs.
A. Sensitivity or Intolerance?

The various meanings of the term “sensitivity” contribute B. Proposed Case Definitions
to the confusion surrounding the condition(s). The word
sensitivity is used in three relatively distinct ways (Ashford Despite the differing opinions and semantic difficulties in
et al., 1995): this area, several case definitions for this phenomenon
have been proposed (summarized in Ashford and Miller,
1. The heightened responses of certain individuals to
1998), some of which may prove useful, e.g., for research,
known toxicants or allergens, i.e., the responses of
medical evaluation purposes, or compensation (AOEC,
people who are especially susceptible to toxic sub-
1992; Bartha et al., 1999; Nethercott et al., 1993; NRC,
stances like mercury or carbon monoxide or to
1992). The original Cullen case definition, used in some
allergens like housedust mites or bee venom.
early studies, unfortunately excludes “diagnosable” condi-
2. The responses of certain individuals to identifiable
tions such as asthma or depression (Cullen, 1987), when in
exposures which cannot be explained by disease
fact chemical intolerance might underlie certain cases of
mechanisms generally understood by doctors. This
asthma or depression. One operational case definition calls
category includes: (a) sick building syndrome,
for sick individuals to be removed from background chem-
involving individuals who respond adversely to one
ical, food, and drug exposures to determine whether their
or several air contaminants which may or may not
symptoms clear, and, if they do, administering single, dou-
be identifiable. Evidence for sick building syn-
ble-blinded, placebo-controlled chemical challenges to see
drome’s existence rests on the fact that affected
which, if any, trigger symptoms. The latter, patient-focused
individuals’ symptoms resolve when they leave the
approach to “defining” multiple chemical intolerance is
problem building; (b) sensitivity, such as that
considered the “gold standard” for the field and has
induced by toluene diisocyanate (TDI), which starts
emerged as a principal research recommendation from sev-
out as hypersensitivity to a specific chemical, or
eral scientific meetings (AOEC, 1992; Miller et al., 1997;
a single chemical class, but evolves into nonspecific
NRC, 1992).
hyperresponsiveness (further described in category
Bartha et al. (1999) offer six “consensus criteria” for
3 below).
multiple chemical sensitivity culled from a survey of 89
3. The heightened, extraordinary, or unusual
clinicians and researchers familiar with the illness but
responses of certain individuals to structurally
whose opinions concerning its origins differed
unrelated chemicals at exposure levels orders of
(Nethercott et al., 1993): (1) a chronic condition (2) with
magnitude below those affecting most people.
symptoms that recur reproducibly (3) in response to low
Multiple chemical sensitivity fits in this third category. levels of exposure (4) to multiple unrelated chemicals
Synonyms and related terms for MCS include environ- and (5) improve or resolve when incitants are removed
mental illness (EI), chemical intolerance, ecological ill- (6) with symptoms that occur in multiple organ systems.
ness, idiopathic environmental intolerance (IEI), The authors urge that multiple chemical intolerance be
universal allergy, and toxicant-induced loss of tolerance formally diagnosed “in addition to any other diagnosable
(TILT). A bright line needs to be drawn between MCS disorders (e.g., migraine, asthma, depression) in all
and antibody-mediated sensitivities or allergies. patients in whom the above six criteria are met and for
Allergists use the term “chemical intolerance” (not whom no single other organic disorder can account for all
“chemical sensitivity”) to distinguish this third category the signs and symptoms. . . .” Patients who experience
from classical allergies. The word “sensitivity” poses a short-lived symptoms associated with a particular odor or
problem: it implies that sensitization has occurred, when, exposure, e.g., tobacco smoke, mothballs, or paint
in fact, the loss of tolerance these individuals experience vapors, but whose symptoms stop when the exposure
might arise from something entirely different, e.g., cell ends with no recurrence or spreading to other substances
membrane disruption or gene activation. So instead of should not be labeled as having multiple chemical
“sensitization,” the term “chemical intolerance” is used intolerance.
Multiple Chemical Intolerance 529

III. PHENOMENOLOGY odorants, there are anecdotal reports of anosmic individuals

suffering from multiple chemical intolerance.
About 50–60% of chemically intolerant individuals say Various studies and surveys suggest that patients who
their illness began following a specific chemical exposure systematically avoid problem exposures find some relief
(or a series of exposures), referred to as an initiating event, (Johnson, 1996; Lax and Henneberger, 1995), but compre-
e.g., a chemical spill, repeated exposure to solvents, a pes- hensive avoidance is challenging, as well as socially isolat-
ticide application, indoor air contaminants associated with ing. Common odors involving low-level volatile organic
new construction, combustion products (Fiedler et al., chemicals (VOC) in the parts per billion or parts per trillion
1996a; Miller, 1994). Only a subset of those exposed range are near-ubiquitous. Making matters worse,
appear to develop chronic symptoms and intolerances. physicians, and even the patients themselves, may fail to
What makes some individuals more susceptible remains a discern symptom-exposure relationships (Ashford and
mystery. Initially, patients may describe “flu-like” symp- Miller, 1998). Several factors may contribute to this. For
toms that fail to resolve, or feeling as though they are in a example, habituation can occur with chronic or repeated
“perpetual fog.” Next to develop are multisystem symptoms exposure to the same substance(s), e.g., volatile organic
that seem to wax and wane unpredictably, followed by a chemicals in a sick office building. Second, apposition, i.e.,
dawning awareness of specific intolerances, frequently overlapping symptoms resulting from various exposures
involving alcoholic beverages or medications at first. Over (chemicals, foods, drugs), may hide or “mask” the effects
time, these intolerances spread to include a wide variety of of particular exposures. According to this scenario, the
everyday exposures—chemical odors (low levels of volatile intolerant individual who applies hairspray and fragrances
compounds), foods, drugs, caffeine, alcoholic beverages, in the morning, cooks breakfast on a gas stove, and drives
and skin contactants. Food intolerances may appear but not through heavy traffic to a sick office building may experi-
be recognized as such. Instead, patients may complain of ence near-continuous symptoms (highly masked) and fail to
digestive difficulties, feeling ill after meals, or becoming recognize any single exposure as causal (Miller 1996,
irritable if a meal is missed or delayed. 1997). In effect, background symptom “noise” might hide
These intolerances may begin within weeks of an acute, any particular “signal.”
high-level exposure or, as in the case of a sick office build- “Withdrawal” symptoms reportedly develop when
ing, emerge insidiously over months or years. Symptoms patients avoid problem exposures for several days, e.g.,
may be triggered via any exposure route—inhalation, over a weekend or while on vacation. Such avoidance can
ingestion, injection (e.g., drugs), or skin or mucosal con- be inadvertent or deliberate, e.g., a physician-recom-
tact. Some patients report that breathing through their mended trial avoidance period (see Sec. V. E). Later, with
mouths instead of their noses, e.g., around traffic exhaust or reexposure, as on a Monday morning after a weekend
people wearing fragrances, mitigates their symptoms some- away from work, symptoms may return “with a
what. Particular odors or exposures (see Table 1)—whether vengeance.” Some chemically intolerant individuals quit
fragrances, chemicals outgassing from new furnishings or their jobs so as to avoid coworkers’ fragrances, carbonless
carpeting, traffic exhaust, cleaning agents, etc.—may trig- copy paper, cleaning agents, etc. Others switch
ger different constellations of symptoms in different indi- employers, occupations, and residences in search of safer
viduals. An individual patient often reports different surroundings.
responses with different exposures. Symptom intensity may In science, anomalies expose the limitations of existing
range from mild (e.g., nasal congestion, nausea, or slight paradigms and drive the search for new ones. In the late
headache) to severe (e.g., mental confusion, depression or 1800s, physicians observed that certain illnesses seemed to
seizures) (Table 2). There is consistency, however: a partic- spread from sick, feverish individuals to their families and
ular exposure, e.g., diesel exhaust or a certain fragrance, in neighbors. These anomalous observations paved the way for
a particular person tends to elicit a characteristic constella- the germ theory of disease. This germ theory, so obvious to
tion of symptoms, a so-called “signature response.” us today, enabled scientists and the public to grasp for the
Responses may occur at below-olfactory-threshold concen- first time the origins of dozens of seemingly unrelated ill-
trations, with symptoms developing within seconds to nesses affecting every organ system. Today, we are witness-
hours after a triggering exposure and persisting minutes to ing another anomaly—a repeating pattern of illness
days. Hyperresponsiveness to physical stimuli, including appearing in groups of people, including Gulf War veterans,
light, noise, and touch, is commonly reported (Miller and from more than a dozen countries following chemical expo-
Prihoda, 1999 a,b). Patients may wear sunglasses indoors sures. What unites the Gulf War veterans and these civilian
or dim the room to keep bright light from reaching their groups is their common experience of an initiating chemical
eyes. Although some patients report hypersensitivity to exposure event followed by newly acquired intolerances and
530 Miller

multisystem symptoms. These observations provide com- one at a time, to determine the source of their symptoms.
pelling scientific evidence for a shared, underlying disease One limitation of the TILT theory is that it may not explain
mechanism—one involving a fundamental breakdown in every case of chemical intolerance: not every patient is able
natural tolerance. This two-stage process—an initial chemi- to identify an initiating event or events. An initiating event,
cal exposure (initiation) leading to newly acquired intoler- e.g., a pesticide application, may go unnoticed.
ances, with symptoms subsequently triggered by multiple Alternatively, genetic, psychological, nutritional and other
common exposures (triggering)—has been referred to as factors may underlie their intolerances.
toxicant-induced loss of tolerance, or “TILT” (Fig. 1).
It does not appear to matter which exposure causes the IV. PREVALENCE
breakdown in tolerance—be it pesticides, solvents, indoor
air contaminants, smoke from oil well fires, or medica- Several large surveys suggest that 15–30% of the U.S. pop-
tions. It is the aftermath of these exposures, the new-onset ulation consider themselves “especially” or “unusually”
intolerances to various substances, that appears to perpetu- sensitive to certain chemical odors, while approximately
ate the symptoms. Four observations suggest that toxicant- 2–6% claim a physician’s diagnosis of “multiple chemical
induced loss of tolerance might be a new theory of sensitivity,” “environmental illness,” or significant daily
chemically induced disease: impairment from chemical exposures (Table 3) (Kreutzer
1. The appearance of the same pattern of symptoms and et al., 1999; Meggs et al., 1996; Voorhees, 1998). In the
new-onset odor and other intolerances in demo- largest of these studies, a statewide randomized telephone
graphically diverse groups worldwide following interview survey conducted by the California Department of
well-defined exposures to pesticides, solvents, Health Services, 15.9% of participants said they were
indoor air contaminants, etc. “allergic or unusually sensitive to everyday chemicals,”
2. The fact that these groups’ new-onset intolerances 11.9% identified two or more chemicals that made them
involve not only chemical inhalants, but also var- sick, and 6.3% reported doctor-diagnosed “environmental
ious foods, medications, caffeine, alcoholic bev- illness” or “multiple chemical sensitivity” (Kreutzer et al.,
erages, and skin contactants. These observations 1999). Female gender and Hispanic ethnicity were associ-
in particular constitute what Kuhn (1970) called ated with greater self-reporting of sensitivity (adjusted odds
a “compelling” or “critical” anomaly. Just as ratios of 1.63 and 1.82, respectively). Neither self-reported
fever is the hallmark symptom for infection, a chemical sensitivity nor doctor-diagnosed multiple chemi-
signal that sends doctors down certain diagnostic cal sensitivity was associated with employment, educational
pathways, new chemical, food, and drug intoler- level, marital status, geographic location, or income. The
ances are the hallmark symptom for TILT. similar rates seen in California, New Mexico, and North
3. Recent animal models replicating key features of Carolina suggest that multiple chemical intolerance could
TILT (see Sec. VI). be among the most prevalent, if not the most prevalent,
4. The striking parallels between this phenomenon and chemically caused medical condition(s) in the United
addiction, suggesting shared neural mechanisms States. The California study concluded that “surprising
likely involving multiple neurotransmitter path- numbers” of people believe that common chemical expo-
ways (see Sec. VI). sures make them sick and that “the homogeneity of
responses across race-ethnicity, geography, education, and
TILT has the potential to explain certain cases of asthma, marital status is compatible with a physiological response
migraine headaches, and depression, as well as chronic or with widespread societal apprehensions in regard to
fatigue, fibromyalgia, and Gulf War syndrome (Fig. 2). But chemical exposure” (Kreutzer et al., 1999). While the media
both stages of TILT—initiation and triggering—are in need have fanned public fears of environmental exposures, this
of testing. Some argue that TILT’s second stage, triggering, by no means proves the problem is psychogenic. Only care-
should be studied first and that to accomplish this, a special ful studies can settle questions concerning etiology.
scientific “apparatus” needs to be built—an environmentally More women than men have participated in clinical
controlled in-patient hospital unit (environmental medical studies of chemical intolerance to date (4:1 female:
unit) in which patients can reside for a week or longer, male ratio), with an average age in the fourth decade and
allowing them reach a “clean” exposure baseline (Miller educational level of at least 2 years of college (Fiedler and
et al, 1997; NRC, 1992). Assuming this occurs and patients’ Kipen, 1997b). In contrast, the California statewide survey
exposure-related symptoms resolve, subjects could then be cited above found a 5:3 female:male ratio in a random gen-
exposed to various potential triggers, including caffeine, eral population sample. Among military and industrial
gasoline, perfume, foods, medications, and tobacco smoke, populations, more males report the problem, likely reflect-
Multiple Chemical Intolerance 531

Figure 1 Phenomenology of toxicant-induced loss of tolerance (TILT). Illness appears to develop in two stages: (1) initiation, i.e., loss
of prior, natural tolerance resulting from an acute or chronic exposure (pesticides, solvents, indoor air contaminants, etc.), followed by
(2) triggering of symptoms by small quantities of previously tolerated chemicals (traffic exhaust, fragrances), foods, drugs, and food/drug
combinations (alcohol, caffeine). The physician sees only the tip of the iceberg—the patient’s symptoms—and formulates a diagnosis
based on them (e.g., asthma, chronic fatigue, migraine headaches). Masking hides the relationship between symptoms and triggers. The
initial exposure event causing breakdown in tolerance may also go unnoticed. (©UTHSCSA 1996.)

Figure 2 Conditions that may result from toxicant-induced loss of tolerance. Illnesses like depression, migraine, arthritis, and chronic
fatigue may have various underlying mechanisms, one of which might be TILT.
532 Miller

Table 3 Frequency of Self-Reported Chemical Intolerance from Several Large Surveys

Those reporting physician-
Those considering themselves diagnosed multiple chemical
especially or unusually sensitive intolerance or daily symptoms
Population (Ref.) Number of people studied to certain chemicals (%) triggered by chemicals(%)
EPA office workers 3948 31 Not evaluated
(Wallace et al., 1993)
Rural North Caroliniansa 1027 33 3.9
(Meggs et al., 1996)
California residentsa 4046 15.9 6.3
(Kruetzer et al., 1999)
New Mexico residentsa 1814 17 1.9
(Voorhees, 1998)
a Randomly sampled.

ing underlying gender ratios (Miller and Prihoda, 1999a,b; logical grounds. The former currently account for nearly
Simon et al., 1990;). In sick buildings, the condition is half of all pesticides used in the United States. These rela-
commonly reported by college-educated white females in tively new neurotoxic compounds, developed for agricul-
the 30- to 50-year age range and of middle- to upper-mid- tural use, also worked well in homes, schools, and office
dle-class socioeconomic status (Ashford and Miller, 1998). buildings, controlling pests with few “callbacks” for exter-
Why more women then men report the problem and why minators. Today, thousands of different synthetic organic
more chemically intolerant patients work in office build- chemicals are released into the air from fuels, paints, cloth-
ings and service industries than in heavy industry remains ing, and consumer products of every description. Ninety
unknown (Black et al., 1990; Lax and Henneberger, 1995; percent of the U.S. population spends more than 90% of
Miller and Mitzel, 1995;). The gender disparity may reflect the day indoors, exposed to complex mixtures of volatile
women’s greater willingness to report symptoms, some- organic chemicals (VOCs) emitted by new carpet, pesti-
thing unique about indoor air pollutants that women may cides, cleaners, fragrances, particle board, and furnishings.
be less able to escape, e.g., as secretaries or realtors, or These airborne chemicals tend to accumulate in modern,
gender-based biological response differences. The paradox sealed indoor environments. Even the air inside vehicles,
that more multiple chemical intolerance cases arise from especially new ones, contains myriad VOCs emitted by
the service sector than heavy industry may be due to (1) plastic interiors, rubber floor mats, and other components.
“the healthy worker” selection effect, i.e., individuals VOCs are entrained from traffic exhaust and freshly tarred
bothered by chemical exposures would tend to choose non- roads. Thus, the vast majority of the population is exposed
chemical jobs, (2) the fact that women, who may be bio- near-continuously to thousands of biologically “foreign”
logically more vulnerable, are less apt to work in heavy VOCs, albeit at low levels (parts per billion or trillion).
industry, mining, construction, etc., or (3) some unknown Add to this the oil embargo of the 1970s, when U.S.
but insidious effect of indoor air pollution mixtures. home owners received tax credits for installing insulation,
Over the past decade, occupational medicine physicians caulking cracks, etc., thus sealing in indoor air contami-
have witnessed a surge in the numbers of these patients in nants. Responding to the energy crunch, the American
their practices (Ashford and Miller, 1998)—another puz- Society of Heating, Refrigeration and Air Conditioning
zle. While increased media coverage of environmental Engineers (ASHRAE) further reduced U.S. recommenda-
exposures has fueled our awareness, at the same time tions for fresh outside air entering commercial buildings,
major increases have occurred in the variety and quantity schools, public spaces, etc., which were 30 cubic feet
of chemicals people encounter each day, all within a few per minute (cfm) per occupant in 1900, down to 5 cfm per
generations. For example, U.S. production and use of occupant—a sixfold decrease in fresh air in commercial
synthetic organic chemicals have risen exponentially since and public buildings.* Home ventilation systems generally
World War II, rising from a billion pounds produced annu-
ally in the early 1940s to 400 billion pounds in the 1980s
(U.S. International Trade Commission). And the nature of * In recent years, ASHRAE fresh air requirements for public and
the chemicals has changed. For example, cholinesterase- commercial spaces have been raised to a minimum of 15 cfm per
inhibiting pesticides (organophosphates and carbamates) occupant, 20 cfm in offices, because of health complaints associ-
have supplanted chlorinated pesticides like DDT, on eco- ated with the 5 cfm recommendation (ASHRAE 1999).
Multiple Chemical Intolerance 533

bring in no fresh outside air; occupants breathe only what be affected (especially early in the illness), multisystem
little leaks through cracks, crevices, open doors, or win- involvement is the norm. The probability that the
dows. With more tightly sealed, “energy-efficient” struc- practitioner is dealing with multiple chemical intolerance
tures and less fresh air brought in from outside, VOC rises if (1) an identifiable chemical exposure, e.g., to
contaminants inside homes and workplaces have crept to solvents, pesticides, combustion products, or volatile
unprecedented levels, often orders of magnitude greater organic chemicals from a sick building, remodeling, or new
than outdoor levels. As a consequence, sick building syn- construction, clearly preceded onset of the symptoms; (2) a
drome has spread across the United States, most notori- major change in the patient’s health occurred, ideally
ously affecting the Environmental Protection Agency’s documented by increased health care utilization and/or
(EPA) Washington, D.C., headquarters, where several hun- absenteeism; (3) clinical signs or abnormal laboratory tests
dred individuals fell ill following remodeling and new car- appear postexposure, e.g., increased liver function tests, a
pet installation (Hirzy and Morison, 1989). Several dozen depressed white blood cell count, or decreased
EPA employees subsequently developed multiple chemical cholinesterase level; (4) new-onset depression, asthma,
intolerances. More than a decade later, many of these indi- severe headaches, etc., appear in the absence of other clear
viduals remain disabled (Ashford and Miller, 1998). causes; (5) others who shared the same exposure event
became ill, especially but not necessarily with similar
symptoms; or (6) previously tolerated chemical exposures
V. CLINICAL RECOGNITION AND now provoke symptoms. (7) If the patient reports newly
EVALUATION acquired intolerances for medications, alcohol, caffeine, or
foods (or feeling ill after meals), and (8) if clinical labora-
Physicians may be reluctant to diagnose multiple chemical tory abnormalities (e.g., pulmonary function tests) improve
intolerance even when it provides the most parsimonious with trial avoidance and/or worsen with reexposure, then
description for a patient’s problems. The diagnosis of multi- the likelihood increases further. No one of these features
ple chemical sensitivity is frequently challenged by work- “proves” the diagnosis, but the more the patient manifests,
ers’ compensation boards, employers, and others.* Some the more the practitioner should suspect toxicant-induced
doctors opt to apply “piecemeal” but recognized and com- loss of tolerance.
pensable diagnostic labels such as asthma, toxic Physicians need to discuss with these patients the polar
encephalopathy, or migraine headache. There are no symp- opinions surrounding the illness’s origins, steering them to
toms, clinical signs, or laboratory tests that are pathogno- sympathetic specialists and helping them explore treat-
monic for multiple chemical intolerance. As for other ment options, including psychological therapies and envi-
medical conditions for which no objective diagnostic tests ronmental interventions, while warning that all therapies
are available, e.g., depression, schizophrenia, and migraine currently are uproven, the underlying mechanism remains
headaches, health care providers must rely on careful history a mystery, and no test(s) are diagnostic. Much research
taking and observation over time to make the diagnosis. will be needed before these patients’ most pressing ques-
Circumstances permitting, patients’ intolerances should be tions can be answered with any degree of scientific
assessed through trial avoidance of suspected substances certainty. Patience on the parts of both patient and practi-
and, if improvement occurs, judicious reintroduction of sin- tioner is essential. “The evaluation of a patient presenting
gle exposures under the supervision of a knowledgeable with MCS may take several hours and it is necessary to
practitioner. allot sufficient time, even if inadequately reimbursed”
Where symptoms and circumstances warrant, doctors (Sparks et al., 1994b). It is not unusual for these individu-
should discuss with patients the possibility of multiple als to consult 10 or more practitioners before the problem
chemical intolerance. Commonly reported complaints is recognized (Miller and Mitzel, 1995). In one study,
include hyperosmia, fatigue, memory and concentration chemically intolerant patients averaged 23 health care
difficulties, mood changes, and multisystem health provider visits per year (Buchwald and Garrity, 1994).
problems (Table 2). Although only one organ system may Physicians can easily underestimate the illness’s multi-
system impact. Chemically intolerant patients tend to
migrate from specialist to specialist, accumulating diagnos-
*“Toxicant-induced loss of tolerance,” which describes the break- tic labels like toxic encephalopathy, chronic fatigue syn-
down in tolerance resulting from exposure—a phenomenon that drome, psychosomatic illness, migraines, and fibromyalgia
has been widely witnessed by reputable scientists and reported on while remaining oblivious to the underlying dynamic.
in numerous, peer-reviewed medical articles—has not been simi- Hyperosmia, a hallmark symptom, may be overlooked in
larly scrutinized or challenged. patients reporting a profusion of symptoms. Physicians’
534 Miller

professional opinions, even when science to support them is reluctant to deal with chemical exposures. Chemically
lacking, greatly influence insurors’ determinations, com- intolerant patients “fall in the crack” between these two
pensation boards, and disability reviewers, as well as specialties.
employers, friends, and family. Consequently, patients tend
to avoid doctors who seem skeptical or ill-informed. B. Symptoms and Intolerances

A. Exposure History Patients typically report multisystem symptoms, with

fatigue being most common (Table 2). Symptoms often
Busy doctors seldom take occupational or environmental mimic chronic fatigue syndrome or fibromyalgia, diag-
histories even when circumstances warrant (IOM, 1995). noses many patients eventually acquire (Ashford and
The process is time-consuming, and physicians often feel Miller, 1998; Buchwald and Garrity, 1994; Chester and
ill-equipped to interpret the information gathered. But Levine, 1994; Miller and Mitzel, 1995). Mood changes,
here the value of a careful exposure history cannot be including irritability, anxiety, and depression, are
overemphasized. Standard forms for collecting basic commonly reported. Exposure-triggered memory and con-
occupational and environmental exposure information are centration difficulties have led some patients to abandon
available and should be part of every patient’s chart (IOM, cognitively demanding careers. Affected groups report
1995). However, these alone are not sufficient for evaluat- strikingly similar symptoms, whether the “initiating” event
ing chemically intolerant individuals, since many involved pesticides or construction-related indoor air
common exposures (e.g., home remodeling, pesticide use) contaminants (Miller and Mitzel, 1995). Central nervous
that bother these patients may be omitted from these system symptoms predominate. Gastrointestinal (e.g.,
forms. One approach is to have patients construct their problems digesting food), respiratory (e.g., shortness of
own symptom/exposure time lines (one line per year), breath or being unable to get enough air), and muscu-
with symptoms and medical problems recorded along loskeletal (e.g., muscle/joint pain) problems are commonly
the top of the line and life events (e.g., changes in jobs, reported (Miller and Mitzel, 1995).
residences, military service, surgeries, pregnancies, Besides chemical intolerances, almost invariably these
remodeling, pesticide use, etc.) along the bottom. A patients report responding adversely to various foods,
clear, concise chronology, preferably in this format, may drugs, alcoholic beverages, or caffeine. Ninety-seven per-
ferret out contributory exposures. cent of 112 self-reported chemically intolerant individuals
A standardized and validated screening questionnaire, in one study reported significant food intolerances, causing
the Quick Environmental Exposure and Sensitivity them to avoid their problem foods or follow elimination
Inventory (QEESI), has been described in the medical lit- diets (Miller and Mitzel, 1995). Food intolerances often
erature and is available for evaluating these patients’ symp- involve frequently eaten items and those commonly craved,
toms and chemical/odor intolerances, monitoring their e.g., bread (wheat), corn chips, chocolate, caffeinated bev-
clinical course, and measuring their treatment response erages, milk, etc. Patients who experience an immediate,
(Miller and Prihoda, 1999a). This 50-item inventory per- sharp reaction after eating a particular food tend to avoid
mits patients to rate their symptoms and intolerances that food, particularly if the food is unusual (e.g., avocado
before and after an exposure event. It can also help affected or cashews) and eaten only occasionally (Randolph, 1956).
individuals identify potential home and workplace triggers On the other hand, foods eaten frequently (twice weekly or
and determine whether they may be more susceptible to more) may precipitate fatigue, headaches, mood problems,
various alcoholic beverages, caffeine, foods, and drugs. If digestive difficulties, etc., but go unrecognized, due to
intolerances to any of these preceded the alleged initiating masking. Because patients may experience a slight
exposure event, it may help clarify why this person, rather “pickup” shortly after these foods are eaten, they may actu-
than others who were similarly exposed, fell ill. ally crave these foods (e.g., “chocoholics”). Food-intoler-
Whenever possible, material safety data sheets ant patients may insist upon having their meals on time. If
(MSDSs) should be obtained to clarify the nature of the they miss a meal or eat late, they may report feeling tired,
patients’ exposures. Occupational medicine physicians, jittery, achy, etc. Some snack between meals, eat just before
toxicologists, and industrial hygienists may help with bedtime (to avoid nighttime withdrawal symptoms or
collecting and evaluating exposure data. Occupational awakening), stash food by the bedside or in the car, or carry
medicine doctors routinely take detailed exposure histo- large cups of coffee or tea wherever they go, “titrating”
ries, but most do not delve into patients’ food intoler- themselves throughout the day—unconscious ploys to
ances. On the other hand, allergists who are trained to ward off food-withdrawal symptoms. Randolph called this
evaluate food intolerances using elimination diets may be phenomenon “food addiction.”
Multiple Chemical Intolerance 535

Chemically intolerant individuals may report acute symp- C. Past Medical History and Associated Conditions
toms with minimal caffeine intake and/or caffeine with-
drawal symptoms lasting a week or longer, including Whenever possible, past medical records should be exam-
headaches, irritability, abdominal cramps, heart pounding, ined, even if voluminous and difficult to obtain. A past his-
insomnia, or fatigue. Blinded, crossover studies have demon- tory of unexplained illness(es) is common in this patient
strated that certain people who consume as little as one cup population. Aerospace workers who developed multiple
of regular coffee per day (about 100 mg of caffeine) reliably chemical intolerances when a new composite plastic was
develop caffeine withdrawal symptoms when they stop introduced into their workplace averaged 6.2 unexplained
(Silverman et al., 1992). Even a few sips of decaffeinated physical symptoms preceding the change in process versus
coffee (about 10 mg of caffeine per cup) are said to precipi- only 2.9 unexplained symptoms in unaffected coworker
tate symptoms in some chemically intolerant patients. controls (Simon et al., 1990). Fifty-four percent of the ill
Initially, these individuals may consume copious quantities workers had histories of anxiety or depression that pre-
of caffeine in an effort to stave off withdrawal symptoms. ceded their exposure, compared with 4% of controls. Other
Caffeine intolerance can be confirmed by stopping all investigators have found that past psychiatric history does
xanthines (coffee, tea, cola, chocolate, caffeinated soft not explain the illness (Fiedler et al., 1992). Even if some
drinks) for 1–2 weeks to determine whether symptoms chemically intolerant individuals have histories of depres-
improve and, if improvement occurs, reintroducing caffeine sion that predate their “initial exposure event,” the question
while observing whether symptoms return. Withdrawal remains whether depression causes chemical intolerance,
symptoms may persist 7–10 days after all xanthines are whether depressed individuals are more susceptible to
stopped. Because sudden xanthine cessation can precipitate chemical exposures (i.e., have vulnerable neurochemistry),
severe withdrawal symptoms in patients prone to debilitating or whether the earlier depression may have been due to
headaches, depression, etc., gradual tapering may be prefer- prior, unidentified intolerances (Davidoff and Fogarty,
able in such cases, albeit withdrawal may be protracted. 1994). Medical and psychiatric diagnoses reported more
Alcohol intolerance may be the first intolerance a patient frequently by chemically intolerant college students
notices (Miller and Prihoda, 1999a,b), perhaps because of include nasal allergies, hives, breast cysts, premenstrual
ethanol’s rapid absorption rate and the fact that most people disorder, and childhood hyperactivity (Bell et al., 1996),
tend to use it intermittently, often on an empty stomach. As and among community-based chemically intolerant mid-
little as one drink may precipitate acute symptoms or may dle-aged individuals, rhinitis, migraines, irritable bowel,
be followed by several days’ withdrawal symptoms (hang- ovarian cysts, menstrual dysfunction, depression, anxiety,
over). Patients rarely tell physicians they are alcohol intol- and panic disorder (Bell et al., 1995). Another community
erant unless the practitioner asks. Smokers may report, if sample found chemically intolerant individuals more likely
queried, that smoking one more cigarette than usual or bor- to seek medical help for heart problems, bronchitis,
rowing someone else’s stronger brand precipitates acute asthma, and pneumonia, and to report parental heart
symptoms such as headache, lightheadedness, shortness of disease, asthma, and diabetes (Baldwin and Bell, 1998).
breath, gagging, coughing, nervousness, or nausea. Some Black et al. (1999) reported that chemically intolerant
individuals quit or switch to lighter brands. patients versus controls reported more first-degree rela-
Patients commonly report adverse reactions to med- tives with major depression, alcoholism, panic disorder,
ications, for example, flu-like symptoms persisting days obsessive-compulsive disorder, and antisocial personality
after methacholine challenge; becoming extremely irrita- disorder and who had made more suicide attempts and
ble and eating ravenously after a steroid injection; chest received more psychiatric treatment.
tightness and chills following radiographic contrast dye
injection; or panic attacks or floating feelings with D. Physical Examination and Laboratory Evaluation
antidepressants. Elevated liver function tests with some
drugs, e.g., piroxicam, may occur. Previously well-toler- Abundant anecdotal evidence suggests that chemically
ated drugs may no longer be tolerated, e.g., over-the- intolerant individuals improve when they identify and
counter decongestants. Skin and/or systemic symptoms learn to avoid exposures that trigger their symptoms.
may occur with skin or mucosal contactants, e.g., Various federally sponsored consensus groups have rec-
adhesive tape, topical creams or ointments, jewelry, ommended research/diagnostic studies using an environ-
soaps, shampoos, plastic mouth appliances, toothpaste, mentally controlled, inpatient hospital unit analogous to a
contact lenses, various fabrics (wool, polyester), drug detoxification unit, in which patients could be taken
condoms, spermicides, cosmetics, deodorants, laundry to a “clean” baseline, allowing exposure-related symptoms
soaps, fabric softeners, and chlorinated pool water. to resolve (Fig. 3) (AOEC, 1992; Miller et al., 1997; NRC,
536 Miller

1992). Subjects could then be exposed to various potential physician, needs to oversee the entire evaluation and treat-
triggers, including caffeine, gasoline, perfume, individual ment, integrating specialists’ advice as appropriate.
foods, medications, and tobacco smoke, one at a time, to To date, no consistently abnormal laboratory findings
determine what is causing their symptoms, but at this time have been demonstrated in these patients. Various studies
no such facility is available in the United States. have reported abnormalities in T- and B-lymphocyte
Although a comprehensive physical examination is counts; helper/suppresser T-cell ratios; immunoglobulin
essential, findings frequently are unremarkable. Baseline levels; autoimmune antibodies (e.g., anti-nuclear, anti-
laboratory tests such as a complete blood count and chemi- smooth muscle, anti-thyroid, anti-parietal cell, etc.); acti-
stry profile can be helpful, as well as tests suggested by vated T lymphocytes (TA1 or CD26); quantitative EEGs;
history or physical findings, such as thyroid function tests, evoked potentials; SPECT and other brain scans (Heuser
pulmonary function tests, peak flow monitoring over time, and Mena, 1997; Hu et al., 1999; Mayberg, 1994; Rossi
tests for collagen-vascular disease, and neuropsychologi- et al., 1999; Waxman, 2000); vitamin, mineral, amino acid,
cal evaluations in some patients (Weaver, 1996). Blood and detoxification enzyme levels; and blood or tissue lev-
tests for environmental chemicals should be used only if els of pesticides, solvents, and other chemicals. Common
there is reason to suspect specific exposures, and these flaws in studies conducted to date include failures to
substances can reasonably be expected to persist in tissues (1) define the study population (no case definition used),
(e.g., a chlorinated pesticide or recent organophosphate (2) compare cases with age- and sex-matched controls,
pesticide exposure, but not most solvent or indoor air (3) blind specimens, and (4) document the test method’s
volatile organic chemical exposures). Referrals to special- accuracy and reproducibility. Some illness proponents
ists are often necessary, given the multisystem nature of claim that different immunological abnormalities occur in
the illness and the need to rule out contributing or coexist- different patients. However, if enough tests are done, sta-
ing conditions, such as an autoimmune disease, endocrine tistically a certain number can be expected to be abnormal
disorder, demyelinating disease, brain tumor, etc. (e.g., 1 in 20), a fact frequently forgotten. Mitchell et al.
(Moorhead and Suruda, 2000). While specialists’ evalua- (2000) recommend that “[p]atients should be informed that
tions can be reassuring for patients and referring physi- subtle differences in individual immunological parameters,
cians, unnecessary invasive testing and polypharmacy are especially if observed in only one laboratory, are common
potential pitfalls. One physician, preferably a primary care and do not necessarily indicate the presence of a systemic

Figure 3 Use of an environmental medical unit (EMU) in the evaluation of health effects from low level chemical exposures. The figure
illustrates stages in the evaluation of a patient in an EMU. At left, prior to entering the EMU, a patient is experiencing overlapping symp-
toms in response to everyday exposures and is unable to discern the effects of any particular exposure. Background symptom “noise” is
high, and, to the patient, symptoms seem to wax and wane unpredictably over time. (1) With entry into the EMU and the avoidance of
all chemical, food, and drug triggers simultaneously, remission of symptoms should begin. During the first few days of “withdrawal,” irri-
tability, headaches, and depression are expected complaints. Within a week, the individual should be at a clean baseline and ready for
challenge testing. (2) Following a chemical, food, or drug challenge (arrows), the patient should report a specific constellation of symp-
toms. (3) When the challenge ends, patients should gradually return to baseline and be free of symptoms. If the individual is re-challenged
with the same substance 4 to 7 days after the first challenge, the same constellation of symptoms should occur. Challenges involving unre-
lated chemicals or foods may take place in the interim.
Multiple Chemical Intolerance 537

disorder, unless there are additional clinical correlates and provided nutrition is maintained. If multiple food intolerances
the results are verified.” are found, achieving adequate nutrition may be difficult.
Long-standing digestive difficulties and/or use of restrictive
E. Trial of Avoidance diets can result in nutrient deficiencies.
It may be helpful to have patients maintain a daily log
There are anecdotal reports of patients diagnosed with this or diary (one page per day), recording each symptom
condition early in its course who avoided further exposure (scored on a 0–10 intensity scale) and listing all inhalant
and apparently recovered (Hileman, 1991), suggesting that and ingestant exposures and the times at which exposures
timely recognition and avoidance might reverse the illness and symptoms occurred. Review of this log may help
and prevent disability. Our current inability to treat the con- reveal problem exposures. It is important to recognize that
dition once entrenched underscores the importance of early patients who smoke, use alcohol or caffeine on a regular
intervention. Physicians need to recognize patients who basis, or have ongoing workplace or other chemical expo-
may be in the initiation phase of the illness and who may sures may be unable to link their symptoms with specific
have ongoing exposures to pesticides, remodeling, solvents, exposures due to masking.
etc. These individuals may benefit from a period of trial
avoidance to determine whether they improve, followed by
judicious, medically supervised reexposure to determine VI. PROPOSED MECHANISMS
whether their symptoms recur. Unfortunately, at this time
most patients are diagnosed long after their illness begins. Under the assumption that most of the symptoms described
Almost uniformly, patients report that identifying and in this chapter stem from a unitary disease process, then the
avoiding chemical and food triggers benefit them most unifying dynamic remains a mystery. Even if one assumes
(Lax and Henneberger, 1995). While conceptually simple, these symptoms arise from multiple causes or different dis-
avoiding potential incitants is no simple task, particularly ease processes, their underlying physiological bases are dif-
for patients with cognitive difficulties. As a rule, patients ficult to imagine. Some physicians and researchers view
need help identifying and minimizing exposures to volatile these patients’ symptoms as psychogenic phenomena,
organic chemicals released by fragrances, cleaners, and resembling depression, somatoform disorder, or posttrau-
other products. Simply telling them to avoid exposures that matic stress disorder. Others see them as chemically
trigger their symptoms is not sufficient. Initially, they may induced and offer various physiological explanations
be loathe to relinquish their favorite fragrances, soaps, can- (Ashford and Miller, 1998; summarized in Sorg, 2000). To
dles, gas stoves, foods, etc., even on a trial basis. date, surprisingly little research has been done in this area
Because avoidance of alcoholic beverages, nicotine, caf- when one considers the fact that 2–6% of the U.S. popula-
feine, medications, and problem foods can precipitate tion think they suffer from it (see Sec. IV).
severe withdrawal symptoms in susceptible individuals, Early mechanistic studies focused on immune dysregu-
e.g., severe headaches, disorientation, depression, and lation as a potential explanation for these patients’ prob-
malaise, such regimens should not be attempted on an out- lems. These studies suffered from major methodological
patient basis unless supervised by a knowledgeable medical limitations, and no consistent immunological abnormality
practitioner. Patients with histories of severe asthma, was found (reviewed in Mitchell et al., 2000). To assess
headaches, depression, or other medical conditions (espe- inter- and intra-laboratory reliability, Mitchell et al. (2000)
cially if emergency management was ever required) should sent split blood specimens to six laboratories. Good agree-
be hospitalized in an environmentally controlled medical ment was found for T-cell subsets, but not for more diffi-
unit during the initial stages of avoidance. cult tests, e.g., immune activation markers. Subsequently,
The vast majority of patients find that avoiding problem patient and control samples were sent to two validated lab-
foods makes them feel better (Johnson, 1996; LeRoy et al., oratories for T-cell subset and antibody analyses. Despite
1996; Miller and Mitzel, 1995). Various elimination diets these validated laboratories and tests, most of the statisti-
have been used with these patients, most often the rotary cally significant differences between the chemically intol-
diversified elimination diet, in which no food or food group is erant subjects and healthy controls were observed in only
consumed more than once every 4 days (Rinkel, 1944; Rinkel one of the two laboratories used, the only exception being
et al., 1951). Popular versions of this diet have appeared that CD4 lymphocyte percentages were higher and
(Mandell and Scanlon, 1979). Adhering to restrictive diets is CD8 percentages lower in the chemically intolerant
difficult, and research is needed to determine their efficacy. group. Data analysis is ongoing.
However, modifying patients’ timing of food intake in order A key question is why these people are bothered by
to uncover possible food intolerances is relatively innocuous, low-level exposures that the vast majority tolerate. Doty
538 Miller

et al. (1988) asked whether chemically intolerant individu- Features of this animal model that fit the clinical picture
als might have lower olfactory thresholds. Eighteen (Sorg, 2000) include the progressive increase in drug/chem-
patients and 18 controls received graded, low-level con- ical response over time and with additional exposures, the
centrations of rose oil (phenyl ethyl alcohol) and methyl apparent permanence of the sensitivity, the absence of
ethyl ketone via olfactometer. There were no significant symptoms in the absence of chemical/drug exposure, the
differences in olfactory thresholds; however, during stimu- greater sensitivity of females versus males, and the spread-
lus presentations, the patients had significantly higher ing of sensitivity to chemically dissimilar substances.
nasal resistance and respiration rates. Fiedler et al. (1996a) Sorg (1996) hypothesized that sensitization to
found no difference in odor identification ability between formaldehyde and other chemicals might affect the same
chemically intolerant patients and controls. brain pathways involved in cocaine sensitization, e.g., the
The heightened nasal resistance seen by Doty et al. sug- mesolimbic dopamine system. Rats inhaling 1ppm of
gests an inflammatory process, a mechanism proposed by formaldehyde 1 hour daily for 20 days, as compared to
others (Bascom, 1991; Meggs, 1994; Meggs and Cleveland, sham-exposed rats, later showed heightened responses to
1993). Fiberoptic rhinolaryngoscopic examination of 10 cocaine. This sensitivity persisted 4–6 weeks
chemically intolerant patients whose illness began after a post–formaldehyde exposure (when the experiment ended)
chemical exposure revealed “cobblestoning” of the pharyn- (Sorg, 1996). Kay (1996) showed that rats exposed to
geal mucosa, resembling lymphoid hyperplasia, in 6 patients toluene vapors, but not most food odors (except mint),
and pale mucosal foci in 8 patients (Meggs and Cleveland, exhibited narrow-band, high-amplitude 15–30 Hz oscilla-
1993). Nasal biopsies of 13 chlorine dioxide–exposed work- tions in the olfactory-limbic tract. Repeated low-level
ers who developed chemical intolerances likewise showed exposure to toluene (below OSHA legal exposure limits) in
inflammation and nerve fiber proliferation (Meggs et al., rats adversely affected their performance of complex
1996). In some specimens, mucosal epithelial cells were learning and spatial memory tasks (Rogers et al., 1999;
detached from the basement membrane and intraepithelial Von Euler et al., 1993).
cell junctions were disrupted. Eosinophils did not appear Consistent with these limbic changes in animals,
elevated. Further studies incorporating control subjects and chemically intolerant individuals score higher on the
blinded assessments are needed. McLean limbic system checklist (Teicher et al., 1993), a
German investigators have been examining the questionnaire probing symptoms associated with temporal
chemosensory event–related potentials of chemically intol- lobe seizures which frequently originate in the limbic sys-
erant patients in response to olfactory (hydrogen sulfide) tem (amygdala) (Bell, 1996). Notably, women with tem-
and trigeminal (carbon dioxide) cues (Hummel et al., poral lobe epilepsy have increased rates of self-reported
1996), both before and after exposure to a common solvent polycystic ovary disease, possibly due to amygdala-regu-
(2-propanol). While solvent exposure did not affect sub- lated release of hypothalamic reproductive hormones.
jects’ odor thresholds, their ability to discriminate odors Chemically intolerant women also self-report more men-
appeared heightened after solvent versus clean air expo- strual problems than do controls (Bell et al., 1995). In a
sure. No normal subjects have yet been tested. Other inves- series of inhalation challenge studies using various low
tigators propose that odor conditioning may play a role in level exposures, Bell et al. (1996, 1997a,b,c) found evi-
multiple chemical intolerance (Bolla-Wilson et al., 1989; dence for sensitization of heart rate, blood pressure,
Guglielmi et al., 1994; Schottenfeld and Cullen, 1986; plasma beta-endorphins, and EEG activity in chemically
Schusterman and Dager, 1991; Siegel and Kreutzer, 1997). intolerant individuals, but not in controls.
Bell et al. (1999) and Sorg (2000) propose that multiple Cholinesterase-inhibiting pesticide exposure has been
chemical intolerance results from central neural sensitiza- implicated in the initiation of multiple chemical intoler-
tion processes. Neural sensitization is the increased neu- ance and the Gulf War veterans’ unexplained illnesses
ronal responsiveness to a stimulus following repeated (Cone and Sult, 1992; Haley et al., 1999; Miller and
exposures to the same stimulus or a different stimulus, a Mitzel, 1995). Rats bred for organophosphate sensitivity
form of learning being explored in laboratory animals show greater susceptibility to various cholinergic ago-
repeatedly exposed to drugs of abuse, especially cocaine nists but, interestingly, are also less tolerant of nicotine,
and amphetamines. The limbic and mesolimbic brain serotonin agonists, dopamine antagonists, diazepam,
regions are particularly susceptible to sensitization, a ethanol, and sucrose—structurally unrelated substances
process that appears to involve excitatory amino acids, (Djuric et al, 1995; Overstreet, 1996). These specially
which may alter pain perception, olfaction, learning, and bred rats have about 20% more cholinergic receptors in
memory. In this rodent model, physical stressors, e.g., foot certain limbic regions, including the hippocampus and
shock and tail pinching, augment stimulant drug responses. striatum. When sensitized to egg protein (ovalbumin) via
Multiple Chemical Intolerance 539

intraperitoneal injection and subsequently fed ovalbumin, remain addicted—to avoid withdrawal symptoms. The the-
these cholinergically sensitive rats showed greater gut ory is that toxicant-induced loss of tolerance leads to
permeability than did control rats (Djuric et al., 1995). amplified stimulatory and withdrawal symptoms and that
Increased gut permeability in humans is thought to the outwardly polar behaviors of the drug addict and the
underlie various food intolerances. Likewise, inhaled chemically intolerant patient may both be strategies for
methacholine and inhaled ovalbumin following ovalbu- avoiding withdrawal symptoms (Miller, 1999, 2001a;
min presensitization via intraperitomal injection resulted Newlin, 1997).
in greater airway reactivity in cholinergically sensitive Proposed psychological mechanisms for multiple
rats than controls (Djuric et al., 1998). chemical intolerance include odor conditioning, physician-
Differences in the ability to absorb, metabolize, and induced (iatrogenic) beliefs, panic disorder, toxic agora-
excrete various chemicals might explain why some people phobia, posttraumatic stress disorder (e.g., illness resulting
are more vulnerable to developing multiple chemical from a traumatic chemical spill or childhood sexual
intolerance, for example, decreased sulfation capacity abuse), somatoform disorder, and depression (Binkley and
(McFadden, 1996), abnormal porphyrin metabolism Kutcher, 1997; Göthe et al., 1995; Gots, 1995; Guglielmi
(Morton, 1995), or paraoxonase (organophosphate-detoxi- et al., 1994; Kurt, 1995; Pennebaker, 1994; Simon, 1994;
fying enzyme) deficiency (Costa et al., 1999; Haley et al., Sparks et al., 1994a, b; Spyker, 1995; Staudenmayer, 1999;
1999). To date, no adequately controlled studies exploring Staudenmayer and Selner, 1987; Staudenmayer et al.,
these possibilities have been published. 1993). Davidoff and Fogarty (1994) examined 10 pub-
Other proposed explanations include: (1) neurogenic lished studies that explored possible psychogenic theories
inflammation—e.g., increased c-fiber neuron density in for chemical intolerance. All drew scientifically unsup-
affected tissues; greater neuropeptide and prostanoid portable conclusions concerning cause and effect, and
production by susceptible subjects; increased and pro- erroneously assumed that psychological symptoms were
tracted response to c-fiber activators like capsaicin; psychogenic when chemical exposures might also explain
increased central autonomic response following c-fiber them.
stimulation; and decreased mucosal neutral endopepti- Some symptoms these patients report mimic panic dis-
dase; (2) nonneurogenic inflamation—e.g., increased order, including hyperventilation, lightheadedness, chest
inflammation in affected tissues resulting in augmented discomfort, palpitations, paresthesias, and impaired men-
neurosensory response; increased inflammatory tation, leading researchers to wonder whether anxiety
response to chemical exposure; and (3) altered percep- underlies chemical intolerance. Eleven of 15 chemically
tual and central integration—e.g., differences in adapta- intolerant patients exposed to their own problem triggers,
tion, habituation, cortical representation, perception, e.g., hairspray or the yellow pages, reported that these
cognition, and/or hedonics; different qualitative and reproduced their characteristic responses, including
quantitative interactions between trigeminal and olfac- tachycardia, tremor, and pallor (Leznoff, 1997). Exhaled
tory systems; different higher integration of sensory CO2 in all 11 responders decreased, consistent with hyper-
inputs (Bascom et al., 1997). ventilation due to an anxiety reaction. A few responded to
Randolph first noted striking parallels between chemi- odorless stimuli, e.g., sugar or water, as well. In a single-
cal intolerance and drug addiction (Randolph, 1962), blind study, Binkley and Kutcher (1997) administered
including the presence of stimulatory and withdrawal saline (placebo) intravenous sodium lactate (known to
symptoms, cravings, and cross-addiction/intolerances to elicit attacks in persons with panic disorder) to 5 chemi-
structurally diverse substances (summarized in Table 4). In cally intolerant patients, all of whom responded with
this model, both addiction and chemical intolerance panic-like symptoms. In another single-blind case-control
involve a fundamental breakdown in tolerance that occurs study, 31 chemically intolerant patients and 31 healthy
only in certain susceptible individuals following repeated controls inhaled a single breath of air alone or air spiked
exposures, whether to drugs of abuse or to chemicals. It is with CO2 (35%) (Poonai et al., 2000). Inhaled CO2 elicits
hypothesized that repeated exposures to drugs or chemi- panic symptoms in the majority of patients with panic dis-
cals result in amplified stimulatory and withdrawal symp- order, but only 5% of healthy controls. Significantly more
toms (Miller 1997, 1999). Subsequently, in order to chemically intolerant patients (71%) than controls (26%)
prevent unpleasant withdrawal symptoms, drug abusers fulfilled panic criteria after CO2 inhalation. Both groups
take another “hit,” and then another, leading to addiction. showed significant changes in heart and breathing rates
In contrast, chemically intolerant individuals who link after CO2 versus air, with no significant difference
their withdrawal symptoms to alcohol, caffeine, etc., shun between patients and controls. The high percentage of con-
these substances, but perhaps for the same reason addicts trols responding in this study and the fact that the CO2
540 Miller

Table 4 Parallel and Opposing Features of Addiction and Abdiction (Chemical Intolerance)
Feature Addiction Abdiction
Multisystem symptoms, especially  
central nervous system symptoms
Multiple, chemically unrelated substances  
affecting same individual (cross-tolerance) (cross-tolerance or “spreading”)
Caffeine, alcohol, nicotine, drugs implicated  
Size of doses “tolerated” vs. those tolerated large small
by general population
Inhalation, ingestion, injection or  
transmucosal routes
Stimulatory and withdrawal symptoms  
Heightened sensitivity to physical stimuli  
(noise, light, heat, cold, touch, vibration)
during withdrawal phase
Cravings, bingeing   (caffeine, foods)
Habituation  
Heightened sensitivity following period  
of avoidance (e.g., tobacco)
Genetic predisposition  
Demographics Poorly educated males, College educated females, middle
lower socioeconomic status to upper socioeconomic
status
Gender ratio (M:F) 2:1 3:5
Age of onset Teens, 20–30 years 30–50 years
Ill-defined physiological mechanisms  
Lack of biological markers  
Lack of effective drugs for treating condition  
Primary therapeutic strategy Abstinence Avoidance
Detox/withdrawal requiring 4–7 days  
Societal views concerning nature of problem Disease vs. lack of willpower Disease vs. belief system leading
to avoid substances (underavoidance) to avoidance of substances
(overavoidance)
Patients viewed as difficult, demanding  
Linked to violence, physical/sexual abuse, suicide  a
Disruption of work, family and social relationships  
a If chemical intolerance and addiction are interrelated and tend to cluster in families, then the higher rates of childhood abuse described by some inves-
tigators (Staudenmayer et al., 1993) among chemically intolerant women conceivably could be the consequence of alcoholic or drug-addicted elders
who abused them.

mixture “had a marked taste” suggests study design prob- capacity. Carefully conducted studies are needed to untan-
lems. The other two panic studies cited above did not use gle the current confusion of competing hypotheses.
controls. Many symptoms of chemical intolerance deviate Funding for such studies has been scant.
from the panic profile. Hyperventilation during chemical
(or even placebo) exposure in a setting where patients are
being scrutinized by researchers and administered chemi- VII. MEDICAL MANAGEMENT
cal challenges does not prove that the illness is psy-
chogenic. A sizable percentage of healthy controls appear Forty percent of chemically intolerant patients in one study
to respond similarly. Results of the lactate infusion study had consulted 10 or more medical practitioners (Miller and
do not rule out possible metabolic problems in these Mitzel, 1995). Some see clinical ecologists, while others
patients, e.g., abnormal red blood cell oxygen-carrying remain with their family doctors. Still others seek out aca-
Multiple Chemical Intolerance 541

demic occupational and environmental medicine doctors, While the vast majority of patients view avoidance of
particularly when workplace exposures are involved. Even odor/chemical and food triggers as paramount (Johnson,
physicians familiar with the phenomenon have difficulty 1996; LeRoy et al., 1996; Miller, 1995), psychological
managing these cases. Patients are apt to ask about various support (to be distinguished from traditional psychother-
unorthodox therapies. At this time, controlled studies for apy) can be helpful in any illness, whether the condition is
any treatments are lacking. Ideally, physicians should not psychogenic or physical in origin. Some authors have
chastise patients who attempt alternative treatments, but advocated psychological interventions as the preferred or
display their desire to see the patients improve, help them only acceptable treatment modality for multiple chemical
avoid potentially dangerous interventions, and serve as intolerance (Sparks et al., 1994b), expressing concern that
their advocates during a perplexing illness. “Increasingly, the illness may be iatrogenic. At this time, placing sole
in difficult circumstances, the reasonable trend in medicine reliance on psychological therapies, to the exclusion of
is to explain the options and allow the patient to decide” trial avoidance, is at best premature and, at worst, poten-
(Vasey, 1995). Ample time should be allotted for visits tially harmful (Miller, 1995). A multifaceted approach is
and/or telephone consultations if patients cannot travel. called for, one that includes the identification and avoid-
Often, other patients and support groups can offer assis- ance of odor/chemical and food triggers, low- or no-cost
tance. alterations of patients’ workplace and home environments,
A multidisciplinary approach, paralleling that used for and appropriate psychological support.
chronic pain, has been suggested (Weaver, 1996). Weaver In the United States, affected individuals must chart
advises: “Regardless of the treatment chosen, it is impor- their own course to recovery. Balanced medical informa-
tant to emphasize that functional improvement and tion and knowledgable, sympathetic physicians are diffi-
increased patient control, not cure, are the goals. Complete cult to find. Fatigue and concentration difficulties may
resolution of odor sensitivity may not be possible.” undermine care seeking, the ability to identify triggers,
Patients can exhaust their financial resources, their fami- implementing lifestyle changes, obtaining social support
lies, and themselves pursuing alternative treatments, when services, etc.
time and money might be better spent modifying their Chemically intolerant patients frequently report adverse
home environment. reactions to medications and poor tolerance for standard
Consistently, the single most helpful intervention drug doses. They appear to be at increased risk of develop-
reported by these patients has been avoidance of problem ing both known drug side effects (even at low doses) and
exposures. In a survey of 206 MCS patients with an aver- idiosyncratic reactions (McLellan, 1987). When new symp-
age educational level of nearly 4 years of college, 71% toms surface, it can be difficult to determine whether they
rated avoidance of problem chemicals, and 54% avoidance are due to a medication (the drug itself, a dye, excipient, or
of problem foods, as “very helpful.” Although 52% had diluent), an environmental exposure, or some new medical
tried psychological or psychiatric therapies, only 17% of problem, e.g., cancer or coronary artery disease.
those who had tried them rated them as “very helpful” Exhaustively evaluating each new symptom quickly
(Miller, 1995). becomes costly and exposes patients to other risks, includ-
DePaul University researchers found that at least three ing anesthetic agents or x-ray contrast dye, potentially
fourths of 305 chemically intolerant subjects who had tried adding more layers of “unexplained” symptoms. Patients
the following interventions found them to be of “enor- may improve as medications are removed from their regi-
mous” or “major” help: avoiding chemicals that cause mens. Some patients benefit from lower-than-normal drug
reactions (93%), creating an environmentally safe (odor- doses (e.g., analgesics or antidepressants, but not antibi-
free) living space (86%), moving to a less polluted area otics), a noteworthy clinical observation consistent with
(76%), and avoiding foods that provoke reactions (75%) their amplified responses to environmental pollutants.
(LeRoy et al., 1996). In a grass-roots survey of 243 chemi- Some find they can tolerate certain drugs better than others
cally intolerant patients, nearly half of whom were on dis- and make special requests, e.g., dental anesthetics without
ability, at least three fourths reported these interventions to epinephrine. Any reasonable request should be honored.
be of “enormous help” or “major help”: avoidance of Environmental testing rarely contributes to understand-
chemical exposures (95%), relocation to avoid pollution ing these patients’ health problems. However, professional
(79%), and avoidance of problem foods (76%) (Johnson, industrial hygienists or indoor air consultants sensitive to
1996). Given the patients’ clear consensus that chemical these concerns can help detect contaminant sources and
and food avoidance benefit them most, increasing numbers make recommendations for minimizing exposures follow-
of physicians have begun to recommend avoidance strate- ing a walk-through workplace or home evaluation. These
gies for these patients. consultants need to be nonsmokers with an excellent sense
542 Miller

of smell. Smokers and passive smokers (as well as workers printers) from the immediate work environment, providing
routinely exposed to petroleum products) are less able to an alternative work space, removing carpeting, selecting
perceive various low-level odors, perhaps as a conse- odorless and less toxic cleaning agents, adopting integrated
quence of sensory deficit or habituation (Ahlstrom et al., pest management, and allowing employees to work from
1986, 1987). Indoor air contaminant concentrations of home (Miller 2001; Miller et al., 1999). Physicians can
importance for chemically intolerant patients appear to be facilitate the accommodation process by sending reason-
orders of magnitude below those prescribed by law able written requests to employers or schools on their
(OSHA) for occupational environments. Unless exceeded, patient’s behalf, but only with the patient’s full knowledge
legally adequate occupational exposure levels have little and written consent, so as to protect the physician should
relevance and should not be invoked. the patient’s job be eliminated.
As for any chronic illness, psychological support may Multiple chemical intolerance is increasingly being
be helpful, depending upon the patient’s needs and will- viewed as a disability (Winterbauer, 1997). Internal mem-
ingness. Tremendous disruption of work, family, and oranda of the Social Security Administration and
social lives can occur with this illness, e.g., divorces from Department of Housing and Urban Development recognize
spouses who smoke. Suicides have been reported. Support the illness for purposes of compensation and housing
can be provided by psychologists, psychiatrists, social accommodation, respectively. Recent Equal Employment
workers, or primary care doctors, although patients rarely Opportunity Commission (EEOC) statistics show that
find that psychological interventions alter their intoler- from November 1, 1993, through March 31, 2000, 564
ances (Johnson, 1996; Miller 1995). Actively psychotic or multiple chemical intolerance discrimination–related com-
suicidal patients need to be taken seriously and treated and plaints were filed against employers, 60% alleging failure
protected accordingly. to provide reasonable accommodation and nearly 50%
Various investigators advocate psychological or psychi- wrongful discharge (EEOC, 2000).
atric interventions, but only anecdotal data suggest that The courts have struggled over whether the illness
these approaches may be helpful (Amundsen et al., 1996; should be viewed as a disability, issuing conflicting opin-
Bolla-Wilson et al., 1989; Guglielmi et al., 1994; ions. Current law would make it difficult for an employer
Schottenfeld and Cullen, 1985; Spyker, 1995). Several to claim that a condition that so greatly restricts daily
authors claim their patients improved as a result of psy- activity is not a disability (Winterbauer, 1997). The
chotherapeutic interventions, but adequate follow-up is not Americans with Disabilities Act (ADA) obligates employ-
provided (Amundsen et al., 1996; Bolla-Wilson et al., ers to seek inexpensive, practical solutions that will reduce
1989; Guglielmi et al., 1994; Schottenfeld and Cullen, troublesome odorous and odorless (e.g., pesticide) expo-
1985; Spyker, 1995). Staudenmeyer et al. (1993) similarly sures (Winterbauer, 1997). It does not require that a chemi-
attest, without providing follow-up or appropriate study cal-free workplace be provided.
design, that among their patients who agreed to undergo
psychotherapy, 75% had a successful outcome.
Among 243 chemically intolerant individuals who had VIII. ILLNESS COURSE, PROGNOSIS, AND
taken an antidepressant, 10–20% found them of “major” or PREVENTION
“enormous” help; 50–65% reported harmful effects; and
10–30% reported they were not helpful or their effect was Few patients report full recoveries, even decades after they
unclear (Johnson, 1996). Two single-case reports have become sick. There are those rare individuals whose illness
appeared in the literature involving individuals whose odor was recognized early in its course, who avoided further
intolerances apparently resolved while they were taking exposure, and who appear to have recovered (Hileman,
selective serotonin-reuptake inhibitor antidepressants, one 1991). At this time, early recognition and avoidance of fur-
case involving psychological desensitization as an adjunct ther exposure offer the greatest potential to prevent dis-
(Stenn and Binkley, 1998) and the other using drug alone ability. Difficulties in treating the illness, once entrenched,
(Andiné et al., 1997). underscore the importance of prompt action. In one clini-
Accommodating these individuals in the workplace can cal series, chemically intolerant individuals who suc-
be challenging but vital to their self-esteem and livelihoods. ceeded in avoiding at least half of their self-reported
Some patients are able to continue working provided that incitants reported feeling better than did nonavoiders at
they avoid problem exposures, e.g., coworkers’ fragrances follow-up 6 months to 21/2 years after their initial visit (Lax
or copier machines. Workplace accommodations can and Henneberger, 1995).
include increasing fresh air supply and air circulation, In the future, understanding multiple chemical intol-
removing business machines (fax machines, copiers, laser erance will be essential for establishing sound national
Multiple Chemical Intolerance 543

and corporate environmental policy. If there is a subset AOEC (Association of Occupational and Environmental Clinics)
of the population that is especially susceptible to various (1992). Advancing the understanding of multiple chemical
“odors,” i.e., common low-level chemical exposures, sensitivity. Toxicol. Ind. Health 8(4):1–257.
strategies to protect these people must be found. If only Ashford, N., and Miller, C. (1989). Chemical sensitivity: a
report to the New Jersey State Department of Health. Trenton,
certain chemical exposures initiate the illness process,
NJ.
then the focus should be on avoiding or minimizing
Ashford, N., and Miller, C. (1998). Chemical Exposures: Low
those exposures, for example, preventing chemical Levels and High Stakes. John Wiley and Sons, Inc., New York.
spills, prohibiting building occupancy during finish-out, Ashford, N., Heinzow, B., Lütjen, K., Marouli, C., Mølhave, L.,
and using integrated pest management. Besides regulat- Mönch, B., Papadopoulos, S., Rest, K., Rosdahl, D., Siskos,
ing exposures that may initiate the illness, notifying peo- P., and Velonakis, E. (1995). Chemical sensitivity in selected
ple in advance of anticipated exposures, e.g., reroofing European countries: an exploratory study. A Report to the
or painting, should be standard practice. Such strategies European Commission. Ergonomia, Athens, Greece.
may protect more vulnerable individuals from becoming ASHRAE (American Society of Heating, Refrigeration and Air
sick in the first place and thereby prevent unnecessary Conditioning Engineers). (1999). Ventilation for acceptable
and costly overregulation of environmental exposures in indoor air quality, ASHRAE Standard 62–99, Atlanta, GA.
ATSDR (Agency for Toxic Substances and Disease Registry)
the future.
(1994). Proceedings of the Conference on Low-Level
Exposure to Chemicals and Neurobiologic Sensitivity.
Toxicol. Ind. Health 10(4/5):253–669.
REFERENCES
Baldwin, C., and Bell, I. (1998). Increased cardiopulmonary dis-
ease risk in a community-based sample with chemical odor
AAAAI (American Academy of Allergy, Asthma and intolerance: implications for women’s health and health-care
Immunology) (1981). Position statements: controversial tech- utilization. Arch. Environ. Health 53:347–353.
niques. J. Allergy Clin. Immunol. 67(5):333–338. Bartha, L., Baumzweiger, W., Buscher, D., Callender, M., Dahl,
AAAAI (American Academy of Allergy, Asthma and K., Davidoff, A., et al. (1999). Multiple chemical sensitivity:
Immunology) (1986). Position statements: clinical ecology. a 1999 consensus. Arch. Environ. Health 54(3):147–149.
J. Allergy Clin. Immunol. 72(8):269–271. Bascom, R. (1989). Chemical hypersensitivity syndrome study:
AAAAI (American Academy of Allergy, Asthma and options for action. A literature review and a needs assessment.
Immunology) (1999). Idiopathic environmental intolerances. Prepared for the State of Maryland Department of Health.
J. Allergy Clin. Immunol. 103:36–40. February 7.
ACOEM (American College of Occupational and Environmental Bascom, R. (1991). Multiple chemical sensitivity: a respiratory
Medicine) (1999). Multiple chemical sensitivities: idiopathic disorder. Toxicol. Ind. Health 8(4):221–228.
environmental intolerance. J. Occ. Environ. Med. Bascom, R., Meggs, W., Frampton, M., Hudnell, K., Killburn, K.,
41(11):940–942. Kobal, G., Medinsky, M., and Rea, W. (1997). Neurogenic
ACP (American College of Physicians) (1989). Clinical ecology: inflammation: with additional discussion of central and per-
position statement. Ann. Intern. Med. 111:168–178. ceptual integration of nonneurogenic inflammation. Environ.
ACS (American Chemical Society) (1999). Special issue on multi- Health Perspect. 105(suppl. 2): 531–537.
ple chemical sensitivity. Toxicol. Ind. Health 15(3–4):283–437. Bell, I. R. (1996). Clinically relevant EEG studies and psy-
Ahlstrom, R., Berglund, B., Berglund, U., Lindvall, T., and chophysiological findings: possible neural mechanisms for
Wennberg, A. (1986). Impaired odor perception in tank clean- multiple chemical sensitivity. Toxicology 111:101–117.
ers. Scand. J. Work Environ. Health 12(6): 574–581. Bell, I., Peterson, J., and Schwartz, G. (1995). Medical histories
Ahlstrom, R., Berglund, B., Berglund, U., Engen, T., and and psychological profiles of middle-aged women with and
Lindvall, T. (1987). A comparison of odor perception in without self-reported illness from environmental chemicals.
smokers, nonsmokers, and passive smokers. Am. J. J. Clin. Psychiatry 56(4): 151–160.
Otolaryngol. 8(1):1–6. Bell, I., Miller, C., Schwartz, G., Peterson, J., and Amend, D.
AMA (American Medical Association, Council of Scientific (1996). Nueropsychiatric and somatic characteristics of young
Affairs) (1992). Clinical ecology. JAMA 268:3465–3467. adults with and without self-reported chemical odor intolerance
Amundsen, M., Hanson, N., Bruce, B., Lantz, T., Schwartz, M., and chemical sensitivity. Arch. Environ. Health 51(1):9–21.
and Lukach, B. (1996). Odor aversion or multiple chemical Bell, I., Rossi, J., Gilbert, M., Kobal, G., Morrow, L., Newlin, D.,
sensitivities: recommendations for a name change and Sorg, B., and Wood, R. (1997a). Testing the neural sensitiza-
description of successful behavioral medicine treatment. Reg. tion and kindling hypothesis for illness from low levels of
Toxicol. Pharmacol. 24:S116–S118. environmental chemicals. Environ. Health Perspect.
Andiné, R., Rönnbäck, L., and Järyholm, B. (1997). Successful 105(suppl. 2):539–47.
use of a selective serotonin reuptake inhibitor in a patient with Bell, I., Schwartz, G., Baldwin, C., Hardin, E., Klimas, N., Kline,
multiple chemical sensitivities. Acta Psychiatr. Scand. J., Patarca, R., and Song, Z. (1997b). Individual differences in
96:82–83. neural sensitization and the role of context in illness from
544 Miller

low-level environmental chemical exposures. Environ Health Djurić, V., Overstreet, D., Bienenstock, J., and Perdue, M. (1995).
Perspect. 105(suppl. 2):457–66. Immediate hypersensitivity in the Flinders rat: further evi-
Bell, I., Schwartz, G., Bootzin, R., and Wyatt, J. (1997c). Time- dence for a possible link between susceptibility to allergies
dependent sensitization of heart rate and blood pressure over and depression. Brain Behav. Immun. 9(3): 196–206.
multiple laboratory sessions in elderly individuals with Djurić, V., Cox, G., Overstreet, D., Smith, L., Dragomir, A., and
chemical odor intolerance. Arch. Environ. Health 52(1): Steiner M. (1998). Genetically transmitted cholinergic hyper-
6–17. responsiveness predisposes to experimental asthma. Brain
Bell, I., Schwartz, G., Hardin, F., Baldwin, C., and Kline, J. Behav. Immun. 12:272–284.
(1998). Differential resting quantitative electroencephalo- Doty, R.L., Deems, D.A., Frye, R.E., Pelberg, R., and Shapiro, A.
graphic alpha patterns in women with environmental chemi- (1988). Olfactory sensitivity, nasal resistance, and autonomic
cal intolerances, depressives, and normals. Biol. Psychiatry function in patients with multiple chemical sensitivities. Arch.
43:376–388. Otolaryngol. Head Neck Surg. 114:1422–1427.
Bell, I., Baldwin, C., Fernandez, M., and Schwartz, G. (1999). EEOC (2000). Statistics on MCS-related complaints. May 31,
Neural sensitization model for multiple chemical sensitivity: 2000. Personal communication from the Director of the
overview of theory and empirical evidence. Toxicol. Ind. EEOC Office of Research, Information and Planning, Deidre
Health 15(3–4):295–304. Flippen, Washington, DC.
Binkley, K., and Kutcher, S. (1997). Panic response to sodium Fiedler, N., and Kipen, H. (1997a). Experimental approaches to
lactate infusion in patients with multiple chemical sensitivity chemical sensitivity. Environ. Health Perspect. 105(suppl.
syndrome. J. Allergy Clin. Immunol. 99(4):570–574. 2):405–547.
Berglund, B., Lindvall, T., and Nordin, S. (1992). Fiedler, N., and Kipen, H. (1997b). Chemical sensitivity: the sci-
Environmenally induced changes in sensory sensitivities. entific literature. Environ. Health Perspect. 105(suppl.
Ann. NY Acad. Sci. 641:304–321. 2):409–415.
Black, D., Rathe, A., and Goldstein, R. (1990). Environmental ill- Fiedler, N., Maccia, C., and Kipen, H. (1992). Evaluation of
ness: a controlled study of 26 subjects with “20th century dis- chemically-sensitive patients. J. Occup. Med. 34:529–538.
ease.” JAMA 264:3166–3170. Fiedler, N., Kipen, H., DeLucia, J., Kelly-McNeil, K., and
Black, D., Okiishi, C., Gabel, J., and Schlosser, S. (1999). Natelson, B. (1996a). A controlled comparison of multiple
Psychiatric illness in the first-degree relatives of persons chemical sensitivities and chronic fatigue syndrome.
reporting multiple chemical sensitivities. Toxicol. Ind. Health Psychosom. Med. 58:38–49.
15:410–414. Fiedler, N., Kipen, H., Natelson, B., and Ottenweller, J. (1996b).
Bolla-Wilson, K., Wilson, R., and Bleecker, M. (1989). Chemical sensitivities and the Gulf War: Department of
Conditioning of physical symptoms after neurotoxic expo- Veterans Affairs Research Center in basic and clinical science
sure. J. Occup. Med. 31(9):684–686. studies of environmental hazards. Reg. Toxicol. Pharmacol
Buchwald, D., and Garrity, D. (1994). Comparison of patients 24(1 Pt 2):S129–S138.
with chronic fatigue syndrome, fibromyalgia, and multiple Genton, M. (1998). Shedding light on darkroom disease: progress
chemical sensitivities. Arch. Intern. Med. 154:2049–2053. and challenges in understanding radiology workers’ occupa-
Chester, A, and Levine, P. (1994). Concurrent sick building syn- tional illness. Can. J. Med. Radiation Tech. 2(2):60–66.
drome and chronic fatigue syndrome: epidemic neuromyas- Göthe, C., Molin, C., and Nilsson, C. (1995). The environmental
thenia revisited. Clin. Infect. Dis., 18(suppl.1): S43–S48. somatization syndrome. Psychosomatics 36(1):1–11.
Cone, J., and Sult, T. (1992). Acquired intolerance to solvents fol- Gots, R. (1995). Multiple chemical sensitivities—public policy
lowing pesticide/solvent exposure in a building: a new group (editorial). J. Toxicol. Clin. Toxicol 33(2):111–113.
of workers at risk for multiple chemical sensitivity. Toxicol. Guglielmi, R., Cox, D., Spyker, D. (1994). Behavioral treatment
Ind. Health 8(4):29–39. of phobic avoidance in multiple chemical sensitivity. J. Behav.
Costa, L., Li, W., Richter, R., Shih, D., Lusis, A., and Furlong, C. Ther. Exp. Psychiatry 25(3):197–209.
(1999). The role of paraoxonase (PON1) in the detoxication Haley, R., Billecke, S., and La Du, B. (1999). Association of low
of organophosphates and its human polymorphism. Chemico- PON1 type Q (type A) arylesterase activity with neurologic
Biol. Interact. 119–120:429–438. symptom complexes in Gulf War veterans. Toxicol. Appl.
Cullen, M., ed. (1987). Workers with Multiple Chemical Pharmacol. 157(3):227–233.
Sensitivities: An Overview, Occupational Medicine: State of Heuser, G., and Mena, L. (1998). Neurospect in neurotoxic
the Art Reviews. Hanley & Belfus, Philadelphia. chemical exposure demonstration of long-term functional
Davidoff, A., and Fogarty, L. (1994). Psychogenic origins of mul- abnormalities. Toxicol. Ind. Health 14(6):813–827.
tiple chemical sensitivity syndrome: A critical review of the Hileman, B. (1991). Multiple chemical sensitivity. Chem. Eng.
research literature. Arch. Environ. Health 49(5):316–325. News 69(29):26–42.
Davidoff, A., and Keyl, P. (1996). Symptoms and health status in Hirzy, J., and Morrison, R. (1989). Carpet/4-phenylcyclohexene
individuals with multiple chemical sensitivities syndrome toxicity: the EPA headquarters case. Presented at the Annual
from four reported sensitizing exposures and a general Meeting of the Society for Risk Analysis, San Francisco.
population comparison group. Arch. Environ. Health Hu, H., Johnson, K., Heldman, R., Jones, K., Komaroff, A. L.,
51(3):201–213. Schacterle, R., Barsky, A., Becker, A., and Holman, L. (1999).
Multiple Chemical Intolerance 545

A comparison of single photon emission computed tomography Health, April 27–29, 1994. Bethesda, MD: National Institutes
in normal controls, in subjects with multiple chemical sensitiv- of Health.
ity syndrome, and in subjects with chronic fatigue syndrome. Miller, C. (1995). Letter to the editor. J. Occup. Med.
Department of Labor and Industries, State of Washington. 37(12):1323.
Hummel, T., Roscher, S., Jaumann, M., and Kobal, G. (1996). Miller, C. (1996). Chemical sensitivity: symptom, syndrome or
Intranasal chemoreception in patients with multiple chemical mechanism for disease? Toxicology 111:69–86.
sensitivities: a double-blind investigation. Reg. Toxicol. Miller, C. (1997). Toxicant-induced loss of tolerance: an emerg-
Pharm. 24:S79–S86. ing theory of disease? Environ. Health Perspect. 105(suppl 2):
Institute of Medicine (IOM) (1995). Environmental Medicine: 445–453.
Integrating Missing Elements into Medical Education, A. Miller, C. (1999). Are we on the threshold of a new theory of dis-
Pope and D. Rall (Eds.). National Academy Press, ease? Toxicant-induced loss of tolerance and its relationship
Washington, DC. to addiction and abdiction. Toxicol. Ind. Health 15: 284–294.
Johnson, A. (1996). MCS Information Exchange. Brunswick, Miller, C. (2000). Toxicant-induced loss of tolerance: addiction,
ME. abdiction, and the chemical environment. Addiction
Kay, L. (1996). Support for the Kindling Hypothesis in multiple 96(1):115–139.
chemical sensitivity syndrome (MCSS) induction. Miller, C. (2001). Multiple chemical intolerance and indoor air
Presentation at the 26th Annual meeting of the Society for quality. In Handbook of Indoor Air Quality, Samet, J., and
Neuroscience, Washington, DC, November 16–21. Spengler, J. (Eds.). McGraw-Hill, NY, pp. 27.1–27.32
Kruetzer, R., Neutra, R., and Lashuay, N. (1999). Prevalence of Miller, C., and Mitzel, H. (1995). Chemical sensitivity attributed
people reporting sensitivities to chemicals in a population- to pesticide exposure versus remodeling. Arch. Environ.
based survey. Am. J. Epidemiol. 150(1):1–12. Health 50(2):119–129.
Kuhn, T. 1970. The Structure of Scientific Revolutions. The Miller, C., and Prihoda, T. (1999a). The Environmental Exposure
University of Chicago Press, Chicago, IL. and Sensitivity Inventory (EESI): a standardized approach for
Kurt, T. (1995). Multiple chemical sensitivities—a syndrome of measuring chemical intolerances for research and clinical
pseudotoxicity manifest as exposure perceived symptoms. applications. Toxicol. Ind. Health 15:370–385.
J. Toxicol. Clin. Toxicol. 33(2):101–105. Miller, C., and Prihoda, T. (1999b). A controlled comparison of
Lax, M., and Henneberger, P. (1995). Patients with multiple symptoms and chemical intolerances reported by Gulf War
chemical sensitivities in an occupational health clinic: presen- veterans, implant recipients and persons with multiple chemi-
tation and follow-up. Arch. Environ. Health 50(6):425–431. cal sensitivity. Toxicol. Ind. Health 15:386–397.
LeRoy, J., Davis, T., and Jason, L. (1996). Treatment efficacy: a Miller, C., Ashford, N., Doty, R., Lamielle, M., Otto, D., Rahill,
survey of 305 MCS patients. CFIDS Chronicle. Winter:52–53. A., and Wallace, L., (1997). Empirical approaches for the
Leznoff, A. (1997). Provocative challenges in patients with investigation of toxicant-induced loss of tolerance. Environ.
multiple chemical sensitivity. J. Allergy Clin. Immunol. 99: Health Perspect. 105(suppl. 2):515–519.
438–442. Miller, C., Gammage, R., and Jankovic, J. (1999). Exacerbation
Mandell, M., and Scanlon, L. (1979). Dr. Mandell’s 5-Day of chemical sensitivity: a case study. Toxicol. Ind. Health
Allergy Relief System. Thomas Y. Crowell, New York. 15:398–402.
Mayberg, H. (1994). SPECT studies of multiple chemical sensi- Mitchell, C., Donnay, A., Hoover, D., and Margolick, J. (2000).
tivity. Toxicol. Ind. Health 10(4–5):661–666. Immunologic parameters of multiple chemical sensitivity. In
McFadden, S. (1996). Phenotype variation in xenobiotic metabo- Multiple Chemical Sensitivity/Idiopathic Environmental
lism and adverse environmental response: focus on sulfur- Intolerance, Occupational Medicine State of the Art Reviews,
dependant detoxification pathways. Toxicology 111:43–65. Sparks, P., (Ed.). Hanley and Belfus, Inc., Philadelphia,
McLellan, R. (1987). Biological interventions in the treatment of pp. 647–665.
patients with multiple chemical sensitivities. In Workers with Monk, J. (1996). Farmers fight chemical war. Chem. Ind.
Multiple Chemical Sensitivities, Occupational Medicine State February 5:108.
of the Art Reviews. Cullen M (Ed.). Hanley & Belfus Moorhead, J., and Suruda, A. (2000). Occipital lobe meningioma
Philadelphia, pp. 755–777. in a patient with multiple chemical sensitivities. Am. J. Ind.
Meggs, W. (1994). RADS and RUDS—the toxic induction of Med. 37:443–446.
asthma and rhinitis. Clin. Toxicol. 32(5):487–501. Morton, W. E. (1995). Redefinition of abnormal susceptibility to
Meggs, W., and Cleveland, C. (1993). Rhinolaryngoscopic exam- environmental chemicals. Presented at the Second
ination of patients with the multiple chemical sensitivity syn- International Congress on Hazardous Waste: Impact on
drome. Arch. Environ. Health 41(1):14–18. Human Ecological Health, Atlanta, Georgia, June 6.
Meggs, W., Dunn, K., Bloch, R., Goodman, P., and Davidoff, L. Nethercott, J., Davidoff, L., Curbow, B., and Abbey, H. (1993).
(1996). Prevalence and nature of allergy and chemical sensi- Multiple chemical sensitivities syndrome: toward a working
tivity in a general population. Arch. Environ. Health case definition. Arch. Environ. Health 48:19–26.
51(4):275–282. Newlin, D. (1997). A behavior-genetic approach to multiple
Miller, C. (1994). Multiple chemical sensitivity and the Gulf War chemical sensitivity. Environ. Health Perspect 105(suppl.
veterans. NIH Workshop on the Persian Gulf Experience and 2):505–508.
546 Miller

NIEHS (National Institute of Environmental Health Sciences) Simon, T., Hickey, D., Fincher, C., Johnson, A., Ross, G., and
(1997). Chemical sensitivity. Environ. Health Perspect. Rea, W. (1994). Single photon emission computed tomogra-
105(suppl. 2):405–547. phy of the brain in patients with chemical sensitivities. Toxicol
NRC (National Research Council) (1992). Multiple Chemical Ind. Health 10:573–577.
Sensitivities: Addendum to Biologic Markers in Sorg, B. (1996). Proposed animal model for multiple chemical
Immunotoxicology, National Research Council, National sensitivity in studies with formalin. Toxicology 111:135–145.
Academy of Sciences. National Academy Press, Sorg, B. (1999). Multiple chemical sensitivity: potential role for
Washington, DC. neural sensitization. Crit. Rev. Neurobiol. 13(3):283–316.
Overstreet, D. H. (1996). Alternatives to naltrexone in animal Sparks, P., Daniell, W., Black, D., Kipen, H., Altman, L.,
models. Alcohol Clin. Exp. Res. 20:231A–235A. Simon, G., and Terr, A. (1994a). Multiple chemical sensitiv-
Pennebaker, J. (1994). Psychological bases of symptom report- ity syndrome: a clinical perspective. I: Case definition, the-
ing: perceptual and emotional aspects of chemical sensitivity. ories of pathogenesis, and research needs. J. Occup. Med.
Toxicol. Ind. Health 10(4/5):497–511. 36(7):718–730.
Poonai, N., Antony, M., Binkley, K., Stenn, P., Swinson, R., Sparks, P., Daniell, W., Black, D., Kipen, H., Altman, L., Simon, G.,
Corey, P., Silverman, F., and Tarlo, S. (2000). Carbon dioxide and Terr, A. (1994b). Multiple chemical sensitivity syndrome: a
inhalation challenges in idiopathic environmental intolerance. clinical perspective. II: Evaluation, diagnostic testing, treatment,
J. Allergy Clin. Immunol. 106(2 part 1):358–363. and social considerations. J. Occup. Med. 36(7):731–737.
Randolph, T. (1956). The descriptive features of food addiction. Spyker, D. (1995). Multiple chemical sensitivities—syndrome
Q. J. Stud. Alcohol 17:198–224. and solution. J. Toxicol. Clin. Toxicol. 33(2):95–99.
Randolph, T. G. (1962). Human Ecology and Susceptibility to the Staudenmayer, H. (1999). Environmental Illness. Myth and
Chemical Environment. Charles C. Thomas, Springfield, IL. Reality. Boca Raton: Lewis Publishers.
Randolph, T. G., and Moss, R. W. (1980). An Alternative Staudenmayer, H., and Selner, J. (1987). Post-traumatic stress
Approach to Allergies. Lippincott and Crowell, New York. syndrome (PTSS): escape in the environment. J. Clin.
Rinkel, H. (1944). Food allergy: the role of food allergy in inter- Psychol. 43(1):156–157.
nal medicine. Ann. Allergy 2:115–124. Staudenmayer, H., Selner, M., and Selner, J. (1993). Adult seque-
Rinkel, H., Randolph, T., and Zeller, M. (1951). Food Allergy. lae of childhood abuse presenting as environmental illness.
Charles C. Thomas, Springfield, IL. Ann. Allergy 71:538–546.
Rogers, W., Miller, C., and Bunegin, L. (1999). A rat model of Stenn, P., and Binkley, K. (1998). Successful outcome in a patient
neurobehavioral sensitization to toluene. Environ. Health with chemical sensitivity: treatment with psychological
Perspect. 152:356–369. desensitization and selective serotonin reuptake inhibitor.
Rossi, G., Rea, W., Johnson, A., Hickey, D., and Simon, T. Psychosomatics 39(6):547–550.
(1999). Neurotoxicity in single photon emission computed Stephens, R., Spurgeon, A., Calvert, I., Beach, J., Lery, L., Berry,
tomography brain scans of patients reporting chemical sensi- H., and Harrington, J. (1995). Neuropsychological effect of
tivities. Toxicol Ind. Health 15(3–4):415–420. long-term exposure to organophosphates in sheep dip. Lancet
Ryan, C., Morrow, L., and Hodgson, M. (1988). Cacosmia and 345:1135–1139.
neurobehavioral dysfunction associated with occupational Teicher, M., Glod, C., Surrey, J., and Swett, C. (1993). Early
exposure to mixtures of organic solvents. Am. J. Psychiatry childhood abuse and limbic system ratings in adult psychiatric
145(11):1442–1445. outpatients. J. Neuropsychiatry Clin. Neurosci. 5:301–306.
Schottenfeld, R., and Cullen, M. (1985). Occupational-induced Thomson, G. (1985). Report of the Ad Hoc Committee on
posttraumatic stress disorders. Am. J. Psychiatry Environmental Hypersensitivity Disorders, Ontario, Canada.
142(2):198–202. Vasey, F. (1995). Observation on women with breast implants.
Schottenfeld, R., and Cullen, M. (1986). Recognition of occu- J. Fla. Med. Assoc. 82(5):348–51.
pational induced posttraumatic stress disorders. J. Occ. Von Euler, G., Ogren, S., Li, X., Fuxe, K., and Gustafsson, J.
Med. 28:365–369. (1993). Persistent effects of subchronic toluene exposure on
Schusterman, D., and Dager, S. (1991). Prevention of spatial learning and memory, dopamine-mediated locomotor
psychological disability after occupational respiratory expo- activity and dopamine D2 agonist binding in the rat.
sures. Occup. Med. 6:11–27. Toxicology 77:223–232.
Siegel, S., and Kreutzer, R. (1997). Pavlovian conditioning and Voorhees, R. E. (1998). Information on multiple chemical sensi-
multiple chemical sensitivity. Environ. Health Perspect. tivity. Memorandum from the New Mexico Department of
Suppl. 105:521–526. Health to the Office of the Governor, March 18.
Silverman, K., Evans, S., Stain, E., and Griffiths, R. (1992). Waddell, W. (1993). The science of toxicology and its relevance
Withdrawal syndrome after the double-blind cessation of caf- to MCS. Reg. Toxicol. Pharmacol. 18:13–22.
feine consumption. N. Engl. J. Med. 327(16):1109–1114. Wallace, L., Nelson, C., Highsmith, E., and Dunteman, G.
Simon, G., Katon, W., and Sparks, P. (1990). Allergic to life: psy- (1993). Association of personal and workplace characteristics
chological factors in environmental illness. Am. J. Psychiatry with health, comfort and odor: a survey of 3948 office work-
147:901–906. ers in the building. Indoor Air 3:193–205.
Multiple Chemical Intolerance 547

Waxman, A. (2000). Functional brain imaging in the assessment Webster’s Third New International Dictionary of the English
of multiple chemical sensitivities. In Multiple Chemical Sensi- Language (Unabridged) (1986). Merriam-Webster,
tivity/Idiopathic Environmental Intolerance, Occupational Springfield, MA.
Medicine State of the Art Reviews, (Sparks, P. (Ed.). Hanley Winterbauer, S. (1997). Multiple chemical sensitivity and the
and Belfus, Inc., Philadelphia, pp. 411–414. ADA: taking a clear picture of a blurry object. Employee
Weaver, V. (1996). Medical management of the multiple chemical Relations Law J. 23(2):69–104.
sensitivity patient. Reg. Toxicol. Pharmacol. 24:S111–S115.
26

The Olfactory System and the Nasal Mucosa as Portals

of Entry of Viruses, Drugs, and Other Exogenous Agents
into the Brain

Harriet Baker
Cornell University, White Plains, New York, U.S.A.

Mary Beth Genter

University of Cincinnati, Cincinnati, Ohio, U.S.A.

I. INTRODUCTION circulation from the nasal cavity. The first route occurs via
internalization of xenobiotics into the receptor neurons of
The important role the olfactory system plays in food the olfactory epithelium. Substances are transported into
acquisition and mating behaviors is reflected in the sub- the brain through the axons of olfactory receptor neurons
stantial development of this sensory modality in the major- in a process also called the olfactory nerve pathway by
ity of vertebrate species with, perhaps, the exception of Mathison et al. (1998). In an older publication, McMartin
primates. In view of these life-sustaining roles, the com- et al. (1987) referred to an analogous mechanism in their
mon dogma posits that a system as important to survival as reference to both transcytosis (uptake of the agent into
olfaction must have mechanisms to protect the sensory vesicles with subsequent discharge into interstitial space)
cells within the olfactory receptor epithelium from the and transcellular transport (passage of agents via pores or
materials in inspired air, both toxicants and odorants. Also, carriers across the cell membrane, diffusion across the
a means should exist to prevent access of extrinsic sub- cytoplasm, and transport out of the cell to the glomerular
stances to the central nervous system (CNS). As discussed region of the olfactory bulb). Within the glomerulus, axo-
below (see also Chapter 3), several mechanisms have been dendritic contacts between receptor cells and mitral cells
proposed that would protect the epithelium from toxic serve as one transport pathway. For some xenobiotics,
exposure. Included among these mechanisms are intracel- transneuronal transport has been observed in widespread
lular detoxification of molecules, removal of substances by areas of the brain. Both anterograde and retrograde trans-
ligand-specific binding proteins contained in mucosal port have been reported, the latter perhaps a result of the
secretions, surveillance by immune system cells, as well as large centrifugal afferent innervation of the olfactory
degeneration and replacement of damaged receptor neu- glomerular region where receptor afferent fibers terminate
rons with new cells derived from basal stem cells. Recent (see Chapter 7). Anterograde labeling can then be observed
data indicate, however, that these protective mechanisms in mitral cell terminal fields such as the piriform cortex.
can be overwhelmed, resulting in access of xenobiotics Retrograde transport is also observed in brain regions with
from the nares to the CNS and to the systemic circulation. projections to the olfactory bulb including the horizontal
Figure 1 (Fig. 1; see color plate) indicates three major limb of the diagonal band, substantia nigra, locus ceruleus,
routes of entry for agents to the CNS and systemic and dorsal raphe nucleus.
549
550 Baker and Genter

Figure 1 (A) Sagittal view of the rat brain showing olfactory epithelium (OE) receptor cells and their axonal projections to the olfac-
tory glomeruli (G). In the glomerulus, receptor cell axons contact the dendrites of periglomerular (PG) and mitral (M) cells. Mitral cells
project to the piriform cortex (PC). Centrifugal afferent innervation comes from the horizontal limb of the diagonal band (HLDB), the
substantia nigra (SN), the dorsal raphe (DR) and the locus ceruleus (LC). (B) The olfactory mucosa includes an epithelial cell layer (OE)
and the lamina propria (LP) separated by the basal lamina (BL). The OE contains sustentacular (S), basal (B), and receptor (R) cells.
Receptor cell axons fasciculate to form the olfactory nerve (ON) that crosses the cribriform plate (CP) to enter the CNS. The axon and
nerve are surrounded by a perineural sheath that forms the perineural space (PN). The lamina propria contains mucus-secreting Bowman’s
glands (BG), axons of receptor cells, and numerous blood vessels (BV). Red and green dots depict possible entry pathways through neu-
rons, glands, and blood vessels. (C) Respiratory epithelium consists of columnar ciliated (C), goblet (G), and basal (B) cells and is highly
vascular (BV). (D) H & E stained section illustrating the layers of the OE and LP showing the numerous blood vessels in the lamina
propria. Olfactory marker protein (OMP) immunostained section shows that only mature receptor neurons and their axons, not basal or
sustentacular cells, contain OMP. (Panels B and C adapted from Lewis and Dahl, 1995.) (See color insert.)

Xenobiotics also may gain access to the CNS by two This pathway provides access of intranasally applied
other routes, as indicated in Figure 1. A number of drugs agents to the systemic circulation and thus to the brain.
(see below) and metals, including gold, are hypothesized Another consequence of the extensive vascularity of the
to reach the subarachnoid space by movement through olfactory epithelium is that some viruses (Charles et
perineural spaces, a process also called the olfactory al., 1995; Oliver and Fazakerley, 1998) and a variety
epithelial pathway or paracellular transport (Jackson of organic compounds (Brandt et al., 1990; Brittebo,
et al., 1979; Mathison et al., 1998; McMartin et al., 1988; Feng et al., 1990) are concentrated in the
1987). In fact, many drugs that otherwise would not have olfactory epithelium following systemic application.
access to the brain can be applied intranasally, including Subsequently, there is either transport to the brain or
antiviral drugs for the treatment of human immunodefi- destruction of the epithelium. The substances that have
ciency virus (HIV) infections that have spread to the been reported to be internalized, transported, and accu-
CNS. The third route is a consequence of the highly vas- mulated through the olfactory system are quite varied,
cular nature of the respiratory and olfactory epithelia. including lectins, dyes, solvents, metals, viruses, amino
Exogenous Agent Uptake 551

Table 1 Substances Analyzed for Transport Through Olfactory Receptor Cells to the Brain
Species Application Transneuronal
Substance studied method transport Ref.
Metals
Aluminum lactate Rabbit Intranasal Indirect pathological Perl and Good, 1987
evidence
Aluminum silicate Rabbit Bedding No Hayek and Waite, 1991
Aluminum acetyl- Rat Inhalation Yes Zatta et al., 1993
acetonate
Cadmium Pike Intranasal No Gottofrey and Tjälve, 1991
Cadmium chloride Rat Intranasal No Tjälve and Henriksson, 1999;
Tjälve et al., 1996
Cadmium oxide Rat Aerosol No Hastings and Evans, 1988;
Hastings, 1990
Cobalt Salmon Intranasal Possible Bazer et al., 1987
Gold Squirrel Intranasal Yes DeLorenzo, 1970
Monkey
Rabbit Mucosal No Czerniawska, 1970
Manganese Rat Intranasal Yes Gianutsos et al., 1997; Tjälve and
Henriksson, 1999; Tjälve et al.,
1996;
Mercury Pike Intranasal No Borg-Neczak and Tjälve, 1996.
Mercury Rat Intranasal No Henriksson and Tjälve, 1998
Lectins
WGA-HRP Rat Intranasal Yes Baker and Spencer, 1986;
Broadwell and Balin, 1985;
Balin et al., 1986; Itaya, 1987;
Shipley, 1985; Stewart,
1985; Thorne, 1995
Mouse Intranasal Yes Baker, 1995
Barley Mouse Genetic Yes Horowitz et al., 1999
Concanavilin A Rat Intranasal No Wiley et al., 1984
Viruses
Adeno (recombinant) Rat Intranasal Yes Draghia et al., 1995; Zhao et al.,
1996
Borna disease Rat Intranasal Yes Morales et al., 1988
Equine Herpes Pig Intranasal Yes Narita et al., 2001
Herpes simplex Rat Intranasal Yes Esiri and Tomlinson, 1984;
McLean et al., 1989
Mouse Corneal Yes Stroop et al., 1984
Mouse Facial skin Limited Stroop et al., 1984
Hepatitis Mouse Intranasal Yes Barnett and Perlman, 1993; Lavi
et al., 1988; Perlman et al.,
1990
Rabies Mouse Intranasal Yes Astic, 1993; Lafay et al., 1991
St. Louis encephalitis Hamster Intraperitoneal Yes Monath et al., 1983
Sendai Mouse Intranasal Limited Mori et al., 1995
Semliki forest Mouse Intranasal Age-dependent Oliver and Fazakerley, 1998
Venezuelan equine Mouse Subcutaneous Yes Charles et al., 1995; Ryshikov
et al., 1995
Dyes
Procion yellow Catfish Intranasal No Holl, 1980
Eel Intranasal No Holl, 1981
(continued)
552 Baker and Genter

Table 1 (continued)
Species Application Transneuronal
Substance studied method transport Ref.
Lucifer yellow Lamprey Intranasal No Suzuki, 1984
Frog Intranasal No Suzuki, 1984
Evans blue Mouse Intranasal No Kristensson and Olsson, 1971
Prussian blue Mouse Intranasal No Rake, 1937.
Trypan blue Mouse Intranasal No Seki, 1941
Bullhead Intranasal No Holl, 1965
Amino acids
Leucine Toad Intranasal No Weiss and Holland, 1967
Garfish Intranasal No Gross and Beidler, 1973
Rabbit Intranasal No Land and Shepherd, 1974
Pike Intranasal No Weiss and Buchner, 1988
Alanine Mouse Intranasal No Margolis and Grillo, 1977
Hamster Intranasal No Burd et al., 1982
Taurine Mouse Intranasal No Brittebo and Ericksson, 1995;
Lindquist et al., 1983
Drugs
Cocaine Mouse Intravenous No Brittebo, 1988
Valproic acid Rat Intravenous No Hoeppner, 1990
Miscellaneous
Aflatoxin B1 Rat Intranasal Limited Larsson and Tjälve, 2000
Ameba Mouse Intranasal Yes Jarolim et al., 2000
Horseradish peroxidase Rat Intranasal No Balin et al., 1986; Kristensson
and Olsson, 1971
Mouse Intranasal No Meredith and O’Connell, 1988;
Stewart, 1985
Immunoglobulins Mouse Intranasal No Baker and Maruniak, 1990
Polychlorinated biphenyls Ferret Ambient air ND Apfelbach et al., 1998
PCB
Solvents Mouse Inhalation No Ghantous et al., 1990
ND, not determined.

Table 2 Systemically Administered Agents Associated with Olfactory Mucosal Damage in Rodents
Compound Route Endpoint Ref.
Acetaminophen i.p., oral OMD Jeffrey et al., 1988; Genter,
1998
Alachlor oral Tumors U.S. Environmental Protection
Agency, 1986; Genter et al., 2000
Bromobenzene i.v., i.p. OMD Brittebo et al., 1990
Caffeine-derived N-nitroso oral Tumors Ivankovic et al., 1998
compounds
Carbimazole i.p. OMD Genter, 1998
Coumarin i.p., oral OMD Gu et al., 1997
p-Cresidine oral Tumors National Cancer Institute, 1979
2,6-Dichlorobenzamide i.p. OMD Brittebo et al., 1991
2,6-Dichlorobenzonitrile i.p., i.v., dermal OMD Brandt et al., 1990; Deamer et al.,
1994
2,6-Dichlorothiobenzamide i.p. OMD Brittebo et al., 1991
Dihydropyridines i.p. Protoporphyria Reed et al., 1989
Diethyldithiocarbamate i.p. OMD Deamer and Genter, 1995
Disulfiram i.p. OMD Deamer and Genter, 1995
1,4-Dithiane gavage OMD Schiferstein et al., 1988
Exogenous Agent Uptake 553

Table 2 (continued)
Compound Route Endpoint Ref.
N-bis(2-Hydroxypropyl)nitrosamine s.c. Tumors Koujitani et al., 1999
N-Hydroxy-IDPN i.p. OMD Crofton et al., 1996
IDPN i.p. OMD Genter et al., 1992
Methimazole i.p., oral OMD Genter et al., 1995
3-Methylindole i.p. OMD Turk et al., 1986
Methylsulfonyl-2,6-dichlorobenzene i.p. OMD, metaplasia Bahrami et al., 2000
N-Nitrosodiethylamine i.p. OMD Jensen and Sleight, 1987
NNK s.c. OMD, tumors Belinsky et al., 1987
2-Pentenenitrile i.p. OMD Genter and Crofton,
2000
Phenacetin oral OMD, tumors Bogdanffy et al., 1989
Isaka et al., 1979
Routes of administration: i.p.
intraperitoneal injection; i.v.
intravenous injection; s.c.
subcutaneous. OMD
olfactory mucosal degeneration;
IDPN
3,3ⴕ iminodipropionitrile; NNK
4-(N-methyl-N-nitrosamino)-1-(3-pyridyl)-1-butanone.

Table 3 Drugs Administered Via the Nasal Cavity for CNS and Systemic Effects
Purpose Drug Model Ref.
Drugs administered via the nasal
cavity for CNS effects
Relief of migraine headaches Sumatriptan (Imitrex®) Dog, human Barrow, 1997; Dahlof et al., 1998
BMS-181885 Monkey Srinivas et al., 1998
Ergotamine derivatives Human Gallagher, 1996; Lipton, 1997;
Rabbit Ziegler et al., 1994
Marttin et al., 1997
Butorphanol Human Melanson et al., 1997
Lidocaine Human Lane, 1996; Mills and Scoggin,
1997; Sachs, 1996
Capsaicin Human Levy, 1995
Sedation Midazolam Human Björkman et al., 1997; Louon and
Reddy, 1994
Diazepam Rabbit Bechgaard et al., 1997
Seizure control Midazolam Human Jeannet et al., 1999; Wallace,
1997
Analgesia Oxycodone Human Takala et al., 1997
Oxymorphone Dog Hussain and Aungst, 1997
Acetylsalicylic acid Rat Hussain et al., 1992
Cancer chemotherapy 5-Fluorouracil Rat Sakane et al., 1999
Treatment/prevention of Lipophilic analog of vaso- Rat Gozes et al., 1996, 1997
neurodegenerative diseases active intestinal peptide
([St-Nle17]VIP)
Dextromethorphan Rat Char et al., 1992
Neostigmine Human DiCostanzo et al., 1993
Nerve growth factor Rat Frey et al., 1997
Synthetic acetylcholine Mouse Karachunski et al., 1998
receptor epitopes
Treatment/prevention of Zidovudine (AZT) Rat Seki et al., 1994
CNS infections D4T Rat Yajima et al., 1998
Cephalexin Rat Sakane et al., 1991
Increased brain activity/ Arginine-vasopressin Human Perras et al., 1997;
memory improvement Pietrowsky et al., 1996

(continued)
554 Baker and Genter

Table 3 (continued)
Purpose Drug Model Ref.
Appetite control Cholecystokinin Dog Pierson et al., 1997
(CCK) agonists
Drugs administered via the nasal
cavity for systemic effects
Diabetes Insulin Human, rabbit Fernandez-Urrusuno et al., 1999;
Hilsted et al., 1995; Valensi et al.,
1996
Smoking cessation Nicotine nasal sprays Human Gourlay and Benowitz, 1997; Jones
et al., 1998; Schneider et al., 1995
Vaccines Cold-adapted influenza Liu, 1998; Maassab and DeBorde,
viruses. 1985
Outer membrane Human Haneberg et al., 1998
vesicles (N. meningitides).
Diptheria-tetanus boosters Human Aggerbeck et al., 1997
Normalization of Hydroxycobalamine Human Slot et al., 1997; Swain, 1995
plasma vitamin B12
Motion sickness Scopolamine Human Putcha et al., 1996
Promethazine Dog Ramanathan et al., 1998
Nocturia Desmopressin Human Butler et al., 1998; Kallas et al.,
1999
Contraception Gonadotrophin-releasing Human Bergquist et al., 1979
hormone agonist
Endometriosis, controlled Nafarelin, buserelin Human Dantas et al., 1994; Jacobson et al.,
ovarian hyperstimulation 1994; Lemay et al., 1984
Prostate cancer, Buserelin Human Matsumiya et al., 1998; Tolis et al.,
oligoasthenozoospermia 1983
Prevention of bone resorption Calcitonin Human Silverman, 1997
Enhanced lactation Oxytocin Human Ruis et al., 1981
Anemia Erythropoietin Rat Shimoda et al., 1995

acids, and drugs. Table 1, a compilation of the range of istered by this route (see Table 3) for either direct deliv-
substances transported, also indicates those for which ery to the CNS or systemic access.
the evidence suggests transneuronal transport. The con- First to be discussed will be the specific substances and
sequences of such transport have only begun to be pathogens transported through olfactory receptor cells.
assessed (see below). One hypothesis suggests that the Second, the proposed mechanisms of internalization and
olfactory system may be a route by which some, as yet transport will be reviewed with special reference as to why
undefined, pathogen or toxin reaches the brain to pro- transneuronal transport occurs only in limited instances.
duce degenerative disorders such as Alzheimer’s disease Intranasal application of drugs will be discussed as a
(Baker and Margolis, 1986; Crapper McLachlan, 1986) means of administering a number of agents for direct entry
and Parkinson’s disease (Tjälve et al., 1996), as well as to the brain. Finally, the consequences of xenobiotic access
learning and behavioral abnormalities (Becker, 1995) to the brain will be assessed from the disease perspective.
(see Chapter 24). This internalization may have effects
reaching through the food chain as some fish apparently
II. NATURE OF SUBSTANCES TRANSPORTED
can accumulate materials such as metals found in water
by this route (Bazer et al., 1987). The significance of A. Metals
accumulation in the food chain is apparent in Table 2, a
list of the compounds that when administered systemi- Aluminum, cadmium, cobalt, gold, and manganese are
cally either destroy or produce tumors in the olfactory among the metals internalized in the olfactory receptor
epithelium. There are, however, benefits to the nasal epithelium and transported either directly to the brain via
route of application since many drugs have been admin- olfactory axons or perhaps, as reviewed by Jackson et al.
Exogenous Agent Uptake 555

(1979), through either the subarachnoid space or perineu- et al., 1996). In rats, exposure to both cadmium oxide and
rally. The evidence, both direct and circumstantial, for chloride, in aerosol form, resulted in accumulation of cad-
their transneuronal transport to brain has supported the mium in the olfactory bulb (Evans and Hastings, 1992;
controversial hypothesis that some of these metals, for Hastings, 1990; Hastings and Evans, 1988). Cadmium oxide
example, aluminum, may be etiologically significant in did not produce anosmia (Hastings, 1990; Hastings and
pathological syndromes such as Alzheimer’s and Evans, 1988). Hastings (1990) suggested that the lack of a
Parkinson’s disease (Crapper McLachlan, 1986; Roberts, functional deficit in rodents following cadmium exposure,
1986; Tjälve et al., 1996). Described below is the method in view of the purported effects in humans, might reflect a
of application and evidence for uptake and transport of requirement for protracted exposure to produce anosmia.
each of the metals. (Table 1)
3. Cobalt
1. Aluminum Intranasal irrigation with cobalt lysine in king salmon fry
(Oncorhynchus tshawytscha, Walbaum) resulted in trans-
Several studies have shown that aluminum can reach the port of the conjugate to the olfactory bulb as well as to both
central nervous system after intranasal application (Hayek ventral and lateral regions of the ventral telencephalon
and Waite, 1991; Perl and Good, 1987; Zatta et al., 1993). (Bazer et al., 1987). The latter projections may represent
Aluminum silicate, presented as a powder in the bedding, transneuronal transport of the metal conjugate (Bazer
was enriched specifically in the rabbit olfactory epithelium et al., 1987), as no direct epithelial connections to this
and olfactory bulb, but not in cerebellum (Hayek and Waite, brain region have previously been reported. There is evi-
1991). Perl and Good (1987) demonstrated that, following dence that cobalt is able to cross neuronal membranes
intranasal application of aluminum lactate, granulomas (Fredman and Jahan-Pawar, 1980).
were observed in the olfactory bulb and cerebral cortex.
The pathological changes observed in brain suggested that
aluminum reaches these brain regions by transport through 4. Gold
olfactory neurons. Inhalation of aluminum acetylacetonate Electron microscopic studies demonstrated that colloidal
resulted in widespread dissemination of the metal complex gold was internalized by olfactory receptor cells within 15
in diverse brain regions (Zatta et al., 1993). Aluminum was minutes of intranasal application in squirrel monkeys (De
also observed in the neurofibrillary tangles associated with Lorenzo, 1970). Gold was thought to be incorporated into
Alzheimer’s disease (for review see Crapper McLachlan, receptor cells by the pinocytotic vacuoles normally
1986; Good et al., 1992), and on intraventricular adminis- observed in the olfactory rod region, more commonly
tration was shown to produce neurofibrillary degeneration referred to as the dendritic knob (De Lorenzo, 1970).
in rabbit and rat brain (Kowall et al., 1989; Shigematsu and Subsequently, gold particles were observed in the cyto-
McGeer, 1992). Additionally, the demonstration that brain plasm of olfactory axons and within 1 hour in the olfactory
regions having direct connections to the olfactory system glomerulus. Evidence was also presented for transneuronal
are primarily affected in Alzheimer’s disease has suggested transport of the gold particles from receptor cell axon ter-
the provocative hypothesis that transport of aluminum via minals within the glomeruli to the dendrites of mitral cells
the olfactory system may play a role in this degenerative (De Lorenzo, 1970). Within the olfactory bulb the particles
syndrome (Roberts, 1986). were preferentially associated with mitrochondria. The
rate of appearance in the glomeruli was consistent with the
2. Cadmium movement of the gold by fast axonal transport (estimated
in these experiments at about 2.5 mm/hr). Czerniawska
Cadmium transport from the olfactory epithelium to the
(1970) also demonstrated uptake of gold into the cere-
olfactory bulb has been demonstrated in both fishes and
brospinal fluid, especially in the region of the cribriform
rodents. A number of reports has suggested an association of
plate and the olfactory bulbs, following either mucous
cadmium exposure in humans with anosmia, and by infer-
membrane or intravenous administration of the metal.
ence, accumulation and transport of the metal (Hastings,
1990) (see Chapter 27). In the pike (Esox lucius), cadmium
5. Manganese
appeared to move at a velocity consistent with processing by
the fast axonal transport system (Gottofrey and Tjälve, A recent series of papers documents both possible routes
1991). Cadmium accumulated in anterior regions of the of entry of manganese into the CNS and the neurological
olfactory bulb, but did not appear to move transsynaptically consequences of exposure. Definitive evidence now
to mitral cell dendrites (Tjälve and Henriksson, 1999; Tjälve exists in rats for transport of manganese from the
556 Baker and Genter

olfactory epithelium into the brain (Tjälve and 1984). A number of investigators (Baker and Spencer,
Henriksson, 1999; Tjälve et al., 1996). In contrast to 1986; Balin et al., 1986; Broadwell and Balin, 1985; Itaya,
other metals applied by intranasal irrigation (e.g., nickel 1987; Shipley, 1985; Stewart, 1985; Thorne et al., 1995)
and cadmium), manganese was transported to all regions have shown that WGA conjugated to horseradish peroxi-
of the CNS including the spinal cord and could still be dase (WGA-HRP) is internalized by olfactory receptor
detected 12 weeks after initial exposure. Manganese may cells following binding to surface receptors and trans-
also reach the CNS when administered by parenteral ported to the olfactory bulb. Uptake and transport occurred
routes but in much lower concentrations. Manganese has primarily ipsilaterally following intranasal application
been shown to produce a neurological syndrome with (Baker and Spencer, 1986; Shipley, 1985). Both antero-
features of Parkinson’s disease (PD). Occupational expo- grade and retrograde transneuronal transport have been
sure occurs in the mining industry, in steel manufactur- observed from the olfactory bulb. Labeled primary olfac-
ing, and in welding and is primarily through inhalation of tory axons containing WGA-HRP reaction product were
metal-containing dusts and fumes (Gorell et al., 1999; observed in the nerve and glomerular layers of the main
Tjälve and Henriksson, 1999). Taken together with the olfactory bulb within 6 hours of intranasal irrigation
widespread CNS distribution following nasal exposure, (Broadwell and Balin, 1985; Itaya, 1987; Shipley, 1985).
the association with PD is consistent with specific toxic- At longer survival times (4–7 days) labeled neurons and
ity towards dopamine neurons. One proposed mechanism terminals have been reported in a number of brain regions
(Tjälve and Henriksson, 1999) suggests that the metal known to either receive input from or send axons to the
may be involved in the formation of free radicals in the olfactory bulb. Anterograde and retrograde label were
presence of catecholamines. observed in the anterior olfactory nucleus (Itaya, 1987;
Shipley, 1985). The piriform and entorhinal cortices, as
6. Other Metals well as the olfactory tubercle, also contained significant
label (Baker and Spencer, 1986; Itaya, 1987; Shipley,
Several studies investigated the transport of nickel and 1985). Brain regions that send centrifugal afferent innerva-
mercury presented intranasally. Inorganic mercury exhib- tion primarily to the glomerular region of the olfactory
ited limited transneuronal transport restricted to the bulb also contained retrogradely labeled cells. These
olfactory bulb following nasal application in rats regions included the anterior olfactory nucleus, the hori-
(Henriksson and Tjälve, 1998). Long-term occupational zontal limb of the diagonal band, the raphe nucleus, and, in
exposure was not associated with increased risk for some studies, the locus ceruleus (Baker and Spencer, 1986;
Parkinson’s disease (Gorell et al., 1999). Transneuronal Shipley, 1985). These data indicate the extent of transport
transport of nickel was studied in rats and pike. The that occurs from the olfactory epithelium to those brain
transport rate for this metal was 20 times slower than for regions with connections to the olfactory bulb. However,
cadmium and manganese. Nickel could be demonstrated the significance, generality, and specificity of lectin trans-
in several forebrain regions, suggesting that the metal is port remain to be delineated.
transported slowly transsynaptically (Tallkvist et al.,
1998; Tjälve and Henriksson, 1999). Nickel exposure has C. Viruses
been associated with cell loss in the olfactory epithelium
but did not alter olfactory function as indicated by mea- From a historical perspective, a neural route of viral inva-
surement of either odor threshold or discrimination sion of the CNS was postulated as a mechanism for the
(Evans et al., 1995). development of rabies as early as the eighteenth century
and demonstrated in the late nineteenth century (see
B. Lectins Stroop, 1995, for review and references). The nasal route
as a site of viral entry to the CNS was established during
Although native and derivatized lectins have been utilized the 1920s and 1930s (Clark, 1929; Hurst, 1936).
by a number of investigators to delineate neuronal connec- Poliomyelitis virus was subsequently shown to reach the
tivity, especially in the visual system (Itaya and Van olfactory bulbs through the olfactory neurons following
Hausen, 1982; Spencer et al., 1982), transport studies of intranasal application in primates (Bodian and Howe,
these oligosaccharide-specific proteins in the olfactory 1941a,b). Reports that monkeys could be protected against
system have been limited primarily to wheatgerm agglu- intranasal inoculation with poliomyelitis virus by nasal
tinin (WGA). A brief report showed that concanavalin A lavage with solutions of alum, picric acid, or zinc sulfate
was also transported in the olfactory system, but suggested (Armstrong and Harrison, 1935; Schultz and Gebhardt,
that transneuronal transport did not occur (Wiley et al., 1936) led to prophylactic treatment of children with nasal
Exogenous Agent Uptake 557

zinc sulfate during poliomyelitis epidemics (Peet et al., D. Dyes

1937; Schultz and Gebhardt, 1937; Tisdall et al., 1937). An
1. Procion and Lucifer Dyes
undesirable consequence of these treatments was long-
lasting anosmia in many individuals (Tisdall et al., 1938), Several dyes, including procion, lucifer, and Evans blue,
presumably produced by irreversible destruction of the were internalized into olfactory epithelial cells. In those
olfactory sensory epithelium. species tested, procion and lucifer dyes specifically labeled
Subsequently, transport to the central nervous system olfactory receptor neurons. Two procion dyes, brilliant
has been reported for several types of viruses following yellow M4RAN and yellow M4R, labeled the receptor
intranasal inoculation, including herpes simplex (Esiri and neurons of the catfish, Ictalurus nebulosus (Holl, 1980).
Tomlinson, 1984; McLean et al., 1989; Stroop et al., Similar intravital staining was observed in olfactory
1984), murine coronavirus (Barnett et al., 1993; Perlman et receptor neurons of the eel, Anguilla anguilla (Holl, 1981).
al., 1988, 1989), Borna disease virus (Morales et al., In both the lamprey, Entosphenus, and the grass frog,
1988), Nile virus (Nir et al., 1965), Venezuelan equine Rana, procion yellow MX4R and lucifer yellow VS exhib-
encephalitis virus (Charles et al., 1995), vesicular stomati- ited internalization specifically by olfactory receptor neu-
tis virus (Huneycutt et al., 1994), Semliki virus (Oliver and rons in contrast to either supporting cells or the glandular
Fazakerley, 1998), Sendai virus (Mori et al., 1995), and components of the epithelium (Suzuki, 1984). Since the
rabies virus (Astic et al., 1993; Lafay et al., 1991). stained receptor neurons retained their dye through
Olfactory bulb ablation prevented the spread of the dissociation procedures, Suzuki (1984) suggested that
neurotropic coronavirus mouse hepatitis virus into the these fluorescent dyes may be useful cell markers for iden-
brain (Perlman et al., 1990). Two neurotropic viruses, her- tifying viable receptor cells in dissociated epithelial prepa-
pes simplex virus (HSV) type 1 and mouse hepatitis virus, rations. Interestingly, not all dyes showed specific uptake.
spread along different neural pathways following Lucifer yellow CH stained the receptor epithelium non-
intranasal inoculation, suggesting that uptake in specific specifically, and procion red H3B at high concentrations
neurotransmitter systems influences viral distribution in resulted in dye aggregations on the surface of the olfactory
the CNS (Barnett et al., 1993). Selective lesions of neural epithelium. The basis for the difference in labeling
pathways also contributed to spatial learning deficits between the dyes was thought to reflect binding to specific
(McLean et al., 1993). Learning and behavioral deficits membranous molecules.
have recently been associated with HSV-1 brain infection,
presumably by the olfactory nerve route (Becker, 1995).
2. Evans Blue
Olfactory labeling was observed in several instances where
the viruses were applied at nonolfactory sites, including Inhalation of a complex of Evans blue dye with albumin
intraperitoneal (Monath et al., 1983) and subcutaneous resulted in strong red fluorescence within numerous cells of
routes (Charles et al., 1995). Virus titers were observed the olfactory epithelium of mice (Kristensson and Olsson,
first in the olfactory epithelium (4 days) and then in the 1971). After 24 hours, the filia olfactoria of the mice exhib-
olfactory bulb (5 days) followed by the rest of the brain. ited fluorescent label. Although labeling was found in the
Olfactory labeling was observed following corneal inocu- olfactory bulbs, especially in young mice, there was no evi-
lation and intradermal application in the mouse (Stroop et dence of transneuronal transport to secondary neurons such
al., 1984). In the latter instances the virus may have as mitral cells. Evidence for label within the submucosal
reached the olfactory epithelium through tears entering the area and leptomeninges surrounding the olfactory bulb sug-
naris or nasopharynx and then entering the nasal receptor gested that some of the dye could reach the CNS by other
epithelium. Applying sensitive immunohistochemical routes in addition to transport through the olfactory nerves
techniques, herpes simplex virus produced Golgi-like, (Kristensson and Olsson, 1971).
transneuronal labeling in the CNS (Blessing et al., 1994;
McLean et al., 1989), and the authors suggested that virus
3. Trypan Blue
injection may be useful for tracing neuronal pathways.
Many recent reports demonstrate that pseudorabies virus Trypan blue dye was studied in both the brown bullhead,
also produced significant transneuronal transport and can Ictalurus natalis (Holl, 1965), and the mouse (Seki, 1941).
be used to label CNS pathways (Graf et al., 2002; Sams et The investigations were undertaken to demonstrate the
al., 1995; Strack and Loewy, 1990; Ugolini, 1995). These time course and conditions for specific uptake of the dye
and many other studies demonstrate that the olfactory sys- into olfactory receptor cells and not into other cell types in
tem may serve as an efficient route of entry of viruses into the olfactory epithelium. Transneuronal transport was not
the brain. investigated.
558 Baker and Genter

E. Amino Acids Tritiated -alanine exhibited a pattern of transport con-

1. Leucine sistent with a less specific internalization in the olfactory
epithelium and incorporation into a number of proteins
The transport of leucine in the olfactory system has been after intranasal administration. Both sensory and nonsen-
studied by a number of investigators (for review, see Weiss sory cells and the lamina propria were labeled. Peak label-
and Buchner, 1988). For the most part, uptake, metabolism, ing in the olfactory bulb was less intense and delayed as
and transport of the amino acid were used to characterize compared to -alanine, suggesting that the protein was
the kinetics of protein movement in normal and regenerat- carried by slow axonal transport. Trans-neuronal transport
ing nerve fibers. The olfactory system was an ideal model was not observed for either - or -alanine (Burd et al.,
since in many fish species the nerve is long and regenerates 1982).
readily. One of the pioneering studies used histochemical
techniques to demonstrate axoplasmic transport of labeled 3. Taurine
proteins following intranasal application of tritiated leucine
in the toad, Bufo americanus (Weiss and Holland, 1967). Taurine accumulated specifically in the olfactory bulb
Gross and Beidler (1973) studied fast axonal transport in after intravenous injection (Lindquist et al., 1983). The
C-fibers of the garfish, Lepisosteus osseus. Experiments ability of the olfactory system to concentrate this amino
performed in the pike, Esox lucius, demonstrated similar acid was even more pronounced following intranasal
dynamics of fast axonal transport (Gross and Kreutzberg, irrigation. An intra-axonal transport mechanism, as
1978; Weiss et al., 1978). The long olfactory nerve of this opposed to a perineural one, was suggested by the
species was used to study temperature dependence of restricted accumulation in the bulb ipsilateral to the
axonal transport, especially slow axonal transport in normal application.
and regenerating olfactory nerve (Cancalon, 1979a, b,
1988; Cancalon et al., 1988). The relative uniformity of F. Drugs
transport in both the antero- and retrograde direction was
shown in the pike olfactory nerve (Weiss and Buchner, Valproic acid (Hoeppner, 1990), cocaine (Brittebo,
1988). Transport of tritiated leucine was also utilized to 1988), and a number of other drugs (Dahl and Hadley,
analyze the projections of olfactory receptor neurons in the 1991) were concentrated in the olfactory epithelium fol-
adult rabbit (Land and Shepherd, 1974). lowing intravenous administration. The relative
In all studies the majority of the leucine was incorpo- enrichment of cocaine in the epithelium was maintained
rated into amino acids following a short delay. Transport for at least 24 hours. The mechanism for the concentra-
was monitored either autoradiographically or biochemi- tion of these drugs in the epithelium is unknown. The
cally within dissected nerve segments. Although both slow presence of D2 dopamine receptors in the receptor
(Cancalon, 1979 a,b) and fast axonal transport (Gross and epithelium (Guthrie et al., 1991; Shipley et al., 1991)
Beidler, 1973, 1975) were demonstrable using these tech- suggests that cocaine, a dopamine uptake blocker, binds
niques, transneuronal transport was not observed under the to a specific receptor site in the epithelium. The presence
experimental conditions utilized (Land and Shepherd, of numerous drug-metabolizing enzymes (Dahl
1974; Weiss and Holland, 1967). and Hadley, 1991) (see also Chapter 3) also may con-
tribute to the binding and metabolism of drugs in the
2. Alanine olfactory epithelium. Transport has not been studied fol-
lowing intranasal administration of the drugs.
In the olfactory system the dipeptide carnosine (-
alanyl-L-histidine) may be either a neurotransmitter or a
G. Miscellaneous Substances
neuromodulator of the primary olfactory neurons.
Therefore, -alanine presented at the mouse external
1. Horseradish Peroxidase
naris was incorporated specifically into carnosine
(Margolis and Grillo, 1977). Subsequent experiments in The transport of HRP differed from that observed for the
hamsters demonstrated that the radioactivity associated conjugate, WGA-HRP. HRP, like WGA-HRP, was inter-
with the -alanine was converted to carnosine ( 82% nalized by receptor cells and labeled the olfactory
of the radioactivity was in the carnosine fraction at 24 hr) glomeruli (Balin et al., 1986; Kristensson and Olsson,
and transported to the olfactory bulb (Burd et al., 1982). 1971; Meredith and O’Connell, 1988; Stewart, 1985).
The rate of appearance in the olfactory bulb was consis- However, transneuronal transport was not observed.
tent with movement of the -alanine by the fast axonal Movement also occurred through the intercellular junc-
transport system. tions to label the leptomeninges (Balin et al., 1986). HRP
Exogenous Agent Uptake 559

uptake was not observed in the accessory olfactory system there is partial to nearly full recovery of olfactory mucosa
unless a large dose of epinephrine was administered to that has undergone chemically induced degeneration (see
activate the vomeronasal organ pumping mechanism chemicals labeled “OMD,” referring to those causing
(Meredith and O’Connell, 1988). Mechanistic considera- olfactory mucosal degeneration in Table 2). However, a
tions discussed below may underlie the differences toxicant that causes persistent metaplasia of the olfactory
between WGA-HRP and HRP. mucosa has recently been described (Bahrami et al., 2000).
A number of drugs (e.g., phenacetin), industrial and agri-
2. Immunoglobulins cultural chemicals (e.g., p-cresidine, alachlor), and envi-
ronmental pollutants (e.g., NNK) are associated with the
Only one report exists documenting the uptake of
development of nasal cancer in rodents (Table 2). It is
immunoglobulins (Baker and Maruniak, 1990). The study
interesting to note that humans are exposed to low levels of
demonstrated that mouse olfactory receptor neurons inter-
many of these, or related, agents every day, but none has
nalized IgG molecules from a number of species. Twenty-
been linked definitively to human nasal cancer; the most
four hours after intranasal application, receptor neurons,
widely recognized risk factors for human nasal cancer are
microvillar cells, and nerve bundles in the olfactory
exposure to wood dust or nickel, both via the inhalation
mucosa contained IgG, but transport was not observed into
route (Barceloux, 1999; Hayes et al., 1986).
the olfactory bulbs.

3. Solvents III. MECHANISMS OF TRANSPORT

Uptake and metabolism were investigated for several sol- A. Uptake
vents, including benzene, toluene, xylene, and styrene
(Ghantous et al., 1990). All the aromatic hydrocarbons The preceding discussion demonstrates that olfactory
were found in the olfactory epithelium, but only toluene, receptor cells can internalize many types of substances.
xylene, and styrene were found in the olfactory bulb. The issue is how and why these cells should internalize a
Transport of the three solvents in olfactory receptor cells wide range of xenobiotics. The answer to both these ques-
was thought to require conversion to either their aromatic tions may be the same. Since the olfactory mucosa comes
acids or their conjugates. In support of this hypothesis, in contact with many substances in inspired air, a means of
benzene does not exhibit biotransformation to aromatic either elimination or detoxification has likely evolved as a
acids and was not transported (Ghantous et al., 1990). The protective mechanism. In fact, the olfactory epithelium
cytochrome P-450 enzymes necessary for these biotrans- contains high concentrations of a wide variety of xenobi-
formations are known to be present in the olfactory epithe- otic-metabolizing enzymes, even when compared to levels
lium (Dahl, 1988; Dahl and Hadley, 1991). in the liver (Dahl, 1988; Dahl and Hadley, 1991; Thornton-
Manning and Dahl, 1997). A number of enzymes have
H. Olfactory Toxicants Delivered via the been found in the olfactory epithelium, including numer-
Bloodstream ous isozymes of P450 monooxygenases, aldehyde dehy-
drogenases, alcohol dehydrogenase, carboxyesterases,
Multiple xenobiotics that are systemically administered epoxide hydrolases, UDP-glucuronyl transferase, glu-
(i.e., by noninhalation routes of exposure) have toxic tathione S-transferase, and rhodanese (Dahl and Hadley,
effects on the olfactory mucosa of experimental animals 1991) (see also Chapter 3). For some of these enzymes,
(Table 2). For example, oral administration of the anal- isozymes have been found that are specific to the olfactory
gesic acetaminophen or the antihyperthyroid drugs mucosa (Dahl, 1989; Jones and Reed, 1989; Lazard et al.,
methimazole or carbimazole to rodents causes degenera- 1991; Nef et al., 1989; Zupko et al., 1991). Thus, uptake
tion of the olfactory mucosa (Genter, 1998; Genter et al., into receptor, supporting, and glandular cells of xenobi-
1995, 1998; Jeffery and Haschek, 1988). The antihyper- otics, including odorants, may serve a detoxification as
thyroid drugs have been associated with impaired olfaction well as a clearance function. These enzymes also may play
in humans (Schiffman, 1983). The herbicide dichlobenil a role in defining the characteristic odor of a substance
(2,6-dichlorobenzonitrile) is particularly interesting from through the production of active metabolites, thereby cre-
the point of view that intravenous (i.v.), intraperitoneal ating new odorants (Dahl, 1988; Dahl and Hadley, 1983;
(i.p.), or dermal exposure results in olfactory mucosal Getchell et al., 1984).
degeneration, with dermal exposure representing an occu- The specificity of the uptake mechanism has not been
pationally relevant route of exposure (Bergman et al., established. For WGA-HRP, receptor-mediated processes
2002; Brandt et al., 1990; Deamer et al., 1994). In general, have been suggested, since the conjugate, which is known
560 Baker and Genter

to bind to surface membrane oligosaccharides, is found both in vivo and in vitro techniques (Allen and Akeson,
associated with vesicles (Baker and Spencer, 1986; 1985; Barber, 1989; Foster et al., 1991; Key and Akeson,
Broadwell and Balin, 1985). Unconjugated HRP, on the 1990; Key and Giorgi, 1986 a,b; Lundh et al., 1989; Mori
other hand, is thought to be internalized via bulk endocy- et al., 1985; Polak et al., 1989; Shirley et al., 1983).
tosis (Broadwell and Balin, 1985; Kristensson and Olsson, Uptake of anti IgG antibodies into receptor cells has been
1971). Surprisingly, one study reported that some prepara- reported (Baker and Maruniak, 1990; Baker and Spencer,
tions of WGA-HRP were not transneuronally transported, 1986). However, immunoreactivity for constituents of the
suggesting that the method of HRP conjugation may alter immune system was not found in receptor cells, but pri-
the ability of WGA to bind with specific membrane recep- marily in Bowman’s glands, the mucociliary complex,
tors (Russell et al., 1991). The association of gold particles lymphocytes clustered near the basement membrane and
with vesicles suggested that uptake of gold occurred sub- scattered in the lamina propria, as well as in mast cells of
sequent to the formation of vesicles by pinocytosis (De the human olfactory mucosa (Mellert et al., 1992). A sim-
Lorenzo, 1970). Demonstration of a vesicular association ilar distribution of the secretory immune system compo-
of several intranasally applied tracers indicated that the nents was observed in salamanders and rats (Getchell and
olfactory epithelium possesses a well-developed system of Getchell, 1991). Monoclonal antibodies also recognize
endocytic vesicles, perhaps as a result of a rapid rate of different cell types within the olfactory mucosa
membrane turnover (Bannister and Dodson, 1992). These (Hempstead and Morgan, 1985), suggesting differences
authors suggested that materials may be trapped during in cell surface constituents that might direct differential
normal membrane processing. The internalization of mate- binding to olfactory receptor as opposed to either sup-
rials into the olfactory receptor cells thus may take place porting or glandular cells.
by either receptor- or non–receptor-mediated processes. In Finally, odorant-binding proteins synthesized by
the former case, internalization may occur at low xenobi- mucus-secreting glands of the olfactory and respiratory
otic concentrations, and in the latter, high concentrations mucosa have been demonstrated (Baldaccini et al., 1986;
may be necessary. Dal Monte et al., 1991; Pevsner et al., 1986). The binding
Some evidence exists for the presence of specific cell proteins may be necessary for either the transport of odor-
surface receptors in the olfactory system. A superfamily ants to olfactory receptors or their clearance after signal
of putative odorant receptors has been cloned (Buck and transduction. The odorant specificity of these proteins sug-
Axel, 1991). The large number of these receptors indi- gests that a similar binding specificity on the receptor cells
cates the presence of cell type–specific surface molecules may underlie ligand (either odorant or other xenobiotic)
that interact with odorants (see Chapter 4). The binding uptake into receptor, supporting, and glandular cells.
of odorants to the receptor proteins may result in their Transfer between these compartments also cannot be ruled
internalization. Specific transport of odorant receptor out, especially since many of the degradative enzymes
mRNA to the olfactory bulb was used to demonstrate that described above are found in glandular and supporting
cells expressing a specific odorant receptor project to a cells and not receptor cells (Dahl and Hadley, 1991;
single glomerulus (Ressler et al., 1993, 1994; Sullivan et Thornton-Manning and Dahl, 1997).
al., 1995; Vassar et al., 1994). Olfactory marker protein
mRNA shows similar transport to the glomerular layer B. Transneuronal Transport
but no transneuronal transport (Wensley et al., 1995).
Interestingly, transneuronal transport was demonstrated Once internalized, transport in the olfactory receptor cells
for barley lectin (a close relative of WGA) synthesized in can occur by either the slow or rapid transport systems
olfactory receptor cells of a transgenic mouse expressing characteristic of neuronal systems (Weiss and Buchner,
the lectin under control of the olfactory marker protein 1988). The distinction between molecules transported only
gene promoter (Horowitz et al., 1999). The function of to the olfactory bulb and those that are transneuronally
the intra-axonal mRNA is unknown, but the ability of transported may lie in the nature of their internalization
olfactory receptor neurons to transport numerous other and thus the organelles with which the molecules are asso-
molecules may be a consequence of this endogenous ciated. For example, WGA-HRP, which is internalized by
transport capacity. Also, the D2 dopamine receptors receptor-mediated endocytosis, is transported intra-axon-
reported in olfactory receptor neurons may play a role in ally in tubulovesicular profiles (Baker and Spencer, 1986)
the uptake of molecules that bind to this receptor and transneuronally following processing in the transmost
(Guthrie et al., 1991; Mansour et al., 1990; Shipley et al., Golgi saccule (Baker and Spencer, 1986; Broadwell and
1991). Glycoproteins, as indicated by lectin binding, also Balin, 1985). The latter compartment processes vesicles
have been demonstrated on olfactory receptor cells using destined for axonal terminals (Broadwell and Balin, 1985).
Exogenous Agent Uptake 561

Gold is vesicle-associated and is transneuronally trans- route of administration can be particularly useful for drugs
ported (De Lorenzo, 1970). Viruses are transported in that are readily degraded in the gastrointestinal tract (e.g.
axonal vacuoles within receptor cells and appear to move small, readily digestible peptides) or drugs that are exten-
transsynapically to invade other brain regions synaptically sively inactivated by liver metabolism following oral
connected to the olfactory bulb (Lavi et al., 1988; Monath ingestion. This is not to say that the nasal respiratory and
et al., 1983). The relative lack of glial labeling for WGA- olfactory epithelia are metabolically inert; in fact, there is
HRP and some viruses (Baker and Spencer, 1986; Lavi a vast body of literature documenting the presence of mul-
et al., 1988) indicates limited release into the extracellular tiple Phase I and Phase II metabolic enzymes in rodent,
space and accounts for the restricted transfer to olfactory- dog, and human nasal mucosa (Dahl and Hadley, 1991;
related brain regions. In contrast, HRP, which does not Gervasi et al., 1991; Sarkar, 1992; Thornton-Manning and
exhibit transneuronal transport, is taken up by bulk endo- Dahl, 1997).
cytosis and is not processed through the Golgi saccule. Of the three major routes of entry of drugs from the
Leucine, which also does not appear to be transported nasal cavity and into the brain or systemic circulation, the
transneuronally (Land and Shepherd, 1974; Weiss and first, transneuronal transport, has been dealt with to this
Holland, 1967), is primarily processed into cellular protein point in this chapter. Another putative mechanism for
components (Weiss and Buchner, 1988). Taken together transport of agents from the nasal cavity and into the CNS
these data suggest that transneuronal transport requires has been termed the olfactory epithelial pathway
receptor-mediated uptake into the olfactory sensory cells, (Mathison et al., 1998) or paracellular transport
followed by vesicular transport of unmodified materials (McMartin et al., 1987). Regardless of the designation, the
through the axon, with subsequent release and uptake in concept is that molecules access the CNS from the nasal
association with synaptic specializations. cavity via supporting cells, cell-to-cell junctions, and/or
spaces between cells. A great deal of research effort has
gone into our understanding of the factors that are impor-
C. Specificity of CNS Transport tant in this process. For example, molecular weight,
lipophilicity, and ionization state/pH of the agent are all
As mentioned above, transport from the olfactory bulb to variables that appear to contribute to the ability of agents
other brain regions occurs along specific pathways. The to access the central nervous system from the nasal cavity
ability of centrifugal afferents to support specific transport (McMartin et al., 1987; Sakane et al., 1991a, 1994;
is suggested by data demonstrating selectivity of retro- Shimoda et al., 1995). The concentration of sulfonamides
grade axonal transport of radioactive transmitters from the in the cerebrospinal fluid (CSF) of the rat resulting from
olfactory bulb to their nuclei of origin (Araneda et al., intranasal administration was demonstrated to increase
1983; Bonnet-Font and Bobillier, 1990; Watanabe and with the lipophilicity of the sulfonamide derivative
Kawana, 1984). Intrabulbar application of radioactive (Sakane et al., 1991a). In contrast, lipophilicity did not
serotonin or norepinephrine results in specific labeling, enhance the absorption of a series of peptides from the
respectively, of the dorsal raphe nucleus and the locus nasal cavity (Donnelly et al., 1997; Sakane et al., 1991).
coeruleus (Araneda et al., 1983; Bonnet-Font and Using fluorescein isothiocyanate–labeled dextran with
Bobillier, 1990). Similarly, the receptor-mediated retro- molecular weights ranging from 4.4 to 40 kDa, it was
grade transfer of nerve growth factor from the olfactory determined that drugs with a molecular weight of 20 kDa
bulb to forebrain cholinergic nuclei indicates the ability of or less could reasonably be expected to be transported
these neurons to transport exogenously applied substances from the nasal cavity to the CSF (Sakane et al., 1995).
(Altar and Bakhit, 1991).
B. Enhancement of Delivery of Intranasally
Administered Drugs
IV. THE NASAL CAVITY AS A ROUTE OF
Transport of therapeutic drugs across the nasal epithelia
ADMINISTRATION FOR THERAPEUTIC
and into the blood is also a desired outcome, as is evi-
DRUGS
denced by the increasing number of drugs being developed
A. Background for this route of administration. Considerable effort has
gone into identifying agents that can enhance the absorp-
There is currently considerable interest in the concept of tion of drugs from the nasal cavity without causing
administration of therapeutic drugs via the nasal cavity for clinically relevant damage to the mucosal surface of the
delivery into the brain or into the systemic circulation. This nasal cavity. The putative absorption enhancers, sodium
562 Baker and Genter

tauro-24,25-dihydrofusidate (STDHF) and dimethyl-- (Gallagher, 1996; Lipton, 1997; Marttin et al., 1997), newer
cyclodextran (DM CD), exhibited some epithelial toxic- generation drugs which act as 5-HT1D receptor agonists
ity, whereas ethyleneamine diamine tetraacetic acid (e.g., sumatriptan or Imitrex®) are also proving to be potent,
(EDTA) was not associated with enhanced absorption or highly specific antimigraine agents when administered via
epithelial toxicity (Donnelly et al., 1997). Methylated - the intranasal route. The bioavailability of intranasally
cyclodextrins were less toxic to nasal epithelia than administered sumatriptan in dogs was equivalent to an oral
sodium glycocholate, STDHF, laureth-9 (a detergent) and dose (Barrow et al., 1997). Similarly, BMS-181885, another
L--phosphatidylcholine (Marttin et al., 1998). Free amine 5-HT1-like receptor agonist, was demonstrated to be quanti-
chitosans and soluble chitosan salts were evaluated for tatively absorbed from the nasal cavity of cynomolgus mon-
their efficacy and safety as nasal enhancers of peptide keys, with the highest observed plasma concentration
absorption and were found to be comparable in their attained in 30 minutes (Srinivas et al., 1998). There has also
action, if not more potent, than cyclodextrins been smaller-scale use of intranasal lidocaine or butorphanol
(Tengamnuay et al., 2000). Cyclodextrins are believed to as abortive treatments for migraines, as well as a case study
enhance absorption of intranasally administered drugs by reporting the effectiveness of intranasal capsaicin in
transiently opening tight junctions between nasal epithelial migraine relief (Levy, 1995; Melanson et al., 1997; Mills
cells (Marttin et al., 1998). High concentrations of certain and Scoggin, 1997; Sachs, 1996).
cyclodextrins are associated with ciliary damage, an end-
point also associated with benzalkonium chloride, which b. Sedatives and Analgesics. The intranasal route is
has been used as a preservative in many nasal formulations also a practical alternative to injections/intravenous adminis-
(Bernstein, 2000; Uchenna Agu et al., 2000). tration of sedatives and medicines for chronic pain.
Midazolam is administered in several medical situations,
C. Examples of Intranasally Administered including induction of anxiolysis and sedation, particularly
Therapeutic Drugs for endoscopic and dental procedures. Midazolam absorp-
tion from the nasal mucosa was nearly complete in a study
Intranasally administered drugs can be roughly divided into involving adult surgical patients who were administered
three main categories: (1) drugs administered via the nasal midazolam in small intranasal doses to preclude swallowing.
cavity for central nervous system effects; (2) drugs adminis- This study was important in that it revised previous observa-
tered via the nasal cavity for systemic effects; and (3) drugs tions that the bioavailability following intranasal administra-
administered via the nasal cavity for nasal and/or respiratory tion was approximately 55%, thus reducing the risk of
tract effects. Although the development of drugs for accidental overdoses (Björkman et al., 1997). Nasal midazo-
treatment of respiratory tract symptoms is undoubtedly the lam also appeared to be safe and effective in treatment of
strategy most intuitively associated with intranasal drug epileptic seizures in children (Jeannet et al., 1999).
delivery, it will not be dealt with in this chapter because of Intranasally administered oxycodone and oxymorphone are
our interest here in transport. The development of drugs effectively absorbed from the nasal cavity for treatment of
administered via the nasal cavity for central nervous system chronic pain (Hussain and Aungst, 1997; Takala et al., 1997).
and/or systemic effects has exploded in recent years (Table 3). Experiments to investigate the feasibility of administering
acetyl salicylic acid by the nasal route have also been con-
1. Drugs Administered via the Nasal Cavity for Central ducted in rat with promising results (Hussain et al., 1992).
Nervous System Effects
a. Antimigraine Medicines. Therapeutic drugs formu- c. Antivirals and Antibiotics. Drugs with activity
lated for intranasal administration for CNS effects include toward the HIV, namely zidovudine (AZT; 3ⴕ-azido-2ⴕ,
migraine headache treatments, sedatives/antiepileptics, neu- 3ⴕ-dideoxythymidine) and D4T (2ⴕ,3ⴕ-didehydro-
roprotective agents, and drugs to suppress cravings for 3ⴕ-deoxythymidine), are absorbed from the rat nasal
smoking and eating. While some efforts are still largely epithelia and transported into the CSF (Seki et al., 1994;
in the experimental stages, migraine patients using Yajima et al., 1998). Under normal administration regi-
medications administered via the intranasal route have real- mens, these drugs cannot cross the blood-brain barrier and
ized significant relief. Due to the high incidence of nausea, therefore are ineffective against acquired immunodefi-
vomiting, and visual disturbances during migraine attacks, ciency syndrome (AIDS) dementia when administered
intranasal administration of migraine medicines is proving orally or by injection. The antibiotic cephalexin appeared
to be superior to pills or self-injections. While ergotamine rapidly in the CSF following intranasal administration
derivatives have been extensively used in the past (Sakane et al., 1991b).
Exogenous Agent Uptake 563

d. Treatment/Prevention of Neurodegenerative promising in the effort to improve systemic delivery of

Diseases. Vasoactive intestinal peptide (VIP) is widely dis- insulin following intranasal administration (Fernandez-
tributed in regions of the brain associated with learning and Urrusuno et al., 1999). Intranasal calcitonin is a potent
memory. VIP antagonists are associated with learning impair- agent for reducing bone resorption (Silverman, 1997).
ments in experimental systems, and in vitro studies have
demonstrated that VIP exerts neuroprotective effects, includ- b. Hormone Analog. Buserelin and nafarelin are
ing protection of cortical neurons from the cytotoxic effects of being used as nasal sprays for treatment of endometriosis,
-amyloid peptide. Intranasal administration of a lipophilic prostate cancer, and low sperm count and show favorable
analog of VIP ([St-Nle17]VIP) to rats has been shown to results as contraceptives (Bergquist et al., 1979). Oxytocin
deliver unmetabolized ([St-Nle17]VIP to the brain and to pre- nasal sprays are associated with high blood levels of the
vent impairments in spatial learning and memory associated drug and enhanced lactation (Landgraf, 1985; Ruis et al.,
with cholinergic blockade in rats (Gozes et al., 1996, 1997). 1981).
Similarly, dextromethorphan and neostigmine were found to
be well absorbed from the rat and human nasal cavities, with c. Vaccines. The intranasal route is also becoming
the latter more effective in relieving symptoms of myasthenia recognized as a route of immunization against agents such
gravis as a nasal spray than when administered intravenously as influenza viruses, Neisseria meningitides, diphtheria,
(Char et al., 1992; Di Costanzo et al., 1993). Administration and tetanus (Liu, 1998; Maassab and DeBorde, 1985).
of vasopressin to humans to increase brain activity and
improve memory has resulted in reports of marginal d. Others. Vitamin B12 derivatives are well
improvement (Perras et al., 1997; Pietrowsky et al., 1996). absorbed following intranasal administration, as are drugs
for treatment of motion sickness and bedwetting (Butler et
e. Smoking Cessation and Appetite Suppression. al., 1998; Kallas et al., 1999; Putcha et al., 1996;
Nicotine nasal sprays have been evaluated as agents to assist Ramanathan et al., 1998; Slot et al., 1997; Swain, 1995).
individuals in smoking cessation. It is possible that the sprays
are efficacious due to the direct delivery of nicotine to the
brain as well as maintenance of blood nicotine levels such V. CONSEQUENCES OF TRANSPORT
that the majority of symptoms of withdrawal are avoided.
While there have been reports that the irritation associated One of the early changes associated with several neurode-
with nicotine nasal sprays precluded their use to attain thera- generative disorders, including Parkinson’s and
peutic blood levels (Gourlay and Benowitz, 1997), another Alzheimer’s diseases, is a substantial deficit in olfactory
study showed that successful “quitters” who most frequently function of a discriminatory nature (Doty et al., 1987,
used nicotine nasal sprays as part of a smoking cessation pro- 1988, 1991) (see Chapters 23 and 24). These data, in con-
gram had higher blood nicotine and cotinine levels than those junction with recent findings of pathological changes in
using the sprays occasionally or not at all (Jones et al., 1998). the olfactory epithelium (Tabaton et al., 1991; Talamo et
Experimentally, cholecystokinin (CCK) agonists show al., 1989) and olfactory bulb (Esiri and Wilcock, 1984;
promise as appetite suppressants (Pierson et al., 1997) Ohm and Braak, 1987), as well as brain areas with close
synaptic relationships to the olfactory bulb (Ferreya-
2. Drugs Administered via the Nasal Cavity for Moyano and Barragan, 1989; Saper et al., 1987), formed
Systemic Effects the basis for the hypothesis that the olfactory system may
be a route of entry for either a toxin or a virus that is
There is a similarly long and varied list of therapeutic important to the etiology of these diseases.
drugs which are designed for intranasal delivery to achieve For most substances, the evidence is at best circumstan-
a desired systemic effect (see recent reviews by Jones et tial. Several reviews have outlined the data, suggesting that
al., 1997; Mathison et al., 1998). environmental aluminum could contribute to the loss of
olfactory functioning and the ensuing disease process in
a. Peptides. Delivery of peptides such as insulin and Alzheimer’s disease. The findings of high levels of alu-
calcitonin via the nasal route would in theory improve their minum in plaques and the ability to produce lesions follow-
bioavailability over oral administration. There have been ing inhalation of aluminum support this hypothesis. The data
mixed results with insulin administration via the nasal could reflect, however, other abnormalities in Alzheimer’s
route (Hilsted et al., 1995), possibly because of peptidases disease, such as changes in the blood-brain barrier that per-
in nasal secretions that have high activity toward insulin mit accumulation of this metal in neurons (Crapper
peptides. Methods to encapsulate insulin peptides appear McLachlan, 1986; Roberts, 1986). Recent studies have
564 Baker and Genter

provided anatomical demonstration of transport to the CNS nomenon? What are the characteristics of cell surface
of intranasally applied manganese (Tjälve and Henriksson, receptors that lead to binding and transport of specific
1999) as well as epidemiological evidence for a Parkinson’s- xenobiotics and viruses? Is internalization concentration-
like syndrome following environmental exposure to the dependent, and are the concentrations utilized in experi-
metal (Gorell et al., 1999). While there is no definitive evi- mental application overwhelming mechanisms that
dence that the olfactory system serves as a route of entry for normally remove odorants and other substances? Is there
metals and other toxins in humans, there is strong support for age-dependence? Recent data showed that viral spread
an environmental role in sporadic Parkinson’s disease following inhalation exposure was influenced by neuronal
(Betarbet et al., 2001; Engel et al. 2001; Langston, 1998; maturity (Oliver and Fazakerley, 1998) and that in the
Tanner and Langston, 1990). olfactory mucosal damage induced by 3,3ⴕ-iminodipropi-
Currently, there is no evidence for viral involvement in onitrile is age-dependent (Genter and Ali, 1998). What
neurodegenerative disorders such as Alzheimer’s or happens to the receptor cells following uptake? There is
Parkinson’s diseases. To postulate that the olfactory system experimental evidence to suggest that excessive uptake
serves as a route of entry for a virus in these diseases is would lead to neuronal degeneration and replacement since
only conjectural. The fact that in some degenerative disor- exposure to normal laboratory environments produced
ders the affected neuronal populations appear to be greater turnover of receptor cells than exposure of animals
restricted to interconnected pathways has been clearly to purified air such as in a laminar flow hood (Hinds et al.,
established (Saper et al., 1987), and transneuronal trans- 1984). In contrast, intranasal WGA-HRP resulted in a
port of viruses does occur along some of these routes espe- reversible thinning of the olfactory epithelium produced
cially in the olfactory system. Trophic substances also primarily as a result of apoptotic receptor cell loss (Moon
were shown to be transported along these same pathways. and Baker, 1998, 2002) (see also Chapter 5). What is the
In the cholinergic system, for example, which innervates role of the immune system in protecting the epithelium and
the olfactory bulb, nerve growth factor is retrogradely brain from viruses and exogenous agents? Are the drug-
transported (Altar and Bakhit, 1991). Thus, an abnormality metabolizing enzymes present in sufficient concentrations
in transport of a trophic factor, either inborn or toxin to protect the epithelium against toxin exposure, and, in
(viral?)-induced, could result in neuronal degeneration. fact, is this the role of these enzymes? We appear to survive
Alternations in peripheral afferent innervation of the olfac- exposure to many substances without disease. However,
tory bulb, both denervation and odorant deprivation, pro- the human olfactory epithelium appears to degenerate with
duce profound anatomical and biochemical consequences. age, so the question remains as to how effective the epithe-
Both neonates and adults show reductions in bulb size lium is at protecting itself and the brain from whatever
(Maruniak et al., 1989 a,b; Meisami, 1976), granule cell enters the nares. From a positive perspective, recent exper-
spine densities (Benson et al., 1984), number of granule imental and clinical data show that nasal application may
cells (Frazier and Brunjes, 1988), and neurotransmitter provide an ideal route for delivery of drugs to the CNS.
expression, including substance P and dopamine, in juxta-
glomerular cells (Baker et al., 1983; Kream et al., 1984).
ACKNOWLEDGMENTS
The latter changes occur without apparent cell death, indi-
cating the presence of phenotypic plasticity (Baker et al.,
Supported by in part grant number ES08799 from NIEHS
1984, 1993; Cho et al., 1996; Stone et al., 1990, 1991) (See
to MBG and grant number AG09686 from NIA to HB.
also Chapter 29). These data indicate the importance of
afferent innervation as a trophic regulator of bulb function
and suggest that either exogenous agents or viral internal- REFERENCES
ization may alter this trophic balance and result in the loss
of olfactory function with eventual consequences in synap- Aggerbeck, H., Gizurarson, S., Wantzin, J., Heron, I. Intranasal
tically interconnected brain regions. booster vaccination against diphtheria and tetanus in man.
Vaccine 1997; 15:307–316.
Allen, W. K., Akeson, R. Identification of a cell surface glyco-
protein family of olfactory receptor neurons with a mono-
VI. CONCLUSIONS
clonal antibody. J. Neurosci 1985; 5:284–296.
Altar, C. A., Bakhit, C. Receptor-mediated transport of human
There is no question that olfactory receptor cells can inter- recombinant nerve growth factor from olfactory bulb to fore-
nalize and transport a variety of materials, both inorganic brain cholinergic nuclei. Brain Res. 1991; 541:82–88.
and organic, including viruses. However, a number of Apfelbach, R., Engelhart, A., Behnisch, P., Hagenmaier, H.
issues remain to be clarified. First, how general is the phe- The olfactory system as a portal of entry for airborne
Exogenous Agent Uptake 565

polychlorinated biphenyls (PCBs) to the brain? [letter]. Arch. Barnett, E. M., Cassell, M. D., Perlman, S. Two neurotropic
Toxicol. 1998; 72:314–317. viruses, herpes simplex virus type 1 and mouse hepatitis
Araneda, S., Font, C., Pujol, J. F., Bobillier, P. Retrograde axonal virus, spread along different neural pathways from the main
transport after radioactive hydroxyindole injections into the olfactory bulb. Neuroscience 1993; 57:1007–1025.
olfactory bulb-an autoradiographic study. Neurochem. Int. Barrow, A., Dixon, C. M., Saynor, D. A., Perren, M. J., Stowe, R.,
1983; 5:741–750. Smith, I. The absorption, pharmacodynamics, metabolism and
Armstrong, C., Harrison, W. T. Prevention of intranasally inocu- excretion of 14C-sumatriptan following intranasal administration
lated poliomyelitis of monkeys by instillation of alum into to the beagle dog. Biopharm. Drug Dispos. 1997; 18: 443–458.
nostrils. Public Health Rep. 1935; 50:725–730. Bazer, G. T., Ebbesson, S. O. E., Reynolds, J. B., Bailey, R. P. A
Astic, L., Saucier, D., Coulon, P., Lafay, F., Flamand, A. The cobalt-lysine study of primary olfactory projections in king
CVS strain of rabies virus as transneuronal tracer in the olfac- salmon fry (Oncorhynchus tshawytscha Walbaum). Cell
tory system of mice. Brain Res. 1993; 619:146–156. Tissue Res. 1987; 248:499–503.
Bahrami, F., Bergman, U., Brittebo, E. B., Brandt, J. Persistent Bechgaard, E., Gizurarson, S., Hjortkjaer, R. K. Pharmacokinetic
olfactory mucosal metaplasia and increased olfactory bulb and pharmacodynamic response after intranasal administra-
glial fibrillary acidic protein levels following a single dose of tion of diazepam to rabbits. J. Pharm. Pharmacol. 1997;
methylsulfonyl-dichlorobenzene in mice: comparison of the 49:747–750.
2,5- and 2,6-dichlorinated isomers. Toxicol. Appl. Pharmacol. Becker, Y. HSV-1 brain infection by the olfactory nerve route and
2000; 162:49–59. virus latency and reactivation may cause learning and behav-
Baker, H. Transport phenomena within the olfactory system. In: ioral deficiencies and violence in children and adults: a point
Doty, R. L., ed. Handbook of Clinical Olfaction and of view. Virus Genes 1995; 10:217–226.
Gustation. New York: Marcel Dekker, 1995:173–190. Belinsky, S., Walker, V., Maronpot, R., Swenberg, J., Anderson,
Baker, H., Margolis, F. L. Deafferentation induced alterations in M. Molecular dosimetry of DNA adduct formation and cell
olfactory bulb as a model for the etiology of Alzheimer’s dis- toxicity in rat nasal mucosa following exposure to the tobacco
ease. Neurobiol. Aging 1986; 7:568–569. specific nitrosamine 4-(N-methyl-N-nitrosamino)-1-(3-pyri-
Baker, H., Spencer, R. F. Transneuronal transport of peroxidase- dyl)-1-butanone and their relationship to induction of neopla-
conjugated wheat germ agglutinin (WGA-HRP) from the sia. Cancer Res. 1987; 47:6058–6065.
olfactory epithelium to the brain of the adult rat. Exp. Brain. Benson, T. E., Ryugo, D. K., Hinds, J. W. Effects of sensory
Res. 1986; 63:461–473. deprivation on the developing mouse olfactory system: a light
Baker, H., Kawano, T., Margolis, F. L., Joh, T. H. Transneuronal and electron microscopic, morphometric analysis. J.
regulation of tyrosine hydroxylase expression in olfactory Neurosci. 1984; 4:238–253.
bulb of mouse and rat. J. Neurosci. 1983; 3:69–78. Bergman, U., Ostergren, A., Gustafson, A. L., Brittebo, B.
Baker, H., Kawano, T., Albert, V. R., Joh, T. H., Reis, D. J., Differential effects of olfactory toxicants on regeneration.
Margolis, F. L. Olfactory bulb dopamine neurons survive Arch. Toxicol. 2002; 76:104–112.
deafferentiation induced loss of tyrosine hydroxylase. Bergquist, C., Nillius, S. J., Wide, L. Intranasal gonadotropin-
Neuroscience 1984; 11:605–615. releasing hormone agonist as a contraceptive agent. Lancet
Baker, H., Morel, K., Stone, D. M., Maruniak, J. A. Adult naris 1979; 2:215–217.
closure profoundly reduces tyrosine hydroxylase expression Bernstein, I. L. Is the use of benzalkonium chloride as a preserv-
in mouse olfactory bulb. Brain Res. 1993; 614:109–116. ative for nasal formulations a safety concern? A cautionary
Baker, T. A., Maruniak, J. Uptake of immunoglobulins by olfac- note based on compromised mucociliary transport. J. Allergy
tory receptor neurons. Chem. Senses 1990; 15:549. Clin. Immunol. 2000; 105:39–44.
Baldaccini, N. E., Gagliardo, A., Pelosi, P., Topazzini, A. Betarbet, R., Sherer, T. B., Mackenzie, G., Garcia-Osuna, M.,
Occurrence of a pyrazine binding protein in the nasal mucosa Panov, A.V., Greenamyre, J. T. Chronic systemic pesticide
of some vertebrates. Comp. Biochem. Physiol. [B] 1986; exposure reproduces features of Parkinson’s disease. Nat.
84:249–253. Neuro. Sci. 2000; 3:1301–1306.
Balin, B. J., Broadwell, R. D., Salcman, M. el-Kalliny, M. Björkman, S., Rigemar, G., Idvall, J. Pharmacokinetics of mid-
Avenues for entry of peripherally administered protein to the azolam given as an intranasal spray to adult surgical patients.
central nervous system in mouse, rat, and squirrel monkey. J. Br. J. Anaesth. 1997; 79:575–580.
Comp. Neurol. 1986; 251:260–280. Blessing, W. W., Ding, Z. Q., Li, Y. W., Gieroba, Z. J., Wilson, A.
Bannister, L. H., Dodson, H. C. Endocytic pathways in the olfac- J., Hallsworth, P. G., Wesselingh, S. L. Transneuronal
tory and vomeronasal epithelia of the mouse: ultrastructure labelling of CNS neurons with herpes simplex virus. Prog.
and uptake of tracers. Microsc. Res. Tech. 1992; 23:128–141. Neurobiol. 1994; 44:37–53.
Barber, P. C. Ulex europus agglutinin I binds exclusively to pri- Bodian, D., Howe, H. A. Experimental studies on intraneural
mary olfactory neurons in the rat nervous system. spread of poliomyelitis virus. Bull. Johns Hopkins Hosp.
Neuroscience 1989; 30:1–9. 1941a; 68:248–267.
Barceloux, D. G., Nickel, J. Toxicol. Clin. Toxicol. 1999; Bodian, D., Howe, H. A. The rate of progression of poliomyelitis
37:239–258. virus in nerves. Bull. Johns Hopkins Hosp. 1941b; 69:79–85.
566 Baker and Genter

Bogdanffy, M. S., Mazaika, T. J., Fasano, W. J. Early cell prolif- [14C]dextromethorphan hydrochloride in rats: levels in
erative and cytotoxic effects of phenacetin on rat nasal plasma and brain. J. Pharm. Sci. 1992; 81:750–752.
mucosa. Toxicol. Appl. Pharmacol. 1989; 98:100–112. Charles, P. C., Walters, E., Margolis, F., Johnston, R. E.
Bonnet-Font, C., Bobillier, P. Retrograde axonal transport speci- Mechanism of neuroinvasion of Venezuelan equine encephali-
ficity in the locus coeruleus neurons after [3H]Noradrenaline tis virus in the mouse. Virology 1995; 208:662–671.
injection into the rat olfactory bulb. Neurochem. Int. 1990; 16: Cho, J. Y., Min, N., Franzen, L., Baker, H. Rapid down-regulation
523–532. of tyrosine hydroxylase expression in the olfactory bulb of
Borg-Neczak, K., Tjälve, H. Uptake of 203Hg2 in the olfactory naris-occluded adult rats. J. Comp. Neurol. 1996; 369:264–276.
system in pike. Toxicol. Lett. 1996; 84:107–112. Clark, W. E. L. Anatomical investigation into the routes by which
Brandt, I., Brittebo, E. B., Feil, V. J., Bakke, J. E. Irreversible infections may pass from the nasal cavities into the brain. Rep.
binding and toxicity of the herbicide dichlobenil (2,6- Public Health Med. Subjects No. 54. London 1929:1–27.
dichlorobenzonitrile) in the olfactory mucosa of mice. Crapper McLachlan, D. R. Aluminum and alzheimer’s disease.
Toxicol. Appl. Pharmacol. 1990; 103:491–501. Neurobiol. Aging. 1986; 7:525–532.
Brittebo, E. B. Binding of cocaine in the liver, olfactory mucosa, Crofton, K. M., Zhao, X., Sayre, L. M., Genter, M. B.
eye, and fur of pigmented mice. Toxicol. Appl. Pharmacol. Characterization of the effects of N-hydroxy-IDPN on the
1988; 96:315–323. auditory, vestibular, and olfactory systems in rats.
Brittebo, E. B., Eriksson, C. Taurine in the olfactory system: Neurotoxicol. Teratol. 1996; 18:297–303.
effects of the olfactory toxicant dichlobenil. Neurotoxicology Czerniawska, A. Experimental investigations on the penetration
1995; 16:271–280. of 198Au from nasal mucous membrane into cerebrospinal
Brittebo, E. B., Eriksson, C., Brandt, I. Activation and toxicity of fluid. Acta. Otolaryngol. 1970; 70:58–61.
bromobenzene in nasal tissue in mice. Arch. Toxicol. 1990; Dahl, A. R. The effect of cytochrome P-450-dependent metabo-
64:54–60. lism and other enzyme activities on olfaction. In: Margolis, F.
Brittebo, E. B., Eriksson, C., Feil, V., Bakke, J., Brandt, I. L., Getchell, T. V., eds. Molecular Neurobiology of the
Toxicity of 2,6-dichlorothiobenzamide (chlorthiamid) and Olfactory System. New York: Plenum Press, 1988:51–70.
2,6-dichlorobenzamide in the olfactory nasal mucosa of mice. Dahl, A. R. The cyanide-metabolizing enzyme rhodenase in rat nasal
Fundam. Appl. Toxicol. 1991; 17:92–102. respiratory and olfactory mucosa. Toxic Lett. 1989; 45:199–205.
Broadwell, R. D., Balin, B. J. Endocytic and exocytic pathways Dahl, A. R., Hadley, W. M. Formaldehyde production promoted
of the neuronal secretory process and trans-synaptic transfer by rat nasal cytochrome P-450-dependent monooxygenases
of wheat germ agglutinin-horseradish peroxidase in vivo. J. with nasal decongestants, essences, solvents, air pollutants,
Comp. Neurol. 1985; 242:632–250. nicotine, and cocaine as substrates. Toxic Appl. Pharmacol.
Buck, L., Axel, R. A novel multigene family may encode odorant 1983; 67:200–205.
receptors: a molecular basis for odor recognition. Cell 1991; Dahl, A. R., Hadley, W. M. Nasal cavity enzymes involved in
65:175–187. xenobiotic metabolism: effects on the toxicity of inhalants.
Burd, G. D., Davis, B. J., Macrides, F., Grillo, M., Margolis, F. L. Toxicology 1991; 21:345–372.
Carnosine in primary afferents of the olfactory system: an Dahlof, C. G., Boes-Hansen, S., Cederberg, C. G., Hardebo, J. E.,
autoradiographic and biochemical study. J. Neurosci. 1982; Henriksson, A. How does sumatriptan nasal spray perform in
2:244–255. clinical practice? Cephalalgia 1998; 18:278–282.
Butler, R., Holland, P., Devitt, H., Hiley, E., Roberts, G., Dal Monte, M., Andrieni, I., Revoltella, R., Pelosi, P. Purification
Redfern, E. The effectiveness of desmopressin in the treat- and characterization of two odorant-binding proteins from
ment of childhood nocturnal enuresis: predicting response nasal tissue of rabbit and pig. Comp. Biochem. Physiol. 1991;
using pretreatment variables. Br. J. Urology. 1998; 2:445–451.
81:29–36. Dantas, Z. N., Vicino, M., Balmaceda, J. P., Asch, R. H., Stone,
Cancalon, P. Influence of temperature on the velocity and on the S. C. Comparison between nafarelin and leuprolide acetate for
isotope profile of slowly transported labeled proteins. J. in vitro fertilization: preliminary clinical study. Fertil. Steril.
Neurochem. 1979a; 32:997–1007. 1994; 61:705–708.
Cancalon, P. Subcellular and polypeptide distributions of slowly Deamer, N. J., Genter, M. B. Olfactory toxicity of diethyldithio-
transported proteins in the garfish olfactory nerve. Brain Res. carbamate (DDTC) and disulfiram and the protective effect of
1979b; 161:115–130. DDTC against the olfactory toxicity of dichlobenil. Chem.
Cancalon, P. F. Axonal transport in the garfish optic nerve: compari- Biol. Interact. 1995; 95:215–226.
son with the olfactory system. J. Neurochem. 1988; 51: 266–276. Deamer, N. J., O’Callaghan, J. P., Genter, M. B. Olfactory toxic-
Cancalon, P., Brady, S. T., Lasek, R. J. Slow transport in a nerve ity resulting from dermal application of 2,6-dichlorobenzoni-
with embryonic characteristics, the olfactory nerve. Brain trile (dichlobenil) in the C57B1 mouse. Neurotoxicology
Res. 1988; 466:275–285. 1994; 15:287–293.
Char, H., Kumar, S., Patel, S., Piemontese, D., Iqbal, K., Malick, De Lorenzo, A. J. D. The olfactory neuron and the blood brain bar-
A. W., Salvador, R. A., Behl, C. R. Nasal delivery of rier. In: Wolstenholme G. E. W, Knight, J. eds. Taste and Smell
in Vertebrates, London: Churchill Livingstone, 1970:151–175.
Exogenous Agent Uptake 567

Di Costanzo, A., Toriello, A., Mannara, C., Benvenuti, C., Fredman, S. M., Jahan-Pawar, B. Cobalt mapping of the nervous
Tedeschi, G. Intranasal versus intravenous neostigmine in system: evidence that cobalt can cross a neuronal membrane.
myasthenia gravis: assessment by computer analysis of J. Neurobiol. 1980; 11:209–214.
saccadic eye movements. Clin. Neuropharmacol. 1993; Frey, W. H. I., Liu, J., Chen, X., Thorne, R. G., Fawcett, J. R.,
16:511–517. Ala, T. A. Delivery of 125I-NGF to the brain via the olfactory
Donnelly, A., Kellaway, I. W., Taylor, G., Gibson, M. Absorption route. Drug Deliv. 1997; 4:87–92.
enhancers as tools to determine the route of nasal absorption Gallagher, R. M. Acute treatment of migraine with dihydroergo-
of peptides. J. Drug Target. 1997; 5:121–127. tamine nasal spray. Dihydroergotamine Working Group. Arch.
Doty, R. L., Reyes, P., Gregor, T. Presence of both odor identifi- Neurol. 1996; 53:1285–1291.
cation and detection deficits in Alzheimer’s disease. Brain Genter, M. B. Evaluation of olfactory and auditory system effects
Res. Bull. 1987; 18:597–600. of the antihyperthyroid drug carbimazole in the Long-Evans
Doty, R. L., Deems, D. A., Stellar, S. Olfactory dysfunction in rat. J. Biochem. Mol. Toxicol. 1998; 12:305–314.
parkinsonism: a general deficit unrelated to neurologic signs, Genter, M. B., Ali, S. F. Age-related susceptibility to 3,3´-imin-
disease stage, or disease duration. Neurology 1988; odipropionitrile-induced olfactory mucosal damage.
38:1237–1244. Neurobiol. Aging 1998; 19:569–574.
Doty, R. L., Perl, D. P., Steele, J. C., Chen, K. M., Pierce, J. D., Genter, M. B., Crofton, K. M. 2-Pentenenitrile. New York:
Reyes, P., Kurland, L. T. Olfactory dysfunction in three neu- Oxford University Press, 2000.
rodegenerative diseases. Geriatrics 1991; 1:47–51. Genter, M. B., Llorens, J., O’Callaghan, J. P., Peele, D. B.,
Draghia, R., Caillaud, C., Manicom, R., Pavirani, A., Kahn, A., Morgan, K. T., Crofton, K. M. Olfactory toxicity of
Poenaru, L. Gene delivery into the central nervous system by beta,beta´-iminodipropionitrile in the rat. J. Pharmacol. Exp.
nasal instillation in rats. Gene Ther. 1995; 2:418–423. Ther 1992; 263:1432–1439.
Engel, L. S., Checkoway, H., Kiefer, M. C., Seixas, N. S., Genter, M. B., Deamer, N. J., Blake, B. L., Wesley, D. S., Levi, P.
Longstreth, W. T., Jr., Scott, K. C., Hudnell, K., Anger, W. K., E. Olfactory toxicity of methimazole: dose-response and
Camicioli, R. Parkinsonism and occupational exposure to pes- structure-activity studies and characterization of flavin-con-
ticides. Occup. Environ. Med. 2001; 58:582–589. taining monooxygenase activity in the Long-Evans rat olfac-
Esiri, M., Wilcock, G. The olfactory bulbs in Alzheimer’s dis- tory mucosa. Toxicol. Pathol. 1995; 23:477–486.
ease. J. Neurol. Neurosurg. Psychiatry 1984; 47:56–60. Genter, M. B., Liang, H. C., Gu, J., Ding, X., Negishi, M.,
Esiri, M. M., Tomlinson, A. H. Immunohistological demonstra- McKinnon, R. A., Nebert, D. W. Role of CYP2A5 and 2G1 in
tion of spread of virus via olfactory and trigeminal pathways acetaminophen metabolism and toxicity in the olfactory
after infection of facial skin in mice. J. Neurol. Sci. 1984; mucosa of the Cypla2(-/-) mouse. Biochem. Pharmacol. 1998;
64:213–217. 55:1819–1826.
Evans, J., Hastings, L. Accumulation of Cd(II) in the CNS Genter, M. B., Burman, D. M., Dingeldein, M. W., Clough, I.,
depending on the route of administration: intraperitoneal, Bolon, B. Characterization of cell proliferation and immuno-
intratracheal, or intranasal. Fundam. Appl. Toxicol. 1992; histochemical markers of alachlor-induced olfactory mucosal
19:275–278. tumors in the Long-Evans rat. Toxicol. Pathol. 2000;
Evans, J. E., Miller, M. L., Andringa, A., Hastings, L. Behavioral, 28:770–781.
histological, and neurochemical effects of nickel (II) on the Gervasi, P. G., Longo, V., Naldi, F., Panattoni, G., Ursino, F.
rat olfactory system. Toxicol. Appl. Pharmacol. 1995; Xenobiotic-metabolizing enzymes in human respiratory nasal
130:209–220. mucosa. Biochem. Pharmacol. 1991; 41:177–184.
Feng, P. C., Wilson, A. G., McClanahan, R. H., Patanella, J. E., Getchell, M. L., Getchell, T. V. Immunohistochemical localization
Wratten, S. J. Metabolism of alachlor by rat and mouse liver of components of the immune barrier in the olfactory mucosae
and nasal turbinate tissues. Drug Metab. Dispos. 1990; of salamanders and rats. Anat. Rec. 1991; 231:358–374.
18:373–377. Getchell, T. V., Margolis, F. L., Getchell, M. L. Perireceptor and
Fernandez-Urrusuno, R., Calvo, P., Remunan-Lopez, C., Vila- receptor events in vertebrate olfaction. Prog. Neurobiol. 1984;
Jato, J. L., Alonso, M. J. Enhancement of nasal absorption of 23:317–345.
insulin using chitosan nanoparticles. Pharm. Res. 1999; Ghantous, H., Dencker, L., Gabrielsson, J., Danielsson, B. R.,
16:1576–1581. Bergman, K. Accumulation and turnover of metabolites of
Ferreyra-Moyano, H., Barragan, E. The olfactory system and toluene and xylene in nasal mucosa and olfactory bulb in the
Alzheimer’s disease. Int. J. Neurosci. 1989; 49:157–197. mouse. Pharmacol. Toxicol. 1990; 66:87–92.
Foster, J. D., Getchell, M. L., Getchell, T. V. Identification of Gianutsos, G., Morrow, G. R., Morris, J. B. Accumulation of
sugar residues in secretory glycoconjugates of olfactory manganese in rat brain following intranasal administration.
mucosae using lectin histochemistry. Anat. Rec. 1991; Fundam. Appl. Toxicol. 1997; 37:102–105.
229:525–544. Good, P. F., Olanow, C. W., Perl, D. P. Neuromelanin-containing
Frazier, L. I., Brunjes, P. C. Unilateral odor deprivation: early neurons of the substantia nigra accumulate iron and aluminum
postnatal changes in olfactory bulb cell density and number. J. in Parkinson’s disease: a LAMMA study. Brain Res. 1992;
Comp. Neurol. 1988; 269:355–370. 593:343–346.
568 Baker and Genter

Gorell, J. M., Johnson, C. C., Rybicki, B. A., Peterson, E. L., Hayek, R., Waite, P. M. E. The olfactory pathway as a possible
Kortsha, G. X., Brown, G. G., Richardson, R. J. Occupational route for aluminum entry to the brain. J. Neurochem. 1991;
exposure to manganese, copper, lead, iron, mercury and zinc 57:S113.
and the risk of Parkinson’s disease. Neurotoxicology 1999; Hayes, R. B., Gerin, M., Raatgever, J. W., de Bruyn, A. Wood-
20:239–247. related occupations, wood dust exposure, and sinonasal can-
Gottofrey, J., Tjälve, H. Axonal transport of cadmium in the olfac- cer. Am. J. Epidemiol. 1986; 124:569–577.
tory nerve of the pike. Pharmacol. Toxicol. 1991; 69:242–252. Hempstead, J. L., Morgan, J. I. A panel of monoclonal antibod-
Gourlay, S. G., Benowitz, N. L. Arteriovenous differences in ies to the rat olfactory epithelium. J. Neurosci. 1985; 5:
plasma concentration of nicotine and catecholamines and 438–449.
related cardiovascular effects after smoking, nicotine nasal Henriksson, J., Tjälve, H. Uptake of inorganic mercury in the
spray, and intravenous nicotine. Clin. Pharmacol. Ther. 1997; olfactory bulbs via olfactory pathways in rats. Environ. Res.
62:453–463. 1998; 77:130–140.
Gozes, I., Bardea, A., Reshef, A., Zamostiano, R., Zhukovsky, S., Hilsted, J., Madsbad, S., Hvidberg, A., Rasmussen, M. H.,
Rubinraut, S., Fridkin, M., Brenneman, D. E. Neuroprotective Krarup, T., Ipsen, H., Hansen, B., Pedersen, M., Djurup, R.,
strategy for Alzheimer disease: intranasal administration of a Oxenboll, B. Intranasal insulin therapy: the clinical realities.
fatty neuropeptide. Proc. Natl. Acad. Sci. USA 1996; Diabetologia 1995; 38:680–684.
93:427–432. Hinds, J. W., Hinds, P. L., McNelly, N. A. An autoradiographic
Gozes, I., Bardea, A., Bechar, M., Pearl, O., Reshef, A., study of the mouse olfactory epithelium: evidence for long-
Zamostiano, R., Davidson, A., Rubinraut, S., Giladi, E., lived receptors. Anat. Rec. 1984; 210:375–383.
Fridkin, M., Brenneman, D. E. Neuropeptides and neuronal Hoeppner, T. J. The anticonvulsant valproic acid concentrates in
survival: neuroprotective strategy for Alzheimer’s disease. the olfactory bulb: selective laminar localization. Brain Res.
Ann. NY Acad. Sci. 1997; 814:161–166. 1990; 532:326–328.
Graf, W., Gerrits, N., Yatim-Dhiba, N., Ugolini, G. Mapping the Holl, A. Vital staining by trypan blue; its selectivity for olfactory
oculometer system: the power of transneuronol labeling with receptor cells of the brown bullhead, Ictalurus natalis. Stain.
rabies virus. Eur. J. Neurosci. 2002; 15:1557–1562. Technol. 1965; 40:269–273.
Gross, G. W., Beidler, L. M. Fast axonal transport in the c-fibers Holl, A. Selective staining by procion dyes of olfactory sensory
of the garfish olfactory nerve. J. Neurobiol. 1973; 4: 413–428. neurons in the catfish Ictalurus nebulosus (author’s transl). Z.
Gross, G. W., Beidler, L. M. A quantitative analysis of isotope Naturforsch. [C] 1980; 35:526–528.
concentration profiles and rapid transport velocities in the C- Holl, A. Marking of olfactory axons of fishes by intravital staining
fibers of the garfish olfactory nerve. J. Neurobiol. 1975; with procion brilliant yellow. Stain Technol. 1981; 56:67–70.
6:213–232. Horowitz, L. F., Montmayeur, J. P., Echelard, Y., Buck, L. B. A
Gross, G. W., Kreutzberg, G. W. Rapid axoplasmic transport in genetic approach to trace neural circuits. Proc. Natl. Acad.
the olfactory nerve of the pike: I. Basic transport parameters Sci. USA 1999; 96:3194–3199.
for proteins and amino acids. Brain Res. 1978; 139: 65–76. Huneycutt, B. S., Plakhov, I. V., Shusterman, Z., Bartido, S. M.,
Gu, J., Walker, V. E., Lipinskas, T. W., Walker, D. M., Ding, X. Huang, A., Reiss, C. S., Aoki, C. Distribution of vesicular
Intraperitoneal administration of coumarin causes tissue- stomatitis virus proteins in the brains of BALB/c mice fol-
selective depletion of cytochromes P450 and cytotoxicity in lowing intranasal inoculation: an immunohistochemical
the olfactory mucosa. Toxicol. Appl. Pharmacol. 1997; analysis. Brain Res. 1994; 635:81–95.
146:134–143. Hurst, E. W. New knowledge of virus diseases of nervous system:
Guthrie, K. M., Pullara, J. M., Marshall, J. F., Leon, M. Olfactory review and interpretation. Brain 1936; 59:1–34.
deprivation increases dopamine D2 receptor density in the rat Hussain, A. A., Iseki, K., Kagoshima, M., Dittert, L. W.
olfactory bulb. Synapse 1991; 8:61–70. Absorption of acetylsalicylic acid from the rat nasal cavity. J.
Haneberg, B., Dalseg, R., Oftung, F., Wedege, E., Hoiby, E. A., Pharm. Sci. 1992; 81:348–349.
Haugen, I. L., Holst, J., Andersen, S. R., Aase, A., Meyer Hussain, M. A., Aungst, B. J. Intranasal absorption of oxymor-
Naess, L., Michaelsen, T. E., Namork, E., Haaheim, L. R. phone. J. Pharm. Sci. 1997; 86:975–976.
Towards a nasal vaccine against meningococcal disease, and Isaka, H., Yoshii, H., Otsuji, A., Koike, M., Nagai, Y., Koura, M.,
prospects for its use as a mucosal adjuvant. Dev. Biol. Stand. Sugiyasu, K., Kanabayashi, T. Tumors of Sprague-Dawley
1998; 92:127–133. rats induced by long-term feeding of phenacetin. Gann. 1979;
Haseman, J. K., Hailey, J. R. An update of the National 70:29–36.
Toxicology Program database on nasal carcinogens. Mutat. Itaya, S. K. Anterograde transsynaptic transport of WGA-HRP in
Res. 1997; 380:3–11. rat olfactory pathways. Brain Res. 1987; 409:205–214.
Hastings, L. Sensory neurotoxicology: use of the olfactory sys- Itaya, S. K., Van Hausen, G. W. WGA-HRP as a transneuronal
tem in the assessment of toxicity. Neurotoxicol. Teratol. 1990; marker in the visual pathways of monkey and rat. Brain Res.
12:455–459. 1982; 236:199–204.
Hastings, L., Evans J. Transaxonal transport of cadmium in the Ivankovic, S., Seibel, J., Komitowski, D., Spiegelhalder, B.,
olfactory system. Chem. Senses 1988; 13:696. Preussmann, R., Siddiqi, M. Caffeine-derived N-nitroso com-
Exogenous Agent Uptake 569

pounds. V. Carcinogenicity of mononitrosocaffeidine and dini- subset of neurons in rabbit cerebral cortex, basal forebrain and
trosocaffeidine in bd-ix rats. Carcinogenesis 1998; 19:933–937. upper brainstem. Neuroscience 1989; 29:320–337.
Jackson, R. T., Tigges, J., Arnold, W. Subarachnoid space of the Kream, R. M., Davis, B. J., Kawano, T., Margolis, F. L.,
CNS, nasal mucosa, and lymphatic system. Arch. Macrides, F. Substance P and catecholaminergic expression in
Otolaryngol. 1979; 105:180–184. neurons of the hamster main olfactory bulb. J. Comp. Neurol.
Jacobson, J., Harris, S. R., Bullingham, R. E. Low dose intranasal 1984; 222:140–154.
nafarelin for the treatment of endometriosis. Acta. Obstet. Kristensson, K., Olsson, Y. Uptake of exogenous proteins in
Gynecol. Scand. 1994; 73:144–150. mouse olfactory cells. Acta. Neuropathol. 1971; 19:145–154.
Jarolim, K. L., McCosh, J. K., Howard, M. J., John, D. T. A light Lafay, F., Coulon, P., Astic, L., Saucier, D., Riche, D., Holley, A.,
microscopy study of the migration of Naegleria fowleri from Flamand, A. Spread of the CVS strain of rabies virus and of
the nasal submucosa to the central nervous system during the the antivirulent mutant AvOl along the olfactory pathways of
early stage of primary amebic meningoencephalitis in mice. J. the mouse after intranasal inoculation. Virology 1991;
Parasitol. 2000; 86:50–55. 183:320–330.
Jeannet, P. Y., Roulet, E., Maeder-Ingvar, M., Gehri, M., Jutzi, A., Land, L. J., Shepherd, G. M. Autoradiographic analysis of olfac-
Deonna, T. Home and hospital treatment of acute seizures in tory receptor projections in the rabbit. Brain Res. 1974; 70:
children with nasal midazolam. Eur. J. Paediatr. Neurol. 1999; 506–510.
3:73–77. Landgraf, R. Plasma oxytocin concentrations in man after differ-
Jeffery, E. H., Haschek, W. M. Protection by dimethylsulfoxide ent routes of administration of synthetic oxytocin. Exp. Clin.
against acetaminophen-induced hepatic, but not respiratory Endocrinol. 1985; 85:245–248.
toxicity in the mouse. Toxicol. Appl. Pharmacol. 1988; Lane, R. J. Intranasal lidocaine for treatment of migraine [letter].
93:452–461. JAMA 1996; 276:1553; discussion 1554.
Jensen, R. K., Sleight, S. D. Toxic effects of N-nitrosodiethy- Langston, J. W. Epidemiology versus genetics in Parkinson’s dis-
lamine on nasal tissues of Sprague-Dawley rats and golden ease: Progress in resolving an age-old debate. Ann. Neurol.
Syrian hamsters. Fundam. Appl. Toxicol. 1987; 8:217–229. 1998; 44(suppl 1):S45–S52.
Jones, D. T., Reed, R. R., Golf: An olfactory neuron specific-G Larsson, P., Tjalve, H. Intranasal instillation of aflatoxin B(1) in
protein involved in odorant signal transduction. Science 1989; rats: bioactivation in the nasal mucosa and neuronal transport
244:790–795. to the olfactory bulb. Toxicol. Sci. 2000; 55:383–391.
Jones, N. S., Quraishi, S., Mason, J. D. The nasal delivery of sys- Lavi, E., Fishman, P. S., Highkin, M. K., Weiss, S. R. Limbic
temic drugs. Int. J. Clin. Pract. 1997; 51:308–311. encephalitis after inhalation of a murine coronavirus. Lab
Jones, R. L., Nguyen, A., Man, S. F. Nicotine and cotinine Invest. 1988; 58:31–36.
replacement when nicotine nasal spray is used to quit smok- Lazard, D., Zupko, K., Poria, Y., Nef, P., Lazarovits, J., Horn, S.,
ing. Psychopharmacology (Berl) 1998; 137:345–350. Khen, M., Lancet, D. Odorant signal termination by olfactory
Kallas, H. E., Chintanadilok, J., Maruenda, J., Donahue, J. L., UDP glucuronosyl transferase. Nature 1991; 349: 790–793.
Lowenthal, D. T. Treatment of nocturia in the elderly. Drugs Lemay, A., Maheux, R., Faure, N., Jean, C., Fazekas, A. T.
Aging 1999; 15:429–437. Reversible hypogonadism induced by a luteinizing hormone-
Karachunski, P. I., Ostlie, N. S., Okita, D. K., Conti-Fine, B. M. releasing hormone (LH-RH) agonist (Buserelin) as a new
Nasal administration of synthetic acetylcholine receptor T therapeutic approach for endometriosis. Fertil. Steril. 1984;
epitopes affects the immune response to the acetylcholine 41:863–871.
receptor and prevents experimental myasthenia gravis. Ann. Levy, R. L. Intranasal capsaicin for acute abortive treatment of
NY Acad. Sci. 1998; 841:560–564. migraine without aura [letter]. Headache 1995; 35:277.
Key, B., Akeson, R. A. Olfactory neurons express a unique gly- Lewis, J. L., Dahl, A. R. Olfactory mucosa: composition, enzy-
cosylated form of the neural cell adhesion molecule (N- matic localization and metabolism. In: Doty, R. L., ed.
CAM). J. Cell Biol. 1990; 110:1729–1743. Handbook of Olfaction and Gustation. New York: Marcel
Key, B., Giorgi, P. P. Selective binding of soybean agglutinin to Dekker, 1995:33–52.
the olfactory system of Xenopus. Neuroscience 1986a; Lindquist, N. G., Lyden, A., Narfstrom, K., Samaan, H.
18:507–515. Accumulation of taurine in the nasal mucosa and the olfactory
Key, B., Giorgi, P. P. Soybean agglutinin binding to the olfactory bulb. Experientia 1983; 39:797–799.
systems of the rat and mouse. Neurosci. Lett. 1986b; 69: Lipton, R. B. Ergotamine tartrate and dihydroergotamine mesy-
131–136. late: safety profiles. Headache 1997; 37:S33–41.
Koujitani, T., Yasuhara, K., Kobayashi, H., Shimada, A., Liu, M. A. Vaccine developments. Nat. Med. 1998; 4:515–519.
Onodera, H., Takagi, H., Hirose, M., Mitsumori, K. Tumor- Louon, A., Reddy, V. G. Nasal midazolam and ketamine for pae-
promoting activity of 2,6-dimethylaniline in a two-stage nasal diatric sedation during computerised tomography. Acta.
carcinogenesis model in N-bis(2-hydroxypropyl)nitrosamine- Anaesthesiol. Scand. 1994; 38:259–261.
treated rats. Cancer Lett. 1999; 142:161–171. Lundh, B., Brockstedt, U., Kristensson, K. Lectin-binding pattern
Kowall, N. W., Pendlebury, W. W., Kessler, J. B., Perl, Beal, M. of neuroepithelial and respiratory epithelial cells in the mouse
F. Aluminum-induced neurofibrillary degeneration affects a nasal cavity. Histochem. J. 1989; 21:33–43.
570 Baker and Genter

Maassab, H. F., DeBorde, D. C. Development and characteriza- Mills, T. M., Scoggin, J. A. Intranasal lidocaine for migraine
tion of cold-adapted viruses for use as live virus vaccines. and cluster headaches. Ann. Pharmacother. 1997;
Vaccine 1985; 3:355–369. 31:914–915.
Mansour, A., Meador-Woodruff, J. H., Bunzow, J. R., Civelli, O., Monath, T. P., Cropp, C. B., Harrison, A. K., Mode of entry of a
Akil, H., Watson, S. J. Localization of dopamine D2 receptor neurotropic arbovirus into the central nervous system. Lab
mRNA and D1 and D2 receptor binding in the rat brain and Invest. 1983; 48:399–410.
pituitary: an in situ hybridization-receptor autoradiographic Moon, Y. W., Baker, H. Induction of cell division in olfactory
analysis. J. Neurosci. 1990; 10:2587–2600. basal epithelium following intranasal irrigation with wheat
Margolis, F. L., Grillo, M. Axoplasmic transport of carnosine (- germ agglutinin-horseradish peroxidase. J. Comp. Neurol.
alanyl-L-histidine) in the mouse olfactory pathway. 1998; 393:472–481.
Neurochem. Res. 1977; 2:507–519. Moon, Y. W., Baker, H. Lectin-induced apoptosis of mature olfac-
Marttin, E., Romeijn, S. G., Verhoef, J. C., Merkus, F. W. Nasal tory receptor cells. J. Neurosci. Res. 2002; 68:398–405.
absorption of dihydroergotamine from liquid and powder for- Morales, J. A., Herzog, S., Kompter, C., Frese, K., Rott, R.
mulations in rabbits. J. Pharm. Sci. 1997; 86:802–807. Axonal transport of Borna disease virus along olfactory path-
Marttin, E., Verhoef, J. C., Merkus, F. W. Efficacy, safety and ways spontaneously and experimentally infected rats. Med.
mechanism of cyclodextrins as absorption enhancers in nasal Microbiol. Immunol. 1988; 177:51–68.
delivery of peptide and protein drugs. J. Drug Target 1998; Mori, I., Komatsu, T., Takeuchi, K., Nakakuki, K., Sudo, M.,
6:17–36. Kimura, Y. Parainfluenza virus type 1 infects olfactory neu-
Maruniak, J. A., Lin, P. J., Henegar, J. R. Effects of unilateral rons and establishes long-term persistence in the nerve tissue.
naris closure on the olfactory epithelia of adult mice. Brain J. Gen. Virol. 1995; 76:1251–1254.
Res. 1989a; 490:212–218. Mori, K., Fujita, S. C., Imamura, K., Obata, K.
Maruniak, J. A., Taylor, J. A., Henegar, J. R., Williams, M. B. Immunohistochemical study of subclasses of olfactory nerve
Unilateral naris closure in adult mice: atrophy of the deprived- fibers and their projections to the olfactory bulb in the rabbit.
side olfactory bulbs. Dev. Brain. Res. 1989b; 47:27–33. J. Comp. Neurol. 1985; 242:214–229.
Mathison, S, Nagilla, R, Kompella, U. B. Nasal route for direct Narita, M., Uchimura, A., Kawanabe, M., Fukushi, H., Hirai, K.
delivery of solutes to the central nervous system: fact or fic- Invasion and spread of equine herpesvirus 9 in the olfactory
tion? J. Drug Target 1998; 5:415–441. pathway of pigs after intranasal inoculation. J. Comp. Pathol.
Matsumiya, K, Kitamura, M, Kishikawa H, Kondoh, N, 2001; 124:265–272.
Fujiwara, Y, Namiki, M, Okuyama, A. A prospective compar- National Cancer Institute. Bioassay of p-cresidine for possible
ative trial of a gonadotropin-releasing hormone analogue with carcinogenicity. Technical Report No. 142. Bethesda, MD:
clomiphene citrate for the treatment of oligoasthenozoosper- Carcinogenesis Testing Program, NCI, National Institutes of
mia. Int. J. Urol. 1998; 5:361–363. Health 1979.
McLean, J. H., Shipley, M. T., Bernstein, D. I. Golgi-like, Nef, P., Heldman, H., Lazard, D., Margalit, T., Jaye, M.,
transneuronal retrograde labeling with CNS injections of her- Hanukoglu, I., Lancet, D. Olfactory-specific cytochrome P-
pes simplex virus type 1. Brain Res. Bull. 1989; 22:867–881. 450. J. Biol. Chem. 1989; 264:6780–6785.
McLean, J. H., Shipley, M. T., Bernstein, D. I., Corbett, D. Nir, Y., Beemer, A., Goldwasser, R. A. West Nile virus infection
Selective lesions of neural pathways following viral inocula- in mice following exposure to a viral aerosol. Br. J. Exp.
tion of the olfactory bulb. Exp. Neurol. 1993; 122:209–222. Pathol. 1965; 46:443–448.
McMartin, C, Hutchinson, L. E., Hyde, R., Peters, G. E. Analysis Ohm, T., Braak, H. Olfactory bulb changes in Alzheimer’s dis-
of structural requirements for the absorption of drugs and ease. Acta. Neuropathol. 1987; 73:365–369.
macromolecules from the nasal cavity. J. Pharm. Sci. 1987; Oliver, K. R., Fazakerley, J. K., Transneuronal spread of Semliki
76:535–540. Forest virus in the developing mouse olfactory system is
Meisami, E. Effects of olfactory deprivation on postnatal growth determined by neuronal maturity. Neuroscience 1998;
of the rat olfactory bulb utilizing a new method for production 82:867–877.
of neonatal anosmia. Brain Res. 1976; 107:437–444. Peet, M. M., Echols, D. H., Richter, H. J. The chemical prophy-
Melanson, S. W., Morse, J. W., Pronchik, D. J., Heller, M. B., laxis for poliomyelitus: the technique of applying zinc sulfate
Transnasal butorphanol in the emergency department man- intranasally. JAMA 1937; 108:2184–2187.
agement of migraine headache. Am. J. Emerg. Med. 1997; Perl, D. P., Good, P. F. Uptake of aluminum into central ner-
15:57–61. vous system along nasal-olfactory pathways. Lancet 1987;
Mellert, T. K., Getchell, M. L., Sparks, L., Getchell, T. V., 1(8540):1028.
Characterization of the immune barrier in human olfactory Perlman, S., Jacobsen, G., Moore, S. Regional localization of
mucosa. Otolaryngol. Head Neck Surg. 1992; 106: virus in the central nervous system of mice persistently
181–188. infected with murine coronavirus JHm. Virology 1988;
Meredith, M., O’Connell, R. J., HRP uptake by olfactory and 166:328–338.
vomeronasal receptor neurons: use as an indicator of incom- Perlman, S., Jacobsen, G., Afifi, A. Spread of a neurotropic
plete lesions and relevance for non-volatile chemoreception. murine coronavirus into the CNS via the trigenial and olfac-
Chem. Senses 1988; 13:487–515. tory nerves. Virology 1989; 170:556–560.
Exogenous Agent Uptake 571

Perlman, S., Evans, G., Affifi, A. Effect of olfactory bulb ablation Sakane, T., Akizuki, M., Yoshida, M., Yamashita, S., Nadai, T.,
on spread of a neurotropic coronavirus into the mouse brain. Hashida, M., Sezaki, H. Transport of cephalexin to the cere-
J. Exp. Med. 1990; 172:1127–1132. brospinal fluid directly from the nasal cavity. J. Pharm.
Perras, B., Droste, C., Born, J., Fehm, H. L., Pietrowsky, R. Verbal Pharmacol. 1991b; 43:449–451.
memory after three months of intranasal vasopressin in healthy Sakane, T., Akizuki, M., Yamashita, S., Sezaki, H., Nadai, T.
old humans. Psychoneuroendocrinology 1997; 22:387–396. Direct drug transport from the rat nasal cavity to the cere-
Pevsner, J., Sklar, P. B., Snyder, S. H., Odorant-binding protein: brospinal fluid: the relation to the dissociation of the drug. J.
localization to nasal glands and secretions. Proc. Natl. Acad. Pharm. Pharmacol. 1994; 46:378–379.
Sci. USA 1986; 83:4942–4946. Sakane, T., Akizuki, M., Taki, Y., Yamashita, S., Sezaki, H.,
Pierson, M. E., Comstock, J. M., Simmons, R. D., Kaiser, F., Nadai, T. Direct drug transport from the rat nasal cavity to the
Julien, R., Zongrone, J., Rosamond, J. D., Synthesis and bio- cerebrospinal fluid: the relation to the molecular weight of
logical evaluation of potent, selective, hexapeptide CCK-A drugs. J. Pharm. Pharmacol. 1995; 47:379–381.
agonist anorectic agents. J. Med. Chem. 1997; 40:4302–4307. Sakane, T., Yamashita, S., Yata, N., Sezaki, H. Transnasal deliv-
Pietrowsky, R., Struben, C., Molle, M., Fehm, H. L., Born, J. ery of 5-fluorouracil to the brain in the rat. J. Drug. Target.
Brain potential changes after intranasal vs. intravenous 1999; 7:233–240.
administration of vasopressin: evidence for a direct nose-brain Sams, J. M., Jansen, A. S., Mettenleiter, T. C., Loewy, A. D.
pathway for peptide effects in humans. Biol. Psychiatry 1996; Pseudorabies virus mutants as transneuronal markers. Brain
39:332–340. Res. 1995; 687:182–190.
Polak, E. H., Shirley, S. G., Dodd, G. H. Concanavalin A reveals Saper, C. B., Wainer, B. H., German, D. C. Axonal and transneu-
olfactory receptors which disciminate between alkane odor- ronal transport in the transmission of neurological disease:
ants on the basis of size. Biochem. J. 1989; 262:475–478. potential role in system degenerations, including Alzheimer’s
Putcha, L., Tietze, K. J., Bourne, D. W., Parise, C. M., Hunter, R. disease. Neuroscience 1987; 23:389–398.
P., Cintron, N. M. Bioavailability of intranasal scopolamine in Sarkar, M. A. Drug metabolism in the nasal mucosa. Pharm. Res.
normal subjects. J. Pharm. Sci. 1996; 85:899–902. 1992; 9:1–9.
Rake, G. The rapid invasion of the body through the olfactory Schieferstein, G. J., Sheldon, W. G., Cantrell, S. A., Reddy, G.
mucosa. J. Exp. Med. 1937; 65:303–315. Subchronic toxicity study of 1,4-dithiane in the rat. Fundam.
Ramanathan, R., Geary, R. S., Bourne, D. W., Putcha, L. Appl. Toxicol. 1988; 11:703–714.
Bioavailability of intranasal promethazine dosage forms in Schiffman, S. S. Taste and smell in disease (first of two parts). N.
dogs. Pharmacol. Res. 1998; 38:35–39. Engl. J. Med. 1983; 308:1275–1279.
Reed, C. J., van den Broeke, L. T., De Matteis F. Drug-induced Schneider, N. G., Olmstead, R., Mody, F. V., Doan, K., Franzon,
protoporphyria in the olfactory mucosa of the hamster. J. M., Jarvik, M. E., Steinberg, C. Efficacy of a nicotine nasal
Biochem. Toxicol. 1989; 4:161–164. spray in smoking cessation: a placebo-controlled, double-
Ressler, K. J., Sullivan, S. L., Buck, L. B. A zonal organization of blind trial. Addiction 1995; 90:1671–1682.
odorant receptor gene expression in the olfactory epithelium. Schultz, E. W., Gebhardt, L. P. Prevention of intranasally inocu-
Cell 1993; 73:597–609. lated poliomyelitis in monkeys by previous intranasal irriga-
Ressler, K. J., Sullivan, S. L., Buck, L. B. Information coding in tion with chemical agents. Proc. Soc. Exp. Biol. Med. 1936;
the olfactory system: evidence for a stereotyped and highly 34:133–135.
organized epitope map in the olfactory bulb. Cell 1994; Schultz, W. W., Gebhardt, L. P. Zinc sulfate prophylaxis in
79:1245–1255. poliomyelitis. JAMA 1937; 108:2182–2184.
Roberts, E. Alzheimer’s disease may begin in the nose and may Seki, M. Vitalfarbung des Riechepithels der Maus mit
be caused by aluminosilicates. Neurobiol. Aging 1986; Trypanblau. Z Zellforsch 1941; 31:218–223.
7:561–567. Seki, T., Sato, N., Hasegawa, T., Kawaguchi, T., Juni, K. Nasal
Ruis, H., Rolland, R., Doesburg, W., Broeders, G., Corbey, R. absorption of zidovudine and its transport to cerebrospinal
Oxytocin enhances onset of lactation among mothers deliver- fluid in rats. Biol. Pharm. Bull. 1994; 17:1135–1137.
ing prematurely. Br. Med. J. (Clin Res Ed) 1981; 283:340–342. Shigematsu, K., McGeer, P. L. Accumulation of amyloid precur-
Russell, M. J., Liu, H, Nunes, R. J., Vijayan, V. Transsynaptic and sor protein in damaged neuronal processes and microglia fol-
non-transsynaptic forms of WGA-HRP. Soc. Neurosci. Abstr. lowing intracerebral administration of aluminum salts. Brain
1991; 17:60. Res. 1992; 593:117–123.
Ryzhikov, A. B., Ryabchikova, E. I., Sergeev, A. N., Tkacheva, N. Shimoda, N., Maitani, Y., Machida, Y., Nagai, T. Effects of dose,
V. Spread of Venezuelan equine encephalitis virus in mice pH and osmolarity on intranasal absorption of recombinant
olfactory tract. Arch. Virol. 1995; 140:2243–2254. human erythropoietin in rats. Biol. Pharm. Bull. 1995; 18:
Sachs, C. J. Intranasal lidocaine for treatment of migraine [letter; 734–739.
comment]. JAMA 1996; 276:1553–4; discussion 1554. Shipley, M. T. Transport of molecules from nose to brain:
Sakane, T., Akizuki, M., Yamashita, S., Nadai, T., Hashida, M., Transneuronal anterograde and retrograde labeling in the rat
Sezaki, H. The transport of a drug to the cerebrospinal fluid olfactory system by wheat germ agglutinin-horseradish per-
directly from the nasal cavity: the relation to the lipophilicity of oxidase applied to the nasal epithelium. Brain. Res. Bull.
the drug. Chem. Pharm. Bull. (Tokyo) 1991a; 39:2456–2458. 1985; 15:120–142.
572 Baker and Genter

Shipley, M. T., Nickell, W. T., Norman, A. B., Gerfen, C. Takala, A., Kaasalainen, V., Seppala, T., Kalso, E., Olkkola, K. T.
Localization of D2 DA receptor mRNA in primary olfactory Pharmacokinetic comparison of intravenous and intranasal
neurons. Soc. Neurosci. Abst. 1991; 17:1091. administration of oxycodone. Acta. Anaesthesiol. Scand.
Shirley, S., Polak, E., Dodd, G. Selective inhibition of rat olfac- 1997; 41:309–312.
tory receptors by concanavalin A. Biochem. Soc. Trans. 1983; Talamo, B. R., Rudel, R. A., Kosik, K., Lee, V. M-Y., Neff, S.,
11:780–781. Edelman, L., Kauer, J. S. Pathology of olfactory neurons in
Silverman, S. L. Nasal calcitonin. Endocrine 1997; 6:199–202. patients with Alzheimer’s disease. Nature 1989; 37:736–739.
Slot, W. B., Merkus, F. W., Van Deventer, S. J., Tytgat, G. N. Tallkvist, J., Henriksson, J., d’Argy, R., Tjälve, H. Transport and
Normalization of plasma vitamin B12 concentration by subcellular distribution of nickel in the olfactory system of
intranasal hydroxocobalamin in vitamin B 12-deficient pikes and rats. Toxicol. Sci. 1998; 43:196–203.
patients. Gastroenterology 1997; 113:430–433. Tanner, C. M., Langston, J. W. Do environmental toxins cause
Spencer, R. F., Baker, H., Baker, R. Evaluation of wheat germ Parkinson’s disease? A critical review. Neurology 1990; 40:
agglutinin immunohistochemistry as a neuroanatomical 17–30.
method for retrograde, anterograde, and anterograde transy- Tengamnuay, P., Sahamethapat, A., Sailasuta, A., Mitra, A. K.
naptic labeling in the cat visual oculomotor systems. Soc. Chitosans as nasal absorption enhancers of peptides: compar-
Neurosci. Abst. 1982; 8:785. ison between free amine chitosans and soluble salts. Int. J.
Srinivas, N. R., Shyu, W. C., Soong, C. W., Greene, D. Absolute Pharm. 2000; 197:53–67.
bioavailability and dose proportionality of BMS-181885, an Thorne, R. G., Emory, C. R., Ala, T. A., Frey, W. H. 2nd.
antimigraine agent, following the administration of single Quantitative analysis of the olfactory pathway for drug deliv-
intranasal doses to cynomolgus monkeys. J. Pharm. Sci. ery to the brain. Brain Res. 1995; 692:278–282.
1998; 87:1170–1172. Thornton-Manning, J. R., Dahl, A. R. Metabolic capacity of nasal
Stewart, W. B. Labelling of olfactory bulb glomeruli following tissue interspecies comparisons of xenobiotic-metabolizing
horseradish peroxidase lavage of the nasal cavity. Brain Res. enzymes. Mutat. Res. 1997; 380:43–59.
1985; 347:200–203. Tisdall, F. F., Brown, A., Defries, R. D., Ross, M. A., Sellers, A.
Stone, D. M., Wessel, T., Joh, T. H., Baker, H. Decrease in tyro- H. Nasal spraying as preventitive of poliomylitis. Can. Public
sine hydroxylase, but not aromatic L-amino acid decarboxy- Health J. 1937; 28:431–434.
lase, messenger RNA in rat olfactory bulb following Tisdall, F. F., Brown, A., Defries, R. D. Persistent anosmia fol-
neonatal, unilateral odor deprivation. Mol. Brain Res. 1990; lowing zinc sulfate nasal spraying. J. Pediatr. 1938; 13:60–62.
8:291–300. Tjälve, H., Henriksson, J. Uptake of metals in the brain via olfac-
Stone, D. M., Grillo, M., Margolis, F. L., Joh, T. H., Baker, H. tory pathways. Neurotoxicology 1999; 20:181–195.
Differential effect of functional olfactory bulb deafferentation Tjälve, H., Henriksson, J., Tallkvist, J., Larsson, B. S., Lindquist,
on tyrosine hydroxylase and glutamic acid decarboxylase N. G. Uptake of manganese and cadmium from the nasal
messenger RNA levels in rodent juxtaglomerular neurons. J. mucosa into the central nervous system via olfactory path-
Comp. Neurol. 1991; 311:223–233. ways in rats. Pharmacol. Toxicol. 1996; 79:347–356.
Strack, A. M., Loewy, A. D. Pseudorabies virus: a highly specific Tolis, G., Faure, N., Koutsilieris, M., Lemay, A., Klioze, S.,
transneuronal cell body marker in the sympathetic nervous Yakabow, A., Fazekas, A. T. Suppression of testicular steroido-
system. J. Neurosci. 1990; 10:2139–2147. genesis by the GnRH agonistic analogue Buserelin (HOE-766)
Stroop, W. G. Viruses and the olfactory system. In: Doty, R. L., in patients with prostatic cancer: studies in relation to dose and
ed. Handbook of Olfaction and Gustation. New York: Marcel route of administration. J. Steroid. Biochem. 1983; 19:995–998.
Dekker, 1995:367–393. Turk, M. A., Flory, W., Henk, W. G. Chemical modulation of 3-
Stroop, W. G., Rock, D. L., Fraser, N. W. Localization of herpes methylindole toxicosis in mice: effect on bronchiolar and
simplex virus in the trigeminal and olfactory systems of the olfactory mucosal injury. Vet. Pathol. 1986; 23:563–570.
mouse central nervous system during acute and latent infec- U.S. Environmental Protection Agency. Special Review of cer-
tions by in situ hybridization. Lab Invest. 1984; 51:27–38. tain pesticide products. Alachlor: Position document 1.
Sullivan, S. L., Ressler, K. J., Buck, L. B. Spatial patterning and Springfield, VA: Office of Pesticide Programs, 1985.
information coding in the olfactory system. Curr. Opin. Uchenna Agu, R., Jorissen, M., Willems, T., Van den Mooter, G.,
Genet. Dev. 1995; 5:516–523. Kinget, R., Verbeke, N., Augustijns, P. Safety assessment of
Suzuki, N. Anterograde fluorescent labeling of olfactory recep- selected cyclodextrins-effect on ciliary activity using a human
tor neurons by Procion and Lucifer dyes. Brain Res. 1984; cell suspension culture model exhibiting in vitro ciliogenesis.
311:181–185. Int. J. Pharm. 2000; 193:219–226.
Swain, R. An update of vitamin B12 metabolism and deficiency Ugolini, G. Specificity of rabies virus as a transneuronal tracer of
states. J. Fam. Pract. 1995; 41:595–600. motor networks: transfer from hypoglossal motoneurons to
Tabaton, M., Cammarata, S., Mancardi, G. L., Cordone, G., connected second-order and higher order central nervous sys-
Perry, G., Loeb, C. Abnormal tau-reactive filaments in olfac- tem cell groups. J. Comp. Neurol. 1995; 356:457–480.
tory mucosa in biopsy specimens of patients with probable Valensi, P., Zirinis, P., Nicolas, P., Perret, G., Sandre-Banon, D.,
Alzheimer’s disease. Neurology 1991; 41:391–394. Attali, J. R. Effect of insulin concentration on bioavailability
Exogenous Agent Uptake 573

during nasal spray administration. Pathol. Biol. (Paris) 1996; found in axons of olfactory receptor neurons. J Neurosci.
44:235–240. 1995; 15:4827–4837.
Vassar, R., Chao, S. K., Sitcheran, R., Nunez, J. M., Vosshall, L. Wiley, R., Baker, H., Baker, R. Concavalin A-horseradish perox-
B., Axel, R. Topographic organization of sensory projections idase (CON A-HRP) conjugate is a useful neuroanatomical
to the olfactory bulb. Cell 1994; 79:981–991. tracer. Soc. Neurosci. Abst. 1984; 10:420.
Wallace, S. J. Nasal benzodiazepines for management of acute Yajima, T., Juni, K., Saneyoshi, M., Hasegawa, T., Kawaguchi, T.
childhood seizures? Lancet 1997; 349:222. Direct transport of 2,3-didehydro-3-deoxythymidine (D4T)
Watanabe, K., Kawana, E. Selective retrograde transport of triti- and its ester derivatives to the cerebrospinal fluid via the nasal
ated D-aspartate from the olfactory bulb to the anterior olfac- mucous membrane in rats. Biol. Pharm. Bull. 1998; 21:
tory nucleus, pyriform cortex and nucleus of the lateral 272–277.
olfactory tract in the rat. Brain Res. 1984; 296:148–151. Zatta, P., Favarato, M., Nicolini, M. Deposition of aluminum in
Weiss, D. G., Buchner, K. Axoplasmic transport in olfactory brain tissues of rats exposed to inhalation of aluminum acetyl-
receptor neurons. In: Margolis F. L., Getchell, T. V., eds. acetonate. Neuroreport 1993; 4:1119–1122.
Molecular Neurobiology of the Olfactory System. New York Zhao, H., Otaki, J. M., Firestein, S. Adenovirus-mediated gene
Plenum Press, 1988:217–236. transfer in olfactory neurons in vivo. J. Neurobiol. 1996; 30:
Weiss, D. G., Krygier-Brevart, V., Gross, G. W., Kreutzberg, G. 521–530.
W. Rapid axoplasmic transport in the olfactory nerve of the Ziegler, D., Ford, R., Kriegler, J., Gallagher, R. M., Peroutka,
pike: II. Analysis of transported proteins by SDS gel elec- S., Hammerstad, J., Saper, J., Hoffert, M., Vogel, B., Holtz,
trophoresis. Brain Res. 1978; 139:77–87. N., et al. Dihydroergotamine nasal spray for the acute treat-
Weiss, P., Holland, Y. Neuronal dynamics and axonal flow in the ment of migraine [see comments]. Neurology 1994; 44:
olfactory nerve as model test object. Proc. Natl. Acad. Sci. 447–453.
USA 1967; 57:258–264. Zupko, K., Poria, Y., Lancet, D. Immunolocalization of
Wensley, C. H., Stone, D. M., Baker, H., Kauer, J. S., Margolis, cytochromes P-450olfl and P-450olf2 in rat olfactory mucosa.
F. L., Chikaraishi, D. M. Olfactory marker protein mRNA is Eur. J. Biochem. 1991; 196:51–58.
27

Influence of Environmental Toxicants on Olfactory Function

Lloyd Hastings
University of Pennsylvania, Philadelphia, Pennsylvania, U.S.A.

Marian L. Miller
University of Cincinnati College of Medicine, Cincinnati, Ohio, U.S.A.

I. INTRODUCTION In this chapter, we review the literature noting diminu-

tion or loss of the sense of smell from occupational
The sense of smell, besides being essential for the enjoy- exposure to airborne chemicals. Additionally, we provide
ment of food and contributing to various aspects of quality an overview of the mucosal pathophysiology that occurs
of life, serves the sentinel function of warning the organism after exposure to a number of toxic compounds.
of dangerous elements in the air. Hence, damage to this sys-
tem is of considerable consequence. Among the causes of
such damage is exposure to volatile toxic chemicals, most II. HISTORICAL BACKGROUND
commonly in industrial environments. Unfortunately, it is
often difficult to establish a causal relationship between Ramazzini (1713), in one of the first books written on occu-
such exposure and dysfunction. This is due, in part, to the pationally related diseases, noted only 1 of over 50 profes-
paucity of sound research concerning the effects of specific sions as being associated with “loss of smell” —that of
chemical agents on the ability to smell. painters. Exposure to extremely purulent odors was an inte-
Exposure to toxic volatiles at levels sufficient to rapidly gral part of many of the occupations examined, as was expo-
impair olfactory function usually occurs by accident. In sure to excessive concentrations of heavy metals, dusts, and
most such cases, the acute exposure is at a sufficiently high toxic chemicals. The often morbid or lethal effects resulting
concentration that physical damage to the olfactory epithe- from exposure to these toxic compounds, such as emphy-
lium and related tissues rapidly ensues. More insidious are sema after exposure to mine dusts, perhaps overshadowed
situations where subtle cumulative damage follows any concern about the viability of the sense of smell.
chronic exposure to relatively low levels of compounds In the centuries since Ramazzini’s early work, more than
that produce no noticeable irritation or untoward sensa- 100 anecdotal reports and/or case histories of olfactory loss
tions. In such cases, the pathological changes may be so resulting from exposure to toxic compounds have been
gradual that inferences of causality are difficult to make. published. However, fewer than two dozen studies have
Moreover, the chemical exposure may exacerbate damage even attempted to address this question quantitatively, and
to the olfactory system from other causes (e.g., viral infec- these studies only appeared within the last 25 years. This
tions) and vice versa, obscuring the specific influences of dearth of knowledge, and perhaps interest, is reflected in a
the exposure. recent review of the cognitive, psychomotor, and sensory

575
576 Hastings and Miller

effects resulting from exposure to solvents, in which the Table 1 Substances Affecting Olfaction for Which There Are
sense of smell was not assessed in any of the 15 studies Some Supportive Scientific Data
cited (Gamble, 2000). This is in spite of the fact that the Metallic compounds acetophenone
nasal passages are an obvious target site for airborne sol- cadmium compounds benzene
vents and that sensory measurement is usually less ambigu- cadmium oxide benzine
ous in terms of outcome than cognitive measurement. dichromates butyl acetate
Two relatively recent reviews have listed numerous nickel hydroxide chloromethanes
volatile chemicals or manufacturing processes that reportedly zinc chromate ethyl acetate
adversely influence olfaction (Amoore, 1986; Naus, 1975). alum menthol
Those listed in Amoore’s review, which subsumes most of arsenic compounds pentachlorophenol
chromic acid trichloroethylene
those reviewed by Naus, are presented in Table 1. Of the 75
silver nitrate acetic acid
or more reports noted by Amoore of relationships between
copper arsenite acetonitrile
exposure to toxic compounds and loss of olfactory function, Metallurgical processes benzoic acid
most were descriptive accounts occurring before 1970. chromium plating benzaldehyde
Rarely were industrial hygiene data pertinent to airborne lead smelting chloroform
exposures available, and even when available, they were, at magnet production dimethyl sulfate
best, crude estimates. Furthermore, the indications of smell mercury (chronic intoxication) furfural
dysfunction were largely subjective, as accurate olfactory nickel plating formaldehyde
measurement was generally lacking. In many of these studies, nickel refining (electrolytic) trichloroethane
workers were clearly exposed to exceedingly high concentra- silver plating m-xylene
tions of toxicants in the workplace. Such nonhygienic factory steel production Dusts
zinc production cement
conditions are much less likely to occur today in developed
aluminum fumes chemicals
industrial societies because of improved worker education
copper fumes hardwoods
and increased governmental regulations. Unfortunately, this manganese fumes lime
is not the case in many developing countries. tin fumes printing
Three additional reviews have appeared in the decade arsenic cotton
after the account of Amoore that have focused on more Nonmetallic inorganic cyanides
contemporary studies (Doty and Hastings, 2001; Schwartz, compounds flax
1991; Shusterman and Sheedy, 1992). Shusterman and ammonia flour
Sheedy reviewed the effects of toxic compounds on the carbon disulfide potash
special senses, including olfaction, while Schwartz carbon monoxide silicon dioxide
focused on the use of epidemiology to discern the rela- chlorine Manufacturing processes
hydrazine acids
tionship between workplace exposure to toxic agents and
nitrogen dioxide asphalt
olfactory sensory dysfunction. Doty and Hastings focused
sulfur dioxide cutting oils
on nasal anatomy, methods of assessment, and more con- fluorides fragrances
temporary empirical studies. In the early 1990s, the first of sulfuric acid lead paints
a series of studies investigating the effects of exposure to bromine paprika
airborne toxicants on upper airway function appeared flue gas coal tar fumes
(Cometto-Muniz and Cain, 1991), examining in detail the hydrogen chloride rubber vulcanization
effects of airborne contaminants on trigeminal (CN V) free hydrogen fluoride perfumes (concentrated)
nerve endings within the nose and, in some cases, the eye. nitric acid spices
Exposure to irritants often causes a reflexive depression in phosgene tobacco
respiratory rate, and this reflexive response has frequently selenium dioxide varnishes
sewer gas wastewater
been used as a guide in establishing permissible exposure
Organic compounds tanning
limits (PELs) and threshold limit values (TLVs). The TLV,
acetaldehyde
much like the PEL, reflects limits to airborne concentra-
tions of substances “under which it is believed that nearly Source: Adapted from Amoore, 1986.
all workers may be repeatedly exposed day after day
without adverse effect” (American Conference of
Governmental Industrial Hygienists, 1994). It has not been
adequately determined, however, whether protracted expo-
sure under the TLV standards produces olfactory deficits.
Influence of Environmental Toxicants on Olfactory Function 577

The problem of acquiring reliable data to be used to dis- larger airborne particles, volatile toxic vapors can still
cern cause and effect, as well as to set standards, is oner- reach this vulnerable area. Even when active sniffing
ous. Severe limitations are present in human research and does not occur, e.g., during conditions of normal resting
epidemiological studies, and animal research, which can inspiration (0.15–0.25 L/sec), 10–15% of the airflow is
be more tightly controlled, fails to provide an easy extra- shunted towards the olfactory cleft (Swift and Procter,
polation to humans, because of differences in the structure 1977). The remainder flows over the lower turbinates,
and function of the nasal passages and olfactory epithe- through the nasopharynx, and into the trachea. This flow
lium. Nevertheless, there is increasing awareness that results in not only warming, cleansing, and humidifying
olfaction should be examined whenever there is airborne the major portion of the airstream, but also exposes the
exposure to toxic compounds (Suruda, 2000). turbinates and the olfactory mucosa to xenobiotics
contained in the air.
III. OLFACTION AND TOXIC INSULT
A. The Olfactory System B. Mechanisms of Toxic Insult

In recent years, much progress has been made in under- As pointed out previously, the process of nasal inspiration
standing the basic mechanisms of olfaction, especially in ensures that the olfactory receptor neurons in the nasal
the area of molecular biology (see Chapters 3–9). mucosa will come into contact with deleterious agents in
However, the effects of toxic compounds on this system the inspired air. To counter this vulnerability, the system
are still poorly understood. Even less well known are the employs a number of defense mechanisms. As a first line
specific compounds that actually pose a risk to humans, as of defense, free nerve endings of the trigeminal nerve
well as the actual prevalence of incidents of toxicant- detect and respond to airborne sensory irritants. Usually a
induced dysfunction occurring in the workplace. Toxic flight response is initiated by the organism upon detection
compounds can affect olfactory function either indirectly, of an irritant. If escape is not possible, the breathing rate
e.g., by causing irritation in the nasal passage, blocking the and/or pattern are altered in an attempt to minimize the
flow of air to the olfactory epithelium, or directly, e.g., by entry of the irritant into the airways. As a second line of
altering the viability of the sensory epithelium. Either way, defense, the nasal passages actively participate in the
airflow patterns within the nasal cavity are important fac- metabolism of these xenobiotics through the presence of
tors. In the first case, the air carrying the odorant stimuli is phase I and phase II enzymes—cytochrome P-450s, glu-
prevented, at least to some degree, from reaching the sen- tathione, and related enzymes—which are found there in
sory receptor cells; in the second, the distribution of the concentrations that can be greater, on a per gram basis,
toxic agents within the inspired air onto the neuroepithe- than in the liver or lungs (Reed, 1993). Since the nasal pas-
lium is dependent on the local airflow patterns. sages are the first site of internal exposure, these tissues are
In the human, the olfactory neuroepithelium lines the understandably involved in the detoxification process.
cribriform plate and sectors of the superior turbinate, However, it should be pointed out that, in some cases, the
superior septum, and middle turbinate.* The total area of resultant metabolites can be toxic themselves, especially to
the neuroepithelium has been estimated to be approxi- the olfactory receptor neurons (Brandt et al., 1990) The
mately 2 cm2 (for both sides) and is positioned well away nasal mucosa is also capable of secreting antibodies, as
from the main airstream. However, the total surface area well as antimicrobial proteins such as lactoferrin and
of the receptive elements, the sensory cilia, is, perhaps, lysozyme, to deal with inhaled pathogens (Mellert et al.,
10 times as great (Doty, 1998). Even though the epithe- 1992) (see also Chapter 3). Unfortunately, these protective
lium is situated well away from the major flow of air, mechanisms in the olfactory epithelium may fail or be cir-
affording it some degree of protection from exposure to cumvented, as evidenced by the occurrence of olfactory
deficits after some toxic exposures.
It should also be acknowledged that while the periph-
eral olfactory mucosa may be the site of insult for many
*The olfactory system (CN I) is not the only sensory system pre-
airborne toxins, the possibility exists that, in some cases,
sent in the nasal cavity that responds to chemical stimuli. Besides
the fine nerve endings of CN V, which respond to chemical irri- the site of damage may be more central. A variety of com-
tants, there is also the vomeronasal organ (Jacobson’s organ), the pounds, including metals and solvents, can pass from the
septal organ, and the nervus terminalis (CN 0). Other than the CN nasal lumen via the olfactory receptor neurons to the olfac-
V, the exact functions of the remaining systems are ill defined, as tory bulb and even, in some cases, to higher brain regions.
are their contributions to olfactory function. The effects of entry of toxic compounds into the brain via
578 Hastings and Miller

this route on olfactory function and higher brain functions being metabolized into reactive intermediates (Reed, 1993).
are largely unknown (see Chapter 26). Genetic variations in the ability to metabolize such com-
Lesions in the olfactory epithelium induced by toxins pounds, and the state of induction of constituent enzymes in
have distribution patterns that reflect the patterns of chemi- the olfactory epithelium, play major roles in their ultimate
cal uptake, local clearance processes, and cell types pre- toxicity. The competency of the host also modulates the
sent within, around, and beneath the epithelium, as well as severity of the response to toxins through genes regulating
a host of other factors. Some of these conditions are physi- such systems as immune responses, chemokine production,
cal, such as air-phase delivery of chemicals and airflow vascular integrity, age, gender, and growth factors, etc.,
patterns, which can exhibit considerable across-and which govern regeneration and wound healing.
within-species variation. Some are related to the interac- A typical toxic insult to the rodent epithelium damages
tions between the olfactory tissue and the chemical agent, the cells that lie above the basement membrane (see
and others are related to the metabolic fate of the chemi- Fig. 1). The degree and depth of this injury is specific for
cals, indirectly affecting the surrounding tissue each compound and dose. Recovery of the epithelium is in
(Brenneman et al., 2000). To assess potential effects of part related to the degree of damage and in part to the cell
nasal airflow on lesion location and severity, Kimbell and types that have been damaged. For example, olfactory
coworkers (1993, 1997) used computational fluid dynam- epithelium can be denuded down to the basement mem-
ics to map the regional dosimetry of inhaled chemicals in brane, while adjacent respiratory epithelium remains
the upper respiratory tract. By using two physiochemical intact. In cases of complete damage to the olfactory
principles, water solubility and reactivity, in conjunction epithelium, save the basal cell layer, regeneration may
with simulated regional flow velocities, they developed begin within 24 hours, with the covering of the naked
models that predict if and where lesions will develop in the basement membrane with squamous cells from surround-
nasal cavity after exposure to a specific agent. The devel- ing areas. By 4 days this epithelium becomes more
opment of such models holds great promise, as it allows cuboidal, and by 30 days it increases in differentiation and
rapid testing of many compounds without the expense thickness to become a more or less mature olfactory
associated with conducting animal studies. epithelium (Hastings et al., 1991). The rapidity with
To determine the effects of toxic compounds on the
which the rodent olfactory epithelium regains its architec-
peripheral olfactory system, the olfactory epithelium
tural integrity and functionality depends to some degree
should be examined, ideally, after exposure to the toxicant.
upon how heavily the tissues beneath the basement mem-
In the case of humans, two square centimeters of tissue
brane, i.e., in the lamina propria, are damaged. In this
would appear, on the surface, to provide sufficient olfac-
respect, when there is a loss of Bowman’s glands beneath
tory epithelium to permit biopsy and examination.
the olfactory epithelium, the regeneration is considerably
However, because of its relative inaccessibility, its irregu-
slower. The participation of Bowman’s glands and ducts
lar and often spotty distribution, and inherent surgical risks
(which surface above the basement membrane of the
(penetrating the cribriform plate), human olfactory epithe-
lial biopsy has rarely been done to address histopathologi- olfactory epithelium) appears to be critical in the efficient
cal issues. To worsen matters, histological examination of and complete regeneration of the epithelium in animals
biopsied material can be hampered by the limited amount exposed to olfactotoxins (Uraih et al., 1987).
of material sampled and the loss of associated structures A factor that may influence the outcome from expo-
and orientation. Similarly, physiological measures in the sure of the olfactory epithelium to toxins is the health
nasal cavity, such as mucociliary flow, are influenced by status of the host. For example, an upper respiratory
all elements of the nasal epithelium, of which the olfactory infection that, in and of itself, may not inflict severe dam-
epithelium contributes only a small portion, and therefore age to the olfactory epithelium may do so when present
they do not provide specific data on the structural and at the same time as a toxic exposure. In other words,
functional status of the olfactory epithelium. synergism can produce greater damage than that which
For these reasons, experimental animals have been used would result from either insult alone. Conversely, just as
to determine mechanisms of action of toxic compounds in exposure to a toxic compound may work in a synergistic
the nasal cavity. Many specific “olfactotoxins” have been fashion with weakened defense mechanisms to cause
identified using animals. Some act immediately and directly greater than normal damage, induction of some enzymes
(e.g., methyl bromide), causing the sloughing of the olfac- by exogenous agents (e.g., as a result of cigarette
tory epithelium down to the basement membrane within 24 smoking) may protect against olfactory insult resulting
hours of exposure (Hastings et al., 1991). Others (e.g., from exposure to some chemicals. Clearly, the specific
3-methylfuran and dimethylamine) act indirectly only after factors that increase the susceptibility of the olfactory
Influence of Environmental Toxicants on Olfactory Function 579

few patients of a larger number who were treated with

vitamin A seemed to regain function. The latter observa-
tion failed to take into account the fact that spontaneous
recovery occurs in a number of patients without treat-
ment. The development of an effective treatment strategy
for olfactory dysfunction is one of the most pressing
needs that faces clinicians working in the area of smell
and taste disorders.

D. Nomenclature of Olfactory Dysfunction

The degree of severity of olfactory dysfunction varies

along a continuum. Based upon quantitative testing (see
Chapter 10), olfactory loss can range from mild hyposmia
(decreased sensitivity) to total anosmia (complete loss)
(see Chapter 22). Although anosmia can occur as a result
of exposure to toxic compounds, hyposmia is the more fre-
Figure 1 (Left) Olfactory mucosa from a control rat shows a quent outcome. A toxicant can also produce altered or dis-
normal olfactory epithelium (E) and lamina propria. Bowman’s torted perception of odors (dysosmia) that occurs either in
glands (G) and nerve bundles (N) are located below the basement response to a stimulus (parosmia) or independently (phan-
membrane (dotted line). (Middle) After methyl bromide expo- tosmia). Among the particularly annoying dysosmias are
sure, the epithelium (E) is almost completely denuded, but even
those described as being like the odor of feces or rot
at the point of most severe damage, the lamina propria remains
(cacosmia), medicine, or burnt toast. Although rare, there
essentially intact with glands (G) and nerve bundles (N) being
present. (Right) After 3-methyl indole however, the endothelium are anecdotal reports of some dysosmias resolving by
(E) is reduced in height, but the lamina propria remains severly removing a specific chemical from the patient’s environ-
atrophic, with a loss of glands and neural elements, and recovery ment. For example, Emmet (1976) described the case of a
is protracted; (V) blood vessel. (250 .) pipefitter exposed chronically to tetrahydrofuran-contain-
ing pipe cement who complained of a constant unpleasant
smell distinct from pipe cement (parosmia); recovery
occurred after removal from occupational exposure to the
epithelium to permanent damage from exposure to toxic cement. Shusterman and Sheedy (1992) reported that a
compounds require further investigation. woman exposed to chloramine gas complained of a “sting-
ing” in the nasopharynx in response to normally unoffen-
C. Therapeutic Approaches sive household odors; recovery occurred without therapy.

Therapeutic approaches to ameliorate olfactory dysfunc-

IV. TOXIC METALS
tion resulting from upper respiratory infections, head
trauma, etc., are very limited. The same is true when dys- A. Cadmium
function results from exposure to toxic agents. Once the
olfactory mucosa has been severely traumatized, neither Cadmium is often cited as the “textbook” example of a
oral nor topical steroid therapy has proven to be particu- toxic compound to which exposure results in anosmia
larly beneficial. While zinc supplementation has been (World Health Organization, 1986). In one of the first cad-
suggested as a therapeutic remedy, no positive value has mium/olfactory studies, Friberg (1950) reported that 37%
been substantiated in double-blinded, placebo-controlled of workers in an alkaline battery factory showed impaired
studies of patients suffering from olfactory dysfunction olfaction after an exposure history of 20 years or more.
(Henkin et al., 1976). Another suggested, but unproven, Thus, cadmium-exposed workers scored significantly
therapeutic strategy is intramuscular injections of beta- poorer on tests of olfactory function than age-matched,
carotene (vitamin A) (Duncan and Briggs, 1962) or oral nonexposed controls; 27% of these workers required 200
retinoid therapy (Roydhouse, 1988). This therapy arose, times the stimulus concentration required by normals to
however, from two questionable sources: the mistaken detect the odor of phenol. Interestingly, about half of these
notion that pigmentation plays a role in olfactory trans- workers were unaware of their dysfunction, conceivably
duction in humans and a physician’s observation that a reflecting its slow development. Physical examination of
580 Hastings and Miller

the nasal respiratory mucosa failed to yield any relation- actual exposure levels were still relatively low. Olfactory
ship between degree of irritation and the olfactory function was assessed using the controversial blast injec-
dysfunction. tion procedure (Elsberg and Levy, 1935). Anosmia or
Potts (1965) found olfactory deficits in approximately hyposmia was found in 45.2% of the exposed group but in
60% of workers exposed to cadmium for 10–29 years and only 4.6% of the controls. The authors presented some evi-
deficits in over 90% of those exposed for more than 30 dence that cigarette smoking, which is a considerable
years. These workers were exposed to extremely high con- source of cadmium in itself, may intensify the dysfunction
centrations of cadmium in a setting of poor industrial associated with occupational cadmium exposure. This is
hygiene, where measurements of airborne cadmium contrary to what is seen with exposure to solvents and irri-
ranged from 0.600 to 236.0 mg/m3. A contemporaneous tant gases, where smoking appears to elicit some protective
study of cadmium smelter workers reported no change in mechanism (Schwartz et al., 1989).
olfactory function, although exposure was sufficient to The results of the last three studies, despite involving
result in proteinuria (Tsuchiya, 1967). Both the duration exposure concentrations that are much lower than those in
(less than 12 years) and the exposure concentration (0.133 studies that took place in the 1950s and 1960s, are in
mg/m3), however, were significantly less than in previous accord with the results of earlier ones and further imply
studies. that, even at low concentrations, cadmium exposure dam-
In a more recent study of a cohort of individuals who ages olfactory function. This conclusion is further sup-
had worked for 5 years or longer in a cadmium-refining ported by a recent case report of anosmia in two patients
plant, 28% reported being anosmic (Yin-Zeng et al., 1985). after cadmium exposure that emphasized the need for
The classification of anosmia, however, was limited to qualitative olfactory testing after exposure to toxic com-
self-report, as no qualitative measures of olfactory function pounds (Suruda, 2000).
were obtained. At the time of this study, the average con-
centration of airborne cadmium was between 0.004 and B. Chromium
0.187 mg/m3; earlier concentrations of cadmium ranged
between 0.21 and 0.95 mg/m3. This work suggests the pos- Chromium, like cadmium, is often employed with nickel
sibility that even moderate exposure to cadmium adversely and other metals in industrial situations, most notably in the
affects olfactory function. In a relatively extensive quanti- manufacture of high-quality steel alloys. Although
tative study, 55 workers exposed to cadmium fumes (from chromium has long been suspected as having an adverse
brazing refrigerator coils) for a duration of approximately influence on olfaction, there has been relatively little
12 years were evaluated using both a butanol odor–detec- research on the effects of chromium on the ability to smell.
tion threshold and odor discrimination tests (Rose et al., Using the T & T olfactometer (Oka, 1981) (see Chapter 10),
1992). Thirteen years after production began, the first Watanabe and Fukuchi (2000) examined odor detection and
measurements of airborne cadmium in the workplace were recognition thresholds of 26 male and 7 female employees
determined to be 0.300 mg/m3. Those workers with the of a chromate-producing factory who had worked there for
highest body burden of cadmium, determined by the mea- at least 7 years. Over half of the subjects (51.4%) were noted
surement of 2-microglobulin in their urine, showed mod- as having perforated nasal septa; 54.5% exhibited elevated
erate to severe hyposmia, but not anosmia. No deficits in smell thresholds to the five odorants evaluated (phenyl
odor discrimination, however, were observed in these ethanol, cyclopentenolone, isovaleric acid, r-undecalactone,
workers. Since the cadmium-related hyposmia was spe- and skatole), with two reportedly being anosmic. While the
cific to detection threshold, and not to discrimination degree of olfactory function was not related to the presence
behavior, it was concluded that the peripheral olfactory of the septal perforations, a common symptom associated
receptor neurons were damaged, but the more central com- with chromium exposure, it was related to the duration of
ponents of the system were not. While this interpretation is employment in the factory.
often evoked when deficits are found in threshold, but not
identification, measures, the rationale for doing so is ques- C. Mercury
tionable (see Chapters 10 and 21).
The olfactory function of 73 workers at a plant in Metals other than cadmium and chromium can produce
Poland involved in the production of nickel-cadmium bat- olfactory deficits. Exposure to high levels of organic mer-
teries was recently compared with that of 43 nonexposed, cury in utero, as in Minamata disease, both raised detec-
age- and smoking-matched controls (Sulkowski et al., tion thresholds for serial dilutions of phenyl ethyl alcohol
2000). While the exposure levels exceeded the maximum and decreased olfactory identification ability, as measured
allowable concentration (Polish MAC
20 g/m3), the by the University of Pennsylvania Smell Identification Test
Influence of Environmental Toxicants on Olfactory Function 581

(UPSIT) (Furuta et al., 1994). The olfactory epithelium not considered by the authors as meaningful, since they
and olfactory bulb were reported as being grossly normal concluded that olfactory function was unaffected by lead
at autopsy in these patients, although some gliosis and exposure, (Bolla et al., 1995).
neuronal degeneration were evident in the bulb. Since most
bulb alterations are secondary to epithelial damage, quan- E. Other Metals
titative assessment of the epithelium of such patients is
sorely needed. Over a dozen metals or metallurgical processes are con-
sidered as adversely influencing olfaction (Tables 1, 2),
D. Lead although, as noted above, only cadmium has received
extensive investigation. There is circumstantial evidence
Exposure to inorganic or organic lead can perturb the cen- that several other metals, such as aluminum and man-
tral nervous system in many ways, from subtle changes in ganese, need to be investigated in detail. Thus, workers
cognitive function to frank encephalopathy. While the involved in aluminum reclamation were found to self-
effects of lead exposure on various sensory systems, report a reduction in the sense of smell twice as often as
including the visual, auditory, and somatosensory, are well controls (Kilburn, 1998). Odor-detection thresholds for
documented (Chuang et al., 2000; He et al., 2000; Wu n-butanol were paradoxically reported in one study to be
et al., 2000), little attention has been paid to whether such lower (i.e., higher sensitivity) in a group of 115 man-
exposure effects olfactory function. Even more surprising ganese-exposed iron alloy workers than in controls
is the fact that no studies have been undertaken to assess (Mergler et al., 1994). The overall airborne manganese
the effects of pediatric lead exposure on olfaction, since concentration in total dust ranged from 0.014 to 11.48
children are much more sensitive than adults to the neuro- mg/m3 (mean
0.225 mg/m3; TLV
1 mg/m3). The
toxic effects of low level lead exposure. authors suggested that this enhanced olfactory acuity was
Schwartz et al. (1993) tested 222 employees involved in temporary and may reflect an excitatory phase of early
the manufacture of tetraethyl lead on a neurobehavioral test manganese intoxication, which might be followed by a
battery and on their ability to identify odors employing the subsequent impairment of olfactory perception. In con-
UPSIT. The testing was performed on site and before a trast, a second study examining n-butanol detection
work shift to minimize any acute effects that the lead might thresholds of 35 manganese workers exposed to roughly
exert. Comprehensive personal industrial hygiene data were comparable levels and durations of manganese
available, allowing accurate exposure histories to be recon- (0.026–0.750 mg/m3, mean
0.193 mg/m3) as the previ-
structed. Mean differences in the test scores were estimated ous one found no effects of the metal on smell function
by comparing the average scores of moderate, high, and (Lucchini et al., 1997). Nevertheless, a significant nega-
highest exposure groups to those of the low exposure (ref- tive correlation was found between manganese levels in
erence) group, and adjusting for the confounding influences the urine and olfactory perception threshold.
of age, intellectual ability, race, and alcohol consumption.
No influence of lead on UPSIT scores was apparent, even
though lead exposure was associated with decrements in a F. Utility of Animal Studies
number of the neuropsychological measures.
In a companion study, Bolla et al. (1995) compared a Animal studies, necessitated by the ethical issues sur-
subset of the tetraethyl lead–exposed workers discussed rounding human exposure studies, allow for the assess-
above to workers exposed to solvents as well as to workers ment of histopathological changes in the olfactory
exposed to neither compound. The same olfactory test and epithelium after known exposures to specific compounds.
neurobehavioral test battery were administered to all par- In rats exposed to 0.500 mg/m3 cadmium for 20 weeks,
ticipants. As before, the overall results showed that, after cardiac and pulmonary toxicity was evident, but the expo-
appropriate adjustments for confounding variables, the sure regimen was insufficient to produce an effect on res-
UPSIT scores of lead-exposed workers did not differ sig- piratory or olfactory tissues (Sun et al., 1996). Nor was
nificantly from those of the reference group (respective there any evidence of olfactory dysfunction when assessed
mean scores
36.4 and 36.9). However, when duration of using an operant detection threshold measure. Similar
exposure was included in the analysis, the UPSIT scores of studies have been conducted using nickel and chromium.
the middle exposure group (11–17 years) were signifi- Although exposure to nickel sulfate at 6 times the TLV
cantly lower than the UPSIT scores of the reference group. produced damage in the rodent olfactory epithelium, olfac-
This was not the case, however, for the other two groups tory function was not affected (Evans et al., 1995). In a
(11 and 17 years). This nonlinear effect was apparently similar paradigm, no apparent consequences of exposure
Table 2 Effects of Toxic Agents on Olfactory Mucosa of Common Laboratory Rodents (Mice, Rats, Hamsters)
Compound,
concentration Duration Effects on histology and function Ref.
Acetaldehyde, 6 hr/d 5 d/wk Degeneration, metaplasia, loss of Appelman et al.,
400–5000 ppm 1–28 mo Bowman’s glands and nerve bundles, 1982; Woutersen
adenomas, squamous cell carcinomas et al., 1984
Woutersen et al.,
1986
Acrolein, 6 hr/d 5d Hypertrophy, hyperplasia, erosion, Buckley et al., 1984
1.7 ppm ulceration, necrosis inflammation
Acrylic acid, 6 hr/d 5 d/wk Degeneration, replacement with respiratory Miller et al., 1981
5–75 ppm 13 wk epithelium, inflammation, hyperplasia
of Bowman’s glands
Benomyl, 6 hr/d 6 d/wk Degeneration Warheit et al., 1989
50–200 mg/m3 13 wk
Bromobenzene, [5 min–3 d] Degeneration of olfactory epithelium and Brittebo et al., 1990
25 mol/kg ip Bowman’s glands
Cadmium, 5 hr/d 5 d/wk Little change Sun et al., 1996
250–500 g/m3 20 wk
Chlorine gas, 6 hr/d 5 d/wk Degeneration, septal perforations, intracellular Wolf et al., 1995
0.4–11 ppm 16 wk deposits of eosinophilic material, mucus cell
hypertrophy
Chloroform, 6 hr/d 7 d Degeneration of Bowman’s glands, cell proli- Mery et al., 1994
300 ppm feration in periosteum and bone
Chloropicrin, 6 hr/d 5 d Hypertrophy, hyperplasia, ulceration, necrosis, Buckley et al., 1984
8 ppm inflammation
Coumarin, [48 hr] Necrosis, cell loss, and basal cell metaplasia Gu et al., 1997
50 mg/kg ip in the olfactory mucosa
Chlorthiamid, [8 hr-7 d] Degeneration of olfactory epithelium and Brittebo et al., 1991
6–50 mg/kg ip Bowman’s glands, replacement with respiratory
epithelium fibrosis in lamina propria
Dibasic esters, 4 hr/d Degeneration, sustentacular cells injured Keenan et al., 1990;
20–900 mg/m3 7–13 wk initially, cell proliferation Bogdanffy and
Frame, 1994
1,2-Dibromo-3- 6 hr/d 5 d/wk Degeneration, metaplasia, hyperplasia Reznik et al.,1980
chloropropane, 13 wk
5–60 ppm
1,2-Dibromo- 6 hr/d 5 d/wk Degeneration, metaplasia, hyperplasia Reznik et al.,1980
ethane, 13 wk
3–75 ppm
Dichlobenil [8 hr-7d] Degeneration of olfactory epithelium and Brandt et al., 1990
12–50 mg/kg ip Bowman’s glands
1,3-Dichloropropene, 6 hr/d 5 d/wk Degeneration and /or metaplasia Scott et al. 1988,
30–150 ppm 6–24 mo Lomax et al., 1989
Dimethylamine 6 hr/d 5 d/wk Degeneration, loss of nerve bundles, Buckley et al., 1984,
10–511 ppm 6–12 mo hypertrophy of Bowman’s glands 1985
1,4-Dithiane, [90 d] Anisotrophic crystals in giant cells Schieferstein et al.,
105–420 (undetermined chemical composition) 1988
mg/kg ig
Epichlorohydrin, 6 hr/d 5 d Ulceration, necrosis Buckley et al., 1984
687 ppm
Ferrocene, 6 hr/d 5 d/wk Iron accumulation, necrotizing inflammation, Nikula et al., 1993
3–30 mg/m3 13 wk metaplasia
Formaldehyde, 6 hr/d 5 d Decrease number of bipolar cells, increase Apfelbach et al.,
0.25–15 ppm 4 mo number of basal cells, degeneration of 1991, 1992
nerve bundles, reduced odor discrimination)
Furfural, 7 hr/d 5 d/wk Disorientation of sensory cells, degeneration Feron and Kruysse,
250–400 ppm 52 wk of Bowman’s glands, cyct-like structures 1978
in lamina propria
Table 2 (continued)
Compound,
concentration Durationa Effects on histology and function Ref.
Furfural alcohol, 13 wks Squamous and respiratory metaplasia of Miller et al., 1991
2–250 ppm olfactory epithelium, inflammation, hyaline
droplets, squamous metaplasia of ducts
Hexamethylene 6 hr/d 5 d/wk Degeneration, mucus hyperplasia Foureman et al.,1994
diisocyanate, 12 mo
0.005–.175 ppm
,ⴕ-Iminodi- [6 hr-56 d] Degeneration of axon bundles, increase of Genter et al., 1992
propionitrile, glial fibrillary acidic protein
200–400 mg/kg ip
Methyl bromide, 4 hr/d 4 d/wk Degeneration, decreased carnosine, Hastings et al., 1991
200 ppm behavioral deficits
3-Methylfuran, 1 hr Degeneration, more severe in rats than Morse et al., 1984
148–322 mol/l hamsters
3-Methylindole, [7–90d] Degeneration, fibrous adhesions, osseous Turk et al., 1987;
100–400 mg/kg ip remodeling, Bowman’s gland hypertrophy, Peele et al.,1991
behavioral deficits
Methyl 2 hr Degeneration of the respiratory and olfactory Uraih et al., 1987
isocyanate, epithelium
10, 30 ppm
Napthalene, [24 hr] Cytotoxicity (mice and hamsters), Plopper et al., 1992
400–1600 mg/kg ip necrosis (rats)
Nickel subsulfide, 6 hr/d 5 d/wk Atrophy Dunnick et al., 1989
0.11–1.8 mg/m3 13 wk
Nickel sulfate, 6 hr/d 12–16 Atrophy, degeneration, decrease in Evans et al., 1995
3.5–635 mg/m3 carnosine consecutive
N-Nitroso- [6 hr-30 d] Degeneration of olfactory epithelium Rangga-Tabbu and
dimethylamine, and Bowman’s glands Sleight, 1992
20–80 mg/kg ip
N-Nitroso- [6 hr-30 d] Degeneration of olfactory epithelium Rangga-Tabbu and
pyrrolidine, and Bowman’s glands Sleight, 1992
30–100 mg/kg ip
Propylene glycol 2 weeks Slight to moderate degeneration of Miller et al., 1984
monomethyl olfactory epithelium
ether acetate,
3000 ppm
Propylene oxide, 4 weeks Degeneration of the olfactory epithelium Eldridge et al. 1995
0–525 ppm
Pyridine, 6 hr- 4 d Degeneration of olfactory epithelium Nikula and Lewis,
5–444 ppm 1994
RP 73401, 1 hr-5 d Degeneration of olfactory epithelium Pino et al., 1999
1 mg/kg/day and Bowman’s glands
Sulfur dioxide, 72 hr, or Necrosis, edema, destruction, hyperplasia, Giddens and
10–117 ppm 6 hr/d 5d hypertrophy Fairchild,1972;
Buckley et al., 1984
Sulfuryl fluoride, 6 hr/d 5d Inflammation Eisenbrandt and
0–600 ppm 2 wk Nitschke, 1989
Tetramethoxy- 6 hr/d 5 d Ulceration, inflammation, and necrosis Kolesar et al., 1989
silane 1–45 ppm 28 days of epithelium
2,4-Toluene 6 hr/d 5d Ulceration, necrosis, inflammation, Buckley et al., 1984
diisocyanate, degeneration
0.4 ppm
3-Trifluoromethyl 6 hr/d Degeneration, reduced Bowman’s Gaskell et al., 1988
pyridine, 10–90 d activity
0.1–329 ppm

aDuration in brackets are times postexposure until killed.

584 Hastings and Miller

to chromium (sodium dichromate) at six times the TLV assumption seem reasonable, since exposure of mice to
were observed (Hastings et al., 1994). SO2 has been shown to cause extensive damage in the
Other rodent studies have found numerous industrial olfactory epithelium (Giddens and Fairchild, 1972). The
compounds that damage the olfactory mucosa (Table 2). authors also pointed out that deficits were found in work-
In general, the rodent olfactory system is both robust and ers who had not been exposed in recent years, suggesting
resilient; while exposure to many toxic compounds can that the damage was more or less permanent in nature.
nearly obliterate the olfactory mucosa, it almost always In one of the few human laboratory studies on this
reconstitutes itself, at least to some degree. In the major- topic, volunteers were exposed to ozone (0.400 ppb,
ity of the studies conducted, however, the emphasis usu- 4 hr/day for 4 days), a component of both indoor and
ally has been on structural or biochemical/molecular outdoor pollution. This resulted in an initial increase in
changes; only rarely has a focus been on behavioral the olfactory threshold to the one compound they exam-
indices of olfactory function (Barrow, 1986; Miller, 1995) ined, butyl alcohol (Prah and Benignus, 1979). However,
(Table 2). More studies, in addition to the ones cited by the end of the 4 days of exposure, thresholds had
above, are sorely needed that assess in greater detail the returned to normal, suggesting that the system may have
relationship between structural damage and functional compensated for the initial insult. This situation is quite
effects. While the animal data are important in determin- different from most occupational chronic exposures,
ing which toxins have the potential for causing damage to where the exposure may span years. In the latter situa-
the olfactory system in humans, the resilience of the tion, there may be an early adaptive phase, which is fol-
rodent system leads to valid skepticism as to whether the lowed, after longer continued exposure, by loss of the
human olfactory epithelium reacts similarly. ability of the system to compensate, perhaps as the result
of cumulative damage to the epithelium. Thus, both
short-and long-term exposure data from various concen-
V. IRRITANT GASES trations of airborne chemicals are needed to accurately
establish their toxicity.
There is little doubt that accidental exposure to high Probably the most comprehensive investigation
concentrations of irritant gases can have an adverse effect addressing the effects of exposure to toxic compounds on
on human olfactory function. However, the toll taken on olfaction was a study by Schwartz et al. (1989). Olfactory
the olfactory system by chronic exposure to low levels of function of 731 chemical workers exposed to acrylic acid
airborne contaminants that are sufficient to at least mildly and a variety of acrylates and methacrylates, at levels
activate CN V afferents is not known. Such exposure rou- below the TLVs, was assessed using the UPSIT. While
tinely occurs in some workplaces and, in the case of such analysis of the cross-sectional data revealed no relation-
agents as ozone and formaldehyde, are common compo- ship between chemical exposure and olfactory deficits, a
nents of both indoor and outdoor air pollution. Such expo- nested case-control study designed to examine cumulative
sure greatly expands the potential population that may be effects of exposure uncovered several associations: first,
exposed to such compounds, including those that may be olfactory dysfunction increased with cumulative expo-
more sensitive to adverse effects, i.e., the very young, the sure; second, the effects appeared to be reversible; and
sick, the elderly, and those with genetic predispositions. third, the highest relative risk of olfactory dysfunction
Animal studies, primarily rodent ones, support the idea occurred in the group of workers who had never smoked.
that, in fact, chronic exposure to a number of irritants can The more pronounced effects found in the never-smoking
seriously damage the olfactory epithelium if present at group may be due to the fact that smoking induces meta-
high enough concentration (see Table 2). bolic enzymes in the nasal mucosa that may provide pro-
In a study investigating the effects of exposure to SO2 tection against the toxic effects of exposure to other
and/or ammonia (concentrations unknown) on olfactory chemicals (Reed, 1993). Recall that, in contrast, smoking
function, 25% of the workers surveyed reported an in conjunction with cadmium exposure appeared to exac-
increase in subjective complaints involving olfactory dis- erbate the olfactory deficit (Sulkowski et al., 2000). One
orders (Harada et al., 1983). When tested empirically using final point stressed by Schwartz et al. (1989) was that
a T & T olfactometer (Oka, 1981), odor-detection thresh- although the study suggested that exposure to acrylates
olds for five different odors were elevated, but only for the and methacrylates was associated with a significant
group exposed to SO2. As with cadmium (Rose et al., decrease in olfaction, the impairment may not always be
1992), the deficit was assumed to be peripheral in nature, of sufficient magnitude to be clinically important.
i.e., related to injury of the nasal mucous membrane. Another well-designed study investigated the effects of
Although workplace exposure data were lacking, this styrene exposure on olfactory function (Dalton et al., 2000).
Influence of Environmental Toxicants on Olfactory Function 585

Styrene exposure in animals has been shown to cause exten- ppm formaldehyde (alone or in combination with wood
sive damage to the olfactory epithelium (Ohashi et al., 1986), dust) showed a slight but significant elevation in detection
but whether this was also true in humans was not known. thresholds (serial dilutions of pyridine) when compared to
Dalton et al. examined olfactory function in a group of work- controls (14.2 vs. 15.6) (Holstrom and Wilhelmsson,
ers with a minimum of 4 years of exposure to styrene and a 1988). In a questionnaire study, 68% of histology techni-
historic mean dose of 154.1 ppm-years (range: 13.8–328 cians exposed to 0.200–1.900 ppm formaldehyde reported
ppm-years) with that of a group of age- and gender-matched decreased odor perception; only 9% of the control group
unexposed controls. Olfactory function was assessed using a reported the same (Kilburn et al., 1985).
detection threshold tests for phenylethyl alcohol and styrene,
as well as identification tests employing ortho- and
retronasally presented odorants. Individual exposure histo- VI. SOLVENTS
ries were reconstructed, and olfactory function was
examined with regards to the individual exposure profiles. Many solvents have been found to be neurotoxic, producing
Contrary to the findings from the animal literature in effects on cognitive function as well as on such sensory sys-
which comparable exposure concentrations had been used, tems as vision and audition (Dick, 1995). Due to their
there was no evidence of any long-term alteration in olfac- lipophilic nature, most solvents readily penetrate the under-
tory function due to styrene exposure. This discrepancy in lying cellular membranes after traversing the mucus.
findings could be due to a number of factors. For example, Workers exposed to petroleum products (fuel oil vapors)
rats and mice are obligate nose breathers, and thus the have been found to be less able to detect decreasing con-
olfactory epithelium may experience a much higher expo- centrations of n-butanol and fuel oil than controls, but were
sure than that which occurs in humans. Furthermore, the no different than controls in detecting thresholds for pyri-
ability to metabolize styrene in olfactory tissue may differ dine and dimethyl disulfide (Ahlstrom et al., 1986). The ele-
between the two species. Since odor-detection thresholds vated thresholds, however, were still within the “normal”
for styrene, but not for phenyl ethyl alcohol, were signifi- range. These workers also displayed “odor intensity recruit-
cantly but reversibly elevated among the exposed workers, ment,” i.e., normal perception of strong stimuli but impaired
exposure-induced adaptation may have been present. perception of weak stimuli. Not surprisingly, the largest
Another toxic compound for which there exists both decrement was found for detection of fuel oil vapor, which
animal and human data is methyl bromide. Methyl bro- supports the concept of “industrial anosmia”—the phenom-
mide has been widely used as a fumigant and is one of the enon whereby exposure to strong odors in the workplace
most potent olfactotoxins identified in the animal litera- results in a reduction in sensitivity for those specific odors,
ture. The olfactory function of 123 structural fumigation while sensitivity to other odors remain the same. Unlike the
workers exposed to either methyl bromide or sulfuryl flu- deficits associated with exposure to cadmium, the effects
oride (another fumigant) was tested by Calvert et al. due to solvents were transient and reversible.
(1998) using the UPSIT. While methyl bromide appeared Schwartz et al. (1990) investigated the effects of solvent
to have no effect on olfactory function, high sulfuryl fluo- exposure on olfactory function in 187 workers at a paint
ride exposure caused a significant decrease in the UPSIT formulation plant. Historical industrial hygiene sampling
score—33.1 compared to 34.4 for the controls—represent- data were available, which provided accurate exposure his-
ing a drop from normal (34 and above) to mild hyposmia tories. Olfactory function was measured using the UPSIT,
(33 and lower) (Doty, 1995). The lack of effect seen with and the data revealed a subtle, dose-related effect of sol-
the methyl bromide may reflect enhancement of the meta- vent exposure on olfactory function, but only in never-
bolic activity of the phase 1 enzymes from low exposure, smokers. The similarity of these results to those found for
thereby preventing damage from occurring when higher, the chemical workers (Schwartz et al., 1989), but not for
more toxic concentrations were encountered. Thus, only if the cadmium workers (Sulkowski et al., 2000), illustrates
the initial exposure was high will damage likely result. how ancillary factors can influence the toxicity of com-
Formaldehyde, a ubiquitous and highly reactive gas, is pounds on olfactory function. Once again, although the
employed widely in a variety of manufacturing processes results were statistically significant, the deficits were,
and is also frequently encountered in biology classes, his- overall, very subtle in nature. In spite of the many limita-
tology departments, and mortuaries. It is absorbed mainly tions inherent in field studies, these results also suggest
in the nose, particularly in the anterior portion. While that chronic low-level exposure to solvents adversely
exposure to formaldehyde is often alleged to decrease affects olfactory function.
olfactory acuity (Spearman, 1954), the supportive data are In some cases, painters may be exposed to higher
very sparse. In one study, workers exposed to 0.075–0.750 concentrations of solvents than chemical plant workers.
586 Hastings and Miller

This is due to the fact that airborne contaminants are usu- related symptoms can nonetheless occur. Probably the
ally closely monitored in commercial plants, as required by most widespread problem related to odors, in general, is
government regulations. On the other hand, painters work annoyance. Annoyance can range from the discomfort
in confined spaces and without monitoring equipment. experienced by an individual upon being exposed to a
Using the UPSIT and a threshold test based on serial dilu- strong perfume or malodor to the distress experienced by
tions of pyridine, Sandmark et al. (1989) compared the a whole community located downwind from a chemical
olfactory function of a group of 54 painters exposed to plant, paper mill, waste treatment plant, or similar facility.
organic solvents with that of 42 unexposed controls. No dif- Such exposures often involve some form of sulfur-contain-
ferences in olfactory function were found between the two ing compound, e.g., hydrogen sulfide, mercaptans,
groups. The authors attributed the lack of effect to a low- thiopenes, etc. (Shusterman, 1992). These compounds
to-moderate degree of exposure to solvents. In contrast, often produce such effects at concentrations greater than
floor-layer workers, who are exposed to high levels of olfactory detection threshold, which may or may not
toluene, reported significantly more smell disturbances exceed the TLV. However, for some individuals, the
than controls when responding by questionnaire (Hotz response is much greater than mere annoyance and can
et al., 1992). become highly aversive and debilitating.
Airborne contaminants found in the workplace almost The olfactory system may play a role in mediating or
invariably consist of a mixture of compounds, making it triggering a constellation of symptoms collectively known
difficult to link specific compounds with identified effects. as idiopathic environmental intolerances (Staudenmayer,
Testing volunteers in laboratory studies under tightly con- 1999). These symptoms include autonomic arousal (light-
trolled conditions circumvents many of these limitations, headedness, nausea, anxiety, tachycardia), upper and lower
but such studies, in turn, are compromised by restrictions respiratory tract irritation, neurological symptoms (short-
on duration and concentration of exposure, as well as term memory loss), and others. A more limited version of
safety and ethical constraints (many compounds are sus- this phenomenon with a well-defined set of criteria is
pected carcinogens). Mergler and Beauvais (1992) known generally as multiple chemical sensitivity (MCS)
exposed volunteers for 7-hour periods to either toluene, (see Chapter 25). MCS is characterized by an aberrant
xylene, or a mixture according to a Latin square design, response elicited primarily by an odorant stimulus, in con-
and evaluated their olfactory function. Detection thresh- trast to the situation where chemical exposure causes
olds for toluene or phenylethyl methylethyl carbinol were abnormal perception of odor stimuli. Exposure to a solvent
measured. A sixfold shift in the detection threshold for or pesticide is usually the precipitating event (a “sensitiz-
toluene was found immediately following exposure to ing” experience) that leads to MCS. In general, in
toluene, xylene, or a mixture of the two. No differences individuals experiencing MCS, exposure to many airborne
were found in phenylethyl methylethyl carbinol thresholds contaminants including solvents, perfumes, cleaning
after exposure to any of the compounds. agents, gasoline, pesticides, and paint, called “triggers”,
As previously noted, exposure to a compound may elicits the multimodal symptoms.
affect detection of the same compound but have no effect Numerous complications exist in understanding the eti-
on detection of other odorants. This would imply that ology of MCS, if indeed it is a homogeneous disease or
thresholds for several different compounds should be disorder. One is that within an individual, a similar set of
obtained when testing for olfactory deficits or, at the very symptoms is caused by a diverse range of chemically and
least, a test such as the UPSIT, which incorporates a large structurally dissimilar synthetic compounds. Conversely,
number of different odorants, should be employed. The exposure to the same chemical can elicit different symp-
fact that the alterations in olfactory function seen in most toms in different MCS patients. Most problematic is the
of these studies were reversible suggests the possibility fact that the exposure levels of the chemicals that “trigger”
that the shift in detection threshold was due to short-term the responses are usually far below the TLVs for the com-
effects such as olfactory fatigue or adaptation, rather than pounds and have no effect on the vast majority of people.
to toxic insult to the olfactory receptor neurons. In most circumstances, the level of exposure sufficient to
induce the symptoms needs to be only at the threshold of
odor perception.
VII. ODORS, ANNOYANCES, AND MULTIPLE Since the etiology of these symptoms does not appear to
CHEMICAL SENSITIVITY based on traditional toxocological mechanisms, a number
of nontoxicological odor-related mechanisms have been
In the absence of exposure levels that produce irritation proposed to explain the elicitation of such acute, odor-trig-
or other traditional toxicological endpoints, acute odor- gered symptoms (Shusterman, 1992; Siegel and Kreutzer,
Influence of Environmental Toxicants on Olfactory Function 587

1997). These include odor-related aversive conditioning, system reactivity resulting from time-dependent sensi-
and odor-related, stress-induced illness; both rely heavily tization. A recent review of these conflicting views
on learned associations mediated by classical conditioning (psychogenic vs. biological) concluded that sufficient
processes. Usually odor-related aversive conditioning information is not currently available to say unequivocally
involves a traumatic exposure episode, while the odor- whether either is correct and that neither should be dis-
related, stress-induced illness paradigm represents the con- missed in determining both etiology and treatment strate-
ditioning of an odor cue with autonomic arousal symptoms gies (Gots, 1996). The role played by perceived olfactory
resulting from stress. The stress is usually chronic in cues and the olfactory system in the etiology of MCS is
nature and may result from the workplace environment still largely unknown.
(job pressures, employment insecurity, etc.) or from
perceived dangers, such as working with “unknown chem-
VIII. CONCLUSIONS
icals” or living near a toxic waste site. Boxer (1990) stated,
“It has been observed clinically that psychologic reactions
Although many reports in the literature suggest that
to exposure to neurotoxins can be more serious than the
exposure to a wide variety of toxic compounds can have a
direct neurotoxic effects.” In any event, odors are very
deleterious effect upon olfactory function, there is little
salient sensory cues, and it is postulated in such thinking
specific information relating to dose effect, duration, mech-
that they quickly become associated with the subjective
anisms of action, etc., for these compounds. Only in recent
symptoms. After conditioning, the odor cue alone becomes
years have there been reliable industrial hygiene data of
sufficient to elicit the symptoms. Other proposed mecha-
value for linking exposure to specific compounds with
nisms include innate odor aversion and odor-related exac-
olfactory dysfunction. Similarly, practical tests for assessing
erbation of underlying conditions. The latter is suggested
olfactory function in the workplace have only recently been
by the finding that MCS patients display an increased
developed. Consequently, very few studies have demon-
prevalence of depression and psychosomatic disorders
strated, with any certainty, altered olfactory function as a
compared to persons from the normal population (Stewart
result of toxic exposure. Furthermore, most of these show
and Ruskin, 1985).
that under normal circumstances exposure to high levels of
Experimental data exist that suggest, contrary to patient
a toxic compound is required to produce observable deficits.
complaints, that odor sensitivity is not altered in most MCS
However, the number of compounds actually tested for
patients. In an odor-detection threshold experiment, Doty
potential olfactotoxicity is minuscule compared to the large
et al. (1988) found that odor thresholds were no lower
number of chemicals used in the workplace. Many more
among MCS symptomatics than among control subjects. In
epidemiological studies are needed which examine the
a similar study, Hummel et al. (1996) found no differences
relationship between workplace exposure to specific toxic
in odor-detection thresholds between MCS symptomatics
agents and olfactory toxicity. Animal studies, while
and normal controls. Paradoxically, they found that the
suggestive, cannot be definitive, given the major species dif-
MCS patients were impaired, relative to controls, in identi-
ferences in the deposition and fate of chemicals that enter
fying and discriminating among suprathreshold odors, yet
the nose.
exhibited smaller odor-induced event-related potentials. If
these counterintuitive findings are valid (no control group
was tested in the event-related potential phase of the study), REFERENCES
the suprathreshold olfactory function of MCS patients may,
in fact, be decreased, not increased. Ahlstrom, R., Berglund, B., Berglund, U., Lindvall, T., and
In contrast to the notion that MCS results from a pri- Wennberg, A. (1986). Impaired odor perception in tank clean-
mary emotional (psychogenic) response to perceived ers. Scand. J. Work Environ. Health 12:574–581.
chemical exposure is the view that the symptoms result American Conference of Governmental Industrial Hygienists.
from pathological interactions between chemicals and the (1994). Threshold Limit Values for Chemical Substances and
biological system. Ziem and McTamney (1997) have sug- Physical Agents and Biological Exposure Indices. ACGIH,
Cincinnati.
gested that toxic porphyria may be the basis for MCS.
Amoore, J. E. (1986). Effects of chemical exposure on olfaction
Meggs (1993; see also Bascom et al, 1997) has advanced in humans. In Toxicology of the Nasal Passages, Barrow, C.
the concept that neurogenic inflammation arising from S. (Ed.). Hemisphere Publishing, Washington, DC, pp.
stimulation of chemical irritant receptors may explain the 155–190.
occurrence of chemical sensitivities. Bell et al. (1992) Apfelbach, R. (1991). Sensitivity to odors in Wistar rats is
have proposed that increased sensitivity to chemical expo- reduced after low-level formaldehyde-gas exposure.
sure seen in MCS patients results from increased limbic Naturwissenschaften 78:221–223.
588 Hastings and Miller

Apfelbach, R., Reibenspies, M., Schmidt, R., Weiler, E., Binding, Steenland, K. (1998). Health effects associated with sulfuryl
N., and Camman, K. (1992). Behavioral and structural modi- fluoride and methyl bromide exposure among structural fumi-
fications in the olfactory epithelium after low level formal- gation workers. Am. J. Public Health 88:1774–1780.
dehyde-gas exposure. In Proceedings of the 3rd International Chuang, H. Y., Schwartz, J., Tsai, S.Y., Lee, M. L., Wang, J. D.,
Conference on Role of Formaldehyde in Biological Systems. and Hu, H. (2000). Vibration perception thresholds in workers
Methylation and Demethylation Processes. Hungarian with long term exposure to lead. Occup. Environ. Med.
Biochemical Society, Sopron, Hungary. pp. 57–62. 57:588–594.
Appelman, L. M., Woutersen, R. A., and Feron, V. J. (1982). Cometto-Muniz, J. E., and Cain, W. S. (1991). Influence of air-
Inhalation toxicity of acetaldehyde in rats. I. Acute and sub- borne contaminants on olfaction and the common chemical
acute studies. Toxicology 23:293–307. sense. In Smell and Taste in Health and Disease, Getchell,
Barrow, C. S. (1986). Toxicology of the Nasal Passages. T. V., Doty, R. L., Bartoshuk, L. M. and Snow, J. B. (Eds.).
Hemisphere Publishing, Washington, DC. Raven Press, New York.
Bascom, R., Meggs, W. J., Frampton, M., Hudnell, K., Killburn, Dalton, P. (2000). Olfactory function and adaptation following
K., Kobal, G., Medinsky, M., and Rea, W. (1997). Neurogenic long-term occupational exposure to styrene. Chem. Senses
inflammation: with additional discussion of central and per- 25:658.
ceptual integration of nonneurogenic inflammation. Environ. Dick, R. B. (1995). Neurobehavioral assessment of occupation-
Health Perspect. 105 (suppl 2): 531–537. ally relevant solvents and chemicals in humans. In Handbook
Bell, I. R., Miller, C. S. and Schwartz, G. E. (1992). An olfactory- of Neurotoxicology, Chang, L.W., and Dyer, R.S. (Eds.).
limbic model of multiple chemical sensitivity syndrome: pos- Marcel Dekker, Inc., New York, pp. 171–212.
sible relationships to kindling and affective spectrum Doty, R. L. (1995). The Smell Identification TestTM Administration
disorders. Biol. Psychiatry 32:218–242. Manual. Sensonics Haddon Heights, NJ.
Bogdanffy, M. S., and Frame, S. R. (1994). Olfactory mucosal Doty, R. L. (1998). Cranial nerve I: olfaction. In Textbook of
toxicity. Integration of morphological and biochemical data in Clinical Neurology, Goetz, C. G., and Pappert, E. J. (Eds.).
mechanistic studies: dibasic ester as an example. Inhal. Philadelphia, Saunders, pp. 90–101.
Toxicol. 6 (suppl):205–219.
Doty, R. L., Deems, D. A., Frye, R. E., Pelberg, R, and Shapiro,
Bolla, K. I., Schwartz, B. S., Stewart, W., Rignani, J., Agnew,
A. (1988). Olfactory sensitivity, nasal resistance, and auto-
J., and Ford, D. P. (1995). Comparison of neurobehavioral
nomic function in patients with multiple chemical sensitiv-
function in workers exposed to a mixture of organic and inor-
ities. Arch. Otolaryngol. Head Neck Surg. 114:1422–1427.
ganic lead and in workers exposed to solvents. Am. J. Ind.
Doty, R. L. and Hastings, L. (2001). Neurotoxic exposure
Med. 27:231–246.
and olfactory impairment. Clinic Occupat. Environ. Med.
Boxer, P. A. (1990). Indoor air quality: a psychosocial perspec-
Neurotoxicology 1:547–575.
tive. J. Occup. Med. 32:425–428.
Duncan, R. B., and Briggs, M. (1962). Treatment of uncomplicated
Brandt, I., Brittebo, E. B., Feil, V. J., and Bakke, J. E. (1990).
anosmia by vitamin A. Arch. Otorhinolaryngol. 75:116–124.
Irreversible binding and toxicity of the herbicide dichlobenil
(2,6-dichlorobenzonitrile) in the olfactory mucosa of mice. Dunnick, J. K., Elwell, M. R., Benson, J. M., Hobbs, C. H., Hahn,
Toxicol. Appl. Pharmacol. 103:491–501. F.F., Haly, P. J., Cheng, Y. S., and Eidson, A. F. (1989). Lung
Brenneman, K. A., James, R. A., Gross, E. A., and Dorman, D. C. toxicity after 13-week inhalation exposure to nickel oxide,
(2000). Olfactory neuron loss in adult male CD rats following nickel subsulfide, or nickel sulfate hexahydrate in F344/N rats
subchronic inhalation exposure to hydrogen sulfide. Toxicol. and B6C3F1 mice. Fundam. Appl. Toxicol. 12:584–594.
Pathol. 28:326–333. Eisenbrandt, D. L., and Nitschke, K. D. (1989). Inhalation toxic-
Brittebo, E.B., Eriksson, C., and Brandt, I. (1990). Activation and ity of sulfuryl fluoride in rats and rabbits. Fundam. Appl.
toxicity of bromobenzene in nasal tissues in mice. Arch. Toxicol. 12:540–557.
Toxicol. 64:54–60. Eldridge, S. R., Bogdanffy, M. S., Jokinen, M. P., and Andrews,
Brittebo, E. B., Eriksson, C., Feil, V., Bakke, J., and Brandt, I. L.S. (1995). Effects of propylene oxide on nasal epithelial cell
(1991). Toxicity of 2,6-dichlorothiobenzamide (chlorthiamid) proliferation in F344 rats. Fundam. Appl. Toxicol. 27:25–32.
and 2,6-dichlorobenzamide in the olfactory nasal mucosa of Elsberg, C., and Levy, I. (1935). The sense of smell: I. A new and
mice. Fundam. Appl. Toxicol. 17:92–102. simple method of quantitative olfactometry. Bull. Neurol. Inst.
Buckley, L. A., Jiang, X. Z., James, R. A., Morgan, K. T., and NY 4:4–19.
Barrow, C.S. (1984). Respiratory tract lesions induced by sen- Emmett, E. A. (1976). Parosmia and hyposmia induced by sol-
sory irritants at the RD50 concentration. Toxicol. Appl. vent exposure. Br. J. Indust. Med. 33:196–198.
Pharmacol. 74:417–429. Evans, J. E., Miller, M. L., Andringa, A., and Hastings, L. (1995).
Buckley, L. A., Morgan, K. T., Swenberg, J. A., James, R. A., Behavioral, histological, and neurochemical effects of nickel
Hamm, T.E.J., and Barrow, C.S. (1985). The toxicity of (II) on the rat olfactory system. Toxicol. Appl. Pharmacol.
dimethylamine in F-344 rats and B6C3F1 mice following a 1- 130:209–220.
year inhalation exposure. Fundam. Appl. Toxicol. 5: 341–352. Feron, V. J., and Kruysse, A. (1978). Effects of exposure to furfural
Calvert, G. M., Mueller, C. A., Fajen, J. M., Chrislip, D. W., vapour in hamsters simultaneously treated with benzo[alpha]
Russo, J., Briggle, T., Fleming, L.E., Suruda, A.J., and pyrene or diethylnitrosamine. Toxicology 11:127–144.
Influence of Environmental Toxicants on Olfactory Function 589

Foureman, G. L., Greenberg, M. M., Sangha, G. K., Stuart, B. P., Hummel, T., Roscher, S., Jaumann, M. P., and Kobal, G. (1996).
Shiotsuka, R.N., and Thyssen, J.H. (1994). Evaluation of Intranasal chemoreception in patients with multiple chemical
nasal tract lesions in derivation of the inhalation reference sensitivities: a double-blind investigation. Reg. Toxicol.
concentration for hexamethylene diisocyanate. Inhal. Toxicol. Pharmacol. 24:S79–S86
6 (suppl):341–355. Keenan, C. M., Kelly, D. P., and Bogdanffy, M. S. (1990).
Friberg, L. (1950). Health hazards in the manufacture of akaline Degeneration and recovery of rat olfactory epithelium follow-
accumulators with, special reference to chronic cadmium poi- ing inhalation of dibasic esters. Fundam. Appl. Toxicol.
soning. Acta Med. Scand. 138:1–124. 15:381–393.
Furuta, S., Nishimoto, K., Egawa, M., Ohyama, M., and Kilburn, K. H. (1998). Neurobehavioral impairment and symp-
Moriyama, H. (1994). Olfactory dysfunction in patients with toms associated with aluminum remelting [see comments].
Minamata disease. Am. J. Rhinol. 8:259–263. Arch. Environ. Health 53:329–335.
Gamble, J. F. (2000). Low-level hydrocarbon solvent exposure and Kilburn, K. H., Seidman, B. C., and Warshaw, R. (1985).
neurobehavioural effects. Occup. Med. (Oxford) 50:81–102. Neurobehavioral and respiratory symptoms of formaldehyde
Gaskell, B. A., Hext, P. M., Pigott, G. H., Hodge, M. C., and and xylene exposure in histology technicians. Arch. Environ.
Tinston, D.J. (1988). Olfactory and hepatic changes following Health 40:229–233.
inhalation of 3-trifluoromethyl pyridine in rats. Toxicology Kimbell, J. S., Gross, E. A., Joyner, D. R., Godo, M. N., and
50:57–68. Morgan, K. T. (1993). Application of computational fluid
Genter, M. B., Llorens, J., O’Callaghan, J. P., Peele, D. B., dynamics to regional dosimetry of inhaled chemicals in the
Morgan, K. T., and Crofton, K. M. (1992). Olfactory toxicity upper respiratory tract of the rat. Toxicol. Appl. Pharmacol.
of beta, beta’-iminodipropionitrile in the rat. J. Pharmacol. 121:253–263.
Exp. Ther. 263:1432–1439. Kimbell, J. S., Godo, M. N., Gross, E. A., Joyner, D. R.,
Giddens, W. E., and Fairchild, G. A. (1972). Effects of sulfur Richardson, R. B., and Morgan, K. T. (1997). Computer sim-
dioxide on the nasal mucosa of mice. Arch. Environ. Health ulation of inspiratory airflow in all regions of the F344 rat
25:166–173. nasal passages. Toxicol. Appl. Pharmacol. 145:388–398.
Gots, R. E. (1996). Multiple chemical sensitivities: distinguish- Kolesar, G. B., Siddiqui, W. H., Geil, R. G., Malczewski, R. M.,
ing between psychogenic and toxicodynamic. Reg. Toxicol. and Hobbs, E. J. (1989). Subchronic inhalation toxicity of
Pharmacol. 24:S8–S15. tetramethoxysilane in rats. Fundam. Appl. Toxicol. 13:
Gu, J., Walker, V. E., Lipinskas, T. W., Walker, D. M., and Ding, X. 285–295.
(1997). Intraperitoneal administration of coumarin causes tis- Lomax, L. G., Stott, W. T., Johnson, K. A., Calhoun, L. L., Yano,
sue-selective depletion of cytochromes P450 and cytotoxicity B.L., and Quast, J.F. (1989). The chronic toxicity and onco-
in the olfactory mucosa. Toxicol. Appl. Pharmacol. 146: genicity of inhaled technical-grade 1,3-dichloropropene in
134–143.
rats and mice. Fundam. Appl. Toxicol. 12:418–431.
Harada, N., Fujii, M., and Dodo, H. (1983). Olfactory disorders
Lucchini, R., Bergamaschi, E., Smargiassi, A., Festa, D., and
in chemical plant workers exposed to SO2 and/or NH3. J. Sci.
Apostoli, P. (1997). Motor function, olfactory threshold, and
Labour 59:17–23.
hematological indices in manganese-exposed ferroalloy
Hastings, L., Miller, M. L., Minnema, D., Evans, J., and Radike,
workers [published erratum appears in Environ Res 1997
M. (1991). Effects of methyl bromide on the rat olfactory sys-
Nov;75(2):187]. Environ. Res. 73:175–180.
tem. Chem. Senses 16:43–55.
Meggs, W. J. (1993). Neurogenic inflammation and sensitivity to
Hastings, L., Andringa, A., and Miller, M. L. (1994). Exposure of
environmental chemicals. Environ. Health Perspect.
the olfactory system to toxic compounds: Structural and func-
101:234–238.
tional consequences. Inhal. Toxicol. 6(suppl):437–440.
Mellert, T. K., Getchell, M. L., Sparks, L., and Getchell, T.V.
He, L., Poblenz, A. T., Medrano, C. J., and Fox, D. A. (2000).
Lead and calcium produce rod photoreceptor cell apoptosis by (1992). Characterization of the immune barrier in human
opening the mitochondrial permeability transition pore. olfactory mucosa. Otolaryngol. Head Neck Surg.
J. Biol. Chem. 275:12175–12184. 106:181–188.
Henkin, R. I., Schecter, P. J., Friedewald, W. T., Demets, D. L., Mergler, D., and Beauvais, B. (1992). Olfactory threshold shift
and Raff, M. (1976). A double blind study of the effects of following controlled 7-hour exposure to toluene and/or
zinc sulfate on taste and smell dysfunction. Am. J. Med. Sci. xylene. Neurotoxicology 13:211–215.
272:285–299. Mergler, D., Huel, G., Bowler, R., Iregren, A., Belanger, S.,
Holstrom, M., and Wilhelmsson, B. (1988). Respiratory symp- Baldwin, M., Tardif, R., Smargiassi, A., and Martin, L.
toms and pathophysiological effects of occupational exposure (1994). Nervous system dysfunction among workers with
to formaldehyde and wood dust. Scand. J. Work Environ. long-term exposure to manganese. Environ. Res. 64:151–180.
Health 14:306–311. Mery, S., Larson, J. L., Butterworth, B. E., Wolf, D.C., Harden,
Hotz, P., Tschopp, A., Soderstrom, D., Holtz, J., Boillat, M. A., R., and Morgan, K. T. (1994). Nasal toxicity of chloroform in
and Gutzwiller, F. (1992). Smell or taste disturbances, neuro- male F-344 rats and female B6C3F1 mice following a
logical symptoms, and hydrocarbon exposure. Int. Arch. 1-week inhalation exposure. Toxicol. Appl. Pharmacol. 125:
Occup. Environ. Health 63:525–530. 214–227.
590 Hastings and Miller

Miller, F. J. (1995). Nasal Toxicity and Dosimetry of Inhaled Rangga-Tabbu, C., and Sleight, S. D. (1992). Sequential study in
Xenobiotics. Taylor & Francis, Washington, DC. rats of nasal and hepatic lesions induced by N-nitrosodi-
Miller, R. R., Ayres, J. A., Jersey, G. C., and McKenna, M. J. methylamine and N-nitrosopyrrolidine. Fundam. Appl.
(1981). Inhalation toxicity of acrylic acid. Fundam. Appl. Toxicol. 19:147–156.
Toxicol. 1:271–277. Reed, C. J. (1993). Drug metabolism in the nasal cavity: rele-
Miller, R. R., Hermann, E. A., Young, J. T., Calhoun, L. L., and vance to toxicology. Drug Metab. Rev. 25:173–205.
Kastl, P.E. (1984). Propylene glycol monomethyl ether Reznik, G., Stinson, S. F., and Ward, J. M. (1980). Respiratory
acetate (PGMEA) metabolism, disposition, and short-term pathology in rats and mice after inhalation of 1,2-dibromo-3-
vapor inhalation toxicity studies. Toxicol. Appl. Pharmacol. chloropropane or 1,2 dibromoethane for 13 weeks. Arch.
75:521–530. Toxicol. 46:233–240.
Miller, R. A., Mellick, P. W., Leach, C. L., Chou, B. J., Irwin, R. D., Rose, C. S., Heywood, P. G., and Costanzo, R. M. (1992).
and Roycrott, J. H. (1991). Nasal toxicity in B6C3F1 mice inhal- Olfactory impairment after chronic occupational cadmium
ing furfuryl alcohol for 2 or 13 weeks. Toxicologist 11:669A. exposure. JOM 34:600–605.
Morse, C.C., Boyd, M.R., and Witschi, H. (1984). The effect of Roydhouse, N. (1988). Retinoid therapy and anosmia [letter]. NZ
3-methylfuran inhalation exposure on the rat nasal cavity. Med. J. 101:465.
Toxicology 30:195–204. Sandmark, B., Broms, I., Lofgren, L., and Ohlson, C. G. (1989).
Naus, A. (1975). Olphactoric properties of industrial matters, Olfactory function in painters exposed to organic solvents.
Charles University, Prague. Scand. J. Work Environ. Health 15:60–63.
Nikula, K. J., and Lewis, J. L. (1994). Olfactory mucosal lesions in Schieferstein, G. J., Sheldon, W. G., Cantrell, S. A., and Reddy,
F344 rats following inhalation exposure to pyridine at threshold G. (1988). Subchronic toxicity study of 1,4-dithiane in the rat.
limit value concentrations. Fundam. Appl. Toxicol. 23:510–517. Fundam. Appl. Toxicol. 11:703–714.
Nikula, K. J., Sun, J. D., Barr, E. B., Bechtold, W. E., Haley, P. J., Schwartz, B. S. (1991). Epidemiology and its application to olfac-
Benson, J. M., Eidson, A. F., Burt, D. G., Dahl, A. R., and tory dysfunction. In The Human Sense of Smell, Laing, D.G.,
Henderson, R. F. (1993). Thirteen-week, repeated inhalation Doty, R.L. and Breipohl, W., (Eds.). Springer-Verlag, Berlin.
exposure of F344/N rats and B6C3F1 mice to ferrocene. Schwartz, B. S., Doty, R. L., Monroe, C., Frye, R., and Barker, S.
Fundam. Appl. Toxicol. 21:127–139. (1989). Olfactory function in chemical workers exposed to
Ohashi, Y., Nakai, Y., lkeoka, H., Koshimo, H., Nakata, J., Esaki, acrylate and methacrylate vapors. Am. J. Public Health 79:
Y., Horiguchi, S., and Teramoto, K. (1986). Degeneration and 613–618.
regeneration of respiratory mucosa of rats after exposure to Schwartz, B. S., Ford, D. P., Bolla, K. I., Agnew, J., Rothman, N.,
styrene. J. Appl. Toxicol. 6:405–412. and Bleecker, M. L. (1990). Solvent-associated decrements in
Oka, T. (1981). [Studies on the olfactory fatigue by T & T olfac- olfactory function in paint manufacturing workers. Am. J. Ind.
tometer (author’s transl)]. [Japanese]. Nippon Jibiinkoka Med. 18:697–706.
Gakkai Kaiho 84:850–857. Schwartz, B. S., Bolla, K. I., Stewart, W., Ford, D. P., Agnew, J.,
Peele, D. B., Allison, S. D., Bolon, B., Prah, J. D., Jensen, K. F., and Frumkin, H. (1993). Decrements in neurobehavioral per-
and Morgan, K. T. (1991). Functional deficits produced by 3- formance associated with mixed exposure to organic and
methylindole-induced olfactory mucosal damage revealed by inorganic lead. Am. J. Epidemiol. 137:1006–1021.
a simple olfactory learning task. Toxicol. Appl. Pharmacol. Shusterman, D. (1992). Critical review: the health significance of
107:191–202. environmental odor pollution. Arch. Environ. Health
Pino, M. V., Valerio, M. G., Miller, G .K., Larson, J. L., Rosolia, 47:76–87.
D. L., Jayyosi, Z., Crouch, C. N., Trojanowski, J. Q., and Shusterman, D. J., and Sheedy, J. E. (1992). Occupational and
Geiger, L.E. (1999). Toxicologic and carcinogenic effects of environmental disorders of the special senses. Occup. Med.
the type IV phosphodiesterase inhibitor RP 73401 on the (London) 7:515–542.
nasal olfactory tissue in rats. Toxicol. Pathol. 27:383–394. Siegel, S., and Kreutzer, R. (1997). Pavlovian conditioning and
Plopper, C. G., Suverkropp, C., Morin, D., Nishio, S., and Buckpitt, multiple chemical sensitivity. Environ. Health Perspect. 105
A. (1992). Relationship of cytochrome P-450 activity to Clara (Suppl. 2):521–526.
cell cytotoxicity. I. Histopathologic comparison of the respira- Spearman, C. R. (1954). Odors, odorants, and deodorants in avi-
tory tract of mice, rats and hamsters after parenteral administra- ation. Ann. NY Acad. Sci. 58:40–43.
tion of naphthalene. J. Pharmacol. Exp. Ther. 261:353–363. Staudenmayer, H. (1999). Environmental Illness: Myth and
Potts, C. L. (1965). Cadmium proteinuria—the health of battery Reality. Lewis Publishers, Boca Raton, FL.
workers exposed to cadmium oxide dust. Ann. Occup. Hyg. Stewart, D., and Ruskin, J. (1985). Psychiatric assessment of
8:55–61. patients with “20th-century disease” (“total allergy syn-
Prah, J. D., and Benignus, V. A. (1979). Decrements in olfactory drome”). Can. Med. Assoc. J. 133:1001–1006.
sensitivity due to ozone exposure. Percept. Motor Skills Stott, W. T., Young, J. T., Calhoun, L. L., and Battjes, J. E.
48:317–318. (1988). Subchronic toxicity of inhaled technical grade 1,3-
Ramazzini, B. (1713). Diseases of Workers (De Morbis dichloropropene in rats and mice. Fundam. Appl. Toxicol.
Artificum). Hafner Publishing Company, New York. 11:207–220.
Influence of Environmental Toxicants on Olfactory Function 591

Sulkowski, J. W., Rydzewski, B., and Miarzynska, M. (2000). of the olfactory sense of chromate producing workers. Jpn.
Smell impairment in workers occupationally exposed to cad- J. Ind. Health 23:606–611.
mium. Acta Otolaryngol. 120:316–318. Wolf, D. C., Morgan, K. T., Gross, E. A., Barrow, C., Moss, O. R.,
Sun, T. J., Miller, M. L., and Hastings, L. (1996). Effects of James, R. A., and Popp, J. A. (1995). Two-year inhalation expo-
inhalation of cadmium on the rat olfactory system: behavior sure of female and male B6C3F1 mice and F344 rats to chlo-
and morphology. Neurotox. Teratol. 18:89–98. rine gas induces lesions confined to the nose. Fundam. Appl.
Suruda, A. J. (2000). Measuring olfactory dysfunction from cad- Toxicol. 24:111–131.
mium in an occupational and environmental medicine office World Health Organization (1986). Principles and Methods for the
practice. J. Occup. Environ. Med. 42:337–337. Assessment of Neurotoxicity Associated with Exposure to
Swift, D. L., and Procter, D. F. (1977). Access of air to the respi- Chemicals. World Health Organization, Geneva.
ratory tract. In Respiratory Defense Mechanisms. Marcel Woutersen, R. A., Appelman, L. M., Feron, V. J., and Van der
Dekker New York. Heijden, C. A. (1984). Inhalation toxicity of acetaldehyde in
Tsuchiya, K. (1967). Proteinuria of workers exposed to cadmium rats. II. Carcinogenicity study: interim results after 15 months.
fume. The relationship to concentration in the working envi- Toxicology 31:123–133.
ronment. Arch. Environ. Health 14:876–880. Woutersen, R. A., Appelman, L. M., Van Garderen-Hoetmer, A.,
Turk, M. A., Henk, W. G., and Flory, W. (1987). 3-Methylindole- and Feron, V.J. (1986) Inhalation toxicity of acetaldehyde in
induced nasal mucosal damage in mice. Vet. Pathol. rats. III. Carcinogenicity study. Toxicology 41: 213–231.
24:400–403. Wu, T. N., Shen, C. Y., Lai, J.S., Goo, C. F., Ko, K. N., Chi, H. Y.,
Uraih, L. C., Talley, F. A., Mitsumori, K., Gupta, B. N., Bucher, Chang, P.Y., and Liou, S.H. (2000). Effects of lead and noise
J.R., and Boorman, G. A. (1987). Ultrastructural changes in exposures on hearing ability. Arch. Environ. Health 55:
the nasal mucosa of Fischer 344 rats and B6C3F1 mice fol- 109–114.
lowing an acute exposure to methyl isocyanate. Environ. Yin-Zeng, L., Jin-Xiang, H., Cheng-Mo, L., Bo-Hong, X., and Cui-
Health Perspect. 72:77–88. Juan, Z. (1985). Effects of cadmium on cadmium smelter work-
Warheit, D. B., Kelly, D. P., Carakostas, M. C., and Singer, A. W. ers. Scand. J. Work Environ. Health 11:29–32.
(1989). A 90-day inhalation toxicity study with benomyl in Ziem, G., and McTamney. (1997). Profile of patients with chemi-
rats. Fundam. Appl. Toxicol. 12:333–345. cal injury and sensitivity. Environ. Health Perspect. 105 (Suppl.
Watanabe, S., and Fukuchi, Y. (2000). Occupational impairment 2):417–436.
28

Evaluation of Olfactory Deficits by Structural Medical Imaging

Cheng Li, Richard L. Doty, and David W. Kennedy

University of Pennsylvania, Philadelphia, Pennsylvania, U.S.A.

David M. Yousem
The Johns Hopkins University School of Medicine, Baltimore, Maryland, U.S.A.

I. INTRODUCTION 1996, 1997, 1998, 1999). In this chapter we comprehen-

sively review the pertinent medical literature on this gen-
Olfactory dysfunction can generally be classified into (1) eral topic and detail our own experience.
conductive disorders caused by interference with the
access of odorants to the olfactory receptors, (2) peripheral
sensorineural disorders resulting from injury to the olfac- II. IMAGING MODALITIES AND TECHNIQUES
tory receptors (within the olfactory mucosa), and (3) cen-
tral neural disorders of the olfactory bulb or tract or related Major advances in pinpointing the anatomical and patholog-
parts of the central nervous system such as the prefrontal ical changes of many disorders of the sinonasal cavity and
lobe, septal nuclei, amygdala, and temporal lobe. For brain have become possible as a result of the development
medical imaging and the anatomical approach, we catego- and refinement of imaging techniques (Carter and Runge,
rize olfactory dysfunction into two major groups: periph- 1988; Healy, 1992; Jagust and Eberling, 1991; Jolles et al.,
eral causes—sinonasal tract disorders—and central 1989; Reiman and Mintun, 1990; Shapiro and Som, 1989;
causes— intracranial disorders. It is important to relate Vogl, 1990; Yousem et al. 1996a, 1997b, 1998). Even
olfactory deficits to the appropriate anatomical and patho- though the imaging evaluation is not the diagnostic equiva-
logical changes. Unfortunately, clinical olfactory testing, lent to histological study, anatomical imaging, such as high-
whether psychophysical or electrophysiological, is rarely resolution computed tomography (CT) and magnetic
capable of localizing the source (aside from determining resonance imaging (MRI), can not only map regional
whether it is on the right or left) or identifying the specific lesions, but may also suggest a differential diagnosis (Carter
cause of decreased smell function. and Runge, 1988; Shapiro and Son, 1989; Som and Shapiro,
Modern medical imaging techniques offer a valuable 1988). On the other hand, functional imaging (PET, SPECT,
means for assessing the basis of some disorders of olfac- fMRI), which is reviewed in Chapter 12, affords one the
tion. Although revolutionary changes in medical imaging potential to explore regional pathophysiological changes in
techniques have occurred in the last few decades, only a the living brain (Healy, 1992; Jagust and Eberling, 1991;
few articles have dealt with imaging studies related to Jolles et al., 1989; Reiman and Mintun, 1990; Yousem et al.,
chemosensory disorders (Doty et al., 1999; Goodspeed 1997b, 1999b, c). The relevant imaging modalities which
et al., 1987; Kimmelman, 1991; Klingmuller et al., 1987; may be helpful in the evaluation of common causes of olfac-
Li et al., 1994; Schellinger et al., 1983; Yousem et al., tory deficits are reviewed in this section.

593
594 Li et al.

A. Plain Radiographs C. Magnetic Resonance Imaging

Plain film radiography, i.e., the “sinus series,” including MRI’s multiplanar capability is especially advantageous in
the Caldwell view, the Waters view, the lateral view, and the evaluation of sinonasal tract neoplasms and brain dis-
the base view, has long been a standard method of diag- orders. MRI, however, is less sensitive for the detection of
nosing nasal and paranasal sinus inflammatory disease. bony cortical abnormalities and landmarks. Soft tissue dis-
Problems of overlap and nonspecific findings are impossi- crimination, on the other hand, is more clearly illustrated
ble to avoid with plain films, and thus the study has been by MRI than by CT. Most soft tissue disease processes can
largely replaced by CT. The most important deficit of the be accurately localized with a minor degree of tissue dif-
plain film is its inability to provide the road map of the ferentiation, i.e., infection vs. tumor vs. hemorrhage. The
ostiomeatal complex, which may guide endoscopic surgi- anatomical discrimination of the brain is much better using
cal intervention (Zinreich et al., 1987). Plain radiographs MRI than CT. One can use thin sections, large matrices,
and conventional plain film tomography have virtually no and smaller fields of view to improve resolution, yet main-
role in the imaging evaluation of olfactory dysfunction. tain, high contrast to noise using T1-weighted scans (T1W)
or fast spin echo T2-weighted (T2W) images. T2-weighted
B. Computed Tomography scans can better delineate the contrast between normal and
inflammatory or neoplastic tissue (Shapiro and Som,
CT is well suited to the investigation of the sinonasal cav- 1989). New phase sensitive inversion recovery pulse
ities. Because CT scanning is as sensitive to soft tissue dis- sequences or standard spoiled gradient echo sequences can
ease as to bony changes, each scan can be photographed at highlight the gray-white matter differentiation and allow
an appropriate window width and level to optimally see better assessment of the hippocampus, parahippocampus,
insidious soft tissue differences in attenuation and fine gyrus rectus, and entorhinal cortex regions. Segmentation
bony detail. To study soft tissue, the window widths range of images to separate cortical volume from whole brain
from 150 to 400 Hounsfield units. Conversely, the bony volume is customary for volumetric studies nowadays.
detail is best observed at wide window settings—from For the evaluation of skull base invasion by sinonasal
2000 to 4000 Hounsfield units. The basic CT scanning pro- tumors, MRI is superior to CT (Paling et al., 1987).
tocol should include all of the nasal cavity, paranasal Gadolinium enhanced scans are particularly useful at the
sinuses, hard plate, anterior skull base, orbits, and skull base to detect dural or leptomeningeal involvement.
nasopharynx. The brain should be included if central Gadolinium-DTPA, a paramagnetic contrast agent, has
causes of olfactory dysfunction are suspected. The scans been widely utilized for distinguishing solidly enhancing
are commonly performed in both the axial and coronal tumor from rim-enhancing inflammatory processes
planes for optimal assessment of the complex paranasal (Brasch, 1992; Vogl et al., 1990).
anatomy, but coronal scans are the most valuable for the With regard to the olfactory system, CT and MRI play
anterior naso-ethmoid (ostiomeatal) region. Alternatively, complementary roles in evaluating sinonasal tract neo-
thin sections in one plane with multiplanar reconstructions plasms (Shapiro and Som, 1989; Som et al., 1990).
may be adequate. For practical purposes, slice thicknesses However, MRI is the study of choice to directly visualize
of 3–5 mm are often employed. For the evaluation of the the olfactory bulbs, olfactory tracts, and intracranial causes
ostiomeatal complex (the maxillary sinus ostium, of olfactory dysfunction (Klingmuller et al., 1987; Suzuki
infundibulum, uncinate process, and middle meatus), et al., 1989; Yousem et al., 1993, 1998, 1999a).
3-mm-thick coronal sections are fairly standard unless
three-dimensional (3D) reconstructions are requested. The D. Nuclear Medicine
quality of the 3D images is improved by utilizing 1-mm-
thick sectioning, which is rapidly performed with the new In general, conventional radionuclide imaging plays no
spiral scanners (minutes) and multidetector scanners (sec- significant role in the diagnostic work-up of patients with
onds). Intravenous contrast enhancement is usually suspected sinonasal tract disease (peripheral causes of
reserved for the identification of vascular lesions, tumors, olfactory deficits), except in the case of cerebrospinal fluid
meningeal or parameningeal processes, and abscess cav- (CSF) leaks. Functional imaging studies, such as positron
ities (Carter and Runge, 1988). Intrathecal contrast may be emission tomography (PET) and single photon emission
employed when cerebrospinal fluid leaks accompany the computed tomography (SPECT), are valuable in detecting
olfactory deficits. High-resolution CT is the most useful alterations of regional brain function and biochemistry in
and cost-effective screening tool for the evaluation of vivo (Alavi & Hirsch, 1991; Fowler et al., 1988; Jagust and
sinonasal tract inflammatory disorders. Eberling, 1991; Jolles et al., 1989; Reman and Mintun,
Medical Imaging of Olfactory Deficits 595

1990). Recent studies have suggested that functional imag- airflow to the olfactory receptors. Besides the obstructive
ing is more sensitive than anatomical imaging in detecting effect, lesions located in the upper nasal vault and/or crib-
abnormalities of the brain related to disorders such as riform plate region may also directly damage the olfactory
Alzheimer’s disease and Parkinson’s disease—conditions epithelium and olfactory neurons (Kern, 2000). The com-
associated with loss of olfactory function (Jagust and mon peripheral sinonasal tract causes of olfactory deficits
Eberling, 1991; Jolles et al., 1989). include infections, tumors, allergic rhinosinusitis, congen-
ital or developmental abnormalities, etc.
III. BASIC ANATOMY AND PHYSIOLOGY OF
A. Sinonasal Infectious Disease
THE OLFACTORY SYSTEM
Paranasal sinusitis is a relatively common disorder affecting
Since the anatomy and physiology of the olfactory system
approximately 30% of the population at some time in their
is discussed elsewhere in this volume (see Chapters 1–9),
lives (Allphin et al., 1991). One of the common symptoms
we only briefly mention this topic here. The sensation of
of acute and chronic paranasal sinusitis is decreased smell
smell is induced by the stimulation of olfactory receptor
sensation, which is generally reversible. The prompt diag-
cells by volatile chemicals. The olfactory receptor cells,
nosis and treatment of sinusitis are important for restoring
i.e., the primary olfactory neurons, are encompassed in the
olfactory function. Though the exact cause of chemosensory
neuroepithelium, which is located at the top of the nasal
dysfunction secondary to sinusitis is elusive, alterations in
vault, the upper portion of the nasal septum, the superior
nasal air flow and mucociliary clearance or obstruction from
surface of the superior nasal turbinate, sectors of the mid-
secretory products, polyps, or retention cysts may contribute
dle turbinate, and the region of the cribriform plate.
to olfactory dysfunction (Loury and Kennedy, 1991).
Afferent information from the receptors is transmitted by
In the diagnosis and evaluation of paranasal sinusitis,
the olfactory nerves, which course through the cribriform
medical imaging plays an important role. At present, high-
plate of the ethmoid bone to terminate in the glomeruli of
resolution CT is the preferred imaging technique, preceded
the olfactory bulb. In the olfactory bulb, the olfactory
by nasal endoscopic examination. Radiographic manifesta-
nerves make synaptic contact with dendrites of mitral and
tions of sinusitis have been well documented. In general,
tufted cells. From there, the efferent neurons of the olfac-
air-fluid levels are usually indicative of acute sinusitis,
tory bulb give rise to fibers forming the olfactory tracts,
whereas mucoperiosteal thickening or sinus opacification
which lie just under the gyrus rectus region in the olfactory
can be seen in acute and chronic disease. Polyps, mucous
sulcus of the frontal lobes. Axons from mitral and tufted
retention cysts, sinus expansion, and bony thickening of
cells project to central brain limbic system components
the walls of the sinus might also indicate chronicity of dis-
including the pyriform cortex and adjacent corticomedial
ease. CT is an excellent modality for the evaluation of
amygdala (which together form the uncus), the ventral
bony abnormalities, such as osteitis or remodeling, seen in
striatum, the parahippocampus area, and the anterior olfac-
some inflammatory lesions. CT also will identify the
tory nuclei. From these areas there are widespread inter-
infundibulum, the maxillary sinus ostium, the middle mea-
connections with many parts of the brain, including the
tus, the uncinate process, and the individual ethmoid air
mediodorsal thalamus, hypothalamus, orbitofrontal and
cells that make up the ostiomeatal complex. This will help
dorsolateral frontal cortex, temporal cortex, and other
the functional endoscopic sinus surgeon in his ability to
areas of the limbic system (see Chapters 8 and 9).
plan effective surgery to restore normal mucociliary clear-
ance. On the other hand, MRI is also highly sensitive for
IV. PERIPHERAL CAUSES OF OLFACTORY detecting mucosal thickening and other soft tissue abnor-
DISTURBANCE malities (Shapiro and Som, 1989). By and large, inflamed
mucosa is usually high in signal intensity on T2-weighted
Sinonasal tract disease is one of the common causes of MR images and low in intensity on T1-weighted scans. The
olfactory disturbance (Deems et al., 1991; Doty and signal intensity of the sinus secretions will vary with the
Mishra, 2001). The etiology of the olfactory deficits concentration of protein within the sinus (Barat, 1990;
among patients with nasal and paranasal sinus disease is Drutman et al. 1991; Shapiro and Som, 1989).
due, in many cases, to nasal airway obstruction. The influ-
ence of nasal obstruction on olfaction has been compre- B. Tumors of the Nasal Cavity and Paranasal Sinuses
hensively reviewed (e.g., Doty and Frye, 1989; Doty and
Mirshra, 2001) (see also Chapter 21). Any cause of bilat- Neoplasms of the sinonasal tract are uncommon. Malignant
eral obstruction can decrease smell sensations by limiting tumors of the nasal cavity and paranasal sinuses account for
596 Li et al.

only 0.2–0.8% of all human malignancies (Som, 1991). A recently described imaging finding characteristic
Early symptoms of sinonasal tract tumors, such as nasal of olfactory neuroblastomas is the presence of peripheral
discharge, unilateral nasal obstruction, and minor intermit- peritumoral cysts along the intracranial portion of the
tent epistaxis, may simulate low-grade chronic infection. tumor. If stippled calcifications are also seen on CT, the
Subsequent symptoms depend on the tumor’s location and diagnosis is assured.
pattern of growth. Neoplasms arising in the upper nasal
cavity and extending through the cribriform plate or into
2. Inverted Papillomas and Other Sinonasal Tumors
the ethmoid sinuses are often accompanied by frontal
headache, visual disturbances, and decreased smell sensa- The inverted papilloma is a relatively rare and locally
tion. Almost all sinonasal tract tumors and tumor-like con- aggressive sinonasal tumor. It constitutes 0.5–4% of pri-
ditions that grow to a large size may cause a decline in mary nasal tumors and occurs predominantly in males in
olfactory acuity by interfering with patency of the nasal air- the fifth and sixth decades of life (Phillips et al., 1990).
way or directly destroying the olfactory receptors. The most The most common presenting symptoms are nasal
common malignancies of the sinonasal system are squa- obstruction, epistaxis, and hyposmia. Subsequent sinusitis
mous cell carcinoma and adenocarcinoma, but lymphoma, and tumor extension into the sinuses and orbits can cause
melanoma, adenoid cystic carcinoma, and chondrosarco- purulent nasal discharge, pain, and diplopia (Som, 1991).
mas also populate the nasal cavity. Two examples of intrin- Radiographic findings of inverted papilloma can vary
sic sinonasal tract tumors relatively unique to the sinuses from a small nasal polypoid nodule to an expansile large
(the olfactory neuroblastoma and the inverted papilloma, mass, which may remodel the nasal vault and extend into
both of which often cause hyposmia or anosmia) may serve the sinuses, orbits, or even the anterior skull base. CT and
as prototypes for masses in this region. MRI are very useful in defining the location and extension
of the tumor (Buchwald et al., 1990; Yousem et al., 1992)
(Fig. 1.). Calcification is not uncommon in this tumor.
1. Olfactory Neuroblastoma
Other sinonasal tract tumors, such as squamous cell car-
Olfactory neuroblastoma, or esthesioneuroblastoma, is a cinoma, adenocarcinoma, melanoma, etc., can also cause
rare nasal tumor originating from the olfactory neuroepi- hyposmia or anosmia during their late stage. Squamous
thelium lining the roof of the nasal vault and in close prox- cell carcinoma accounts for 80% of paranasal sinus malig-
imity to the cribriform plate. There have been less than nances, is most commonly seen in the maxillary sinus, and
300 reported cases in the world literature. Olfactory neu- usually demonstrates bone destruction at the time of pre-
roblastomas occur in all age groups with a peak incidence sentation. Adenocarcinomas occur most frequently in the
in the 11–20 and 51–60 year groups. There is a slight pre- ethmoid sinus while melanoma is usually seen intranasally.
ponderance of the tumor in women. The incidence of Additional benign neoplasms known to affect the
olfactory neuroblastoma has been estimated to range from sinonasal cavity include osteomas, enchondromas, schwan-
2 to 3% of all malignant intranasal neoplasms. The most nomas, and juvenile angiofibromas. Osteomas are usually
common symptoms are unilateral nasal obstruction and identified in the frontal sinus and may be a source for recur-
recurrent epistaxis. Hyposmia or rhinorrhea is not rent headache and /or recurrent sinusitis. The classic story
unusual. Extension into the orbit, paranasal sinuses, or of a frontal sinus osteoma narrowing the sinus opening is a
anterior cranial fossa may cause vision disturbances and patient who has severe sinus pain associated with takeoffs
headache (Elkon et al., 1979; Li et al., 1993; Newhill et al., from airplane flights. This is a benign mass, which is often
1985). In the detection and staging of olfactory neuroblas- completely invisible on MRI due to the presence of dense
toma, CT and/or MRI play an important role. Generally compact bone making up the mass. On the other hand, it is
speaking, MRI is more accurate than CT in showing the easily identified on CT as a markedly hyperdense bony
tumor’s intracranial extent. MRI is also exquisitely useful mass protruding in the sinus. Occasionally, the osteoma
for differentiating neoplasm from postobstructed secre- will result in mucocele formation and/or pneumocephalus
tions because of the difference in the signal intensity as the posterior wall of the frontal sinus is breached.
(secretions are bright on T2, tumor intermediate). Enchondromas are less common neoplasms of the
Unfortunately, signal intensity characteristics of various sinonasal cavity which, on CT, often have a popcorn calci-
sinonasal tract tumors overlap each other, so MRI cannot fication appearance different from the stippled calcifica-
usually predict specific tumor histology. However juvenile tion of inverted papillomas. This lesion, because of its
angiofibroma can usually be distinguished from other characteristic calcification, is best evaluated with CT.
tumors on the basis of its high vascularity and marked Schwannomas of the fifth cranial nerve are the most
enhancement. common to affect the sinonasal cavity. They will typically
Medical Imaging of Olfactory Deficits 597

Figure 1 A 40-year-old woman with 3-month history of decreasing smell sensation and left nasal obstruction. (A) Bone-targeted coro-
nal CT shows an expanded opacified left nasal cavity with bowing of the lateral nasal wall (arrows) and opacification of the left maxil-
lary and both sphenoid sinuses. (B) Axial contrast-enhanced CT scan shows erosion through the left lamina papyracea (arrow) with
displacement of the medial rectus and globe laterally. The differentation between tumor and obstructed secretions is not readily apparent
with CT. Histological diagnosis: nasal cavity carcinoma arising within a dysplastic inverted papilloma.

follow the course of the nerve and can expand skull base intraorbital spread. One of the advantages of MRI is the
foramina through which they travel. The signal intensity of ability to distinguish sinus neoplasm from postobstructive
schwannomas varies according to the content of the dense secretions. This may be difficult by CT if the secretions are
Antoni A tissue or loose Antoni B tissue, the latter being isodense to the mass and if the malignancy does not
brighter on T2W scans. Schwannomas enhance avidly, enhance dramatically. If one was forced to study the
although they may have inhomogeneity to the enhancement. patient with a single modality, the literature supports MRI
Finally, one has the juvenile angiofibroma, a fascinating as the best study for the staging of sinonasal malignancies
benign neoplasm, which appears to arise in the region of (Hunink et al., 1990; Kraus et al., 1992; Paling et al., 1987;
the sphenopalatine foramen and/or the pterygopalatine Sisson et al., 1989).
fossa The lesion accounts for 0.5% of head and neck Som et al. (1991) noted that squamous cell carcinoma
masses and is typically seen in adolescent males who pre- (low in T2 intensity) could be distinguished from inflam-
sent with epistaxis and/or a nasal mass (Mehra, 1989). The mation (high in T2 intensity). They compared CT to MRI
lesion is highly vascular as exemplified on MRI by the sig- for mapping sinonasal tumors. They found that MRI and
nal flow voids within the lesion and its marked contrast CT were equivalent in 23 of 53 patients in defining tumor
enhancement. Because of its propensity for spreading via extent and that MRI was superior to CT in 26 patients. Of
the canals and foramina at the skull base, MRI is probably the 4 cases in which CT was superior, subtle bony erosion
the study of choice for the evaluation of this neoplasm. (2) and osteo(1)-cartilaginous(1) lesions accounted for the
Embolization of these lesions will assist the surgeon in “misses” on MRI. Of 60 inflammatory lesions, MRI was
limiting blood loss if resection is considered. superior (Bonte et al., 1993) or equivalent (Everall et al.,
1991) to CT in all cases. Inflammation (bright) and neo-
plasm (intermediate) could be distinguished in 95% of
3. Malignant Neoplasms
cases based on T2W signal intensity. Even when the sinus
CT and MRI probably play complementary roles in the secretions become increasingly inspissated and the signal
evaluation of sinonasal malignancies because of CT’s intensity on T2W scans decreases, the neoplasm can be dis-
superiority in defining bony margins and MRI’s superior tinguished from the obstructed secretions by its typical
soft tissue resolution and ability to define intracranial or heterogeneity as opposed to the smooth homogenous
598 Li et al.

appearance of sinus secretions. This is also true in the squamous cell carcinoma signal intensity characteristics
cases of mucoceles, which may occur after or in associa- on MRI, the lesion is characterized by a low signal inten-
tion with sinus neoplasms. Additionally, MRI has shown sity on T2W scans. This is why differentiation with
that most squamous cell carcinomas of the sinonasal cav- obstructed secretions which are typically bright in signal
ity enhance with gadolinium in a solid fashion as opposed intensity on T2W scans is so easy on MRI.
to a peripheral rim of enhancement in sinus secretions Because of Som et al.’s early work depicting sinonasal
and/or mucoceles. Unfortunately, lymphomas, undifferen- malignancies as hypointense on T2W scans, people have
tiated carcinomas, inverted papillomas, and some sarco- come to rely on this pulse sequence for mapping cancers
mas may have identical signal intensity and enhancement (Som et al., 1990). Unfortunately, low intensity on T2W
characteristics as squamous cell carcinoma. scans is an inconstant finding in sinonasal malignancies in
Gadolinium is particularly useful for demonstrating general. Hunick et al. found that over 50% of head and
epidural or meningeal invasion of neoplasms. Often, post- neck malignancies had signal intensity on T2W scans that
contrast scans must be combined with fat suppression was brighter than muscle and isointense to brain (Hunink
techniques in order to identify enhancement amidst the et al., 1990). Approximately 25% of benign tumors had the
abundant skull base fat. In one series, 75% of patients with same intensity pattern. Lanzieri et al. (1991) also reported
intracranial extension of sinonasal malignancies had addi- that the signal intensities of tumors, mucoceles, schwan-
tional information about tumor extent demonstrated with nomas, and obstructed secretions may show some overlap.
postcontrast MRI studies (van Tassel et al., 1991). Som et al. (1991) have found that minor salivary gland
Subtraction MRI of pregadolinium scans from post- masses and schwannomas may have T2W signal intensity
gadolinium scans may improve visibility of such subtle similar to that of inflammatory lesions. Minor salivary
enhancement (Lloyd and Barker, 1991). It should be noted gland tumors and melanoma are the next most common
that meningeal enhancement need not necessarily imply malignancies to affect the sinonasal cavity after squamous
neoplastic invasion; just as in cases of meningioma, the cell carcinoma (van Tassel et al., 1991). The minor salivary
dura may enhance because of reactive fibrovascular gland tumors represent a wide variety of histological types
changes alone. including adenoid cystic carcinoma, mucoepidermoid car-
When one encounters a sinonasal mass that is eroding cinoma, adenocarcinoma, and undifferentiated carcinoma.
intracranially, one must consider carcinoma, olfactory neu- Of these tumors, adenoid cystic carcinoma is the most
roblastoma, sarcomas, lymphomas, sinonasal polyposis, common variety. Its signal intensity may be high or low on
and inverted papillomas. Twelve percent of patients with T2W scans, possibly related to the degree of tubular or
polyposis and mucoceles eventually erode the skull base cribriform histological pattern as well as cystic spaces,
(Som et al., 1991). The pattern of bone destruction may be necrosis, and tumor cell density. Tissue specificity is not
similar between malignant and benign lesions at the readily achievable with MRI or CT. Gadolinium is of par-
non–sinus bearing skull base. Bone remodeling in this ticular use with adenoid cystic carcinomas, which have a
location is a rarity; a permeative pattern is the norm for all propensity for perineural spread (Graamans and Slootweg,
lesions. Som et al. (1988) have suggested that a lesion with 1989). With sinonasal cavity malignancies one should
homogeneous signal intensity invading intracranially is always attempt to trace back the branches of the fifth cra-
more likely to be a malignancy, whereas heterogeneity nial nerve via the pterygopalatine fossa, foramen rotun-
suggests an inflammatory cause. Unfortunately, necrosis, dum, foramen ovale, and orbital fissures in order to
hemorrhage, or calcification in carcinomas, olfactory neu- identify perineural neoplastic spread.
roblastomas, or sarcomas may cause signal heterogeneity. Adenocarcinomas of the paranasal sinuses have a
Polyps generally enhance in a peripheral pattern; true neo- predilection for the ethmoid sinuses and appear more com-
plasms enhance solidly. Malignancies have a broad flat monly in woodworkers. This tumor also tends to have low
base of skull erosion; benign conditions have a rounded signal intensity on T2W MRI images but may have high
polypoid intracranial excrescence. signal intensity in a small percentage of cases.
Squamous cell carcinomas account for 80% of Sarcomas of the sinonasal cavities are very rare, with
the malignancies to affect the paranasal sinuses and 80% in chondrosarcoma being the most common. Again, the his-
the maxillary sinus. The hallmark of malignancies of the tological diagnosis is probably better suggested by CT
sinonasal cavity is bony destruction, seen in approximately based on the characteristic whorls of calcification.
80% of CT scans of sinonasal squamous cells carcinoma at However, for staging, MRI is competitive with CT, and,
initial presentation. The lesion is confined to the maxillary particularly if repeat examinations are going to be
antrum in only 25% of cases at presentation (Lyons and required, follow-up with MRI to avoid the radiation expo-
Donald, 1991). In most series documenting sinonasal sure of CT is recommended.
Medical Imaging of Olfactory Deficits 599

Melanoma is a tumor that is usually identified in the odorant to the olfactory receptor area. The sense of smell
nasal cavity as opposed to the paranasal sinuses. It has is probably less than normal in many patients with cranio-
been associated with melanosis in which there is field facial anomalies (Crysdale, 1981). Congenital develop-
deposition of melanin along the mucosal surface of the mental abnormalities include choanal atresia, hereditary
sinonasal cavity. Therefore, multiplicity of lesions nasal septal deviation, facial hypoplasia, cleft palate, nasal
becomes a problem when dealing with melanomas. dermoids and epidermoids, cephaloceles, and gliomas, etc.
Neither CT nor MRI is particularly helpful in identifying Medical imaging techniques, especially high-resolution
the field “cancerization” of melanoma. When melanoma CT, play a key role to detect and evaluate the facial and
contains melanin there is paramagnetism which causes bony changes (Barkovich et al., 1991; Klein et al., 1987).
T1 and T2 shortening accounting for high signal intensity CT is most useful because surgical correction requires
on T1W scans and low signal intensity on T2W scans identification of and closure of the osseous abnormalities.
(Atlas et al., 1990). However, an amelanotic melanoma MRI is most effective in defining soft tissue masses such
may have bright signal intensity on T2W scans. The pres- as cephaloceles and nasal gliomas.
ence of hemorrhage associated with the melanoma, a Congenital anosmia can be associated with a number of
common occurrence because of the coincidence of epis- developmental and inflammatory conditions. Kallmann’s
taxis, may further obfuscate the signal intensity pattern syndrome, also known as hypogonadotrophic hypogonadism
(Yousem et al., 1996c). with anosmia, is a congenital X-linked disorder in which the
Lymphoma does occur in the paranasal sinuses and may olfactory bulbs and tracts are not formed. This is not associ-
have variable signal intensity as well. It is characterized by ated with holoprosencephaly, and the usual deficits are
homogeneous signal intensity without necrosis and the related to hormonal abnormalities in the pituitary gland with
association with cervical lymphadenopathy. the loss of sense of smell. Infertility often coexists. In 1993,
Metastatic disease to the paranasal sinuses is extremely an MR study of the olfactory system in Kallmann’s disease
rare. Of the primary causes of metastases to the sinuses, showed absence of the olfactory bulbs and tracts in 17 of 18
renal cell carcinoma is probably the most common. This patients while confirming the presence of the olfactory bulbs
tumor also has a propensity for hemorrhage and may also and tracts in all 10 studied patients with idiopathic hypo-
have a variable signal intensity depending upon the stage gonadotropic hypogonadism (Yousem et al., 1993, 1996a).
of hemorrhage. Some patients have absence of the olfactory bulbs and tracts
without Kallmann’s syndrome. It is unclear whether this rep-
C. Allergic Reactions resents congenital absence or whether an inflammatory con-
dition early in infancy destroys the olfactory bulbs and tracts.
Allergic rhinitis is a common upper airway condition Certain viruses have a propensity for injuring the olfactory
affecting about 30 million Americans with peak prevalence system. A recent study has noted the incomplete formation of
in the age group from 35–54 years (Baroody and Naclerio, olfactory sulci in patients with congenital anosmia as well as
1991). Hyposmia or anosmia is common with allergic a variable percentage of aplastic olfactory bulbs, tracts, and
rhinitis, mainly caused by nasal obstruction by polyps or tubercles (Di Rienzo et al., 2002). Still others may have con-
inflamed mucosa, which limit access of inspired air to the genital absence of sense of smell on the basis of early head
roof of the nasal vault (Cowart et al., 1993). The diagnos- trauma where the ciliary nerves as they crossed the cribiform
tic work-up begins with a careful history, which attempts plate may be sheared and the olfactory system is affected.
to identify offending allergens. Skin testing of specific Infectious causes may also affect the sense of smell in early
antigens is often used to confirm the diagnosis. Medical childhood, usually secondarily to viruses. In these cases one
imaging studies play a supplementary role in the evalua- sees the olfactory bulbs and tracts; but they are not functional.
tion of sinonasal airway status and differential diagnosis. Holoprosencephaly is a congenital, multiple midline
CT and MRI are also important for detecting any compli- malformation disorder that has a known association with
cations such as sinusitis, mucoceles, and aggressive polyps sensory deficits of vision and olfaction. Although variable
in patients with allergic rhinitis. Rounded excrescences amounts of aplasia and hypoplasia of the olfactory appara-
and enlargement of ostia are seen in the airway of patients tus may be identified, the most common MR finding is
with polyposis. complete absence of the olfactory bulbs, occurring in 92%
of patients. A high association with absence of the olfac-
D. Congenital or Developmental Abnormalities tory nerves and tubercles is also seen. There does appear to
be some, albeit poor, differentiation of the olfactory sulci
It is generally accepted that normal variations in the nasal and gyri recti, which were absent only in a little over half
anatomy may play a role in preventing the access of an of the subjects (Barkovich and Quint, 1993).
600 Li et al.

E. Other Peripheral Causes measure to assess naso-ethmoid trauma (Daly et al.,

1990; Kassel, 1988).
It is estimated that 30 million Americans have used
cocaine and 5 million use it regularly (Gregler and Mark,
1986). Intranasal use of cocaine and heroin has reached V. CENTRAL CAUSES OF OLFACTORY
epidemic proportions in the United States. Although DISEASES
hyposmia or anosmia has been suggested to occur often
in cocaine abusers, few studies using quantitative mea- There are numerous CNS disorders that are associated
sures of olfactory function have confirmed such reports. with olfactory dysfunction. The most common types fall in
A sole study on this topic reported that of 11 cocaine the categories of degenerative neuropsychiatric disorders,
abusers who underwent detailed olfactory testing, only hereditary conditions, trauma, and central neoplasms. Of
one was found to be anosmic and another had mild olfac- course, in some disorders the involvement of both periph-
tory discrimination dysfunction (Gordon et al., 1990). eral and central neural processes may occur.
These authors note that most cocaine abusers do not
develop permanent olfactory dysfunction. If, in fact, A. Alzheimer’s Disease
olfactory disturbance occurs as a result of heavy cocaine
use, it could be due to associated conductive disorders, It has been well documented that olfaction is significantly
nasal airway obstruction, alteration in sinonasal aerody- altered in Alzheimer’s (AD). Nearly all studies of olfactory
namics, damage to the olfactory epithelium, damage to function in patients with AD have reported decreased smell
the central olfactory system, or osteolysis of the cribri- relative to age-matched controls (see Chapter 23). These
form plate (Kuriloff, 1989). studies demonstrate marked impairment of smell function
Concerning the conductive disorders, several reports of in early AD, whether measured by identification, discrim-
osteolytic sinusitis and extensive osteocartilaginous necro- ination, or threshold sensitivity (Doty, 1991; Doty et al.,
sis of the nasal septum in cocaine abusers have been 1987; Serby et al., 1991).
described (Newman, 1988; Schweitzer, 1986). Erosion of Recent neuropathological studies have correlated well
nasal septal cartilage is a known complication of cocaine with these clinical findings. The anterior olfactory nuclei
abuse. Within the differential diagnosis for cartilaginous in AD patients contain senile plaques, neurofibrillary tan-
destruction, one should include Wegener’s granulomatosis, gles, granulovascular degeneration, and cell loss
syphilis, leprosy, lymphoma, rhinoscleroma (a klebsiella (Averback, 1983; Esiri and Wilcock, 1984). The olfactory
infection), and fungal invasion. CT, preferably in the coro- bulbs also show involvement (Esiri and Wilcock, 1984;
nal plane, can provide excellent views of septal perfora- Ohm and Braak, 1987), as does nasal sensory epithelium
tion, osteolysis, and sinusitis. (Jafek et al., 1992). In addition, central olfactory struc-
To evaluate intracranial disorders associated with tures, especially the amygdala and the entorhinal, pyri-
cocaine, MRI is the study of choice. Vasculitic infarcts, form, and temporal cortices, are frequently affected by
hypertensive hemorrhages, and white matter ischemic foci Alzheimer’s disease (Harrison, 1986; Pearson and Powell,
may be seen with MRI. Recently Tumeth and colleagues 1989). Besides the above findings, devastating nerve cell
demonstrated multifocal cortical deficits with a special loss and gliosis in the region of the hippocampal formation
predilection for the frontal and temporal lobes on SPECT have been observed at autopsy in AD patients (Ball et al.,
perfusion brain scans in chronic cocaine abusers (Tumeth 1985; Hyman et al., 1984).
et al., 1990). Similar findings have been reported by others Neuroimaging has played an important role in detecting
(Holman et al., 1991; Kolow et al., 1988). These findings some of the pathological changes of AD patients in vivo,
may suggest a central basis for some cases of cocaine- and its uses are growing, both for clinical evaluation and as
related decreased olfaction. Some studies also have a research tool. Early CT studies in AD patients demon-
revealed that cerebral atrophy develops in chronic cocaine strated generalized enlargement of the ventricular system
abusers and that the severity correlates with the duration of and sulci (George et al, 1981; Naser et al., 1980). Several
abuse (Pascual-Leone et al., 1991). reports have noted that ventricular and sulcal enlargement
Anosmia or hyposmia is a frequent sequela of high- correlate with the severity of AD (Albert et al., 1984; George
level midface fractures in which the olfactory nerves et al., 1983). However, these findings are not specific and
may be severed at the level of the cribriform plate have relatively weak correlations. de Leon and colleagues
(Kassel, 1988; Mathog, 1992). Because ethmoid com- (1989) have emphasized the rate of change in ventricular
plex and cribriform plate fractures are difficult to detect size with repeated CT scans as an important index in the
on plain radiographs, thin-section coronal CT is the best diagnosis of AD. Recently, several investigators have recog-
Medical Imaging of Olfactory Deficits 601

nized that CT and/or MRI delineation of atrophic changes in regional oxygen, and glucose metabolism, which may pro-
the temporal lobe and the hippocampus with enlargement of vide evidence supportive of the diagnosis of AD (Jagust
hippocampal-choroidal fissures strongly support the diagno- and Eberling, 1991). The above-mentioned structural
sis of AD (de Leon et al., 1988; George et al., 1990; Kesslak atrophic changes by CT and MRI are also supported by
et al., 1991; Kido et al., 1989). functional imaging studies (McDonald et al., 1991;
McDonald and colleagues (1991) reviewed MRI scans in Ohnishi et al., 1991). The major findings of functioning
22 patients with early-onset AD. The results showed that imaging studies in patients with AD are abnormal regional
patients with AD were significantly more likely than age- cerebral blood flow pattern and flow reduction. The com-
matched controls to have MR evidence of periventricular mon sites of blood flow reduction are in the temporopari-
hyperintensities on T2W scans. This study suggested that the etal region and the frontal areas. In one report (Bonte et al,
increased frequency of periventricular hyperintensities may 1993), seven patients with possible diagnosis of AD stud-
have a relationship to the disease process. Our own experi- ied by SPECT showed only frontal flow abnormalities. Is
ence with MRI studies of AD patients is that most of the this an early imaging finding which may suggest a patho-
cases with AD have, in addition to ventriculomegaly and physiologic basis to explain the decreasing smell sensation
sulcal widening, significantly reduced volume of the tempo- in AD? Of course, more studies are needed for further dis-
ral lobe and slight atrophy of olfactory bulbs. (Fig. 2). covering the nature of AD. We believe that early and cor-
Besides CT and MRI, SPECT and PET techniques are rect diagnosis of AD in vivo by neuroimaging techniques
also useful for evaluating regional cerebral blood flow, will be possible in the near future.
There is a dose-related association between apolipopro-
tein E-4 (APOE-4) allelic frequency and the development
of AD (APOE-2 may confer protection). Recent studies
have shown a decline in resting parietal, temporal, and pre-
frontal PET glucose metabolism in cognitively intact
patients with APOE-4. It remains to be seen whether this,
and/or an analogous fMRI study, may serve to be a predic-
tor of development of AD.
Recently some investigators have used dynamic con-
trast susceptibility contrast imaging MR to try to duplicate
the nuclear medicine flow studies. Indeed they have found
that relative values of temporoparietal regional cerebral
blood volume (as a percentage of cerebellar rCBV) were
reduced by a factor of 20% bilaterally in the patients with
Alzheimer disease compared to normals. Using left and
right temporoparietal rCBV as index measures, specificity
was 96% and sensitivity was 95% in moderately AD and
88% in mild AD (Harris, 1998).

B. Parkinson’s Disease

Odor detection and identification are significantly

impaired in Parkinson’s disease (PD) patients (Doty et al.,
1988, 1995; Montgomery et al., 2000) (see Chapter 23). It
is unclear whether the olfactory deficits in PD and AD
share the same cause. Not surprisingly, PD research into
the cause of smell dysfunction has focused on dopaminer-
Figure 2 A 60-year-old woman with Alzheimer’s disease.
UPSIT scores revealed severe bilateral anosmia. (A) Normal
gic changes. Brooks and colleagues (1991) have demon-
olfactory bulbs are seen (arrows) on coronal MR. Dilation of the strated by using PET that patients with PD show
olfactory sulci (arrowheads) reflects generalized atrophy. (B) significantly reduced mean uptake of 18F-dopa in the cau-
Coronal MR scan through the temporal lobes shows temporal date and putamen, especially in the posterior part of the
horn enlargement and atrophic changes, slightly worse on the putamen. Previous functional imaging studies have also
right side. indicated a reduction of striatal dopamine storage in PD.
602 Li et al.

PET technique with 18F-dopa in PD patients has also olfactory dysfunction, but the exact mechanism for hypos-
demonstrated reduced basal ganglia activity (Alavi and mia in HD patients remains to be worked out.
Hirsch, 1991). However, the olfactory deficit is unrelated
to severity of motor or cognitive symptoms and is not D. Korsakoff’s Psychosis
improved by L-dopa therapy (Doty et al., 1992), so the
underlying causes of olfactory dysfunction in PD still Patients with Korsakoff’s psychosis (KP) exhibit impaired
requires more study. odor detection, identification, and intensity estimation
CT scanning has little role in establishing the diagnosis (Jones et al., 1975; Mair et al., 1986). An animal model
of PD other than to exclude mass lesions in the brain. In study has shown that the behavior of rats recovering from
general, CT shows no specific striatal abnormalities and pyrithiamine-induced thiamine deficiency share several
occasionally only mild, nonspecific ventricular and sulcal important features with KP patients, including the impair-
enlargement. The major feature of PD on MRI appears to ments observed for smell, hearing, learning, and memory
be a trend towards a decreased width of the pars compacta (Mair et al., 1991). The mechanism of hyposia and/or
of the substantia nigra (Braffman et al, 1989). There is a dysosmia in patients with KP is unclear and still under
lateral to medial gradient of loss of the normal signal of the investigation. Olfactory perception may be selectively
pars compacta as well as volume loss. Moderate or marked impaired in KP by the diencephalon lesions that are char-
cortical atrophy tends to occur more frequently in PD acteristic of this disease. Degeneration in the mediodorsal
patients than in controls. MRI may occasionally show thalamic nucleus (the common neuropathological lesion in
abnormal decreased T2W intensity in the putamen and to a KP) and atrophy in the prefrontal areas may also cause the
lesser degree in the caudate nuclei and substantia nigra, olfactory dysfunction (Mair et al., 1986).
suggestive of iron deposition (Drayer et al, 1986). A quantitative neuropathological study of the human
cerebral cortex has shown that the number of cortical neu-
C. Huntington’s Disease rons in the superior frontal lobe in chronic alcoholic
patients is significantly reduced (Happer et al., 1987).
Patients with Huntington’s disease (HD) evidence olfac- Chronic alcoholism is also associated with smaller vol-
tory dysfunction (Doty, 1991; Moberg et al., 1987). umes of cortical white and gray matter relative to controls
Neuropathological studies in HD have demonstrated pre- (Pfefferbaum et al., 1995). Traditional neuropsychological
mature neuronal cell death and reactive gliosis occurring tests and functional imaging studies have also demon-
most markedly in the head of the caudate nuclei bilaterally strated disturbances of frontal-lobe function and metabolic
(Myers et al., 1991; Vonsattel et al., 1985). A loss of deficits in patients with KP (Joyce and Robbins, 1991;
70–80% of the striatal neurons may occur before func- Kopelman, 1991; Metter et al., 1989).
tional impairment is obvious. Similar but less extensive Brain CT scans have demonstrated that KP patients
changes also affect the putamen. Later, atrophy of the cere- show more pronounced third and lateral ventricular dilata-
bral cortex occurs as well. All of these progressive atrophic tion and wider interhemispheric fissures than matched
changes can be identified on CT and MRI scans, especially groups of normal controls and non-Korsakoff alcoholics
in the caudate nuclei, where the volume of the caudate (Jacobson and Lishman, 1990; Ron, 1983; Ron et al.,
head decreases and the intercaudate distance increases 1982). Shrinkage in the frontal lobes and cerebellum
(Simmons et al., 1986; Starkstein et al., 1989). Increased appears to be especially pronounced (Jacobson and
signal intensity in the putamen and globus pallidus has Lishman, 1990). A MRI study (Jernigan et al., 1991) has
been described in the juvenile form of Huntington’s dis- revealed that patients with KP show widespread reductions
ease, and frontal atrophy is usually present. in gray matter volumes in addition to CSF increases, with
PET studies of patients with HD have consistently the greatest reductions observed in diencephalic structures.
demonstrated hypometabolism in the caudate nuclei, often The volume losses that best differentiate the KP patients
before the development of atrophy on CT (Hayden et al., from the alcoholic controls included losses in anterior por-
1986). SPECT studies involving HD patients have also tions of the diencephalon, mesial temporal lobe structures,
revealed decreased uptake in the caudate nuclei, including and orbitofrontal cortices (areas involved in olfaction per-
the caudates of one patient with early disease and no evi- ception). Several other studies (Donnal et al., 1990;
dence of atrophy on MRI (Nagel et al., 1988; Reid et al., Gallucci et al., 1990; Squire et al., 1990; Victor, 1990) have
1988). Thus, functional imaging with PET or SPECT may also demonstrated that MRI is highly sensitive in detecting
contribute to the early diagnosis of HD. reversible diencephalon (medial thalamic) and mesen-
Theoretically, the input of caudate/putaminal fibers to cephalic (periaqueductal) lesions. In addition to general-
the limbic system and striatum may be altered, leading to ized cortical and cerebellar vermian atrophy seen on CT
Medical Imaging of Olfactory Deficits 603

and MR, recent reports have noted the presence of high evaluated the volume of the temporal lobes in schizophre-
signal intensity areas in the periaqueductal gray matter of nias by a quantitative MRI study. The results showed that
the midbrain (40%), the paraventricular thalamic regions the volume of temporal lobe gray matter was 20% smaller
(46%), the mamillothalamic tract, and in tissue surround- in the patients than in the control subjects, and lateral ven-
ing the third ventricle on T2W MR scans (T2WI). tricular volume was 67% larger in the schizophrenia group
Reversible thalamic lesions in the dorsal medial nuclei than in the control group. Schizophrenic patients tend to
have also been reported. These areas may or may not have smaller hippocampi that matched controls. schizo-
enhance (in some cases the enhancement may be dramatic, phrenias are also reported to have cavum septum pellu-
almost sarcoid-like) and may be associated with mamillary cidum more frequently than controls. In a recent study,
body atrophy. Mamillary body enhancement may be the Turetsky et al. (2000) reported that patients with schizo-
sole manifestation of Wernicke’s encephalopathy. Myelin phrenia exhibited 23% smaller olfactory bulb volume
degeneration, mamillary body volume loss, intracellular bilaterally than comparison subjects by a quantitative MRI
edema, and microglial proliferation are seen pathologically study.
(but may be present in alcoholics without Wernicke’s).
MRI findings in patients with KP may enable early F. Congenital Anosmia
diagnosis of the disease, which may have a positive effect
on both treatment and prognosis (Gallucci et al., 1990). Congenital anosmia, which traditionally has been defined
as anosmia present from a patient’s earliest recollection,
E. Schizophrenia has been recognized for centuries. The most common form
of congenital anosmia is Kallmann’s syndrome or olfac-
Impaired olfactory function has been reported in schizo- tory dysplasia, which is characterized by hypo-
phrenics, especially males (see Chapters 23 and 24). These gonadotropic hypogonadism and anosmia (Kallmann et
olfactory deficits, which are not of the same magnitude as al., 1944; Lieblich et al., 1982). The incidence of
those seen in AD and PD, are perhaps not unexpected given Kallmann’s syndrome is about 1:100,000 in men and
the occurrence of olfactory hallucinations as symptoms in a 1:50,000 in women. There has been increasing interest in
number of patients with schizophrenia and the evidence the pathology, pathophysiology, and genetics of this disor-
linking both to temporal lobe dysfunction (Rausch et al., der. Pathological and surgical studies of patients with
1977; Roberts, 1988). Neuropathological studies in schizo- Kallmann’s syndrome have shown agenesis of the olfac-
phrenic patients have reported neuronal loss in the entorhi- tory bulbs (DeMorsier and Gauthier, 1963; Males et al.,
nal region and prefrontal cortex, gliosis in the basal limbic 1973). Laboratory findings include decreased serum folli-
structures of the forebrain, and atrophy in temporolimbic cle-stimulating hormone and luteinizing hormone as well
structures (Benes et al., 1986; Falkai et al., 1988). as decreased urinary gonadotropins (Lieblich et al., 1982).
Neurophysiological function studies (including regional In medical imaging studies, CT is a limited tool for the
cerebral blood flow, brain electrical activity mapping, and demonstration of sinonasal and intracranial abnormalities
regional metabolic activity in the brain) in patients with in patients with congenital anosmia (Klein et al., 1987;
schizophrenia have demonstrated prefrontal cortex and Moorman et al., 1984). Surface coil MRI is the optimal
temporal lobe dysfunction (Mesulam, 1990). Functional modality to reveal the intricate details of the olfactory
imaging, such as PET or SPECT, in the study of schizo- bulbs, tracts, and rhinencephalon in vivo. Klingmuller and
phrenia is limited and inconclusive. However, functional colleagues (1987) have clearly demonstrated the olfactory
imaging has provided some evidence that certain schizo- sulci in a normal control group by MRI, but not in the
phrenic patients have decreased blood flow and metabolism patients with olfactory dysplasia. More recently, the
in the frontal lobes (hypofrontality) (Alavi and Hirsch, authors have studied two cases with Kallmann’s syndrome
1991). by MRI. Both showed no olfactory bulb at all and flatten-
Anatomical imaging findings have basically paralleled ing of the gyrus recti (Yousem et al, 1993, 1996a); frontal
the neuropathological changes in the brains of patients and temporal lobe volumes were normal (Fig. 3).
with schizophrenia. The most consistent finding on both In a mixed population of patents with congenital anos-
CT and MRI is an increase in the size of the cerebral ven- mia, we found olfactory bulb and tract absence (68–84%)
tricular system, especially in the frontal and temporal and hypoplasia (16–32%) in all 25 cases studied. Eight
horns, and corresponding decreases in cerebral tissue, individuals had Kallmann’s syndrome (hypogonadotropic
especially in the prefrontal cortex and in medial tem- hypogonadism with anosmia). Temporal and/or frontal
porolimbic structures (Mesulam, 1990; Suddath et al., lobe volume loss were noted in 5 individuals, mild in all
1989; Young et al., 1991). Suddath and colleagues (1989) but one individual. We concluded that congenital anosmia
604 Li et al.

G. Head Trauma

Craniofacial trauma can alter olfactory ability through one

of several mechanisms: (1) damage to the nose, sinuses, or
both with resultant mechanical obstruction to odorants, (2)
shearing of olfactory filaments as they course through the
cribriform plate, (3) contusion to the olfactory bulb, and
(4) contusion or shearing injury of the cerebral cortex, par-
ticularly the frontal and temporal lobes (see Chapter 30).
The incidence of anosmia or hyposmia after head trauma
has been reported quite variably from 2 to 38%, (Deems et
al., 1991; Doty et al., 1997; Hagan, 1967; Leigh, 1943;
Levin et al., 1985; Schechter and Henkin, 1974; Summer,
1964; Zusho, 1982) and increases with the severity of
injury (Levin et al., 1985; Summer, 1964). However, even
a minor injury can sometimes result in anosmia or hypos-
mia (Schechter and Henkin, 1974; Summer 1964). Recent
evidence has shown that the location of the hematoma or
contusion of the brain after head trauma is one of the most
important factors leading to olfactory dysfunction
(Costanzo and Zasler, 1991; Doty et al., 1997; Levin et al.,
1985; Yousem et al., 1996b). Specifically, diminished
olfactory discrimination has been confirmed in patients
with prefrontal lesions (Potter and Butters, 1980). Animal
studies have shown that the prefrontal olfactory area plays
a prominent role in the fine and specific discrimination of
odors (Tanabe et al., 1975). Besides prefrontal lesions,
temporal lobe structures are also involved in the odor pro-
cessing of odor perception (Rausch and Serafetinides,
1975; Rausch et al., 1977). Indeed frontal or temporal lobe
hematomas or contusions are now believed to be one of the
most common causes of olfactory dysfunction after head
injury (Costanzo and Zaster, 1991; Doty et al., 1997; Levin
et al., 1985; Schellinger et al., 1993; Yousem et al., 1996b)
(Fig. 4).
It has been established that plain skull radiography
plays only a small role in the evaluation of head trauma
(Masters et al., 1987). CT currently is the study of choice
when diagnostic imaging is necessary after acute head
trauma (Cohen, 1990; Kelly et al., 1988). CT can detect
Figure 3 (A) Coronal 500/20 scan from normal volunteer
(64-year-old woman with normal smell function) demonstrates
subarachnoid hemorrhage, fractures, and intraventricular
normal olfactory bulbs (arrows). (B) Coronal 500/17 scan of blood, lesions for which MRI is less sensitive acutely. CT
27-year-old woman with congenital anosmia without can be performed with close patient monitoring in a rapid
Kallmann’s syndrome shows extremely atrophic olfactory fashion. However, MRI is superior to CT in the detection
bulbs (arrows). (C) Coronal 500/14 scan of 29-year-old male and characterization of subacute injuries, hemorrhage
patient with Kallmann’s syndrome evidences absence of olfac- outside the subarachnoid space as in subdural hematomas,
tory bulbs and tracts with flattened gyrus rectus (arrow) on the cortical contusion, and shearing injuries. MRI is exquis-
right side, but with normal-appearing gyrus rectus on the left itely sensitive to diffuse axonal injuries leading to
side. demyelination. MRI is also useful in the follow-up of brain
contusion and/or hemorrhage, thereby eliminating the
or hyposmia appears to be an olfactory bulb tract phenom- radiation exposure associated with CT (Cohen, 1990;
enon rather than a central process (Yousem et al., 1996a). Zimmerman et al., 1986).
Medical Imaging of Olfactory Deficits 605

tory test scores (Doty et al., 1997b; Yousem et al.,

1996b). Abnormalities on MR in patients with posttrau-
matic olfactory dysfunction occur at a very high rate
(88%), predominantly in the olfactory bulbs, tracts, and
inferior frontal lobes. Qualitative and quantitative assess-
ments of damage show little correlation with olfactory
tests probably due to multifocal injury, ciliary nerve dam-
age, and the constraints of small sample size.

H. Brain Tumors

The incidence of chemosensory changes caused by

intracranial tumors has rarely been investigated. In a study
of 750 consecutive patients presenting with chemosensory
disorders to the University of Pennsylvania Smell and
Taste Center, only two cases (0.3%) were induced by
brain tumors (Deems et al., 1991). In one study anosmia
was reportedly present in 19 of the 26 cases of Foster-
Kennedy syndrome (retrobulbar optic neuritis, central
scotoma, optic atrophy on the side of the lesion and con-
tralateral papilledema usually occurring in tumors of the
frontal lobe of the brain which press downward) (Jarus
and Feldon, 1982). Bakay (1984) emphasized that loss of
smell perception is one of the first signs of olfactory
meningiomas.
In general, tumors or other destructive lesions involving
the olfactory bulb, tract, or prefrontal lobe may cause
olfactory deficits. Temporal lobe tumors usually cause
olfactory hallucinations. It is estimated that approximately
20% of the tumors of the temporal lobe produce some form
of olfactory disturbance (Furstenberg et al., 1943). The
presence of olfactory deficits correlates more with the
location of tumors than the histology (Fig. 5).

I. Acquired Immunodeficiency Syndrome

Figure 4 A 20-year-old woman with posttraumatic anosmia.
Olfactory deficits of patients with human immunodefi-
(A) A small olfactory tract is seen on the right side (arrow), but
none is seen on the left. Severe inferior frontal lobe encephalo-
ciency virus (HIV) infection have been reported (Brody
malacia is soon on this coronal T1W MR scan. (B) et al., 1991; Heald et al., 1998). These authors suggest that
Encephalomalacia is well seen on the T2W MR scan where impaired olfaction might serve as a marker of early central
hyperintense signal (S) has replaced the inferior frontal lobes nervous system HIV involvement. The principal
(where smell processing occurs). histopathological abnormalities in the brain of acquired
immunodeficiency syndrome (AIDS) patients are in the
At at our institution, 25 patients with posttraumatic subcortical structures, predominantly in the central white
smell dysfunction were evaluated by olfactory testing and matter, deep gray structures including the basal ganglia,
MR. Quantitative and qualitative gradings for olfactory the thalamus, and the brain stem (Petito et al., 1986; Price
bulb, tract, subfrontal region, hippocampus, and temporal et al., 1988). Everall et al., (1991) have found that the neu-
lobe damage correlated with olfactory test results. Twelve ronal numerical density in the frontal cortex is signifi-
patients were anosmic, 8 had severe impairment, and 5 cantly lower in HIV patients than in controls—a loss of
were mildly or impaired. Olfactory bulb and tract (88% about 38% of neurons in the superior frontal gyrus in AIDS
of patients), subfrontal (60%), and temporal lobe (32%) patients. This may account for the olfactory deficits in
injuries were found but did not correlate well with olfac- these patients.
606 Li et al.

Therefore, the possibility of peripheral cause of olfactory

deficits in AIDS patients also has to be taken into account
in certain cases.

J. Multiple Sclerosis

Multiple sclerosis (MS), a markedly debilitating

neurological disease, affects millions of Americans in the
prime of their lives. Though the influence of MS on the
sense of smell has long been controversial, recent MRI
studies (Doty et al., 1997, 1999) have demonstrated that
the olfactory function in patients with MS is closely corre-
lated with the number of demyelinating plaques within
central olfactory processing areas of the brain, as deter-
mined by MRI (Fig. 6, 7). A strong negative relationship
(Spearman r
0.94) was found between scores on the
University of Pennsylvania Smell Identification Test
(UPSIT) and the number of plaques within the inferior
frontal and temporal lobe regions (Doty et al., 1997). A
close association was present, longitudinally, between the
remission and exacerbation of plaque numbers and UPSIT
scores, with lower UPSIT scores occurring during periods
of exacerbation (Doty et al., 1999).

K. Other Central Causes

There are also reports of olfactory dysfunction in

hypochondriasis, amyotrophic lateral sclerosis, epilepsy,
and migraine (Doty et al., 1991b; Mott and Leopold,
1991). Although the pathogenesis of olfactory dysfunction
in these disorders is still unclear, it appears that a central
mechanism is involved, rather than a peripheral one.

VI. OVERVIEW AND DISCUSSION

Figure 5 Temporal lobe mass in a 62-year-old woman with
olfactory hallucinations. (A) T2W MR scan reveals a relatively It is apparent from the studies reviewed in this chapter and
well-defined right temporal lobe mass with mild mass effect. (B) the information presented elsewhere in this volume that
Contrast-enhanced T1W MR image shows peripheral enhance- olfactory dysfunction can be due to numerous causes.
ment of the tumor with a satellite nodule laterally. Sulci are Once an olfactory disorder has been recognized, the most
effaced and the temporal horn is obliterated. important step in the diagnostic process is to determine the
site of the lesion, i.e., anatomical localization.
Neuroradiological study has found that patients with Unfortunately, current clinical olfactory testing is unable
HIV infection show widened cortical sulci, enlarged ven- to localize the site of morphological changes (Doty et al.,
tricles, cerebral atrophy, and brain stem atrophy when 1984). Modern medical imaging techniques can be of great
compared with controls (Brun et al., 1986; Elovaara et al., value in the anatomical classification and localization of
1990; Post et al., 1988). Opportunistic infections and CNS the common causes of olfactory dysfunction (Li et al.,
lymphoma may be superimposal on these changes. The 1994). The most common source of olfactory dysfunction
pathogenesis of the olfactory deficits of AIDS patients is the peripheral pathway (Goodspeed et al., 1987; Mott
needs further investigating but most likely will relate to and Leopold. 1991). In the evaluation of peripheral causes,
disease in the prefrontal lobe. In addition to CNS changes, the “sinus series” radiographs offer limited information. At
sinusitis in HIV-infected patients is common and severe. present, high-resolution CT, especially coronal scans, is
Medical Imaging of Olfactory Deficits 607

Figure 7 Axial T2W MRI scans from a severely micrsomic

(UPSIT
20) 50-year-old man with an 8-year history of MS. The
place of section in (A) is 6 mm below that of (B). Note the promi-
Figure 6 A 55-year-old MS patient with no significant olfac- nent plaques (10  5 mm each) within the posterior part of the
tory dysfunction, as measured by the UPSIT. Axial T2W MRI white matter of the gyrus rectus of the L and R subfrontal lobe
scan shows no obvious plaques in the inferior frontal and tempo- regions (arrows 1 and 2, respectively), and the relatively high signal
ral lobe regions (A). Numerous plaques were identified in supra- intensity plaques in the subtemporal lobe regions (arrows 3 and 4).
periventricular regions (B).
abnormalities of the brain parenchyma revealed by neu-
roimaging studies in patients with AD, KP, or schizophre-
the study of choice to look at the bony sinonasal structures nia involve central brain areas that contain netrons of
and the ostiomeatal complex. CT can also provide impor- olfactory projections including areas of the prefrontal lobe,
tant information as a road map, which may be needed for temporal lobe, hippocampus, and thalamus.
surgical treatment. Recent studies have provided a clear physiological
MRI possesses special ability in soft tissue discrimina- explanation for decreased olfactory function in patients
tion and offers multiplanar capabilities. In the evaluation with MS (Doty et al., 1997, 1998, 1999). Current studies
of the central causes of olfactory disturbances, MRI has a from our laboratory suggest that MS, with its relatively
paramount role. Neuroimaging studies of patients with discrete focal regions of demyelination lesion, may be of
olfactory deficits related to neuropsychiatric problems value in studying brain regions involved in sensory per-
have revealed interesting findings and possibly clues for ception in addition to olfaction.
understanding some of the links between olfactory deficits It is much more difficult to explain the olfactory dys-
and pathophysiological changes of the brain. The neu- function in PD patients, and presently imaging studies have
roimaging findings of patients with AD, KP, or schizo- been of little use in clarifying this matter. Loss of olfaction
phrenia share some similarities. Thus, almost all of the in these patients may be related to factors with dopamine
608 Li et al.

and dopamine receptors, although, as noted earlier, no REFERENCES

return of function accompanies L-dopa treatment. In addi-
tion, pathological changes in the areas of putamen and cau- Alavi, A., and Hirsch, I. J. (1991). Studies of central nervous sys-
date nuclei, which have fibers connected with limbic tem disorders with single photon emission computed tomo-
system and striatum, may contribute to the loss of the sense graphy and positron emission tomography: Evolution over the
of smell. In this hypothesis, the olfactory dysfunction in PD past 2 decades. Semin. Nucl. Med. 21:58–81.
Albert M, Naser M. A, Levine H. L, et al. (1984). Ventricular size
patients might share a similar etiology to patients with HD.
in patients with presenile dementia of the Alzheimer’s type.
In congenital disorders, such as Kallmann’s syndrome, Arch. Neurol. 41:1258–1263.
the cause of anosmia can be seen on MRI studies as the Allphin AL, Strauss M, and Abdul-Karin F. W (1991). Allergic
absence of olfactory bulbs (Yousem et al., 1993, 1996a). fungal sinusitis: Problems in diagnosis and treatment.
Other congenital abnormalities, such as choanal atresia Laryngoscope 101:815–820.
and meningoencephaloceles, also can be detected by imag- Atlas S. W, Braffman B. H, LoBrutto R, Elder D. E, Herlyn D
ing studies (Klein et al., 1987; Moorman et al., 1984). (1990). Human malignant melanomas with varying degrees of
In the categories of head trauma and brain tumors, imag- melanin content in nude mice: MRI imaging, histopathology,
ing studies have shown strong links between olfactory dys- and electron paramagnetic resonance. I. Comput. Assist.
function and the location of the damaged brain. The Tomog. 14:547–554.
histology of the tumor or traumatic injury is less critical than Averback, P. (1983). Two new lesions in Alzheimer’s disease.
Lancet 2:1203.
its location (Costanzo and Zasler, 1991; Jarrus and Feldon,
Bakay L. (1984). Olfactory meningiomas: The missed diagnosis.
1982; Schellinger et al., 1983; Yousem et al., 1996b). JAMA 251:53–55.
Hyposmia or anosmia induced by occupational or acci- Ball M. J, Fishman M, Hachinski V, et al. (1985). A new definition
dental exposure to toxins, as well as that induced by of Alzheimer’s disease: A hippocampal dementia. Lancet 1:
intranasal use of drugs such as cocaine, has been tradition- 14–16.
ally thought to be due to damage to the peripheral path- Barat J L (1990). Mucoceles of the sphenoidal sinus. Report of six
ways. However, one study has suggested that olfactory cases and review of the literature. J. Neuroradiol. 17: 135–151.
deficits caused by occupational exposure to toxins may Barkovich A J, Vandermark P, Edwards MSB, and Cogen P H
have both peripheral toxic and CNS effects (Schwartz (1991). Congenital nasal masses: C T and M R imaging fea-
et al., 1989). Imaging studies have shown CNS complica- tures in 16 cases. AJR 156:587–598.
tions in cocaine abusers (Holman et al., 1991; Kalkow et Barkovich A J, and Quint, D. J. (1993). Middle interhemispheric
fusion: an unusual variant of holoprosencephaly. AJNR
al., 1988; Pascual-Leone et al., 1991; Tumeth et al., 1990),
14:431–440.
and one report of anosmia as a sequela of hydrogen sulfide Baroody F. M, and Naclerio RM (1991). Allergic rhinitis. In
(H2S) inhalation suggested the loss to be due to central Smell and Taste in Health and Disease, Getchell TV, et al.
brain damage (Tvedt et al., 1991). (Eds.). Raven Press, New York, pp. 529–552.
Benes F. M, Davidson J, and Bird E. D (1986). Quantitative
cytoarchitectural studies of the cerebral cortex of schizo-
VII. SUMMARY phrenics. Arch. Gen. Psychiatry 43:31–35.
Bonte B. L, Tintner R, Weiner MF, et al. (1993). Brain blood flow
Medial imaging is an essential part of the evaluation of in the dementias: SPECT with histopathologic correlation.
patients with olfactory disorders. In the assessment of the Radiology 186:361–365.
peripheral causes of olfactory deficits, medical imaging stud- Braffman B. H, Grossman R. I, Goldberg H. I, et al. (1989). MR
ies, especially CT and/or MRI, can reveal anatomical infor- imaging of Parkinson disease with spin-echo and gradient-
mation and structural changes, suggest differential diagnosis, echo sequences. AJR 152:159–165.
and provide the road map that may be needed for surgical Brasch R. C (1992). New directions in the development of MR
intervention. On the other hand, in the evaluation of the cen- imaging contrast media. Radiology 183:1–11.
Brody D, Serby M, Etenne N, et al. (1991). Olfactory identifica-
tral causes, MRI, fMRI, PET, or SPECT can provide infor-
tion deficits in HIV infection. Am. J. Psychiatry 148:248–250.
mation elucidating the links between olfactory dysfunction Brooks D. J, Ibanez V, Playford E. D, et al. (1991). Presynaptic
and the structural or functional changes in the living brain. and postsynaptic striatal dopaminergic function in neuroacan-
thocytosis: A positron emission tomographic study. Ann.
Neurol. 30:166–171.
ACKNOWLEDGMENTS Brun B, Boesen F, Gerstoft J, et al. (1986). Cerebral computed
tomography in men with acquired immunodeficiency syn-
Supported, in part, by Grants RO1 DC04278, RO1 drome. Acta Radiol. 27:385–387.
AG17496, RO1 DC 02974, and PO1 DC00161 (R. L. Buchwald C, Nielsen L. H, Ahlgren P, et al. (1990). Radiologic
Doty, Principal Investigator). aspects of inverted papilloma. Eur. I. Radiol. 10:134–139.
Medical Imaging of Olfactory Deficits 609

Carter B. L, and Runge V. S (1988). Imaging modalities for the Doty R. L, Bartoshuk L. M, and Snow J. B Jr (1991). Causes of
study of paranasal sinuses and nasopharynx. Otolaryngol. olfactory and gustatory disorders. In Smell and Taste in
Clin. North Am. 21:395–420. Health and Disease, Getchell TV, et al. (eds.). Raven Press,
Cohen W (1990). Recent developments in the imaging of neu- New York, pp. 449–462.
raxis trauma. Curr. Opin. Radiol. 2:34–39. Doty R. L, Kimmelman C. P, and Lesser R. P (1992a). Smell and
Costanzo R. M, and Zasler N. D (1991). Head trauma. In Smell taste and their disorders. In Diseases of the Nervous System:
and Taste in Health and Disease, Getchell TV, et al. (Eds.). Clinical Neurobiology, Vol. 1, Asbury AK, Mckhann GM, and
Raven Press, New York, pp. 711–730. McDonald WI (Eds.), W.B. Saunders, Philadelphia, pp.
Cowart B. J, Flynn-Rodden K, McGeady S. J, and Lowery L. D 390–403.
(1993). Hyposomia in allergic rhinitis. J. Allergy Clin. Doty, R. L, Stern M. B, Pfeiffer C, et al. (1992b). Bilateral olfac-
Immunol. 91:747–751. tory dysfunction in early stage medicated and unmedicated
Crysdale WS (1981). Otorhinolaryngologic problems in patients idiopathic Parkinson’s disease. J. Neurol. Neurosurg.
with craniofacial anomalies. Otolaryngol. Clin. North Am. Psychiatry 55:138–142.
14:145–155. Doty, R. L, Bromley, S. M, Stern, MB (1995). Olfactory testing as
Daly B. D, Russell J. L, Davidson M. J, and Lamb J. T (1990). an aid in the diagnosis of Parkinson’s disease: development of
Thin section computed tomography in the evaluation of naso- optimal discrimination criteria. Neurodegeneration 4:93–97.
ethmoid trauma. Clin. Radiol. 41:1267–1272. Doty R. L, Li C, Mannon L. J, and Yousem D. M (1997a).
De Morsier G, and Gauthier G (1963). La dysplasie olfacto-gen- Olfactory dysfunction in multiple sclerosis [published erra-
itale. Path. Biol. 11:(1)267–272. tum appears in N Engl J Med 1997; 337(7):507]. N. Engl. J.
de Leon M. J, McRae T, Tsai JR, et al. (1988). Abnormal cortical Med. 336:1918–1919.
response in Alzheimer’s disease linked to hippocampal atro- Doty R. L, Yousem D. M, Pham L. T, et al. (1997b). Olfactory
phy. Lancet 2:391–392. dysfunction in patients with head trauma. Arch. Neurol.
de Leon M. J, George A. E, Reisberg B, et al. (1989). Alzheimer’s 54:1131–1140.
disease: Longitudinal CT studies of ventricular change. AJR Doty R. L, Li C, Mannon L. J, and Yousem D. M (1999).
153:1257–1262. Olfactory dysfunction in multiple sclerosis: Relation to longi-
Deems D. A, Doty R. L, Settle R. G, et al. (1991). Smell and taste tudinal changes in plaque numbers in central olfactory struc-
disorders: A study of 750 patients from the University of tures. Neurology 53:880–882.
Pennsylvania Smell and Taste Center. Arch. Otolaryngol. Drayer B. P, Olanow W, Burger P, et al. (1986). Parkinson plus
Head Neck Surg. 117:519–528. syndrome diagnosis using high field MR imaging of brain
Di Rienzo, L., Artuso, A., and Colosino, C. (2002). Isolated iron. Radiology 159:493–498.
congenital agenesis of olfactory bulbs and tracts in a child Drutman J, Babbel RW, Harnsberger HR, et al. (1991). Sinonasl
without Kallman’s syndrome. Ann. Otol. Rhinol-Larygol. polyposis. Semin. Ultrasound CT MR 12:561–574.
111:657–660. Elkon D, Hightower S. I, Lim ML, et al. (1979). Esthesio-
Donnal J. F, Heinz E. R, and Burger P. C (1990). MR of neuroblastoma. Cancer 44:1087–1094.
reversible thalamic lesions in Wernick’s syndrome. AJNR 11: Elovaara I, Poutiainen E, Raininko R, et al. (1990). Mild brain
893–894. atrophy in early HIV infection: The lack of association with
Doty RL (1991). Olfactory dysfunction in neurodegenerative dis- cognitive deficits and HIV-specific intrathecal immune
orders. In Smell and Taste in Health and Disease, Getchell response. I. Neurol. Sci. 99:121–136.
TV, et al. (Eds.). Raven Press, New York, pp. 735–751. Esiri M. M, and Wilcock G. M (1984). The olfactory bulbs in
Doty R. L, Frye R (1989). Influence of nasal obstruction on smell Alzheimer’s disease. J. Neurol. Neurosurg. Psychiatry 47:56–60.
function. Otolaryngol. Clin. North Am. 22:397–411. Everall I. P, Lutherat P. J, and Lantos P. L (1991). Neuronal loss
Doty, R.L. & Mishra, A. (2001). Influences of nasal obstruction, in the frontal cortex in HIV infection. Lancet 337:1119–1121.
rhinitis, and rhinosinusitis on the ability to smell. Falkai P, Bogerts B, and Rozumek M (1988). Limbic pathology
Laryngoscoppe 111:409–423. in schizophrenia: The entorhinal region — a morphometric
Doty R. and Practice of Rhinology, Goldman J (Ed.). John Wiley study. Biol. Psychiatry 24:515–521.
& Sons, New York, pp. 761–785. Fowler J. S, Hoffman E. J, Larson S. M, et al. (1988). Positron
Doty RL, Shaman P, and Dann M (1984). Development of the emission tomography: A new approach to brain chemistry.
University of Pennsylvania Smell Identification Test: A stan- JAMA 206:2704–2710.
dardized microencapsulated test of olfactory function. Furstenberg A. C, Crosby E, and Farrior B (1943). Neurologic
Physiol. Behav. (Monogr.) 32:489–502. lesions which influence the sense of smell. Arch. Otolaryngol.
Doty R. L, Reyes P. F, and Gregor T (1987). Presence of both 38:529–530.
odor identification and detection deficits in Alzheimer’s dis- Gallucci M, Bozzao A, Splendiani A, et al. (1990). Wernicke
ease. Brain Res. Bull. 18:597–600. encephalopathy: MR findings in five patients. AJR
Doty R. L, Deems D. A, Stellar S (1988). Olfactory dysfunction 155:1309–1314.
in parkinsonism: A general deficit unrelated to neurologic George A. E, deLeon M. J, Ferris S. H, et al. (1981). Parenchymal
signs, disease stage, or disease duration. Neurology 38: CT correlates of senile dementia (Alzheimer’s disease): Loss
1237–1244. of grey-white discriminality. AJNR 2:205–213.
610 Li et al.

George A. E, de Leon MJ, Rosenbloom S, et al. (1983). Jacobson, R. R, and Lishman, W. A (1990). Cortical and dien-
Ventricular volume and cognitive deficit: A computed tomo- cephalic lesions in Korsakoff’s syndrome: A clinical and CT
graphic study. Radiology 149:493–498. scan study. Psychol. Med. 20:63–75.
George A. E, de Leon M. J, Stylopoulos L. A, et al. (1990). CT Jafek, B. W, Eller, P. M, Johnson, E. W, et al. (1992).
diagnostic features of Alzheimer disease: Importance of the Ultrastructural changes of the olfactory epithelium in
choroidal/hippocampal fissure complex. AJNR 11:101–107. Alzheimer’s disease. Am. J. Rhinol. 6:219–225.
Goodspeed, R. B, Gent, J. F, Leonard G, et al. (1987). The preva- Jagust, W. J, and Eberling, J. L (1991). MRI, CT, SPECT, PET:
lence of abnormal paranasal sinus x-rays in patients with Their use in diagnosing dementia. Geriatrics 46:28–35.
olfactory disorders. Conn. Med. 51:1–3. Jarus, G. D, and Feldon, S. E (1982). Clinical and computer
Gordon, A. S, Moran, D. T, Jajek, VW, et al (1990). The effect of tomographic findings in the Foster Kennedy syndrome. Am. J.
chronic cocaine abuse on human olfaction. Arch. Otolaryngol. Ophthalmol. 93:317–322.
Head Neck Surg. 116:1415–1418. Jernigan, T. L, Schafer, K, Butters, N, and Cermak, LS (1991).
Graamans, K, and Slootweg, P. J. (1989). Orbital exenteration in Magnetic resonance imaging of alcoholic Korsakoff patients.
surgery of malignant neoplasms of the paranasal sinuses. The Neuropsychopharmacology 4:175–186.
value of preoperative computed tomography. Arch. Jolles, P. R, Chapman, P. R, and Alavi, A (1989). PET, CT, and
Otolaryngol. Head Neck Surg. 115:977–980. MRI in the evaluation of neuropsychiatric disorders: Current
Gregler, L. L, and Mark, H (1986). Medical complications of applications. J. Nucl. Med. 30:1589–16.
cocaine abuse. N. Engl. J. Med. 315:1495–1500. Jones, B. P, Moskowitz, H. R, Butters N, et al. (1975).
Gussack, G. S, and Hudgins, P. A. (1991). Imaging modalities in Psychophysical scaling of olfactory visual and auditory stim-
recurrent head and neck tumors. Laryngoscope 101:119–124. uli by alcoholic Kosakoff’s patients. Neuropsychologica 13:
Hagan, P. J (1967). Post-traumatiac anosmia. Arch. Otolaryngol. 387–393.
85:107–111. Joyce, E. M, and Robbins, T. W (1991). Frontal lobe function in
Happer, C, Kril, J, Daly J (1987). Are we drinking our neurones Korsakoff and non-Korsakoff alcoholics’ planning and spatial
away? Br. Med. J. 294:534–536. working memory. Neuropsychologica 29:709–723.
Harris, G. J, Schlaepfer, T. E, Peng, L. W, Lee S, Federman, E. B, Kallmann, F. J, Schoenfeld, W. A, and Barrera, S. E (1944). The
and Pearlson G. D (1994). Magnetic resonance imaging eval- genetic aspects of primary eunuchoidism. Am. J. Ment. Defic.
uation of the effects of ageing on grey-white ratio in the 48:203–221.
human brain. Neuropathol. Appl. Neurobiol. 20: 290–293. Kassel, E. E (1988). Traumatic injuries of the paranasal sinuses.
Harris, GJ, Aylward, E. H, Peyser C. E, et al. (1996). Single pho- Otolaryngol. Clin. North Am. 21:455–493.
ton emission computed tomographic blood flow and magnetic Kelly, A. B, Zimmerman, R. D, Snow, R. B, et al. (1988). Head
resonance volume imaging of basal ganglia in Huntington’s trauma: Comparison of MR and CT—experience in 100
disease. Arch. Neurol. 53:316–324. patients. AJNR 9:699–708.
Harris, G. J, Lewis, R. F, Satlin, A, et al. (1998). Dynamic sus- Kern, R. C. (2000). Chronic sinusitis and anosmia: pathologic
ceptibility contrast MR imaging of regional cerebral blood changes in the olfactory mucosa. Laryngoscope 110:
volume in Alzheimer disease: a promising alternative to 1071–1077.
nuclear medicine. AJNR 19:1727–1732. Kesslak, J. P, Nalcioglu, O, and Cotman, C. W (1991). Qualification
Harrison, P. J (1986). The pathogenesis of Alzheimer’s disease: of magnetic resonance scans for hippocampal and parahip-
Beyond the cholinergic hypothesis. J. Roy. Soc. Med. pocampal atrophy in Alzheimer’s disease. Neurology 41: 51–54.
79:347–351. Kido, D. K, Caine, E. D, LeMay M, et al. (1989). Temporal lobe
Hayden, M. R, Martin, W. R. W, Stoessl, A. J, et al. (1986). atrophy in patients with Alzheimer’s disease: A CT study.
Positron emission tomography in the early diagnosis of AJNR 10:551–555.
Huntington’s disease. Neurology 36:888–894. Kimmelman, C. P (1991). Medical imaging of smell and taste
Heald A. E, Piper C. F, and Schiffman SS (1998). Taste and smell disorders. In Smell and Taste in Health and Disease, Getchell
complaints in HIV-infected patients. AIDS 12:1667–1674. TV, et al. (Eds.). Raven Press, New York, pp. 471–479.
Healy B (1992). From form to function — better imaging techniques Klein, V. R, Friedman, J. M, Brookshire, G. S, et al. (1987).
extend study of living system (from NIH). JAMA 267: 2863. Kallmann’s syndrome associated with choanal atresia. Clin.
Holman B. L, Carvalho P. A, Mendelson J, et al. (1991). Brain Genet. 31:224–227.
perfusion is abnormal in cocaine-dependent polydrug users: A Klingmuller, D, Dewes W, Krabe T, et al. (1987). Magnetic reso-
study using Technetium-99m-HMPAO and ASPEC. J. Nucl. nance imaging of the brain in patients with anosmia and hypo-
Med. 32:1206–1210. thalamic hypogonadism (Kallmann’s syndrome). J. Clin.
Hunink, M. G, de Slegte, R. G, Gerritsen, G. J, Speelman, H Endocrinol. Metab. 65:581–584.
(1990). CT and MRI assessment of tumors of the nose and Kolkow, N. D, Mullani, N, Gould, K. L, et al. (1988). Cerebral
paranasal sinuses, the nasopharynx and the parapharyngeal blood flow in chronic cocaine users: A study with positron
space using R0C methodology. Neuroradiology 32:220–225. emission tomog. Br. J. Psychiatry 152:641–648.
Hyman, B. T, Van Hoesen, G. W, and Damasio, A. R (1984). Kopelman, M. D (1991). Frontal dysfunction and memory
Alzheimer’s disease: Cell-specific pathology isolates the hip- deficits in the alcoholic Korsakoff syndrome and Alzheimer-
pocampel formation. Science 225:1168–1170. type dementia. Brain 114:117–137.
Medical Imaging of Olfactory Deficits 611

Kraus, D. H, Lanzieri, C. F, Wanamaker, J. R, Little, J. R, and Mehra, Y. N; Mann, S. B; Dubey, S. P; Suri S. (1989). Computed
Lavertu P (1992). Complementary use of computed tomogra- tomography for determining pathways of extension and a
phy and magnetic resonance imaging in assessing skull base staging and treatment system for juvenile angiofibromas. Ear
lesions. Laryngoscope 102:623–629. Nose Throat J. 68:576–589.
Kuriloff, D. B (1989). Nasal septal perforation and nasal obstruc- Mesulam, M. M (1990). Schizophrenia and the brain. N. Engl. J.
tion. Otolaryngol. Clin. North Am. 22:333–350. Med. 322:842–845.
Lanzieri, C. F., Shah, M., Krauss, D., and Lavertu, P. (1991). Use Metter, E. J, Kempler D, Jackson C, et al. (1989). Cerebral glu-
of gadolinium-enhanced MR imaging for differentiating cose metabolism in Wernickes’s, Broca’s, and conduction
mucoceles from neoplasms in the paranasal sinuses. aphasia. Arch. Neurol. 46:27–34.
Radiology 178:425–428. Moberg, P. J, Pearlson, G. D, Speedie, L. J, et al. (1987).
Leigh, A. D (1943). Defects of smell after head injury. Lancet Olfactory recognition: Differential impairments in early and
1:38–40. late Huntington’s and Alzheimer’s dtsease. J. Clin. Exp.
Levin, H. S, High, W. M, and Eisenberg, H. M (1985). Neuropsychol. 9:650–664.
Impairment of olfacory recognition after closed head injury. Montgomery, E. B, Koller, W. C, LaMantia, T. J, et al. (2000).
Brain 108:579–591. Early detection of probable idiopathic Parkinson’s disease:
Li C, Yousem, D. M, Hayden RE, and Doty RL (1993). Olfactory Development of a diagnostic test battery. Movement Disord.
neuroblastomas: MR evaluation. AJNR 14:1167–1171. 15:467–473.
Li C, Yousem DM, Doty RL, and Kennedy DW (1994). Moorman, J. R, Crain B, and Osborne D (1984). Kallmann’s syn-
Neuroimaging in patients with olfactory dysfunction. AJR drome with associated cardiovascular and intracranial anom-
162:411–418. alies. Am. J. Med. 77:369–372.
Lieblich, J. M, Rogol, A. D, White, B. J, et al. (1982). Syndrome Mott, A. E, and Leopold, D. A (1991). Disorders in taste and
of anosmia with hypogonadism Kallmann’s syndrome) Am. J. smell. Med. Clin. North Am. 75:1321–1353.
Med. 73:506–519. Myers, R. H, Vonsattell, J. P, Paskevich, P. A, et al (1991).
Loury, M. C, and Kennedy, D. W (1991). Chronic sinusitis and Decreased neuronal and increased oligodendroglial densities
nasal polyposis. In Smell and Taste in Health and Disease, in Huntington’s disease caudate nucleus. J. Neuropathol. Exp.
Getchell TV, et al. (Eds.). Raven Press, New York, pp. 517–528. Neurol. 50:729–742.
Lloyd, G. A, and Barker, P. B. (1991). Subtraction magnetic res- Nagel, J. S, Johnson, K. A, Ichise M, et al. (1988) Decreased
onance for tumors of the skull base and sinuses: a new imag- iodine-123 IMP caudate nucleus uptake in patients with
ing technique. J. Laryngol. Otol. 105:628–631. Huntington’s disease. Clin. Nucl. Med. 13:486–496.
Lloyd, G. A. S, Lund, V. J, Phelps, P. D, Howard, D. J. (1987). Naser, M. A, Gebhardt C, and Levine, H. L (1980). Decrease
MRI in the evaluation of nose and paranasal sinus diseases. computerized tomography numbers in patients with presenile
Br. J. Radiol. 60:957–968. dementia. Arch. Neurol. 37:401–409.
Lyons, B. M, Donald, P. J. (1991). Radical surgery for nasal cav- Newbill, E. T, Johns, M. E, Cantrel,l R. W (1985).
ity and paranasal sinus tumors. Otolaryngol. Clin. North Am. Esthesioneuroblastoma: Diagnosis and management. South.
24:1499–1521. Med. J. 78:275–282.
Mair, R. G, Doty, R. L, Kelly, K. M, et al. (1986). Multimodel Newman, N. M, DiLoreto, D. A, Ho, T. I, et al. (1988). Bilateral
sensory discrimination deficits in Korsakoff’s psychosis. optic neuropathy and osteolytic sinusitis. Complications of
Neuropsychologica 24:831–839. cocaine abuse. JAMA 259:72–74.
Mair, G. R, Knoth, R. L, Rabchenuk, S. A, et al. (1991). Ohm, T. G, Braak H (1987). Olfactory bulb changes in
Impairment of olfactory auditory, and spatial serial reversal Alzheimer’s disease. Acta Neuropathol. (Berlin) 73:365–369.
learning in rats recovered from pyrithiamine-induced thi- Ohnishi, T, Hoshi, H, Nagamachi S, et al. (1991) Regional cere-
amine deficiency. Behav. Neurosci. 105:360–374. bral blood flow study with 123I-IMP in patients with degener-
Males, J. L, Townsend, J. L, and Schneider, R. A (1973). ative dementia. AJNR 12:513–520.
Hypogonadotrophic hypogonadism with anosmia — Paling, M. R, Black, W. C, Levine, P. A, et al. (1987). Tumor
Kallmann’s syndrome. Arch. Int. Med. 131:501–507. invasion of the anterior skull base: A comparison of MR and
Masters, S. J, McClean, P. M, Arcarese, J. S, et al. (1987). Skull CT studies. J. Comput. Assist. Tomogr. 11:824–830.
x-ray examinations after head trauma. Recommendations by Parker, G. D, and Harnsberger, H. R (1991). Clinical-radiologic
multidisciplinary panel and validation study. N. Engl. J. Med. issues in perineural tumor spread of malignant diseases of the
316:84–91. extracranial head and neck. RadioGraphics 11:383–399.
Mathog, R. H (1992). Atlas of Craniofacial Trauma. W.B. Pascual-Leone A, Dhuna A, and Anderson D (1991). Cerebral
Saunders Co., Philadelphia, pp. 377–379. atrophy in habitua cocaine abusers: A planimetric CT study.
McAlister, W. H, Lusk R, Muntz, H. R (1989). Comparison of Neurology 41:34–38.
plain radiographs and coronal CT scans in infants and chil- Pearson, R. C. A, and Powell, T. P. S (1989). The neuroanatomy
dren with recurrent sinusitis. AJR 153:1259–1264. of Alzheimer’s disease. Rev. Neurosci. 2:101–122.
McDonald, W. M, Krishnan, K. R, Doraiswamy, P. M, et al Petito, C. K, Cho E-s, Lemann W, et al. (1986). Neuropathology
(1991). Magnetic resonance findings in patients with early- of acquired immunodeficiency syndrome (AIDS): An autopsy
onset Alzheimer’s disease. Biol. Psychiatry 29:799–810. review. J. Neuropathol. Exp. Neurol. 45:635–646.
612 Li et al.

Pfefferbaum, A., Sullivan, E. V., Mathalon, D. H., Shear, P. K., Serby M, Larson P, and Kalkstein, D (1990). Olfactory sense in
Rosenbloom, M. J., and Lim, K. O. (1995). Longitudinal psychoses. Biol. Psychiatry 28:830.
changes in magnetic resonance imaging brain volumes in Serby, M, Larson, P, and Kalkstein, D (1991). The nature and
abstinent and relapsed alcoholics. Alcoholism: Clinical and course of olfactory deficits in Alzheimer’s disese. Am. J.
Experimental Research 19:1177–1191. Psychiatry 148:357–360.
Phillips, P. P, Gustafson, R. O, and Facer, G. W (1990). The clin- Shapiro, M. D, and Som, P. M (1989). MRI of the paranasal
ical behavior of inverting papilloma of the nose and paranasal sinuses and nasal cavity. Radiol. Clin. North Am. 27: 447–475.
sinuses: Report of 112 cases and review of the literature. Simmons, J. T, Pastakia B, Chase, T. N, et al. (1986). Magnetic
Laryngoscope 100:463–469. resonance imaging in Huntington’s disease. Am. J.
Post, M. J. D, Tate, L. G, Quencer, R. M, et al. (1988). CT, MR, Neuroradiol. 7:25–28.
and pathology in HIV encephalitis and meningitis. AJR 151: Sisson, G. A Sr, Toriumi, D. M, and Atiyah, R. A. (1989).
373–380. Paranasal sinus malignancy; a comprehensive update.
Potter, H, and Butters, N (1980). An assessment of olfactory Laryngoscope 99:143–150.
deficits in patients with damage to prefrontal cortex. Som, P. M (1991). Tumors and tumor-like conditions. In Head
Neuropsychologica 18:621–628. and Neck Imaging, 2nd ed., Som PM (Ed.). Mosby-Year Book
Price, J. L (1985). Beyond the primary olfactory cortex: Inc., St. Louis, pp. 169–227.
Olfactory-related areas in the neocortex, thalamus and hypo- Som, P. M, Shapiro, M. D, Biller, H. F, et al. (1988). Sinonasal
thalamus. Chem. Senses 10:239–258. tumors and inflammatory tissues: Differentiation with MR
Price, J. L (1990). Olfactory system. In The Human Nervous imaging. Radiology 167:803–808.
System, Paxinos, G. (Ed.). Academic Press, Inc., San Diego Som, P. M, Dillon, W. P, Curtin, H. D, et al. (1990). Hypointense
pp. 979–998. paranasal sinus foci: Differential diagnosis with MR imaging
Price, R. W, Brew B, Sidtis J, et al (1988). The brain in AIDS: and relation to CT findings. Radiology 176:777–781.
Central nervous system HIV-I infetion and AIDS dementia Squire, L. R, Amaral, D. G, and Press, G. A (1990). Magnetic res-
complex. Science 239:586–592. onance imaging of the hippocampal formation and mammil-
Rausch R, and Serafetinides, E. A (1975). Specific alteration of lary nuclei distinguish medial temporal lobe and diencephalic
olfaction in humans with temporal lobe lesions. Nature amnesia. J. Neurosci. 10:3109–3117.
255:557–558. Starkstein, S. E, Folstein, S. E, Brandt J, et al. (1989) Atrophy in
Rausch R, Serafetinides, E. A, Crandall PH (1977). Olfactory Huntington’s disease: A CT-scan study. Neuroradiology
memory in patients with anterior temporal lobectomy. Cortex 31:156–159.
13:445–452. Suddath, R. L, Casanova, M. F, Goldberg, T. E, et al. (1989).
Reid, I. C, Besson, J. A. O, Best, P. V, et al. (1988). Imaging of Temporal lobe pathology in schizophrenia. A quantitative mag-
cerebral blood flow markers in Huntington’s disese using sin- netic resonance imaging study. Am. J. Psychiatry 146:464–472.
gle photon emission computed tomography. J. Neurol. Sumner, D (1964). Post-traumatic anosmia. Brain 87:107–120.
Neurosurg. Psychiatry 51:1264–1268. Suzuki, M, Takashima T, Kadoya M, et al. (1989). MR imaging
Reiman, E. M, and Mintun, M. A (1990). Positron emission of olfactory bulbs and tracts. AJNR 10:955–957.
tomography (editorials). Arch. Intern. Med. 150:729–731. Tanabe, T, Jino M, Takagi SF (1975). Discremination of odors in
Roberts, G. W (1988). Abnormalities in brain structure in schiz- olfactory bulb, pyriform-amygdaloid areas, and orbitofrontal
ophrenia. Curr. Opin. Psychiatry 1:83–89. cortex of the monkey. J. Neurophysiol. 38:1284–1296.
Ron, M. A (1983). The alcoholic brain: CT scan and psycholog- Tumeth, S. S, Nagel, J. S, English RJ, et al. (1990). Cerebral
ical findings. Psychol. Med. Monograph, Supplement 3. abnormalities in cocaine abusers: demonstration by SPECT
Ron, M. A, Acker W, Shaw GK, et al. (1982). Computerized perfusion brain scintigraphy. Radiology 176:821–824.
tomography of the brain in chronic alcoholics: A survey and Turetsky, B. I, Moberg, P. J, Yousem, D. M, et al. (2000).
follow-up study. Brain 105:497–514. Reduced olfactory bulb volume in patients with schizophre-
Schechter PJ, and Henkin RI (1974). Abnormalities of taste and nia. Am. J. Psychiatry 157:828–830.
smell after head trauma. J. Neurol. Neurosurg. Psychiatry 37: Tvedt, B. Skyberg, K, Aaserud, O, et al. (1991). Brain damage
802–810. caused by hydrogen sulfide: A follow-up study of six patients.
Schellinger D, Henkin RT, and Smirniotopoulos JG (1983). CT Am. J. Int. Med. 20:91–101.
of the brain in taste and smell dysfunction. AJNR 4:7 Van Tassel, P, Lee, Y. Y. (1991). Gd-DTPA enhanced MRI for
52–754. detecting intracranial extension of sinonasal malignancies. J.
Schwartz, B, Doty, R. L, Frye, R. E, et al. (1989). Olfactory func- Comput. Assist. Tomogr. 15:387–392.
tion in chemical workers exposed to acrylate and methacrylate Victor, M (1990). MR in the diagnosis of Wernicke-Korsakoff
vapors. Am. J. Public Health 79:613–618. syndrome. AJR 1315–1316.
Schweitzer, V. G (1986). Osteolytic sinusitis and pneumomedi- Vogl, T, Dresel S, Bilaniuk LT, et al. (1990) Tumors of the
astinum: Deceptive otolaryngologic complications of cocaine nasopharynx and adjacent areas: MR imaging with Gd-DTPA.
abuse. Laryngoscope 96:206–210. AJNR 11:187–194.
Medical Imaging of Olfactory Deficits 613

Vonsattel, J. P, Myers, R. H, Stevens, T. J, et al. (1985). imaging during odor stimulation: Preliminary data.
Neuropathological classification of Huntington’s disease. J. Neuroradiology 204:833–838.
Neuropathol. Exp. Neurol. 44:559–577. Yousem, D. M., Geckle, R. J., Bilker, W. B. & Doty, R. L. (1998).
Young AH, Blackwood DHR, Roxborough H, et al. (1991). A Olfactory bulb and tract and temporal lobe volumes:
magnetic resonance imaging study of schizophrenia: Brain Normative data across decades. Ann. NY Acad. Sci.
structure and clinical symptoms. Br. J. Psychiatry 158–164. 855:546–555.
Yousem, D. M, Fellows, D. W, Kennedy, D. W, et al. (1992). Yousem, D. M., Geckle, R. J., Bilker, W. B., Kroger, H. & Doty,
Inverted papilloma: MR evaluation. Radiology 185:501–505. R.L. (1999a). Posttraumatic smell loss: Relationship to psy-
Yousem, D. M, Li, C, Turner, W. J. D, et al. (1993). Kallmann’s chophysical tests and volumes of the olfactory bulbs and
syndrome: MR evaluation of olfactory system. AJNR tracts and the temporal lobes. Acad. Radiol. 6:264–272.
14:839–843. Yousem, D. M., Maldjian, J. A., Hummel, T., Alsop, D. C.,
Yousem, D. M, Geckle, R. J, Bilker W, McKeown DA, Doty RL. Geckle, R. J., Kraut, M. A., Doty, R. L. (1999b). The effect of
(1996a). MR evaluation of patients with congenital hyposmia age on odor-stimulated functional magnetic resonance imag-
or anosmia. AJR 166:439–443 ing. Am. J. Neuroradiol. 20:600–608.
Yousem, D. M, Geckle R. J, Bilker WB, McKeown DA, and Doty Yousem, D. M., Maldjian, J. A., Siddiqi, F., Hummel, T., Alsop,
RL. (1996b). Posttraumatic olfactory dysfunction: MR and D. C., Geckle, R. J., Bilker, W. B., and Doty, R. L. (1999c).
clinical evaluation. AJNR 17:1171–1179 Gender effects on odor-stimulated functional magnetic reso-
Yousem, D. M., Li, C., Montone, K.T., et al. (1996c). Primary nance imaging. Brain Res. 818:480–487.
malignant melanoma of the sinonasal cavity: MR imaging Zimmerman, R A, Bilaniuk, L. T, Hackney, D. B, et al. (1986).
evaluation. RadioGraphics 16:1101–1110. Head injury: early results of comparing CT and high-field
Yousem, D. M., Geckle, R. J., Doty, R. L. & Bilker, W. B. (1997a). MR. AJR 147:1215–1222.
Reproducibility and reliability of volumetric measures of Zinreich, S. J., Kennedy, D. W., Roenbaum, A. E., et al. (1987).
olfactory eloquent structures. Acad. Radiol. 4:264–269. Paranasal sinuses: CT imaging requirements for endoscopic
Yousem, D. M., Williams, S. C. R., Howard, R. O., Andrew, C., surgery. Radiology 163:769–775.
Simmons, A., Allin, M., Geckle, R.J., Suskin, D., Bullmore, Zusho, H. (1982). Post-traumatic anosmia. Arch. Otolaryngol.
E. T., Brammer, M. J. & Doty, R. L. (1997b). Functional MRI 108:90–92.
29

Plasticity Within the Olfactory Pathways: Influences of

Trauma, Deprivation, Stem Cells, and Other Factors

Joel Maruniak
University of Missouri, Columbia, Missouri, U.S.A.

I. INTRODUCTION within 90 days. Schultz (1941, 1942) discovered the regen-

erative properties of the olfactory epithelium while con-
The olfactory pathways of terrestrial mammals are excep- ducting experiments aimed at finding a way to protect
tional for their extraordinary plasticity. For most compo- humans from polio virus. In those studies he found that
nents of the adult nervous system, plasticity is limited to the destruction of the monkey olfactory epithelium by zinc
standard capabilities of individual neurons such as extend- sulfate conferred protection against intranasal exposure to
ing and retracting axons and dendrites, changing synaptic polio virus. However, a few months later he determined
strength, and regenerating severed processes. While this that susceptibility of the monkeys to intranasal virus deliv-
ordinary type of plasticity is important and interesting, it ery had returned. He concluded that destruction of the
pales in comparison to the more global capabilities of the olfactory epithelium prevented infection by removing the
olfactory pathways. For example, the olfactory system of pathway into the central nervous system (CNS) but that it
adults shows a unique ability to respond to and recover must regenerate within a few months. In 1960 he published
from trauma and sensory deprivation. This striking plastic- the first definitive paper documenting the regeneration of
ity is largely attributable to the presence of stem cells, the olfactory epithelium after experimental destruction
which normally supply replacements for certain neurons in (Schultz, 1960). In the ensuing decade other researchers
the olfactory epithelia and bulbs. In response to trauma or confirmed Schultz’s findings (Andres, 1965; Takagi,
deprivation, the production and survival of these stem cells 1969). In the 1970s it was established that the plasticity of
can be increased or decreased. In many ways these proper- the adult olfactory epithelium arose from its property of
ties of the adult olfactory system are reminiscent of those of normal turnover of receptor neurons (Moulton, 1974). For
the other sensory systems during development. the next 15 years the olfactory epithelium was thought to
The unusual regenerative properties of the olfactory be unique in being the only part of the mammalian nervous
epithelia were known for decades before the plasticity of system where neurons were routinely replaced during
the bulbs was established. The first indication that the adulthood (see Chapter 5).
olfactory epithelium might possess remarkable plasticity The idea that the olfactory bulbs might also possess
arose from studies in the early 1940s. Nagahara (1940) cut unusual plasticity arose from Meisami’s early studies of
the olfactory nerves in mice and reported that most cells in the effects of unilateral naris closure on their develop-
the epithelium had degenerated by 3 days and regenerated ment (Meisami, 1976). At the time, the first reports of the

615
616 Maruniak

negative effects of early visual deprivation had just been suffice it to say that these basal cells endow the olfactory
published (Barlow, 1975). In order to test the effect of epithelium with its peerless responses to trauma by
odor deprivation on postnatal development of the olfac- providing replacements for lost receptor neurons.
tory system, Meisami pioneered a technique for closing a Even after seemingly complete destruction of the entire
naris in newborne rats (Meisami, 1976). He found that population of receptor neurons, the olfactory epithelium is
after a month of closure, the deprived-side bulbs were able to reconstitute itself and restore behavioral function.
28% smaller than the nondeprived bulbs. Many subse- In animal experiments, receptor neurons are typically
quent studies have confirmed this striking effect of destroyed either by cutting their axons or by exposure of
neonatal naris closure on bulb size (see Brunjes and the nasal cavity to a toxic chemical such as ZnSO4, Triton-
Frazier, 1986). X, or methyl bromide (Fig. 1). In most cases acute destruc-
Because of the normal turnover of receptor neurons, tion of receptor neurons leads to a dramatic increase in
researchers in the 1980s began touting the olfactory mitosis of the basal cells (Camara and Harding, 1984;
epithelium as a part of the adult nervous system in which Schwartz-Levy et al., 1991; Carr and Farbman, 1992;
development could be studied. This concept led my lab Schwob et al., 1992) and restoration of the olfactory
to wonder if adult olfactory bulbs might similarly retain epithelium and odor-guided behavior within a month or
the kind of plasticity they had been shown to possess two (Harding and Wright, 1979; Monti-Graziadei et al.,
during early postnatal development. In the late 1980s we 1980; Matulionis et al., 1982; Samanen and Forbes, 1984;
performed unilateral naris closures on adult mice and Costanzo, 1985; Yee and Costanzo, 1995; Schwob et al.,
found that their bulbs responded almost identically to the 1999; Cummings et al., 2000).
bulbs of neonatally closed mice (Maruniak et al., 1989). In humans, a number of diseases may cause loss of
That study indicated that the olfactory bulbs were unlike receptor neurons, but it is generally believed that viral
any other sensory component of the CNS in retaining infection is one of the leading causes (Douek et al., 1975;
inordinate sensitivity to sensory deprivation through- Doty, 1979) (see Chapter 22). Another major cause of
out life. anosmia is injury or severance of the receptor axons during
In subsequent studies we showed that the unique plas- head trauma (Doty et al., 1997) (see Chapter 30). Even
ticity of the adult olfactory bulbs had its origin in the nor- though the olfactory epithelium is able to regenerate new
mal turnover of its granule cells (Corotto, et al., 1993). We receptor neurons, for reasons not yet understood there is
reported that neuronal precursor cells continued to be pro- often incomplete or no recovery of olfactory acuity after
duced throughout life in the subependymal layer around posttraumatic loss. Explanations have centered around the
the ventricles and migrate into the olfactory bulbs where possibility that glial scar formation might inhibit regener-
they differentiate into granule cells. That and subsequent ation or block the choanae in the cribiform plate between
studies overturned the long-held belief (Rakic, 1985; the epithelia and bulbs (Pasterkamp et al., 1998).
Morshead and van der Kooy, 1992) that the adult brain pro-
duces no new neurons.
In this chapter we will first examine the origins and
hallmarks of the extraordinary plasticity of the olfactory
pathways and then will consider the factors that can
impact this plasticity. Because of the role that stem cells
play in enabling plasticity, the origin and utility of neural
and embryonic stem cells will be addressed in Sec. VI of
this chapter.

II. PLASTICITY OF THE OLFACTORY

EPITHELIUM

The incredible plasticity of the olfactory epithelium is

largely attributable to its possession of a population of Figure 1 A section through the nasal septum showing normal
stem cells, the basal cells, which allow turnover of the olfactory epithelium on the left and the great loss of receptor neu-
receptor neurons. Chapter 5 of this book discusses the role rons from the epithelium on the right after the olfactory nerves on
of these stem cells in the maintenance of the normal olfac- that side were cut. Stained with an antibody to olfactory marker
tory epithelium. For the purpose of the present review, protein. (Scale bar
30 m.)
Plasticity Within the Olfactory Pathways 617

III. PLASTICITY OF THE OLFACTORY CNS Benson et al., 1984; Brunjes and Frazier, 1986; Maruniak
STRUCTURES et al., 1989). This effect is of a much larger magnitude than
has been reported for the other senses. Recently, Brunjes’
A. Olfactory Bulbs
group showed that the olfactory system is almost com-
pletely plastic in being able to recover even from the effects
The plasticity of the olfactory CNS pathways, though not
of neonatal naris closure (Cummings et al., 1997).
as great as that of the olfactory epithelia, surpasses that of
While the resiliency of the olfactory bulbs is great, it is
other parts of the adult brain. The extraordinary elements
limited because interneurons are the only type of cell that is
of olfactory CNS plasticity can be traced in large part to
replaced. Thus, if an insult or degenerative process causes
the olfactory bulbs, which are unique in the scope of their
loss of mitral or tufted cells—the output neurons of the
responses to trauma and sensory deprivation throughout
bulbs—then the function of the bulb is irretrievably
life. This unique plasticity is largely endowed by the con-
impaired. By contrast, the recovery of the olfactory epithe-
tinual turnover of the granule cells, which are resupplied
lium can occur after virtually complete destruction because
by precursors migrating from a population of stem cells
all of its cell types appear to be able to be regenerated
around the lateral ventricles (Fig. 2). These stem cells will
(Huard et al., 1998) (see Sec. II) (see also Chapter 5).
be addressed in Sec. VI of this chapter.
The dramatic changes that occur in the deprived bulb
Sense-appropriate stimulation during postnatal devel-
are manifold and all lamina of the bulbs seem to be
opment is essential for full elaboration of the CNS compo-
affected (Brunjes and Frazier, 1986; Henegar and
nents of sensory systems. For most senses such stimulation
Maruniak, 1991). The layer that is most markedly affected
must occur within a circumscribed period of time after
is the external plexiform layer, which contains mainly the
birth called the critical period. A critical period is a devel-
processes of, and synaptic interactions between, the
opmental window of time during which stimulus-driven
mitral/tufted cells and granule cells (Fig. 3). The granule
innervation of CNS sensory structures normally occurs. If
cell interneurons are by far the most numerous type of neu-
an animal is deprived of appropriate stimulation during the
ron in the bulbs and are involved in various sorts of odor
critical period, then the CNS fails to develop properly. In
processing. In fact, most of the gross anatomical effects of
most cases this leads to irreversible sensory deficits. In
naris closure can be traced to the loss of granule cells.
contrast, deprivation outside a critical period typically has
However, in possums, which are born in a very immature
little or no permanent effects.
state and in which mitral cell formation continues postna-
Studies from our and other labs have shown that olfac-
tally, neonatal deprivation also decreases the number of
tion responds quite differently to deprivation than the other
mitral cells (Cummings et al., 1997).
senses. The most striking difference is that there appears to
In adult mice we ascertained that naris closure reduced
be no critical period. Deprivation at any postnatal time
the number of granule cells by 30% (Henegar and
leads to similar negative changes in the anatomy and
Maruniak, 1991). Using tritiated thymidine autoradiogra-
physiology of the system. Thus, unilateral naris closure for
phy and quantification of apoptotic cells, we established
longer than one month causes the deprived-side bulbs of
that these losses were due to decreased proliferation and
rodents to atrophy by about 25% compared to the open-side
reduced survival of granule cell precursors within the
bulbs (Fig. 1) (Meisami, 1976; Brunjes and Borror, 1983;
deprived bulbs (Corotto et al., 1994).

1. Olfactory Bulb Neurochemistry

In addition to morphological changes, there are also sig-
nificant changes in the neurochemistry and metabolism of
odor-deprived bulbs. Levels of succinate dehydrogenase
and cytochrome oxidase are depressed by neonatal closure
(Cullinan and Brunjes, 1987), while NADPH diaphorase
(nitric oxide synthase) immunoreactivity is not affected
(Croul-Ottman and Brunjes, 1988). The amount of
dopamine and tyrosine hydroxylase messenger RNA in
deprived bulbs is reduced by at least half (Stone et al.,
1990; Wilson and Wood, 1992). To compensate for these
Figure 2 Path of the rostral migratory stream from the prolif- reduced dopamine levels in the deprived bulb, dopamine
erative areas around the lateral ventricle into the olfactory bulb. D2 receptors are upregulated by 32% (Guthrie et al.,
618 Maruniak

decreases in some bulb enzymes such as tryrosine

hydroxylase and neurotransmitters such as dopamine and
norepinephrine, while not affecting others (Nadi et al.,
1981; Kawano and Margolis, 1982; Baker et al., 1983).

B. Olfactory Cortex

The most common methods for assessing plasticity of the

olfactory cortex are unilateral naris closure, olfactory bul-
bectomy, and lateral olfactory tract lesions. Of these, olfac-
tory bulbectomy is, by far, the most common (Alberts,
1974; Brunjes, 1992). In this section, we first examine the
relatively scant data available on the effects of deprivation
on the physiology and anatomy of the olfactory cortex.
Next, we discuss the effects of bulbectomy and lateral
Figure 3 Effects of unilateral naris closure of 2 months on the olfactory tract lesion.
morphology of the olfactory bulbs of a mouse. There is a striking
reduction in the overall size of the deprived bulb on the left. The
1. Deprivation
largest decreases are in the granule cell layer (GCL) and the
external plexiform layer (EPL). The rostral migratory stream is The kinds of striking morphological changes seen in the
located in the middle of the bulbs between the arrows within the deprived olfactory bulbs do not extend to higher-order cor-
right bulb. (Scale bar
100 m.) tical structures. Brunjes’ group found that size and other
features of the anterior olfactory nucleus (the second-order
structure in the olfactory CNS) were relatively unaffected
1991). In contrast, beta-adrenergic receptor levels are sig-
by neonatal naris closure (Brown and Brunjes, 1990).
nificantly reduced in the deprived bulb (Woo and Leon,
However, a more detailed study of the anterior piriform
1995), possibly in compensation for a transient increase in
cortex revealed that there were some subtle reductions in
bulbar norepinephrine (Wilson and Wood, 1992).
size and other morphological features (Wilson et al.,
In bulbs deprived of olfactory stimulation during adult-
2000).
hood, dopamine and tyrosine hydroxylase levels are also
markedly decreased while catecholamine and GAD levels
2. Bulbectomy
are unaffected (Kosaka et al., 1987; Stone et al., 1991;
Baker et al., 1993; Philpot et al., 1998). Finally, depriva- In the early postnatal olfactory system, bulbectomy appears
tion appears to rapidly and significantly depress bulbar to have little or no effect on the morphology of the olfac-
metabolism (Korol and Brunjes, 1990). Over the long tory cortex (Friedman and Price, 1986). By contrast, bul-
term, metabolic effects may be further exaggerated by a bectomy at later times causes a number of striking effects
decrease in vascularization that occurs in deprived bulbs on the olfactory cortex and other parts of the brain. For
(Korol and Brunjes, 1992). example, in the adult, bulbectomy results in rapid transneu-
Destruction of the receptor neurons by axotomy or ronal degeneration of pyramidal neurons in the piriform
chemical lesion causes a greater loss of bulb size than the cortex (Heimer and Kalil, 1978; Capurso et al., 1997).
25% seen after naris closure. Axotomy leads to a 33% loss While ablation of the olfactory bulbs causes loss of
of bulb weight after a month (Baker et al., 1984), while olfactory capabilities, it additionally leads to other deficits
ZnSO4 irrigation results in a 40–75% decrease within a that cannot be attributed solely to loss of the ability to
month (Margolis et al., 1974; Meisami and Manoochehri, smell (Alberts, 1974; Edwards, 1974; Miro et al., 1982;
1977). Some of this excess loss can be attributed to the dis- Hall and Macrides, 1983; Brunjes, 1992). One of the more
appearance of the olfactory nerve layer as a result of interesting of such collateral effects is the finding that the
destruction of the receptor neurons or their axons. olfactory bulbs play a role in chronobiology. For example,
Additional loss may be due to neuronal degeneration olfactory bulbectomy has been shown to unmask the pho-
within the olfactory bulb following lesion of the olfactory toperiodic response (i.e., light-controlled seasonal repro-
receptor neurons (Pinching and Powell, 1971; Gozzo and ductive changes) in animals that do not normally show
Fülöp, 1984). photoperiodism (Nelson and Zucker, 1981). Furthermore,
As with deprivation following naris closure, axotomy removal of the bulbs or transection of the lateral olfactory
or chemical lesion of the receptor neurons causes striking tracts (the main output pathway of the bulbs) causes an
Plasticity Within the Olfactory Pathways 619

increase in basal levels of gonadotropin secretion (Pieper attributable to problems in the olfactory epithelium, it is
et al., 1989). important to understand the factors that control neurogen-
Another interesting nonolfactory effect of bulbectomy esis, recovery, and maintenance of the epithelium.
is the similarity of the animal’s affect and neurochemistry The basal cells of the olfactory epithelia are like other
to that seen in depression. In fact, the bulbectomized ani- stem cells in requiring fibroblast growth factor (FGF), epi-
mal has become accepted as an animal model for depres- dermal growth factor (EGF), and transforming growth fac-
sion since many of the affective sequelae of bulbectomy tor (TGF) for survival and differentiation (Reynolds and
can be ameliorated by antidepressants (Van Riezen et al., Weiss, 1992; Richards et al., 1992; Kuhn et al., 1997).
1977). One of the characteristic consequences of bilateral Herzog and Otto (1999) assessed the ability of these three
bulbectomy is abnormally high circulating levels of growth factors to facilitate recovery of the epithelium after
corticosterone, which are typically associated with depres- destruction by ZnSO4 irrigation. They found that all three
sion and stress (Steckler et al., 1999). Bulbectomy also enhanced reinnervation of the bulbs in a dose-dependent
causes marked increases in the density of innervation by manner, with TGF being the most effective.
serotonergic fibers of the frontal cortex (Zhou et al., 1998). A role for the neurotropins in olfactory epithelial devel-
In addition, it elevates the expression of NMDA receptors opment and plasticity has been inferred from the expression
in the prefrontal cortex (Petrie et al., 2000; Webster et al., of their receptor subtypes, Trks A, B, and C and nerve
2000). Finally, bulbectomy causes long-term increases in growth factor receptor (NGFR), at different stages in devel-
neuropeptide Y expression in the piriform cortex and den- opment and after various insults. The olfactory receptor
tate gyrus, suggesting a possible role for this peptide in neurons hold the distinction of being the only lineage of
depression (Holmes et al., 1998). neurons to sequentially express all of those receptors dur-
Perhaps all of these effects of bulbectomy are related ing its life cycle (Roskams et al., 1996). Destruction of the
since humans with seasonal affective disorder (SAD), suf- olfactory epithelium by Triton X upregulates expression of
fer from seasonal depression. Thus, in those afflicted with NGFR in the olfactory nerve layer of the bulb, and levels
SAD, the olfactory bulbs might function abnormally or return to pre-lesion values within 16 weeks (Turner and
unusually, causing the individual to become somewhat Perez-Polo, 1994)
photoperiodic and experience depression in the wintertime. Buckland and Cunningham (1999) compared the expres-
However, a recent test of olfactory acuity in SAD patients sion of several neurotropins before and after bulbectomy and
found no significant deficits (Postolache et al., 1999). found the following. Glial cell line–derived neurotrophic fac-
tor (GDNF) is normally expressed in mature receptor neu-
rons and mitral cells but disappears from receptor neurons
IV. INFLUENCE OF GROWTH FACTORS after bulbectomy. Ciliary neurotrophic factor (CNTF)
ON PLASTICITY OF THE OLFACTORY immunoreactivity is normally expressed strongly by neurons
PATHWAYS at all stages but after bulbectomy it is also largely lost. Brain-
derived neurotrophic factor (BDNF) is seen only in horizon-
The neurotropins (members of the nerve growth factor gene tal basal cells and is unaffected by bulbectomy.
family) play an essential role in the activity-dependent Two additional compounds appear to play roles in
development and survival of neurons in sensory systems development, neurogenesis, and regeneration of the olfac-
(Thoenen, 1995) and appear to play similar roles in the tory epithelium. The first of these, insulin like growth fac-
olfactory system throughout life. Some are expressed by tor-1 (IGF-1), seems to function in the support of
olfactory neurons at different times in their life cycles, neurogenesis in the epithelium (Pixley et al., 1998). The
while others appear to be target-derived factors supplied by second, nitric oxide (which also functions as a second mes-
target tissues. senger and neurotransmitter), appears to play a part in the
We will first consider the growth factors that are known development and regeneration of the olfactory epithelium.
to play a role in the olfactory epithelium and then review In fact, the highest levels of nitric oxide synthase are seen
those that function in the olfactory bulbs. in regenerating receptor neurons (Roskams et al., 1994).

A. Olfactory Epithelium B. Olfactory Bulbs

While the olfactory epithelium has a robust capacity for It is clear that the neurotropins are also important for the
recovery from injury, it frequently experiences varying production and survival of neuronal precursors for the
degrees of loss of its sensory neurons following trauma olfactory bulbs. For example, infusion of BDNF into the
(see Sec. II). Because the loss of olfactory acuity is often lateral ventricles of adult rats leads to a 100% increase in
620 Maruniak

neuronal precursors arriving in the olfactory bulbs (Zigova limited data suggesting that dysosmia and hyposmia occur
et al., 1998). In BDNF mutant mice there is excessive cell in some hypothyroid humans (McConnell et al., 1975).
death in the granule, periglomerular, and subventricular One possible way in which thyroid hormone may exert
layers of the bulb (Linnarsson et al., 2000). Naris closure its effects is through its ability to increase neurotropin levels
reduces bulbar levels of the orphan nuclear receptor NGFI- in the CNS (Giordano et al., 1992). It also has been shown
B (Liu and Baker, 1999), which may mediate interactions to synergize with NGF in the olfactory bulbs to support nor-
between NGF and the retinoic acid pathways (Katagiri mal growth and maturation (Clos and Legrand, 1990).
et al., 2000). While the role of NGF in the bulbs is not
clear, destruction of the olfactory epithelium by Triton X
causes a dramatic decline in NGFR in the glomeruli VI. THE ROLE OF STEM CELLS IN
(Turner and Perez-Polo, 1994). PLASTICITY OF THE OLFACTORY
In keeping with the generally accepted role of EGF as BULBS
a mitogen, Craig et al. (1996) found that infusion of
EGF into the ventricles of mice caused an increase in total The stem cells that supply a continuous stream of granule
number of neuronal precursor cells and enhanced their cell precursors to the adult olfactory bulbs are located
migration. FGF-1 was found to be selectively present in around the lateral ventricles (Smart, 1961; Moorshead and
glomeruli and the external plexiform layer of the adult van der Kooy, 1992; Morshead et al., 1998). They appear to
bulb, perhaps providing trophic support for the innervating be a remnant of the stem cells whose progeny formed the
receptor neurons (Key et al., 1996). forebrain during development (Garcia-Verdugo et al., 1998).
Our study of the effects of unilateral naris closure on In fact, the proliferating progeny still traverse the same path-
the adult olfactory bulb showed that deprivation caused way, the rostral migratory stream, that furnished all the cells
decreased neurogenesis and increased cell death in the ros- for the development of the olfactory bulbs (Fig. 2).
tral migratory stream within the ipsilateral bulb (Corotto Besides the olfactory bulbs, there is only one other
et al., 1994). The most obvious explanation for those find- component of the adult mammalian CNS that undergoes
ings is that normal levels of odor stimulation and conse- turnover of its neurons: the hippocampus. Interestingly, in
quent electrical activity in the bulbs causes the production both structures the class of neurons that undergoes
of one or more growth factors that support the production replacement is granule cell interneurons. In the case of the
and survival of granule cells and their precursors. In fact, hippocampus the stem cells reside in the dentate gyrus
some studies suggest that both growth factors and electri- (Gage et al., 1998).
cal activity must be present for full support of neuronal There is currently some controversy about the actual
survival (Ghosh et al., 1994; Meyer-Franke et al., 1995). location of the stem cells in the lateral ventricles. One camp
contends that the ependymal cells that line the ventricles
are the true stem cells (Johansson et al., 1999b), while the
V. INFLUENCE OF THYROID HORMONES other believes that it is in the overlying subependymal layer
ON THE OLFACTORY PATHWAYS that the stem cells reside (Chiasson et al., 1999). Whatever
the case, the rostral migratory stream is believed to be an
Thyroid hormone is crucial for the growth and development extension of these proliferative zones that passes through
of animals (Meisami, 1984) and particularly for elaboration the forebrain and into the center of the olfactory bulbs.
of the proper biochemistry, physiology, and anatomy of the Once inside the olfactory bulbs migrating precursors
developing CNS, including the olfactory pathways change direction and move radially out of the rostral migra-
(Dussault and Ruel, 1987). A number of studies have shown tory stream to destinations in the granule cell layer or
that hypothyroidism drastically affects the development of around the glomeruli (Zigova et al., 1996).
the peripheral olfactory system, as well (Mackay-Sim and The discovery by our lab and others that a functional
Beard, 1987; Pasternostro and Meisami, 1991)—an alter- population of neural stem cells remains around the lateral
ation that can be reversed by thyroxine treatment ventricles of adult rodents stimulated research and brought
(Paternostro and Meisami, 1993). The most important post- great hope for their use in repair of the nervous system.
natal action of thyroid hormone in the olfactory periphery is The ensuing studies on neural stem cells were performed
in support of neurogenesis and maturation (Paternostro and at the same time that great strides in our knowledge of the
Meisami, 1994, 1996). In fact, odor preferences of adult power of embryonic stem cells was occuring. These stud-
mice made experimentally hypothyroid are markedly altered ies have shown that the lines, if any, between the proper-
(Beard and Mackay-Sim, 1987), although olfactory sensitiv- ties of adult neural and embryonic stem cells are quite
ity, per se, may not be (Brosvic et al., 1996). There are blurred.
Plasticity Within the Olfactory Pathways 621

Embryonic stem cells have been found to possess stem cells for research and clinical application, the journal
almost miraculous powers because they are completely Science designated stem cell research as the breakthrough
undifferentiated and appear to be able to become any type of the year in 1999.
of cell. For example, it is now clear that embryonic stem A reasonable question at this point is why one would
cells can be made to differentiate into virtually any type of even bother with adult neural stem cells when cultured
cell including neural cells or glia in culture or in vivo (Liu embryonic stem cells can seemingly do everything. One
et al., 2000). Similarly, the cells around the ventricles, reason to use adult cells would be to perform autotrans-
labeled neural stem cells because they were thought to be plantation back into an adult host’s body without worrying
capable of giving rise to only neurons or glia, are now about rejection. In most situations, removal of host stem
known to be also undifferentiated and capable of becoming cells, expansion in vitro, and then autotransplantation
almost any kind of cell (Flax et al., 1998; Clarke et al., would seem to be safest route. In addition, this strategy
2000). One of the surprising properties of stem cells is that, would circumvent the ethical issues of creating chimeric
when injected into an animal, they seem to mostly differ- humans by implantation of stem cells from another human
entiate into the correct site-specific type of cell (Lewis, or even animal. Thus, it would seem to be appropriate to
2000). For example, neural stem cells have been shown to identify and study all potential sources of stem cells,
be able to differentiate into blood cells when placed in the including neural stem cells, in the adult human so that their
bone marrow (Bjornson et al., 1999). Furthermore, Gage’s utility under different circumstances can be ranked.
group has shown that stem cells taken from the hippocam- Several sources of viable neural stem cells have already
pus will differentiate into olfactory bulb neurons when been demonstrated in adult humans. Not surprisingly, the
implanted into the rostral migratory stream and allowed to lateral ventricles and the hippocampus have been used to
migrate into the bulbs (Suhonen et al., 1996). Thus, it obtain stem cells that can be grown and expanded in cul-
seems that a stem cell is a stem cell is a stem cell, and all ture (Johansson et al., 1999a; Kukekov et al., 1999). In
are undifferentiated and totipotent. addition, the olfactory bulbs themselves have been shown
Continuous cultures of both embryonic and neural stem to be a source of culturable neural stem cells. Recently,
cells have been established. Embryonic stem cell lines stem cells were obtained from the bulbs of patients under-
have been initiated by collecting cells of the inner cell going neurosurgery, placed in culture, expanded, and dif-
mass from blastocyst-stage embryos and growing them in ferentiated into neurons and glia (Pagano et al., 2000).
culture (Evans and Kaufman, 1981; Martin, 1981). Such Thus, while the olfactory bulbs may be able to be used as
stem cells can be induced to differentiate into a neuronal a source of stem cells in adults, the procedure necessary to
phenotype by retinoic acid treatment (Bain et al., 1995). harvest such cells would undoubtedly be a risky one.
Alternatively, treatment with bone morphogenic protein For researchers and clinicians in olfaction there are a
(BMP) inhibits differentiation into the neuronal phenotype number of unanswered questions about the use of stem
(Finley et al., 1999). Neural stem cells can be harvested cells. Experiments still need to be performed addressing
from the lining of the lateral ventricles and grown indefi- questions of the limits of stem cell transplantation in our
nitely as a cell line in culture (Gage et al., 1995). Such system. Can they be used to generate mitral or tufted cells
cells have even been derived from adult human tissue in olfactory bulbs where those cell types have been lost? A
(Johansson et al., 1999a; Kukekov et al., 1999). To facili- good research animal in which to answer the latter ques-
tate detection in experiments, neural and embryonic stem tion might be the PCD mouse, which loses its mitral cells
cell lines are usually transfected with a reporter gene that over a fairly short period of time during young adulthood
allows them to be easily visualized in host tissues (e.g., by (Greer and Shepherd, 1982).
immunohistochemistry or fluorescence). Another important question is whether stem cells can be
When stem cells from such cultured lines are injected used to repair the olfactory epithelia in patients who have
back into the blastocyst stage of an embryo, they are lost the sense of smell because of peripheral damage. A
widely incorporated into the resulting animal (Bradley recent experiment suggests that this approach will be fruit-
et al., 1984; Gossler et al., 1989). In fact, even stem cells ful. Goldstein et al. (1998) found that stem cells harvested
of other species are readily incorporated at this stage and from the olfactory epithelium of bulbectomized rats could
form chimeric adults (Brustle et al., 1998). In the develop- be used to produce both neuronal and nonneuronal cell
ing brain, injection of stem cells into the ventricles leads to replacements when implanted in an olfactory epithelium
widespread incorporation of their progeny into all parts of that had been destroyed by methyl bromide. While these
the CNS. This procedure has allowed the amelioration of precursors were clearly multipotent, it is not known
neurological diseases in mutant mice (e.g., Yandava et al., whether they possess the same kind of totipotency seen in
1999). As a result of the tremendous utility and potential of stem cells from the CNS. If they do, they might provide an
622 Maruniak

easily accessible and harvestable source of stem cells for the receptor neuron turnover mentioned above to provide a
culture, expansion, and subsequent reimplantation back coordinated adjustment of olfactory sensitivity to seasonal
into an adult. changes in odors (e.g., those related to food sources, repro-
The future appears to be quite bright for stem cell duction, aggression, and parenting). In fact, several lines of
research and application. Their presence and importance in evidence suggest that such a mechanism is plausible. First,
the extraordinary plasticity of the peripheral and central we have shown that on a global level odor deprivation
olfactory pathways should direct much future work to leads to reduced accumulation of newly produced granule
olfaction. cells and increased death of existing ones (Corotto et al.,
1994). Our study suggests that the normal electrical
activity of receptor neurons maintains a balance between
VII. THE ORIGIN OF OLFACTORY
addition of new granule cells to, and subtraction of exist-
PLASTICITY
ing granule cells from, the olfactory bulbs.
Other studies, at a more local level, suggest that an indi-
I would like to conclude this chapter with a discussion of
vidual odor stimulates a unique and characteristic set of
hypotheses about why olfactory pathways express such a
functional columns in the bulbs (Onoda, 1992; Sallaz and
high degree of plasticity. As previously stated, the extra-
Jourdan, 1993; Guthrie et al., 1993). A functional column
ordinary components of olfactory plasticity can be traced
is an odor-processing unit consisting of glomeruli, which
to the presence of stem cells, which normally supply
are activated by inputs from the receptor neurons of the
replacements for receptor neurons and granule cells. Since
nose, the mitral cells that receive this input, and a group of
other sensory systems lose their ability to replace neurons
granule cells that modify the responses of the mitral cells
with maturation, the question becomes why the olfactory
(Guthrie et al., 1993). If the effects of global deprivation
system does not.
are extrapolated to the level of a functional column then it
While the rationale for replacement of receptor neurons
seems logical that chronic activity or inactivity within indi-
is not clear, there are two generally accepted contributing
vidual columns might determine whether associated gran-
factors. First, because the receptor neurons are among the
ule cells increase or decrease in number. In other words,
most, if not the most, exposed nerve cells in the body
odor-stimulated columns would tend to add new granule
(Nakashima et al., 1984), it is believed that under normal
cells and odor-deprived columns would tend to lose gran-
conditions they are so vulnerable that they have to be rou-
ule cells. This would provide a mechanism by which the
tinely replaced. Even in their relatively protected location
bulbs could contribute to adjustments in odor sensitivity.
in the recesses of the nasal cavity, they come in contact
with environmental pathogens, toxins, and particulate mat-
ter and are exposed to wide ranges of temperature and VIII. SUMMARY
humidity. Second, there is evidence that turnover of recep-
tor neurons allows the epithelium to change its sensitivity The olfactory pathways of adults possess a degree of plas-
to odors. For example, research on rats has found that ticity that approaches that of the developing nervous sys-
repeated exposure to an odor can increase the sensitivity of tem. The olfactory epithelia display an unsurpassed ability
the olfactory epithelium to that odor (Wang et al., 1993). to recover from trauma, while the bulbs show an unusually
Furthermore, in both humans and rats, repeated exposure strong response to, and ability to recover from, sensory
to an odor can increase behavioral sensitivity to that odor deprivation. Neurotropins, other growth factors, and some
and even other odors (e.g., Doty et al., 1981; Doty and hormones appear to mediate different aspects of these plas-
Ferguson-Segall, 1989; Stevens and O’Connell, 1995; tic responses. The resiliency of the olfactory pathways
Wysocki et al., 1989; Yee and Wysocki, 2001). arises from their possession of two populations of neurons
It is not known why the olfactory bulbs and hippocam- that turn over in adulthood: the olfactory receptor neurons
pus continue to replace interneurons throughout life while of the periphery and the granule cells of the brain.
the rest of the brain remains static. Certainly, neither is Replacements for the receptor neurons are supplied by the
exposed to any more trauma than other neurons of the basal cells of the epithelium. New granule cells originate
brain and so that cannot be a factor. Hypotheses for the as the progeny of stem cells which are located around the
hippocampus mainly center around the role that new lateral ventricles. These precursors migrate into the bulbs
interneurons could play in learning (Gould et al., 2000). via the rostral migratory stream. Much work is currently
Selective replacement of granule cells may enable the being directed at determining the utility of using the two
bulbs also to contribute to changes in odor sensitivity (a types of olfactory stem cells to repair damaged parts of the
form of learning, as well). This could act in concert with nervous system.
Plasticity Within the Olfactory Pathways 623

REFERENCES Camara, C. G., and Harding, J. W. (1984). Thymidine incorporation

in the olfactory epithelium of mice: early exponential response
Alberts, J. R. (1974). Producing and interpreting experimental induced by olfactory neurectomy. Brain Res. 308:63–68.
olfactory deficits. Physiol. Behav. 12:657–670. Capurso, S. A., Calhoun, M. E., Sukhov, R. R., Mouton, P. R.,
Andres, K. H. (1965). Differentiation and regeneration of sen- Price, D. L., and Koliatsos, V. E. (1997). Deafferentation
sory cells in the olfactory region. Naturwissenschaften causes apoptosis in cortical sensory neurons in the adult rat.
52:500. J. Neurosci. 17:7372–7384.
Bain, G., Ray, W. J., Yao, M., and Gottlieb, D. I. (1996). Retinoic Carr, V. M., and Farbman, A. I. (1992). Ablation of the olfactory
acid promotes neural and represses mesodermal gene expres- bulb up-regulates the rate of neurogenesis and induces preco-
sion in mouse embryonic stem cells in culture. Biochem. cious cell death in olfactory epithelium. Exp. Neurol.
Biophys. Res. Comm. 223:691–694. 115:55–59.
Baker, H., Kawano, T., Albert, V., Joh, T. H., Reis, D. J., and Chiasson, B. J., Tropepe, V., Morshead, C. M., and van der Kooy,
Margolis, F. L. (1984). Olfactory bulb dopamine neurons D. (1999). Adult mammalian forebrain ependymal and
survive deafferentation-induced loss of tyrosine hydroxylase. subependymal cells demonstrate proliferative potential, but
Neuroscience 11:605–615. only subependymal cells have neural stem cell characteristics.
Baker, H., Morel, K., Stone, D. M., and Maruniak, J. A. (1993). J. Neurosci. 19:4462–4471.
Adult naris closure profoundly reduces tyrosine hydroxylase Clarke, D. L., Johansson, C. B., Wilbertz, J., Veress, B., Nilsson,
expression in mouse olfactory bulb. Brain Res. 614:109–116. E., Karlstrom, H., Lendahl, U., and Frisen, J. (2000).
Barlow, H. B. (1975). Visual experience and cortical develop- Generalized potential of adult neural stem cells. Science 288:
ment. Nature 258:199–204. 1660–1663.
Beard, M. D. and Mackay-Sim, A. (1987). Loss of sense of smell Clos, J., and Legrand, C. (1990). An interaction between thyroid
in adult, hypothyroid mice. Brain Res. 433:181–189. hormone and nerve growth factor promotes the development
Benson, T. E., Ryugo, D. K., and Hinds, J. W. (1984). Effects of of hippocampus, olfactory bulbs and cerebellum: a compara-
sensory deprivation on the developing mouse olfactory sys- tive biochemical study of normal and hypothyroid rats.
tem: a light and electron microscopic, morphometric analysis. Growth Factors 3:205–220.
J. Neurosci. 4:638–653. Corotto, F. S., Henegar, J. A., and Maruniak, J. A. (1993).
Bjornson, C. R., Rietze, R. L., Reynolds, B. A., Magli, M. C., and Neurogenesis persists in the subependymal layer of the adult
Vescovi, A.L. (1999). Turning brain into blood: a hematopoi- mouse brain. Neurosci. Lett. 149:111–114.
etic fate adopted by adult neural stem cells in vivo. Science Corotto, F. S., Henegar, J. R., and Maruniak, J. A. (1994). Odor
283:534–537. deprivation leads to reduced neurogenesis and reduced neu-
Bradley, A., Evans, M., Kaufman, M. H., and Robertson, E. ronal survival in the olfactory bulb of the adult mouse.
(1984). Formation of germ-line chimaeras from embryo- Neuroscience 61:739–744.
derived teratocarcinoma cell lines. Nature 309:255–256. Costanzo, R. M. (1985). Neural regeneration and functional
Brosvic, G. M., Risser, J. M., Mackay-Sim, A., and Doty, R. L. reconnection following olfactory nerve transection in hamster.
(1996). Odor detection performance in hypothyroid and Brain Res. 361:258–266.
euthyroid rats. Physiol. Behav. 59:117–121. Craig, C. G., Tropepe, V., Morshead, C. M., Reynolds, B. A.,
Brown, J. L., and Brunjes, P. C. (1990). Development of the ante- Weiss, S., and van der Kooy, D. (1996). In vivo growth factor
rior olfactory nucleus in normal and unilaterally odor expansion of endogenous subependymal neural precursor cell
deprived rats. J. Comp. Neurol. 301:15–22. populations in the adult mouse brain. J. Neurosci. 16:
Brunjes, P. (1992). Lessons from lesions: the effects of olfactory 2649–2658.
bulbectomy. Chem. Senses 17:729–763. Croul-Ottman, C. E., and Brunjes, P. C. (1988). NADPH
Brunjes, P. C., and Borror, M. J. (1983). Unilateral odor depriva- diaphorase staining within the developing olfactory bulbs of
tion: differential effects due to time of treatment. Brain Res. normal and unilaterally odor-deprived rats. Brain Res.
Bull. 11:501–503. 460:323–328.
Brunjes, P. C., and Frazier, L. L. (1986). Maturation and plastic- Cullinan, W. E., and Brunjes, P. C. (1987). Unilateral odor depri-
ity in the olfactory system of vertebrates. Brain Res. vation: effects on the development of staining for olfactory
396:1–45. bulb succinate dehydrogenase. Brain Res. 432:35–42.
Brustle, O., Choudhary, K., Karram, K., Huttner, A., Murray, K., Cummings, D. M., Henning, H. E., and Brunjes, P .C. (1997).
Dubois-Dalcq, M., and McKay, R. D. (1998). Chimeric brains Olfactory bulb recovery after early sensory deprivation. J.
generated by intraventricular transplantation of fetal human Neurosci. 17:7433–7440.
brain cells into embryonic rats. Nat. Biotech. 16:1040–1044. Cummings, D. M., Emge, D. K., Small, S. L., and Margolis, F. L.
Buckland, M. E., and Cunningham, A. M. (1999). Alterations in (2000). Pattern of olfactory bulb innervation returns after
expression of the neurotrophic factors glial cell line-derived recovery from reversible peripheral deafferentation. J. Comp.
neurotrophic factor, ciliary neurotrophic factor and brain- Neurol. 421:362–373.
derived neurotrophic factor, in the target deprived olfactory Doty, R. L. (1979). A review of olfactory dysfunctions in man.
neuroepithelium. Neuroscience 90:333–347. Am. J. Otolaryngol. 1:57–79.
624 Maruniak

Doty, R. L., Snyder, P., Huggins, G., and Lowry, L. D. (1981). Gould, E., Tanapat, P., Rydel, T., and Hastings, N. (2000).
Endocrine, cardiovascular, and psychological correlate of Regulation of hippocampal neurogenesis in adulthood. Biol.
olfactory sensitivity changes during the human menstrual Psychiatry 48:715–720.
cycle. J. Comp. Physiol. Psychol. 95:45–60. Gozzo, S., and Fulop, Z. (1984). Transneuronal degeneration in
Doty, R. L., and Ferguson-Segall, M. (1989). Influence of castra- different inbred strains of mice: a preliminary study of olfac-
tion on the odor detection performance of male rats. Behav. tory bulb events after olfactory nerve lesion. Int. J. Neurosci.
Neurosci. 103:691–693. 23:187–194.
Doty, R. L., Yousem, D. M., Pham, L. T., Kreshak, A. A., & Lee, Greer, C. A., and Shepherd, G. M. (1982). Mitral cell degenera-
W.W. (1997). Olfactory dysfunction in patients with head tion and sensory function in the neurological mutant mouse
trauma. Arch. Neurol. 54:1131–1140. Purkinje cell degeneration (PCD). Brain Res. 235:156–161.
Douek, E., Bannister, L. H., and Dodson, H. C. (1975). Recent Guthrie, K. M., Pullara, J. M., Marshall, J. F., and Leon, M.
advances in the pathology of olfaction. Proc. R. Soc. Med. 68: (1991). Olfactory deprivation increases dopamine D2 receptor
467–470. density in the rat olfactory bulb. Synapse 8:61–70.
Dussault, J. H., and Ruel, J. (1987). Thyroid hormones and brain Guthrie, K. M., Anderson, A. J., Leon, M., and Gall, C. (1993).
development. Ann. Rev. Physiol. 49:321–334. Odor-induced increases in c-fos mRNA expression reveal an
Edwards, D. A. (1974). Non-sensory involvement of the olfactory anatomical “unit” for odor processing in olfactory bulb. Proc.
bulbs in the mediation of social behaviors. Behav. Biol. 11: Natl. Acad. Sci. 90:3329–3333.
287–302. Hall, R. D., and Macrides, F. (1983). Olfactory bulbectomy impairs
Evans, M. J., and Kaufman, M. H. (1981). Establishment in cul- the rat’s radial maze behavior. Physiol. Behav. 30: 797–803.
ture of pluripotential cells from mouse embryos. Nature Harding, J. W., and Wright, J. W. (1979). Reversible effects of
292:154–156. olfactory nerve section on behavior and biochemistry in mice.
Finley, M. F., Devata, S., and Huettner, J. E. (1999). BMP-4 Brain Res. Bull. 4:17–22.
inhibits neural differentiation of murine embryonic stem cells. Heimer, L., and Kalil, R. (1978). Rapid transneuronal degenera-
J. Neurobiol. 40:271–287. tion and death of cortical neurons following removal of
Flax, J. D., Aurora, S., Yang, C., Simonin, C., Wills, A. M., the olfactory bulb in adult rats. J. Comp. Neurol. 178:559–609.
Billinghurst, L. L., Jendoubi, M., Sidman, R. L., Wolfe, J. H., Henegar, J. R., and Maruniak, J. A. (1991). Quantification of the
Kim, S. U., and Snyder, E. Y. (1998). Engraftable human effects of long-term unilateral naris closure on the olfactory
neural stem cells respond to developmental cues, replace bulbs of adult mice. Brain Res. 568:230–234.
neurons, and express foreign genes. Nat. Biotech. 16: Herzog, C., and Otto, T. (1999). Regeneration of olfactory recep-
1033–1039. tor neurons following chemical lesion: time course and
Friedman, B., and Price, J. L. (1986). Age-dependent cell death enhancement with growth factor administration. Brain Res.
in the olfactory cortex: lack of transneuronal degeneration in 849:155–161.
neonates. J. Comp. Neurol. 246:20–31. Holmes, P.V., Davis, R. C., Masini, C. V., and Primeaux, S. D.
Gage, F. H., Ray, J., and Fisher, L. J. (1995). Isolation, character- (1998). Effects of olfactory bulbectomy on neuropeptide gene
ization, and use of stem cells from the CNS. Ann. Rev. expression in the rat olfactory/limbic system. Neuroscience
Neurosci. 18:159–192. 86:587–596.
Gage, F. H., Kempermann, G., Palmer, T. D., Peterson, D. A., and Huard, J. M., Youngentob, S. L., Goldstein, B. J., Luskin, M. B.,
Ray, J. (1998). Multipotent progenitor cells in the adult den- and Schwob, J.E. (1998). Adult olfactory epithelium contains
tate gyrus. J. Neurobiol. 36:249–266. multipotent progenitors that give rise to neurons and non-
Garcia-Verdugo, J. M., Doetsch, F., Wichterle, H., Lim, D. A., neural cells. J. Comp. Neurol. 400:469–486.
and Alvarez-Buylla, A. (1998). Architecture and cell types of Johansson, C. B., Svensson, M., Wallstedt, L., Janson, A. M., and
the adult subventricular zone: in search of the stem cells. J. Frisen, J. (1999a). Neural stem cells in the adult human brain.
Neurobiol. 36:234–248. Exp. Cell Res. 253:733–736.
Ghosh, A., Carnahan, J., and Greenberg, M. E. (1994). Johansson, C. B., Momma, S., Clarke, D. L., Risling, M.,
Requirement for BDNF in activity-dependent survival of cor- Lendahl, U., and Frisen, J. (1999b). Identification of a neural
tical neurons. Science 263:1618–1623. stem cell in the adult mammalian central nervous system. Cell
Giordano, T., Pan, J. B., Casuto, D., Watanabe, S., and Arneric, 96:25–34.
S.P. (1992). Thyroid hormone regulation of NGF, NT-3 and Katagiri, Y., Takeda, K., Yu, Z. X., Ferrans, V. J., Ozato, K., and
BDNF RNA in the adult rat brain. Mol. Brain Res. Guroff, G. (2000). Modulation of retinoid signalling through
16:239–245. NGF-induced nuclear export of NGFI-B. Nat. Cell Biol. 2:
Goldstein, B. J., Fang, H., Youngentob, S. L., and Schwob, J. E. 435–440.
(1998). Transplantation of multipotent progenitors from the Kawano, T., and Margolis, F. L. (1982). Transsynaptic regulation
adult olfactory epithelium. Neuroreport 9:1611–1617. of olfactory bulb catecholamines in mice and rats. J.
Gossler, A., Joyner, A. L., Rossant, J., and Skarnes, W. C. (1989). Neurochem. 39:342–348.
Mouse embryonic stem cells and reporter constructs to detect Key, B., Treloar, H. B., Wangerek, L., Ford, M. D. and
developmentally regulated genes. Science 244:463–465. Nurcombe, V. (1996). Expression and localization of FGF-1
Plasticity Within the Olfactory Pathways 625

in the developing rat olfactory system. J. Comp. Neurol. 366: Meisami, E. (1976). Effects of olfactory deprivation on postnatal
197–206. growth of the rat olfactory bulb utilizing a new method for
Korol, D. L., and Brunjes, P. C. (1990). Rapid changes in 2- production of neonatal unilateral anosmia. Brain Res. 107:
deoxyglucose uptake and amino acid incorporation following 437–444.
unilateral odor deprivation: a laminar analysis. Dev. Brain Meisami, E. (1984). Complete recovery of growth deficits after
Res. 52:75–84. reversal of PTU induced postnatal hypothyroidism in the female
Korol, D. L., and Brunjes, P.C. (1992). Unilateral naris closure rat: a model for catch-up growth. Life Sci. 34:1487–1496.
and vascular development in the rat olfactory bulb. Meisami, E., and Manoochehri, S. (1977). Effects of early bilateral
Neuroscience 46:631–641. chemical destruction of olfactory receptors on postnatal growth,
Kosaka, T., Kosaka, K., Hama, K., Wu, J. Y. and Nagatsu, I. Mg-ATPase and Na K-ATPase activity of olfactory and non-
(1987). Differential effect of functional olfactory deprivation olfactory structures of the rat brain. Brain Res. 128: 170–175.
on the GABAergic and catecholaminergic traits in the rat Meyer-Franke, A., Kaplan, M. R., Pfrieger, F. W., and
main olfactory bulb. Brain Res. 413:197–203. Barres, B. A. (1995). Characterization of the signaling inter-
Kuhn, H. G., Winkler, J., Kempermann, G., Thal, L. J., and Gage, actions that promote the survival and growth of developing
F.H. (1997). Epidermal growth factor and fibroblast growth retinal ganglion cells in culture. Neuron 15:805–819.
factor-2 have different effects on neural progenitors in the Miro, J. L., Canguilhem, B., Schmitt, P., and Koch, A. (1982).
adult rat brain. J. Neurosci. 17:5820–5829. Hyperphagia and obesity after olfactory bulbectomy per-
Kukekov, V. G., Laywell, E. D., Suslov, O., Davies, K., Scheffler, formed at different times of the year in the European hamster.
B., Thomas, L. B., O’Brien, T. F., Kusakabe, M., and Physiol. Behav. 29:681–685.
Steindler, D.A. (1999). Multipotent stem/progenitor cells with Monti Graziadei, G. A., Karlan, M. S., Bernstein, J. J., and
similar properties arise from two neurogenic regions of adult Graziadei, P. P. (1980). Reinnervation of the olfactory bulb
human brain. Exp. Neurol. 156:333–344. after section of the olfactory nerve in monkey (Saimiri sci-
Lewis, R. (2000). A paradigm shift in stem cell research? ureus). Brain Res. 189:343–354.
Scientist 14:1–3. Morshead, C. M., and van der Kooy, D. (1992). Postmitotic death
Linnarsson, S., Willson, C. A., and Ernfors, P. (2000). Cell death is the fate of constitutively proliferating cells in the
in regenerating populations of neurons in BDNF mutant mice. subependymal layer of the adult mouse brain. J. Neurosci.
Mol. Brain Res. 75:61–69. 12:249–256.
Liu, N., and Baker, H. (1999). Activity-dependent Nurrl and Morshead, C. M., Craig, C. G., and van der Kooy, D. (1998). In
NGFI-B gene expression in adult mouse olfactory bulb. vivo clonal analyses reveal the properties of endogenous
Neuroreport 10:747–751. neural stem cell proliferation in the adult mammalian fore-
Liu, S., Qu, Y., Stewart, T. J., Howard, M. J., Chakrabortty, S., brain. Development 125:2251–2261.
Holekamp, T.F., and McDonald, J.W. (2000). Embryonic stem Moulton, D. G. (1974). Dynamics of cell populations in the olfac-
cells differentiate into oligodendrocytes and myelinate in cul- tory epithelium. Ann. NY Acad. Sci. 237:52–61.
ture and after spinal cord transplantation. Proc. Natl. Acad. Nadi, N. S., Head, R., Grillo, M., Hempstead, J., Grannot-
Sci. 97:6126–6131. Reisfeld, N., and Margolis, F. L. (1981). Chemical deaf-
Mackay-Sim, A., and Beard, M.D. (1987). Hypothyroidism dis- ferentation of the olfactory bulb: plasticity of the levels of
rupts neural development in the olfactory epithelium of adult tyrosine hydroxylase, dopamine and norepinephrine. Brain
mice. Brain Res. 433:190–198. Res. 213:365–377.
Margolis, F. L., Roberts, N., Ferriero, D., and Feldman, J. Nagahara, Y. (1940). Experimentelle Studien über die histologis-
(1974). Denervation in the primary olfactory pathway of chen Veränderungen des Geruchsorgans nach der
mice: biochemical and morphological effects. Brain Res. 81: Olfactorisdurchschneidung. Beiträge zur Kenntnis des
469–483. feineren Baus des Geruchsorgans. Jpn. J. Med. Sci. V Pathol.
Martin, G. R. (1981). Isolation of a pluripotent cell line from 5:165–199.
early mouse embryos cultured in medium conditioned by ter- Nakashima, T., Kimmelman, C. P. and Snow, J. (1984). Structure
atocarcinoma stem cells. Proc. Natl. Acad. Sci. 78: of human fetal and adult olfactory neuroepithelium. Arch.
7634–7638. Otolaryngol. 110:641–664.
Maruniak, J. A., Taylor, J. A., Henegar, J. R., and Williams, M. B. Nelson, R. J., and Zucker, I. (1981). Photoperiodic control of repro-
(1989). Unilateral naris closure in adult mice: atrophy of the duction in olfactory-bulbectomized rats. Neuroendocrinology
deprived-side olfactory bulbs. Dev. Brain Res. 47:27–33. 32:266–271.
Matulionis, D. H., Breipohl, W., and Bhatnagar, K.P. (1982). Onoda, N. (1992). Odor-induced fos-like immunoreactivity in the
Degeneration and regeneration of olfactory epithelium in the rat olfactory bulb. Neurosci. Lett. 137:157–60.
mouse. A scanning electron microscopic study. Ann. Otol. Pagano, S. F., Impagnatiello, F., Girelli, M., Cova, L., Grioni, E.,
Rhinol. Laryngol. Suppl. 89:1–12. Onofri, M., Cavallaro, M., Etteri, S., Vitello, F., Giombini, S.,
McConnell, R. J., Menendez, C. E., Smith, F. R., Henkin, R. I., Solero, C.L., and Parati, E.A. (2000). Isolation and character-
and Rivlin, R. S. (1975). Defects of taste and smell in patients ization of neural stem cells from the adult human olfactory
with hypothyroidism. Am. J. Med. 59:354–364. bulb. Stem Cells 18:295–300.
626 Maruniak

Pasterkamp, R. J., De Winter, F., Holtmaat, A. J., and Verhaagen, nections in developing and regenerating olfactory receptor
J. (1998). Evidence for a role of the chemorepellent sema- neurons. Neuron 13:289–299.
phorin III and its receptor neuropilin-1 in the regeneration of Roskams, A. J., Bethel, M. A., Hurt, K. J., and Ronnett, G. V.
primary olfactory axons. J. Neurosci. 18:9962–9976. (1996). Sequential expression of Trks A, B, and C in the regen-
Paternostro, M. A., and Meisami, E. (1991). Lack of thyroid hor- erating olfactory neuroepithelium. J. Neurosci. 16:1294–1307.
mones but not their excess affects the maturation of olfactory Sallaz, M., and Jourdan, F. (1993). c-fos expression and
receptor neurons: a quantitative morphologic study in the 2-deoxyglucose uptake in the olfactory bulb of odour-stimu-
postnatal rat. Int. J. Dev. Neurosci. 9:439–452. lated awake rats. Neuroreport 4:55–58.
Paternostro, M. A., and Meisami, E. (1993). Developmental Samanen, D. W., and Forbes, W. B. (1984). Replication and dif-
plasticity of the rat olfactory receptor sheet as shown by com- ferentiation of olfactory receptor neurons following axotomy
plete recovery of surface area and cell number from extensive in the adult hamster: a morphometric analysis of postnatal
early hypothyroid growth retardation. Dev. Brain Res. neurogenesis. J. Comp. Neurol. 225:201–211.
76:151–161. Schultz, E. W. (1941). Regeneration of olfactory cells. Proc. Soc.
Paternostro, M. A., and Meisami, E. (1994). Quantitative Exp. Biol. Med. 46:41–43.
[3H]thymidine autoradiography of neurogenesis in the olfac- Schultz, E. W. (1960). Repair of the olfactory mucosa. Am. J.
tory epithelium of developing normal, hypothyroid and Pathol. 37:1–19.
hypothyroid-rehabilitated rats. Dev. Brain Res. 83:151–162. Schultz, E.W., and Gebhardt, S. P. (1942). Studies on chemical pro-
Paternostro, M. A., and Meisami, E. (1996). Essential role of thy- phylaxis of experimental poliomyelitis. J. Infect. Dis. 70: 7–50.
roid hormones in maturation of olfactory receptor neurons: an Schwartz Levey, M., Chikaraishi, D. M., and Kauer, J. S. (1991).
immunocytochemical study of number and cytoarchitecture Characterization of potential precursor populations in the
of OMP-positive cells in developing rats. Int. J. Dev. mouse olfactory epithelium using immunocytochemistry and
Neurosci. 14:867–880. autoradiography. J. Neurosci. 11:3556–3564.
Petrie, R. X., Reid, I. C., and Stewart, C. A. (2000). The N- Schwob, J. E., Szumowski, K. E., and Stasky, A. A. (1992).
methyl-D-aspartate receptor, synaptic plasticity, and depres- Olfactory sensory neurons are trophically dependent on the
sive disorder. A critical review. Pharmacol. Ther. 87:11–25. olfactory bulb for their prolonged survival. J. Neurosci.
Philpot, B. D., Men, D., McCarty, R., and Brunjes, P.C. (1998). 12:3896–3919.
Activity-dependent regulation of dopamine content in the Schwob, J. E., Youngentob, S. L., Ring, G., Iwema, C. L., and
olfactory bulbs of naris-occluded rats. Neuroscience Mezza, R. C. (1999). Reinnervation of the rat olfactory bulb
85:969–977. after methyl bromide-induced lesion: timing and extent of
Pieper, D. R., Newman, S. W., Lobocki, C. A., and Gogola, G. reinnervation. J. Comp. Neurol. 412:439–457.
(1989). Bilateral transection of the lateral olfactory tract but Smart, I. (1961). The subependymal layer of the mouse brain and
not removal of the vomeronasal organs inhibits short-pho- its cell production as shown by autoradiography after
toperiod-induced testicular regression in golden hamsters. 3H-thymidine injection. J. Comp. Neurol. 116:325–347.
Brain Res. 485:382–390. Steckler, T., Holsboer, F, and Reul, J. M. (1999). Glucocorticoids
Pinching, A. J., and Powell, T. P. (1971). Ultrastructural features and depression. Baillieres Best Pract. Res. Clin. Endocrinol.
of transneuronal cell degeneration in the olfactory system. J. Metab. 13:597–614.
Cell Sci. 8:253–287. Stevens, D. A., and O’Connell, R. J. (1995). Enhanced sensitivity
Pixley, S. K., Dangoria, N. S., Odoms, K. K., and Hastings, L. to androstenone following regular exposure to pernenone.
(1998). Effects of insulin-like growth factor 1 on olfactory Chem. Senses 20:413–419.
neurogenesis in vivo and in vitro. Ann. NY Acad. Sci. Stone, D. M., Wessel, T., Joh, T. H., and Baker, H. (1990).
855:244–247. Decrease in tyrosine hydroxylase, but not aromatic L-amino
Postolache, T. T., Doty, R. L., Wehr, T. A., Jimma, L. A., Han, L., acid decarboxylase, messenger RNA in rat olfactory bulb fol-
Turner, E. H., Matthews, J. R., Neumeister, A., No, C., lowing neonatal, unilateral odor deprivation. Brain Res Mol.
Kroger, H., Bruder, G. E., and Rosenthal, N.E. (1999). Brain Res. 8:291–300.
Monorhinal odor identification and depression scores in Stone, D.M., Grillo, M., Margolis, F. L., Joh, T.H., and Baker, H.
patients with seasonal affective disorder. J. Affect. Disord. 56: (1991). Differential effect of functional olfactory bulb deaf-
27–35. ferentation on tyrosine hydroxylase and glutamic acid decar-
Rakic, P. (1985). Limits of neurogenesis in primates. Science boxylase messenger RNA levels in rodent juxtaglomerular
227:1054–1056. neurons. J. Comp. Neurol. 311:223–233.
Reynolds, B. A., and Weiss, S. (1992). Generation of neurons and Suhonen, J. O., Peterson, D. A., Ray, J., and Gage, F.H. (1996).
astrocytes from isolated cells of the adult mammalian central Differentiation of adult hippocampus-derived progenitors into
nervous system. Science 255:1707–1710. olfactory neurons in vivo. Nature 383:624–627.
Richards, L. J., Kilpatrick, T. J., and Bartlett, P.F. (1992). De Takagi, S. (1969). Degeneration and regeneration of the sensory
novo generation of neuronal cells from the adult mouse brain. neuron: studies on the olfactory epithelium. Shinkei Kenkyu
Proc. Natl. Acad. Sci. 89:8591–8595. No Shimpo 13:152–163.
Roskams, A. J., Bredt, D. S., Dawson, T.M., and Ronnett, G.V. Thoenen, H. (1995). Neurotrophins and neuronal plasticity.
(1994). Nitric oxide mediates the formation of synaptic con- Science 270:593–598.
Plasticity Within the Olfactory Pathways 627

Turner, C.P., and Perez-Polo, J.R. (1994). Changes in expression ostensibly anosmic people. Proc. Nat. Acad. Sci. USA
of the low affinity receptor for neurotrophins, p75NGFR, in 86:7976–7978.
the regenerating olfactory system. Int. J. Dev. Neurosci. Yandava, B. D., Billinghurst, L. L., and Snyder, E. Y. (1999).
12:767–773. “Global” cell replacement is feasible via neural stem cell
van Riezen, H., Schnieden, H., and Wren, A.F. (1977). Olfactory transplantation: evidence from the dysmyelinated shiverer
bulb ablation in the rat: behavioural changes and their rever- mouse brain. Proc. Natl. Acad. Sci. 96:7029–7034.
sal by antidepressant drugs. Br. J. Pharmacol. 60:521–528. Yee, K. K., and Costanzo, R.M. (1995). Restoration of olfactory
Wang, H. W., Wysocki, C. J., and Gold, G. H. (1993). Induction mediated behavior after olfactory bulb deafferentation.
of olfactory receptor sensitivity in mice. Science Physiol. Behav. 58:959–968.
260:998–1000. Yee, K. K., and Wysocki, C. J. (2001). Odorant exposure
Webster, H. H., Flores, G., Marcotte, E. R., Cecyre, D., Quirion, increases olfactory sensitivity: olfactory epithelium is impli-
R., and Srivastava, L.K. (2000). Olfactory bulbectomy alters cated. Physiol. Behav. 72:705–711.
NMDA receptor levels in the rat prefrontal cortex. Synapse Zhou, D., Grecksch, G., Becker, A., Frank, C., Pilz, J., and
37:159–162. Huether, G. (1998). Serotonergic hyperinnervation of the
Wilson, D. A., and Wood, J. G. (1992). Functional consequences frontal cortex in an animal model of depression, the bulbec-
of unilateral olfactory deprivation: time-course and age sensi- tomized rat. J. Neurosci. Res. 54:109–116.
tivity. Neuroscience 49:183–192. Zigova, T., Betarbet, R., Soteres, B. J., Brock, S., Bakay, R. A.,
Wilson, D. A., Best, A. R., and Brunjes, P. C. (2000). Trans-neu- and Luskin, M.B. (1996). A comparison of the patterns of
ronal modification of anterior piriform cortical circuitry in the migration and the destinations of homotopically transplanted
rat. Brain Res. 853:317–322. neonatal subventricular zone cells and heterotopically trans-
Woo, C. C., and Leon, M. (1995). Early olfactory enrichment planted telencephalic ventricular zone cells. Dev. Biol.
and deprivation both decrease beta-adrenergic receptor density 173:459–474.
in the main olfactory bulb of the rat. J. Comp. Neurol. 360: Zigova, T., Pencea, V., Wiegand, S. J., and Luskin, M. B. (1998).
634–642. Intraventricular administration of BDNF increases the num-
Wysocki, C.J., Dorries, K.M., and Beauchamp, G. K. (1989). ber of newly generated neurons in the adult olfactory bulb.
Ability to perceive androstenone can be acquired by Mol. Cell. Neurosci. 11:234–245.
30

Head Injury and Olfaction

Richard M. Costanzo, Laurence J. DiNardo, and Evan R. Reiter

Virginia Commonwealth University, Richmond, Virginia, U.S.A.

I. INTRODUCTION and taste centers for evaluation and treatment of their

chemosensory dysfunction (Cain, 1989; Ikeda et al., 1999;
The first published report of post-traumatic anosmia Kondo et al., 1998). Standardized tests have played an
appeared in volume 1 of the London Hospital Report important role in the development of new data on the epi-
(Jackson, 1864). In this article, John Hughlings Jackson demiology and incidence of this sensory disorder (Deems
describes the case of a 50-year-old patient who lost his et al., 1991). For example, Doty and colleagues used the
sense of smell after being knocked off his horse by a high- University of Pennsylvania Smell Identification Test
wayman. The patient, who received a blow to the head, (UPSIT) to measure changes in olfactory function in
suffered a severe concussion and afterwards never regained 66 patients with head trauma over periods ranging from
his sense of smell. Subsequent reports of post-traumatic 1 month to 13 years (Doty et al., 1997). They found that
anosmia in the late 1800s also involved blows to the head 36.6% improved, 18% worsened, and 45% had no change.
or falls from horses or horse-drawn vehicles (Ferrier, 1876; However, UPSIT testing revealed that only 3 patients (5%
Legg, 1873; Notta, 1870; Ogle, 1870; Rotch, 1878). of the study sample) regained normal olfactory function on
Costanzo and Zasler (1991) noted that approximately retesting. In addition, this study demonstrated that among
one of every 400 people in the United States suffers from those patients who believed they had improved, less than
the sequelae of head injury. Most of these injuries occur half showed a statistical improvement in their UPSIT
among young adult males between the ages of 15 and 24 scrore.
years. Although the major cause of head injury among this Several studies suggest that the likelihood of anosmia is
age group is vehicular accidents, in older populations falls dependent on the severity of the injury (Costanzo and
and assaults play a prominent role as well. Early studies of Becker, 1986; Doty et al., 1997; Sumner, 1975). Heywood
posttraumatic olfactory loss suggest a 4–7% incidence of et al. (1990) matched olfactory test scores with patients’
anosmia following head injury (Kraus et al., 1986; Leigh, Glasgow coma scale (GCS) ratings and found a correlation
1943; Mifka, 1964; Zusho, 1982). However, more recent between the severity of head injury and the amount of
studies report a higher incidence of anosmia—reaching olfactory disturbance. In mild injury (GCS 13–15), 13% of
60% in cases of severe head injury (Costanzo et al., 1987; patients were totally anosmic and 27% showed difficulty
Deems et al., 1991; Doty et al., 1997; Ogawa and Rutka, with odorant detection or identification. In the moderate
1999). This might be accounted for in part by increased head injury group (GCS 9–12), 11% were anosmic while
awareness of olfactory loss in head trauma, the use of stan- 67% showed some degree of olfactory impairment. Among
dardized tests for the quantitative measurement of olfac- those with severe head injury (GCS 3–8), 25% were anos-
tory function (Cain et al., 1983), or a selection bias mic and 67% had some degree of impairment. Partial or
introduced when studying patients who present to smell complete unilateral loss of olfactory function can also

629
630 Costanzo et al.

occur and is more likely to go unnoticed than complete cleft with direct injury to the olfactory neuroepithelium, (2)
bilateral anosmia. scarring with synechia formation or fractures of the nasal
Parosmia, abnormal odor sensations or the presence of a skeleton or septum with subsequent airflow alterations that
strange odor in the absence of a stimulus, has been reported prevent odorants from reaching the olfactory cleft (Fig. 1A),
in 25–33% of patients with head trauma (Doty et al., 1997; or (3) posttraumatic rhinosinusitis with resultant alterations
Duncan and Seiden, 1995). Although the mechanisms in nasal airflow or mucus quantity and viscosity. Although
underlying parosmia are not fully understood, it has been this cause of hyposmia is not common, it is important to rec-
shown that the number of patients experiencing parosmia ognize it since it is potentially treatable. Often reduction of
decreases with time following injury. This improvement in mucosal edema, repair of nasal or septal fractures with relief
function may reflect a true resolution of olfactory function of airway obstruction, or the treatment of sinusitis can
or an adaptation process in which the patient learns to improve olfaction. Resolution of olfactory deficits is more
ignore or tune out these abnormal olfactory sensations. likely to occur when there has been no direct trauma and,
consequently, no irreversible scarring or damage to the
olfactory cleft.
II. MECHANISMS OF INJURY

Traumatic olfactory dysfunction may be caused by several B. Shear Injuries

mechanisms: (1) sinonasal contusions or fractures with or
without direct damage to the olfactory apparatus, (2) tear- The delicate axons of olfactory receptor cells pass
ing or shearing of olfactory nerve filaments, or (3) contu- through small foramina of the cribriform plate at the
sion or hemorrhage within the olfactory-related brain base of the skull and synapse directly with cortical cells
regions (Fig. 1). in the olfactory bulb. Tearing or shearing of these axons
may occur with fractures of the naso-orbito-ethmoid
A. Sinonasal Tract Alterations region involving the cribriform plate or with rapid trans-
lational shifts in the brain secondary to coup or contra-
Zusho (1982) studied a group of 212 patients with posttrau- coup forces generated by blunt head trauma (Fig. 1B).
matic anosmia who came for follow-up evaluation of head Fracture of the cribriform plate is actually an uncommon
and facial injuries. He found that 44.8% had facial or skull cause of anosmia.
fractures and 11.3% had facial contusions with fractures of Anosmia is more likely to be produced by posterior-
the nasal bone. Following sinonasal tract alterations, unilat- anterior coup and contra-coup forces (Sumner, 1975;
eral olfactory loss may be present and hyposmia, rather than Zusho, 1982). This is due to a shearing effect on the deli-
anosmia, often results. Potential mechanisms include (1) cate olfactory axon fibers as the brain shifts with respect to
mucosal hematoma, edema, or avulsion within the olfactory the cranial base. This frequently results in complete bilat-

Figure 1 Mechanisms of posttraumatic olfactory dysfunction. (A) Injury to the sinonasal tract with obstruction of airflow to the olfac-
tory cleft. (B) Shearing of the olfactory nerves at the cribriform region. (C) Cortical contusions and brain hemorrhage involving the
olfactory cortex.
Head Injury and Olfaction 631

eral anosmia. It is of interest to note that nasal endoscopy reduction, nasal intubation, and administration of numer-
and the use of biopsy techniques have produced clinical ous medications that might also affect olfaction, thus com-
ultrastructural correlates of posttraumatic anosmia that plicating evaluation of patients’ olfactory complaints. A
support the notion of shearing of olfactory nerve fibers detailed history, therefore, is an important first step.
(Moran et al., 1985). Specifically, the fila olfactoria have The patient should be asked about pretraumatic olfac-
been found to be traumatically severed at the cribriform tory function, and any previous deficits, either transient or
plate. Regenerating axons apparently fail to reach the prolonged, should be explored in an effort to secure the eti-
olfactory bulb due to scar formation and obliteration of the ology. Consideration must be given to olfactory loss asso-
foramina in the cribriform plate. ciated with an upper respiratory tract infection. Even
without a history of previous impairment, common causes
C. Brain Contusion and Hemorrhage of olfactory disturbance such as rhinosinusitis, previous
head trauma, or nasal obstruction should be ruled out.
Head trauma often results in traumatic brain injury in the The patient should be asked about present olfactory
form of cortical contusion or intraparenchymal hemor- function. The severity and nature of the olfactory dysfunc-
rhage (Fig. 1C). Contusion of the olfactory bulbs is one tion should be documented. Factors such as the complete-
plausible explanation for posttraumatic anosmia. Damage ness of loss, uni- or bilaterality of deficits, and the
localized to the olfactory bulbs or hemorrhage in the vicin- presence of qualitative changes in olfaction (e.g., alter-
ity can occur without involvement of other brain regions ations in known smells, or dysosmias, or olfactory halluci-
(Costanzo and Zasler, 1992). This can lead to direct neu- nations, or phantosmias) should be ascertained. This helps
ronal injury, or secondary ischemic insult. Inamitsu and the physician estimate the potential impact of the olfactory
coworkers (1990) demonstrated retrograde degeneration of loss on the patient. The time course of the olfactory distur-
the olfactory neuroepithelium in rats after surgical ablation bance yields important information about its etiology. As
of the olfactory bulbs. Despite subsequent neuronal regen- discussed above, most posttraumatic olfactory losses are
eration, the number of mature neural elements identified at the result of shear injury to the olfactory nerve fibers. Such
the olfactory neuroepithelium was diminished. injuries typically, but not always, result in immediate
Posttraumatic disorders of odor discrimination are often olfactory loss. The patient should be asked about clear rhi-
associated with cortical lesions (Costanzo and Zasler, norrhea or “salty-tasting” postnasal drip, especially with-
1991; Yee and Costanzo, 1998). Higher-order olfactory out an antecedent history of rhinosinusitis. This might be
neurons project mostly to the anterior pyriform cortex, indicative of a cerebrospinal fluid (CSF) leak resulting
amygdala, and temporal lobe region. Other projections ter- from fracture of the anterior cranial base, potentially
minate in areas of the frontal lobe. Projections are bilateral involving the cribriform. Posttraumatic CSF rhinorrhea
and extensively distributed (Yousem et al., 1996). does not necessarily indicate a breach in the cribriform
Therefore, it is difficult to attribute complete anosmia to plate, because fractures of the adjacent ethmoid sinus roof
cerebral damage alone. Levin and colleagues (1985) sug- are often the cause of dural disruptions (Raveh et al.,
gest that impaired olfactory recognition without anosmia 1988). Gradual, progressive losses are typical of inflam-
may result from local or diffuse injury to the orbitofrontal matory disease, as can result following sinonasal trauma.
and temporal lobe regions. It is consequently not surpris- Immediate but transient improvement with topical vaso-
ing that behavior disturbances and memory disorders constriction further supports an obstructive etiology.
frequently accompany impaired olfactory recognition. The direction and severity of head injury should be
ascertained. Zusho (1982) found that the direction of the
external force resulting in anosmia at the time of injury was
III. CLINICAL ASSESSMENT most often posterior-anterior. In a more recent study, Doty
and colleagues found that, on average, a single focused
A. History impact to the back of the head produced only slightly larger
deficits (39.3%, 66 of 168 cases studied) than an impact to
Because of the potentially life-threatening nature of head the front (36.3%, 61 of 168 cases studied) of the head (Doty
trauma and the frequent co-occurrence of other visceral or et al., 1997). Transverse, superior, and inferior forces were
orthopedic injuries requiring immediate medical attention, less common. The severity of head injury is also associated
both the patient and his or her treating physicians might with the incidence of posttraumatic anosmia. In particular,
not recognize posttraumatic olfactory loss until well after the Glasgow coma scale, or perhaps more accurately the
the time of injury. This time lag often encompasses treat- duration of posttraumatic amnesia, seems to correlate with
ments such as neurosurgical procedures, facial fracture the occurrence of posttraumatic anosmia (Levin et al.,
632 Costanzo et al.

1985; Sumner, 1975). It is important to note that anosmia B. Physical Examination

may result from even minor head injuries. The physician
should thus make every possible effort to reconstruct the The treatment of any life-threatening injuries obviously takes
mechanism of injury. If the injury resulted in loss of con- precedence in the management of head trauma. The details of
sciousness or amnesia and the patient is unable to provide evaluation and stabilization of the head-injured patient are
specific details, any available family members or witnesses complex, and will not be reviewed here. Subsequently, the
should be consulted (with the patient’s permission) to pro- head-injured patient rarely receives an immediate sensory
vide additional history. As the physician working up the evaluation, with the possible exception of ophthalmological
olfactory complaint is seldom seeing the patient at the time evaluation in cases where screening neurological examina-
of trauma, the emergency department or inpatient hospital tion reveals visual compromise. Frequently, either the patient
records should be obtained and carefully evaluated for does not recognize an olfactory deficit until some time after
clues. These clues might include mention of the location of injury, or evaluation of such deficits is deferred while other
head or facial lacerations or ecchymosis, the presence of treatment or rehabilitation issues are addressed. Whether
epistaxis or clear rhinorrhea, or coexisting cranial nerve occurring at the time of injury or some time later, evaluation
deficits. Results of any radiographic workup including of patients with olfactory deficits should include a thorough
plain films, magnetic resonance (MR), or computed tomog- examination by an otorhinolaryngologist. In addition to eval-
raphy (CT) scans of the head or maxillofacial area should uation for potential causes of olfactory impairment, the
be reviewed, and whenever possible the films themselves examination should not neglect auditory, vestibular, and
should be reviewed with a neuroradiologist. Subtle findings other cranial nerve testing, given previously discussed poten-
in the sinonasal and cribriform area might be overlooked in tial associations with olfactory loss. In addition, the
films obtained solely to evaluate intracranial injury. chemosensory function of the trigeminal nerve must be con-
Operative notes from any surgical procedures performed, in sidered separately when evaluating olfaction.
particular neurosurgical procedures or maxillofacial frac- The examining physician must keep in mind the com-
ture reduction, should be reviewed in detail. Lastly, both the mon pathophysiological mechanisms of posttraumatic
inpatient chart and the outpatient office notes from any olfactory loss. Signs of blunt or severe head trauma might
other treating physicians should be reviewed for mention of suggest generation of coup-contra-coup forces and a shear-
olfactory complaints, or denial of the same. This is cer- ing effect on the olfactory fibers. The presence of facial or
tainly important in determining the mechanism of a scalp lacerations, ecchymosis, edema, or tenderness are
patient’s olfactory loss, but may prove even more critical suggestive of such injuries. Direct injury or even fracture
when evaluating a patient involved in legal proceedings sur- of the cribriform area may be associated with midfacial or
rounding the traumatic event. Discrepancies in the record nasal fractures. Traumatic telecanthus, described as widen-
might suggest a patient is seeking to magnify their level of ing or lateral displacement of the medial canthus, is seen
olfactory impairment for personal gain. This would indicate with injuries to the naso-orbito-ethmoid complex that
a need for more thorough olfactory testing to document might also involve fracture of the cribriform plate and CSF
malingering. fistula.
Other sequelae of head injury should also be considered Nasal endoscopy is critical in the evaluation of post-
since they may functionally influence olfactory disorders. traumatic olfactory deficits. A 4.0 mm 30 degree rigid
Gustatory disturbance and cognitive dysfunction are par- endoscope is typically used, although examination of
ticularly important. Zusho (1982) observed a 16.5% inci- patients with significant septal deviation or other anatomi-
dence of gustatory complaints in posttraumatic anosmic cal narrowing may best be performed with a 2.4 mm
patients. Formal olfactory and gustatory testing is required endoscope. The intranasal examination is conducted both
to confirm this finding because taste and smell disturbance before and after topical nasal vasoconstriction with either
are easily confused by the patient. Mott and Leopold phenylephrine (0.25%) or oxymetazoline (0.025%) to
(1991) suggest obtaining a psychosocial history for all determine the degree of reversible mucosal edema.
patients with traumatic olfactory loss. Neuropsychological Routine evaluation should be systematic and include
testing should be considered since frontal lobe contusions inspection of the inferior, middle, and superior meati as
are frequently cited in posttraumatic anosmia and can well as the nasopharynx. Purulent discharge from the mid-
cause subtle changes in cognitive function (Zasler et al., dle or superior meati is indicative of rhinosinusitis.
1990). Associations between posttraumatic anosmia and Specific abnormalities potentially obstructing airflow,
hearing impairment (41%), tinnitus (22.6%), disequilib- such as hematoma, septal deviation, turbinate hypertro-
rium (14.2%), and visual disturbances (2.8%) have also phy, neoplasia, or polyposis, should be sought. Inspection
been described (Zusho, 1982). of the olfactory cleft itself is critical. Given the typically
Head injury and Olfaction 633

narrow dimensions of this space, which might be further cavities producing airflow obstruction, posttraumatic
compromised by deflections of the middle turbinate or sinusitis, and fractures of the cribriform plate are easily
septum, access can be challenging. This may be facilitated demonstrated without need for intravenous contrast. While
by use of the 2.4 mm endoscope or by gentle lateralization fractures of the anterior cranial base may suggest CSF leak,
of the middle turbinate after application of topical anes- the localization of a suspected leak may be achieved using
thesia. The injured olfactory cleft may manifest mucosal intrathecal metrizimide, a water-soluble contrast agent.
edema, hematoma or ecchymosis, mucosal lacerations, or Intracranial soft tissue injuries are best evaluated by MR
CSF leaks. scanning. Hematoma and contusion are possible intracra-
nial causes of posttraumatic anosmia and may be sought
C. Radiography using axial and coronal MRI (Fig. 2). We find the greatest
yield in identifying treatable causes of posttraumatic olfac-
High-resolution CT and MR scans are valuable adjuncts in tory loss can be achieved by thin-cut CT of the maxillofa-
the diagnosis of posttraumatic anosmia. Plain-film radio- cial region including the cribriform fossa. This obviates the
graphs may be of some value in the acute setting to assist in need for intravenous contrast and the greater expense of
the diagnosis of skull or facial fractures. However, plain MR imaging. MR is reserved for cases in which other neu-
films have limited ability to diagnose intracranial abnor- rological findings in addition to anosmia require workup.
malities or fractures of the delicate bones of the cribriform
fossa and have thus been supplanted by CT and MR in the D. Olfactory Function Testing
workup of posttraumatic olfactory loss. Modern imaging
technologies are most useful when properly chosen and All patients reporting olfactory deficits following head
performed (Yousem et al., 1996, 1999). CT scanning trauma should undergo olfactory testing to confirm and
should utilize thin cuts (1–2 mm) through the skull base in quantify the degree of loss (see Chapters 10 and 22). The
both axial and coronal planes. Abnormalities of the nasal development and standardization of olfactory screening
tests (OSTs) has improved identification of posttraumatic
olfactory losses considerably (Cain, 1989; Costanzo and
Zasler, 1991; Doty and Kobal, 1995). Before the use of
OSTs, it is likely that most olfactory disturbances result-
ing from head trauma went unrecognized. There are now
several standardized tests available for evaluation of
olfactory function. The UPSIT consists of four booklets
containing 10 “scratch-and-sniff ” microencapsulated
odorants each. The analysis of response data can be of use
in the detection of malingerers (Doty et al., 1984).
Another test developed at the University of Connecticut
employs both odor-detection and odor-identification sub-
tests (Cain et al., 1988). Detection threshold levels have
utility in evaluating receptor cell function, and identifica-
tion scores are useful in uncovering cortical contusions
and brain injury (Costanzo and Zasler, 1991). In Japan,
testing is often performed using a graded series of odorant
concentrations presented on strips of blotting paper to
deliver different concentrations of odor stimuli to subjects
(Ikeda et al., 1999; Takagi, 1987). Recently, a device that
delivers odorants into the nose, termed the “Jet Stream
Olfactometer,” has been introduced for evaluating smell
function (Ikeda et al., 1999). Using differing concentra-
tions of various odorants, both detection and recognition
thresholds may be determined. Although comprehensive
olfactory testing is routine at specialized smell and taste
Figure 2 Magnetic resonance image showing bilateral centers, this is not the case in most emergency rooms,
orbitofrontal contusions in a patient with bilateral anosmia neurosurgical wards, or head injury rehabilitation facili-
following head trauma. ties. However, screening tests such as the 3-item Pocket
634 Costanzo et al.

Smell TestTM and the 12-item Brief Smell Identification eration (Cain et al., 1987; Cone and Shusterman, 1991;
TestTM are available commercially (Sensonics, Inc., Hoffman et al., 1998). Both the medical and lay communi-
Haddon Heights, NJ) and are likely to be more suitable for ties have historically underestimated the impairment attrib-
emergency room use or bedside testing. Although olfac- utable to posttraumatic olfactory deficits. Olfaction has
tory function testing can be extremely valuable in the important safety functions such as the early detection of
assessment of the head-injured patient, it may be some fire, gas leaks, spoiled foods, or dangerous fumes, as well
time before testing kits gain widespread acceptance and as hedonic functions such as the assessment of the palata-
testing becomes routine. bility of foods and beverages and the detection of fra-
grances or aromas. While these functions are less vital to a
E. Treatment and Prognosis person’s well-being and functionality than vision or hear-
ing, their loss still negatively impacts a person’s quality of
Many posttraumatic olfactory disturbances are irre- life and potentially the level of disability. The following
versible. Recovery may occur in some cases but is depen- section provides the clinician with practical information to
dent on the mechanism of olfactory injury. The repair of address the functional ramifications of posttraumatic anos-
sinonasal obstruction or the resolution of cerebral hemor- mia.
rhage or contusion may improve olfaction. Sumner (1975)
suggested that early recovery from posttraumatic anosmia A. Disability Evaluation and Activities
could result in these instances. As with other neuronal of Daily Living
injuries, the prognosis for recovery from posttraumatic
anosmia worsens with time. Sumner (1964) noted that The American Medical Association impairment rating
39% of those who recover olfactory function do so within system, Guides to the Evaluation of Permanent
the first 10 weeks after injury. Costanzo and Becker (1986) Impairment, lists total bilateral loss of smell as a 3%
emphasized that if recovery of olfaction does not occur impairment of the whole person. Unilateral or partial
within 6 months to 1 year following injury, recovery is bilateral loss precludes any permanent impairment rating
unlikely. They reported that 33% of all moderately to from being assigned (AMA, 1993). We feel this rating
severely head-injured patients showed some improvement system may not fully represent the functional disability
in olfaction, whereas 27% worsened. In a more recent associated with olfactory loss following head trauma
study, Doty et al. (1997) retested 66 patients with olfactory and/or brain injury. In order to evaluate the degree of
dysfunction several years following head trauma and functional disability as well as resultant handicap, the
found that 24 (36%) improved slightly, 30 (45%) had no evaluating professional must first consider the particular
change, and 12 (18%) worsened. Improvement of anosmia patient’s activities of daily living (ADLs), vocational sta-
over a long duration may be explained by the regeneration tus, and avocational pursuits. Such evaluation is, in most
or recovery of damaged olfactory neurons. Costanzo and cases, difficult if not impossible to accomplish during a
colleagues have demonstrated recovery of olfaction in clinic visit and may require in-depth questioning and
hamsters following transection of the olfactory nerve on-site assessment. From a functional standpoint, the
fibers at the cribriform. Regenerating olfactory receptor main areas that must be considered relative to potential
cell axons have been shown to grow through the cribriform adverse effects of smell loss are safety issues, vocational
plate and reestablish contact with cells in the olfactory pursuits, appetite and nutrition, hygiene, homemaking,
bulb (Costanzo, 1983; Koster and Costanzo, 1996; Yee and child care, and hobbies and leisure activities.
Costanzo, 1995). The capacity for olfactory receptor cell Personal safety is a major consideration for individuals
recovery in humans has not been established. with impaired olfactory function. The inability to detect
smoke, gas leaks, spoiled foods, and chemical vapors in the
home or workplace presents a major risk to health and
IV. SPECIAL CONSIDERATIONS safety. Appropriate provisions must be provided to ensure
maximal safety for the patient as well as other persons in
Although loss of the sense of smell may seem to some the home environment. Patients should be advised concern-
like an insignificant functional impairment, such deficits ing the installation and use of smoke detectors and the
can be disabling and have a negative impact on patients’ routine monitoring of gas appliances. Perishable foods
quality of life (Miwa et al. 2001). The impact of olfactory should all be dated when purchased. Foods should be stored
dysfunction on an individual’s health, safety, ability to optimally to minimize the chance for spoilage. Precautions
function in the workplace or at home, and the effect on should be exercised in the use and storage of noxious sub-
quality of life are all important issues that require consid- stances used in household cleaning and gardening.
Head Injury and Olfaction 635

Patients employed as firefighters, chemists, plumbers, the child’s optimal welfare. Pet care may also present a
bakers, cosmeticians, florists, food technologists, and challenge to the anosmic patient. Scheduled litter box and
cooks might be significantly impaired in their job perfor- animal pen inspection may prove a useful compensatory
mance if they could no longer smell. Appropriate job technique. Anosmic patients will be unable to detect their
analysis is required prior to making final recommendations own body odor, the body odor of others, the smell of cloth-
regarding whether or not a given patient should attempt ing if soiled or musty, or the smell of colognes or per-
vocational reentry. Employee and employer education may fumes. A schedule should be established for hygiene tasks
be critical, as might job or workplace modifications to such as washing clothes and bathing. Appropriate use of
increase patient safety. Some individuals may not be able body fragrances can be encouraged by training the person
to return to prior jobs due to inherent safety risks not to measure out specified quantities of cologne or perfume.
addressable via compensatory strategies. Varney (1988) Avocational activities may also be adversely affected by
found a significant negative correlation between the pres- loss of smell following head trauma. Activities such as
ence of anosmia and successful vocational reintegration entertaining, gardening, painting, wood burning, leather-
although the frequency of vocational dysfunction among work, and cooking may all lose at least a component of
anosmic and hyposmic patients in general is unlikely as their inherent hedonic value secondary to the inability to
high as previously reported (Correia et al., 2000). perceive the associated smells and odors.
Unfortunately, there are no large-scale well-controlled From a quality-of-life standpoint, we seldom think what
studies on the effect of anosmia on vocational reentry fol- it would be like not to be able to smell the grass, trees, and
lowing traumatic brain injury, it would seem quite appar- flowers when we step outside, appreciate the smell of
ent that certain vocational pursuits might become more bacon on the griddle and coffee percolating on a weekend
difficult, if not dangerous, with complete anosmia. morning, smell our children or spouse, or breathe in crisp
Anosmic patients may report alterations in the taste of pine-scented mountain air or salt-laden air at the seashore.
foods rather than a loss of smell. In a study of 750 patients Individuals with posttraumatic anosmia may never again
with complaints of abnormal smell or taste perception, appreciate these smells and the associated pleasures that
Deems and colleagues found that fewer than 4% of those accompany them. Studies have shown a significant relation
with an olfactory loss and gustatory complaints actually between olfactory disturbances and clinical depression
had demonstrable gustatory loss (Deems et al., 1991). (Deems et al., 1991) and other psychological effects
Also, patients who complained only of gustatory loss were (Toller, 1999).
three times more likely to show objective evidence of
olfactory loss than gustatory loss. B. Medicolegal Considerations
Given the link between appetite and eating with the aro-
mas of food, it is not surprising that anosmic patients com- Assessment of olfaction in the head-injured patient may be
plain of “not enjoying food anymore.” Excess use of salt, an important factor in medicolegal matters involving per-
seasoning, or sweeteners in an attempt to find missing sonal injury litigation. Since subjective reports or simple
stimulation is not uncommon. Nutrition as well as medical bedside testing may be unreliable in objectively docu-
status can be compromised by anosmic patients who seek menting loss of olfactory function and identifying malin-
out select compensatory food items that may be exces- gerers, special protocols for sensory testing should be
sively salty, sweet, or crunchy. The astute clinician can employed to validate reproducibility of results (Doty et al.,
assist the anosmic patient in gaining greater enjoyment 1995; Heywood and Costanzo, 1986). In addition, a thor-
from their meals by remembering that the palatability of ough medical and psychiatric history should be obtained to
food is derived from more than just smell; taste, texture, rule out disorders that may affect smell but are unrelated to
and flavor all play a role. Altering flavor via modification the history of head trauma, including depression and med-
of texture, vision, temperature, or taste may prove a useful ication-induced dysosmia (Schiffman, 1983, 1994;
compensatory tool for the anosmic or hyposmic patient Schiffman et al., 1999). Complete evaluation including
relative to eating-satisfaction issues (Schiffman, 2000). neurological, otolaryngological, and radiological examina-
The anosmic individual may have problems with cook- tions as described previously should also be performed to
ing, child care, or pet care. Patients with anosmia who support or refute any patient claims of injury.
cook should be instructed to follow recipes and avoid over-
or underseasoning. Close monitoring of cooking is essen- C. Counseling and Information Resources
tial to avoid burning or overcooking. Child care issues are
multiple and include education regarding timely diaper In many cases of posttraumatic anosmia there is no medi-
changes, hygiene timing, and food preparation to ensure cal treatment available to the patient to help restore
636 Costanzo et al.

olfactory function. However, the physician can provide Table 2 Resources for Information on Olfactory Impairment
important advice and counseling about the risks and haz- and Head Injury
ards associated with anosmia. This includes safety issues 1. National Institute on Deafness and Other Communication
such as the inability to detect gas leaks, chemical vapors, Disorders (NIDCD)
and smoke, as well as routine ADLs such as food prepara- Address: Office of Health Communication and Public
tion, child care, personal hygiene, and the proper use of Liaison, 31 Center Drive, MSC 2320,
household chemicals such as cleaning agents, bleach, sol- Bethesda, MD 20892-2320
vents, and pesticides. A list of sample questions for Voice: (301) 496-7243
patients with olfactory impairment is given in Table 1. FAX: (301) 402-0018
Often patient counseling and reassurance are important in E-mail: marin_allen@nih.gov
Internet: http://www.nih.gov/nidcd/health/st.htm
helping patients to cope with the psychological burden of
their olfactory loss. By helping patients understand that 2. Association for Chemoreception Sciences (AChemS)
they have a recognized medical condition, the physician Address: 744 Duparc Circle, Tallahassee, FL 32312
can usually allay patients’ fears or frustrations, even in the Voice: (850) 531-0854
absence of definitive treatment. Resources for obtaining FAX: (850) 531-0854
additional information on olfactory impairment and brain E-mail: mmered@neuro.fsu.edu
Internet: http://neuro.fsu.edu/achems/TasteAndSmell.htm
injury are provided in Table 2.
3. American Academy of Otolaryngology–Head and Neck
Surgery (AAO-HNS)
V. SUMMARY Address: One Prince Street, Alexandria, VA 22314
Voice: (703) 519-1589
Olfactory loss is a frequent sequela of traumatic head FAX: (703) 299-1125
E-mail: entinfo@aol.com
injury. Patients are typically unaware of their loss until
Internet: http://www.entnet.org/smelltaste.html
some time after injury, and it is often overlooked during
initial clinical evaluation. It is important that the phy- 4. American Association of Neurological Surgeons
sician understand the mechanisms that can cause post- 22 South Washington Street, Park Ridge, IL 60068
traumatic anosmia and be familiar with current Voice: (847) 692-9500
methodologies for clinical assessment. A thorough his- FAX: (847) 692-2589
E-mail: info@aans.org
tory and physical exam including otolaryngological
Internet:http://www.neurosurgery.org/pubpages/patres/
evaluation are essential. Modern imaging technologies
headinjury.html

can play an important role in the patient work-up.

Specific olfactory function testing may be performed by
a variety of techniques and is critical in verifying the
Table 1 Sample Questions for Patients with Olfactory
Impairment
presence of olfactory loss and determining its severity.
Although there is often no therapeutic intervention that
Does anyone in your home smoke cigars or cigarettes?
can reverse posttraumatic anosmia, spontaneous recov-
Do you have smoke detectors in your home? How many and
where are they located?
ery may occur in some groups of patients, including
Do you have gas appliances (stove, furnace, water heater, grill, those whose dysfunction is secondary to hematomas or
fireplace)? whose overall function is not severely compromised.
Do you have a wood-burning fireplace or kerosene heater? Olfactory impairment may increase risk for personal
Are you responsible for the care of children, pets, or animals? injury due to inability to detect gas leaks, smoke, spoiled
Have you experienced changes in your appetite? foods, and other potentially harmful conditions. It is
Do you add excess salt or sugar to enhance the taste of food? important for the treating physician to assess for any per-
Are you concerned about bad breath or body odor? sonal and work-related disabilities, as patients may ben-
Do you check and monitor foods for spoilage? efit from counseling or compensatory strategies.
Do you use household chemicals such as cleaners, bleach, sol- Although loss of olfactory function is rarely a life-threat-
vents, and pesticides?
ening disability, it nevertheless adversely affects the
Is your sense of smell important in your job or work environ-
ment?
quality of life and can in some cases lead to changes in
mood, appetite, and behavior.
Head Injury and Olfaction 637

ACKNOWLEDGMENTS Doty, R. L., Shaman, P., and Dann, M. (1984). Development of

the University of Pennsylvania Smell Identification Test: a
This work was supported by Grant DC00165 from the standardized microencapsulated test of olfactory function.
National Institute on Deafness and Other Communication Physiol. Behav. 32:489–502.
Doty, R. L., McKeown, D. A., Lee, W. W., and Shaman, P. (1995).
Disorders.
A study of the test-retest reliability of ten olfactory tests.
Chem. Senses 20:645–656.
Doty, R. L., Yousem, D. M., Pham, L. T., Kreshak, A. A., Geckle,
REFERENCES R., and Lee, W. W. (1997). Olfactory dysfunction in patients
with head trauma. Arch. Neurol. 54:1131–1140.
AMA (1993). Guides to the Evaluation of Permanent Duncan, H. J., and Seiden, A. M. (1995). Long-term follow-up of
Impairment. American Medical Association, Chicago. olfactory loss secondary to head trauma and upper respiratory
Cain, W. S. (1989). Testing olfaction in a clinical setting. Ear tract infection. Arch. Otolaryngol. Head Neck Surg.
Nose Throat J. 68:316, 322–316, 328. 121:1183–1187.
Cain, W. S., Gent, J., Catalanotto, F. A., and Goodspeed, R. B. Ferrier, D. (1876). The hemispheres considered physiologically.
(1983). Clinical evaluation of olfaction. Am. J. Otolaryngol. In The Functions of the Brain. G. P. Putnam’s Sons, New
4:252–256. York, pp. 181–211.
Cain, W. S., Leaderer, B. P., Cannon, L., Tosun, T., and Ismail, H. Heywood, P. G., and Costanzo, R. M. (1986). Identifying nor-
(1987). Odorization of inert gas for occupational safety: psy- mosmics: a comparison of two populations. Am. J.
chophysical considerations. Am. Ind. Hyg. Assoc. J. 48:47–55. Otolaryngol. 7:194–199.
Cain, W. S., Gent, J. F., Goodspeed, R. B., and Leonard, G. Heywood, P. G., Zasler, N. D., and Costanzo, R. M. (1990).
(1988). Evaluation of olfactory dysfunction in the Olfactory screening test for assessment of smell loss follow-
Connecticut Chemosensory Clinical Research Center. ing traumatic brain injury. Proceedings of the 14th Annual
Laryngoscope 98:83–88. Conference on Rehabilitation of the Brain Injured
Cone, J. E., and Shusterman, D. (1991). Health effects of indoor Williamsburg, Virginia, 1990.
odorants. Environ. Health Perspect. 95:53–59. Hoffman, H. J., Ishii, E. K., and Macturk, R. H. (1998). Age-
Correia, S., Faust, D., and Doty, R. L. (2000). A reexamination of related changes in the prevalence of smell/taste problems
the rate of vocational dysfunction among patients with anos- among the United States adult population—results of the
mia and mild/moderate closed head injury. J. Clin. Exp. 1994 disability supplement to the National Health Interview
Neuropsychol. 15:1–12. Survey (NHIS). Ann. NY Acad. Sci. 855:716–722.
Costanzo, R. M. (1983). Neural regeneration and functional Ikeda, K., Tabata, K., Oshima, T., Nishikawa, H., Hidaka, H., and
reconnection following olfactory nerve transection in hamster. Takasaka, T. (1999). Unilateral examination of olfactory
Soc. Neurosci. Abstr. 9:699. threshold using the Jet Stream Olfactometer. Auris Nasus
Costanzo, R. M., and Becker, D. P. (1986). Smell and taste disor- Larynx 26:435–439.
ders in head injury and neurosurgery patients. In Clinical Inamitsu, M., Nakashima, T., and Uemura, T. (1990).
Measurements of Taste and Smell, H. L. Meiselman and R. S. Immunopathology of olfactory mucosa following injury to the
Rivlin (Eds.). MacMillian Publishing Company, New York, olfactory bulb. J. Laryngol. Otol. 104:959–964.
pp. 565–578. Jackson, J. H. (1864). Illustrations of diseases of the nervous sys-
Costanzo, R. M., and Zasler, N. O. (1991). Head trauma. In Smell tem. London Hospital Report 1:470–471.
and Taste in Health and Disease, T. V. Getchell, R. L. Doty, Kondo, H., Matsuda, T., Hashiba, M., and Baba, S. (1998). A
L. M. Bartoshuk, and J. B. Snow, Jr. (Eds.). Raven Press, New study of the relationship between the T & T olfactometer and
York, pp. 711–730. the University of Pennsylvania Smell Identification Test in a
Costanzo, R. M., and Zasler, N. D. (1992). Epidemiology and Japanese population. Am. J. Rhinol. 12:353–358.
pathophysiology of olfactory and gustatory dysfunction in Koster, N. L., and Costanzo, R. M. (1996). Electrophysiological
head trauma. J. Head Trauma Rehabil. 7:15–24. characterization of the olfactory bulb during recovery from
Costanzo, R. M., Heywood, P. G., Ward, J. D., and Young, H. F. sensory deafferentation. Brain Res. 724:117–120.
(1987). Neurosurgical applications of clinical olfactory Kraus, W., Fife, D., and Ramstein, K. (1986). The relationship of
assessment. NY Acad. Sci. 510:242–244. family income to the incidence, external causes, and outcomes
Deems, D. A., Doty, R. L., Settle, R. G., Moore Gillon, V., Shaman, of serious brain injury, San Diego County, California. Am.
P., Mester, A. F., Kimmelman, C. P., Brightman, V. J., and J. Public Health Vol 76:1345–1347.
Snow, J. B., Jr. (1991). Smell and taste disorders, a study of 750 Legg, J. W. (1873). A case of anosmia following a blow. Lancet
patients from the University of Pennsylvania Smell and Taste 2:659–660.
Center. Arch. Otolaryngol. Head Neck Surg. 117:519–528. Leigh, A. D. (1943). Defects in smell after head injury. Lancet
Doty, R. L., and Kobal, G. (1995). Current trends in the meas- 1:38–40.
urement of olfactory function. In Handbook of Olfaction and Levin, H. S., High, W. M., and Eisenberg, H. M. (1985).
Gustation, R. L. Doty (Ed.). Marcel Dekker, Inc., New York, Impairment of olfactory recognition after closed head injury.
pp. 191–225. Brain 108:579–591.
638 Costanzo et al.

Mifka, P. (1964). Post-traumatische anosmie. Wien. Med. Schiffman, S., S. Zervakis, J., Suggs, M. S., Shaio, E., and
Wochenschr. 114:793. Sattely-Miller, E. A. (1999). Effect of medications on taste:
Miwa, T., Furukawa, M., Tsukatani, T., Costanzo, R. M., example of amitriptyline HCl. Physiol Behav. 66:183–191.
DiNardo L. J. and Reiter E. R. (2001). Impact of olfactory Sumner, D. (1964). Post-traumatic anosmia. Brain 87:107–120.
impairment on quality of life and disability. Arch. Sumner, D. (1975). Disturbance of the senses of smell and taste
Otolaryngol. Head Neck Surg. 127:497–503. after head injuries. In Handbook of Clinical Neurology, P. J.
Moran, D. T., Jafek, B. W., Rowley, J. C. 3d., and Eller, P. M. Vinken and G. W. Bruyn (Eds.). North-Holland Publishing
(1985). Electron microscopy of olfactory epithelia in two Co., Amsterdam, pp. 1–25.
patients with anosmia. Arch. Otolaryngol. 111:122–126. Takagi, S. F. (1987). A standardized olfactometer in Japan. A
Mott, A. E., and Leopold, D. A. (1991). Disorders in taste and review over ten years. Ann. NY Acad. Sci. 510:113–118.
smell. Med. Clin. North Am. 75:1321–1353. Toller, S. V. (1999). Assessing the impact of anosmia: review of
Notta, A. (1870). Recherches sur la perte de l´odorat. Arch. Gen. a questionnaire’s findings. Chem. Senses 24:705–712.
Med. 15:385–407. Yee, K. K., and Costanzo, R. M. (1995). Restoration of olfactory
Ogawa, T., and Rutka, J. (1999). Olfactory dysfunction in head mediated behavior after olfactory bulb deafferentation.
injured workers. Acta Otolaryngol. Suppl. 540:50–57. Physiol. Behav. 58:959–968.
Ogle, W. (1870). Anosmia or cases illustrating the physiology Yee, K. K., and Costanzo, R. M. (1998). Changes in odor quality
and pathology of the sense of smell. Med. Chir. Trans. discrimination following recovery from olfactory nerve tran-
53:263. section. Chem. Senses 23:513–519.
Raveh, J., Vuillemin, T., and Sutter, F. (1988). Subcranial man- Yousem, D. M., Geckle, R. J., Bilker, W. B., McKeown, D. A., and
agement of 395 combined frontobasal-midface fractures. Doty, R. L. (1996). Posttraumatic olfactory dysfunction: MR
Arch. Otolaryngol. Head Neck Surg. 114:1114–1122. and clinical evaluation. Am J. Neuroradiol. 17:1171–1179.
Rotch, T. M. (1878). A case of traumatic anosmia and ageusia, Yousem, D. M., Geckle, R. J., Bilker, W. B., Kroger, H., and
with partial loss of hearing and sight. Bost. Med. Surg. J. 99: Doty, R. L. (1999). Posttraumatic smell loss: relationship of
130–132. psychophysical tests and volumes of the olfactory bulbs and
Schiffman, S. S. (1983). Taste and smell in disease. N. Engl. J. tracts and the temporal lobes. Acad. Radiol. 6: 264–272.
Med. 308:1337–1343. Zasler, N. D., Costanzo, R. M., and Heywood, P. G. (1990).
Schiffman, S. (1994). Changes in taste and smell: drug interac- Neuroimaging correlates of olfactory dysfunction after trau-
tions and food preferences. Nutr. Rev. 52:S11–S14. matic brain injury. Arch. Phys. Med. Rehabil. 71:814.
Schiffman, S. S. (2000). Intensification of sensory properties of Zusho, H. (1982). Posttraumatic anosmia. Arch. Otolaryngol.
foods for the elderly. J. Nutr. 130:927S–930S. 108:90–92.
31

Saliva: Its Role in Taste Function

Robert M. Bradley
University of Michigan, Ann Arbor, Michigan, U.S.A.

Lloyd M. Beidler
Florida State University, Tallahassee, Florida, U.S.A.

I. INTRODUCTION particularly important in taste function because they pro-

vide the environment for the majority of the tongue’s taste
Saliva is a complex fluid secreted by glands that drain into buds (Bradley, 1995).
the oral cavity. This fluid coats the whole surface of the Salivary secretion is under the control of the autonomic
oral mucosa, including the apical extensions of the taste nervous system (Garrett and Proctor, 1998). The output of
receptor cells, providing the external milieu of the taste saliva is reflexively controlled and is therefore very depen-
receptor. Thus, chemicals that eventually interact with the dent on afferent sensory input. Even though there is a base-
taste receptive membranes have to traverse this fluid layer; line flow of saliva, increased flow of saliva requires reflex
hence, the initial events in taste transduction that are some- activation of the autonomic innervation.
times referred to as perireceptor events involve saliva Normally the total electrolyte content of whole saliva is
(Bradley, 1991; Matsuo 2000; Spielman, 1990). only about 10% of serum concentrations. However, the
Salivary glands are composed of a large number of osmolality may vary greatly under different physiological
identical, blindly ending, tubes. The closed end consists of conditions and salivary flow rates. The acini of the glands
a cluster of pyramid-shaped acinar cells forming the gland secrete a solution called primary saliva that has about the
acini. The lumen of the acini are contiguous with a duct same Na and K concentrations as blood plasma.
system that converge on the main excretory duct that However, as the saliva passes through the various ducts
drains saliva (called final saliva) into the oral cavity and tubules, the primary saliva is modified, Na being
(Baum, 1993). There are one or two types of acinar cells reabsorbed from the saliva and K secreted into the saliva.
and four or five cell types in the duct system. Both the parasympathetic and sympathetic branches of
The bulk of saliva is secreted by three paired salivary the autonomic nervous system are involved in the control of
glands: the parotid, submandibular, and sublingual glands. secretion. The cells of origin of the parasympathetic innerva-
However, a series of smaller salivary glands in the mucosa tion are located along the medial border of the solitary tract
of the lips (labial), cheeks (buccal), palate (palatine), and nucleus in the brainstem in the superior and inferior saliva-
tongue (lingual) also contribute significantly to the overall tory nuclei, while the sympathetic innervation originates in
volume of saliva. The lingual salivary glands that drain the intermediolateral nucleus situated in the upper thoracic
into the clefts of the circumvallate and foliate papillae on segments of the spinal cord (Bradley, 1991). The parasympa-
the posterior tongue (von Ebner’s glands) are thought to be thetic secretomotor fibers travel to the glands with various

639
640 Bradley and Beidler

Figure 1 Brainstem and spinal cord autonomic circuits involved in the reflex control of salivary secretion. VII, facial nerve; IX, glos-
sopharyngeal nerve. (From Bradley, 1995).

cranial nerve branches, while the sympathetic innervation B. Proteins

distributes with blood vessels. Before synapsing with the
basolateral membranes of the gland acini, the parasympa- Recent advances in electrophoresis, DNA, and protein
thetic fibers first synapse in peripheral autonomic ganglia, sequencing and identification with labeled antibodies
and the sympathetic fibers synapse in the superior cervical have greatly increased our knowledge of the organic
ganglion (Fig. 1). composition of saliva. It is estimated that over 200 dif-
ferent proteins and peptides are contained in the saliva of
human and mouse (see Table 1). Many of these, however,
II. CHEMICAL COMPOSITION OF SALIVA
may be either isozymes or members of the same protein
family. They may range in size from small peptides to
The chemical composition of the secretion of each type of
large immunoglobins and include highly acidic, as well
salivary gland differs. The concentration of a given sub-
as basic, proteins. The heterogeneity of the salivary pro-
stance in whole saliva varies with flow rate, species, sex,
tein population is primarily due to peptide or protein
physical activity, pharmacological state, and time of day.
manufacture in the acini with polygenic determinants.
While 99% of saliva is water, the remaining 1% consists of
This may be followed by further modifications during the
ions and organic chemicals.
transcription and translation process, such as glycosyla-
tion or phosphorylation during the passage through the
A. Ions tubules. A variety of other molecules, including steroids
and a number of organic molecules, are included in
The most important cations of saliva are sodium and saliva. A given substance or a whole family of proteins
potassium, while the predominant anions are chloride may be separated with the aid of binding columns or gels,
and bicarbonate. Other electrolytes also occur in saliva purified, sequenced, and then computer-compared to
including calcium, phosphate, fluoride, thiocyanate, other known proteins. A more elegant method of protein
magnesium, sulfate, and iodide. Salivary water and ions identification is to isolate salivary RNA and derive a spe-
are derived from blood plasma, but their concentrations cific DNA clone whose gene can be sequenced and
in final saliva depends on a number of factors, of computer-compared.
which the most important is flow rate (Fig. 2) (Schneyer Many substances in the blood serum are also found in
et al., 1972). saliva. The salivary concentration of some of these is
Saliva: Its Role in Taste Function 641

Figure 2 Electrolyte composition, pH, and osmolality of human parotid gland saliva at resting (0.01 mL/min) and stimulated
(1.0 mL/min) flow rates. (Data from Shannon et al., 1974.)

proportional to the serum concentration. Since it is more sent 21%, the dorsum of the tongue 12%, and the combined
convenient to obtain a sample of saliva than of blood, the ventral surface of the tongue and floor of the mouth 12.3%
former is often used in routine tests to follow changes in of the total surface area, or 16% of the oral mucosa. This
serum concentrations of agents of interest (e.g., chemi- thin aqueous salivary layer is only a small barrier to pas-
cals, viruses), such as human immunodeficiency virus sage of water-soluble molecules into the epidermal layer of
(HIV). Salivary tests have been utilized for a number of the oral cavity. The speed of tissue permeation is more
steroid hormones, for drugs including cocaine and mari- dependent on the lipid character of the epidermal layer.
juana, and for a measure of cigarette smoking. Small mol- However, the thin aqueous layer on the surface of the teeth
ecules [molecular weight (MW) 100] pass into the is little protection for the exposed dental plaque that nor-
saliva from the blood compartment by filtration. The mally coats various surfaces of the teeth, and substances
speed of transfer of many larger molecules is related to the such as sugars and acids have access since no lipid barrier
lipophilicity or partition coefficient and the pK. The sali- exists between the plaque and oral cavity. Passage of
vary flow rate and pH are also important in determining chemicals through the more highly keratinized tissue of
the drug transfer. Most drugs are not bound to salivary the dorsal tongue surface has been studied with animal
proteins, whereas they appear in both bound and unbound models. Penetration was shown to be proportional to the
(active) forms in the blood plasma. magnitude of the lipid:aqueous partition coefficient
(Mistretta, 1971)
The high molecular weight glycoproteins such as
III. GENERAL FUNCTIONS OF SALIVA
the mucins adhere to the cell surfaces and contribute to
A. Tissue Permeability and Lubrication the slippery and tenacious character of the surfaces of
the oral cavity. They also give saliva its rather high vis-
Saliva coats the entire tissue area of the oral cavity. The cosity. Both viscosity and slipperiness contribute to
mean total surface area of the adult mouth is about 215 cm2, saliva’s unusual properties, which distinguish it from
and the average thickness of the mobile salivary film is other aqueous solvents. The water retention expressed by
0.07–0.10 mm (Collins and Dawes, 1987). The teeth repre- the solvation of the carbohydrate residues of the glyco-
642 Bradley and Beidler

Table 1 Proteins and Peptides in Mammalian Whole Salivaa starch for enzyme reaction and the length of time the starch
remains in the oral cavity before being swallowed. Since
N-Acetyl-D-glucosaminidase 6-Hydroxymethylpterin the stomach acidity is too high for amylase enzymatic
Agglutinagen A IgA Prostaglandins
activity, it is thought that little further starch digestion
Agglutinagen B IgG Pterin
occurs until the food enters the small intestine and is in
Agglutinagen C IgM Renin
Albumin Kallikrein contact with pancreatic amylase (Rosenblum et al., 1988).
Aldolase Lactoferrin However, starch digestion is not totally eliminated in
Amylase family Lactoperoxidase pathological conditions of the pancreas in which pancre-
Angiotensin II Lethal factor atic amylase activity is impaired. Young infants can digest
Anticomplementary factor Lingual lipase starch in spite of an effective absence of pancreatic activ-
Antileukoprotease Lipoproteins ity during the first 6–8 months of life, and since amylase is
Antitrypsin Lysozyme present in the saliva of term infants it is likely that this non-
Aproerytherin Mesodermal growth factor pancreatic source of amylase is responsible for starch
Biopterin Metalloprotease digestion (Lebenthal and Lee, 1984; Murray et al., 1986;
Bone marrow colony- Monopterin
Zoppi et al., 1972). Moreover, salivary amylase activity in
stimulating factor Mucins
the stomach is protected by the substrates of amylase as
Calmodulin Neopterin
Carbonic anhydrase Nerve growth factor well as the end products of amylase digestion (Rosenblum
Cathepsin Neural tube growth factor et al., 1988). Thus, salivary amylase may play a more
Collagenase Parotid glycoprotein family important role in starch digestion not only in the mouth,
Cystatin family Peroxidase but in other parts of the digestive tract as well.
Elastase Plasminogen activator The adhering of products of starch digestion to the
Endothelial growth- Platelet-activating factor dental plaque proceeds over a much longer time period
stimulating factor Proline-rich proteins than that of the free starch in the saliva of the oral cavity.
Esterase Sialomucin The longer period of adhesion affords amylase sufficient
Esterase B Sialotonin time to break the starch into its smaller sugar compo-
Esteropeptidase Somatostatin
nents. These are then metabolized by the bacteria, and
Ferritin Statherin
acid is released, which attacks the enamel surface. If
Fibronectin Tissue plasminogen activator
Fucosyltransferase Tonin sticky foods containing mono- or disaccharides are taken
Gastrin Transferrin into the mouth, acid will also accumulate at the surface of
Glucagon VEG protein the teeth over a long time period. This is in contrast to the
Glutamine–glutamic acid Vitamin B–binding proteins intake of sugar-containing drinks, which are much more
protein family Wound contraction factor rapidly removed from the oral cavity, resulting in a low
Granulocytosis-inducing factor Xanthopterin pH state of short duration and producing less tooth decay.
aIf individual protein family members and isoenzymes are included sep- The von Ebner salivary glands secrete lipase. Fat diges-
arately, the total number approaches 200. tion can begin in the mouth and is a very important reac-
tion in those infants having little pancreatic function or
those with cystic fibrosis. Furthermore, lingual lipase can
proteins contributes to lubrication under light load and function after being swallowed and passed into the stom-
high sliding speed. Lubrication under heavy loads is ach, where the pH is low (Hamosh, 1990).
provided by salivary statherin due to its amphipathic Perhaps the most important function of saliva in the
properties. The polar end of statherin adorbs to the digestive process is solubilization of the ingested food. The
hydroxyapatite surface of the tooth, and its less polar tail largest component of saliva is water. Many of the compo-
extends outward to form an oriented film, which nents of food are water-soluble and are released into the
increases horizontal sliding (Douglas et al., 1991). oral cavity during chewing. Solubilization also increases
Salivary lubrication is also very important in bolus for- dispersion, and both increase the efficiency of enzyme
mation, swallowing, and speech production. action.

B. Digestion C. Wound Healing

Amylase is a major protein component of human saliva. It Licking of wounds is a common observation in many
initiates the breakdown of starch in the oral cavity. The mammals. This instinctive behavior has been shown to be
extent of such digestion depends on the availability of the important in assisting wound healing. Saliva contains a
Saliva: Its Role in Taste Function 643

number of substances that have a role in wound healing calcium phosphate precipitation and controls the mineral-
such as epidermal growth factor (EGF), nerve growth fac- ization of the tooth.
tor (NGF), and transforming growth factor–alpha (TGF)
(Bodner, 1991; Humphreys-Beher, et al., 1994; Schultz F. Buffering Capacity
et al., 1991). The tongue, in particular, is constantly
exposed to destructive mechanical and chemical trauma, The pH of resting (or unstimulated salivary secretion)
and the EGF of saliva has an important role in promoting saliva is about 5.7–6.2. This value is very dependent on
wound healing within the oral cavity (Noguchi et al, 1991). salivary flow rate, rising with the magnitude of flow rate to
More recently, EGF in saliva has been shown to play a become more alkaline (7.6) (Fig. 2). The major compo-
role in maintaining taste buds. Removal of the sub- nents of the salivary buffering system are bicarbonates and
mandibular and sublingual salivary glands results in the phosphates. The buffering capacity is dependent not only
loss of taste buds in the fungiform papillae, which can be on the concentrations of phosphates and carbonates at any
prevented if EGF is supplied in the drinking water (Morris- given time, but also on their rate of replacement by
Wiman et al., 2000), indicating that salivary EGF has an changes in flow rate and by the total amount of saliva in
important role in taste bud maintenance. the oral cavity at any given time.
Small pH electrodes have been used to follow the
D. Temperature Control changes in dental plaque pH after a sucrose rinse. The
initial pH of the plaque fell soon after the sucrose rinse
A number of different mammals, including rats, that do not and then returned to the control value after 20 minutes as
lose heat by sweating or panting, spread saliva on their the buffering role of the saliva became effective.
body surface, resulting in heat loss by evaporative cooling. The buffering capability of saliva became apparent if the
In this means of cooling, conversion of body water to access of saliva to the dental plaque was restricted. The
water vapor results in a heat loss by the process of the initial pH fell and remained low for over 20 minutes
latent heat of evaporation (Mitchell, 1974). The impor- (Englander et al., 1959).
tance of saliva in temperature control is demonstrated in
rats that have been desalivated. Rats without saliva can tol- G. Role of Saliva in Taste Function
erate heat exposure for a much shorter time than normal
rats (Hainsworth, 1968). In the rat the submaxillary and Saliva plays a dominant role in the initial events in taste
sublingual gland provide the most significant contribution transduction. It is the medium in which solid food must
to evaporative cooling and the parotid gland provides sup- dissolve and be transported to the taste bud receptor cell
plemental secretion during extreme hyperthermia membranes. Since a layer of saliva always coats the tongue
(Hainsworth and Stricker, 1969, 1971). surface and fills the taste pores, taste stimuli have to dif-
fuse through this layer to gain access to the receptor sites
E. Control of Oral Flora and Dentition on the apical portion of the taste bud cells. This process has
been modeled by Beidler (1961) and DeSimone and Heck
Saliva regulates the bacterial and fungal populations that (1980). When calculations based on this model were com-
reside in the oral cavity. Lactoferrin, lactoperoxidase, pared to measures of the latency of the electrophysiologi-
N-acetyl-D-glucosaminidase, and lysozyme all have cal response (Marowitz and Halpern, 1977), diffusion time
antimicrobial action. The mucin glycoproteins have a high was not a major component in response latency.
viscosity, which helps clear the bacteria from the oral cav- Taste buds in the fungiform papillae are bathed in saliva
ity as well as modulate the bacterial colonization on vari- derived from the major salivary glands. However, the
ous surfaces. The histidine-rich proteins also help control largest population of taste buds is located in the circum-
fungal growth. vallate and foliate papillae. The taste bud density along the
Excessive growth of bacterial and fungal populations papilla walls is very high, amounting to 60% of the total
may lead to oral infections. The immunoglobins aid in the lingual taste buds in rats (Bradley, 1995). The clefts of the
control of such infections. In addition, bacterial accumula- circumvallate and foliate papillae are very narrow and
tions on the surface of the tooth localize acid production deep, isolating them from saliva of the major glands. Thus,
due to sugar metabolism. The salivary calcium phosphate slow diffusion cannot account for the rapid entry of
supersaturated levels help control the hydroxyapatite crys- tastant-laden saliva, and a pumping action during tissue
tal formation on the surface of the tooth. The cystatins, movement has been suggested as a mechanism. The fluid
statherin, and proline-rich proteins help control the level of milieu of the clefts is provided by saliva entering the base
salivary calcium phosphate. Statherin in particular inhibits of the grooves from the von Ebner salivary glands.
644 Bradley and Beidler

Electrophysiological recordings from circumvallate taste The salivary PRPs also chemically interact with many
buds concurrent with stimulation of von Ebner gland tastants in foods. For example, the taste threshold of tannic
secretion has demonstrated that fluid from these glands acid in mice parallels the ability of PRPs to bind tannic acid
can effectively rinse away a taste stimulus (Gurkan and (Glendinning, 1992). It is interesting that when tannin-con-
Bradley, 1988). Thus, the lack of saliva can greatly impede taining foods are ingested by many wild animals, their bod-
the sensation of taste by decreasing tastant solubilization ies react physiologically to increase the concentration of
as well as tastant transport. In addition to solubilization PRPs in the saliva and thus bind more tannin, which in turn
and transport, saliva has a number of other roles in taste decreases its salivary concentration and increases the taste
sensation. threshold. The concentration of salivary PRPs in many ani-
mals can be greatly altered by the administration of -
adrenergic antagonists such as isoproterenol (Bannister
1. Saliva as an Adapting Solution
et al., 1989; Mehansho et al., 1985).
Since saliva is a solution containing ions that can be tasted,
the taste receptor membranes are constantly exposed to a 3. Role of Saliva in Sour Taste
low concentration of these tastants. Of these ions, sodium,
The saliva’s ambient pH, its ability to maintain it (buffer-
potassium, chloride and bicarbonate can achieve concen-
ing capacity), and the type of acid introduced into the oral
trations above their taste-detection level (Fig. 2). Sodium
cavity are all important determinants of the magnitude of
ions have been shown to influence saltiness (Delwiche and
sourness of acids. A reduction in the perceived sourness of
O’Mahony, 1996; McBurney and Pfaffmann, 1963) result-
HCl has been shown to occur as a result of the buffering
ing in elevated taste thresholds. Moreover, since the ambi-
action of salivary bicarbonate (Christensen et al., 1987;
ent concentrations of sodium and potassium vary with
Helm et al., 1982; Norris et al., 1984). Since the pH of
salivary flow rate, their taste detection thresholds may also
saliva varies with a number of physiological parameters,
vary. Thus, when measuring taste thresholds, it is impor-
sourness and acid detection can vary.
tant to rinse the adapting solution away.
Similar results have been reported in electrophysio-
The significance of saliva as an adapting solution was
logical recordings from the rat chorda tympani.
demonstrated by Matsuo and coworkers, who compared
Responses of the chorda tympani to HCl differ when the
taste responses in anesthetized and unanesthetized animals
tongue is adapted to saliva or water. The response to HCl
(Matsuo et al., 1994; Matsuo and Yamamoto 1990, 1992).
is less when the tongue is bathed in saliva compared to
In acute recordings from the chorda tympani in rats in
responses recorded in water adaptation (Matsuo and
which the saliva is effectively washed off the tongue, the
Yamamoto, 1992). Chorda tympani responses to HCl
response to 0.1 M NaC1 was of much greater magnitude
were 40% lower in awake rats licking 0.01 M HCl than in
than the response to 0.5 M sucrose. In contrast, the
anesthetized rats in which saliva was absent (Matsuo et
responses to these two stimuli were of similar magnitude
al., 1994).
in unanesthetized animals in which the tongue was not
rinsed. Similar results can be obtained in anesthetized ani-
mals if the tongue is adapted to saliva rather than distilled
4. Von Ebner Gland Proteins
water. However, in experiments on the hamster chorda
tympani, tongue adaptation to saliva has no effect on the A salivary protein was found in human and rat von Ebner
response to sucrose, perhaps reflecting a species difference salivary glands that belongs to the lipocalin superfamily of
(Rehnberg et al., 1992). Moreover, the dramatic effect of lipophilic ligand carrier proteins (Bläker et al., 1993;
saliva on sweet responses in the rat are not observed at Schmale et al., 1990). The protein is expressed in the
either the solitary nucleus (Nakamura and Norgren, 1991) acinar cells of the gland and is secreted into the saliva. It
or pontine taste relay nuclei (Nishijo and Norgren, 1991). was postulated that this salivary protein binds bitter
hydrophobic tastants and thus concentrates them in the
saliva adjacent to the taste buds in the clefts of the circum-
2. Role of Saliva in Bitter Taste
vallate and foliate papillae. The hydrophobic tastant carri-
The most studied bitter taste is that of sucrose octaacetate ers would increase the availability of the bitter substances
(SOA) in mice. A close linkage of mouse genes for SOA to the taste receptor microvilli.
bitter taste and salivary proline-rich proteins (PRP) was The role of von Ebner protein in bitter taste has been
found by Azen et al. (1986). This is of interest in the study tested in mice. Because the von Ebner protein is not
of bitterness in humans since PRPs represent up to 70% of synthesized in mice, it was possible to introduce the gene
the total proteins in human parotid saliva. from the rat responsible for expressing the protein in
Saliva: Its Role in Taste Function 645

mice and then perform behavioral tests. The bitter-tasting Table 2 Electrochemical Seriesa
substance used in the behavioral test was dentatonium
Gold  1.7 volts
benzoate, and tests were performed on transgenic mice Mercury  0.91 volt
expressing the von Ebner protein and on control mice. Palladium  0.83 volt
While both transgenic and nontransgenic mice were aver- Silver  0.80 volt
sive to the higher concentrations of dentatonium ben- Tin  0.14 volt
zoate, at 0.01 mM the preference ratios were significantly Aluminum  1.7 volts
higher in the transgenic animals (Kock et al., 1994). aValues as reduction potentials.
Thus, the control mice appeared to be more sensitive to Source: From Hunsberger, 1974.
dentatonium benzoate, throwing some doubt on the role
of von Ebner protein in bitter taste. Further doubts about
the role of von Ebner protein in bitter taste function H. Saliva and Galvanic Currents
comes from its absence in some species despite an almost
universal ability to taste bitter compounds, its lack of dif- A source of electrical current is formed when two dissim-
ferential expression when stimulation with the four taste ilar metals are placed in an ionic medium such as a body
modalities (Spielman et al., 1993), and its identical neu- fluid. Galvani utilized such an arrangement to stimulate
clotide sequence to a protein present in tears known as nerve and muscles in his studies of electricity (Herness,
tear prealbumin (van’t Hof et al., 1997). Lastly, com- 1985), and Volta studied the effect of metals placed into the
pounds known to be extremely bitter in humans do not oral cavity which stimulated the taste buds (Piccolino,
bind to the von Ebner protein (Schmale et al., 1993). 1997). Today it is common for adults to have dental amal-
Thus, the physiological function of the von Ebner protein gam fillings or gold crowns, which, under certain condi-
remains an enigma. It may contribute to the clearance of tions, may act as a source of electrical current (Hugoson,
taste stimuli (Kock et al., 1994) or may play a role in the 1986). There may be two consequences. First, the currents
control of inflammatory processes (van’t Hof et al., may stimulate sensory nerves and initiate sensations of
1997). A further protein that is located in the ducts, but taste, pain, or other unusual sensory qualities. Second,
not the acini of von Ebner glands, has been described (Li electrolysis at the interface of the metal and ionic saliva
and Snyder, 1995), which also has no known function but may release metallic ions injurious to the surrounding tis-
serves to reinforce the unique character of the salivary sue and cause corrosive damage. Recently, the release of
glands associated with the foliate and circumvallate mercury from amalgam fillings has led to public debate
papillae. concerning their safety.
The magnitudes of galvanic currents are dependent on
the ionic content of saliva, including its pH, and the nature
of the metals. The saliva can vary with many conditions,
5. Gustin including the salivary flow rate. However, the magnitude
An acidic zinc-containing salivary protein named gustin of the voltages produced is mostly influenced by the
was described in human parotid saliva (Henkin et al., choice of the two metals involved, the magnitude being
1975). Gustin has been implicated in taste bud growth func- proportional to the difference in their reduction potentials
tions, and a role in taste bud maintenance was indicated and (see Table 2). Note that a large potential difference may be
was decreased in concentration in patients with taste disor- generated when an individual with a gold crown takes a
ders (Shatzman and Henkin, 1981). Because gustin con- piece of aluminum foil into the mouth. Since the path of
current may pass down the root canal and enter the nerve,
tained zinc, it was suggested that zinc is an important ion in
the generated pain is both immediate and strong! The
normal taste function (Henkin, 1984). Moreover, loss of
return current path is by way of the bony fluid and the
taste function was thought to be related to zinc deficiency,
saliva in contact with the foil (Schreiver and Diamond,
and zinc supplement was suggested for those patients who
1952). Note also that a gold crown should never be in
complained of loss or alteration of taste function. However,
direct contact with an amalgam filling!
the role of zinc in taste disorders has been questioned since
in a double-blind study the use of zinc was effectively
equivalent to a placebo in the treatment of taste disorders IV. SJÖGREN’S SYNDROME AND
(Anonymous 1979; Henkin et al., 1976). Moreover, it has SALIVARY FUNCTION
recently been shown that gustin is actually carbonic anhy-
drase VI and is therefore not a protein uniquely associated Sjögren’s syndrome (SS) is an autoimmune disease charac-
with taste buds (Thatcher et al., 1998). terized by exocrine gland dysfunction. One of the symptoms
646 Bradley and Beidler

of this disorder is decreased function of the salivary glands, Table 3 Drugs that Effect Salivary Secretion
reducing salivary flow and resulting in a dry mouth or xeros- Category Number of drugs
tomia (Sreebny and Broich, 1987; Van der Reijden et al.,
1999). SS is more frequent in females than males (9:1) and Antihistamine 56
is most often found in individuals over 40 years of age. Anticholinergic 48
Antihypertensive 45
Since saliva plays such an important role in taste func-
Anorexic 44
tion, it is not surprising to find that SS patients often com-
Diuretic 33
plain of impaired taste function. Patients with SS were Antidepressant 25
found to have a decreased taste acuity (Henkin, 1972). Antipsychotic 24
However, individuals with chronic and complete absence Analgesic 15
of the salivary glands have normal taste function when Antinauseant 11
tested, indicating that the taste system itself is normal Antiparkinsonism 11
(Weiffenbach, 1986). Moreover, in a more recent investi- Psychotherapeutic 11
gation of SS patients with complaints of impaired taste Cough and cold preparation 9
function, judgments of taste intensity were normal despite Antidiarrheal 8
a decreased threshold sensitivity (Weiffenbach et al., Anti-inflammatory 7
1995). Thus, the complaints probably do not reflect
changes in the peripheral taste system but may be due to
the problems with taste stimulus solubilization and deliv-
ery to the taste receptors. In addition, since taste transduc- suggest substitutes. Since dryness is particularly noted at
tion involves apical ion channels on the taste receptor night, the drug schedule should be such that maximum
membranes, the absence of a source of permeating ions dosage does not occur at that time.
may impair taste transduction.
VI. REMEDIES FOR LOW SALIVARY
FLOW RATES
V. EFFECTS OF DRUGS ON SALIVARY
SECRETION A. Changing Patient Habits

Although SS is often related to salivary dysfunction, a Very slow salivary flow rates greatly decrease the buffer-
large number of other diseases may also be involved, ing capacity of the fluid in the mouth, its ability to remove
including often recognized diseases such as hypertension, the starch or sugars from the surface of the tongue, and the
rheumatoid diseases, diabetes, and hyperlipidemia. Those amount of antibacteriosides, antibodies, growth factors,
patients with cystic fibrosis, sarcoidosis, Bell’s palsy, pan- enamel-supporting agents, and numerous other beneficial
creatitis, and thyroiditis also have symptoms, as do those proteins that normally pour into the oral cavity. No longer
with adrenocortical diseases. Since many of these diseases can the patient depend on saliva to maintain a healthy oral
are observed in the elderly population and since many of environment.
the associated medications also have an effect on saliva Mastication initiates the salivary response. Human con-
production, complaints of dry mouth are common in the sumption of a liquid diet (Metrecal) rather than normal
elderly. Nearly one in five older adults has xerostomia solid food for 1 week was shown to decrease parotid flow
(Rhodus and Brown, 1990). Over 400 drugs have been by 30% (Hall et al., 1967). A similar 2-week experiment
identified as producing xerostomic side effects (Sreebny with rats showed a 39–52% decrease in parotid gland
and Schwartz, 1986). The drugs that produce dry mouth weight accompanied by atrophy (Hall and Schneyer,
can be categorized according to therapeutic function 1964). Such experiments suggest that persons with low
(Table 3). salivary flow rate should change their diet to increase the
The number of people aged 65 years or older in the magnitude and length of time of chewing. Crisp vegetables
United States is about 50 million. Many of these are treated and fruits should replace mashed potatoes and other soft
for several conditions requiring several drugs to be taken foods, and nuts and hard pretzels should be taken as snacks
daily. Dry mouth is a common complaint. Fortunately, most for those with moderate but not severe dry mouth. Between
patients with dry mouth have no irreversible damage to the meals a noncaloric object can be held in the mouth. It is
major salivary glands. Furthermore, many such patients interesting to note that American Indians were reported to
take four or more drugs daily. The physician should recon- place a pebble in the mouth to stimulate salivary flow
sider the number of drugs given and their dose and should when water was not immediately available.
Saliva: Its Role in Taste Function 647

The patient should be advised to gargle with a water normal salivary components involved in remineralization
rinse immediately following a meal or snack to slow the (Douglas, 1991).
progression of tooth decay. Chewing gum for 30 minutes The antimicrobial function of normal saliva is very
or more after meals enhances remineralization of caries- important and is served by secretory IgA, histidine-rich
like lesions in human enamel (Manning and Edgar, 1992). proteins, lactoferrin, and lactoperoxidase. Artificial salivas
In addition, it is important to rinse the mouth with a suit- do not adequately address this issue. Since both oral bac-
able buffer before retiring or taking a nap. If a tablet is terial and fungal infections are not uncommon and
used instead of a liquid, it is important not to concentrate acquired immune deficiency is of increasing importance,
the tablet in any given area of the oral cavity or else dam- Levine et al. (1987) suggested that separate artificial sali-
age may occur. It is preferable to follow the tablet with liq- vas with appropriate oral medications be developed to
uid a few minutes after the remains of the tablet disappear. serve people with these ailments.
The use of commercial mouthwashes that contain high
concentrations of alcohol is to be discouraged.
Since the concentration of those agents in the oral fluid C. Salivary Stimulation
that normally aid in guarding against tissue damage is
greatly decreased, the patient should have his dentist check Many patients with salivary deficits still have some sali-
the mouth frequently. A continuous dialogue between vary tissue functioning. The flow rate is normally increased
medical doctor and dentist is warranted. by mechanical and gustatory stimulation (Kjeilen et al.,
Continued use of tobacco in any form as well as of alco- 1987). The myoepithelial cells respond to chewing and ini-
hol should be discouraged. Use of the two together is even tiate the reflex leading to salivary flow. The latter is directly
more deadly since many of the injurious chemicals associ- related to the number of teeth involved.
ated with tobacco are soluble in alcohol, which easily pen- The gustatory-salivary reflex is best initiated with sour
etrates the barrier of the outer cell layer of the oral tissue (citric acid) and sweet stimuli. Thus, chewing sweet-sour
and interacts with the dividing basal cells beneath. gum or lozenges is helpful for many with xerostomia.
Unfortunately, maximum stimulation is usually short-
B. Saliva Substitutes lived. De Rossi et al. (1987) reported the parotid glands’
resting flow rate as 0.08 mL/min for men aged 41–60,
One of the most important aspects of saliva is mouthfeel which rose to 0.59 mL/min after they chewed fruit gum
and associated bolus movement. Water alone is not a good and 1.7 mL/min after ingestion of lemon juice.
substitute except for temporary relief (Gans et al., 1990). Sreebny and Broich (1987) indicate that unstimulated flow
Considering the large number of different chemicals found of whole saliva of normal people is about 0.3–0.5 mL/min;
in whole mouth saliva, one would expect a good saliva most of this is of submandibular/sublingual origin. After 2%
substitute to be rather complex. The high molecular weight citric acid stimulation the whole saliva flow rate rises to about
mucins are the primary contributors to the saliva’s viscos- 1–3 mL/min, with the parotid and submandibular/sublingual
ity. A substitute of about 1% carboxymethylcellulose glands contributing a large fraction of the total saliva, the for-
(CMC) is often used in commercially available salivas to mer slightly more than the latter. The stimulated saliva is very
produce an acceptable viscosity. However, lubrication of important in eating and in speech. However, the feeling of oral
hard surfaces is poor with such a simple fluid. dryness may be more correlated with unstimulated saliva that
Replacement of CMC with mucin enhances consumer is primarily mucus in character. Thus, most commercial arti-
acceptability. The proline-rich glycoproteins of saliva ficial salivas, which do not contain mucin-like substances, are
serve to lubricate hard surfaces, and the addition of serum not very effective in relieving the dry mouth sensation
albumin enhances the effect (Hatton et al., 1987). Because (Sreebny and Broich, 1987).
of their formulation simplicities, most artificial salivas do Salivary gland secretion can be stimulated directly by
not produce a long-lasting lubricating effect (Olsson and parasympathetic drugs. Fox et al. (1991) used 5-mg cap-
Axell, 1991). Recently, polymer-based saliva substitutes sules of pilocarpine hydrochloride given three times a day.
have been developed as an artificial saliva with some suc- Two thirds of the patients with salivary hypofunction
cess (van der Reijden et al., 1996). increased their saliva production with only mild and toler-
Most saliva substitutes highlight viscosity but do not able side effects. These trials lasted 5 months.
adequately address the problem of remineralization. The If parasympathetic agonists can increase salivary gland
substrates are usually buffered and contain mineral salts output, then one would expect that antagonists would
including fluorides. However, there is no substitution for decrease salivary gland output. Many dry mouth complaints
statherin and the anionic proline-rich proteins that are the of the elderly are due to side effects of their medication.
648 Bradley and Beidler

ACKNOWLEDGMENTS Douglas, W., Reeb, E., Ramasubbu, N., Raj, P., Bhandary, K., and
Levine, M. (1991). Statherin: a major boundary lubricant of
The preparation of this chapter was supported in part by human saliva. Biochem. Biophys. Res. Commun. 180: 91–97.
NIH grants from the National Institute of Deafness and Englander, H. R., I. L. Shklair, and Fosdick, L. S. (1959). The
other Communication Disdorders numbers DC00288 and effects of saliva on the pH and lactate concentration in dental
plaques I. Caries-rampant individuals. J. Dent. Res.
DC04198 to RMB.
38:848–853.
Fox, P., Atkinson, J., Macynski, A., Wolff, A., Kung, D., Valdez,
REFERENCES I., Jackson, W., Delapenha, R., Shiroky, J., and Baum, B.
(1991). Pilocarpine treatment of salivary gland hypofunction
Anonymous (1979). Ineffectiveness of zinc in treating ordinary and dry mouth (xerostomia). Arch. Intern Med. 151:
taste and smell dysfunctions. Nutr. Rev. 37:283–285. 1149–1152.
Azen, E., Lash, I., and Taylor, B. (1986). Close linkage of mouse Gans, R., Watson, G., and Tabak, L. (1990). A new assessment in
genes for salivary proline-rich proteins (PRPs) and taste. vitro of human salivary lubrication using a compliant sub-
Trends Genet. 2:199–200. strate. Arch. Oral Biol. 35:487–492.
Bannister, A. J., Divecha, N., Ashmore, M., and McDonald, C. J. Garrett, J. R., and Proctor, G. B. (1998). Control of salivation.
(1989). Basic proline-rich proteins of murine parotid glands: Front. Oral Biol. 9:135–155.
induction of mRNA by isoprenaline and post-secretion pro- Glendinning, J. (1992). Effect of salivary proline-rich proteins on
cessing. Eur. J. Biochem. 181:371–379. ingestive responses to tannic acid in mice. Chem. Senses
Baum, B. J. (1993). Principles of saliva secretion. Ann. NY Acad. 17:1–12.
Sci. 694:17–23. Gurkan, S., and Bradley, R. M. (1988). Secretions of von Ebner’s
Beidler, L. M. (1961). Taste receptor stimulation. In Progress in glands influence responses from taste buds in rat circumval-
Biophysics and Biophysical Chemistry, XII, J. A. V. Butler, late papilla. Chem. Senses 13:655–661.
H. E. Huxley, and R. E. Zirkle (Eds.). Pergamon Press, New Hainsworth, F. R. (1968) Evaporative water loss from rats in the
York, pp. 107–151. heat. Am. J. Physiol. 214:979–982.
Bläker, M., K. Kock, C. Ahlers, F. Buck, and Schmale, H. (1993). Hainsworth, F. R., and Stricker, E. M. (1969) Evaporative cool-
Molecular cloning of human von Ebner’s gland protein, a ing in the rat: effects of partial salivation. Am. J. Physiol.
member of the lipocalin superfamily highly expressed in lin- 217:494–497.
gual salivary glands. Biochim. Biophys. Acta Gene Struct. Hainsworth, F. R., and Stricker, E. M. (1971). Evaporative cooling
Expression 1172:131–137. in the rat: differences between salivatory glands as thermoreg-
Bodner, L. (1991). Effect of parotid submandibular and sublin- ulatory effectors. Can. J. Physiol. Pharmacol. 49: 573–580.
gual saliva on wound healing in rats. Comp. Biochem. Physiol. Hall, H., and Schneyer, C. (1964). Salivary gland atrophy in rat
[A] 100A:887–890. induced by liquid diet. Proc. Soc. Exp. Biol. Med. 117: 789–793.
Bradley, R. M. (1991) Salivary secretion. In Smell and Taste in Hall, M., Merig, J., and Schneyer, C. (1967). Metrecal-induced
Health and Disease, T. V. Getchell, R. L. Doty, changes in human saliva. Proc. Soc. Exp. Biol. Med. 124:
L. M. Bartoshuk, and J. B. J. Snow (Eds.). New York: Raven 532–536.
Press, pp. 127–144.
Hamosh, M. (1990). Lingual lipase. In Lingual and Gastric
Bradley, R. M. (1995) Essentials of Oral Physiology. Mosby, St.
Lipases: Their Role in Fat Digestion, M. Hamosh (Ed.). CRC
Louis.
Press, Boca Raton, FL, pp. 35–106.
Christensen C. M., Brand J. G., and Malamud D. (1987). Salivary
Hatton, M., Levine, M., Margarone, J., and Aguirre, A. (1987).
changes in solution pH: a source of individual differences in
Lubrication and viscosity features of human saliva and com-
sour taste perception. Physiol. Behav. 40:221–227.
mercially available saliva substitutes. J. Oral Maxillofae.
Collins, L., and Dawes, C. (1987). The surface area of the adult
human mouth and thickness of the salivary film covering the Surg. 45:496–499.
teeth and oral mucosa. J. Dent. Res. 66:1300–1302. Helm, J. F., Dodds, W. J., Hogan, W. J., Soergel K. H., Egide, M. S.,
Delwiche, J., and O’Mahony, M. (1996). Changes in secreted sali- and Wood, C. M. (1982). Acid neutralizing capacity of human
vary sodium are sufficient to alter salt taste sensitivity: Use of saliva. Gastroenterology 83:69–74.
signal detection measures with continuous monitoring of the Henkin, R. I. (1984). Zinc in taste function. A critical review.
oral environment. Physiol. Behav. 59:605–611. Biol. Tr. Elem. Res. 6:263–280.
De Rossi, G., Focacci, C., and Campioni, P. (1987). Salivary Henkin, R. I., Talal, N., Larson, A. L., and Mattern, C. F. T.
gland physiology and radioisotope uptake. In Radioisotope (1972). Abnormalities of taste and smell in Sjorgren’s syn-
Study of Salivary Glands, G. De Rossi (Ed.). CRC Press, Boca drome. Ann. Intern. Med. 76:375–383.
Raton, FL, pp. 2–21. Henkin, R. I., Lippoldt, R., Bilstad, J., and Edelhoch, H. (1975).
DeSimone, J. A., and Heck, G. L. (1980). An analysis of the A zinc protein isolated from human parotid saliva. Proc. Natl.
effects of stimulus transport and membrane charge on the salt, Acad. Sci. USA 72:488–492.
acid and water-response of mammals. Chem. Senses Henkin, R. I., Schechter, P. J., Friedewald, W. T., Demets, D. L.,
5:295–316. and Raff, M. (1976). A double blind study of the effects of
Saliva: Its Role in Taste Function 649

zinc sulfate on taste and smell dysfunction. Am. J. Med. Sci. Morris-Wiman, J., Sego, R., Brinkley, L., and Dolce, C. (2000).
272:285–299. The effects of sialoadenectomy and exogenous EGF on taste
Herness, M. S. (1985). Neurophysiological and biophysical evi- bud morphology and maintenance. Chem. Senses 25:9–19.
dence on the mechanism of electric taste. J. Gen. Physiol. Murray, R. D., Kerzner, B., Sloan, H. R., Juhling, M. H., Gilbert,
86:59–87. M., and Ailabouni, A. (1986). The contribution of salivary
Humphreys-Beher, M. G., Macauley, S. P., Chegini, N., van amylase to glucose polymer hydrolysis in premature infants.
Setten, G., Purushotham, K., Stewart, C., Wheeler, T. T., and Pediatr. Res. 20:186–191.
Schultz, G. S. (1994). Characterization of the synthesis and Nakamura, K., and Norgren, R. (1991). Gustatory responses of
secretion of transforming growth factor- from salivary neurons in the nucleus of the solitary tract of behaving rats.
glands and saliva. Endocrinology 134:963–970. J. Neurophysiol. 66:1232–1248.
Hugoson, A. (1986). Results obtained from patients referred for Nishijo, H., and Norgren, R. (1991). Parabrachial gustatory
the investigation of complaints related to oral galvanism. neural activity during licking by rats. J. Neurophysiol.
Swed. Dent. J. 10:15–28. 66:974–984.
Kjeilen, J., Brodin, P., Aars, H., and Berg, T. (1987). Parotid sali- Noguchi, S., Ohba, Y., and Oka, T. (1991). Effect of salivary epi-
vary flow in response to mechanical and gustatory stimulation dermal growth factor on wound healing of tongue in mice.
in man. Acta Physiol. Scand. 131:169–175. Am. J. Physiol. 260:E620–E625.
Kock, K., Morley, S. D., Mullins, J. J., and Schmale, H. (1994). Norris, M. B., Noble, A. C., and Pangborn, R. M. (1984). Human
Denatonium bitter tasting among transgenic mice expressing saliva and taste responses to acids varying in anions titratable
rat von Ebner’s gland protein. Physiol. Behav. 56:1173–1177. acidity and pH. Physiol. Behav. 32:237–244.
Lebenthal, E., and Lee, P. C. (1984). Alternative pathways for Olsson, H., and Axell, T. (1991). Objective and subjective effi-
digestion and adsorption in early infancy. J. Pediatr. cacy of saliva substitutes containing mucin and car-
Gastroenterol. Nutr. 3:1–3. boxymethylcellulose. Scand. J. Dent. Res. 99:316–319.
Levine, M., Aguirre, A., Hatton, M., and Tabak, A. (1987). Piccolino, M. (1997) Luigi Galvani and animal electricity: Two
Artificial salivas: present and future. J. Dent. Res. 66: 693–698. centuries after the foundation of electrophysiology. Trends
Li, X.-J., and Snyder, S. H. (1995). Molecular cloning of Ebnerin, Neurosci. 20:443–448.
a von Ebner’s gland protein associated with taste buds. J. Biol. Rehnberg, B. G., Hettinger, T. P., and Frank, M. E. (1992)
Chem. 270:17674–17679. Salivary ions and neural taste responses in the hamster. Chem.
Manning, R., and Edgar, W. (1992). Salivary stimulation by Senses 17:179–190.
chewing gum and its role in the remineralization of caries-like Rhodus, N., and Brown, J. (1990). The association of xerostomia
lesions in human enamel in situ. J. Clin. Dent. 3:71–74. and inadequate intake in older adults. J. Am. Diet. Assoc.
Marowitz, L. A., and Halpern, B. P. (1977). Gustatory neural 90:1688–1692.
response of the chorda tympani to lick-duration stimuli. Rosenblum, J. L., Irwin, C. L., and Alpers, D. H. (1988). Starch
Chem. Senses Flav. 2:457–485. and glucose oligosaccharides protect salivary-type amylase
Matsuo, R. (2000). Role of saliva in maintaining taste sensitivity. activity at acid pH. Am. J. Physiol. 254:G775–G780.
Crit. Rev. Oral Biol. Med. 11:216–229. Schmale, H., Holtgreve-Grez, H., and Christiansen, H. (1990).
Matsuo, R., and Yamamoto, T. (1990). Taste nerve responses dur- Possible role for salivary gland protein in taste reception indi-
ing licking behavior in rats: Importance of saliva in responses cated by homology to lipophilic-ligand carrier proteins.
to sweeteners. Neurosci. Lett. 108:121–126. Nature 343:366–368.
Matsuo, R., and Yamamoto, T. (1992). Effects of inorganic con- Schmale, H., Ahlers, C., Bläker, M., Koch, A., and Spielman, A.
stituents of saliva on taste responses of the rat chorda tympani I. (1993). Perireceptor events in taste. In The Molecular Basis
nerve. Brain Res. 583:71–80. of Smell and Taste Disorders, D. Chadwick, J. Marsh, and
Matsuo, R., Yamamoto, T., Ikehara, A., and Nakamura, O. (1994). J. Goode (Eds.). John Wiley, Chichester, pp. 167–185.
Effect of salivation on neural taste responses in freely moving Schneyer, L. H., Young, J. A., and Schneyer, C. A. (1972).
rats: Analyses of salivary secretion and taste responses of the Salivary secretion of electrolytes. Physiol. Rev. 52:720–777.
chorda tympani nerve. Brain Res. 649:136–146. Schreiver, W., and Diamond, L. (1952). Electromotive forces and
McBurney, D. H., and Pfaffmann, C. (1963). Gustatory adapta- electric currents caused by metallic dental fillings. J. Dent.
tion to saliva and sodium chloride. J. Exp. Psychol. Res. 31:205–229.
65:523–529. Schultz, G., Rotatori, D. S., and Clark, W. (1991). EGF and TGF-
Mehansho, H., Clements, S., Sheares, B. T., Smith, C., and  in wound healing and repair. J. Cell. Biochem. 45:346–352.
Carlson, D. M. (1985). Induction of proline-rich glycoprotein Shannon, I. L., Suddick, R. P. and Dowd, F. J. (1974). Saliva:
synthesis in mouse salivary glands by isoproterenol and by composition and secretion. In Monographs in Oral Science,
tannins. J. Biol. Chem. 260:4418–4423. Vol. 2. S. Karger, Basel, pp. 1–103.
Mistretta, C. (1971). Permeability of tongue epithelium and its Shatzman, A. R., and Henkin, R. I. (1981). Gustin concentration
relation to taste. Am. J. Physiol. 8:73–78. changes relative to salivary zinc and taste in humans. Proc.
Mitchell, D. (1974) Physical basis of thermoregulation. In Natl. Acad. Sci. USA 78:3867–3871.
Environmental Physiology, D. Robertshaw (Ed.). Spielman, A. I. (1990) Interaction of saliva and taste. J. Dent.
Butterworth, London, pp. 1–32. Res. 69:838–843.
650 Bradley and Beidler

Spielman, A. I., D’Abundo, S., Field, R. B., and Schmale, H. Van der Reijden, W. A., Vissink, A., Veerman, E. C. I., and Nieuw
(1993) Protein analysis of human von Ebner saliva and a Amerongen, A. V. (1999). Treatment of oral dryness and
method for its collection from the foliate papillae. J. Dent. related complaints (xerostomia) in Sjögren’s syndrome. Ann.
Res. 72:1331–1335. Rheum. Dis. 58:465–473.
Sreebny, L. M., and Broich, G. (1987). Xerostomia (dry mouth). Van’t Hof, W., Blankenvoorde, M. F. J., Veerman, E. C. I., and
In The Salivary System, L. M. Sreebny (Ed.). CRC Press, Nieuw Amerongen, A. V. (1997). The salivary lipocalin Von
Boca Raton, FL, pp. 179–202. Ebner’s gland protein is a cysteine proteinase inhibitor. J.
Sreebny, L., and Schwartz, S. (1986). A reference guide to drugs Biol. Chem. 272:1837–1841.
and dry mouth. Gerodontology 57:75–99. Weiffenbach, J. M., Fox, P. C., and Baum, B. J. (1986). Taste
Thatcher, B. J., Doherty, A. E., Orvisky, E., Martin, B. M., and and salivary function. Proc. Natl. Acad. Sci. USA
Henkin, R. I. (1998). Gustin from human parotid saliva is car- 83:6103–6106.
bonic anhydrase VI. Biochem. Biophys. Res. Commun. 250: Weiffenbach, J. M., Schwartz, L. K., Atkinson, J. C., and Fox, P. C.
635–641. (1995). Taste performance in Sjögren’s syndrome. Physiol.
Van der Reijden, W. A., Van der Kwaak, H., Vissink, A., Behav. 57:89–96.
Veerman, E. C. I., and Nieuw Amerongen, A. V. (1996). Zopi, G., Andreotti, G., Pajno-Ferrara, F., Njai, D. M.,
Treatment of xerostomia with polymer-based saliva substi- and Gaburro, D. (1972). Exocrine pancreatic function in
tutes in patients with Sjögren’s syndrome. Arthritis Rheum. premature and full term neonates. Pediatr. Res.
39:57–69. 6:880–889.
32

Morphology of the Peripheral Taste System

Martin Witt
University of Technology Dresden, Dresden, Germany

Klaus Reutter
University of Tübingen, Tübingen, Germany

Inglis J. Miller, Jr.

Wake Forest University School of Medicine, Winston-Salem, North Carolina, U.S.A.

I. INTRODUCTION perception. One aim of this chapter is to provide sufficient

knowledge of peripheral gustatory anatomy as a basis for
Many papillae are evident, I might say, innumerable, understanding other chapters of this book. Some structural
and the appearance is so elegant that they catch the details about the human peripheral taste system are well
view and thoughts of the observer, and control him known, but it is also worthwhile to provide comparative
for a long time and not without enjoyment....All this anatomical information to fill in the gaps or understand and
delights the curious mind, when observing with an establish basic principles. The fundamental question of
engyscope; and if anyone asks to what they are simi- how tastants are perceived has been addressed for more
lar, I am unsure whether I should compare this huge than two millennia, and the concepts, theories, and experi-
number of papillae first with grapes or fruits of the mental “proofs” proposed have been discarded in the past
bay, or innumerable mushrooms emerging between that our present-day concepts can be incorporated.
fine, densely standing blades of grass... Therefore, this chapter also includes important historical
paradigms in taste research which may serve as caveats for
These are the first detailed and enthusiastic words on sculpting the discipline as contemporary concepts emerge.
papillae and their “membranes” in human tongues by Aristotle (384–322 B.C.), applying Platonic concepts,
Lorenzo Bellini (1665) (Fig. 1), who knew that Marcello argued that taste sensation was carried from the tongue via
Malpighi (1628–1694) had reported on lingual papillae the the blood to the liver or heart, which was the common seat
year before. Malpighi discovered mucosal elevations of the soul and all sense perception (Siegel, 1970). Galen’s
associated with nerve fibers, naming these elevations (Claudius Galenus, 129–201 A.D.) anatomical analyses
“papillae,” as the morphological substrates for gustatory challenged this notion: his detailed studies on the innerva-
sensation (Fig. 2). tion of the tongue describe correctly the different functions
The peripheral taste apparatus includes the gustatory of the three principal nerves supplying the tongue (lin-
sensory organs or taste buds and their innervation, and the gual, glossopharyngeal, and hypoglossal nerves) and
specific papillae within which taste buds are assembled. demonstrate their origin at the base of the brain. Galen
What one needs to know about taste buds in relation to how posited that the lingual nerve communicated gustatory
one perceives tastants depends on the approach to taste sensations, a concept yet resonant in contemporary

651
652 Witt et al.

neurobiology. One strand of nerve fibers corresponding to

cranial nerve IX (CN IX) (rediscovered in humans by
Panizza in 1834) was already known to Galen as the
principal gustatory nerve of the tongue. CN IX also carries
some motor fibers to the pharynx. Galen also noted excre-
tory ducts of the lingual and submandibular glands (in ox).
The particular structure of the tongue surface was
described initially by Casserius (1609) and later by
Malpighi (1664; cited: Malpighi, 1687) and Bellini (1665)
(reviewed by Jurisch, 1922). Further evidence of the sig-
nificance of the papillary epithelium and its cells came
from observations in taste organs of the frog (Waller, 1847,
1849; Fixsen, 1857). Taste buds were identified initially on
the barbels and skin of fishes by Leydig (1851) and
described as becherförmige Organe (goblet-shaped
organs), whose function he associated with tactile sensitiv-
ity. Schulze (1863) subsequently suggested they were
chemosensory structures. Similar organs in mammals were
described as Schmeckbecher (taste goblets) (Lovén,
1868) and Geschmacksknospen (taste buds) or
Geschmackszwiebeln (taste onions) (Schwalbe, 1868).
Herrick (1904) translated Geschmacksknospen as “taste
buds.” Their location within lingual papillae, the latter
already associated with the loci of taste perception, lent
credence to their identity as taste sensor organs.
Nineteenth-century studies focused on cytological fea-
tures, nerve supply (Figs. 3, 4), and the development of
taste buds. While the nerve-dependent nature of taste
Figure 1 Cover sheet of Lorenzo Bellini’s book summarizing sensation was known since Galen’s time, its significance
examination of the anatomy of the kidney and taste organs. for sensory organ physiology blossomed in the mid-1800s.
(Langerack, Leiden, 1711.) With development of the neuron doctrine (see Koelliker,
1844), previously described “ganglion globules”
(Ganglienkugeln) (Ehrenberg, 1833) could now be
acknowledged as part of a specialized cellular system (cell
theory of Schleiden and Schwann) (Schwann, 1839). Two
major prerequisites favored the expansion of scientific

Figure 2 Depiction of a bovine tongue

by Marcello Malpighi (1686) showing
“patches” where papillae were observed.
Note the concentration of fungiform
papillae (dots) on the tip of the tongue.
The drawing of vallate and foliate papil-
lae is still rather vague.
Morphology of the Peripheral Taste System 653

Figure 3 These illustrations from Bourgery and Jacob (1839) show the human tongue including nerve and blood supply, as well as the
lingual muscle system. In spite of precise macroscopical observations on the innervation of glands and muscles, it still represents indi-
rectly the (nowadays revised) morphological concept of gustation in the early nineteenth century, according to which the principal taste
nerves are the lingual (p in A and B) and glossopharyngeal (t in B) nerves. The chorda tympani, although depicted near the submandibular
ganglion (q in A), does not reach the lingual dorsum. Ebner’s glands are not known to the authors. (A) D, Submandibular gland with q,
chorda tympani and submandibular ganglion and p, lingual nerve; s, hypoglossal nerve; E, sublingual gland; F, Nuhn’s gland. (B) C,
Styloid process; b, stylopharyngeus muscle, the leading muscle for t, glossopharyngeal nerve; k, lingual artery; p, lingual nerve, inter-
rupted to show the glossopharyngeal nerve; q, part of the chorda tympani nerve; s, hypoglossal nerve; X, trunk of the facial nerve. (See
color insert.)
knowledge in the nineteenth century: (1) replacement Müller (1801–1858) in Germany, and (2) technical
of the often speculative natural philosophy by the experi- advances, which included improved microscopes (K.
mentally based natural sciences, mainly represented by Zeiss, 1816–1888, together with Ernst Abbé, 1840–1905;
Francois Magendie (1783–1855) in France and Johannes and E. Leitz, 1843–1920), as well as use of histological
654 Witt et al.

per contiguitatem.” This first observation of secondary

sensory cells in a taste organ that was clearly different
from primary sensory cells, as had already been described
in the olfactory epithelium, albeit without the understand-
ing of the “contact” nature of taste bud cells and their
innervating afferents as later described ultrastructurally,
was opposed by Retzius (1892). Using the Golgi silver
impregnation method, Retzius attributed a sensory func-
tion only to free nerve endings. Later, Krause (1911)
clearly demonstrated that the finest nerve fibers enter the
taste bud and ascend to the taste pore but do not merge
with taste bud cells whose basal processes end near the
basal cells.

II. LINGUAL PAPILLAE AND TASTE BUD

DISTRIBUTION
A. Nongustatory Papillae

Approximately 60 years before taste buds were identified

as gustatory organs, an illustration of the human tongue by
Soemmering (1806) (Fig. 4) accurately showed the regional
distribution of lingual papillae. A line called the sulcus ter-
minalis (an ontogenetic remnant, see below), which is
located posterior to the vallate papillae, separates the body
of the tongue from the lingual root. It may be seen that the
sulcus terminalis extends laterally to the pharyngeal wall
from the foramen caecum (also an ontogenetic remnant, see
Figure 4 The first precise depiction of a human tongue by below) near the midline (see Figs. 4, 12). The root of the
Samuel Soemmering (1806). (Above) The anterior part (left side) tongue is covered by a papilla-free smooth epithelium, and
of the tongue shows numerous fungiform papillae. Behind the beneath this epithelium lie mucous glands and a reticular
V-like arrangement of vallate papillae (right side), the lingual
connective tissue filled with lymphatic follicles, which led
tonsils and the entrance to larynx with the epiglottis are visible.
(Below) Lateral view of the tongue shows the left lingual artery
to the designation of “lingual tonsil.” Ducts of intralingual
and its ramifications into gustatory papillae, which appear as red salivary glands (Ebner, 1873) empty into the troughs of val-
dots after injection of the artery with a red dye (observation of late and foliate papillae (see below).
Soemmering). The dorsal surface of the tongue is covered with filiform
and conical papillae from the sulcus terminalis to the tongue
tip. Filiform papillae are the most prevalent type, while the
techniques, e.g., introduction of chromic acid for histolog- number of conical papillae may vary. Both types of papillae
ical examination of neural tissue (Hannover, 1840), allow- are sparse along the lingual margin and abundant in the
ing for the distinction between cells and fibers. Further, middle regions. Conical papillae have a cylindrical base and
gold chloride staining facilitated discriminating finely taper to a sharp point at their apex. Filiform papillae
ramifying nerve fibers (Gerlach, 1858). Helmholtz (1842) (L. filum
thread) have a pyramidal shape and a narrow tail
observed a direct continuity between nerve cells (globuli of cornified cells extending from their apical tips as a pen-
gangliosi) and their fibers in evertebrates. The demand for nant. The fila are part of the fibrous mat on the tongue’s
the neuron doctrine in vertebrates was established by surface in the hypertrophic condition called “hairy tongue.”
Koelliker (1844). Introduction of the methylene blue stain-
ing method (Ehrlich, 1886) facilitated observations that
fine nerve fibers in the frog taste disc widened with vari- B. Gustatory Papillae
cosities, penetrated the gustatory epithelium and
approached the sensory cells “with extremely sharp small Taste buds occur in distinct papillae of the tongue, the
knobs, [which]...connect taste cells not continuously, but epithelium of the palate, oropharynx, larynx (epiglottis),
Morphology of the Peripheral Taste System 655

and the upper esophagus. Taste buds of most vertebrates sparse within foliate papillae and the soft palate, including
are bulb-shaped structures composed of about 50–120 the uvula. Hoffmann concluded that the development of
bipolar cells (see Figs. 5, 9, 11). With the exception of taste perception is dependent on the number of taste buds
basal cells, the slender taste bud cells arise from an inter- on a particular location.
rupted basal membrane and converge with their apical Of the approximately 4600 total taste buds in all three
protrusions, the microvilli, into the mucus-filled taste pit. lingual fields in humans, vallate buds comprise about 48%
Together, these cells form the organ’s sensory epithelium. (2200), foliates about 28% (1280), and fungiforms 24%
The nuclei of the cells are located in the lower third of the (1120). However, taste bud numbers vary greatly among
taste bud, which is approximately the region where most individuals (Miller and Reedy, 1990a), with some adults
afferent nerve fiber terminals are distributed. Sensory cells possessing a total of only 500 taste buds (Linden, 1993). The
possess transmembrane receptors and/or ion channels for taste bud density of foliate papillae seems to be constant in
specific taste stimuli at apical and lateral portions of the life (Hou-Jensen, 1933), but age-related differences have
cell membrane. Taste buds are demarcated from surround- been reported for vallate papillae, which are more numerous
ing nongustatory epithelial cells by specialized epithelial and containing more taste buds in younger individuals
cells (marginal cells). (Jurisch, 1922). There are also more marginal fungiform
papillae during the late fetal and newborn period, but they
usually lack taste buds and are referred to as “sucking papil-
1. Distribution of Lingual Taste Buds
lae” (Habermehl, 1952; Yamasaki and Takahashi, 1982).
The pattern of taste bud distribution over the tongue sur-
face is similar among humans and other mammals. Lingual 2. Vallate Papillae
taste buds are found exclusively within gustatory papillae,
Vallate papillae, comprehensively described by Haller
i.e., those bearing taste buds. Similar types of gustatory
(1766) and Soemmering (1806), lie directly anterior to
papillae are located on homologous regions of adult
the sulcus terminalis and extend in a V-shaped line across
mammalian tongues. The gustatory papillae include the
the root of the tongue (Fig. 4). They are round and mea-
vallate, foliate, and fungiform papillae. As the term sug-
sure between 2 and 8 mm in diameter. The pores of taste
gests, typical fungiform papillae are mushroom-shaped,
buds open into the trenches around the bases of each val-
with a slender neck and an enlarged head (Figs. 6–8). But
late papilla (Fig. 6). The papillae are innervated by an
the majority of fungiform papillae vary in form, and the
enormously large nerve fiber plexus originating from the
filiform papillae are intermingled among them. Shortly
glossopharyngeal nerve [cranial nerve, IX (CN IX) see
after the published discovery of taste buds in humans
below] compared to foliate or fungiform papillae. The
(Schwalbe, 1868; Lovén, 1868), the first systematic inves-
number of vallate papillae per human tongue varies
tigations on the distribution of taste buds within the oral
between 4 and 18 (n
2264 tongues), with an average of
cavity in humans were carried out by the medical student,
9.2  1.8 papillae (Munch, 1896). Ninety-eight percent
Hoffmann (1875). He emphasized that taste buds are more
of all tongues have a central median papilla (Fig. 4). The
presence of three or four lateral papillae on each tongue
half was observed most often (20%). Atrophic changes
were observed in papillae of some men 40 years old
and some women 55–60 years old, though Jurisch
(1922) reported that the number of vallate papillae did
not appear to change systematically as a function of age.
The average numbers of taste buds per papilla are sum-
marized in Table 1.

3. Foliate Papillae
Foliate papillae in humans were first reported by Albinus
(1754) and histologically described by Hönigschmied
(1873) but did not become the focus of scientific attention
in humans until the twentieth century. These papillae,
located bilaterally along the posterolateral margins of the
Figure 5 Two taste buds of a rabbit foliate papilla. Some cells, tongue surface, consist of parallel rows of ridges (folia)
as well as some nerve fibers, are reactive for NSE. and valleys, which lie adjacent to the lower molar teeth.
656 Witt et al.

Figure 6 Schematic drawings of taste bud–bearing mammalian lingual papillae (longitudinal sections). (A) Fungiform papilla (papilla
fungiformis) with two apically situated taste buds and their innervation. On the right side, a (taste bud–free) filiform papilla. (B) Foliate
papillae (pp. foliatae). (C) Vallate papilla (p. vallata). In B and C, the taste buds are directed to the lateral trenches of the papillae. Serous
Ebner glands drain to the trenches. Dermal connective tissue is rich in nerve fibers; below the taste buds they form subgemmal plexus.

Ducts located between the folia transmit secretions from tongue. The rostralmost furrows (lateral rugae) contain no
mostly serous lingual glands within the root of the tongue. glandular ducts or taste buds, and their epithelia are more
Scanning electron microscopy of human foliate papillae cornified than that between foliate papillae (Hou-Jensen,
and transmission electron microscopy of their resident 1933). Fungiform papillae can be found on the tops of
taste buds was reported by Svejda and Janota (1974) and these lateral rugae. Although human foliate papillae were
Azzali et al. (1996). The number of taste buds in human thought to be “rudiments” (in comparison to the well
foliate papillae was reported by Hou-Jensen (1933) and developed rabbit foliate papillae), Mochizuki (1939) cal-
Mochizuki (1939) (see Table 1). Confusion in finding taste culated an average of 1300 taste buds per tongue, which
buds within the folds of human foliate papillae reflects exceeds the number of fungiform taste buds [800 buds:
papilla structure. As many as 20 parallel ridges and fur- Braus (1940); 1000 buds: Miller and Bartoshuk (1991)].
rows are found on the posterolateral margin of the human Indeed, contiguous taste buds within the same cleft may

Table 1 Distribution of Taste Bud Numbers in or on Human Oral or Lingual Structures

Vallate papillae Fungiform papillae Foliate papillae Palate Larynx Ref.
– 20–30 (fetal) ? 15–20 per papilla. Hoffmann, 1875
400 Wyss, 1870
252  151a Arey et al., 1935
234  114a 1279 Mochizuki, 1939
240  125a 1000 (total) – Miller and Bartoshuk, 1991
800 (total) Braus, 1940
708–1328 Hou-Jensen, 1933
 2500 (neonate) Lalonde and Eglitis, 1961
 25 (senile) Jowett and Shrestha, 1998
585 Cheng and Robinson, 1991
1120 Miller and Reedy, 1990b
a per papilla.
Morphology of the Peripheral Taste System 657

Figure 8 Human fungiform papilla during development, week

15. One taste pore is visible (arrow). Scale bar
50 m.

Figure 7 Fungiform papillae during development (12th post- over a large area of tongue surface. The anterior portion of
ovulatory week). (Above) Anterior part of the tongue. the tongue extends from the line of vallate papillae to the
Paramedian sagittal section. (Below) Detail of the anterior part at tongue tip (Figs. 4, 7, 8). This region contains about 30 cm2
the tongue’s tip. Only a few papillae contain taste pores. Filiform of surface area, depending on the size of the person, and the
papillae are not developed yet. M, mandible. Scale bar
1 mm fungiform papillae are spread unevenly over it. The number
(above), 250 m (below). of taste buds differs among fungiform papillae, and there
are large variations among human subjects in fungiform
taste bud distribution. Most papillae on the 5 mm margin of
the tongue tip are shorter than those on more posterior
form a more functional unit, since they share access to
regions. Following the surface in a posterior direction from
a common taste stimulus pool. Foliate papillae are inner-
the midline of the tip toward the back of the tongue, fungi-
vated by branches of the glossopharyngeal nerve (CN IX),
form papillae become progressively larger in size.
but the more anterior portion also receives nerve fibers
They also are larger in more posterior lingual regions.
from the chorda tympani (Oakley, 1970; Pritchard, 1991).
Small, rounded papillae are present on the margin. Some
of them contain taste buds, while others are comparable in
4. Fungiform Papillae
size to filiform papillae but lack fila. Some papillae on the
Due to their morphological heterogeneity, fungiform papil- margin of the tip are elongated like conical papillae, and
lae have been variously described as papillae clavatae, these, generally, lack taste buds. Fungiform papillae vary
capitatae, lenticulares, obtusae, majores, mediae (Haller, in size and shape: some are short and cuboidal, and others
1766; Soemmering, 1806). These papillae can be easily are tall with expanded heads like mushrooms. Among
identified as pink elevations about 0.5 mm in diameter on papillae on the margin of shorter height and smaller dia-
the anterior portion of the living human tongue. meter ( 0.5 mm), the distinction between filiform and
Notwithstanding its convenient location, the fungiform fungiform papillae becomes obscure. Fungiform papillae
taste bud population has been difficult to quantify since occasionally have projections on their apices like small fila
fungiform papillae vary in appearance and are distributed on filiform papillae.
658 Witt et al.

Taste buds have been quantified in terms of tongue sur- pharyngolaryngeal water response, possibly mediated by
face areas, referred to as “taste bud density,” or number of receptors signaling the absence of chloride ions (reviewed
taste buds per cm2 of tongue surface. There are about 145 by Lindemann, 1996). Similarly, taste buds of the larynx
gustatory (fungiform) papillae per lateral half of the seem not to play a role in gustation but detect chemicals
tongue, with about 30 papillae per cm2 on the tip, but only that are not saline-like in composition (Bradley, 2000).
about 3 papillae per cm2 on the posterolateral area. There
are about 30 large fungiform papillae in the posteromedial
IV. SALIVARY GLANDS OF THE TONGUE
region, for an estimated total of about 320 fungiform papil-
lae per tongue. On 320 fungiform (gustatory) papillae, we
“Lingua sicca non gustat (A dry tongue does not taste)”:
estimate an average of about 3.5 taste buds per papilla, for
This declaration of Haller (1766) refers to the dependence
a total of 1120 fungiform taste buds (Miller and Reedy,
of taste ability on solutions, within which tastants are dis-
1990a, 1990b) (Table 1).
solved and transported to the taste bud. Extralingual saliva
Most investigators who study human fungiform papil-
is secreted by small, mostly mucous glands embedded in
lae report the existence of papillae without taste buds,
the epithelium of the cheek and palate. More saliva is pro-
which rarely occurs in other mammals (Mistretta and
duced by the serous parotid gland and the muco-serous
Baum, 1984). Studies in humans show that fungiform
sublingual and submandibular glands (Fig. 3), whose
papillae lacking taste pores comprise 12.8–67% of the
secretory ducts open at the tongue frenulum just under-
fungiform population per subject (Arvidson and Friberg,
neath its tip. Intralingual saliva originates from the muco-
1980; Miller and Reedy, 1990a, b; Cheng and Robinson,
serous anterior lingual glands (glands of Blandin and
1991). This wide range may reflect how investigators
Nuhn) (Tandler et al., 1994) and deep posterior serous
decide which papillae are “fungiform.”
salivary glands (Ebner) located in the submucous connec-
tive tissue below the foliate and vallate papillae of the
tongue (Ebner, 1873; Riva et al., 1999). Their excretory
III. EXTRALINGUAL TASTE BUDS
ducts lead to the deepest sites of the papillar furrows (Fig.
6). The gland lobules lie deeply in large patches of con-
There are “extralingual” taste buds in regions of the
nective tissue which, in turn, are separated from each
oral, pharyngeal, and laryngeal cavities. Interestingly,
other by muscle bundles. In addition, adjacent to Ebner
Magendie (1820) and Carus (1849) erroneously associated
glands in vallate papillae lie mucous (Weber’s) glands
the teeth with taste perception. Verson (1868) described the
(Nagato et al., 1997), which in humans open into the
first “goblet-like organs” within the dorsal epithelium of
crypts of the lingual tonsils (Zimmermann, 1927). Neither
the epiglottis, and Davis (1877) and Wilson (1980)
Weber’s nor Blandin-Nuhn glands lie in close proximity
observed taste buds in and the elicitation of taste percep-
to taste buds, and their particular significance for taste
tion from the human larynx. Lalonde and Eglitis (1961)
perception is unknown because of the difficulties in col-
counted more than 2500 taste buds on the epiglottis, soft
lecting saliva from these glands (Tandler et al., 1994).
palate, laryngeal pharynx, and oral pharynx of one human
There is biochemical and histochemical evidence that
neonate. Taste buds are evident in the epiglottis of one
the saliva of Ebner’s glands, as well as that of other non-
neonatal specimen (Rabl, 1895) and esophagus in human
lingual salivary glands, has more functions than that of
fetuses (Ponzo, 1907) and adults (Schinkele, 1942; Burkl,
a serous “washing solution.” Binding proteins such as
1954). Taste buds are also found near the openings of sub-
Ebnerin (Li and Snyder, 1995) are supposed to modulate
lingual salivary gland ducts in some other primates (Hofer
sensations. Schmale et al. (1990) isolated a protein from
et al., 1979) and near the ducts of the molar glands in
rat Ebner’s glands that is structurally similar to odorant
rodents (Iida et al., 1983). In chickens and quails, taste
binding proteins in Bowman’s glands of the olfactory
buds in nonlingual parts of the oral cavity are almost
mucosa. The gland is under autonomic control (Gurkan
always associated with salivary gland ducts (Ganchrow
and Bradley, 1987). For reports on specific ligand-recep-
and Ganchrow, 1987). Miller and Smith (1984) estimate
tor interaction with taste qualities, see Azen et al., 1990;
that about 25% of the hamster’s total taste buds are
Spielman, 1990; Schmale et al., 1993; Toto et al., 1993.
extralingual, and Mistretta and Baum (1984) accounted for
a similar proportion of extralingual taste buds in the rat. It
is not known whether extralingual taste buds are function- V. BLOOD SUPPLY TO GUSTATORY PAPILLAE
ally different from those on the tongue. Taste buds of the
epiglottis and/or uvula could be involved in initiation of The mammalian tongue receives its blood supply from the
upper airway reflexes (Bradley et al., 1983) and in the lingual artery, which is usually a branch of the external
Morphology of the Peripheral Taste System 659

carotid artery (Fig. 3). Study of the tongue’s vascular sys- buds, involved in finding food; in rocklings (Gaidropsarus)
tem historically parallels that of the lingual papillae. For they are assumed to be important for predator avoidance
example, Albinus (1754), Soemmering (1806), and Arnold (Kotrschal, 1996). Since the arginine-like receptor in catfish
(1839) performed intravascular injections in order to visu- taste buds also occurs in solitary chemosensory cells, Finger
alize the papillary surface. More recently, distribution of (1997) suggests that taste buds might include solitary
the blood supply to different regions of the tongue and chemosensory cells within them. During development of
different types of lingual papillae has been described fish, solitary chemosensory cells seem to precede the devel-
by Hellekant (1976). Each type of gustatory papilla is opment of taste buds. In mammals, however, solitary
supplied by a characteristic capillary configuration (rat: chemosensory cell–like cells have been observed only
Ohshima et al., 1990; cat, rabbit: Ojima et al., 1997a,b,c), transiently, during development. In newborn rats, single
and fine capillary networks are found adjacent to taste gustducin-immunopositive cells are seen in locations where
buds. The capillary loops of larger papillae in rats and dogs later-developing vallate papillae will appear (Sbarbati et al.,
often show a constriction, maybe sphincter-like structures, 1999). Individual slender cells, immunopositive for cytoker-
but rarely arteriovenous anastomoses (Selliseth and Selvig, atin 20 (Witt et al., 1999), an intermediate filament protein
1993; Hu et al., 1996). Taste stimuli injected systemically that is exclusively present in taste bud and epidermal Merkel
elicit responses in gustatory nerves as the bolus passes cells (Moll, 1993; Zhang and Oakley 1996), are seen occa-
through the tongue (Bradley, 1973). sionally during early ontogenesis of the human tongue.
However, gustatory epithelia of adult mammals have not yet
been reported to possess solitary chemosensory cells.
VI. SOLITARY CHEMOSENSORY CELLS
VII. CELL TYPES OF VERTEBRATE TASTE
In addition to taste buds and free nerve endings, the solitary BUDS
chemosensory cells comprise another chemosensory system
in vertebrates. They are not assembled in clusters, but are Peripheral taste organs differ in number, size, and shape in
dispersed across the surface of the animal. Solitary different vertebrate taxa, according to their importance for
chemosensory cells are related to taste bud cells in the sense the particular species. For the sake of brevity, the following
that the former are secondary sensory cells with a slender, overview is restricted to some functionally well-
bipolar phenotype (Finger, 1997). “Classical” solitary characterized vertebrate species. Details are available in
chemosensory cells have been studied mostly in teleosts the review by Reutter and Witt (1993). Figure 9 shows a
(Whitear, 1992). The evolutionary benefits of these cells are scheme representing the organization of cells in fish, frog,
still in question: in sea robins (Trigon), they are, beside taste and mammalian taste buds.

Figure 9 Longitudinal sections of taste organs from representatives of three different vertebrate classes. (A) Fish (bullhead, Teleostei);
(B) amphibian (frog, Anura); (C) mammal. In this schematic drawing, each sensory epithelial cell type is represented once with a distinct
grey step. The organs lie in squamous epithelium of different height, on top of dermal papillae, which are also of different height. Each
dermal papilla contains nerve fibers and a capillary vessel.
660 Witt et al.

A. Taste Buds of Lower Vertebrates — Cell Types regarding particular phenotypes: There are only species-
specific taste bud types, and a general “model” seems
1. Fish
difficult to find among vertebrates. This underlines the
In fish, and especially in some teleosts that are well-adapted thesis that morphological phenotypes and the structural
to the dark, the taste organ is significantly more important organization of taste bud cells do not necessarily reveal
for food intake than in amphibians and mammals. Thus, a general blueprint but, rather, reflect specific environmen-
these fish, like the Siluridae, possess many more taste buds tal conditions and/or feeding behaviors (Reutter and Witt,
than representative of the latter classes (Atema, 1971; 1999). The influence of environmental differences has
Miller and Bartoshuk, 1991; Finger et al., 1996). The fish been studied in two closely related teleosts, one of which
taste bud is generally pear-shaped and similar to that of is sighted (Astyanax mexicanus), and the other of which
mammals (Fig. 9). Electron microscopic studies of teleost is a blind cave fish (Astyanax jordani). Whereas taste
taste buds (for reviews see Reutter, 1986; Jakubowski and bud morphology is rather similar, the cave fish compen-
Whitear, 1990; Reutter and Witt, 1993; Boudriot and sates for blindness by significantly more gustatory axon
Reutter, 2001) have attempted, somewhat incompletely, to profiles (Boudriot and Reutter, 2001) and an expanded
relate the ultrastructure and functions of taste bud cell expression of Prox 1 gene in developing taste buds
types. This is also apparent in the nonuniform usage of (Jeffery et al., 2000).
nomenclatures. Most authors refer to elongated “light” and
“dark” taste bud cells, as well as “intermediate” and
2. Amphibians
“degenerative” cells (Desgranges, 1965; Welsh and Storch,
1969; Reutter, 1971, 1978; Connes et al., 1988). Generally, The taste organs of postmetamorphotic Salientia
light cells are supposed to be “sensory” (receptor) cells, (
Anura), unlike piscine and mammalian taste bud bulb-
while the dark cells are regarded as “supporting” cells like formations, are relatively large disk-like epithelial dif-
(e.g., Desgranges, 1965; Hirata, 1966; Whitear, 1970). ferentiations of the dorsal lingual and palatal mucosa
However, Reutter (1992) considers the dark cells also as (Fig. 9). Waller (1847, 1849) and Engelmann (1872) called
sensory because they exhibit synaptic contacts with nerve these structures Geschmacksscheibe or “taste discs.” In
profiles or with basal cells. These observations, however, contrast to Salientia, the taste buds of the urodeles
are not supported by the work of Grover Johnson and [
Caudata, e.g., mudpuppy (Necturus), newt (Triturus), or
Farbman (1976) and Jakubowski and Whitear (1986). Axolotl (Ambystoma)] have a bulb-like shape (Fährmann,
According to these authors the differentiation “light” and 1967; Farbman and Yonkers, 1971; Cummings et al., 1987;
“dark” in combination with functional terms such as “sup- Toyoshima et al., 1987; Toyoshima and Shimamura, 1987;
porting cells” and “sensory (receptor) cells” are misleading Delay and Roper, 1988). The cellular elements of the taste
and should be avoided. According to Merkel (1880), Hirata disc, or Endscheibe (Merkel, 1880), of the frog were sub-
(1966), and Reutter (1973), the taste bud may also have ject to numerous investigations and received various desig-
mechanoreceptive functions, particularly in view of the nations: After Waller (1847, 1849) had first distinguished
morphology of basal cells (Reutter, 1971, 1986). While the between papillae conicae (
P. filiformes) and papillae
existence of synaptic contacts of light and dark taste bud fungiformes, Fixsen (1857) described two different cell
cells has not yet been finally proven, the different lengths types, the so-called “cellulae cylindricae” and “cellulae
of their microvilli perhaps suggest functional differences: fusiformes,” the processes of which pass through the whole
The light taste bud cells have a single long microvillus sensory epithelium to reach the connective tissue core of
reaching far into the mucous layer of the taste bud surface, the papilla. Engelmann (1872) and Merkel (1880) further
the receptor area. By penetrating this layer, longer developed the terminology: Merkel distinguished between
microvilli may be exposed to quite different “perireceptor “cylindrical cells” (Cylinderzellen) situated on the epithe-
events” (Getchell et al., 1984) than the small microvilli of lial surface and surrounding “wing cells” (Flügelzellen),
dark taste bud cells that do not penetrate the mucous layer the nuclei of which lie deeper in the epithelium. After
(Witt and Reutter, 1990). Graziadei and DeHan (1971) had described only two cell
Ultrastructural investigations in nonteleostean fish types (“associate cells” and “sensory cells”) in electron
reveal clearly that taste buds differ within the main verte- microscopy, the close relationship between “rod cells”
brate taxa (Reutter and Witt, 1993). Further, in different (Stäbchenzellen) and cylindrical cells (Merkel, 1880) was
systematic groups of fish, the taste buds do not follow only recently reintroduced by von Düring and Andres (1976). In
one structural design. Thus, taxon-specific taste buds or addition, the latter authors first described the basal cells and
cell types do not exist. Similar differences among mam- Merkel cells of the frog taste disc, which are the only cells
malian taste buds point to a similar, inevitable conclusion that do not contact the epithelial surface.
Morphology of the Peripheral Taste System 661

Tadpoles possess so-called premetamorphic papillae, B. Mammalian Taste Buds — Cell Types
which bear bud-like taste organs at their tops. During meta-
morphosis, these structures wholly disappear and are Early histological investigators of mammalian taste buds
replaced by fungiform papillae with large taste disks described two types of elongated, fusiform cells in taste
(Nomura et al., 1979; Zuwala and Jakubowski, 1991; buds of human vallate papillae (Schwalbe, 1868): “sup-
Zuwata, 1997). Tadpole taste discs consist of sensory and porting” and “taste cells,” the latter divided into
supporting cells, and basal cells are lacking. The taste disc Stiftchenzellen (pin cells) and Stabzellen (rod cells) with
in adult frogs contains up to 8 cell types (Reutter and Witt, differences in contrast and brightness (Fig. 10).
1993): mucus cells, wing cells, two types of sensory cells Classification into “light” and “dark” taste bud cells
(cylindrical and rod-like type), two types of basal cells [stem was also used in early electron microscopic analyses
and Merkel cell–like basal cells (Zancanaro et al., 1995)], (Engström and Rytzner, 1956). Farbman (1965) consid-
and marginal cells and ciliated cells (Toyoshima et al., 1999) ered dark, fusiform taste bud cells of human fungiform
(Fig. 9). The cell types and the history of the nomenclature papillae as sensory cells (type I), whereas other investiga-
have been described in detail (Jaeger, 1976; Reutter and tors (Paran et al., 1975) described a type II cell that con-
Witt, 1993; Witt, 1993; Osculati and Sbarbati, 1995). tains numerous vacuoles and mitochondria, especially in
Taste buds of the mudpuppy (Necturus, Urodela) are apical areas. This latter cell is not believed to be sensory.
similar to those of fishes. They are composed of dark and Cottler Fox et al. (1987) speculated that the different elec-
light cells and possess serotonergic Merkel cell–like basal tron densities are due to irreproducible fixation artifacts. In
cells, which are synaptically connected with either nerve general, ultrastructural and immunohistochemical criteria
fibers or dark and light cells (Delay and Roper, 1988; are considered more important for the classification of
Delay et al., 1993, 1994). cells than the evaluation of the electron density of the cyto-
plasm. Moreover, present understanding in mammalian
taste bud cytology leads to using a rather heterogeneous
3. Reptiles and Birds
nomenclature, based on morphological and functional dif-
Lingual taste buds in reptiles have been described in turtles ferences across species. The basis for the current nomen-
(Korte, 1980; Iwasaki et al., 1996), tortoises (Pevzner and clature was established by the Murrays and collegues
Tikhonova, 1979), and some lizards (Uchida, 1980). (Murray and Murray, 1967, 1971; Murray et al., 1969;
Lizards (Gekkonidae or Anguidae), as well as snakes, have Murray, 1986) in taste bud cells of the rabbit foliate papil-
virtually no taste buds on the tongue, but rather on the buc- lae. However, these cell types are different in some respect
cal floor and oral epithelia of the mandible and maxilla from recent classifications obtained in rodent taste buds
(Toubeau et al., 1994). In shape, reptilian taste buds resem- (Kinnamon et al., 1985, 1993, 1994; Pumplin et al., 1997).
ble those of mammals. There are up to five different types Royer and Kinnamon (1988) observed considerable devi-
of taste bud cells, classified into light and dark cells and as ations in the cytoarchitecture of murine foliate taste buds
types 1, 2, 3, A, B or as types I, II, II and basal cells compared to that of other mammals. For example, they did
(Reutter and Witt, 1993). not find type III cells, and all bud cells had synaptic
Reptiles and birds belong to the same superclass, and connections with nerve fibers. There are only a few ultra-
one might expect a similar organization of avian taste structural studies of human taste buds. Paran et al. (1975)
buds. However, the few examples of bird taste buds show and Azzali (1997) concluded that human taste bud cells are
great variability among species. Buds in grain-eating essentially similar in rabbit and dog (Kanazawa, 1993). An
birds appear in the posterior part of the tongue, near the example for a more generalized taste bud is depicted in
pharynx, and in the distal palatal mucosa (Saito, 1966; Figure 11. The classification in this chapter follows the
Ganchrow and Ganchrow, 1987; Ganchrow et al., description of rabbit taste buds (Murray, 1986; Royer and
1991; Sprissler, 1994). Unlike mammals, avian taste buds Kinnamon, 1991; Reutter and Witt, 1993).
do not reside within lingual bud-bearing papillae. In
addition, taste buds contain tubulus-like channels
1. Type I Cells
circumscribed by elongated cells grouped in a rosette
configuration, with the channel lumen continuous These cells are the most frequent. They are spindle-shaped
apically with the taste pore (Berkhoudt, 1985; Ganchrow and have a basal process that envelops the axons in a
et al., 1993). Taste buds are richly innervated, and Schwann cell–like manner. Type I cells protrude with
synapses are seen between all cell types (light cells and brush-like, long microvilli (1–2 m) into the taste pit.
dark cells) and nerve fibers (Reutter and Witt, 1993; These cells ensheath type II and III cells with cellular pro-
Sprissler and Reutter, 1993; Sprissler, 1994). trusions and may insulate them. Apically, they contain
662 Witt et al.

Figure 10 Schwalbe’s (1868) taste bud description (from vallate papilla of the swine). Minor bundles and fibrils of nerves are lost in
the “interior of taste goblets.” Note the incorrect depiction of basal cells (left side). Isolated cells of the human taste goblets, one of which
(4) shows a large microvillus. 1
Supporting cell (Deckzelle, mostly at the margin); 2, 4
pin cell (Stiftchenzelle); 3
rod cell
(Stabzelle). According to Schwalbe, both cell types might mediate different taste sensations.

large granules, 100–400 nm in diameter (Murray, 1986). nerve fibers (Kinnamon et al., 1985, 1988; Royer and
The nuclei of type I cells are irregularly shaped. According Kinnamon, 1988).
to Murray and Murray (1971) and Murray (1986), these
cells have a secretory (supporting cells) and possibly
3. Type III Cells
phagocytotic function and probably produce the amor-
phous material of the taste pit (e.g., Farbman, 1965; Taste buds contain only 5–7% of type III cells. They have
Menco, 1989; Ohmura et al., 1989; Witt, 1996). unbranched basal and apical processes. Their apical por-
tion protrudes with a single, large microvillus into the taste
2. Type II Cells pit and may reach the taste pore. Type III cells are pre-
sumably the only cells that have synaptic contacts with
Most of these cells are located in the periphery of the taste intragemmal nerve fibers. Near the cell nucleus there are
bud. They are fusiform, but do not possess enveloping numerous serotonin-containing, dense-cored vesicles
processes, or granules. Their cytoplasm is moderately elec- (80–140 nm in diameter) (Nada and Hirata, 1977;
tron-dense, and nuclei are round to oval. Synapses are not Fujimoto et al., 1987). Proteins [VIP (Herness, 1989), spot
observed. Toyoshima and Tandler (1987) describe a modi- 35 protein, and calbindin (Yoshie et al., 1991; Johnson
fied endoplasmic reticulum with specialized subsurface et al., 1992)] were also demonstrated in these cells.
cisterns adjacent nerve profiles. The function of these cells Therefore, these cells are considered by most investigators
may be secretory or chemosensory, but synapses in mon- as gustatory sensory cells.
keys are not evident (Farbman et al., 1985, 1987). In mice,
vallate and foliate “light” (type II) and “dark” (type I and
4. Type IV Cells
type II) taste bud cells exhibit synaptic contacts with nerve
fibers, thus suggesting a gustatory function. However, type Type IV cells (Murray et al., 1969; Murray, 1973) include
I and type II cells do not form synapses with the same basal cells (Nemetschek-Ganssler and Ferner, 1964) or
Morphology of the Peripheral Taste System 663

Figure 11 Mammalian taste bud in longitudinal sec-

tion, idealized schematic drawing according to electron
microscopic findings. Each cell type is depicted once.
Following Murray’s nomenclature, cells of type I, II,
and III are elongated and form the buds sensory epithe-
lium proper. Apically, these cells end with different
types of microvilli within the taste pit and may reach
the taste pore. Type IV cells are basal cells, type V mar-
ginal cells. Synapses are found especially at the bases
of type III cells. Nerve fibers within the short dermal
papilla are slightly myelinated, and within the taste bud
they form an unmyelinated plexus. Note the basal lam-
ina between dermis (which contains a capillary) and the
epithelium.

“pregustatory cells” (Scalzi, 1967). These are relatively and may possibly be taste bud stem cells (Beidler and
small undifferentiated cells that lie at the taste bud’s base, Smallman, 1965; Farbman, 1980) which express partic-
which do not form processes that reach the pore. They ular non–taste receptor proteins, e.g., CD44 isoforms
contain numerous bundles of intermediate filaments (Witt and Kasper, 1998), during human taste bud onto-
(Royer and Kinnamon, 1991) and differ from Merkel genesis.
cell–like basal cells of fishes and amphibians. Type IV
cells are considered to be undifferentiated stem cells of
C. Molecular Markers of Taste Bud Cells — Basis
their bud cell progeny (Murray, 1973, 1986; Roper, 1989).
for a New Classification?
Some authors report on “intermediate cells” (e.g.,
Kinnamon et al., 1985). It has been pointed out by Farbman
One of the most intriguing challenges for suggesting pos-
et al. (1985) that differences in the electron density of the
sible functional properties of taste bud cells is to identify
cytoplasm could also reflect different stages in the maturation
subsets of these cells by morphological features as well as
of the same cell type indicating different states of function.
molecular properties, many of which can be traced even in
improved primary taste bud cell cultures (Kishi et al.,
5. Type V Cells
2001). Histochemical evidence on the neurochemical
“Marginal cells” (also “perigemmal cells” and, in exten- nature of taste cells [see the concept of paraneuron by
sion of Murray’s nomenclature, “type V cells”) have Fujita (1994)] have identified the panneuronal markers,
been described (Beidler and Smallman, 1965; Farbman, neuron-specific enolase (NSE) (see Fig. 5), and protein
1980; Gurkan and Bradley, 1987; Reutter and Witt, gene product 9.5 (PGP 9.5) (Yoshie et al., 1988; Montavon
1993). However, they have nothing in common with the et al., 1996; Astbäck et al., 1997), carbohydrate-binding
secretory marginal cells of taste organs in fish and frog proteins, the lectins (Witt and Reutter, 1988; Witt and
664 Witt et al.

Miller, 1992), and members of the intermediate filament biological approaches, e.g., introduction of green fluores-
family, e.g., cytokeratins 18–20 and vimentin (Zeng et al., cent protein chimeras in vitro (Landin and Chaudhari,
1995; Zhang et al., 1995; Zhang and Oakley, 1996; Witt 2000) or calcium imaging after application of specific
et al., 1999; Witt and Kasper, 1999), as well as microfila- neuropeptidergic stimuli (Lu et al., 2000) might contribute
ments (Höfer and Drenckhahn, 1999). However, the rela- to a solution of this problem.
tion between molecular marker and taste bud cell type has Several authors report on morphological and immuno-
not been firmly established ultrastructurally. histochemical differences between vallate/foliate and
Immunoelectron microscopic studies have tried to fungiform taste buds within the same species. For exam-
match functional parameters with those of conventional ple, mouse taste bud cells of fungiform papillae contain
electron density. For example, cell adhesion molecules more synapses and presynaptic vesicles than those of val-
(Nolte and Martini, 1992; Smith et al., 1993, 1994) and late papillae (Kinnamon et al., 1993), and the number of
several blood group antigens (Pumplin et al., 1997, 1999; taste bud cells containing group H blood antigen and gust-
Smith et al., 1999) characterize subsets of taste bud cells. ducin is three times higher in vallate than in fungiform
Dark cells of rat vallate taste buds have been associated papillae (Smith et al., 1993). In rabbit, lectin carbohydrate
with H blood group antigen (2B8 epitope) expression, profiles of both taste bud populations differ as well (Witt
and light cells with the Lewis b antigen (Pumplin et al., and Miller, 1992). The reasons for and significance of
1997). These authors conclude that light cells in rat these differences between fungiform and vallate/foliate
vallate taste buds have at least two molecular phenotypes, taste cells are not clear, but factors determining their
namely cells with Lewis b and those without. A subset of varying phenotype could include a different local saliva
the light cell phenotype contains partly the G protein composition (Shatzman and Henkin, 1981; Schmale et al.,
gustducin (Ruiz-Avila et al., 1995; Menco et al., 1997), 1990; Schmale and Bamberger, 1997) or morphogenetic
which is involved in the perception of sweet and bitter conditions of local epithelium (Smith et al., 1999).
taste (Wong et al., 1996). Evidence for communication between taste cells
Choline acetyl transferase, an enzyme involved in the includes the presence of cell adhesion molecules (Nolte
synthesis of the neurotransmitter acetylcholine, has been and Martini, 1992; Smith et al., 1993), heparin-binding
identified in rat type II cells (Menco et al., 1997). Whereas proteins (Wakisaka et al., 1998), and membrane receptors
the putative neurotransmitter serotonin is confined to basal that influence the intracellular signal transduction cas-
cells of fish taste buds (Reutter, 1971) and Merkel cell–like cades. For example, the hyaluronan receptor, CD44, was
basal cells of amphibian taste organs (Toyoshima and identified in a subset of human fetal taste bud cells
Shimamura, 1987; Delay et al., 1997; Hamasaki et al., (marginal cells) and most of adult human taste bud cells
1998), serotonin in mammals has been described in type (Witt and Kasper, 1998). This transmembrane protein is
III cells in the rabbit (Fujimoto et al., 1987; Kim and linked to a series of actin-associated microfilaments, e.g.,
Roper, 1995) and human taste buds (Azzali, 1997). This ezrin and ankyrin, which are located in microvilli of type I
led to the hypothesis that these cell types were equivalent cells and might influence the function of ion-translocating
in both taxa (Kim and Roper, 1995). Lindemann (1996) membrane proteins (Höfer and Drenckhahn, 1999). In
suggests the term “serotonergic cells” instead of type III light of efferent neural control, taste bud cell communica-
cells (10% of all cells). tion may be mediated via local axon reflexes between sen-
Generally, neuropeptides are located in intragemmal sory cells (Caicedo et al., 2000). The leaf-like enclosing
nerve fibers rather than in particular bud cells. An excep- morphology of type I (dark) cells suggests an insulating,
tion is vasointestinal peptide (VIP) that has been detected glia-like function.
in a subset of rat taste bud cells (Herness, 1989) and in
light taste bud cells in the carp by electron microscopy
(Witt, 1995). Though most of these markers are expressed VIII. DEVELOPMENT OF THE HUMAN
only in differentiated cells and are not evident after nerve PERIPHERAL TASTE SYSTEM
dissection (Smith et al., 1993; Whitehead et al., 1998),
their functional correlation with taste perception data is Morphogenesis of the mouth cavity is characterized by the
mostly unknown. development of the tongue anlage, which appears prior to,
To avoid the present Babylonian confusion of tongues and is a prerequisite of, the formation of gustatory papillae.
with regard to taste bud cell nomenclature, future research At the embryonic age of 4 weeks, the first structure of
directions should try to associate electrophysiologically the tongue anlage to appear is the tuberculum impar, which
characterized, isolated taste bud cells with a particular cell is situated between the first (mandibular) and second
type based on its specific protein expression. Modern cell (hyoid) branchial arches (Fig. 12). Then, anterolateral to
Morphology of the Peripheral Taste System 665

Figure 12 Development of the human tongue, 5th postovulatory week. The schematic drawing was done by compiling Hinrichsen’s
(1990) and our own data. By horizontal section the floor of the forecoming buccal cavity of a human embryo is removed and viewed from
dorsally. The floor relief is derived from the branchial arches I (mandibular arch), II (hyoidal arch), III (3rd pharyngeal arch), and IV (4th
pharyngeal arch), and by their derivatives which are: 1, the paired lingual swellings; 2, the impar tubercle; and 4, the hypobranchial emi-
nence. These three structures form the tongue anlage. Between 2 and 4, the anlage of the thyroid gland invaginates, and as its remnant
the 3 (foramen caecum) is left; 5 is the anlage of the epiglottis. The branchial arches, as well as the tongue anlage, are innervated by the
cranial nerves CN V3 (mandibular nerve), CN VII (facial nerve), CN IX (glossopharyngeal nerve), and CN X (vagal nerve). CN XII
(hypoglossal nerve) invades the tongue anlage as well and innervates its muscular system. Later, the lingual nerve (from CN V3) and the
chorda tympani (running with CN VII) join each other and supply the anterior two thirds of the tongue with somatosensory and gusta-
tory nerve fibers, whereas CN IX and CN X carry taste fibers for the posterior third of the tongue, the epiglottis, and the pharynx.

the tuberculum impar, the paired lingual swellings (which buds. It was unclear to Hermann if supporting, or neuroep-
derive from the medial parts of the mandibular arches) fuse ithelial, cells were being replaced.
with the tuberculum impar. The tongue’s base is formed by Vallate papillae start to develop earlier than fungiform
the hypobranchial eminence (copula of His) forming papillae, and begin with the appearance of a central mid-
within the third and fourth branchial arches. The border line papilla just behind the foramen caecum around the
between the caudal part and the body of the tongue is 6th postovulatory week. From week 7 on, there develop
demarcated by a V-shaped rim, the sulcus terminalis many hillock-like epithelial elevations on the tongue’s
(Bradley, 1972; Witt and Reutter, 1997). The innervation dorsum, as seen with scanning electron microscopy
pattern of cranial nerves, which later supplies particular (Fig. 13). Some of these elevations are precursors of
lingual regions, reflects the early innervation of branchial fungiform papillae and are especially densely distributed
arches (see Fig. 12). The pretrematic nerve of the first near the midline and the lateral ridges of the tongue
branchial arch is the lingual nerve (from cranial nerve V3); (Habermehl, 1952; Hersch and Ganchrow, 1980; Witt and
that of the second arch, the chorda tympani (from the inter- Reutter, 1997). Analysis of serial sections of the tongue
medio-facial nerve); and that of the third arch constitutes encompassing this critical developmental age (weeks 6–8)
later lingual rami of the glossopharyngeal nerve (for demonstrates that not every dermal elevation will be the
details, see textbooks on anatomy and embryology; e.g., target of nerve fibers. First, around week 7–8, nerve
Williams et al., 1989; Hinrichsen, 1990). fibers, migrating towards the periphery, form a large
The first detailed developmental studies on the surface intragemmal plexus. Our own studies show that there are
appearance of the tongue were carried out by Froriep no taste bud anlagen without approaching nerve fibers
(1828) and continued histologically by Tuckerman (1889), (Witt and Reutter, 1996; Witt and Kasper, 1998). Also,
Gråberg (1898), and Hellman (1922). Hermann (1885) taste bud primordia without dermal papillae are evident,
described the stages of karyokinesis in developing taste as well as individual bipolar epithelial cells resembling
666 Witt et al.

At their apical ends, taste bud cells cannot be distin-

guished by their electron density (week 15) (Fig. 15).
There are cells with long, slender microvilli, but, in
contrast to adult taste buds (Azzali, 1997), there are no
type I cells with typical dark granules believed to secrete
the mucous material that fills the taste pit (Witt and
Reutter, 1996).

IX. INNERVATION OF THE HUMAN TONGUE

AND TASTE BUDS

The tongue is innervated by (1) motor nerve fibers, consti-

tuting the hypoglossal nerve (CN XII), which supplies the
inner (intrinsic) and the hyoidal tongue muscles; (2)
somatosensory nerve fibers, composed of divisions of the
trigeminal (CN V3) and glossopharnygeal (CN IX) nerves;
(3) autonomic nerve fibers, which stem from the interme-
dio-facial nerve (CN VII), the glossopharyngeal nerve, and
the vagus nerve (CN X); and (4) sensory (gustatory) nerve
fibers that transmit taste information centrally from taste
buds, namely (1) the chorda tympani, a branch of the inter-
Figure 13 Scanning electron micrograph of a human embry- mediate nerve (generally considered a part of the facial
onic tongue, 7th postovulatory week. Fine dots in the dorsal nerve; CN VII), (2) glossopharyngeal nerve (CN IX), and
surface demarcate later fungiform papillae. Anlagen of vallate (3) vagus nerve (CN X) (summarized in Fig. 16).
papillae (arrows) lie in front of the sulcus terminalis. Short arrow Classical research papers of the nineteenth century
indicates the median vallate papillae, which originates first. Scale
explored the clinical consequences in taste perception
bar
0.5 mm.
associated with diagnostic features of the cranial nerves.
These papers have formed the bases for the neurological
examination. Two prominent themes were explored: (1)
solitary chemosensory cells. These individual cells are Which cranial nerves are associated with functional attri-
immunopositive for cytokeratin 20, a marker for lingual butes of the taste system? (2) How is taste perception
taste bud cells (Zhang and Oakley, 1996; Witt and Kasper, affected by neurological diseases? In physiological
1999) (see above). Temporal correlation, which would experiments, Magendie (1820) dissected the lingual nerve
suggest dependence of taste bud development on nerve of living, unanaesthetized animals and observed loss of
ingrowth, has not as yet been seen. taste, but intact movement and sensation of palate, gin-
Lingual taste bud primordia first occur around the 7th giva, and buccal mucosa. Panizza (1834) performed dis-
and 8th postovulatory weeks (Bradley and Stern, 1967; section of the glossopharyngeal nerve, resulting in loss of
Bradley, 1972). Taste pores, commonly acknowledged as taste. Alcock (1836) described the role of the chorda tym-
a sign of taste bud maturity, appear between the 10th and pani and the sphenopalatine ganglion. Lussana (1869,
14th weeks. The presence of a taste pore is not always 1872) traced the target tissue of the chorda tympani nerve
associated with a fully mature taste bud because the bot- to the anterior two thirds of the tongue. The dependence
toms of early taste pits may be covered by flat epithelial of taste buds on nerve supply was experimentally shown
cells (Witt and Reutter, 1997). However, transmission by von Vintschgau and Hönigschmied (1877): 40 days
electron microscopic studies show that early taste bud after dissection of the glossopharyngeal nerve, the number
primordia (week 8) synaptically contact nerve fibers, of taste buds dramatically decreased. In similar experi-
suggesting that the potential for neurotransmission ments, Ranvier (1888) observed that taste bud sensory
precedes the exposure of sapid molecules to the apical cells degenerate, and supporting cells pushed through the
surface of the taste bud cell. At its base, the developing pore to the tongue surface. During this process, he
human taste bud (weeks 12–15) contains processes of observed cellules migratrices [phagocytic fibroblasts?
dark and light cells, as well as processes resembling type (Suzuki et al., 1996)] loaded with fed particles, which
III cells (exhibiting synapses with nerve fibers) (Fig. 14). were “probably responsible for removal of old materia”.
Morphology of the Peripheral Taste System 667

Figure 14 Basal portion of a developing human taste bud, week 15. Light cells (L), nerve fiber profiles (N), and dark cells (D) are the
prominent structures at the base. Type III-like cells contain dense-cored vesicles (arrows). BL
Basal lamina; S
Schwann cell; scale
bar
1 m. (From Witt and Reutter, 1996.)

Soon it became evident that experiments based on vivi- observations from the second half of the century, particu-
section were not only painful for the animal (mostly dogs larly those derived from electrophysiological studies.
or cats), but also unreliable in their results (Alcock, 1836; 1. Chorda tympani and greater petrosal nerve: Distal to
Wagner, 1837). Nevertheless, the overall conclusions the intermediate nerve branch of the facial nerve,
derived from these nineteenth-century nerve dissection peripheral axons of some geniculate ganglion somata
experiments (reviewed by Parker, 1922; Jägel, 1991) can- (the chorda tympani nerve) take a recurrent course
not deny their import for current knowledge of cranial within the facial canal in the petrosal part of the tem-
nerve supply and taste sensitivity. poral bone, pass through the middle ear, and exit the
An elegant review of the background of peripheral taste skull via the petrotympanic fissure to join the lingual
pathways in humans was written by Lewis and Dandy division of the trigeminal nerve, the lingual nerve.
(1930). They examined both the neurological and neu- Both intermedio-facial (gustatory) and trigeminal
roanatomical literature on gustatory pathways. The sensory (somatosensory) fibers run in the lingual nerve and
distribution of the facial nerve and its clinical importance distribute to the fungiform papillae on the anterior two
was described by Hunt (1915). He disentangled the over- thirds of the tongue and may reach also the anterior
lapping sensory fields of the facial nerve (including the portion of the foliate papillae. Taste buds on the soft
chorda tympani) from the trigeminal nerve by documenting palate are innervated by the greater petrosal branch of
the distribution of herpes zoster inflammation. The herpetic the intermedio-facial nerve, whose somata also lie
eruptions outlined the sensory fields of the geniculate gan- within the geniculate ganglion (Harris, 1952; Miller
glion on the tongue, soft palate, and ear. Another basis for and Spangler, 1982). Some chorda tympani fibers are
evaluating the involvement of the chorda tympani nerve reported to anastomose with the greater petrosal nerve
with lingual taste buds came from patients who had under- via the otic ganglion (Schwartz and Weddell, 1938;
gone middle ear surgery (Bull, 1965; Borg et al., 1967). Pritchard, 1991). Both the greater petrosal and chorda
Contemporary reviews of human (Norgren, 1990) and tympani nerves also carry parasympathetic fibers to
primate (Pritchard, 1991) taste pathways have incorporated their associated salivary glands: the greater petrosal
668 Witt et al.

Figure 15 Transmission electron micrograph of a human fungiform taste bud during development (week 15). The taste pore (TP) is
already open, and some elongated cells stick into the taste pit. However, differences between cell types in the apical portion of the taste
bud cannot be made yet. There is no mucus in the taste pit. Approximately two thirds of the taste bud are filled with ramifications of nerve
fibers (N). A part of the basal portion is outlined with a rectangle (see Fig. 14). MC
marginal cell; scale bar
10 m. (From Witt and
Reutter, 1996, with permission of John Wiley and Sons.)
Morphology of the Peripheral Taste System 669

Figure 16 Innervation of the human tongue and the taste bud–bearing epithelia (hatched regions), compiled by figures from Feneis (1985)
and Sobotta (1993). The cranial nerves CN VII (which includes the intermediate nerve with its branches, greater petrosal nerve and chorda
tympani, black), CN IX, and CN X contain sensory gustatory fibers. CN V is the trigeminal nerve with its divisions V1, ophthalmic, V2,
maxillary, and V3, mandibular nerves. CN XII, hypoglossal nerve, is the motor nerve of the tongue. gG, Geniculate ganglion; pG, ptery-
gopalatine ganglion; iG, inferior ganglion of the glossopharyngeal (CN IX) and vagal (CN X) nerves, respectively; sG, submandibular gan-
glion with postganglionic autonomic nerve fibers of the chorda tympani to supply the submandibular and sublingual glands.

nerve serves the palatine glands, while the chorda Probably, the glossopharyngeal nerve also supplies
tympani innervates the submandibular and sublingual taste buds in the pharynx.
glands via the submandibular ganglion. 3. Vagus nerve: Taste buds on the laryngeal surface of
2. Glossopharyngeal nerve: Axons of the glossopharyn- the epiglottis, larynx, and proximal portion of the
geal nerve originate from ganglion cells mainly in the esophagus are innervated by the superior laryngeal
inferior (petrosal) glossopharyngeal ganglion. These branch of the vagus nerve, which has the perikarya of
peripheral axons supply both taste buds and general its chemosensory neurons in the inferior (nodose)
sensory innervation to the vallate and foliate papillae. vagal ganglion.
Salivary glands (Ebner) are supplied by parasympa- 4. Trigeminal nerve: The possible role of trigeminal
thetic fibers via an intrinsic ganglion (Remak, 1852). nerve fibers in taste perception has been discussed for
670 Witt et al.

two millenia, and there is no end in sight. Via the lin- Thus, nerve fibers are required to maintain taste buds
gual nerve, this nerve conveys somatosensory infor- once the latter are formed and start to function (e.g., Hosley
mation from the tongue to the trigeminal ganglion et al., 1987), but controversy exists whether nerves are
(Gasseri). In fact, most of the nerve fibers entering the necessary to initiate taste bud development. At present,
fungiform papillae are trigeminal, while a few fibers there are two main doctrines concerning this issue: (1) the
originate from the chorda tympani (25% in rat) presence of nerve fibers, if not the synaptic contact to local
(Farbman and Hellekant, 1978). epithelial cells, is a prerequisite for the neurosensory
Palatal trigeminal fibers may respond to sapid stim- transformation of epithelial cells (Witt and Reutter, 1996;
uli, as revealed by trigeminal transection experiments Oakley, 1998a,b; Oakley et al., 1998); (2) initial taste bud
(Berridge and Fentress, 1985) and electrophysiology development is nerve-independent, suggested by a series of
(Harada and Smith, 1992). Finally, taste qualities may studies in salamanders (Stone, 1940; Barlow et al., 1996;
be influenced by nonsapid stimuli, e.g., temperature: Barlow and Northcutt, 1998a,b). Taste buds seem to
approximately one-half of the nerve fibers involved in develop from local epithelium and not from neurogenic
taste transduction respond to temperature (Cruz and ectoderm [axolotl: Barlow and Northcutt (1997), mouse:
Green, 2000). The interaction of both gustatory and Stone et al. (1995)]. It may be that mechanisms of differen-
somatosensory qualities may be as tight-knit as their tiation of the same receptor organ vary among vertebrate
anatomical proximity. Katz et al. (2000) suggest that taxa. Growth factors other than BDNF may contribute to
gustation should be thought of as an integral part of a the maintenance of gustatory papillae, e.g., epidermal
distributed, interacting multimodal system. growth factor (EGF) supplied by salivary glands (Morris-
Wiman et al., 2000). More detailed studies on developmen-
The observation that taste buds degenerate after dissec- tal aspects of the peripheral gustatory system including
tion of their sensory innervation and, subsequently, reap- whether taste buds may develop without the stimulation of
pear after regeneration of their peripheral nerves has been nerves are described later in this book (see Chapter 36).
a major focus of research in the peripheral taste system.
Nineteenth- and early twentieth-century literature on taste
bud degeneration, regeneration, and development was ACKNOWLEDGMENTS
reviewed comprehensively by Parker (1922). Olmsted
(1920) proposed that trophic maintenance of fish taste The authors are indebted to Mihnea Nicolescu, who pro-
buds depended on the transmission of a putative trophic vided the schematic drawings. Drs. Judith and Donald
material from nerve to epithelium. Cross reinnervation of Ganchrow helped with critical reading of the manuscript
the glossopharyngeal nerve to fungiform taste buds (which and contributed thereby to avoid mistakes, redundancies,
are normally supplied by the chorda tympani) had no effect and senseless phrases.
on the usual immunohistochemical properties of fungiform
versus vallate taste buds (Smith et al., 1999). As a conse-
REFERENCES
quence, these authors believe that the protein expression in
and subsequent function of taste buds depend on the
Albinus. (1754). Academicarum annotationum liber 1, Tab.I.
epithelium from which the cells arise, and not the buds Lugdunum Bataviensis (Leiden).
specific nerve supply. Meanwhile, Nosrat and Olson Alcock, B. (1836). Determination of the question, which are the
(1995) and Nosrat et al. (2000) detected mRNA of brain- nerves of taste. Dublin J. Med. Chem. Sci. 10:256–279.
derived neurotrophic factor (BDNF) and neurotrophins in Arey, L., Tremaine, M., and Monzingo, F. (1935). The numerical
developing anterior tongue epithelium, before nerve fibers and topographical relation of taste buds to human circumval-
were observed. This argues for the hypothesis that trophic late papillae throughout the life span. Anat. Rec. 64:9–25.
factors act as target-derived chemoattractants for the early Arnold, F. (1839). Tabulae Anatomicae. Icones organum sen-
nerve fibers. These, in turn, initiate the formation of taste suum. Organon gustus. Turici, Zürich.
buds. BDNF-null mutant mice failed to develop taste buds Arvidson, K., and Friberg, U. (1980). Human taste: response and
taste bud number in fungiform papillae. Science
(Oakley et al., 1998). Sensory ganglia involved in taste bud
209:807–808.
innervation (see above) are reduced by 40% in volume Astbäck, J., Arvidson, K., and Johansson, O. (1997). An immuno-
compared to about 20% of trigeminal ganglion under the histochemical screening of neurochemical markers in fungi-
same condition (Mistretta et al., 1999). Taste buds do not form papillae and taste buds of the anterior rat tongue. Arch.
develop after injection of -bungarotoxin into the amniotic Oral Biol. 42:137–147.
fluid in fetal mice. This neurotoxin abolishes motor and Atema, J. (1971). Structures and functions of the sense of taste in
sensory nerve development (Morris-Wiman et al., 1999). the catfish (Ictalurus natalis). Brain Behav. Evol. 4:273–294.
Morphology of the Peripheral Taste System 671

Azen, E. A., Hellekant, G., Sabatini, L. M., and Warner, T. F. Bradley, R. M., Stedman, H. M., and Mistretta, C. M. (1983).
(1990). mRNAs for PRPs, statherin, and histatins in von Superior laryngeal nerve response patterns to chemical stimu-
Ebner’s gland tissues. J. Dent. Res. 69:1724–1730. lation of sheep epiglottis. Brain Res. 276:81–93.
Azzali, G. (1997). Ultrastructure and immunocytochemistry of Braus, H. (1940). Handbuch der mikroskopischen Anatomie des
gustatory cells in man. Anat. Anz. 179:37–44. Menschen. Springer, Berlin.
Azzali, G., Gennari, P. U., Maffei, G., and Ferri, T. (1996). Bull, T. R. (1965). Taste and the chorda tympani. J. Laryngol.
Vallate, foliate and fungiform human papillae gustatory cells. Otol. 79:479–493.
An immunocytochemical and ultrastructural study. Minerva Burkl, W. (1954). über das Vorkommen von Geschmacksknospen
Stomatol. 45:363–379. im mittleren Drittel des Oesophagus. Anat. Anz. 100:320–321.
Barlow, L. A., and Northcutt, R. G. (1997). Taste buds develop Caicedo, A., Kim, K. N., and Roper, S. D. (2000). Glutamate-
autonomously from endoderm without induction by cephalic induced cobalt uptake reveals non-NMDA receptors in rat
neural crest or paraxial mesoderm. Development taste cells. J. Comp. Neurol. 417:315–324.
124:949–957. Carus, C. G. (1849). System der Physiologie. Brockhaus,
Barlow, L. A., and Northcutt, R. G. (1998a). The role of innerva- Leipzig.
tion in the development of taste buds: insights from studies of Casserius. (1609). Penthaesteseion (cited after Jurisch, 1922).
amphibian embryos. Ann. NY Acad. Sci. 855:58–69. Cheng, L. H., and Robinson, P. P. (1991). The distribution of
Barlow, L. A., and Northcutt, R. G. (1998b). Vertebrate taste-bud fungiform papillae and taste buds on the human tongue. Arch.
development: Are salamanders the model? Reply. Trends Oral Biol. 36:583–589.
Neurosci. 21:338–338. Connes, R., Granie-Prie, M., Diaz, J. P., and Paris, J. (1988).
Barlow, L. A., Chien, C. B., and Northcutt, R. G. (1996). Ultrastructure des bourgeons du gout du téléostéen marin
Embryonic taste buds develop in the absence of innervation. Dicentrarchus labrax L. Can. J. Zool. 66:2133–2142.
Development 122:1103–1111. Cottler Fox, M., Arvidson, K., Hammarlund, E., and Friberg, U.
Beidler, L. M., and Smallman, R. L. (1965). Renewal of cells (1987). Fixation and occurrence of dark and light cells in taste
within taste buds. J. Cell Biol. 27:263–272. buds of fungiform papillae. Scand. J. Dent. Res. 95:417–427.
Bellini, L. (1665). Gustus organum novissime deprehensum prae- Cruz, A., and Green, B. G. (2000). Thermal stimulation of taste.
missis ad faciliorem intelligentiam quibusdam de saporibus Nature 403:889–892.
[Taste organs newly observed; with certain premises about the Cummings, T. A., Delay, R. J., and Roper, S. D. (1987).
senses given for easier understanding (translation from Latin Ultrastructure of apical specializations of taste cells in the
into German by Jurisch (1922))]. Mangetus Bibliotheca anat. mudpuppy, Necturus maculosus. J. Comp. Neurol.
2, Bologna. 261:604–615.
Berkhoudt, H. (1985). Special sense organs: Structure and func- Davis, C. (1877). Die becherförmigen Organe des Kehlkopfs.
tion of avian taste receptors. In: Form and function in birds, Arch. Mikr. Anat. 14:158–167.
Vol. 3, King, A. S., and McIelland, J. (Eds.). Academic Press, Delay, R. J., and Roper, S. D. (1988). Ultrastructure of taste cells
New York, pp. 463–496. and synapses in the mudpuppy Necturus maculosus. J. Comp.
Berridge, K. C., and Fentress, J. C. (1985). Trigeminal-taste Neurol. 277:268–280.
interaction in palatability processing. Science 228:747–750. Delay, R. J., Taylor, R., and Roper, S. D. (1993). Merkel-like
Borg, G., Diamant, H., Oakley, B., Ström, L., and Zotterman, Y. basal cells in Necturus taste buds contain serotonin. J. Comp.
(1967). A comparative study of neural and psychophysical Neurol. 335:606–613.
responses to gustatory stimuli. In Olfaction and Taste 2, Delay, R. J., Mackay Sim, A., and Roper, S. D. (1994).
Hayashi, T. (Ed.). Pergamon Press, Oxford, pp. 253–265. Membrane properties of two types of basal cells in Necturus
Boudriot, F., and Reutter, K. (2001). Ultrastructure of the taste buds. J. Neurosci. 14:6132–6143.
taste buds in the blind cave fish Astyanax jordani Delay, R. J., Kinnamon, S. C., and Roper, S. D. (1997). Serotonin
(“Anoptichthys”) and the sighted river fish Astyanax mexi- modulates voltage-dependent calcium current in Necturus
canus (Teleostei, Characidae). J. Comp. Neurol. 434: taste cells. J. Neurophysiol. 77:2515–2524.
428–444. Desgranges, J. C. (1965). Sur l’existence de plusieurs types de
Bradley, R. M. (1972). Development of the taste bud and gusta- cellules sensorielles dans les bourgeons du gout des barbil-
tory papillae in human fetuses. In Third Symposium on Oral lons du Poisson-chat. C. R. Acad. Sci. (Paris) [D] 261:
Sensation and Perception, Bosma, J. F. (Ed.). Charles 1095–1098.
C Thomas, Springfield, IL, pp. 137–162. Ebner, V. (1873). Die acinösen Drüsen der Zunge und ihre
Bradley, R. M. (1973). Electrophysiological investigations of Beziehungen zu den Geschmacksorganen. Leuschner &
intravascular taste using perfused rat tongue. Am. J. Physiol. Lubensky, Graz.
224:300–304. Ehrenberg, C. G. (1833). Notwendigkeit einer feineren mecha-
Bradley, R. M. (2000). Sensory receptors of the larynx. Am. nischen Zerlegung des Gehirns und der Nerven vor der
J. Med. 108 (Suppl 4a):47S–50S. chemischen, dargestellt an Beobachtungen von C. G.
Bradley, R. M., and Stern, I. B. (1967). The development of the Ehrenberg. Poggendorfs Annl. Physik Chemie XXVIII: 450.
human taste bud during the foetal period. J. Anat. Ehrlich, P. (1886). Über die Methylenblaureaktion der lebenden
101:743–752. Nervensubstanz. Deutsche Med. Wochenschr. 12:49–52.
672 Witt et al.

Engelmann, T. W. (1872). Die Geschmacksorgane. In Handbuch Getchell, T. V., Margolis, F. L., and Getchell, M. L. (1984).
der Lehre von den Geweben des Menschen und der Thiere, Perireceptor and receptor events in vertebrate olfaction. Prog.
Vol. 2, Stricker, S. (Ed.). Engelmann, Leipzig, pp. 822–838. Neurobiol. 23:317–345.
Engström, H., and Rytzner, C. (1956). The fine structure of taste Gråberg, J. (1898). Beiträge zur Genese des Geschmacksorgans
buds and taste fibers. Ann. Otol. 65:361–375. des Menschen. Morphol. Arb. 8:117–134.
Fährmann, W. (1967). [Light and electron microscopical studies Graziadei, P. P., and DeHan, R. S. (1971). The ultrastructure of
on the taste bud of neotene axolotls (Siredon mexicanum frogs’ taste organs. Acta Anat. (Basel) 80:563–603.
Shaw)]. Z. Mikrosk. Anat. Forsch. 77:117–152. Grover Johnson, N., and Farbman, A. I. (1976). Fine structure of
Farbman, A. I. (1965). Electron microscope study of the devel- taste buds in the barbel of the catfish, Ictalurus punctatus.
oping taste bud in rat fungiform papilla. Dev. biol. Cell Tissue Res. 169:395–403.
11:110–135. Gurkan, S., and Bradley, R. M. (1987). Autonomic control of von
Farbman, A. I. (1980). Renewal of taste bud cells in rat circum- Ebner’s lingual salivary glands and implications for taste sen-
vallate papillae. Cell Tissue Kinet. 13:349–357. sation. Brain Res. 419:287–293.
Farbman, A. I., and Hellekant, G. (1978). Quantitative analyses Habermehl, K. H. (1952). Über besondere Randpapillen an der
of the fiber population in rat chorda tympani nerves and fungi- Zunge neugeborener Säugetiere. Z. Anat. Entwicklgesch.
form papillae. Am. J. Anat. 153:509–521. 116:355–372.
Farbman, A. I., and Yonkers, J. D. (1971). Fine structure of the Haller, A. v. (1766). Gustus. In Elementa physiologiae, Vol. IV, V.
taste bud in the mud puppy, Necturus maculosus. Am. J. Anat. F. Grasset, Lausanne, pp. 99–124.
131:353–369. Hamasaki, K., Seta, Y., Yamada, K., and Toyoshima, K. (1998).
Farbman, A. I., Hellekant, G., and Nelson, A. (1985). Structure of Possible role of serotonin in Merkel-like basal cells of the
taste buds in foliate papillae of the rhesus monkey, Macaca taste buds of the frog, Rana nigromaculata. J. Anat.
mulatta. Am. J. Anat. 172:41–56. 193:599–610.
Farbman, A. I., Ogden Ogle, C. K., Hellekant, G., Simmons, S. R., Hannover, A. (1840). Die Chromsäure, ein vorzügliches Mittel bei
Albrecht, R. M., and van der Wel, H. (1987). Labeling of sweet mikroskopischen Untersuchungen. Müllers Archiv: 549–558.
taste binding sites using a colloidal gold- labeled sweet pro- Harada, S., and Smith, D. V. (1992). Gustatory sensitivities of the
tein, thaumatin. Scanning Microsc. 1:351–357. hamster’s soft palate. Chem. Senses 17:37–51.
Feneis, H. (1985). Pocket Atlas of Human Anatomy. Thieme, New Harris, W. (1952). The fifth and seventh nerves in relation to the
York. nervous mechanism of taste sensation: a new approach. Br.
Finger, T. E. (1997). Evolution of taste and solitary chemorecep- Med. J. 1:831–836.
tor cell systems. Brain. Behav. Evol. 50:234–243. Hellekant, G. (1976). The blood circulation of the tongue. In
Finger, T. E., Bryant, B. P., Kalinoski, D. L., Teeter, J. H., Frontiers of Oral Physiology, Vol. 2 Kawamura, Y. (Ed.).
Bottger, B., Grosvenor, W., Cagan, R. H., and Brand, J. G. S. Karger, Basel, pp. 130–145.
(1996). Differential localization of putative amino acid recep- Hellman, T. J. (1922). Die Genese der Zungenpapillen beim
tors in taste buds of the channel catfish, Ictalurus punctatus. Menschen. Upps. Läkaref. Förh. 26(5–6):1–69.
J. Comp. Neurol. 373:129–138. Helmholtz, H. (1842). De fabrica systematis nervosi evertebrato-
Fixsen, C. (1857). De linguae raninae textura disquisitiones rum. Dissertation, University of Berlin
microscopicae. Dissertation, University Dorpat. Hermann, F. (1885). Beitrag zur Entwicklungsgeschichte des
Froriep, R. (1828). De lingua anatomica quaedam et semiotica. Geschmacksorgans beim Kaninchen. Arch. Mikrosk. Anat.
Dissertation, University Dorpat. Entwicklungsmech. 24:216–229.
Fujimoto, S., Ueda, H., and Kagawa, H. (1987). Herness, M. S. (1989). Vasoactive intestinal peptide-like immunore-
Immunocytochemistry on the localization of 5-hydroxytryp- activity in rodent taste cells. Neuroscience 33:411–419.
tamine in monkey and rabbit taste buds. Acta Anat. Basel Herrick, C. J. (1904). The organ and sense of taste in fishes. U. S.
128:80–83. Fish Comm. Bull. 1902 237–272.
Fujita, T. (1994). Gustatory cells as paraneurons. In Olfaction Hersch, M., and Ganchrow, D. (1980). Scanning electron
and Taste, Vol. XI, Kurihara, K., Suzuki, N., and Ogawa, H. microscopy on human embryonic and fetal tongue. Chem.
(Eds.). Springer, New York, pp. 2–4. Senses 5:331–341.
Ganchrow, D., Ganchrow, J. R., and Goldstein, R. S. (1991). Hinrichsen, K. V. (1990). Humanembryologie. Lehrbuch und
Ultrastructure of palatal taste buds in the perihatching chick. Atlas der vorgeburtlichen Entwicklung des Menschen.
Am. J. Anat. 192:69–78. Springer, Berlin.
Ganchrow, J. R., and Ganchrow, D. (1987). Taste bud development Hirata, Y. (1966). Fine structure of the terminal buds on the bar-
in chickens (Gallus gallus domesticus). Anat. Rec. 218:88–93. bels of some fishes. Arch. Histol. Jpn. 26:507–523.
Ganchrow, J. R., Ganchrow, D., Royer, S. M., and Kinnamon, J. C. Höfer, D., and Drenckhahn, D. (1999). Localisation of actin,
(1993). Aspects of vertebrate gustatory phylogeny: morphology villin, fimbrin, ezrin and ankyrin in rat taste receptor cells.
and turnover of chick taste bud cells. Microsc. Res. Tech. Histochem. Cell Biol. 112:79–86.
26:106–119. Hofer, H., Meinel, W., and Rommel, C. (1979). Taste buds in the
Gerlach, J. (1858). Mikroskopische Studien aus dem Gebiet der epithelium of the plica sublingualis of New World monkeys.
menschlichen Morphologie. F. Enke, Erlangen. Anat. Anz. 145:17–31.
Morphology of the Peripheral Taste System 673

Hoffmann, A. (1875). Ueber die Verbreitung der Kinnamon, J. C., Taylor, B. J., Delay, R. J., and Roper, S. D.
Geschmacksknospen beim Menschen. Arch. Pathol. Anat. (1985). Ultrastructure of mouse vallate taste buds. I. Taste
Physiol. Klin. Med. 62:516–530. cells and their associated synapses. J. Comp. Neurol.
Hönigschmied, J. (1873). Beiträge zur mikroskopischen 235:48–60.
Anatomie über die Geschmacksorgane der Säugethiere. Kinnamon, J. C., Sherman, T. A., and Roper, S. D. (1988).
Zeitschr. Wiss. Zoologie, Abt. A 23:414–434. Ultrastructure of mouse vallate taste buds: III. Patterns of
Hosley, M. A., Hughes, S. E., and Oakley, B. (1987). Neural synaptic connectivity. J. Comp. Neurol. 270:1–10, 56.
induction of taste buds. J. Comp. Neurol. 260:224–232. Kinnamon, J. C., Henzler, D. M., and Royer, S. M. (1993).
Hou-Jensen, H. (1933). Die Papillae foliatae des Menschen. HVEM ultrastructural analysis of mouse fungiform taste
Z. Anat. Entwicklgesch. 102:348–388. buds, cell types, and associated synapses. Microsc. Res. Tech.
Hu, Z. L., Masuko, S., and Katsuki, T. (1996). Distribution and ori- 26:142–156.
gins of nitric oxide-producing nerve fibers in the dog tongue: Kinnamon, J. C., McPheeters, M. M., and Kinnamon, S. C.
correlated NADPH-diaphorase histochemistry and immunohis- (1994). Structure/Function correlates in taste buds. In
tochemistry for calcitonin gene-related peptide using light and Olfaction and Taste, Vol. XI, Kurihara, K., Suzuki, N., and
electron microscopy. Arch. Histol. Cytol. 59:491–503. Ogawa, H. (Eds.). Springer, New York, pp. 9–12.
Hunt, J. R. (1915). The sensory field of the facial nerve: a further Kishi, M., Emori, Y., Tsukamoto, Y., and Abe, K. (2001). Primary
contribution to the symptomatology of the geniculate gan- culture of rat taste bud cells that retain molecular markers for
glion. Brain 38:418–446. taste buds and permit functional expression of foreign genes.
Iida, M., Yoshioka, I., and Muto, H. (1983). Taste bud papillae on Neuroscience 106:217–225.
the retromolar mucosa of the rat, mouse and golden hamster. Koelliker, R. A. (1844). Die Selbständigkeit und Abhängigkeit
Acta Anat. Basel 117:374–381. des sympathischen Nervensystems durch anatomische
Iwasaki, S., Asami, T., and Wanichanon, C. (1996). Beobachtungen bewiesen. Ein akademisches Programm.
Ultrastructural study of the dorsal lingual epithelium of the Meyer. U. Zeller, Zürich.
soft-shell turtle, Trionyx cartilagineus (Chelonia, Korte, G. E. (1980). Ultrastructure of the taste buds of the red-
Trionychidae). Anat. Rec. 246:305–316. eared turtle, Chrysemys scripta elegans. J. Morphol. 163:231.
Jaeger, C. B., and Hillmann, D. E. (1976). Gustatory system. Kotrschal, K. (1996). Solitary chemosensory cells: Why do pri-
Morphology of gustatory organs. In Frog Neurobiology A mary aquatic vertebrates need another taste system? Trends
Handbook, Llinás, R., and Precht, W. (eds.). Springer, New Ecol. Evol. 11:110–114.
York, pp. 588–606. Krause, R. (1911). Kursus der normalen Histologie. Urban and
Jägel, B. (1991). Zur Geschichte der Anatomie und Physiologie Schwarzenberg, Berlin.
des Geschmackssinnes. Dissertation, University of Kiel. Lalonde, E., and Eglitis, J. (1961). Number and distribution of
Jakubowski, M., and Whitear, M. (1986). Ultrastructure of taste taste buds on the epiglottis, pharynx, larynx, soft palate and
buds in fishes. Folia Histo-chem. Cytobiol. 24:310–311. uvula in a human newborn. Anat. Rec. 140:91–95.
Jakubowski, M., and Whitear, M. (1990). Comparative morphol- Landin, A., and Chaudhari, N. (2000). Expression of genes intro-
ogy and cytology of taste buds in teleosts. Z. Mikrosk.-anat. duced into rat taste cells via liposome-mediated transfection.
Forsch. 104:529–560. 22nd annual meeting of the Association for Chemoreception
Jeffery, W., Strickler, A., Guiney, S., Heyser, D., and Tomarev, S. Sciences (AChemS), Sarasota, FL.
(2000). Prox 1 in eye degeneration and sensory organ com- Lewis, D., and Dandy, W. E. (1930). The course of the nerve
pensation during development and evolution of the cavefish fibers transmitting sensation of taste. Arch. Surg. 21:249–288.
Astyanax. Dev. Genes Evol. 210:223–230. Leydig, F. (1851). über die Haut einiger Süßwasserfische.
Johnson, E. W., Eller, P. M., Jafek, B. W., and Norman, A. W. Z. Wiss. Zool. 3:1–12.
(1992). Calbindin-like immunoreactivity in two peripheral Li, X. J., and Snyder, S. H. (1995). Molecular cloning of Ebnerin,
chemosensory tissues of the rat: taste buds and the a von Ebner’s gland protein associated with taste buds. J. Biol.
vomeronasal organ. Brain Res. 572:319–324. Chem. 270:17674–17679.
Jowett, A., and Shrestha, R. (1998). Mucosa and taste buds of the Lindemann, B. (1996). Taste reception. Physiol. Rev.
human epiglottis. J. Anat. 193:617–618. 76:719–766.
Jurisch, A. (1922). Studien über die Papillae vallatae beim Linden, R. W. (1993). Taste. Br. Dent. J. 175:243–253.
Menschen. Z. Anat. Entwicklungsgesch. 66:1–149. Lovén, C. (1868). Beiträge zur Kenntnis vom Bau der
Kanazawa, H. (1993). Fine structure of the canine taste bud with Geschmackswärzchen der Zunge. Arch. mikrosk. Anat.
special reference to gustatory cell functions. Arch. Histol. IV:96–110.
Cytol. 56:533–548. Lu, S., Kaya, N., and Herness, S. (2000). Cholecystokinin
Katz, D. B., Nicolelis, M. A., and Simon, S. A. (2000). IV. There increases intracellular calcium levels in rat posterior taste
is more to taste than meets the tongue. Am. J. Physiol. receptor cells. 22nd annual meeting of the Association for
Gastrointest. Liver Physiol. 278:G6–G9. Chemoreception Sciences (AChemS), Sarasota, FL.
Kim, D. J., and Roper, S. D. (1995). Localization of serotonin in Lussana, F. (1869). Recherches expérimentales et observations
taste buds: a comparative study in four vertebrates. J. Comp. pathologiques sur les nerfs du gout. Arch. Phys. (Paris)
Neurol. 353:364–370. 2:197–210.
674 Witt et al.

Lussana, F. (1872). Sur les nerfs du gout. Observations et expéri- Morris-Wiman, J., Sego, R., Brinkley, L., and Dolce, C.
ences nouvelles. Arch. Phys. (Paris) 4:334–350. (2000). The effects of sialoadenectomy and exogenous EGF
Magendie, F. (1820). Grundriss der Physiologie (translated from on taste bud morphology and maintenance. Chem. Senses
French by C. F. Heusinger). Baercke, Eisenach. 25:9–19.
Malpighi, M. (1687). Exercitatio epistolica de lingua. In: Opera Munch, F. (1896). Die Topographic der Papillen de Zunge des
omnia, Malpighi, M. (ed.). R. Scott and G. Wells, London, Menschen und der Saugethiere. Morph. Arbeit. 6:605–684.
pp. 13–20. Murray, R. G. (1973). The ultrastructure of taste buds. In The
Menco, B. P. (1989). Olfactory and nasal respiratory epithelia, Ultrastructure of Sensory Organs, Friedmann, I. (ed.). North
and foliate taste buds visualized with rapid-freeze freeze- Holland Publishing Company, Amsterdam, pp. 1–81.
substitution and Lowicryl K11M embedding. Ultrastructural Murray, R. G. (1986). The mammalian taste bud type III cell: a
and initial cytochemical studies. Scanning Microsc. critical analysis. J. Ultrastruct. Mol. Struct. Res. 95:175–188.
3:257–272. Murray, R. G., and Murray, A. (1967). Fine structure of taste buds
Menco, B. P. M., Yankova, M. P., and Simon, S. A. (1997). of rabbit foliate papillae. J. Ultrastruct. Res. 19:327–353.
Freeze-substitution and postembedding immunocytochem- Murray, R. G., and Murray, A. (1971). Relations and possible sig-
istry on rat taste buds: G-proteins, calcitonin gene-related nificance of taste bud cells. Contrib. Sens. Physiol. 5:47–95.
peptide, and choline acetyl transferase. Microsc. Microanal. Murray, R. G., Murray, A., and Fujimoto, S. (1969). Fine struc-
3:53–69. ture of gustatory cells in rabbit taste buds. J. Ultrastruct. Res.
Merkel, F. (1880). Ueber die Endigungen der sensiblen Nerven in 27:444–461.
der Haut der Wirbelthiere. Stiller, Rostock. Nada, O., and Hirata, K. (1977). The monoamine-containing cell
Miller, I. J., Jr., and Bartoshuk, L. M. (1991). Taste perception, in the gustatory epithelium of some vertebrates. Arch. Histol.
taste bud distribution, and spatial relationships. In Smell and Jpn. 40 (Suppl.):197–206.
Taste in Health and Disease, Getchell, T. V., Bartoshuk, L. Nagato, T., Ren, X. Z., Toh, H., and Tandler, B. (1997).
M., Doty, R. L., and Snow, J. B. J., Jr. (Eds.). Raven Press, Ultrastructure of Weber’s salivary glands of the root of the
New York, pp. 205–233. tongue in the rat. Anat. Rec. 249:435–440.
Miller, I. J., Jr., and Reedy, F. E., Jr. (1990a). Variations in human Nemetschek-Ganssler, H., and Ferner, H. (1964). über die
taste bud density and taste intensity perception. Physiol. Ultrastruktur der Geschmacksknospen. Z. Zellforsch.
Behav. 47:1213–1219. 63:155–178.
Miller, I. J., Jr., and Reedy, F. E. J. (1990b). Quantification of Nolte, C., and Martini, R. (1992). Immunocytochemical localiza-
fungiform papillae and taste pores in living human subjects. tion of the L1 and N-CAM cell adhesion molecules and their
Chem. Senses 15:281–294. shared carbohydrate epitope L2/HNK-1 in the developing and
Miller, I. J., Jr., and Smith, D. V. (1984). Quantitative taste bud differentiated gustatory papillae of the mouse tongue.
distribution in the hamster. Physiol. Behav. 32:275–285. J. Neurocytol. 21:19–33.
Miller, I. J., Jr., and Spangler, K. M. (1982). Taste bud distribu- Nomura, S., Shiba, Y., Muneoka, Y., and Kanno, Y. (1979). A
tion and innervation on the palate of the rat. Chem. Senses scanning and transmission electron microscope study of the
7:99–108. premetamorphic papillae: possible chemoreceptive organs in
Mistretta, C. M., and Baum, B. J. (1984). Quantitative study of the oral cavity of an anuran tadpole (Rana japonica). Arch.
taste buds in fungiform and circumvallate papillae of young Histol. Jpn. 42:507–516.
and aged rats. J. Anat. 138:323–332. Norgren, R. (1990). Gustatory system. In The Human Nervous
Mistretta, C. M., Goosens, K. A., Farinas, I., and Reichardt, L. F. System, Paxinos, G. (Ed.). Academic Press, San Diego,
(1999). Alterations in size, number, and morphology of gusta- pp. 845–861.
tory papillae and taste buds in BDNF null mutant mice Nosrat, C. A., and Olson, L. (1995). Brain-derived neurotrophic
demonstrate neural dependence of developing taste organs. factor mRNA is expressed in the developing taste bud-bearing
J. Comp. Neurol. 409:13–24. tongue papillae of rat. J. Comp. Neurol. 360:698–704.
Mochizuki, Y. (1939). Studies of the papilla foliata of Japanese. Nosrat, I. V., Lindskog, S., Seiger, A., and Nosrat, C. A. (2000).
Okajimas Folia Anat. Jpn. 18:334–369. Lingual BDNF and NT-3 mRNA expression patterns and their
Moll, R. (1993). Cytokeratins as markers of differentiation: relation to innervation in the human tongue: similarities and
expression profiles in epithelia and epithelial tumors. In differences compared with rodents. J. Comp. Neurol.
Progress in Pathology, Vol. 142, Seifert, G. (Ed.). G. Fischer, 417:133–152.
New York, pp. 1–197. Oakley, B. (1970). Reformation of taste buds by crossed sensory
Montavon, P., Hellekant, G., and Farbman, A. (1996). nerves in the rat’s tongue, Acta Physiol. Scand. 79:88–94.
Immunohistochemical, electrophysiological, and electron Oakley, B. (1998a). Taste neurons have multiple inductive roles
microscopical study of rat fungiform taste buds after regener- in mammalian gustatory development. Ann. NY Acad. Sci.
ation of chorda tympani through the non-gustatory lingual 855:50–57.
nerve. J. Comp. Neurol. 367:491–502. Oakley, B. (1998b). Vertebrate taste-bud development: are sala-
Morris-Wiman, J., Basco, E., and Du, Y. (1999). The effects of manders the model? Trends Neurosci. 21:337–337.
beta-bungarotoxin on the morphogenesis of taste papillae and Oakley, B., Brandemihl, A., Cooper, D., Lau, D., Lawton, A., and
taste buds in the mouse. Chem. Senses 24:7–17. Zhang, C. X. (1998). The morphogenesis of mouse vallate
Morphology of the Peripheral Taste System 675

gustatory epithelium and taste buds requires BDNF-depen- Remak, R. (1852). Ueber die Ganglien der Zunge bei
dent taste neurons. Dev. Brain Res. 105:85–96. Säugethieren und beim Menschen. Müller’s Archiv.
Ohmura, S., Horimoto, S., and Fujita, K. (1989). Lectin cyto- Retzius, G. (1892). Die Nervenendigungen in dem
chemistry of the dark granules in the type 1 cells of Syrian Geschmacksorgan der Säugethiere und Amphibien. B. Das
hamster circumvallate taste buds. Arch. Oral Biol. 34:161–166. Geschmacksorgan der Amphibien. Samson and Wallin,
Ohshima, H., Yoshida, S., and Kobayashi, S. (1990). Blood vas- Stockholm.
cular architecture of the rat lingual papillae with special ref- Reutter, K. (1971). Die Geschmacksknospen des Zwergwelses
erence to their relations to the connective tissue papillae and Amiurus nebulosus (Lesueur). Morphologische und histo-
surface structures: a light and scanning electron microscope chemische Untersuchungen. Z. Zellforsch. 120:280–308.
study. Acta Anat. (Basel) 137:213–221. Reutter, K. (1973). Typisierung der Geschmacksknospen von
Ojima, K., Matsumoto, S., Takeda, M., Saiki, C., and Takahashi, Fischen. I. Morphologische und histochemische
T. (1997a). Numerical variation and distributive pattern on Untersuchungen an Xiphophorus helleri Heckel (Poeciliidae,
microvascular cast specimens of vallate papillae in the cross- Cyprinodontiformes, Teleostei). Z. Zellforsch. 143:409–423.
bred Japanese cat tongue. Anat. Anz. 179:117–126. Reutter, K. (1978). Taste organ in the bullhead (Teleostei). Adv.
Ojima, K., Takahashi, T., Matsumoto, S., Takeda, M., Saiki, C., Anat. Embryol. Cell Biol. 55:3–94.
and Mitsuhashi, F. (1997b). Angioarchitectural structure of Reutter, K. (1986). Chemoreceptors. In Biology of the
the fungiform papillae on rabbit tongue anterodorsal surface. Integument, Vol. 2, Bereiter-Hahn, J., Matoltsy, A. G., and
Anat. Anz. 179:329–333. Richards, K. S. (Ed.). Springer, Berlin, pp. 586–604.
Ojima, K., Takeda, M., Matsumoto, S., Saiki, C., and Takahashi, Reutter, K. (1992). Structure of the peripheral gustatory organ,
T. (1997c). Functional role of V form distribution seen in represented by the siluroid fish Plotosus lineatus (Thunberg).
microvascular cast specimens of the filiform and fungiform In Fish Chemoreception, Hara, T. J. (Ed.). Chapman and Hall,
papillae on the posterior central dorsal surface of the cat London, pp. 60–78.
tongue. Anat. Anz. 179:321–327. Reutter, K., and Witt, M. (1993). Morphology of vertebrate taste
Olmsted, J. M. D. (1920). The results of cutting the seventh cra- organs and their nerve supply. In Mechanisms of Taste
nial nerve in Amiurus nebulosus (Lesueur). J. Exp. Zool. Transduction, Simon, S. A. and Roper, S. D. (Eds.). CRC
31:369–401. Press, Boca Raton, FL, pp. 29–82.
Osculati, F., and Sbarbati, A. (1995). The frog taste disc: a proto- Reutter, K., and Witt, M. (1999). Comparative aspects of fish
type of the vertebrate gustatory organ. Prog. Neurobiol. taste bud ultrastructure. In Advances in Chemical Signals in
46:351–399. Vertebrates, Johnston, E., Müller-Schwarze, D., and
Panizza, B. (1834). Ricerche sperimentali sopra i nervi. Bizzoni, Sorensen, P. W. (Eds.). Kluwer Academic/Plenum Publishers,
Pavia. Dordrecht, pp. 573–581.
Paran, N., Mattern, C. F., and Henkin, R. I. (1975). Ultrastructure Riva, A., Loffredo, F., Puxeddu, R., and Testa Riva, F. (1999). A
of the taste bud of the human fungiform papilla. Cell Tissue scanning and transmission electron microscope study of the
Res. 161:1–10. human minor salivary glands. Arch. Oral Biol. 44 (Suppl.
Parker, G. H. (1922). Smell, Taste, and Allied Senses in the 1):S27–31.
Vertebrates. Lippincott, London. Roper, S. D. (1989). The cell biology of vertebrate taste recep-
Pevzner, R. A., and Tikhonova, M. A. (1979). Ultrastructural tors. Annu. Rev. Neurosci. 12:329–353.
organization of the taste buds of reptiles. I. Tortoises. Royer, S. M., and Kinnamon, J. C. (1988). Ultrastructure of
Tsitologija 21:1266. mouse foliate taste buds: synaptic and nonsynaptic interac-
Ponzo, M. (1907). Sulla presenza di organi del gusto nella parte tions between taste cells and nerve fibers. J. Comp. Neurol.
laringea della faringe, nel tratto cervicale dellésofago e nel 270:11–24, 58.
palato duro del feto umano. Anat. Anz. 31:570–575. Royer, S. M., and Kinnamon, J. C. (1991). HVEM serial-section
Pritchard, T. C. (1991). The primate gustatory system. In Smell analysis of rabbit foliate taste buds: I. Type III cells and their
and Taste in Health and Disease, Getchell, T. V., Bartoshuk, synapses. J. Comp. Neurol. 306:49–72.
L. M., Doty, R. L., and Snow, J. B. J., Jr. (Eds.). Raven Press, Ruiz-Avila, L., McLaughlin, S. K., Wildman, D., McKinnon,
New York, pp. 109–125. P. J., Robichon, A., Spickofsky, N., and Margolskee, R. F.
Pumplin, D. W., Yu, C., and Smith, D. V. (1997). Light and dark (1995). Coupling of bitter receptor to phosphodiesterase
cells of rat vallate taste buds are morphologically distinct cell through transducin in taste receptor cells. Nature 376:
types. J. Comp. Neurol. 378:389–410. 80–85.
Pumplin, D. W., Getschman, E., Boughter, J. D., Jr., Yu, C., and Saito, I. (1966). Comparative anatomical studies of the oral
Smith, D. V. (1999). Differential expression of carbohydrate organs of the poultry. V. Structures and distribution of taste
blood-group antigens on rat taste-bud cells: relation to the func- buds of the fowl (in Japanese). Bull. Fac. Agric.
tional marker alpha-gustducin. J. Comp. Neurol. 415:230–239. 13:95–102.
Rabl, H. (1895). Notiz zur Morphologie der Geschmacksknospen Sbarbati, A., Crescimanno, C., Bernardi, P., and Osculati, F.
auf der Epiglottis. Anat. Anz. 11:153–156. (1999). Alpha-gustducin-immunoreactive solitary chemosen-
Ranvier, L. (1888). Technisches Lehrbuch der Histologie (German sory cells in the developing chemoreceptorial epithelium of
translation by W. Nicati and H. v. Wyss). F. C. W. Vogel, Leipzig. the rat vallate papilla. Chem. Senses 24:469–472.
676 Witt et al.

Scalzi, H. A. (1967). The cytoarchitecture of gustatory receptors Sprissler, C. (1994). Ultrastruktur der Geschmacksknospe der
from the rabbit foliate papillae. Z. Zellforsch. Mikrosk. Anat. japanischen Wachtel (Coturnix coturnix japonica).
80:413–435. Dissertation, University of Tübingen.
Schinkele, O. (1942). Über das Vorkommen von Sprissler, C., and Reutter, K. (1993). Ultrastructure of taste buds
Geschmacksknospen im kranialen Drittel des Oesophagus. Z. in the quail, Coturnix coturnix japonica. In 20th Göttingen
Mikrosk.-anat.Forsch. 51:498–501. Neurobiology Conference, Elsner, N. and Richter, D. W.
Schmale, H., and Bamberger, C. (1997). A novel protein with (Eds.). G. Thieme, Stuttgart.
strong homology to the tumor suppressor p53. Oncogene Stone, L. M., Finger, T. E., Tam, P. P., and Tan, S. S. (1995). Taste
15:1363–1367. receptor cells arise from local epithelium, not neurogenic
Schmale, H., Holtgreve-Grez, H., and Christiansen, H. (1990). ectoderm. Proc. Natl. Acad. Sci. USA 92:1916–1920.
Possible role for salivary gland protein in taste reception indi- Stone, L. S. (1940). The origin and development of taste organs
cated by homology to lipophilic-ligand carrier proteins. in salamanders observed in the living condition. J. Exp. Zool.
Nature 343:366–369. 83:481–506.
Schmale, H., Ahlers, C., Blaker, M., Kock, K., and Spielman, A. I. Suzuki, Y., Takeda, M., Obara, N., and Nagai, Y. (1996).
(1993). Perireceptor events in taste. Ciba Found. Symp. Phagocytic cells in the taste buds of rat circumvallate papillae
179:167–180; discussion 180–185. after denervation. Chem. Senses 21:467–476.
Schulze, F. E. (1863). über die becherförmigen Organe der Svejda, J., and Janota, M. (1974). Scanning electron microscopy
Fische. Z. Wiss. Zool. 12:218–222. of the papillae foliatae of the human tongue. Oral Surg. Oral
Schwalbe, G. (1868). über die Geschmacksorgane der Med. Oral Pathol. 37:208–216.
Säugethiere und des Menschen. Arch. Mikr. Anat. 4:154–187. Tandler, B., Pinkstaff, C. A., and Riva, A. (1994). Ultrastructure
Schwann, T. (1839). Mikroskopische Untersuchungen über die and histochemistry of human anterior lingual salivary glands
übereinstimmung in der Struktur und dem Wachstum der Thiere (glands of Blandin and Nuhn). Anat. Rec. 240:167–177.
und Pflanzen. Sander’sche Buchhandlung (Reimer), Berlin. Toto, P. D., Nadimi, H., and Martinez, R. (1993). Von Ebner’s
Schwartz, H. G., and Weddell, G. (1938). Observations on the gland, an immunohistochemical study. Ann. NY Acad. Sci.
pathways transmitting the sensation of taste. Brain 61:99–115. 694:322–324.
Selliseth, N. J., and Selvig, K. A. (1993). Microvasculature of Toubeau, G., Cotman, C., and Bels, V. (1994). Morphological
the dorsum of the rat tongue: a scanning electron micro- and kinematic study of the tongue and buccal cavity in the
scopic study using corrosion casts. Scand. J. Dent. Res. lizard Anguis fragilis (Reptilia:Anguidae). Anat. Rec.
101:391–397. 240:423–433.
Shatzman, A. R., and Henkin, R. I. (1981). Gustin concentration Toyoshima, K., and Shimamura, A. (1987). Monoamine-contain-
changes relative to salivary zinc and taste in humans. Proc. ing basal cells in the taste buds of the newt Triturus
Natl. Acad. Sci. USA 78:3867–3871. pyrrhogaster. Arch. Oral Biol. 32:619–621.
Siegel, R. E. (1970). Galen on Sense Perception. His Doctrines, Toyoshima, K., and Tandler, B. (1987). Modified smooth endo-
Observations and Experiments on Vision, Hearing, Smell, plasmic reticulum in type II cells of rabbit taste buds.
Taste, Touch and Pain, and Their Historical Sources. S. J. Submicrosc. Cytol. 19:85–92.
Karger, Basel. Toyoshima, K., Miyamoto, K., and Shimamura, A. (1987). Fine
Smith, D. V., Akeson, R. A., and Shipley, M. T. (1993). NCAM structure of taste buds in the tongue, palatal mucosa and gill
expression by subsets of taste cells is dependent upon inner- arch of the axolotl, Ambystoma mexicanum. Okajimas Folia
vation. J. Comp. Neurol. 336:493–506. Anat. Jpn. 64:99–109.
Smith, D. V., Klevitsky, R., Akeson, R. A., and Shipley, M. T. Toyoshima, K., Seta, Y., Toyono, T., and Takeda, S. (1999).
(1994). Expression of the neural cell adhesion molecule Merkel cells are responsible for the initiation of taste organ
(NCAM) and polysialic acid during taste bud degeneration morphogenesis in the frog. J. Comp. Neurol. 406:129–140.
and regeneration. J. Comp. Neurol. 347:187–196. Tuckerman, F. (1889). On the development of the taste-organs of
Smith, D. V., Som, J., Boughter, J. D., Jr., St. John, S. J., Yu, C., man. J. Anat. Physiol. 23:559–582.
and Christy, R. C. (1999). Cellular expression of alpha-gust- Uchida, T. (1980). Ultrastructural and histochemical studies on
ducin and the A blood group antigen in rat fungiform taste the taste buds in some reptiles. Arch. Histol. Jpn. 43:459–478.
buds cross-reinnervated by the IXth nerve. J. Comp. Neurol. von Düring, M. V., and Andres, K. H. (1976). The ultrastructure
409:118–130. of taste and touch receptors of the frog’s taste organ. Cell
Sobotta, J. (1993). Atlas der Anatomie des Menschen, Vol. 1. Tissue Res. 165:185–198.
Urban and Schwarzenberg, München. von Vintschgau, M., and Hönigschmied, J. (1877). Nervus glos-
Soemmering, S. T. (1806). Abbildungen der menschlichen sopharyngeus und Schmeckbecher. Arch. Physiol.
Organe des Geschmackes und der Stimme. Varrentrapp und 14:443–448.
Wenner, Frankfurt. Verson, E. (1868). Beiträge zur Kenntnis und des Kehlkopfes und
Spielman, A. I. (1990). Interaction of saliva and taste. J. Dent. der Trachea. Sitzungsber. Wiener Acad. Wissenschaft. Math.-
Res. 69:838–843. naturwiss. Klasse 57:1093–1102.
Morphology of the Peripheral Taste System 677

Wagner, R. (1837). Bestätigung des Panizzaschen Lehrsatzes, Witt, M., and Reutter, K. (1990). Electron microscopic demon-
dass das 9te Nervenpaar (n.glossopharyngeus) der stration of lectin binding sites in the taste buds of the
Geschmacksnerv ist. Frorieps Notizen 4:129–131. European catfish Silurus glanis (Teleostei). Histochemistry
Wakisaka, S., Tabata, M. J., Maeda, T., Matsumoto, K., Wanaka, 94:617–628.
A., Muramatsu, H., Muramatsu, T., and Kurisu, K. (1998). Witt, M., and Reutter, K. (1996). Embryonic and early fetal
Immunohistochemical localization of pleiotrophin and mid- development of human taste buds: a transmission electron
kine in the lingual epithelium of the adult rat. Arch. Histol. microscopical study. Anat. Rec. 246:507–523.
Cytol. 61:475–480. Witt, M., and Reutter, K. (1997). Scanning electron microscopi-
Waller, A. (1847). Microscopic examination of the papillae and cal studies of developing gustatory papillae in humans. Chem.
nerves of the tongue of the frog, with observations on the Senses 22:601–612.
mechanism of taste. London Edinburgh Dublin Philosoph. Witt, M., Ganchrow, J. R., and Ganchrow, D. (1999). Distribution
Magazine J. Sci. XXX:277–289. of vimentin in the developing chick taste bud during the
Waller, A. (1849). Minute structure of the papillae and nerves of the perihatching period. Cell. Mol. Biol. (Noisy-Le-Grand)
tongue of the frog and the toad. Communicated by R. Owen. 45:303–316.
Philosoph. Transact. Royal Soc. London Pt.I:139–149. Wong, G. T., Ruiz-Avila, L., Ming, D., Gannon, K. S., and
Welsch, U., and Storch, V. (1969). [Fine structure of the taste Margolskee, R. F. (1996). Biochemical and transgenic
buds of catfish (Clarias batrachus (L) and Kryptopterus bicir- analysis of gustducin’s role in bitter and sweet transduction.
rhis (Cuvier and Valenciennes)]. Z. Zellforsch. Mikrosk. Anat. Cold Spring Harb. Symp. Quant. Biol. 61:173–184.
100:552–559. Wyss, H. v. (1870). Die becherförmigen Organe der Zunge. Arch.
Whitear, M. (1970). The skin surface of bony fishes. J. Zool. mikrosk. Anat. 6:237–260.
London 160:437–454. Yamasaki, F., and Takahashi, K. (1982). A description of the
Whitear, M. (1992). Solitary chemosensory cells. In Fish times of appearance and regression of marginal lingual
chemoreception, Hara, T. (Ed.). Chapman and Hall, London, papillae in human fetuses and newborns. Anat. Rec.
pp. 103–125. 204:171–173.
Whitehead, M. C., Ganchrow, J. R., Ganchrow, D., and Yao, B. Yoshie, S., Wakasugi, C., Teraki, Y., Iwanaga, T., and Fujita, T.
(1998). Neural cell adhesion molecule, neuron-specific eno- (1988). Immunocytochemical localizations of neuron-specific
lase and calcitonin gene-related peptide immunoreactivity in proteins in the taste bud of the guinea pig. Arch. Histol. Cytol.
hamster taste buds after chorda tympani lingual nerve dener- 51:379–384.
vation. Neuroscience 83:843–856. Yoshie, S., Wakasugi, C., Teraki, Y., Iwanaga, T., and Fujita, T.
Williams, P. L., Warwick, R., Dyson, M., and Bannister, L. H. (1991). Fine structure of the taste bud in guinea pigs. II.
(Eds). Gray’s Anatomy, 37th ed., Churchill Livingstone, Localization of spot 35 protein, a cerebellar Purkinje cell-spe-
Edinburgh, 1989, p. 228. cific protein, as revealed by electron-microscopic immunocy-
Wilson, D. B. (1980). Embryonic development of the head and tochemistry. Arch. Histol. Cytol. 54:113–118.
neck: part 4, organs of special sense. Head Neck Surg. Zancanaro, C., Sbarbati, A., Bolner, A., Accordini, C., Piemonte, G.,
2:237–247. and Osculati, F. (1995). Biogenic amines in the taste organ.
Witt, M. (1993). Ultrastructure of the taste disc in the red-bellied Chem. Senses 20:329–335.
toad Bombina orientalis (Discoglossidae, Salientia). Cell Zeng, Q., Lawton, A., and Oakley, B. (1995). Glycoconjugates
Tissue Res. 272:59–70. and keratin 18 define subsets of taste cells. Histochem. J.
Witt, M. (1995). Distribution of vasoactive intestinal peptide-like 27:997–1006.
immunoreactivity in the taste organs of teleost fish and frog. Zhang, C., and Oakley, B. (1996). The distribution and origin of
Histochem. J. 27:161–165. keratin 20-containing taste buds in rat and human.
Witt, M. (1996). Carbohydrate histochemistry of vertebrate taste Differentiation 61:121–127.
organs. Prog. Histochem. Cytochem. 30:1–168. Zhang, C. X., Cotter, M., Lawton, A., Oakley, B., Wong, L., and
Witt, M., and Kasper, M. (1998). Immunohistochemical distribu- Zeng, Q. (1995). Keratin 18 is associated with a subset of
tion of CD44 and some of its isoforms during human taste bud older taste cells in the rat. Differentiation 59:155–162.
development. Histochem. Cell Biol. 110:95–103. Zimmermann, K. W. (1927). Die Speicheldrüsen der Mundhöhle
Witt, M., and Kasper, M. (1999). Distribution of cytokeratin fila- und die Bauchspeicheldrüse. In Handbuch der
ments and vimentin in developing human taste buds. Anat. Mikroskopischen Anatomie des Menschen, Vol 5, Möllendorf,
Embryol. 199:291–299. W. v. (Ed.). Springer, Berlin, pp. 61–244.
Witt, M., and Miller, I. J., Jr. (1992). Comparative lectin histo- Żuwata, K. (1997). Ultrastructure of premetamorphic taste
chemistry on taste buds in foliate, circumvallate and fungiform organs of the Bombina variegata. Rocz. Akad. Med. Bialymst.
papillae of the rabbit tongue. Histochemistry 98:173–182. 42 (Suppl 2):204–207.
Witt, M., and Reutter, K. (1988). Comparative lectin histochemi- Żuwata, K., and Jakubowski, M. (1991). Development of taste
cal investigation of the taste buds of different vertebrates. Acta organs in Rana temporaria. Transmission and scanning elec-
Histochem. Suppl. 36:405–408. tron microscopic study. Anat. Embryol. (Berl) 184: 363–369.
33

Central Taste Anatomy and Neurophysiology

Edmund T. Rolls
University of Oxford, Oxford, United Kingdom

Thomas R. Scott
San Diego State University, San Diego, California, U.S.A.

The aims of this chapter are to describe the anatomy and 1984). Second-order taste neurons project through the
physiology of the central taste system, with reference to central tegmental tract to terminate in the parvicellular
the rodent and to the primate to make the findings rele- division of the ventroposteromedial thalamic nucleus
vant to taste processing and its disorders in humans. (VPMpc) (Beckstead et al., 1980; Norgren, 1984;
Neuroimaging studies in humans are also described. We Pritchard et al., 1989). A remarkable difference from the
also address the convergence of gustatory input with taste system of rodents is this direct projection from the
olfactory, somatosensory, and visual afferents, and how NTS to gustatory thalamus. In rodents there is an obliga-
the central representation of the sensory properties of tory relay from the NTS to the pontine parabrachial taste
food is made relevant to the control of the appetite for a nuclei, which in turn project to the thalamus (Norgren,
food and food intake. Much of the research in rodents 1984; Norgren and Leonard, 1973). The pontine taste
has been carried out at subcortical levels, whereas in pri- nuclei also project to the hypothalamus and amygdala in
mates more research has been carried out on the cortical rodents (Norgren, 1976), providing direct access in
processing of taste. The evidence suggests that not only rodents to these subcortical structures important in moti-
the central connections, but also the principles of pro- vational behavior (e.g., feeding) and learning (Rolls
cessing in the taste pathways, may be different in rodents 1990). In contrast, in primates there appears to be no such
and primates. direct pathway from the brainstem taste areas to the
hypothalamus and amygdala (Norgren, 1984). This fun-
damental difference in the anatomy of the rodent and pri-
I. GUSTATORY ANATOMY mate taste pathways shows that even in a phylogeneti-
cally old system such as taste, the way in which the
Diagrams of the taste pathways in primates are shown in system functions and processes information may be dif-
Figures 1 and 2. As described in detail in the preceding ferent across mammalian orders. This may result from
chapter, three cranial nerves (CN) carry taste information the great development of the cerebral cortex in primates
centrally: CN VII (chorda tympani and greater superficial and the advantage of using extensive cortical processing
petrosal branches), CN IX (lingual branch), and CN X from each sensory modality before the representations
(superior laryngeal branch). Gustatory axons terminate in are integrated in multimodal regions, as suggested below.
the rostral part of the nucleus of the solitary tract (NTS) The thalamic taste area, VPMpc, projects to the primary
in the medulla (Beckstead and Norgren, 1979; Norgren, taste cortex, which forms the rostral part of the frontal

679
680 Rolls and Scott

The orbitofrontal cortex receives inputs from other sen-

sory systems, as shown schematically in Figure 2. There are
direct connections from the primary olfactory cortex (pyri-
form cortex) to area 13a of the posterior orbitofrontal cor-
tex, which in turn has onward projections to a middle part
of the orbitofrontal cortex (area 11) (Barbas, 1993;
Morecraft et al., 1992; Carmichael et al., 1994) (see Fig. 2).
Visual inputs reach the orbitofrontal cortex directly from
the inferior temporal cortex in which representations of
objects are found (Booth and Rolls, 1998; Rolls, 2000c),
the cortex in the anterior part of the superior temporal sul-
cus in which face-responsive neurons are found (Hasselmo
et al., 1989a,b; Wallis and Rolls, 1997), and the temporal
pole (see Barbas, 1988, 1993, 1995; Barbas and Pandya,
1989; Carmichael and Price, 1995; Morecraft et al.,
1992; Seltzer and Pandya, 1989). There are corresponding
auditory inputs from the superior temporal cortex (Barbas,
1988, 1993), and somatosensory inputs from somatosen-
sory cortical areas 1, 2 and SII in the frontal and pericentral
operculum, and from the insula (Barbas, 1988; Carmichael
Figure 1 Drawing of the human brain, with the presumed and Price, 1995). The caudal orbitofrontal cortex receives
afferent limb of the taste system superimposed. nst, nucleus of strong inputs from the amygdala (see Chapter 8) (Price,
the solitary tract; pbn, parabrachial nucleus; VPMpc, ventropos- 2001). The orbitofrontal cortex also receives inputs via the
teromedial nucleus of the thalamus, pars parvocellularis; ins, mediodorsal nucleus of the thalamus, pars magnocellularis,
insula; op, operculum. (Drawing by Mr. Birck Cox, used with which itself receives afferents from temporal lobe struc-
permission.) tures such as the prepyriform (olfactory) cortex, amygdala,
and inferior temporal cortex (see Price et al., 1996). The
operculum and adjoining insula (Figs. 1–3) (Pritchard orbitofrontal cortex projects back to temporal lobe areas
et al., 1986). In macaques, this is at the anterior end of the such as the inferior temporal cortex and, in addition, to the
Sylvian fissure. Cells in the insular-opercular taste cortex entorhinal cortex (or “gateway to the hippocampus”) and
(IO) then project anteriorly to a part of the caudolateral cingulate cortex (Insausti et al., 1987). The orbitofrontal
orbitofrontal cortex (OFC), in which taste-responsive cortex also projects to the preoptic region and lateral hypo-
neurons are found (Baylis et al., 1994; Rolls et al., 1990). thalamus, to the ventral tegmental area (Johnson et al.,
This OFC area does not receive inputs from VPMpc, but 1968; Nauta, 1964), and to the head of the caudate nucleus
instead receives projections from the mediodorsal nucleus (Kemp and Powell, 1970). Reviews of the cytoarchitecture
of the thalamus, which projects to the prefrontal cortex. and connections of the orbitofrontal cortex are provided by
Thus the caudolateral orbitofrontal taste cortex contains a Petrides and Pandya (1994), Pandya and Yeterian (1996),
secondary taste cortical area, which receives gustatory Carmichael and Price (1994, 1995), and Barbas (1995).
inputs from the primary frontal opercular and insular taste The anatomy of taste pathways beyond these is not
cortices. Afferents also reach the caudolateral orbitofrontal known in detail. In primates, taste neurons are found in the
taste cortex from the more ventral part of the rostral insu- hypothalamus to which OFC projects (Burton et al., 1976;
lar cortex, the amygdala, the substantia innominata, the Rolls et al., 1986), and in the amygdala (Rolls, 2000;
rhinal sulcus, and the surrounding orbitofrontal cortex. Sanghera et al., 1979; Scott et al., 1993), which receives
Through some of these pathways visceral information may inputs from the insular cortex (Mesulam and Mufson,
reach the caudolateral orbitofrontal taste cortex. Injections 1982b) and from the OFC (Amaral et al., 1992).
of horseradish peroxidase more anteriorly in the
orbitofrontal cortex have been shown to label neurons in
II. TASTE PROCESSING
the secondary taste cortex, but not in the primary taste
cortex, providing evidence for tertiary cortical taste A. Nucleus of the Solitary Tract
regions in more anterior parts of the primate orbitofrontal
1. Primates
cortex in which taste-responsive neurons are found (Baylis
et al., 1994; Rolls, 1997; Rolls and Baylis, 1994; Thorpe Taste neurons have been found and their responses analyzed
et al., 1983). in the rostral part of the nucleus of the solitary tract of
Central Taste Anatomy and Neurophysiology 681

Figure 2 Schematic diagram of the taste pathways in primates (center) showing how they converge with olfactory, visual, and
somatosensory pathways. The gate functions shown refer to the finding that the responses of taste neurons in the orbitofrontal cortex and
the lateral hypothalamus are modulated by hunger. VPMpc: Ventroposteromedial thalamic nucleus; V1, V2, V4: visual cortical areas.

macaque monkeys (Scott et al., 1986a). Different neurons ent stimuli. This distributed encoding may provide an
were found that responded best to glucose, NaCl, HCl efficient mechanism for information transmission in a rela-
(sour), and quinine HCl (bitter), but the tuning of the tively small number of neurons (see Erickson, 1985; Hinton
neurons was in most cases broad; for example, 84% of the et al., 1986; Rolls and Treves, 1998). In macaques, feeding
neurons had at least some response to 3 or more of these 4 to satiety with glucose has no impact on the responsiveness
prototypical taste stimuli. The mean value of the breadth of of NTS cells to the taste of the glucose (Yaxley et al., 1985).
tuning metric of Smith and Travers (1979) (see Chapter 34),
calculated from the responses to the four prototypical
2. Rodents
stimuli for 52 neurons analyzed in the nucleus of the solitary
tract (NTS), was 0.87. Thus, neurons in this part of the Evidence about gustatory coding in the taste pathways of
primate taste system have relatively broad tuning to differ- rodents is described in Chapter 35. In rodents the NTS is
682 Rolls and Scott

thought to manage a basic level of taste discrimination, to c. Autonomic Reflexes. A second projection from ventral
control somatic reflexes of acceptance or rejection, and to NTS goes caudally, to ramify through viscerosensory NTS
regulate autonomic reflexes that anticipate digestive and the dorsal motor nucleus of the vagus (DMNX), through
processes. Neurons here may also modulate feeding which it may influence autonomic reflexes. There are sev-
behavior according to short-term satiety signals. eral cephalic phase pancreatic and gastrointestinal reflexes
associated with digestion, of which the best delineated is the
a. Basic Discriminations. Lesions in the gustatory cephalic phase insulin release (Powley et al., 1987).
division of the NTS impair a rat’s capacity to discriminate Thousands of  cells in the rats’ islets of Langerhans are
sapid stimuli from water (Shimura et al., 1997a). The con- stimulated to release insulin by fewer than 200 fibers cours-
centration-response function is flattened for all basic taste ing through the gastric and hepatic branches of the vagus
qualities, but the effect is most severe for sucrose and nerve. These originate in the rostromedial division of the
NaCl. The deficit is greater with damage to the NTS than DMNX, which is overlain by the gustatory NTS. Neurons in
to other gustatory relays. Conversely, rats with only the DMNX send apical dendrites into the NTS, functionally fus-
hindbrain intact show typical acceptance-rejection reflexes ing the two structures and offering gustatory input direct
(see below), implying that the taste quality that drives the control over autonomic reflexes (Powley and Berthoud,
reflex is being properly assessed. 1991). Accordingly, a sweet taste elicits insulin release, and
this effect is blocked by vagotomy (Louis-Sylvestre et al.,
b. Acceptance-Rejection Reflexes. Clusters of cells in 1983). The insulin release is also blocked if the sweet stim-
the ventral regions of gustatory NTS send their projections ulus is paired with gastrointestinal distress to create a condi-
ventrolaterally to salivatory and pharyngeal areas of the tioned taste aversion, implying that this profound learning
reticular formation, to the hypoglossal nucleus, and toward experience alters the gustatory code for sweetness.
the facial and ambiguus nuclei, through which acceptance-
rejection reflexes could be orchestrated. These are fully d. Satiety. The acceptance-rejection reflexes men-
integrated within the hindbrain, in both rodents (Grill and tioned above may be modified by short-term satiety signals
Norgren, 1978) and humans (Steiner, 1979), and are in decerebrate rats (Grill and Kaplan, 1990). If satiated, or
stereotypical to each of the basic taste qualities. if blood glucose levels are raised, acceptance reflexes to a

Figure 3 Coronal sections to show the locations of the primary taste cortices in the macaque in the frontal operculum and rostral insula
and of the secondary taste cortex in the caudolateral orbitofrontal cortex. The coordinates are in mm anterior (A) or posterior (P) to sphe-
noid. (From Aggleton and Passingham, 1981.)
Central Taste Anatomy and Neurophysiology 683

sweet stimulus are lost, as they are in an intact rat, imply- subtypes of taste cells, the responsiveness of sucrose-spe-
ing that hindbrain cells are sensitive to these immediate cialized neurons to sucrose were suppressed by 77%. This
signals of satiety. Similarly, when glucose is made is nearly the same as the reduction in response of sugar
systemically more available to neurons by means of intra- cells to glucose in the NTS (74%) after hyperglycemia had
venous glucose infusions (Giza and Scott, 1983), mild been induced (Giza et al., 1997a,b). The result implies that
hyperinsulinemia (Giza and Scott, 1987a), or glucagon a reduced response among cells that would otherwise pro-
administration (Giza et al., 1993), gustatory responses to vide a powerful signal for reward may be one mechanism
glucose in the NTS show some decline, implying a reduc- by which satiety is manifested.
tion in the appetitive nature of the taste signal (Giza et al.,
1992). The decrease of responsiveness occurs within five
minutes, and is reversible. Moreover, the effects are differ- b. Conditioned Taste Aversions. A conditioned taste
ential, depending on the cell type. Giza et al. (1997a,b) aversion (CTA) is formed to a novel taste when it is paired
found that neurons that responded specifically to sugars with gastrointestinal malaise. PBN is central to this
were suppressed by 74% in their response to glucose fol- process. Rats with either electrolytic (Reilly et al., 1993)
lowing hyperglycemia. or ibotenic acid lesions of the PBN could learn neither a
conditioned taste (Reilly, 1998) nor odor aversion
(Grigson et al., 1998), though their ability to develop a
B. Parabrachial Nucleus conditioned taste preference was unimpaired. The failure
resulted neither from a loss of taste nor from visceral sen-
In rodents, gustatory afferents terminate in the medial sitivity, but rather from an inability to form an association
parabrachial nucleus (PBN), whereas afferents communi- between them. There are conflicting data regarding the dis-
cating visceral information synapse more laterally. ruptive effects of PBN lesions on existing CTAs (Grigson
However, the gustatory and visceral modalities converge et al., 1997; Sakai et al., 1998).
on single cells within the PBN, suggesting that while it is Both electrophysiological and c-fos studies have been
involved in gustatory processing, the primary role of the employed to document the effects of a CTA on activity in
PBN may be to mediate the interactions between physio- the PBN. Creation of a CTA to the taste of NaCl resulted
logical condition and taste. The integrity of the PBN is in greater responses to NaCl thereafter (Shimura et al.,
crucial to the formation and retention of a conditioned 1997b), though the impact of that increase on the code for
taste aversion and for the expression of a sodium appetite. NaCl could not be determined. The same result had been
reported using a saccharin-conditioned stimulus in the
1. Processing Taste Information NTS (Chang and Scott, 1984). The application of
immunocytochemical techniques has permitted topo-
Neurons in the medial PBN of the rat respond robustly to graphic analysis of the effects of a CTA. Yamamoto and
taste stimuli (Perrotto and Scott, 1976), and the evoked colleagues report that, when a formerly appetitive taste is
activity is essentially a linear transformation of that evoked used as the CS in a CTA paradigm, the c-fos expression
from NTS cells by the same stimuli (Scott and Perrotto, it elicits shifts from a PBN subnucleus associated with
1980). Studies of c-fos expression suggest that discrete positive hedonics to one normally activated by aversive
subareas of the PBN are activated by visceral stimulation tastes (Yamamoto, 1993). Separate subnuclei of the PBN
and by tastes of various qualities (Yamamoto et al., 1994). may send discrete projections to forebrain areas that
Destruction of PBN neurons, however, has considerably analyze taste and hedonics. If such a CTA-induced
less impact on taste discrimination than do corresponding reorganization of the topographic representation of taste
lesions in the NTS (Flynn et al., 1991a,b). is confirmed, it would provide a basis for the shift in
hedonic value of the CS, established in PBN, but
2. Responses to Physiological Condition manifested more rostrally.
a. Satiety. Lesions of the PBN block the increase in
feeding that is otherwise induced by the fructose analog c. Sodium Appetite. Lesions of the PBN blocked the
2,5-anhydro-D-mannitol (Grill et al., 1995). Since 2,5-AM expression of a sodium appetite in rats (Flynn et al.,
works only peripherally, this implies vagal mediation. 1991a,b). The implication that PBN is involved in sodium
Conversely, when a rat has a satiating load of intralipid appetite is strengthened by electrophysiological results,
administered into the duodenum, taste responsiveness in showing that gustatory responsiveness to NaCl declines
the PBN declines. The decline is selective, with the great- when rats are depleted, just as has been shown in the CT
est effect on the taste activity evoked by sucrose. Among nerve (Contreras and Frank, 1979) and NTS (Jacobs et
684 Rolls and Scott

al., 1988). The nature of that involvement has yet to be responsive before the lesion is made. With the location
clarified. assured, the region of destruction can be sharply limited.
Finally, fibers of passage are spared by the use of cytotoxic
chemicals that destroy only cell bodies. Under these
C. Thalamic Taste Area
conditions, the presumed functions of VPMpc were not
1. Primates confirmed. Rats with lesions in VPMpc showed only
slightly impaired preference-aversion functions across taste
Pritchard et al. (1989) were able to confirm that the par-
qualities and concentrations in both 24-hour, two-bottle and
vicellular division of the VPMpc is the thalamic taste relay
15-minute, single-bottle tests (Reilly and Pritchard, 1996a).
nucleus in primates by showing that single neurons in it
They were willing to expend a normal amount of effort for
responded to taste stimuli in macaque monkeys. The neu-
a gustatory reinforcer (Reilly and Trifunovic, 1999). They
rons were relatively broadly tuned, with a mean breadth of
were capable of forming normal conditioned taste aversions
tuning of 0.73 (measured across responses to sucrose,
and preferences and conditioned odor aversions (Grigson et
NaCl, HCl, and quinine HCl). Responses to sweet and salt
al., 2000; Reilly and Pritchard, 1996b). They were able to
were most common, with 56% responding best to sucrose,
express a sodium appetite, with the only abnormality being
and 24% responding best to NaCl. Relatively few neurons
that the appetite did not increase in strength with multiple
had best responses to HCl or QHCl (14%), and most of the
sodium depletions (Scalera et al., 1997). In sum, the thala-
responses to HCl and QHCl were described as side-band
mic taste relay does not appear necessary in rodents for
responses in NaCl-best neurons. In addition to gustatory
nearly normal perception of and response to taste stimuli,
neurons, some neurons in the nucleus responded to tactile
nor for taste-guided behaviors that rely on integrating taste
stimuli. Rather surprisingly, given the effects of cortical
information with physiological condition, including
lesions on food preferences described below, Reilly and
malaise (CTA), nutritional repletion (CTP), or sodium
Pritchard (1995) reported that VPMpc lesions in macaques
depletion (sodium appetite).
caused only minor changes in acceptance behavior.
Where thalamic-lesioned rats are deficient is in tasks
that require a gustatory memory not associated with phys-
2. Rodents
iological condition. They are unable to show anticipatory
In rats, cells in the parvicellular division of the VPMpc contrast, i.e., the reduction in consumption of a mildly
represent an obligatory synapse between gustatory preferred solution in anticipation of one that is highly
parabrachial nucleus and gustatory cortex. Therefore, these preferred (Riley and Trifunovic, 1999). Intact rats show
neurons were assumed to play a major role in taste per- suppressed drinking of weak saccharin relative to controls
ception. Electrophysiological studies indicate that taste if they have been trained to expect a strong sucrose solu-
cells in VPMpc have the sensitivities to support such a tion immediately afterward. Rats with VPMpc lesions
role. Thalamic taste cells show monotonically increasing show no such suppression despite repeated training
responses with stimulus concentration and differential sessions, implying that they are not capable of forming a
activity to a full range of taste qualities (Scott and gustatory memory. The difference between this training
Erickson, 1971). Moreover, many taste cells in the rat and conditioning based on physiological changes (CTA,
VPMpc are multimodal for touch and thermal stimulation CTP, sodium appetite) is that the latter has been shown to
as well, implying the early phases of the integration of dif- involve actual changes in the responsiveness of the taste
ferent sensory modalities (Verhagen et al., 1999). system (Chang and Scott, 1984; Contreras and Frank,
Consistent with this evidence, early lesion studies 1979; Giza et al., 1993; Jacobs et al., 1988; McCaughey
reported that destruction of VPMpc resulted in elevated and Scott, 2000). Thus, the effect may be established else-
detection thresholds, blunting of preference-aversion func- where, and the resulting altered responsiveness of the taste
tions, and loss of the capacity to generate conditioned taste system may sustain the aversion (CTA) or preference
aversions or to express a sodium appetite (Wolf and (CTP, sodium appetite) response. The implication is that
DiCara, 1974). It now appears that these studies were the thalamic taste relay of rodents is concerned not so
flawed, victims of the persistent technical shortcomings of much with the consummatory aspects of taste-related
early lesion experiments: lesions that were misplaced, behavior as with providing a route necessary for the for-
were inordinately large, or that destroyed fibers of passage mation of the gustatory memories that guide the search for
that underlay functions unrelated to those of the target site. food (Reilly, 1998).
In recent experiments, lesion sites have been identified Finally, discrepant reports about the involvement of
with the aid of electrophysiological guidance to ensure VPMpc in CTA formation may be reconciled by a close
proper placement. Neurons are confirmed as being taste study of the anatomical connections from the parabrachial
Central Taste Anatomy and Neurophysiology 685

nucleus to thalamus (Bester et al., 1999). The more lateral composed of functionally similar neurons. Nor does it
regions of PBN (internal lateral and ventral lateral subnu- imply that gustatory processing areas that are more con-
clei) project to the midline and intralaminar thalamic cerned with the hedonic components of chemicals (e.g.,
complex, which has been shown to be second in impor- orbitofrontal cortex) may not reintroduce topographic
tance only to the PBN itself in creating a CTA. From the organization.
thalamus, fibers project to the basolateral amygdala,
whose integrity is also required for normal CTA formation. b. Coding of Taste Quality. Coding strategies are
In contrast, the more medial divisions of PBN project to discussed in Chapter 15. There, Smith and Scott rely pri-
VPMpc, carrying taste information not subject to involve- marily on data from rodents to conclude that there are
ment in the CTA. The magnitude of lesions performed in classes of taste cells and that taste quality is represented by
earlier studies apparently conflated these two pathways. patterns of activity read across them. Analysis of responses
from some 800 taste cells from IO leads to the same result
D. Insular-Opercular (Primary) Taste Cortex in the insular and opercular primary taste cortices (Scott
and Plata-Salaman, 1999; Scott et al., 1986b; Yaxley et al.,
As one proceeds beyond the thalamic taste area, about 1990).
which little is known, and on to the cortical taste areas, the Figure 4 shows the relationship among response
wealth of the data are from primates. In the following profiles of 399 taste cells in primate IO in the form of a
analysis, we focus on results gained from recordings in dendrogram (whose organization is described in the figure
alert macaques. legend) to four prototypical tastants (Scott and Plata-
Salaman, 1999). It reveals rather sharp distinctions
1. Anatomical Connections between the clusters of 153 glucose-oriented cells to the
left, 137 NaCl-oriented cells in the center, and 89 quinine-
Afferents to the insular-opercular cortex (IO) include
oriented cells to the right, from which a small group of 20
those from the thalamic taste relay in VPMpc, but also
HCl-oriented cells extends. Each group is statistically
connections from primary somatosensory cortex, the
distinct from the others. Therefore, taste neurons may be
(auditory) cortex in the superior temporal sulcus, entorhi-
segregated into statistically independent types, but within
nal cortex, and the basolateral amygdala. Efferents from
them the constituent neurons do not have identical sensi-
this area project to the second somatosensory cortical
tivities. It also appears that the primate taste system
area, the cortex in the superior temporal sulcus, several
devotes ~73% of its neural resources (290 of 399 cells) to
parts of the frontal cortex including the OFC, and the lat-
the detection of stimuli that humans describe as sweet or
eral, central, and corticomedial nuclei of the amygdala
salty, i.e., those that dominate the cuisines of the world.
(Augustine, 1996).
Neurons tuned primarily to quinine constitute ~22% of
cortical cells, while only ~5% are oriented toward the
2. Representation of Taste in the Insular-Opercular
detection of acids.
Cortex
Taste quality may be represented by the similarity
a. Topographic Organization. There are indications of between patterns of activity generated across all taste cells.
a crude topographic arrangement of taste qualities in the By this index, cells in IO most accurately represent taste
nucleus of the solitary tract (NTS) of the macaque (Scott et quality. The relationship between the correlations gener-
al., 1986) that conform to the organization of sensitivities ated from macaque neurophysiological data and from the
in the human mouth (Collings, 1974). Moreover, a preser- descriptions of the same stimulus pairs by humans is
vation of anatomical organization from NTS to VPMpc has closest in IO. This is borne out by a quantitative compari-
been reported (Beckstead et al., 1980). However, this is son between macaque neurophysiology and human
apparently not retained in IO (Pritchard et al., 1986). psychophysics (Fig. 5). Humans judged the similarity
Accordingly, a plot of the location of each taste cell in IO between each of 21 pairs of stimuli, and these judgments
as a function of its most effective stimulus offers no were plotted against the correlation between patterns
evidence for topographic organization by taste quality. elicited by the same two stimuli in macaque cortex. The
Apparently, the gross physical location of a neuron is not a only major discrepancy involved NaCl versus the bitter
salient criterion for coding taste quality, i.e., there is no salts, CaCl2 and MgCl2, a relationship that was closer in
apparent region devoted to perceiving, say, saltiness. the macaque than in humans. Even with this anomaly, the
This conclusion does not eliminate the possibility of correlation between the two plots is 0.91; without it,
fine-grained topographic organization that would not have 0.97. Therefore, the most accurate analysis of perceived
been detected by these studies, e.g., cortical columns taste quality is likely to take place in primary taste cortex.
686 Rolls and Scott

Figure 4 A dendrogram indicating the distribution of similarity among 399 taste cells in the insular-opercular cortex. The basic stim-
ulus that elicited the dominant response from each group of neurons is listed at the bottom, while the degree of similarity among the
profiles of the neurons is shown on the ordinate. Heavy lines indicate the levels at which the groups reach statistical independence. (From
Scott and Plata-Salamán, 1999.)

3. Other Inputs into the Insular-Opercular Cortex Although neuroimaging and evoked potential studies do
not have sufficient spatial resolution to separate activations
A relatively low proportion of neurons in the primary taste produced in the anterior insula from more posterior insular
cortex are apparently activated by taste (Scott and Plata- areas with known somatosensory and other inputs, acti-
Salaman, 1999; Scott et al., 1986b; Yaxley et al., 1990). For vation in this general region can be produced by stimulation
example, across seven studies 6% of insular-opercular of visceral regions of the esophagus (Aziz et al., 1995;
neurons responded to taste, 24% responded to movements of Weusten et al., 1994) and of the throat (Roper et al., 1993),
the jaw, 4% to tactile stimulation of the mouth, 1% to tongue as well as by general visceral arousal induced by phobic
extension, and 1% to the sight of food (Scott and Plata- anxiety (Rauch et al., 1995). Further, activation of this
Salaman, 1999). Some of these effects may be produced by general region has been produced in positron emission
inputs from the primary somatosensory cortex, and there is a tomography (PET) studies when subjects moved their
whole set of somatosensory inputs to parts of the insula just mouths and tongue and during speech, implying involve-
posterior to the primary taste cortex (Mufson and Mesulam, ment in oromotor functions (Raichle, 1991). Fiol et al.
1982; Robinson and Burton, 1980; Schneider et al., 1993), (1988) reported that seizures that began in this general
which is at the anterior (rostral) end of the insular-orbital region elicited vomiting, and electrical stimulation of this
cortex (Pritchard et al., 1986). Oral and facial somatosensory region in experimental animals does the same
fields do appear to be represented in the anterior insula, at (Oppenheimer et al., 1992). Also, activation in this general
least close to taste areas (Schneider et al., 1993). area can be produced by the sight of face expressions of
Central Taste Anatomy and Neurophysiology 687

Figure 5 Relationship between response profiles of stimulus quality generated from human psychophysical responses (---) and from
neural responses in the macaque (—). The correlation between each stimulus pair is nearly identical by these two approaches, with the
exception of those between bitter and salty salts (N: 0.3 M NaCl vs. Ca: 0.3 M CaCl2 and Mg: 0.3 M MgCl2). F: 1 M fructose; K: 0.3 M
KCl; S: 1 M sucrose. (Human psychophysical data are from Kuznicki and Ashbaugh, 1979.)

disgust (literally bad taste) (Phillips et al., 1997). Further by Thorpe et al. (1983). In recordings made from 3120
research is needed to investigate the extent to which indi- single neurons, Rolls et al. (1990) found a secondary corti-
vidual neurons with taste responses in the primary taste cor- cal taste area in the caudolateral part of the orbitofrontal
tex at the rostral end of the insular and adjoining frontal cortex of the macaque. The area is part of the dysgranular
operculum actually show convergence of inputs from these field of the orbitofrontal cortex, OFdg (see Figs. 2 and 3),
other sources. and is situated anterior to the primary taste cortical area in
the frontal opercular and adjoining insular cortices. The
E. Gustatory Responses in the Caudolateral responses of 49 single neurons with gustatory responses in
Orbitofrontal Secondary Cortical Taste Area the caudolateral orbitofrontal taste cortex were analyzed
using the taste stimuli glucose, NaCl, HCl, quinine HCl,
In a study of the role of the orbitofrontal cortex in learning, water, and black currant juice. Examples of the responses of
Thorpe et al. (1983) found a small proportion of neurons one orbitofrontal cortex neuron to the stimuli are shown in
(8%) with gustatory responses in the main part of the Figure 6, and quite sharp tuning is evident, in that the neu-
orbitofrontal cortex. In some cases these neurons were very ron responded primarily to the taste of glucose. The mean
selective for particular gustatory stimuli. Therefore, when breadth of tuning coefficient (calculated over the responses
they set out to search for a secondary taste cortical area, to the four prototypical stimuli) for 49 cells in the cau-
Rolls et al. (1990) started recording at the anterior bound- dolateral orbitofrontal cortex was 0.39. This tuning is much
ary of the opercular and insular cortical taste areas and finer than that of neurons in the nucleus of the solitary tract
worked forward towards the orbitofrontal area investigated of the monkey, and finer than that of neurons in the primary
688 Rolls and Scott

F. Amygdala

The amygdala receives projections from both the primary

(IO) and secondary (OFC) taste cortices. Sanghera et al.
(1979) described taste responses in some amygdala neurons
in macaques, and Wilson and Rolls (1993, 2000) found
more. Nishijo et al. (1988a,b) found ingestion-related
amygdala neurons, at least some of which were probably
taste-related. Taste-responsive cells composed about 7% of
neurons in another study in the primate amygdala (Scott
et al., 1993). As opposed to the precision with which stim-
ulus concentration and quality are represented in IO, the
amygdaloid representation was found to be rather crude.
Inhibitory responses are nearly as prevalent as excitatory
ones, such that the net concentration-response functions are
nearly flat, with increasing inhibition offsetting greater
excitation. When absolute response rates are calculated,
concentration-response functions are monotonically
increasing to most basic stimuli, but are not steep enough to
match human psychophysical functions. The basic taste
qualities are less distinctly separated from one another than
at lower neural levels. Glucose is clearly discriminable
from others; NaCl and MSG are less distinct; HCl and qui-
Figure 6 Examples of the responses recorded from one nine HCl are not separable. Therefore, taste-evoked activity
caudolateral orbitofrontal taste cortex neuron to the six taste in the amygdala does not appear to provide an adequate
stimuli, water, 20% black currant juice (BJ), 1.0 M glucose, neural basis for the discriminative capacity of humans or
1.0 M NaCl, 0.01 M HCl, and 0.001 M quinine HCl (QHCl). monkeys with regard to either stimulus concentration or
quality. Rather, the role of the amygdala may be to impart
hedonic value to the taste experience (see below).
frontal opercular and the insular taste cortices. A cluster
analysis showed that at least seven different groups of neu- G. Hypothalamus
rons were present, the mean profiles of which are shown in
Figure 7. For each of the stimuli glucose, black currant The lateral hypothalamic area receives projections from
juice, NaCl, and water, there was one group of neurons that OFC and from the amygdala and has been reported to pos-
responded much more to that tastant than to the other tas- sess taste-responsive cells in rats (Norgren, 1970). In
tants. The other groups of neurons responded to two or macaques some neurons are found that respond to taste,
more of these tastants, such as glucose and black currant and these neurons are implicated in responses to food in
juice. In this particular region neurons were not found with that they are modulated in a sensory-specific manner by
large responses to HCl or quinine HCl (Rolls et al., 1990). satiety (see below) (Burton et al., 1976; Rolls et al., 1986).
In more recent studies, the area of orbitofrontal cortex from These neurons are likely to be involved in autonomic
which recordings have been made has been extended, and responses to foods and may also be involved in the reward
neurons with large responses to other stimuli such as HCl, value that food has when the organism is hungry (Rolls,
quinine, and monosodium glutamate have been found 1999; Rolls et al., 1980). Oomura et al. (1991) and Karádi
(Baylis and Rolls, 1991; Critchley and Rolls, 1996c; Rolls et al. (1992) have also shown that lateral hypothalamic
and Baylis, 1994; Rolls et al., 1994, 1996a,b, 1999). In neurons that are glucose-sensitive (as shown by micro-
these studies of further regions of the orbitofrontal cortex, electrophoresis) are more likely to respond to taste and
Rolls and colleagues are finding further areas with taste- olfactory stimuli than are glucose-insensitive neurons.
responsive neurons, some of which are broadly tuned, but
as in the earlier studies (Rolls et al., 1990; Thorpe et al., H. Basal Ganglia
1983), some of the neurons are very finely tuned.
Moreover, as described below in Sec. IV, some of these The ventral striatum, including the nucleus accumbens and
neurons are multimodal. the olfactory tubercle, receives afferents from the OFC and
Central Taste Anatomy and Neurophysiology 689

Figure 7 The mean response profiles (  SD) of each of the main clusters of neurons in the secondary taste cortex. The numbers of the
cells included in each cluster are annotated. (From Rolls et al., 1990.)

amygdala and contains some taste-responsive neurons Japanese word umami, which is a taste common to a
(Williams et al., 1993). Some 19% of the neurons in diversity of food sources including fish, meats, mush-
globus pallidus are reported to respond to taste stimulation, rooms, cheese, and some vegetables. Within these food
while 16% are sensitive to olfactory stimulation (Karádi sources, glutamates and 5-nucleotides—sometimes in a
et al., 1995), but their response characteristics are not yet synergistic combination—create the umami taste (Ikeda,
well defined. 1909; Kawamura and Kare, 1987; Yamaguchi, 1967;
Yamaguchi and Kimizuka, 1979). Monosodium
I. Protein Taste L-glutamate (MSG), guanosine-5-monophosphate
(GMP), and inosine-5-monophosphate (IMP) are exam-
An important food taste that appears to be different from ples of umami stimuli.
that produced by sweet, salt, bitter, or sour is the taste These findings raise the question of whether umami
of protein. At least part of this taste is captured by the taste operates through channels in the primate taste system
690 Rolls and Scott

which are separable from those for the “prototypical” corresponding odors, in that humans rated the intensity of
tastes sweet, salt, bitter, and sour. (Although the concept of umami flavor higher if the taste of glutamate was
four prototypical tastes has been used by tradition, there is combined with the odor of methyl furyl disulfide (garlic)
increasing discussion about the utility of the concept, and (Rolls, 2000c; Rolls et al., 1998).
increasing evidence that the taste system is more diverse Thus, neurophysiological evidence in primates does indi-
than this; see, e.g., Kawamura and Kare, 1987.) To investi- cate that there is a representation of umami flavor in the
gate the neural encoding of glutamate in the primate, cortical areas which is separable from that to the four proto-
Baylis and Rolls (1991) made recordings from 190 taste- typical taste qualities sweet, salt, bitter, and sour (see also
responsive neurons in the primary taste cortex and Rolls, 2000c; Rolls et al., 1998). This representation is
adjoining orbitofrontal cortex taste area in macaques. probably important in the taste produced by proteins. These
Single neurons were found that were tuned to respond best findings of cortical neurons tuned to respond to protein
to monosodium glutamate (umami taste), just as other cells (umami) taste are supported by the more recent discovery of
were found with best responses to glucose (sweet), sodium glutamate taste receptors on the tongue (Chaudhari et al.,
chloride (salty), HCl (sour), and quinine HCl (bitter). 2000). In addition, to the extent that certain amino acids taste
Across the population of neurons, the responsiveness to sweet, salty, or sour, they evoke neuronal activity in the
glutamate was poorly correlated with the responsiveness to primary taste cortex, which is similar to that evoked by glu-
NaCl, so that the representation of glutamate was clearly cose, NaCl, or HCl, respectively (Plata-Salaman et al., 1992).
different from that of NaCl. Further, the representation of
glutamate was shown to be approximately as different
from each of the other four tastants as they are from each III. EFFECTS OF HUNGER AND SATIETY
other, as shown by multidimensional scaling and cluster ON TASTE PROCESSING AT DIFFERENT
analysis. Moreover, it was found that glutamate is approx- STAGES OF THE TASTE PATHWAY
imately as well represented in terms of mean evoked
neural activity and the number of cells with best responses Humans report that the hedonic value of tastes declines
to it as the other four stimuli, glucose, NaCl, HCl, and qui- with satiety (Cabanac 1971), but that the intensity of the
nine. It was concluded that in primate taste cortical areas, taste shows little if any decrease (Cabanac, 1971; Rolls and
glutamate, which produces umami taste in humans, is Rolls, 1977, 1982; Rolls et al., 1981a,b, 1982; Thompson
approximately as well represented as are the tastes pro- et al., 1976). These findings show that pleasantness of the
duced by glucose (sweet), NaCl (salty), HCl (sour), and taste of food decreases to zero when the food is eaten to
quinine HCl (bitter) (Baylis and Rolls, 1991). satiety, but that we can still identify the food, and taste it
In a further investigation, these findings have been perfectly well, when we are satiated. In the primate brain,
extended beyond the sodium salt of glutamate to other the evidence described below indicates that these two
umami tastants that have the glutamate ion but do not processes are clearly separated in gustatory information
introduce the sodium ion into the experiment and to a processing, with the nature of the taste (which taste it is,
nucleotide umami tastant (Rolls et al., 1994). In recordings and its intensity) represented in the primary taste cortex,
made mainly from neurons in the orbitofrontal cortex taste and the pleasantness of taste represented only after this
area, it was shown that single neurons that had their best processing, in the secondary taste cortex in the
responses to sodium glutamate also had good responses orbitofrontal cortex, in the lateral hypothalamus, and to
to glutamic acid. The correlation between the responses to some extent in the amygdala. In contrast, in rodents the
these two tastants was higher than between any other pair, separation between the identity of a taste and its hedonic
which included in addition a prototypical set including value seems much less distinct, with modulation of taste
glucose, sodium chloride, HCl, and quinine HCl. processing by hunger early on in the taste pathways. It is
Moreover, the responsiveness to glutamic acid clustered important to examine the effects of satiety on taste
with the response to monosodium glutamate in a cluster processing, because alteration of the reward value of taste
analysis with this set of stimuli, and glutamic acid was by hunger versus satiety is one of the main ways in which
close to sodium glutamate in a space created by multi- food intake is controlled (Rolls, 1975, 1999).
dimensional scaling. It was also shown that the responses
of these neurons to the nucleotide umami tastant inosine- A. Rodents
5-monophosphate were more correlated with their
responses to monosodium glutamate than to any prototyp- Physical and chemical signals of satiety have been shown
ical tastant (Rolls et al., 1994). A psychophysical to have impact on taste responsiveness in the hindbrain of
convergence was also found between umami taste and the rat. Gastric distension by air or with 0.3 M NaCl sup-
Central Taste Anatomy and Neurophysiology 691

presses responses in the NTS, with the greatest effect on attenuated by feeding to satiety (Yaxley et al., 1985). Thus,
glucose (Glenn and Erickson, 1976). Intravenous infusions taste processing at this early stage of the taste system does
of 0.5 g/kg glucose (Giza and Scott, 1983), 0.5 U/kg not appear to be modulated by satiety, the signals for
insulin (Giza and Scott, 1987a), and 40 mg/kg glucagon which include gastric distension (Gibbs et al., 1981) as
(Giza et al., 1993) all cause reductions in taste responsive- well as other signals (see Rolls, 1999). It was also found
ness to glucose in the NTS. The intraduodenal infusion of that in the primary taste cortex, both in the frontal opercu-
lipids cause a decline in taste responsiveness in the PBN, lar part (Rolls et al., 1988) and in the insular part (Yaxley
with the bulk of the suppression borne by glucose cells et al., 1988), hunger does not modulate the responsiveness
(Hajnal et al., 1999). The loss of signal that would other- of single neurons to gustatory stimuli.
wise be evoked by hedonically positive tastes implies that In contrast, in the secondary taste cortex, in the cau-
the pleasure that sustains feeding is reduced, making dolateral part of the orbitofrontal cortex, it was found that
termination of a meal more likely (Giza et al., 1992). the responses of neurons to the taste of glucose decreased
Conversely, the intracerebroventricular administration of to zero while the monkey ate it to satiety, during the course
neuropeptide Y—a peptide that induces feeding—causes of which its behavior turned from avid acceptance to active
an increase in appetitive behavioral reflexes to sucrose rejection (Rolls et al., 1989). This modulation of respon-
(Baird et al., 2000). If taste activity in NTS is affected by siveness of the gustatory responses of the orbitofrontal
the rat’s nutritional state, then intensity judgments in rats cortex neurons by satiety could not have been due to
should change with satiety. There is evidence that they do. peripheral adaptation in the gustatory system or to altered
Rats with conditioned aversions to 1.0 M glucose show efficacy of gustatory stimulation after satiety was reached,
decreasing acceptance of glucose solutions as their because modulation of neuronal responsiveness by satiety
concentrations approach 1.0 M. This acceptance gradient was not seen at the earlier stages of the gustatory system,
can be compared between euglycemic rats and those made including the nucleus of the solitary tract, the frontal oper-
hyperglycemic through intravenous injections (Scott and cular taste cortex, and the insular taste cortex. Evidence
Giza, 1987). Hyperglycemic rats showed greater was obtained that gustatory processing involved in thirst
acceptance at all concentrations from 0.6 to 2.0 M glucose, also becomes interfaced to motivation in the caudolateral
indicating that they perceived these stimuli to be less orbitofrontal cortex taste projection area, in that neuronal
intense than did conditioned rats with no glucose load responses here to water were decreased to zero while water
(Giza and Scott, 1987b). was drunk until satiety was produced (Rolls et al., 1989).
In the secondary taste cortex, it was also found that the
B. Primates decreases in the responsiveness of the neurons were rela-
tively specific to the food with which the monkey had been
In order to provide evidence as to where hunger controls fed to satiety. For example, in 7 experiments in which the
taste processing in primates, the responses of single monkey was fed glucose solution, neuronal responsiveness
neurons at different stages of the taste system have been decreased to the taste of the glucose but not to the taste of
analyzed while macaques are fed to satiety, usually with a black currant juice (see example in Fig. 8). Conversely, in
20% glucose solution. To ensure that the results were rele- two experiments in which the monkey was fed to satiety
vant to the normal control of feeding (and were not due, for with fruit juice, the responses of the neurons decreased to
example, to abnormally high levels of artificially adminis- fruit juice but not to glucose (Rolls et al., 1989).
tered putative satiety signals such as gastric distension or This evidence shows that the reduced acceptance of
plasma glucose), Rolls and colleagues allowed the food that occurs when food is eaten to satiety and the
monkeys to feed until they were satiated and determined reduction in the pleasantness of its taste (Cabanac, 1971;
whether this normal and physiological induction of satiety Rolls, 1986; Rolls and Rolls, 1977, 1982; Rolls et al.,
influenced the responsiveness of neurons in the taste 1981a,b, 1982, 1983) are not produced by a reduction in
system, which were recorded throughout the feeding until the responses of neurons in the nucleus of the solitary tract
satiety was reached. The recordings were made in the or frontal opercular or insular gustatory cortices to
monkey in order to make the results as relevant as possible gustatory stimuli. Indeed, after feeding to satiety, humans
to our understanding of sensory processing and the control reported that the taste of the food on which they had been
of feeding, and its disorders, in humans. satiated tasted almost as intense as when they were hungry,
It was found that this modulation of taste-evoked sig- though much less pleasant (Rolls et al., 1983). This com-
nals by motivation is not a property found in early stages parison is consistent with the possibility that activity in the
of the primate gustatory system. The responsiveness of frontal opercular and insular (primary) taste cortices, as
taste neurons in the nucleus of the solitary tract is not well as in the nucleus of the solitary tract, does not reflect
692 Rolls and Scott

by satiety, and it is presumably in areas such as these that

neuronal activity may be related to whether a food tastes
pleasant and whether the food should be eaten.
In the amygdala, the impact of feeding on taste cells is
intermediate (Yan and Scott, 1996). Some neurons are
fully suppressed by feeding the monkey to satiety, whereas
others are unmodified. The implication is that reward value
is reflected in the activity of amygdaloid neurons, but that
these responses are less responsive than those in the OFC
to a rapidly changing condition which affects taste reward
such as a glucose meal.
The results above help clarify the neural mechanisms
that underlie sensory-specific satiety. Apparently, this phe-
nomenon cannot be attributed to receptor adaptation, nor
to reduced activity in peripheral nerves or in the central
gustatory system through the level of primary taste cortex.
Rather, it rests with higher-order cortical processes of the
OFC and beyond. Gustatory information in primates is
kept distinct from its reward value until it reaches OFC.
Consistent with this, OFC lesions in primates alter food
preferences (Baylis and Gaffan, 1990). Not only do OFC
neurons in primates represent the primary reward or rein-
forcing value of taste stimuli, but they are also involved in
learning and correcting associations of visual and olfactory
with taste reward and punishment stimuli (Rolls, 1999;
Rolls et al., 1996a; Thorpe et al., 1983).

IV. MULTIMODAL REPRESENTATIONS

IN THE TASTE SYSTEM

At some stage in taste processing, it is likely that taste rep-

resentations are brought together with inputs from
different modalities, for example, with olfactory inputs to
form a representation of flavor. Takagi and colleagues
Figure 8 The effect of feeding to satiety with glucose solution on (Tanabe et al., 1975a,b) have found an olfactory area in the
the responses of two neurons in the secondary taste cortex to the
medial orbitofrontal cortex. In a mid-mediolateral part of
taste of glucose and of black currant juice (BJ). The spontaneous
firing rate is also indicated (SA). Below the neuronal response data
the caudal orbitofrontal cortex is the area investigated by
for each experiment, the behavioral measure of the acceptance or Thorpe et al. (1983), in which are found many neurons
rejection of the solution on a scale from 2 to 2 is shown. The with visual and some with gustatory responses. Recordings
solution used to feed to satiety was 20% glucose. The monkey was in the caudolateral orbitofrontal cortex taste area were dif-
fed 50 mL of the solution at each stage of the experiment as indi- ferent from those in the frontal opercular and insular pri-
cated along the abscissa until he was satiated as shown by whether mary taste cortices, in that there were neurons with
he accepted or rejected the solution. Pre: The firing rate of the responses in other modalities within or very close to the
neuron before the satiety experiment started. The values shown are caudolateral orbitofrontal taste cortex (Rolls et al., 1990).
the mean firing rate and its SE. (From Rolls et al., 1989.) It was therefore investigated systematically whether there
are neurons in the secondary taste cortex and adjoining
the pleasantness of the taste of a food, but rather its sensory more medial orbitofrontal cortex that respond to stimuli
qualities independently of motivational state. On the other from other modalities, including the olfactory and visual
hand, the responses of the neurons in the caudolateral modalities, and whether single neurons in this cortical
orbitofrontal cortex taste area and in the lateral hypothala- region in some cases respond to stimuli from more than
mus (Burton et al., 1976; Rolls et al., 1986) are modulated one modality. It was noted that anatomically olfactory
Central Taste Anatomy and Neurophysiology 693

inputs reach this orbitofrontal cortex region in macaques responded best among the tastants to NaCl (N), best
from the pyriform (primary olfactory) cortex, which pro- among the odors to onion odor (On), and well also to
jects to area 13a of the medial orbitofrontal cortex, which salmon (S). The olfactory input to these neurons was fur-
in turn has onward projections to the lateral orbitofrontal ther defined by measuring their responses while the mon-
cortex area, which has been shown to be the secondary key performed an olfactory discrimination task. In the task,
taste cortex, and to more rostral parts of the orbitofrontal if one odor was delivered through a tube close to the nose,
cortex (area 11) (see Fig. 2) (Carmichael et al., 1994; then the monkey could lick to obtain fruit juice (reward tri-
Price, 2001; Price et al., 1991). als). If a different odor was delivered, the monkey had to
avoid licking, otherwise he obtained saline (saline trials).
A. Olfactory and Visual Representations in the The responses of neurons in this task showed that they
Orbitofrontal Cortex Influenced by Taste could have relatively short latency (e.g., 150 msec)
responses to the odor stimuli and did not respond in a
In this investigation of the orbitofrontal cortex taste areas visual discrimination task, so that their responses were
(Rolls and Baylis, 1994), it was found that of 112 single specifically related to the olfactory stimuli, and not just to
neurons that responded to any of these modalities, many performance of a task (Rolls and Baylis, 1994). The
were unimodal (taste 34%, olfactory 13%, visual 21%) but different types of neurons (unimodal in different modali-
were found in close proximity to each other. Some single ties and multimodal) were frequently found close to
neurons showed convergence, responding for example to one another in tracks made into this region (see Fig. 10),
taste and visual inputs (13%), taste and olfactory inputs consistent with the hypothesis that the multimodal
(13%), and olfactory and visual inputs (5%). Some of these representations are actually being formed from unimodal
multimodal single neurons had corresponding sensitivities inputs to this region.
in the two modalities. Thus, one set responded best to To investigate the rules that underlie the formation of
sweet tastes (e.g., 1 M glucose) and responded more in a these multimodal representations, analyses were per-
visual discrimination task to the visual stimulus that signi- formed on responses of these neurons in an olfactory dis-
fied sweet fruit juice than to the one that signified saline. crimination task, in which one set of odors delivered from
Another set responded to sweet taste and responded in an an olfactometer were associated if the correct lick response
olfactory discrimination task to fruit odor. An example of was made with the delivery of glucose taste reward and
one such bimodal neuron is shown in Figure 9. The neuron another set of odors was associated with the delivery of

Figure 9 The responses of a bimodal neuron recorded in the caudolateral orbitofrontal cortex. G, 1.0 M glucose; N, 0.1 M NaCl;
H, 0.01 M HCl; Q, 0.001 M quinine HCl; M, 0.1 M monosodium glutamate; Bj, 20% black currant juice; Tom, tomato juice; B, banana
odor; Cl, clove oil odor; On, onion odor; Or, orange odor; S, salmon odor; C, control no-odor presentation. The mean responses  SE
are shown. The neuron responded best to the tastes of NaCl and monosodium glutamate and to the odors of onion and salmon. (From
Rolls and Baylis, 1994.)
694 Rolls and Scott

the neuron responded reversed when the taste with which

it was associated reversed. Data from one such experiment
performed on a single neuron in the macaque orbitofrontal
cortex are shown in Figure 11. Extinction of the differen-
tial neuronal responses after task reversal was seen in 43%
of these neurons; i.e., these neurons simply stopped dis-
criminating between the two odors after the reversal. These
findings demonstrate directly a coding principle in primate
olfaction whereby the responses of some orbitofrontal cor-
tex olfactory neurons are modified by and depend upon the
taste with which the odor is associated.
It was of interest that this modification was less
complete, and much slower, than the modifications found
for orbitofrontal visual neurons during visual-taste reversal
(Rolls et al., 1996a). This relative inflexibility of olfactory
responses is consistent with the need for some stability in
odor-taste associations to facilitate the formation and
perception of flavors. In addition, some orbitofrontal
cortex olfactory neurons did not code in relation to the
taste with which the odor was associated (Critchley and
Rolls, 1996a), so that there is also a taste-independent rep-
resentation of odor in this region.
Many of the neurons with visual responses in this
region also show dependency of the visual response on the
taste with which the visual stimulus is associated. For
example, many of the visually responsive neurons in the
Figure 10 Examples of tracks made into the orbitofrontal cor-
orbitofrontal cortex responded only to a visual stimulus
tex in which taste (T) and olfactory (O) neurons were recorded associated with a rewarding taste (e.g., glucose as com-
close to each other in the same tracks. Some of the neurons were pared to saline); if the experimenter altered the taste with
bimodal (T/O). (From Rolls and Baylis, 1994.) which the visual stimulus was linked (i.e., reversal of the
visual discrimination task), these visual neurons typically
altered the visual stimulus to which they responded, so that
saline if the monkey incorrectly licked to that set of stim- their visual responses remained associated with a given
uli. It was found that 25% of the neurons responded to the taste. This part of the orbitofrontal cortex thus seems to
taste reward–associated odors more than to the saline pun- implement a mechanism that can alter flexibly the
ishment associated odors; that 35% responded to the saline responses to visual stimuli depending on the reinforcement
punishment–associated odors more than to the reward (e.g., the taste) associated with the visual stimulus (see
taste–associated odors; and that 40% of the orbitofrontal Thorpe et al., 1983). This enables prediction of the taste
cortex olfactory neurons did not categorize the odors based associated with ingestion of what is seen, and thus in the
on the taste with which they were associated (Critchley visual selection of foods (see Rolls, 1986, 1999). This cor-
and Rolls, 1996a).
In a further test of how the taste with which an odor is
associated influences the representation of odor in the
primate orbitofrontal cortex, associated was reversed in the *In an olfactory discrimination experiment, if the monkey makes
taste with which an odor was associated in a Go/NoGo a lick response when one odor, the S, is delivered, he obtains a
drop of glucose reward; if the monkey incorrectly makes a lick
olfactory discrimination task. Rolls et al. (1996a) found
response to another odor, the S, he obtains a drop of aversive
that 68% of orbitofrontal cortex odor-responsive neurons saline. At some time in the experiment, the contingency between
modified their responses in some way following changes the odor and the taste is reversed. The monkey relearns the dis-
in the taste reward associations of the odorants during crimination, showing reversal. It is of interest to investigate in
olfactory-taste discrimination reversals.* Full reversal of which parts of the olfactory system the neurons show reversal, for
the neuronal responses was seen in 25% of the neurons where they do, it can be concluded that the neuronal response to
analyzed. In other words, in these cases the odor to which the odor depends on the taste with which it is associated.
Central Taste Anatomy and Neurophysiology 695

Figure 11 The activity of a single

orbitofrontal olfactory neuron during the
performance of a two odor olfactory dis-
crimination task and its reversal. Before
reversal a lick to the odor of amyl acetate
produced a taste of salt, and after the
reversal a taste of glucose. Before reversal
a lick to the odor of cineole produced a
taste of glucose, and after the reversal a
taste of salt. Each point represents the
mean poststimulus activity of the neuron
to approximately 10 trials of the different
odorants. The standard errors of these
responses are shown. The odorants were
amyl acetate (initially S) and cineole
(initially S). After 80 trials of the task,
the taste reward associations of the stimuli
were reversed. This neuron reversed its
responses to the odorants following the
task reversal, showing that the neuron
learned to respond with a greater increase
of firing rate to whichever odor was asso-
ciated with salt taste. (From Rolls et al.,
1996.)

tical region is implicated more generally in a certain type The interactions between olfactory, visual, and taste stim-
of learning, namely in extinction and in the reversal of uli found in the orbitofrontal cortex may underlie the inter-
visual discriminations. It is suggested that the taste actions found in psychophysical experiments [such as the
neurons in this region are important for these functions, for finding that the rated flavor intensity of umami (protein)
they provide information about whether a reward has been was enhanced by adding methyl furyl disulfide (garlic
obtained (see Rolls, 1986, 1990, 1992a; Thorpe et al., odor) to monosodium glutamate (Rolls, 2000c; Rolls et al.,
1983). The ability of this part of the cortex to perform 1998)].
rapid learning of associations between visual stimuli and
primary reinforcers such as taste provides the basis for the B. Olfactory, Visual, and Taste Sensory-Specific
importance of this part of the brain in food-related and Satiety
emotion-related learning (see Rolls, 1990, 1992a, 1994,
1999). Consistent with this function, parts of the It has also been possible to investigate whether the olfac-
orbitofrontal cortex receive direct projections from the tory representation in the orbitofrontal cortex is affected by
inferior temporal visual cortex ( Barbas, 1988; Seltzer and hunger. In satiety experiments, Critchley and Rolls
Pandya, 1989), a region important in high-order visual (1996b) showed that the responses of some olfactory
information processing (Rolls, 1992b, 1999, 2000d; Rolls neurons to a food odor are decreased when the monkey is
and Treves, 1998). fed to satiety with a food (e.g., fruit juice) with that odor.
These results show that there are regions in the In particular, seven of nine olfactory neurons that were
orbitofrontal cortex of primates where the sensory responsive to the odors of foods, such as black currant
modalities of taste, vision, and olfaction converge; in many juice, were found to decrease their responses to the odor of
cases the neurons have corresponding sensitivities across the satiating food. The decrease was typically at least
modalities. It appears to be in these areas that flavor partly specific to the odor of the food that had been eaten
representations are built, where flavor is taken to mean a to satiety, potentially providing part of the basis for
representation best evoked by a combination of gustatory sensory-specific satiety. It was also found for eight of nine
and olfactory input. This orbitofrontal region does appear neurons that had selective responses to the sight of food
to be an important region for convergence, for there is only that they demonstrated a sensory-specific reduction in their
a low proportion of bimodal taste and olfactory neurons in visual responses to foods following satiation. These find-
the primary taste cortex (Rolls and Baylis, 1994). ings show that the olfactory and visual representations of
696 Rolls and Scott

food, as well as the taste representation of food, in the pri- To investigate whether the sensory-specific reduction in
mate orbitofrontal cortex are modulated by hunger. the responsiveness of the orbitofrontal olfactory neurons
Usually a component related to sensory-specific satiety might be related to a sensory-specific reduction in the plea-
can be demonstrated. sure produced by the odor of a food when it is eaten to sati-
These findings link at least part of the processing of ety, Rolls and Rolls (1997) measured human responses to
olfactory and visual information in this brain region to the the smell of a food eaten to satiety. It was found that the
control of feeding-related behavior. This is further pleasantness of the odor of a food, but much less signifi-
evidence that part of the olfactory representation in this cantly its intensity, was decreased when the subjects ate it to
region is related to the hedonic value of the olfactory satiety. It was also found that the pleasantness of the smell
stimulus and, in particular, that at this level of the olfactory of other foods (i.e., foods not eaten in the meal) showed
system in primates, the pleasure elicited by the food odor much less decrease. This finding has clear implications for
is at least part of what is represented. (1) the control of food intake, (2) ways to keep foods
As a result of the neurophysiological and behavioral presented in a meal appetizing, and (3) effects on odor-
observations showing the specificity of satiety in the pleasantness ratings that could occur following meals. In an
monkey, experiments were performed to determine whether investigation of the mechanisms of this odor-specific
satiety was specific to foods eaten in humans. It was found sensory-specific satiety, Rolls and Rolls (1997) allowed
that the pleasantness of the taste of food eaten to satiety humans to chew a food, without swallowing, for approxi-
decreased more than for foods that had not been eaten in the mately as long as the food is normally in the mouth during
meal (Rolls et al., 1981). One consequence of this is that if eating. They demonstrated a sensory-specific satiety with
one food is eaten to satiety, appetite reduction for other this procedure, showing that the sensory-specific satiety
foods is often incomplete, which will lead to enhanced eat- does not depend on food reaching the stomach. Thus, at least
ing when a variety of foods is offered (Rolls et al., 1981a,b, part of the mechanism is likely to be produced by a change
1984). Because sensory factors such as similarity of color, in processing in the olfactory pathways. The earliest stage of
shape, flavor, and texture are usually more important than olfactory processing at which this modulation occurs is not
metabolic equivalence in terms of protein, carbohydrate, yet known. It is unlikely to be in the receptors, because the
and fat content in influencing how foods interact in this change in pleasantness found was much more significant
type of satiety, it has been termed “sensory-specific satiety” than the change in the intensity (Rolls and Rolls, 1997).
(Rolls, 1990a; Rolls and Rolls, 1977, 1982; Rolls et al., When a variety of foods is available, the enhanced
1981a,b, 1982). It should be noted that this effect is distinct eating due to the operation of sensory-specific satiety may
from alliesthesia, in that alliesthesia is a change in the have been advantageous in evolution by ensuring that
pleasantness of sensory inputs produced by internal signals different foods with important different nutrients were
(such as glucose in the gut) (see Cabanac, 1971; Cabanac consumed. However, for humans today, when a wide
and Duclaux, 1970; Cabanac and Fantino, 1977), whereas variety of foods is readily available, this factor may lead to
sensory-specific satiety is a change in the pleasantness of overeating and obesity. In a test in the rat, it has been found
sensory inputs, which is accounted for at least partly by the that variety itself can lead to obesity (Rolls et al., 1983; see
external sensory stimulation received (such as the taste of a also Rolls and Hetherington, 1989).
particular food), in that, as shown above, it is at least partly
specific to the external sensory stimulation received. C. Taste and Texture, Including Astringency and the
The parallel between these studies of feeding in Texture of Fat
humans and of the neurophysiology of hypothalamic and
1. Texture
orbitofrontal cortex neurons in the monkey has been
extended by the observations that in humans, sensory- The orbitofrontal cortex of primates is also important as an
specific satiety occurs for the sight as well as for the taste of area of convergence for somatosensory inputs, related, for
food (Rolls et al., 1982). Further, to complement the finding example, to the texture of food, including that of fat in the
that in the hypothalamus neurons are found that respond mouth. It has been shown in recent recordings that single
differently to food and to water (see Rolls, 1999) and that neurons influenced by taste in this region can in some cases
satiety with water can decrease the responsiveness of hypo- have their responses modulated by the texture of the food.
thalamic neurons that respond to water, it has been shown This was shown in experiments in which the texture of food
that in humans motivation-specific satiety can also be was manipulated by the addition of methyl cellulose or gela-
detected. For example, satiety with water decreases the pleas- tine or by puréeing a semi-solid food (Rolls, 1997, 1999;
antness of the sight and taste of water but not of food (Rolls Rolls and Critchley, unpublished). The somatosensory
et al., 1983). inputs may reach this region via the rostral insula does pro-
Central Taste Anatomy and Neurophysiology 697

ject into this region, and which Mesulam and Mufson pure hydrocarbon) and to silicone oil [Si(CH3)2O)n]. Some of
(1982a,b; Mufson and Mesulam, 1982) have shown does the fat-related neurons do have convergent inputs from the
receive somatosensory inputs. chemical senses, in that in addition to taste inputs, some of
these neurons respond to the odor associated with a fat, such
as the odor of cream (Rolls et al., 1999). Feeding to satiety
2. Fat
with fat (e.g., cream) decreases the responses of these neu-
Texture in the mouth is an important indicator of whether fat rons to zero on the food eaten to satiety. This is a sensory-
is present in a food, which is important not only as a specific texture-dependent effect in that if the neuron
high-value energy source, but also as a potential source of receives a taste input from, for example, glucose taste, then
essential fatty acids. In the orbitofrontal cortex, Rolls et al. the response of the neuron to glucose is not decreased by
(1999) found a population of neurons that responds when fat feeding to satiety with cream. Thus, there is a representation
is in the mouth. An example of such a neuron is shown in of the macronutrient fat in this brain area, and the activation
Figure 12. The neuron illustrates that information about produced by fat is reduced by eating fat to satiety.
fat, as well as taste, can converge onto the same neuron in The perception of lipids is a function of both taste and
this region. The neuron responded to taste, in that its firing texture. Given that the taste system is crucial for the iden-
rate was significantly different within the group of tastants tification of nutrients, it is curious that it appeared for
sweet, salt, bitter, and sour. However, its response to fat in the many years to take no part in the perception of lipids, the
mouth was larger. The fat-related responses of these neurons most calorically rich of nutrients. Recently, however a
are produced, at least in part, by the texture of the food rather role of taste in lipid detection has been identified in rats
than by receptors sensitive to certain chemicals, in that such (Herness and Gilbertson, 1999). Lingual lipase reduces
neurons typically respond not only to foods such as cream triglycerides to mono- and diglycerides and free fatty
and milk containing fat, but also to paraffin oil (which is a acids, which are transported to the receptors by proteins

Figure 12 A neuron in the primate orbitofrontal cortex responding to the texture of fat in the mouth. The cell (Be047) increased its fir-
ing rate to cream (double and single cream), and responded to texture rather than the chemical structure of the fat in that it also responded
to 0.5 ml of silicone oil (Si(CH3)2O)n) or paraffin oil (hydrocarbon). The cell has a taste input too, in that it had a consistent but small
response to umami taste (monosodium glutamate, MSG). Gluc, glucose; NaCl, salt; HCl, sour; Q-HCL, quinine, bitter. The spontaneous
firing rate of the cell is also shown. (After Rolls et al., 1999.)
698 Rolls and Scott

produced by von Ebner’s glands (Kock et al., 1992). Free Critchley and Rolls (1996c) recorded from taste-respon-
fatty acids have been shown to inhibit delayed-rectifying sive neurons in the orbitofrontal cortex and adjacent insula.
potassium channels in taste cells (Liu et al., 1998), result- Single neurons were found that were tuned to respond to
ing in depolarization, increased intracellular calcium, and tannic acid (0.001 M) and represented a subpopulation of
a presumed release of neurotransmitter. The channels are neurons that was distinct from neurons responsive to the
sensitive only to essential fatty acids, reinforcing their tastes of glucose (sweet), NaCl (salty), HCl (sour), quinine
role in the detection of nutrients. However, this system is (bitter), and monosodium glutamate (umami). In addition,
unlikely to be involved in fat detection in humans, both across the population of taste-responsive neurons, tannic
because of the low concentration of lingual lipase in acid was as well represented as the tastes of NaCl, HCl,
humans (Herness and Gilbertson, 1999) and because fat in quinine, or monosodium glutamate. Multidimensional
the mouth can be detected very rapidly (1–2 sec), whereas scaling analysis of the neuronal responses to the tastants
the free fatty acid–detection system described has a time indicated that tannic acid lies outside the boundaries of the
course of many seconds. four conventional taste qualities (sweet, sour, bitter, and
salty) (Critchley and Rolls, 1996c). Taken together, these
data indicate that the astringent taste of tannic acid should
3. Astringency
be considered as a distinct flavor quality, which receives a
Another somatosensory input to the orbitofrontal cortex separate representation from sweet, salt, bitter, and sour in
that combines with taste inputs is produced by tannic acid, the primate cortical taste areas. Although the sensors for
which produces the “taste” of astringency. Tannic acid is a astringency are probably somatosensory rather that gusta-
member of the class of compounds known as polyphenols, tory, single neurons in the primate orbitofrontal cortex
which are present in a wide spectrum of plant matter, par- activated by tannic acid were also typically activated by
ticularly in foliage, the skin and husks of fruit and nuts, some taste inputs, so that convergence between the astrin-
and the bark of trees. The tannic acid in leaves is produced gency and taste systems does occur, and astringency can
as a defense against insects. There is less tannic acid in contribute to the flavor of food.
young leaves than in old leaves. Large monkeys cannot
obtain the whole of their protein intake from small
animals, insects, etc., and thus obtain some of their protein V. IMAGING STUDIES IN HUMANS
from leaves. Tannic acid binds protein (hence its use in
tanning) and amino acids, and thus prevents their absorp- Human neuroimaging studies using function magnetic
tion, rendering them indigestible. Thus it is adaptive for resonance imaging (fMRI) have shown that taste stimuli
monkeys to be able to taste tannic acid, so that they can activate an area of the anterior insula, which is probably
select food sources without too much tannic acid (Hladik, the primary taste cortex, and part of the orbitofrontal
1978 and personal communication). cortex, which is probably the secondary taste cortex
In humans, tannic acid elicits a characteristic astringent (Francis et al., 1999; Small et al., 1999). The orbitofrontal
taste. Oral astringency is perceived as the feeling of long- cortex taste area is distinct from areas activated by odors
lasting puckering and drying sensations on the tongue and and by pleasant touch (Francis et al., 1999). It has been
membranes of the oral cavity. High levels of tannic acid in shown that within individual subjects separate areas of the
some potential foods makes them unpalatable without orbitofrontal cortex are activated by sweet (pleasant) and
preparative techniques to reduce its presence (Johns and salt (unpleasant) tastes (O’Doherty et al., 2000b). Zald
Duquette, 1991), yet in small quantities tannic acid is com- et al. (1998) found activation of the human amygdala by
monly used to enhance the flavor of food. In this context the aversive taste of salt. Francis et al. (1999) also found
tannic acid is a constituent of a large range of spices and activation of the human amygdala by the taste of glucose.
condiments, such as ginger, chilies, and black pepper Extending this study, O’Doherty et al. (2000b) showed that
(Uma-Pradeep et al., 1993).* Tannic acid is a natural the human amygdala was as much activated by the
antioxidant by virtue of its chemical structure (see affectively pleasant taste of glucose as by the affectively
Critchley and Rolls, 1996c). negative taste of NaCl, and thus provided evidence that the
In order to investigate whether astringency is repre- human amygdala is not especially involved in processing
sented in the cortical taste areas concerned with taste, aversive, as compared to rewarding, stimuli.
It is of interest that, in humans, there is an area of the
*Tannic acid itself is not present in tea, yet a range of related far anterior insula that is activated by olfactory stimuli
polyphenol compounds are, particularly in green tea (Graham, (Francis et al., 1999; O’Doherty et al., 2000). It is not clear
1992). whether this area is separate from the region of the insula
Central Taste Anatomy and Neurophysiology 699

activated by taste. In macaques, the primary taste cortex (in beyond taste to olfactory, textural, thermal, and visual
the anterior insula and adjoining frontal operculum) does components of the stimulus, so that in the secondary taste
not appear to be strongly activated by olfactory stimuli cortex of primates flavor representations are built. The
(though further studies on this are in progress), and the responses of any neuron are usually kept in register across
human anterior insular olfactory area may thus correspond the modalities by associative learning. In particular, olfac-
to what, in macaques, is the caudal transitional area of the tory stimuli and visual stimuli become secondary or
orbitofrontal cortex where it adjoins the insula, area Ofdg, learned reinforcers because of their learned association in
where part of the secondary taste cortex is located (Baylis the primate orbitofrontal cortex with the primary rein-
et al., 1994). In humans, there is strong and consistent acti- forcer of taste or texture (see Secs. IV. A and IV. C).
vation of the right orbitofrontal cortex by olfactory stimuli Because neurons respond to combinations of inputs
(Francis et al., 1999; Zatorre et al., 1992). In an investiga- received through these different modalities, their tuning
tion of where the pleasantness of olfactory stimuli might can become quite selective. In addition, in the primate
be represented in humans, O’Doherty et al. (2000) showed secondary taste cortex, the reward or hedonic value of the
that the activation of an area of the orbitofrontal cortex to taste (and flavor) of food is represented, in that the
banana odor was decreased (relative to a control vanilla responses of neurons here decrease after the monkey has
odor) after bananas were eaten to satiety. Thus activity in a fed to satiety. This decrease is in part specific to the par-
part of the human orbitofrontal cortex olfactory area is ticular food eaten, and this sensory-specific satiety is
related to sensory-specific satiety. implemented in the orbitofrontal cortex. Sensory-specific
satiety may be implemented by a process of synaptic
habituation in the orbitofrontal cortex (but at no earlier
VI. THE NEURAL REPRESENTATION OF stage of processing), which has a time course of the 5–10
TASTE AND FLAVOR minutes that it may take to eat a food to satiety. The effect
of this functional architecture may be to allow in primates
Gustatory neurons are broadly tuned, a fundamental the hedonic or reward value of a food to decrease with
characteristic that demands that the afferent code for taste satiety, while at the same time an assessment of the iden-
quality be read across more than one neuron type. Such a tity and intensity of the taste, performed in the primary
distributed or ensemble code offers advantages of redun- taste cortex, would be largely unmodified. By this archi-
dancy, precision, and high information-carrying capacity tecture, stimulus identification remains distinct from
[discussed in Chapter 35 and by Rolls and Treves hedonics in primates, a phenomenon suggested by the
(1998)]. In addition, a distributed representation offers an results of psychophysical studies in humans.
advantage in forming associative memories, where asso- The evidence from rodents suggests that early analy-
ciations must be learned between stimuli in different ses of the taste input at the hindbrain level control
modalities, using a Hebbian associative synaptic modifi- reflexes for acceptance or rejection and autonomic
cation rule. Such an ensemble-encoded representation reflexes that anticipate digestion (nucleus of the solitary
allows many associations to be stored, yet enables the tract). These early analyses also enable the associative
important properties of generalization, and of graceful processes involved in conditioning (parabrachial
degradation upon sustaining damage, to be expressed nucleus). The thalamic relay may serve gustatory mem-
(Rolls and Treves, 1998). ories that do not arise from taste-gut interactions. The
Gustatory neurons in the primate remain rather broadly functional architecture of the rodent taste system may be
tuned through to the level of the primary taste cortex (see rather different from that of primates, in that rodents do
Rolls, 1997; Scott and Plata-Salamán, 1999). The highly not have obligatory routing of taste information from the
distributed nature of the representation here may be nucleus of the solitary tract to the thalamus and this to the
appropriate for a representation of the identity of the taste cortex, but have additional connections from the nucleus
stimulus, that is, exactly how sweet, bitter, salt, sour, or of the solitary tract to the parabrachial nucleus, which in
umami-tasting it is, together with information about its turn has connections with many subcortical structures.
intensity. Consistent with this, hunger and sensory-spe- Further, and probably related to this difference in the
cific satiety do not affect the firing to tastants of neurons anatomy, the effects of satiety on taste seem to be imple-
in the primate primary taste cortex (see Sec. III. B), so the mented at least in part in lower order parts of the taste
representation can be about the identity and strength of system of rodents such as the nucleus of the solitary tract,
the tastant. In the secondary taste cortex in the OFC, the and consistent with this, the identity and intensity of taste
neurons become more selective to taste quality, but are in rodents do not appear to be kept very distinct from the
also likely to be multimodal, extending their sensitivity hedonic value of the taste.
700 Rolls and Scott

REFERENCES Booth M.C.A., and Rolls E. T. (1998). View-invariant representa-

tions of familiar objects by neurons in the inferior temporal
Aggleton, J. P., and Passingham, R. E. (1981). Stereotaxic visual cortex. Cerebral Cortex 8:510–523.
surgery under x-ray guidance in the rhesus monkey, with Burton, M. J., Rolls, E. T., and Mora, F. (1976). Effects of hunger
special reference to the amygdala. Exp. Brain Res. 44: on the responses of neurones in the lateral hypothalamus to
271–276. the sight and taste of food. Exp. Neurol. 51:668–677.
Amaral, D. G., Price, J. L., Pitkanen, A., and Carmichael, S. T. Cabanac, M. (1971). Physiological role of pleasure. Science
(1992). Anatomical organization of the primate amygdaloid 173:1103–1107.
complex. In The Amygdala, J. P. Aggleton (Ed.). Wiley-Liss, Cabanac, M., and Duclaux, R. (1970). Specificity of internal sig-
New York, pp. 1–66. nals in producing satiety for taste stimuli. Nature
Augustine, J. R. (1996). Circuitry and functional aspects of the 227:966–967.
insular lobe in primates, including humans. Brain Res. Rev. Cabanac, M., and Fantino, M. (1977). Origin of olfacto-gustatory
22:229–244. alliesthesia: Intestinal sensitivity to carbohydrate concentra-
Aziz, Q., Furlong, P. L., Barlow, J., Hobson, A., Alani, S., tion? Physiol. Behav. 10:1039–1045.
Bancewicz, J., Ribands, M., Harding, G. F. A., and Carmichael, S. T., and Price, J. L. (1995). Sensory and premotor
Thompson, D. G. (1995). Topographic mapping of cortical connections of the orbital and medial prefrontal cortex of
potentials evoked by distension of the human proximal and macaque monkeys. J. Comp. Neurol. 363:642–664.
distal oesophagus. Electroencephalogr. Clin. Neurophysiol. Carmichael, S. T., Clugnet, M. C., and Price, J. L. (1994). Central
96:219–229. olfactory connections in the macaque monkey. J. Comp.
Baird, J. P., Travers, J. B., and Travers, S. P. (2000). Effects of Neurol. 346:403–434.
ICV NPY administration on sucrose taste reactivity. Meeting Chaudhari, N., Landin, A. M., and Roper, S. D. (2000).
of the Association for Chemoreception Sciences, Sarasota, A metabotropic glutamate receptor variant functions as a taste
FL. receptor. Nature Neurosci. 3:113–119.
Barbas, H. (1988). Anatomic organization of basoventral and Chang, F. C. T., and Scott, T. R. (1984). Conditioned taste aver-
mediodorsal visual recipient prefrontal regions in the rhesus sions modify neural responses in the rat nucleus tractus soli-
monkey. J. Comp. Neurol. 276:313–342. tarius. J. Neurosci. 4:1850–1862.
Barbas, H. (1993). Organization of cortical afferent input to the Collings, V. B. (1974). Human taste response as a function of
orbitofrontal area in the rhesus monkey. Neuroscience locus of stimulation on the tongue and soft palate. Percept.
56:841–864. Psychophys. 16:169–174.
Barbas, H. (1995). Anatomic basis of cognitive-emotional inter- Contreras, R., and Frank, M. (1979). Sodium deprivation alters
actions in the primate prefrontal cortex. Neurosci. Biobehav. neural response to gustatory stimuli. J. Gen. Physiol.
Rev. 19:499–510. 73:569–594.
Barbas, H., Pandya, D. N. (1989). Architecture and intrinsic con- Critchley, H. D., and Rolls, E. T. (1996a). Olfactory neuronal
nections of the prefrontal cortex in the rhesus monkey. responses in the primate orbitofrontal cortex: analysis in an
J. Comp. Neurol. 286:353–375. olfactory discrimination task. J. Neurophysiol. 75:1659–1672.
Baylis, L. L., and Gaffan, D. (1991) Amygdalectomy and ventro- Critchley, H. D., and Rolls, E. T. (1996b). Hunger and satiety
medial prefrontal ablation produce similar deficits in food modify the responses of olfactory and visual neurons in the
choice and in simple object discrimination learning for an primate orbitofrontal cortex. J. Neurophysiol. 75:1673–1686.
unseen reward. Exp. Brain Res. 86:617–622. Critchley, H. D., and Rolls, E. T. (1996c). Responses of primate
Baylis, L. L., and Rolls, E. T. (1991). Responses of neurons in taste cortex neurons to the astringent tastant tannic acid.
the primate taste cortex to glutamate. Physiol. Behav. 49: Chem. Senses 21:135–145.
973–979. Erickson, R. P. (1985). Grouping in the chemical senses. Chem.
Baylis, L. L., Rolls, E. T., and Baylis, G. C. (1994). Afferent con- Senses 10:333–340.
nections of the orbitofrontal cortex taste area of the primate. Fiol, M. E., Leppik, I. E., Mireles, R., and Maxwell, R. (1988).
Neuroscience 64:801–812. Ictus emeticus and the insular cortex. Epilepsy Res. 2: 127–131.
Beckstead, R. M., and Norgren, R. (1979). An autoradiographic Flynn, F. W., Grill, H. J., Schwartz, G. J., Norgren, R. (1991a).
examination of the central distribution of the trigeminal, Central gustatory lesions. 1. Preference and taste reactivity
facial, glossopharyngeal, and vagal nerves in the monkey. tests. Behav. Neurosci. 105:933–943.
J. Comp. Neurol. 184:455–472. Flynn, F. W., Grill, H. J., Schulkin, J., and Norgren, R. (1991b).
Beckstead, R. M., Morse, J. R., and Norgren, R. (1980). The Central gustatory lesions. 2. Effects of sodium appetite, taste
nucleus of the solitary tract in the monkey: projections to the aversion learning and feeding behaviors. Behav. Neurosci.
thalamus and brainstem nuclei. J. Comp. Neurol. 190: 259–282. 105:944–954.
Bester, H., Bourgeais, L., Villanueva, L., Besson, J. M., and Francis, S., Rolls, E. T., Bowtell, R., McGlone, F., O’Doherty,
Bernard, J. F. (1999). Differential projections to the J., Browning, A., Clare, S. and Smith, E. (1999). The rep-
intralaminar and gustatory thalamus from the parabrachial resentation of pleasant touch in the brain and its relation-
area: A PHA-L study in the rat. J. Comp. Neurol. 405: ship with taste and olfactory areas. NeuroReport 10:
421–449. 453–459.
Central Taste Anatomy and Neurophysiology 701

Gibbs, J., Maddison, S. P., and Rolls, E. T. (1981). Satiety role Hajnal, A., Takenouchi, K., and Norgren, R. (1999). Effect of
of the small intestine examined in sham-feeding rhesus intraduodenal lipid on parabrachial gustatory coding in awake
monkeys. J. Comp. Physiol. Psychol. 95:1003–1015. rats. J. Neurosci. 19:7182–7190.
Giza, B. K., and Scott, T. R. (1983). Blood glucose selectively Hasselmo, M. E., Rolls, E. T., and Baylis, G. C. (1989a). The role
affects taste evoked activity in the rat nucleus tractus solitar- of expression and identity in the face-selective responses of
ius. Physiol. Behav. 31:643–650. neurons in the temporal visual cortex of the monkey. Behav.
Giza, B. K., and Scott, T. R. (1987a). Intravenous insulin Brain Res. 32:203–218.
infusions in rats decrease gustatory-evoked responses to Hasselmo, M. E., Rolls, E. T., Baylis, G. C., and Nalwa, V.
sugars. Am. J. Physiol. 252:R994–R1002. (1989b). Object-centered encoding by face-selective neurons
Giza, B. K., and Scott, T. R. (1987b). Blood glucose level affects in the cortex in the superior temporal sulcus of the monkey.
perceived sweetness intensity in rats. Physiol. Behav. Exp. Brain Res. 75:417–429.
41:459–464. Herness, M. S., and Gilbertson, T. A. (1999). Cellular mechan-
Giza, B. K., Scott, T. R., and Van der Weele, D. A. (1992). isms of taste transduction. Ann. Rev. Physiol. 61:873–900.
Administration of satiety factors and gustatory responsiveness Hinton, G. E., McClelland, J. L., and Rumelhart, D. E. (1986).
in the nucleus tractus solitarius of the rat. Brain Res. Bull. Distributed representations. In Parallel Distributed
28:637–639. Processing, Vol. 1: Foundations, D. E. Rumehlart and J. L.
Giza, B. K., Deems, R. O., Van der Weele, D. A., and Scott, T. R. McLelland (Eds.). MIT Press, Cambridge, MA, pp. 77–109.
(1993). Pancreatic glucagon suppresses gustatory responsive- Hladik, C. M. (1978). Adaptive strategies of primates in relation
ness to glucose. Am. J. Physiol. 265:R1231–R1237. to leaf-eating. In The Ecology of Arboreal Folivores, G. G.
Giza, B. K., Ackroff, K., McCaughey, S. A., Sclafani, A., and Montgomery (Ed.). Smithsonian Institute Press, Washington
Scott, T. R. (1997a). Preference conditioning alters taste DC, pp. 373–395.
responses in the nucleus of the solitary tract of the rat. Am. J. Ikeda, K. (1909). On a new seasoning. J. Tokyo Chem. Soc.
Physiol. 273:R1230–R1240. 30:820–836.
Giza, B. K., McCaughey, S. A., Scott, C. L., Zhang, L., and Scott, Insausti, R., Amaral, D. G., and Cowan, W. M. (1987). The
T. R. (1997b). Intravenous glucose injections suppress entorhinal cortex of the monkey. II. Cortical afferents.
responses of sugar-sensitive taste neurons in the rat NST. J. Compar. Neurol. 264:356–395.
Neurosci. Abstr. 23:1037. Jacobs, K. M., Mark, G. P., and Scott, T. R. (1988). Taste
Glenn, J. F., and Erickson, R. P. (1976). Gastric modulation of responses in the nucleus tractus solitarius of sodium-deprived
gustatory afferent activity. Physiol. Behav. 16:561–568. rats. J. Physiol. 406:393–410.
Graham, H. N. (1992). Green tea composition, consumption and Johnson, T. N., Rosvold, H. E., and Mishkin, M. (1968)
polyphenol chemistry. Prev. Med. 21:334–350. Projections from behaviorally defined sectors of the prefrontal
Grigson, P. S., Shimura, T., and Norgren, R. (1997). Brainstem cortex to the basal ganglia, septum and diencephalon of the
lesions and gustatory function. 3. The role of the nucleus of monkey. Exp. Neurol. 21:20–34.
the solitary tract and the parabrachial nucleus in retention of a Jones, B., and Mishkin, M. (1972). Limbic lesions and the prob-
conditioned taste aversion in rats. Behav. Neurosci. lems of stimulus-reinforcement. Exp. Neurol. 36:362.
111:180–187. Johns, T., and Duquette, M. (1991). Detoxification and mineral
Grigson, P. S., Reilly, S., Shimura, T., and Norgren, R. (1998). supplementation as functions of geophagy. Am. J. Clin. Nutr.
Ibotenic acid lesions of the parabrachial nucleus and condi- 53:448–456.
tioned taste aversion: further evidence for an associative Karadi, Z., Oomura, Y., Nishino, H., Scott, T. R., Lenard, L., and
deficit in rats. Behav. Neurosci. 112:160–171. Aou, S. (1992). Responses of lateral hypothalamic glucose-
Grigson, P. S., Lyuboslavsky, P., and Tanase, D. (2000). Bilateral sensitive and glucose-insensitive neurons to chemical stimuli
lesions of the gustatory thalamus disrupt morphine- but not in behaving rhesus monkeys. J. Neurophysiol. 67:389–400.
LiCl-induced intake suppression in rats: evidence against the Karadi, Z., Faludi, B., Lenard, L., Czurko, A., Niedetsky, C.,
conditioned taste aversion hypothesis. Brain Res. Vida, I., and Nishino, H. (1995). Glucose-sensitive neurons of
858:327–337. the globus pallidus. 2. Complex functional attributes. Brain
Grill, H. J., and Kaplan, J. M. (1990). Caudal brain stem partici- Res. Bull. 37:157–162.
pates in the distributed neural control of feeding. In Handbook Kawamura, Y., and Kare, M. R. (Eds.) (1987). Umami: A Basic
of Behavioral Neurobiology, Vol. 10, Neurobiology of Food Taste. Marcel Dekker, New York.
and Fluid Intake, E. M. Stricker (Ed.). Plenum Press, New Kemp, J. M., and Powell, T. P. S. (1970). The cortico-striate
York, pp. 125–149. projections in the monkey. Brain 93:525–546.
Grill, H. J., and Norgren, R. (1978). The taste reactivity test. II. Kock, K., Bläker, M., and Schmale, H. (1992). Postnatal
Mimetic responses to gustatory stimuli in chronic thalamic development of von Ebner’s glands: accumulation of a protein
and chronic decerebrate rats. Brain Res. 143:281–297. on the lipocalin superfamily in taste papillae of rat tongue.
Grill, H. J., Friedman, M. I., Norgren, R., Scalera, G., and Seeley, Cell Tissue Res. 267:313–320.
R. (1995). Parabrachial nucleus lesions impair feeding Kuznicki, J. T., and Ashbaugh, N. (1979). Taste quality differ-
response elicited by 2,5-anhydro-D-mannitol. Am. J. Physiol. ences within the sweet and salty taste categories. Sensory
37:R676–R682. Proc. 3:157–182.
702 Rolls and Scott

Liu, L., Kim, I., Hu, S., Wang, S., Shang, H., and Gilbertson, T. A. Oppenheimer, S. M., Gelb, A., Girvin, J. P., and Hachinski, V. C.
(1998). Identification of a Shaker Kv 1.5-like K channel in (1992). Cardiovascular effects ofhuman insular cortical stim-
taste cells: the primary target for fatty acid inhibition. Chem. ulation. Neurology 42:1927–1932.
Senses, in press. Pandya, D. N., and Yeterian, E. H. (1996). Comparison of pre-
Louis-Sylvestre, J., Giachetti, I., and LeMagnen, J. (1983). frontal architecture and connections. Phil. Trans. Roy. Soc.
Vagotomy abolishes the differential palatability of food. Lond. B. 351:1423–1431.
Appetite 4:295–299. Perrotto, R. S., and Scott, T. R. (1976). Gustatory neural coding
McCaughey, S. A., and Scott, T. R. (2000). Am. J. Physiol., in in the pons. Brain Res. 110:283–300.
press. Petrides, M., and Pandya, D. N. (1994). Comparative architec-
Mesulam, M. M., and Mufson, E. J. (1982a). Insula of the Old tonic analysis of the human and macaque frontal cortex. In
World monkey. I. Architectonics in the insulo-orbito-temporal Handbook of Neuropsychology, F. Boller and J. Grafman
component of the paralimbic brain. J. Comp. Neurol. 212: 1–22. (Eds.). Elsevier Science, Amsterdam, pp. 17–58.
Mesulam, M. M., and Mufson, E. J. (1982b). Insula of the old Phillips, M. L., Young, A. W., Senior, C., Brammer, M., Andrew,
world monkey. III: Efferent cortical output and comments on C., Calder, A. J., Bullmore, E. T., Perrett, D. I., Rowland, D.,
function. J. Comp. Neurol. 212:38–52. Williams, S. C., Gray, J. A., and David, A. S. (1997). A spe-
Morecraft R. J., Geula C., and Mesulam M. M. (1992). cific neural substrate for perceiving facial expressions of dis-
Cytoarchitecture and neural afferents of orbitofrontal cortex gust. Nature 389:495–498.
in the brain of the monkey. J. Comp. Neurol. 323:341–358. Plata-Salaman, C. R., Scott, T. R. and Smith-Swintosky, V. L.
Mufson, E. J., and Mesulam, M. M. (1982). Insula of the Old (1992). Gustatory neural encoding in the monkey cortex: L-
World monkey. II. Afferent cortical input and comments on amino-acids. J. Neurophysiol. 67:1552–1561.
the claustrum. J. Comp. Neurol. 212:23–37. Powley, T. L., and Berthoud, H. R. (1991). Neuroanatomical
Nakano, Y., Oomura, Y., Lenard, L., Nishino, H., Aou, S., bases of cephalic phase reflexes. In Appetite and Nutrition:
Yamamoto, T., and Aoyagi, K. (1986). Feeding-related activ- Chemical Senses IV, M. I. Freidman, M. G. Tordoff, and M.
ity of glucose- and morphine-sensitive neurons in the monkey R. Kare (Eds.). Marcel Dekker, New York, pp. 391–404.
amygdala. Brain Res. 399:167–172. Powley, T. L., Fox, E. A., and Berthoud, H. R. (1987).
Nauta, W. J. H. (1964). Some efferent connections of the pre- Retrograde tracer technique for assessment of selective and
frontal cortex in the monkey. In The Frontal Granular Cortex total subdiaphragmatic vagotomies. Am. J. Physiol. 253:
and Behavior, J. M. Warren, K. Akert, K (Eds.). McGraw- R361–R370.
Hill, New York, pp. 397–407. Price, J. L. (2001). Anatomy and neurochemistry of central olfac-
Nishijo, H., Ono, T., and Nishino, H. (1988a). Single neuron tory structures. In Handbook of Olfaction and Gustation, 2nd
responses in amygdala of alert monkey during complex sen- ed. Marcel Dekker, New York.
sory stimulation with affective significance. J. Neurosci. Price, J. L., Carmichael, S. T., and Drevets, W. C. (1996).
8:3570–3583. Networks related to the orbital and medial prefrontal cortex; a
Nishijo, H., Ono, T., and Nishino, H. (1988b). Topographic dis- substrate for emotional behavior? Prog. Brain Res.
tribution of modality-specific amygdalar neurons in alert 107:523–536.
monkey. J. Neurosci. 8:3556–3569. Pritchard, T. C., Hamilton, R. B., Morse, J. R., and Norgren, R.
Norgren, R. (1970). Gustatory responses in the hypothalamus. (1986). Projections of thalamic gustatory and lingual areas in
Brain Res. 21:63. the monkey, Macaca fascicularis. J. Comp. Neurol.
Norgren, R. (1976). Taste pathways to hypothalamus and amyg- 244:213–228.
dala. J. Comp. Neurol. 166:17–30. Pritchard, T. C., Hamilton, R. B., and Norgren, R. (1989). Neural
Norgren, R. (1984). Central neural mechanisms of taste. In coding of gustatory information in the thalamus of Macaca
Handbook of Physiology—The Nervous System III, Sensory mulatta. J. Neurophysiol. 61:1–14.
Processes 1, I. Darien-Smith (Ed.). American Physiological Raichle, M. E. (1991). Memory mechanisms in the processing of
Society, Washington DC, pp. 1087–1128. words and word-like symbols. Exploring brain functional
Norgren, R., and Leonard, C. M. (1973). Ascending central gus- anatomy with positron tomography. Ciba Foundation
tatory pathways. J. Comp. Neurol. 150:217–238. Symposium 163:198–204.
O’Doherty, J., Rolls, E. T., Francis, S., Bowtell, R., McGlone, F., Rauch, S. L., Savage, C. R., Alpert, N. M., Miguel, E. C., Baer,
Kobal, G., Renner, B., and Ahne, G. (2000). Sensory-specific L., Breiter, H. C., Fischman, A. J., Manzo, P. A., Moretti, C.,
satiety related olfactory activation of the human orbitofrontal and Jenike, M. A. (1995). A positron emission tomographic
cortex. Neuro Report 11:893–897. study of simple phobic symptom provocation. Arch. Gen.
O’Doherty, J., Rolls, E. T., Francis, S., McGlone, F., and Bowtell, Psychiatry 52:20–28.
R. (2001). Representation of pleasant and aversive taste in the Reilly, S. (1998). The role of the gustatory thalamus in taste-
human brain. J. Neurophysiol. 85:1315–1321. guided behavior. Neurosci. Biobehav. Rev. 22:883–901.
Oomura, Y., Nishino, H., Karadi, Z., Aou, S., and Scott, T. R. Reilly, S., and Pritchard, T. C. (1995). The effect of thalamic
(1991). Taste and olfactory modulation of feeding related neu- lesions on primate taste preference. Exp. Neurol. 135:
rons in the behaving monkey. Physiol. Behav. 49:943–950. 56–66.
Central Taste Anatomy and Neurophysiology 703

Reilly, S., and Pritchard, T. C. (1996a). Gustatory thalamus Rolls, E. T. (1999). The Brain and Emotion. Oxford University
lesions in the rat. 1. Innate taste preferences and aversions. Press, Oxford.
Behav. Neurosci. 110:737–745. Rolls, E. T. (2000a). The orbitofrontal cortex and reward.
Reilly, S., and Pritchard, T. C. (1996b). Gustatory thalamus Cerebral Cortex 10:284–294.
lesions in the rat. 2. Aversive and appetitive taste conditioning. Rolls, E. T. (2000b). Taste, olfactory, visual and somatosensory
Behav. Neurosci. 110:746–759. representations of the sensory properties of foods in the brain,
Reilly, S., and Trifunovic, R. (1999). Gustatory thalamus lesions and their relation to the control of food intake. In Neural and
eliminate successive negative contrast in rats. Behav. Metabolic Control of Macronutrient Intake, H. R. Berthoud
Neurosci. 113:1008–1119. and R. J. Seeley (Eds.). CRC Press, Boca-Raton, FL, pp.
Reilly, S., Grigson, P. S., and Norgren, R. (1993). Parabrachial 247–262.
nucleus lesions and conditioned taste aversion. Evidence sup- Rolls, E. T. (2000c). The representation of umami taste in the
porting an associative deficit. Behav. Neurosci. taste cortex. J. Nutrition 130:9605–9655.
107:1005–1017. Rolls, E. T. (2000d). Functions of the primate temporal lobe cor-
Robinson, C. J., and Burton, H. (1980). Organization of tical visual areas in invariant visual object and face recogni-
somatosensory receptive fields in cortical areas 7b, retroinsu- tion. Neuron 27:1–20.
lar, postauditory and granular insula of M. fascicularis. J. Rolls, E. T. (2000e). Neurophysiology and functions of the pri-
Comp. Neurol. 192:69–92. mate amygdala, and the neural basis of emotion. In The
Rolls, B. J. (1990a). The role of sensory-specific satiety in food Amygdala: A Functional Analysis, J. P. Aggleton (Ed.).
intake and food selection. In Taste, Experience, and Feeding, Oxford University Press, Oxford, pp. 447–478.
E. D. Capaldi and T. L. Powley (Eds.). American Rolls, E. T., and Baylis, L. L. (1994). Gustatory, olfactory and
Psychological Association, Washington, DC, pp. 197–209. visual convergence within the primate orbitofrontal cortex.
Rolls, B. J., and Hetherington, M. (1989). The role of variety in J. Neurosci. 14:5437–5452.
eating and body weight regulation. In Handbook of the Rolls, E. T., and Rolls, B. J. (1977). Activity of neurons in sen-
Psychophysiology of Human Eating, R. Shepherd (Ed.). sory, hypothalamic and motor areas during feeding in the
Chichester, Wiley, pp. 57–84. monkey. In Food Intake and Chemical Senses, Y. Katsui, M.
Rolls, B. J., Rolls, E. T., Rowe, E. A., and Sweeney, K. (1981a). Sato, S. Takagi, and Y. Oomura (Eds.). University of Tokyo
Sensory specific satiety in man. Physiol. Behav. 27:137–142. Press, Tokyo, pp. 525–549.
Rolls, B. J., Rowe, E. A., Rolls, E. T., Kingston, B., and Megson, Rolls, E. T., and Rolls, B. J. (1982). Brain mechanisms involved
A. (1981b). Variety in a meal enhances food intake in man. in feeding. In Psychobiology of Human Food Selection, L. M.
Physiol. Behav. 26:215–221. Barker (Ed.). AVI Publishing Company, Westport, CT, pp.
Rolls, B. J., Rowe, E. A., and Rolls, E. T. (1982). How sensory 33–62.
properties of foods affect human feeding behavior. Physiol. Rolls, E. T., and Rolls, J. H. (1997). Olfactory sensory-specific
Behav. 29:409–417. satiety in humans. Physiol. Behav. 61:461–473.
Rolls, B. J., Van Duikenvoorde, P. M., and Rolls, E. T. (1984). Rolls, E. T., and Treves, A. (1998). Neural Networks and Brain
Pleasantness changes and food intake in a varied four course Function. Oxford University Press, Oxford.
meal. Appetite 5:337–348. Rolls, E. T., Burton, M. J., and Mora, F. (1980).
Rolls, E. T. (1975). The Brain and Reward. Pergamon, Oxford. Neurophysiological analysis of brain-stimulation reward in
Rolls, E. T. (1986). Neuronal activity related to the control of the monkey. Brain Res. 194:339–357.
feeding. In Feeding Behaviour: Neural and Humoral Rolls, E. T., Rolls, B. J., and Rowe, E. A. (1983). Sensory-spe-
Controls, R. C. Ritter, S. Ritter and C. D. Barnes (Eds.). cific and motivation-specific satiety for the sight and taste of
Academic Press, New York, pp. 163–190. food and water in man. Physiol. Behav. 30:185–192.
Rolls, E. T. (1990b). A theory of emotion, and its application to Rolls, E. T., Murzi, E., Yaxley, S., Thorpe, S. J., and Simpson, S.
understanding the neural basis of emotion. Cognition Emotion J. (1986). Sensory-specific satiety: food-specific reduction in
4:161–190. responsiveness of ventral forebrain neurons after feeding in
Rolls, E. T. (1992a). Neurophysiology and functions of the pri- the monkey. Brain Res. 368:79–86.
mate amygdala. In The Amygdala, J. P. Aggleton (Ed.). Wiley- Rolls, E. T., Scott, T. R., Sienkiewicz, Z. J., and Yaxley, S. (1988).
Liss, New York, pp. 143–165. The responsiveness of neurones in the frontal opercular gus-
Rolls, E. T. (1992b). Neurophysiological mechanisms under- tatory cortex of the macaque monkey is independent of
lying face processing within and beyond the temporal hunger. J. Physiol. 397:1–12.
cortical visual areas. Phil. Trans. Roy. Soc. Lond. B 335: Rolls, E. T., Sienkiewicz, Z. J., and Yaxley, S. (1989). Hunger
11–21. modulates the responses to gustatory stimuli of single neurons
Rolls, E. T. (1994). Neurophysiology and cognitive functions of in the orbitofrontal cortex. Eur. J. Neurosci. 1:53–60.
the striatum. Rev. Neurol. (Paris) 150:648–660. Rolls, E. T., Yaxley, S., and Sienkiewicz, Z. J. (1990).
Rolls, E. T. (1997). Taste and olfactory processing in the brain Gustatory responses of single neurons in the orbitofrontal
and its relation to the control of eating. Crit. Rev. Neurobiol. cortex of the macaque monkey. J. Neurophysiol. 64:
11:263–287. 1055–1066.
704 Rolls and Scott

Rolls, E. T., Critchley, H., Mason, R., and Wakeman, E. A. Shimura, T., Grigson, P. S., and Norgren, R. (1997a). Brain stem
(1996a). Orbitofrontal cortex neurons: role in olfactory and lesion and gustatory function. 1. The role of the nucleus of the
visual association learning. J. Neurophysiol. 75:1970–1981. solitary tract during a brief intake test in rats. Behav.
Rolls, E. T., Critchley, H., Mason, R., and Wakeman, E. A. Neurosci. 111:155–168.
(1996b). Responses of neurons in the primate taste cortex to Shimura, T., Tanaka, H., and Yamamoto, T. (1997b). Salient
the glutamate ion and to inosine 5-monophosphate. Physiol. responsiveness of parabrachial neurons to the conditioned
Behav. 59:991–1000. stimulus after the acquisition of taste aversion learning in rats.
Rolls, E. T., Critchley, H. D., Browning, A., and Hernadi, I. Neuroscience 81:239–247.
(1998). The neurophysiology of taste and olfaction in pri- Small, D. M., Zald, D. H., Jones-Gotman, M., Zatorre, R. J.,
mates, and umami flavor. Ann. Acad. Sci. 855:426–437. Pardo, J. V., Frey, S. and Petrides, M. (1999). Human cortical
Rolls, E. T., Critchley, H. D., Browning, A. S., Hernadi, A. and gustatory areas: A review of functional neuroimaging data.
Lenard, L. (1999). Responses to the sensory properties of fat NeuroReport 10:7–14.
of neurons in the primate orbitofrontal cortex. J. Neurosci. Steiner, J. E. (1979). Human facial expressions in response to
19:1532–1540. taste and smell stimulation. In Advances in Child
Roper, S. N., Lévesque, M. F., Sutherling, W. W., and Engle, J. Jr. Development, H. W. Reese and L. Lipsett (Eds.). Academic
(1993). Surgical treatment of partial epilepsy arising from the Press, New York, pp. 57–295.
insular cortex. J. Neurosurg. 79:226–229. Tanabe, T., Yarita, H., Iino, M. Ooshima, Y., and Takagi, S. F.
Sakai, N., and Yamamoto, T. (1998). Role of the medial and (1975a). An olfactory projection area in orbitofrontal cortex
lateral parabrachial nucleus in acquisition and retention of of the monkey. J. Neurophysiol. 38:1269–1283.
conditioned taste aversion in rats. Behav. Brain Res. 93: Tanabe, T., Iino, M., and Takagi, S. F. (1975b). Discrimination of
63–70. odors in olfactory bulb, pyriform-amygdaloid areas, and
Sanghera, M. K., Rolls, E. T. and Roper-Hall, A. (1979). Visual orbitofrontal cortex of the monkey. J. Neurophysiol.
responses of neurons in the dorsolateral amygdala of the alert 38:1284–1296.
monkey. Exp. Neurol. 63:610–626. Thompson, D. R., Moskowitz, H. R., and Campbell, R. G.
Scalera, G., Grigson, P. S., and Norgren, R. (1997). Gustatory (1976). Effects of body weight and food intake on pleasant-
functions, sodium appetite, and conditioned taste aversion ness for a sweet stimulus. J. Appl. Physiol. 41:77–83.
survive excitotoxic lesions of the thalamic taste area. Behav. Thorpe, S. J., Rolls, E. T. and Maddison, S. (1983). Neuronal
Neurosci. 111:633–645. activity in the orbitofrontal cortex of the behaving monkey.
Schneider, R. J., Friedman, D. P., and Mishkin, M. (1993). Exp. Brain Res. 49:93–115.
A modality-specific somatosensory area within the insula of Turner, B. H., Mishkin, M., and Knapp, M. (1980). Organization of
the rhesus monkey. Brain Res. 621:116–120. the amygdalopetal projections from modality-specific cortical
Scott, T. R., and Erickson, R. P. (1971). Synaptic processing of association areas in the monkey. J. Comp. Neurol. 191: 515.
taste-quality information in the thalamus of the rat. Uma-Pradeep, K., Geervani, P., and Eggum, B. O. (1993). Common
J. Neurophysiol. 34:868–884. Indian spices: nutrient composition, consumption and contribu-
Scott, T. R., and Giza, B. K. (1987). A measure of taste intensity tion to dietary value. Plant Foods Hum. Nutr. 44:138–148.
discrimination in the rat through conditioned taste aversions. Verhagen, J. V., Giza, B. K., and Scott, T. R. (1999). Taste in the
Physiol. Behav. 41:315–320. rat thalamus. Neurosci. Abstr. 25:2184.
Scott, T. R., and Perrotto, R. S. (1980). Intensity coding in the Wallis, G., and Rolls, E. T. (1997). Invariant face and object recog-
pontine taste area: gustatory information is processed simi- nition in the visual system. Prog. Neurobiol. 51:167–194.
larly throughout rat’s brainstem. J. Neurophysiol. 44: Weusten, B. L. A. M., Franssen, H., Wieneke, J. H., and Smout,
739–750. A. J. P. M. (1994). Multichannel recording of cerebral poten-
Scott, T. R., and Plata-Salamán, C. R. (1999). Taste in the mon- tials evoked by esophageal balloon distention in humans.
key cortex. Physiol. Behav. 67:489–511. Digest. Dis. Sci. 39:2074–2083.
Scott, T. R., Yaxley, S., Sienkiewicz, Z. J., and Rolls, E. T. Wilkins, L., and Richter, C. P. (1940). A great craving for salt by
(1986a). Taste responses in the nucleus tractus solitarius of the a child with corticoadrenal insufficiency. J. Am. Med. Assoc.
behaving monkey. J. Neurophysiol. 55:182–200. 114:866–868.
Scott, T. R., Yaxley, S., Sienkiewicz, Z. J., and Rolls, E. T. Williams, G. V., Rolls, E. T., Leonard, C. M., and Stern, C.
(1986b). Gustatory responses in the frontal opercular cortex (1993). Neuronal responses in the ventral striatum of the
of the alert cynomolgus monkey. J. Neurophysiol. 56: behaving monkey. Behav. Brain Res. 55:243–252.
876–890. Wilson, F. A. W., and Rolls, E. T. (1993). The effects of stimulus
Scott T. R., Karadi Z., Oomura Y., Nishino, H., Plata-Salaman, novelty and familiarity on neuronal activity in the amygdala
C. R., Lenard, L., Giza, B. K., and Aou, S. (1993). Gustatory of monkeys performing recognition memory tasks. Exp. Brain
neural coding in the amygdala of the alert macaque monkey. Res. 93:367–382.
J. Neurophysiol. 69:1810–1820. Wilson, F. A. W., and Rolls, E. T. (2003). The primate amygdala
Seltzer, B., and Pandya, D. N. (1989). Frontal lobe connections and reinforcement: a dissociation between rule-based and
of the superior temporal sulcus in the rhesus monkey. associatively-mediated memory revealed in amygdala
J. Comp. Neurol. 281:97–113. neuronal activity. (In press.)
Central Taste Anatomy and Neurophysiology 705

Wolf, G., and DiCara, L. V. (1974). Impairments in sodium Yaxley, S., Rolls, E. T., Sienkiewicz, Z. J., and Scott, T. R. (1985).
appetite after lesions of gustatory thalamus—replication and Satiety does not affect gustatory activity in the nucleus of the
extension. Behav. Biol. 10:106–112. solitary tract of the alert monkey. Brain Res. 347: 85–93.
Yamaguchi, S. (1967). The synergistic taste effect of monosodium Yaxley, S., Rolls, E. T., and Sienkiewicz, Z. J. (1988). The
glutamate and disodium 5’-inosinate. J. Food Sci. 32:473–478. responsiveness of neurones in the insular gustatory cortex of
Yamaguchi, S., and Kimizuka, A. (1979). Psychometric studies the macaque monkey is independent of hunger. Physiol.
on the taste of monosodium glutamate. In Glutamic Acid: Behav. 42:223–229.
Advances in Biochemistry and Physiology, L. J. Filer, Yaxley, S., Rolls, E. T., and Sienkiewicz, Z. J. (1990). Gustatory
S. Garattini, M. R. Kare, A. R. Reynolds, and R. J. Wurtman responses of single neurons in the insula of the macaque mon-
(Eds.). Raven Press, New York, pp. 35–54. key. J. Neurophysiol. 63:689–700.
Yamamoto, T. (1993). Neural mechanisms of taste aversion learn- Zald, D. H., Lee, J. T., Fluegel, K. W., and Pardo, J. V. (1998).
ing. Neurosci. Res. 16:181–185. Aversive gustatory stimulation activates limbic circuits in
Yamamoto, T., Shimura, T., Sako, N., Yasoshima, Y., and Sakai, humans. Brain 121:1143–1154.
N. (1994). Neural substrates for conditioned taste aversion in Zatorre, R. J., Jones-Gotman, M., Evans, A. C., and Meyer, E.
the rat. Behav. Brain Res. 65:123–137. (1992). Functional localization of human olfactory cortex.
Yan, J., and Scott, T. R. (1996). The effect of satiety on responses Nature 360:339–340.
of gustatory neurons in the amygdala of alert cynomolgus
macaques. Brain Res. 740:193–200.
34

Molecular Physiology of Gustatory Transduction

Timothy A. Gilbertson
Utah State University, Logan, Utah, U.S.A.

Robert F. Margolskee
Howard Hughes Medical Institute and The Mount Sinai School of Medicine, New York, New York, U.S.A.

I. INTRODUCTION code typically utilizing cyclic nucleotides (cNMPs) and

inositol trisphosphate (IP3). These signaling cascades
The aim of this chapter is to provide an overview of the usually involve effector enzymes downstream of the
molecular mechanisms involved in taste transduction. G proteins to generate or regulate second messengers
Human taste perception can be divided into five major cat- leading to TRC depolarization and Ca2 release. In the
egories: sweet, sour, salt, bitter, and umami (glutamate). case of ionic stimuli (e.g., Na, K, H), the tastant itself
Acids and sodium salts, which elicit sour and salt taste, constitutes all or part of the initial intracellular signal:
respectively, are fluxed by and/or regulate the activities of these ions pass through or block apical ion channels
ion channels expressed in taste receptor cells (TRCs). leading to TRC depolarization and/or hyperpolarization.
Sugars, artificial sweeteners, bitter plant alkaloids, and The great chemical and physical diversity of tastants has
other bitter compounds primarily stimulate G protein–cou- necessitated multiple transduction mechanisms (see Fig.
pled receptors (GPCRs) to initiate one or more second 1), in contrast to visual and olfactory systems where mul-
messenger signaling cascades. Transduction of the taste of tiple receptors in iterated but fundamentally similar trans-
amino acids such as L-glutamate appears to depend on both duction pathways transduce variants of one general type
GPCR cascades and stimulation of amino acid–gated ion of stimulus (photons or small volatile molecules) into an
channels. During the past 10 years major new insights in intracellular signal.
taste transduction have been gained from diverse studies The vertebrate taste bud is a polarized ovoid structure
utilizing electrophysiology, biochemistry, molecular containing between 50 and 150 TRCs. The TRCs are
cloning, and transgenic and gene knockout animal models. elongated neuroepithelial cells that stretch from the
The sensation of taste is initiated by the interaction of bottom of the taste bud to the top. TRCs make contact
sapid molecules (“tastants”) with receptors and ion chan- with the oral cavity through the bud’s apical taste pore.
nels in the apical microvilli of TRCs. TRCs use a variety The apical surface of TRCs is rich in convoluted
of mechanisms to transduce chemical information into microvilli; it is here that receptors and channels involved
intracellular signals. Those taste transduction pathways in taste transduction are presumably concentrated. At the
utilizing G proteins and their coupled receptors convert base of the taste bud are the basal cells that are thought to
chemical information into a cellular second messenger be precursors to mature TRCs (see Chapter 32).

707
708 Gilbertson and Margolskee

II. BITTER TRANSDUCTION MECHANISMS which are detected by humans as bitter in the nanomolar
to micromolar range. Other vertebrates also detect and
A. Diversity of Bitter Compounds and Their
avoid many of these same compounds in similar ranges,
Detection Mechanisms
leading to the inference that the human sense of bitter-
ness correlates with other vertebrate animals’ aversive
The human sense of bitter taste is thought to be an evolu- responses to the same toxic compounds that humans
tionarily selected mechanism for the avoidance of poi- characterize as bitter. Support for this idea comes from
sons, particularly toxic plant alkaloids and glycosides, the identification in humans, rats, and mice of conserved

Figure 1 Proposed transduction mechanisms in vertebrate taste receptor cells. All taste pathways converge on the common elements
of a rise in intracellular Ca2 followed by neurotransmitter (NT) release. Sodium salts depolarize taste cells directly through Na influx
through amiloride-sensitive ENaC. Acids, in the form of protons, also permeate ENaC, activate proton-activated cation channels (MDEG
and, perhaps, ASIC) and inhibit apical K channels. L-Glutamate (L-Glu), which is the stimulus eliciting umami taste, activates the
T1r1/T1r3 heterodimer and the taste form of mGluR4, a GPCR linked to decreases in cAMP concentration via phosphodiesterase (PDE)
activation. The decrease in cAMP may lead to decreased inhibition of cNMP-gated channels and a rise in intracellular Ca2. Other amino
acids, like arginine (L-Arg), activate ionotropic receptors causing TRC depolarization. Artificial sweeteners activate ionotropoic recep-
tors linked to cation channels and GPCR linked via phospholipase C (PLC) to IP3 production and release of Ca2 from intracellular stores.
Natural sugars apparently activate GPCRs linked via adenylyl cyclase (AC) to cAMP production, which in turn may inhibit basolateral
K channels through phospohorylation by cAMP-activated protein kinase A (PKA). Bitter compounds, such as denatonium and propy-
lthiouracil (PROP), activate the GPCR T2R (TRB). Activated T2R stimulates production of PDE via the -gustducin and the concomi-
tant decrease in cAMP. The G subunits of (3, 13) released upon activation of heterotrimeric gustducin by T2R lead to IP3 produc-
tion via PLC2 and eventual rise in intracellular Ca2. Other bitter compounds, including quinine and divalent cations, have been
demonstrated to inhibit apical K channels in some species. DAG, diacylglycerol; AP, action potentials. (Adapted from Gilbertson et al.,
2000.)
Molecular Physiology of Gustatory Transduction 709

families of genes encoding apparent bitter-responsive 1995); and cycloheximide (cyx) (Lush and Holland,
receptors (Adler et al., 2000; Matsunami et al., 2000; 1988). Other data suggest that these are not all indepen-
Chandrashekar et al., 2000). Furthermore, and with some dent loci, but in some cases pleiotropic manifestations of
notable exceptions, detection thresholds for many bitter the soa locus (Harder et al., 1992). The soa locus has been
compounds are similar among humans and other verte- mapped to mouse chromosome 6 (Capeless et al., 1992).
brates. On the whole, bitter agents comprise an incredibly The murine genetic data argue for the existence of three or
diverse group of compounds (e.g., charged-hydrophilic more unlinked loci that affect bitter taste: (1) soa, (2) PTC
denatonium; neutral-hydrophobic quinine; divalent aversion, and (3) several loci that affect quinine aversion
cations Ca2 and Ba2; amino acids L-tryptophan and L- (Harder et al., 1992). SOA nontaster mice display little or
phenylalanine; peptides Ser-Leu-Ala and Phe-Phe-Phe; no whole nerve response to SOA (although they respond
modified sugar and sweetener analogs MAD-diCl-Gal). normally to other bitter compounds), indicating that the
This broad diversity is reflected in the multiple receptors defect is peripheral in nature (i.e., concerning the ability
and multiple transduction mechanisms used by taste of the TRC to generate a depolarizing response to the tas-
cells. tant). In recent genomics-based studies (Adler et al.,
2000; Matsunami et al., 2000), clusters of 40–80 candi-
B. Genetic Studies Suggesting the Existence date taste receptors were identified and mapped to mouse
of Multiple Bitter Detection Mechanisms chromosomes 6 and 15 (see below). One of these GPCRs
was shown to function as a cycloheximide (bitter) recep-
Inherent interindividual differences in the ability to taste tor and may correspond to the cyx locus; presumably the
certain bitter compounds have been demonstrated in other candidate receptors among this large gene family
genetic studies of humans and mice (for reviews see Miller may also be bitter-responsive and correspond to the rua,
and Bartoshuk, 1991). It was reported in 1932 that soa, glb, and qui loci, as well as to other as yet unidenti-
phenylthiocarbamide (PTC) tastes bitter to certain individ- fied bitter response loci.
uals, but is tasteless to others (Fox, 1932). Family studies
indicated that insensitivity to PTC (PTC “taste blindness”) C. Receptor-Mediated Bitter Transduction
is inherited as a simple Mendelian recessive trait. PTC Pathways
nontasters are also insensitive to many other bitter com-
pounds, including 6-n-propyl-2-thiouracil (PROP). The membrane impermeant compound denatonium
PTC/PROP tasters are more sensitive than nontasters to the benzoate is for humans the most intensely bitter
bitter tastes of caffeine, KCl, saccharin, and benzoate. compound known (Saroli, 1984). In rat circumvallate
However, tasters and nontasters do not differ in their (CV) papillae TRCs, denatonium caused the release of
responses to other bitter compounds (e.g., quinine) (Hall et Ca 2 from internal stores (Akabas et al., 1988).
al., 1975; Bartoshuk, 1979). Interestingly, PTC/PROP Denatonium caused a rapid transient increase in IP3
tasters are also more sensitive than nontasters to the sweet levels within dissected CV papillae (Spielman et al.,
tastes of saccharin, sucrose, and neohesperidin (Bartoshuk, 1994), and by autoradiographic visualization it was
1979; Gent and Bartoshuk, 1983). The variety of com- shown that calcium was depleted from the papillae upon
pounds that tasters and nontasters differentially respond to the addition of IP3 (Hwang et al., 1990) (as would be
suggests that the molecular basis for differences between expected if IP3 caused the opening of Ca2 channels
these two populations is not simply due to a single taste within taste cells). It had been thought previously that
receptor; rather, this difference may relate to coordinate the denatonium-responsive receptor activated Gq or a
expression of multiple genetically linked receptors or to a Gq-like G protein (both of which have been identified in
common site of action further along in the transduction rodent TRCs (McLaughlin et al., 1992a,b; Kusakabe
cascade (e.g., a G-protein subunit, a second et al., 1998, 2000), to activate phospholipase C (PLC) to
messenger–regulating enzyme, or an end target such as an generate IP3. Furthermore, the IP3 receptor, PLC2, and
ion channel). other components of the GPCR/G protein-responsive
Polymorphisms in the ability to taste various bitter phosphatidylinositol signaling pathway have been shown
compounds have been identified and mapped in inbred by immunological and/or histochemical methods to be
strains of mice. Independent autosomal dominant loci present in rodent CV papillae TRCs (Hwang et al., 1990;
were identified for strychnine and sucrose octaacetate Rossler et al., 1998; Clapp et al., 2001; Miyoshi et al.,
(SOA) (the soa locus) (Lush, 1981); raffinose undecaac- 2001). It is now known that denatonium-responsive
etate (rua) (Lush, 1986); quinine (qui) (Lush, 1984); cop- GPCRs mediate this response via activation of PLC2
per glycinate (glb) (Lush and Holland, 1988; Lush et al., by gustducin’s  component (see below).
710 Gilbertson and Margolskee

Gustducin is a transducin-like heterotrimeric G protein et al., 1999). A novel G-protein  subunit, G13, was
selectively expressed in ~30% of TRCs (McLaughlin, cloned and shown to colocalize in TRCs with -gustducin
et al., 1992a; Boughter et al., 1997). In the presence of any and G3. G13 was shown in vitro to interact with -
one of several bitter compounds, bovine taste gustducin, G1, and denatonium-responsive taste recep-
receptor–containing membranes activated transducin or tors. It had been shown previously that rat TRCs expressed
gustducin, but not the related G protein Gi (Ruiz-Avila PLC2 (Rossler et al., 1998), an isoform particularly sensi-
et al., 1995; Ming et al., 1998). Among the bitter tive to activation by G protein  subunits (Katz et al.,
compounds that activated transducin/gustducin were dena- 1992). Using quench flow measurements and neutralizing
tonium, quinine, nicotine, atropine, naringen, but not antibodies to PLC2, it was shown that this PLC isoform
caffeine or urea. In vivo assays also have demonstrated mediated the generation of IP3 in taste tissue (Rossler et al.,
clearly that gustducin’s -subunit (-gustducin) plays a 1998). Similar studies showed that the TRC-expressed G
key role in TRC responses to numerous bitter compounds subunits G13 and G3 were also required to mediate the
(Wong et al., 1996). In comparison to their littermate rise in taste tissue IP3 in response to denatonium (Huang
controls, -gustducin knockout mice displayed markedly et al., 1999; Rossler et al., 2000). Thus, it appears that het-
reduced behavioral and nerve responses to the bitter erotrimeric gustducin mediates two responses in TRCs: a
compounds denatonium benzoate and quinine sulfate. decrease in cNMPs via its -subunit and a rise in IP3 via its
Based on its close relatedness to -transducin, particu- G313 component. The subsequent steps in the
larly within the regions of -transducin that are known /PLC2/IP3 pathway are apparently activation of IP3
to interact with PDE6, it was originally proposed that receptors and release of Ca2 from internal stores followed
-gustducin’s role in bitter transduction is mediated by acti- by neurotransmitter release (Bernhardt et al., 1996).
vation of a taste PDE (McLaughlin et al., 1992a).
Consistent with this proposal, Price (1973) had shown that
the bitterness of several compounds correlated well with D. Identification of Gustducin-Coupled Taste
their ability to activate PDE. Furthermore, very high levels Receptors
of PDE activity are present in taste tissue (Kurihara, 1972;
Law and Henkin, 1982) and two cAMP PDEs have been Following up on genetic mapping of bitter response loci of
cloned from taste tissue and shown to have elevated expres- humans and mice and utilizing the partially completed, at
sion in taste tissue (McLaughlin et al., 1994). that time, human genomic DNA sequence databases, a fam-
More recently, it was shown in vitro that -gustducin and - ily of ⵑ25 human GPCRs (named T2Rs or TRBs) was iden-
transducin activate taste-expressed Ca2/calmodulin- tified recently as candidate taste receptors (Adler et al.,
sensitive type I PDEs (Ruiz-Avila et al., 1995; M. M. Bakre, 2000; Matsunami et al., 2000). The T2R/TRB receptors map
R. Lupi, and R. F. Margolskee, unpublished). A direct within multigene clusters on human chromosome regions
demonstration of the importance of the gustducin-PDE 5p15, 12p13, and 7q31 and the syntenic regions of mouse
interaction in bitter taste transduction comes from the recent chromosomes 6 and 15. Consistent with the inference that
work of Yan et al., (2001): several bitter compounds were T2R/TRB receptors might be responsive to bitter com-
shown to cause a decrease in cNMP levels in taste tissue, pounds, PROP sensitivity maps to human 5p15 and 7q31
which was blocked selectively by antibodies against -gust- (Reed et al., 1999), while SOA sensitivity (soa) maps to the
ducin. The subsequent steps in this transduction pathway distal region of mouse chromosome 6 (syntenic with human
are speculative (reviewed in Spielman, 1998; Gilbertson et 12p13) (Capeless et al., 1992; Lush et al., 1995). Members
al., 2000): decreased cNMPs may act on protein kinases, of the T2R/TRB multigene family share 30–70% identity,
which in turn may regulate TRC ion channel activity, or but are only distantly related to other known GPCRs.
cNMP levels may regulate directly the activity of cNMP- T2R/TRB receptors are most highly conserved in their three
gated and cNMP-inhibited ion channels expressed in TRCs. predicted cytoplasmic loops (presumptive sites of G protein
Many of the bitter compounds shown to activate interaction) and the adjacent transmembrane segments;
gustducin in vitro lead to pertussis toxin–sensitive genera- these receptors display the greatest divergence in the extra-
tion of IP3 (Spielman et al., 1994), yet neutralizing cellular regions (predicted regions of ligand binding).
antibodies directed against -gustducin did not block the Using in situ hybridization, it was shown that the
bitter-induced rise in IP3 levels (Yan et al., 2001), suggest- T2R/TRB receptors are only expressed in gustducin-
ing that another G protein -subunit might mediate this positive TRCs. Furthermore, the in situ hybridization signal
response. The resolution of this apparent conflict comes with single vs. multiple T2R/TRB probes identified largely
from the demonstration that the IP3 effect depends on the the same TRCs, implying that most or even all T2R/TRB
activation of PLC2 by gustducin’s  constituents (Huang receptors are expressed in the same gustducin-positive
Molecular Physiology of Gustatory Transduction 711

TRCs. This finding would suggest that activation of human threshold for detection of denatonium, suggesting
gustducin by one or another bitter-responsive T2R/TRB that another ligand is probably the preferred stimulus
receptor would have the same output and that at equipotent for this receptor pair and that another human T2R/TRB
concentrations one bitter compound should not be distin- receptor may be more sensitive to denatonium.
guishable from another—generally the case. Most of the
gustducin-positive TRCs in the CV, foliate, or palate were E. Other G Protein–Dependent and Independent
also T2R/TRB-positive, although most gustducin-positive Bitter Transduction Pathways
TRCs in the fungiform papillae were T2R/TRB-negative. It
may be that other receptors are expressed in the gustducin- G proteins other than gustducin have been inferred to play
positive/T2R/TRB-negative TRCs; given that gustducin is a role in bitter (and sweet) transduction since -gustducin
implicated in bitter and sweet responses, this subset of knockout mice have diminished, but not totally abolished,
gustducin-positive TRCs might express sweet-responsive responses to both bitter and sweet compounds (Wong
taste receptors unrelated to the T2R/TRB receptors. A et al., 1996). Transgenic expression in the gustducin
regionally variable pattern of expression in rat and mouse lineage of TRCs of a dominant-negative form of -gust-
TRCs was noted with several T2R/TRB clones examined. ducin that interferes with T2R/TRB-G protein interactions
Most taste buds of CV, foliate, geschmacksstreifen (“taste further decreased the residual responses to bitter
stripe”), and epiglottis contained T2R/TRB-positive TRCs; compounds of -gustducin knockout mice, consistent with
typically~15–20% of TRCs were positive. However, fewer another G protein expressed in gustducin-positive TRCs
than 10% of fungiform taste buds contained any T2R/TRB- mediating these responses (Ruiz-Avila et al., 2001).
positive TRCs, although in the positive buds ~15% of the Gi2, Gi3, G14, G15, Gq, Gs, and -transducin are
TRCs were positive. possible candidates to mediate these responses since they
Chimeras of two of the T2R/TRB receptors (containing are expressed in TRCs (McLaughlin et al., 1992a;
the N-terminal 39 amino acids of rhodopsin) were Ruiz-Avila et al., 1995; Kusakabe et al., 1998, 2000).
successfully expressed in HEK 293 cells and shown to be Based on in vitro taste assays with bovine taste
activated by bitter compounds (Chandrashekar et al., membranes, the bitter compounds caffeine and theo-
2000). Cells transfected with the mouse receptor mT2R-5 phylline are not transduced by the T2R/TRB-gustducin
responded only to cycloheximide from among 55 tastants pathway (Ming et al., 1998). Based on quench flow studies
tested. Consistent with this receptor mediating in vivo with murine taste tissue, it was determined that caffeine
responses to this bitter compound, the cycloheximide and theophylline, known PDE inhibitors, act directly on
concentration needed to elicit a half-maximal response in taste PDEs to raise TRC cGMP levels (Rosenzweig et al.,
mT2R-5-transfected cells was comparable to the murine 1999). Soluble guanylyl cyclase (GC) is the presumptive
threshold for aversion. Multiple amino acid differences source of the cGMP; both GC and nitric oxide synthase
were noted in mT2R-5 isolates from cycloheximide (NOS) are known to be present in TRCs (Asanuma and
taste-sensitive (CBA/Ca, BALB/c, C3H/He) vs. -insensi- Nomura, 1995; Kretz et al., 1998). Presumably, these two
tive (C57BL/6, 129/Sv) strains of mice. Furthermore, bitter compounds also raise TRC cAMP levels.
transfected cells expressing mT2R-5 from the nontaster In addition to bitter transduction mediated by GPCR/G
strain required a nearly 10-fold higher concentration of protein pathways, there is considerable evidence for bitter
cycloheximide to elicit a response, similar to the difference transduction mediated by ion channels. Several bitter
in in vivo responses between the taster and nontaster compounds are known K channel blockers and cause
strains. mT2R-5 was shown to selectively couple to TRC depolarization by blocking K channels (e.g., TEA,
-gustducin vs. i, s, o, or q, providing additional 4-aminopyridine, and Ba2) (reviewed in Lindemann,
support to it being a cycloheximide-responsive bitter 1996; Herness and Gilbertson, 1999). Quinine leads to
receptor and suggesting that the other T2R/TRB receptors TRC depolarization along with an increase in membrane
are most likely additional bitter-responsive taste receptors. resistance to K (Sato and Beidler, 1982). In patch-
To finally prove that mT2R-5 encodes a (or the) cyclohex- clamped frog TRCs, quinine completely blocked Na
imide-responsive taste receptor will require the generation currents and partially blocked K currents (Avenet and
and testing of the appropriate knockout or transgenic Lindemann, 1987); the addition of EGTA or CaCl2 to the
mouse model. Cells transfected with either hT2R- patch pipette solution did not affect the quinine blockade
4/mT2R-8, an orthologous pair of human/mouse of of currents, implying that Ca2 is not involved. In another
receptors, showed responses to denatonium and PROP. study using intracellular recording of frog TRCs, quinine
However, the concentration of denatonium required was was shown to depolarize the cells by inducing the active
more than three orders of magnitude higher than the secretion of Cl across the basolateral membrane
712 Gilbertson and Margolskee

(Okada et al., 1988). Although denatonium has been 3. GPCR/G protein–independent mechanisms may
shown to activate heterotrimeric gustducin via coupled transduce responses to bitter compounds via direct
T2R/TRB receptors, leading to G13/PLC2-mediated effects upon TRC ion channels. For example, TEA
generation of IP3 and Ca2 release from internal stores, it and quinine block apical K channels, leading to
has also been shown in guinea pig TRCs to cause Ca2 TRC depolarization. These three models are not
influx (Orola et al., 1992), and in the mouse denatonium mutually exclusive and may occur in the same
blocks delayed rectifier K currents (Spielman et al., TRCs.
1989). In bullfrog TRC membrane patches it was shown
For those components of these proposed pathways that
that quinine, denatonium benzoate, and other bitter com-
have been molecularly cloned or otherwise identified in
pounds directly gated a TRC cation channel (Tsunenari et
TRCs, it should be possible to confirm their presence and
al., 1999). In mudpuppy (Necturus maculosus) TRCs
functionality in bitter-responsive TRCs. Among the
denatonium benzoate led to an increase in intracellular
proposed transduction components, T2R/TRB receptors,
Ca2, derived predominantly from internal stores (Ogura
-gustducin, -transducin Gq, G14, G1, G3, G13,
et al., 1997). Thapsigargin, GDPS, and a PLC inhibitor
PLC2, and cNMP-gated channels have been shown
inhibited this response, but IBMX, membrane-permeant
by molecular methods to be present in TRCs. The
cNMPs, and pertussis toxin did not affect this reponse,
cNMP-inhibited channel identified electrophysiologically
arguing against the involvement of pertussis toxin-sensi-
has not yet been cloned. All ~25 of the T2R/TRB receptors
tive G proteins (e.g., gustducin, transducin, or Gi).
are candidates for involvement in bitter transduction,
Dextromethorphan also led to Ca2 release from mud-
although strong functional data exist only for mT2R5 as a
puppy TRCs; this response was independent of G proteins
cycloheximide receptor.
and was not blocked by a PLC inhibitor, suggesting a
direct action on Ca2 stores (Ogura and Kinnamon, 1999).
III. SWEET TRANSDUCTION MECHANISMS
F. Proposed Models for Bitter Transduction
A. Diversity of Sweet Compounds and Sweet
Modifiers
There are at least three possible bitter transduction path-
ways.
Sweet compounds are extremely diverse in their chemical
1. A T2R/TRB-gustducin/transducin-PDE-↓cNMP composition: examples include simple carbohydrates
pathway—bitter compounds such as cycloheximide (monosaccharides and disaccharides), D- and L-amino
bind to and activate one or more T2R/TRB receptors, acids (most D-amino acids are sweet; only some L-amino
which are coupled to heterotrimeric gustducin (and/or acids are sweet), artificial sweeteners (saccharin, cycla-
transducin); activated -gustducin activates PDEla mate, aspartame, acesulfame K, etc.), plant proteins (mon-
isoforms, leading to decreased TRC levels of cNMPs. ellin and thaumatin), chloroform, and simple salts of beryl-
The subsequent steps are speculative: one possibility lium and lead. Sweet transduction is thought to utilize
is that decreased cNMPs lead to opening of cNMP- seven transmembrane-helix receptors coupled to G pro-
inhibited cation channels, causing TRC depolariza- teins and/or an amiloride blockable Na channel. Sugars
tion; another possibility is that decreased cNMPs lead and high-potency artificial sweeteners have been shown to
to closure of cNMP-gated cation channels, causing bind to the taste cell surface and to taste cell membrane
TRC hyperpolarization. fractions (Cagan and Morris, 1979), consistent with there
2. A T2R/TRB-gustducin/transducin-PLC2-↑IP3/DAG being a cell surface receptor for sweet compounds.
pathway—bitter compounds such as denatonium Biochemical studies with membranes derived from the
bind to and activate one or more T2R/TRB recep- anterior tongue of the rat (containing fungiform papillae)
tors, which are coupled to heterotrimeric gustducin; or with intact taste buds from rat CV papillae showed that
activation of gustducin releases its  subunits sweet compounds activated adenylyl cyclase (AC) to
(G3G13), which activates PLC2; PLC2 elevate intracellular levels of cAMP (Striem et al., 1989,
generates IP3 and DAG, which lead to Ca2 release 1991). The competitive sugar antagonist SOA inhibited
from internal stores and TRC depolarization and sweet induced elevation of cAMP (Striem et al., 1991). In
transmitter release from vesicles. It is possible membrane extracts from the rat tongue, AC was activated
that a receptor-mediated/gustducin-independent by sucrose in a GTP-dependent fashion (Striem
pathway may utilize the -subunit of Gq or G14 to et al., 1989). These observations argue for the presence of
activate PLC. specific receptors that upon activation by sweeteners
Molecular Physiology of Gustatory Transduction 713

activate Gs, which in turn activates AC to generate cAMP a transmembrane protein with a readily accessible extra-
as the intracellular second messenger. cellular domain. Receptors with large extracellular ligand-
Monellin and thaumatin are two naturally occurring binding domains, such as T1r1, T1r2, T1r3, and mGluR,
sweet proteins that are on a molar basis 104–105 times as might be expected to be particularly sensitive to these
sweet as sucrose (van der Wel and Loeve, 1972). treatments. Genetic mapping and heterologous expression
Thaumatin consists of 207 amino acids and has a molecu- have identified T1r2 and T1r3 as components of a sweet
lar weight of 22,200 (Iyengar et al., 1979). Monellin con- receptor (see below).
sists of two chains, both of which are required for sweet-
ness. Both thaumatin and monellin are perceived as sweet B. Physical Characterization of Sweet Receptors
only by humans and old world primates: rat, pig, hamster,
dog, and rabbit do not exhibit neural responses to these Direct biochemical analysis of sweet receptors has been
proteins (Brouwer et al., 1973, Hellekant, 1976). hindered by the lack of suitable high-affinity ligands.
Presumably, the sweet-responsive receptor that binds to Sugars are unsuited for biochemical purification of their
these proteins is present only in primates or amino acid dif- cognate receptor(s) since they have apparent Kd of 0.1–1.0
ferences in this receptor in non-primates affect its ability to M (Cagan, 1971). However, more useful ligands may be
bind these protein sweeteners. found among high-potency artificial sweeteners such as
The protein miraculin is tasteless by itself but induces superaspartame and the intensely sweet proteins thaumatin
sweetness in the presence of acid (Bartoshuk et al., 1969; and monellin that are ~100,000-fold sweeter than sucrose
Kurihara et al., 1969). Miraculin, like thaumatin and on a molar basis (van der Wel and Loeve, 1972; Tinti and
monellin, is limited in its action to humans and certain Nofre, 1991). Attempts to use these high-affinity sweeten-
other primates. Miraculin caused sweet responsive nerve ers for the biochemical purification of sweet receptors
fibers to respond to acid stimuli, but it did not affect sour- have not yet yielded a sufficiently pure preparation to
responsive fibers (Brouwer et al., 1983). Responses to determine the sequence and thereby molecularly clone the
monellin, thaumatin, and miraculin cross-adapt, indicating receptor(s). A more fruitful approach to cloning the recep-
that these proteins act at the same site. Perhaps miraculin tors underlying sweet taste responses has been to use the
binding induces a conformational change in the sweet genetics of sweet taste responses in combination with
receptor that does not fully activate the receptor except at genomics-based analysis of candidate genes (Max et al.,
low pH. Alternatively, two independent events may be 2001; Montmayer et al., 2001; Sainz et al., 2001).
required to activate sweet-responsive cells: binding of Labeled thaumatin was used to physically characterize
miraculin to the receptor and lowering of pH. Another the nature of its binding site (Farbman et al., 1987): thau-
protein taste modifier, curculin, has sweet taste by itself, matin bound to microvilli and vesicles shed from
but its sweetness is greatly enhanced in the presence of microvilli of rhesus monkey foliate papillae. It was subse-
acid or in the absence of divalent cations (Yamashita et al., quently shown that thaumatin binding elicited shedding of
1990); presumably curculin and miraculin act by similar microvilli vesicles into the pores of foliate and vallate
mechanisms at the same receptor target. papillae (Farbman and Hellekant, 1989). Repeated stimu-
Several different types of compounds specifically lation with thaumatin led to a decline in the neural
inhibit sweet taste. Gymnemic acid, ziziphin, and holdul- response to thaumatin or sucrose, but not to citric acid
cin are triterpine saponins that preferentially inhibit sweet (sour), suggesting that the thaumatin-and sucrose-binding
(Stocklin, 1968; Saul, Kennedy and Stevens, 1985; sites are in the same TRCs and in close proximity. In
Yamada, Imoto and Yoshiokai, 1985). Other surface-acting another set of studies, monellin was shown to bind to
agents such as the detergent SDS reduce sweetness, but membranes from human and bovine CV papillae with a Kd
SDS also reduces the intensity of salt and bitter tastes of about 105 M (Cagan and Morris, 1979). A photo-affin-
(DeSimone et al., 1980). In addition, low concentrations ity derivative of thaumatin was used to label monkey taste
(0.1 mM) of CuCl2 and ZnCl2 selectively suppress sweet papillae: polyacrylamide gel electrophoresis indicated that
without affecting bitter or salt tastes (Iwasaki and Sato, a 50,000 MW protein present in taste papillae but absent
1984). The sweet receptors are also very sensitive to pro- from nontaste papillae bound to the derivatized thaumatin
teolysis: treatment of the apical surface of the rat tongue (Shimazaki et al., 1986).
with proteases results in the loss of responsiveness to
sucrose, but not to other tastants (as assessed by chorda C. SAR-Inferred Models of Sweet Receptors
tympani nerve recordings) (Hiji, 1975). That detergents
and saponins had their most profound effects upon Based upon common features of sweeteners, several dif-
sweet responses is consistent with the sweet receptor being ferent models have been developed for the structure of the
714 Gilbertson and Margolskee

sweet receptors’ binding pocket. In the A-H/B model of ligand binding properties and determine if these receptors
Schallenberger and Acree (1967), a hydrogen bond donor underlie all sweet responses. Experiments with T1r2/T1r3
within the sweet tastant (A-H: A is an electronegative knockout mice will be particularly informative regarding
atom, H is a hydrogen atom) interacts with an electro- the existence of multiple sweet detection mechanisms.
negative hydrogen bond acceptor within the sweet
receptor; and an electronegative atom within the tastant (B: D. Evidence for Multiple Sweet Receptors and
usually a N or O atom) interacts with an electropositive Multiple Transduction Pathways
hydrogen within the receptor. According to the original
model, an A-H/B pharmacophore must be present in all Studies with mouse taste cells (Tonosaki and Funokosaki,
sweeteners with the distance between the A and B groups 1984), hamster taste buds and taste fibers (Faurion and
between 2.5 and 4 Å. This model also posits that the sweet Vaysettes-Courchay, 1990; Cummings et al., 1993), gerbil
receptor (or a family of closely related receptors) has com- and hamster chorda tympani (Jakinovich and Goldstein,
plementary regions that recognize and bind to the A-H/B 1976, Rehnberg et al., 1992), chimpanzee taste fibers
sites of the sweet compounds: the initial event in sweet (Hellekant and Ninomiya, 1991), and human psy-
detection then is the simultaneous interaction of the A-H/B chophysics (Schiffman et al., 1981) all argue for multiple
sites with their complements in the sweet receptor. sweet receptors and/or mechanisms. In human psy-
Two modifications of the original A-H/B model chophysical experiments, it was found that partial cross
propose the presence of a third binding site: Kier (1972) adaptation or no cross adaptation occurred between high-
suggested that a dispersion interaction () at the receptor is potency sweeteners (neohesperidin dihydrochalcone,
necessary for high potency; Schallenberger (1978) aspartame, and saccharin) and sucrose when the high-
proposed a third interaction () that is hydrophobic in potency sweeteners were the preadapting stimuli
nature and required for high potency. Recognition by the (Schiffman et al., 1981).
receptor in these tripartite models is predicted to be chiral, Electrophysiological studies have shown that sucrose
with specific assigned distances between each of the three causes a subset of mouse TRCs to depolarize and undergo
constituents of the sweet pharmacophore. an increase in membrane resistance (Tonosaki and
One problem with all of these models is that almost Funokoshi, 1988a,b). Injection of cAMP, cGMP, EGTA, or
any organic compound with OH, NH, or CH (A-H the channel blocker TEA into this subset of mouse TRCs
groups), an electronegative group (B), and a hydrophobic also elicited depolarization with decreased K conduc-
group () is predicted to be sweet—and this is not the tance (Tonosaki and Funokoshi 1988a). In about one third
case! Several high-potency sweeteners do not seem to fit of whole-cell recordings from isolated rat TRCs, saccharin
in with these models (DuBois et al., 1993). Furthermore, caused a decrease in outward K current sufficient to
slight structural alterations of these sweeteners that do not depolarize the intact TRC (Behe et al., 1990).
affect the A-H/B groups lead to loss of sweetness or the Furthermore, in on-cell recordings from rat TRCs, saccha-
appearance of bitterness. Based upon ultra high-potency rin caused action potentials in about one third of the cells
guanidine sweeteners such as superaspartame, a model (Behe et al., 1990). In gerbil TRCs, sucrose and saccharin
has been proposed (DuBois et al., 1993) that suggests the caused a decrease in outward K currents (Uchida and
presence of complementary sites within the receptor that Sato, 1997ab). In isolated TRCs from frog (Rana escu-
interact with the following structures: a carboxylate lenta) cAMP, forskolin and methyl xanthines caused a
region, a major N-H region, a minor N-H region, a decrease in outward K currents (Avenet and Lindemann,
hydrophobic region, and a  stacking region. One prob- 1987). It was subsequently shown by whole-cell and
lem for a sweet receptor model involving classical recep- excised patch recordings that a cAMP-dependent kinase
tor-ligand type interactions is the fact that enantiomeric closes a basolateral K conductance (Avenet et al.,
hexoses, including D- and L-glucose, elicit identical taste 1988ab). These results from various vertebrate TRCs
behaviors (Schallenberger et al., 1969); this argues suggest that cNMPs serve as the intracellular second mes-
against stereospecific recognition and led to the sugges- senger for the sweet pathway and elicit depolarization of
tion of a mechanism based upon colligative properties of these cells by K channel closure, followed by Ca2 influx
carbohydrates (DuBois et al., 1993). through voltage-dependent Ca2 channels. It is uncertain
Now that theT1r2/T1r3 sweet-responsive receptors that protein kinase A (PKA) is actually the mediator of
have been molecularly cloned and expressed (Kitagawa cNMP-elicited K channel closure since the PKA inhibitor
et al., 2001; Max et al., 2001; Montmayeur et al., 2001; H-89 did not block sucrose-elicited action responses of
Nelson et al., 2001; Sainz et al., 2001) (see below), it will hamster fungiform TRCs (Varkevisser and Kinnamon,
be possible to biochemically characterize the receptors’ 2000). It may even be that sweetener-elicited cNMPs
Molecular Physiology of Gustatory Transduction 715

directly regulate this channel or act indirectly through 1983); presumably amiloride exerts its effect upon an
some other mechanism independent of PKA. epithelial type Na channel or a structurally related chan-
Using a “loose patch-in situ” recording method, nel. This may explain the psychophysical observation that
Cummings et al., (1993) recorded action currents in very low concentrations of NaCl are perceived by humans
response to sucrose and several artificial sweeteners: with as sweet (Beets, 1978; Bartoshuk et al., 1978). Ion trans-
time (1–3 min ), the action currents abated, suggesting that port studies with isolated canine lingual epithelium
an adaptive response occurred. Subsequent to this adapta- showed that both mono- and disaccharides increased
tion, the same sweetener could not elicit an action current; a transepithelial short circuit current that was partially
however, other sweeteners did elicit currents (i.e., they did blocked by amiloride (DeSimone et al., 1984, Mierson
not cross-adapt). Membrane-permeant cNMP analogs et al., 1988). Amiloride was also found to inhibit the
elicited action currents in sweet-responsive taste buds canine chorda tympani responses to sucrose and fructose
(28/28 with 8 cpt-cAMP; 7/8 for dibutyryl cGMP) and did (Mierson et al., 1988); however, other studies failed to
not elicit such currents in sweet-nonresponsive taste buds demonstrate an amiloride sensitive chorda tympani
(0/12 with 8 cpt-cAMP; 0/8 with dibutyryl cGMP). In this response to sugars in rat, gerbil, and hamster (Brand et al.,
study it was shown that neither amiloride nor the absence 1985; Jakinovich, 1985; Herness, 1987). The canine
of permeant apical cations had an effect upon taste bud response to sucrose was unaffected by cAMP, cGMP, or
action currents; this argues that cNMPs are involved as a Ca2, arguing against these second messengers playing
second messenger in taste cell responses to sweet and that a role in sucrose response in the dog (Simon et al., 1989).
this response does not involve an apical cNMP-responsive It is clear that different species use different mechanisms
channel. In additional studies of hamster fungiform TRCs for sweet transduction; furthermore, some species (e.g.,
using whole-cell recording, it was demonstrated that all humans) use multiple mechanisms for sweet transduction
sweet-responsive cells (~12.5% of the total) were also (see Jakinovich and Sugarman, 1988).
responsive to cNMPs (31/31 cells tested); conversely, all
sweet-unresponsive cells were likewise unresponsive to E. Biochemical and Molecular Biological
cNMPs (0/206 cells tested) (Cummings et al., 1996). Identification of Sweet Transduction Elements
Outward K currents of sweet-responsive TRCs were
reduced by sweet compounds or cNMPs and potentiated In biochemical studies using membrane vesicles derived
by the PDE inhibitor IBMX, suggesting that the sweet from the anterior tip of the rat tongue (containing fungi-
compounds elicited cNMPs that led to reduction of the K form papillae, connective tissue, and some underlying
currents. muscle), sucrose caused a twofold stimulation of AC
In addition to the TRC K channels, which are thought activity (Striem et al., 1989). The effect on AC required
to be indirectly regulated by cNMPs, cNMP-gated chan- GTP and was dependent upon the sucrose concentration.
nels, such as are well known in olfactory receptor neurons Among nonsugar sweeteners, saccharin, but not aspartame
and photoreceptor cells, may also serve in a sweet/cNMP or neohespederidin, elevated cAMP production in the
transduction pathway. Molecular biological (Misaka et al., tongue tip membranes. Sucrose did not elevate cAMP
1997, 1998, 1999) and electrophysiological studies production in membrane vesicles derived from skeletal
(Herness, 1993; Sugimoto, 1997) have provided direct evi- muscle, olfactory cilia, or nonsensory tongue epithelium.
dence for the existence in taste cells of a cNMP-responsive However, vesicles from tongue muscle did show an
channels. A novel channel that is inhibited by cNMPs has elevation of cAMP in response to sucrose. In a subsequent
been identified in TRCs from frog (Kolesnikov series of experiments (Naim et al., 1991), membrane
and Margolskee, 1995) and mouse (S. S. Kolesnikov and vesicles derived from the CV papilla of the pig showed
R. F. Margolskee, unpublished). These observations leave a GTP-dependent elevation of cAMP in response to
open the possibility that cNMPs may act indirectly in sucrose: control experiments with nonsensory epithelium
TRCs via protein kinase A and directly via cNMP-gated from pig tongue showed no rise in cAMP in response to
and cNMP-inhibited ion channels. Additional molecular the addition of sucrose. However, membranes from cow or
biological studies to identify other sweet transduction pig fungiform papillae (devoid of underlying muscle)
components are described in the next section. showed no sucrose-dependent rise in cAMP; instead, the
There is evidence for the presence of an altogether dif- addition of sucrose led to a decrease in cAMP levels (as
ferent sweet transduction mechanism in both dogs and was also found with control nonsensory epithelium). These
humans. Psychophysical studies in humans have shown results suggest that sweet receptors activate AC (via Gs) in
that the diuretic amiloride reduces the perceived intensity the CV papillae of cow and pig, but not in the fungiform
of both NaCl and sweet compounds (Schiffman et al., papillae from these animals. Either the AC response in rat
716 Gilbertson and Margolskee

fungiform differs from those of the cow and pig, or the bitter, it also potentiates the sweetness of certain
muscle present in the rat tongue tip preparation is respon- compounds in humans (Schiffman et al., 1986), perhaps by
sible for much of the sucrose-induced stimulation of AC. inhibition of cAMP PDE activity. The PDE inhibitor
In biochemical studies with intact rat CV taste buds, IBMX causes depolarization of cAMP-responsive TRCs
sucrose caused a dose-dependent rise in cAMP (Striem from frog (Avenet and Lindemann, 1987) and sweet-
et al., 1991). A competitive antagonist of sucrose, methyl- responsive TRCs from hamster (S. C. Kinnamon, personal
dichloro-dideoxy--D-galactopyranoside (MAD-diCl- communication). PDE has been shown to be present at
Gal), inhibited the cAMP elevation by about 65%. high levels in taste tissue (Law and Henkin, 1982); two
MAD-diCl-Gal had previously been shown to competi- cloned cAMP type PDEs (PDE4 isoforms) were shown to
tively inhibit sucrose activation of the chorda tympani have elevated expression in taste tissue (McLaughlin et al.,
response and sucrose stimulation of AC in rat anterior 1994). Most recently, PDE1A, PDE1C, PDE4, and PDE5
tongue tip (Striem et al., 1990). Control nontaste lingual were identified in bovine taste tissue by immunoblots of
epithelium did not respond to either sucrose or the chromatography fractions (M. M. Bakre, R. Lupi, and R. F.
competitive inhibitor. The artificial sweetener saccharin Margolskee, unpublished). PDE1A isoforms were shown
had no significant effect upon cAMP levels (in contrast to by immunohistochemistry to be highly expressed in a sub-
its above-cited effect on tongue tip membranes). In patch- set of -gustducin–positive and–negative TRCs. PDE1A is
clamp experiments with TRCs from rats, saccharin was well known to be activated by Ca2/calmodulin (Zhao et
shown to cause depolarization by a decrease in K perme- al., 1997); it is now known to be activated by transducin
ability (Behe et al., 1990). However, only 40% of the and/or gustducin (Bakre et al., 2001). Given the supposi-
saccharin-responsive cells also responded to cAMP; tion that elevated cNMPs are involved in sweet responses,
supporting the idea that saccharin does not exert its effects transducin/gustducin-activated PDE1A would decrease
via AC. It has been proposed that saccharin (and certain cNMP levels and in all likelihood be antagonistic to sweet
other amphiphilic sweet compounds) directly activates and/or involved in sweet signal termination. This provides
TRC-expressed G proteins (Naim et al., 1994). In other a possible explanation for how the -gustducin knockout
studies it was shown that saccharin acted by taste mem- mouse could be deficient in sweet responsiveness even if
brane receptors to inhibit bitter-responsive receptor activa- -gustducin is not directly involved in the sweet transduc-
tion of gustducin/transducin (Ming, et al., 1999). The tion pathway. If tonic activation of PDE1A by -gustducin
recent demonstration that heterologously expressed helps to set the basal level of TRC cNMPs, then elevation
T1r2/T1r3 responds to saccharin and other sweeteners of TRC cNMP levels in the absence of -gustducin may
(Nelson et al., 2001) confirms that this is a receptor— interfere with normal sweet transduction responses. AC
mediated phenomenon. has been shown to be present in bovine taste tissue (CV
Gs -subunit mRNA and protein are elevated in TRCs and fungiform regions) (Kurihara and Koyama, 1972) and
vs. the surrounding nontaste tissue (Kusakabe et al., 2000; in foliate papillae of the rabbit (Nomura, 1978; Asanuma
S. K. McLaughlin and R. F. Margolskee, unpublished), and Nomura, 1982); however, specific AC isoforms have
consistent with a proposed role for Gs in sweet taste not been cloned from TRCs to date.
transduction. Gs presumably activates AC in response to In addition to the role proposed for cNMPs in sweet
sweeteners, since the only known G proteins that stimulate responses, there is also evidence that IP3 may act as a
AC are Gs and Golf (Jones and Reed, 1989). Golf is known second messenger for some sweet compounds.
not to be expressed in taste tissue (P. McKinnon, unpub- Furthermore, this pathway may also regulate the same K
lished), whereas Gs is expressed in taste buds (McLaughlin conductance targeted by cNMPs generated in response to
et al., 1992a, b; Kusakabe et al., 2000). The Gs gene sugars. The artificial sweeteners saccharin and SC45647
generates a great diversity of transcripts via differential caused a robust increase in IP3 levels in CV tissue and
splicing (Kozasa et al., 1988; Swaroop et al., 1991); a rapid rise in CV TRC Ca2 via release from intracellular
preliminary data indicate that at least one Gs splice prod- stores (Bernhardt et al., 1996). Sucrose elicited only
uct is expressed in taste tissue but not in nonsensory a modest rise in CV IP3 levels and elevated Ca2 in the
portions of the tongue (Kusakabe et al., 2000; S. K. sweetener-responsive TRCs via influx, not store release. In
McLaughlin and R. F. Margolskee, unpublished). Potential this study a clear segregation was noted between
targets for cNMPs generated by Gs/AC are protein kinase sweet-responsive (sucrose, saccharin and SC45647) vs.
A–regulated K channels and cNMP-gated channels, bitter-responsive (denatonium) TRCs. An important
described above. conclusion was that IP3- and cNMP-dependent sweet path-
PDEs and ACs are implicated in sweet and/or bitter ways exist in the same TRC. Recent results with hamster
transduction via cNMP regulation. Although caffeine is taste buds determined that inhibition of protein kinase C
Molecular Physiology of Gustatory Transduction 717

(PKC) blocked artificial sweetener-elicited action currents (Bachmanov, et al., 1997; Blizzard et al., 1999). It was
in fungiform taste buds (Varkevisser and Kinnamon, initially proposed that the candidate taste receptor T1r1
2000). Apparently the same TRC K conductance was sac due to its position on distal 4 (Hoon et al., 1999).
sensitive to cNMPs generated in response to sugars is However, by an analysis of the recombination frequency
phosphorylated and closed by PKC activated in response between T1r1 and those markers closest to sac in F2 mice,
to the artificial sweetener-stimulated taste transduction it is clear that T1r1 is rather distant (~5 cM) from sac
pathway. It is interesting to consider these results in the (D18346 is an STS marker that maps most closely to sac)
context of heterotrimeric gustducin-mediated responses to (Li et al., 2000). Furthermore, by analysis of regions of the
the bitter compound denatonium: -gustducin via PDE1A sequenced human genome syntemic to that of the mouse in
decreases cNMP levels, while G313 via PLC2 the region of D18346 and other markers linked to sac, it
increases IP3 levels. Perhaps in the case of the sweet- was determined that a novel GPCR, termed T1r3, is within
responsive TRCs heterotrimeric Gs activates AC to gener- ~15,000 bp of D18346/sac (Max et al., 2001). From this
ate cAMP, while its  component activates PLC2 to same analysis it was determined that T1R1 is several hun-
generate IP3. dred thousand base pairs away from D18346/sac. Unlike
Although gustducin is clearly implicated in bitter the T2R/TRB receptor family, T1r3 is a single receptor, not
responses based on results from in vitro and in vivo stud- part of a multigene cluster.
ies, it also is thought to play a role in sweet. -Gustducin At the amino acid sequence level T1r3 is ~30% related
null mice have markedly diminished chorda tympani and to T1r1 and T1r2 (Hoon et al., 1999). T1r3 is expressed
glossopharyngeal nerve responses to sucrose and selectively in TRCs in fungiform, foliate and circumvallate
SC45647. In two bottle preference tests the -gustducin papillae and geschmacksstreifen (Max et al., 2001;
null mice showed diminished preference for both sucrose Montmayeur et al., 2001; Sainz et al., 2001). T1r3 is co-
and SC45647. There are also biochemical data to directly expressed with T1r1 in anterior TRCs and with T1r2 in
implicate gustducin in sweet transduction: although posterior TRCs (Nelson et al., 2001; Montmayeur et al.,
neither sucrose nor artificial sweeteners activate gustducin 2001).
in the presence of bovine taste receptor-containing A comparison of the sequence of T1r3 from several
membranes (as is the case with many bitter compounds) strains of mice identified two polymorphisms that differen-
(Ruiz-Avila et al., 1995; Ming et al., 1998), a number of tiated all sweet sensitive “taster” strains of mice from all
artificial sweeteners competitively inhibit bitter receptor sweet-insensitive “nontaster” strains (Bachmanov et al.,
activation of gustducin, suggesting that these compounds 2001; Max et al., 2001; Montmayeur et al., 2001; Nelson et
may act as bitter antagonists (Ming et al., 1999). This type al., 2001; Sainz et al., 2001). These polymorphisms are
of interaction may underlie peripheral aspects of bitter- within the N-terminal extracellular region of T1r3 that based
sweet mixture suppression and the observation that quinine on homology to the mGluR1 glutamate receptor is predicted
suppresses chorda tympani nerve responses to sucrose to be involved in GPCR dimerization and ligand binding
(Formaker and Frank, 1996). (Kunishima et al., 2000; Max et al., 2001). Confirmation
that T1r3 is indeed sac came from the conversion of non-
F. Genetic Mapping of Sweet Taste Loci and taster mice into tasters by the transgenic expression of the
the Search for Sweet Receptors taster form of T1r3 (Nelson et al., 2001; M. Rong, W. He, S.
Damak, and R. F. Margolskee, unpublished). Heterologous
Gs, -gustducin, AC, PLC2, basolateral K channels, expresion of T1r3 together with T1r2 demonstrated that this
and cNMP-gated channels are all potential components combination responds to several natural and artificial sweet-
of one or more sweet transduction pathways. However, the eners, suggesting that T1r2 and T1r3 form a heterodimeric
sweet-responsive receptors themselves were only recently receptor (Nelson et al., 2001). How T1r2/T1r3 can bind to
cloned. A number of laboratories used genomics-based such diverse sweeteners is presently unknown.
approaches such as were used to clone the mammalian
T2R/TRB receptors to identify sweet taste receptors (Max G. Proposed Models for Sweet Transduction
et al., 2001; Montmayeur et al., 2001; Sainz et al., 2001).
In mouse, sac and dpa are genetically mapped loci that There are at least three possible sweet transduction
provide major contributions to differences between sweet- pathways:
sensitive and sweet-insensitive strains (Lush et al., 1995).
Sac has been finely mapped to the distal end of mouse 1. A GPCR-Gs-cNMP pathway—sugars are thought to
chromosome 4, while dpa has been less finely mapped bind to and activate one or more GPCRs (e.g.,
to the proximal portion of mouse chromosome 4 T1r2/T1r3) coupled to Gs; activated Gs would
718 Gilbertson and Margolskee

activate AC to generate cAMP; cAMP activates pro- epithelia (Garty and Palmer, 1997). Thus, it was hypothe-
tein kinase A, which phosphorylates a basolateral K sized that the apically localized epithelial Na channel
channel, leading to closure of the channel, depolar- found in the kidney and in other Na-conserving tissues
ization of the taste cell, voltage-dependent Ca2 was utilized by TRCs in the recognition of Na salts. The
influx, and neurotransmitter release. mechanism involves Na influx through the apical
2. A GPCR-Gq/G-IP3 pathway—artificial sweeteners amiloride-sensitive sodium channels leading to a direct
presumably bind to and activate one or more GPCRs depolarization of the TRCs and subsequent neurotransmit-
(e.g., T1r2/T1r3) coupled to PLC2 by either the  ter release, similar to the model in Na-transporting
subunit of Gq or by G subunits; activated Gq or epithelia proposed decades earlier (Koefoed-Johnsen and
released G would activate PLC2 to generate IP3 Ussing, 1958) (Fig. 2A).
and DAG; IP3 and DAG elicit Ca2 release from The involvement of amiloride-sensitive pathways in the
internal stores, leading to depolarization of the taste transduction of sodium salts subsequently has been docu-
cell and neurotransmitter release. mented in numerous studies carried out in a variety of
3. A ligand-gated cation channel has been identified in species (for detailed reviews, see Lindemann, 1996;
dog and human—sweet compounds lead to cation Boughter and Gilbertson, 1999; Lindemann et al., 1999;
influx via an amiloride-blockable Na channel; this Gilbertson et al., 2000). Two important points have been
pathway is insensitive to second messengers and is demonstrated in these studies. First, the amiloride-sensi-
not known to occur in rats, hamsters, or mice. tive Na channels in TRCs appear structurally and
functionally similar to the well-described epithelial Na
Direct evidence suggests that at least the first two path-
channels (ENaC) found in other tissues. Second,
ways exist in the same subset of TRCs.
amiloride-sensitive pathways cannot account for the entire
For those components of these proposed pathways that
gustatory response to NaCl.
have been molecularly cloned or otherwise identified in
TRCs, it should be possible to confirm their presence and
B. Taste Cells with ENaC-Like Channels
functionality in sweet-responsive TRCs. Among the pro-
posed transduction components, T1r2, T1r3, Gs, Gq,
The properties of ENaC in Na-conserving epithelia
G (of various types), and PLC2 have been shown by
have been described in detail (for review see Palmer, 1992;
molecular methods to be present in TRCs. The amiloride-
Garty and Palmer, 1997). Electrophysiological recordings
blockable Na channel inferred to be involved in sweet
have demonstrated that the amiloride-sensitive currents in
responses has not yet been cloned. No TRC-expressed ACs
TRCs, like the native ENaC in most Na-transporting
have been cloned or otherwise characterized. Eventually it
epithelia, are sensitive to amiloride in the submicromolar
should be possible to reconstitute the entire sweet-respon-
range (inhibition constants  1 M). Benzamil, a more
sive pathway in vitro using molecularly cloned and/or
specific inhibitor of ENaC than amiloride (Kleyman and
purified components.
Cragoe, 1988), also inhibits amiloride-sensitive Na cur-
rents in TRCs (Gilbertson and Fontenot, 1998) consistent
with effects on ENaC-like channels and not other
IV. SODIUM SALT TRANSDUCTION
amiloride-sensitive targets (Kleyman and Cragoe, 1988).
MECHANISMS
Though there has been no detailed biophysical characteri-
A. Sodium Salt Transduction Utilizing zation of the amiloride-sensitive Na channels in mam-
Ion Channels malian TRCs, a preliminary report (Avenet, 1992) showed
that they have the small unitary conductance (~5 pS)
The earliest studies of taste transduction involved those predicted of the common ENaC in the kidney.
aimed at elucidating the mechanism for the detection of Another similarity between the common ENaC and
sodium salts, which are generally thought of as the proto- the amiloride-sensitive Na channels in TRCs is in the
typical ‘salty’ stimuli in humans. These initial reports saturation of the channel with increasing concentrations
using transepithelial current recording (DeSimone et al., of mucosal Na. In transporting epithelia, this saturation
1981), afferent nerve recording (Brand et al., 1985), and has been attributed to feedback inhibition due to
human psychophysics (Schiffman et al., 1983) were con- increases in intracellular Na that activate secondary,
sistent in showing that the response to NaCl was at least Ca2-dependent mechanisms (Ling and Eaton, 1989),
partially sensitive to the diuretic amiloride. Amiloride’s self-inhibition due to a direct inhibitory effect of Na on
action as a diuretic depends on its well-known activity to ENaC (Fuchs et al., 1977), decreases in driving force due
inhibit voltage-insensitive Na channels in transporting to large-scale changes in intracellular Na (Lindemann,
Molecular Physiology of Gustatory Transduction 719

Figure 2 Models for Na current flow in taste cells. (A) Apical amiloride-sensitive Na channels allow influx of Na ions directly
depolarizing the taste receptor cell (transcellular pathway). Net current flow is inward through these ENaC-like channels and outward via
the Na/K ATPase. The closed current loop (dotted line) is through the tight junction permeable to small ions. (B) For influx of Na in
taste cells lacking apical ENaC-like channels, Na ions must first permeate the tight junctions (paracellular pathway). Na ions may then
enter the cell through basolateral amiloride-sensitive channels. Since amiloride cannot permeate the tight junctions, this mechanism
would appear insensitive to mucosal amiloride. In this case two current loops are present: one through the tight junctions as in (A) and
another through the basolateral membrane. (Adapted from Lindemann et al., 1999.)

1984), and saturation at the Na-binding site within and ) are expressed in the taste buds from the fungiform
ENaC itself. Amiloride-sensitive Na conductances papillae in the anterior rat tongue (Lin et al., 1999; Kretz
recorded both in isolated TRCs and from the lingual et al., 1999) and the fungiform taste buds also exhibit
epithelium show evidence of saturation. On the basis of significant amiloride-sensitive Na currents (Gilbertson
the time course of the development of saturation and its and Zhang, 1998; Gilbertson and Fontenot, 1998).
sensitivity to the sulfhydryl reagent p-hydrozymer- Conversely, in the rat, taste buds from the CV papillae
curibenzoate, this phenomenon, which is found only in express primarily the  subunit and relatively little  or
cells containing amiloride-sensitive Na channels, has . These taste buds have voltage-independent Na con-
been attributed to self-inhibition (Gilbertson and Zhang, ductances qualitatively similar to those expected from
1998). The presence of additional mechanisms (e.g., ENaC channels, yet these conductances are insensitive
feedback inhibition, or changes in driving force) has not to amiloride and benzamil (Gilbertson and Fontenot,
been ruled out. Taken together, the amiloride-sensitive 1998; Gilbertson and Zhang, 1998). One interpretation
Na channels in TRCs appear functionally similar in of these data is that the relative expression of ENaC sub-
many respects to the ENaC found in a variety of other units in the gustatory system may be critical for estab-
Na-conserving epithelia lishing the set point for salt responsiveness in the oral
Recent molecular approaches have also demonstrated cavity. Moreover, the regulation of the expression of
the presence of both ENaC-like protein and RNA in these molecular components by natriferic hormones
taste tissue. ENaC, which is a member of the may contribute to the adaptive nature of the gustatory
Degenerin/ENaC superfamily of ion channels (see system, at least in terms of sodium salt recognition.
Benos and Stanton, 1999), consists of 3 homologous Consistent with this interpretation, a recent report
subunits(, , and ), each containing only two trans- demonstrated that aldosterone caused increased expres-
membrane regions, making them the structurally sim- sion of the  and  subunits of ENaC in all papillae and
plest of ion channels found to date. The  subunit is concomitantly led to an increase in the amplitude of the
capable of making functional (i.e., Na-conductive) amiloride-sensitive Na currents recorded electrophysi-
channels when expressed alone, but the amiloride-sensi- ologically (Lin et al., 1999). Thus, the relative expres-
tive Na currents are enhanced by two orders of magni- sion of ENaC subunits in the oral cavity may affect both
tude when the  and  subunits are coexpressed with  the magnitude as well as the pharmacology of the
(Canessa et al., 1994). All of the ENaC subunits (, , response to Na salts.
720 Gilbertson and Margolskee

Differential ENaC subunit expression may also explain Na concentrations around the basolateral membrane of
some of the species differences with respect to the contri- the TRCs, where it may enter TRCs through amiloride-
bution of amiloride-sensitive Na currents to overall salt sensitive Na channels situated on this surface (see Fig.
responsiveness. These differences are significant and range 2B). The differential permeability of the tight junctions to
from those, like rat, in which a large degree of the NaCl various anions has been hypothesized to account for the
response is amiloride-sensitive (Brand et al., 1985; Doolin differences in the intensity and taste of various sodium
and Gilbertson, 1996; Gilbertson and Fontenot, 1998), to salts (i.e., sodium chloride vs. sodium gluconate).
those in some mouse strains that are apparently completely One caveat of this mechanism is that there may be
insensitive to this inhibitor (Ninomiya et al., 1989; amiloride-sensitive Na channels on the basolateral
Miyamoto et al., 1998, 1999). Apparently, those ENaC- membrane of TRCs. Though such high-affinity amiloride-
like channels expressed at the apical membrane may be the sensitive Na channels have been found in frog TRCs
ones important for imparting a high degree of salt sensitiv- (Avenet and Lindemann, 1989), evidence for their pres-
ity. Miyamoto and colleagues (1999) have shown that ence in rodents, for example, is equivocal. As alluded to
C57B1/6 mice, which are more salt responsive than above, antibodies against ENaC have labeled TRC profiles
BALB/c mice, have greater amiloride sensitivity at their on their basolateral margins, indicative of the presence of
apical receptive membranes than do the BALB/c mice. an ENaC-like protein on this surface. Consistent with this,
Despite this difference, when the whole cell (i.e., apical Mierson et al. (1996) found a low-affinity (Ki ~ 50 M),
and basolateral membranes) was exposed to amiloride, no amiloride-sensitive Na conductance using transepithelial
difference between amiloride sensitivity in the two strains recording. Gilbertson and colleagues failed to identify
was found. Within a species this difference is noted as either a high- or low-affinity basolateral amiloride-
well. Recordings from the rat chorda tympani nerve that sensitive conductance in rat or hamster using both transep-
innervates the anterior tongue including the fungiform ithelial and patch clamp recording (Doolin and Gilbertson,
taste buds are more salt responsive than those of the glos- 1996; Gilbertson and Zhang, 1998; Gilbertson and
sopharyngeal nerve innervating the CV taste buds and Fontenot, 1998). The correlation between ENaC expres-
posterior tongue (Harada and Smith, 1992; Ninomiya sion, particularly specific ENaC subunit expression, and
et al., 1994). Both epithelial transport studies (Gilbertson functional amiloride-sensitive Na channels remains an
and Zhang, 1998) and patch clamp recordings (Gilbertson open question that needs to be systematically addressed.
and Fontenot, 1998) have shown that the high specificity, Additional mechanisms, independent of apical and/or
apical amiloride-sensitive conductance found in the fungi- basolateral amiloride-sensitive Na channels, also likely
form TRCs is lacking in those from the CV papillae. contribute to salt transduction. Though the molecular
Clearly, more systematic analyses of the strain and taste components of these mechanisms have not been identified,
bud differences in salt responsiveness and expression several independent investigations have concluded that a
of ENaC subunits is needed to strengthen this intriguing non-specific apically localized cation channel contributes
correlation. to sodium salt transduction (Doolin and Gilbertson, 1996;
Gilbertson and Zhang, 1998; Miyamoto et al., 1999). This
C. Amiloride-Insensitive Pathways in Salt Taste balance between amiloride-sensitive, amiloride-insensitive
Transduction and paracellular pathways appears variable among differ-
ent species and may account for some of the strain and
A common theme emerging in the study of taste transduc- species differences in salt responsiveness reported.
tion is that there is no single mechanism for the detection
of any individual class of taste stimuli. Consistent with this D. Proposed Models for Salt Transduction
idea, apical amiloride-sensitive Na channels cannot com-
pletely account for sodium salt responsiveness. Some of The transduction of sodium salts, like that for other modal-
the evidence suggesting the presence of additional salt ities, involves more than a single transduction scheme. The
transduction pathways has come from experiments that clearest example of a mechanism for sodium salt transduc-
showed an insensitivity of salt responses to mucosal appli- tion in taste receptor cells involves the movement of Na
cations of amiloride (Formaker and Hill, 1991). One of the ions through apically localized epithelial Na channels,
mechanisms that has been proposed to account for these similar to the ENaC channels found in other Na-absorb-
responses involves the movement of Na (and other small ing epithelia. Simply, Na ions in the oral cavity diffuse
ions) through the tight junctions between the TRCs (Elliot down their electrochemical gradient, entering the taste cell
and Simon, 1990; Ye et al., 1991). This “paracellular path- through these open ENaC-like channels leading to a direct
way” for Na flux would lead to changes in interstitial depolarization of the cell. As alluded to earlier, this
Molecular Physiology of Gustatory Transduction 721

depolarization activates voltage-dependent Ca2 influx Cummings and Kinnamon, 1992). This conductance, par-
and eventually release of the taste cell neurotransmitter tially open at the cells’ resting potential, would lead to a
onto gustatory afferents. The involvement of paracellular direct depolarization of the TRC. The applicability of this
pathways in sodium salt transduction may also lead to taste mechanism beyond lower vertebrates remains in doubt,
cell activation. In this mechanism, the selective permeabil- however, since the evidence suggesting that there is an api-
ity of the intercellular tight junctions between taste cells cal K conductance in mammals is scant (Gilbertson et al.,
allows the movement of small ions (Na, Cl, H) into the 1992) and may be highly species dependent (Gilbertson
basolateral margins of the taste buds. Once in the intersti- and Zhang, 1998). Other mechanisms have been proposed
tial space below the tight junctions, Na ions may diffuse in lower vertebrates that similarly have not been demon-
through putative ENaC-like channels present on the strated in higher organisms to date. These include the
basolateral membrane leading to a direct depolarization of ability of acids to alter the electrical coupling between
the taste cell. The entire sodium salt response in taste cells TRCs (Bigiani and Roper, 1994) and to activate a cationic
cannot, however, be completely accounted for by the conductance (Okada et al., 1987, 1994; Miyamoto et al.,
apical ENaC-like channels and by the aforementioned 1988).
paracellular pathway. Thus, though there is little direct
evidence for their existence, additional mechanisms have B. Acid-Sensing ENaC-Like Channels and Acid
been hypothesized to account for these ENaC and paracel- Transduction
lular independent pathways.
To date, none of the molecular components of the salt The recent cloning of several new members of the
transduction pathway have been cloned from taste tissue. Degenerin/ENaC superfamily of ion channels (Benos and
Though the ENaC in other transporting epithelia have been Stanton, 1999) that have the capability to sense and
cloned and functionally expressed, this has not been done respond to changes in acidity implicated their involvement
for the ENaC-like channels of TRCs. Given the increase in in acid sensing in TRCs as well (Lindemann et al., 1999).
molecular approaches being used in the study of the taste The ENaC-like channel found in hamster TRCs had previ-
system, this shortcoming likely will be rectified in the near ously been implicated in acid transduction, owing to its
future. It will be of particular interest to know whether significant proton permeability (Gilbertson et al., 1992,
multiple variants of ENaC-like channels are present in 1993), which is a hallmark of ENaC-like channels every-
taste cells that may explain the data showing the disjunc- where (Palmer, 1992). Significant proton influx through
tion between ENaC expression and functional amiloride amiloride-sensitive Na channels apparently only occurs
sensitivity in TRCs, as well as identify variants of the under conditions of low apical (mucosal) Na concentra-
ENaC that may have different affinities for amiloride. tions, as is typically the case in hamster (Rehnberg et al.,
1992). This may partially explain why some species like
rat and mouse do not show a strong inhibitory effect of
amiloride on acid-induced responses in the taste system
V. ACID TRANSDUCTION MECHANISMS (Kinnamon et al., 1993, DeSimone et al., 1995) while oth-
A. Effect of Acidic Stimuli on Multiple Ion Channels ers do (Gilbertson et al., 1993; Gilbertson and Gilbertson,
and Other Targets 1994). Expression of alternative combinations of ENaC
subunits, as discussed above, may also be predicted to alter
The proton is primary stimulus for the transduction of the properties of acid responses in TRCs.
acids, which humans refer to as sour (Settle et al., 1986). Other amiloride-sensitive elements have also been
Acidic stimuli activate TRCs, afferent nerve fibers, and implicated in the acid response in TRCs. Waldmann et al.
central gustatory neurons in a concentration-dependent (1997), in their report describing the cloning and func-
fashion (for review, see Herness and Gilbertson, 1999). tional expression of another member of the
Protons have widespread actions, affecting a variety of pH- Degenerin/ENaC family, the acid-sensing channel, or
sensitive cellular targets and coupled with their high ASIC, suggested that this channel may be responsible for
permeability through many types of ion channels and inter- acid transduction in the taste system as well. To date,
cellular junctions; it is not surprising that a variety of several ASICs, including ASIC1, ASIC2a, ASIC2b, and
potential transduction mechanisms for these compounds ASIC4, have been identified in taste tissue (Liu and Simon,
have been described (see Fig. 1). One of the first descrip- 2001; Ugawa et al., 1998). Using reverse transcription
tions of a mechanism underlying acid transduction polymerase chain reaction, Liu and Simon (1999) failed to
involved a direct proton-mediated inhibition of an apical find ASIC (ASIC-) and another acid-sensitive member of
K channel in mudpuppy (Kinnamon et al., 1988; the family, DRASIC (dorsal root acid-sensing channel), in
722 Gilbertson and Margolskee

TRC RNA, yet they did find a sensory-specific splice vari- have not been detailed. Finally, a recent study identified
ant of ASIC (Chen et al., 1998), ASIC-, and another the presence of hyperpolarization-activated and cyclic
intriguing candidate for a role in taste transduction, the nucleotide-gated channels (HCN1, HCN4) in taste tissue
acid-sensitive vanillinoid receptor (VR1), that responds to where they were shown to be responsive to low pHs asso-
capsaicin, heat, and low pH (Welch et al., 2000). Another ciated with sour taste (Stevens et al., 2001). As with the
candidate that has been identified in TRCs by in situ case for sodium salt transduction, it has become clear that
hybridization that may play a role in acid sensing is the there are multiple mechanisms responsible for the trans-
mammalian degenerin-1 channel (MDEG1). This channel, duction of acids by the peripheral gustatory system.
also known as the brain-type Na channel or BNaCl, when
expressed in heterologous systems like Xenopus oocytes, C. Proposed Models for Acid Transduction
expresses large acid-induced currents (Ugawa et al., 1998).
Recently, Lin et al. (2002) demonstrated that acid-induced Several different transduction pathways have been
currents in rat vallate taste receptor cells are functionally reported to be involved in acid sensing in TRCs. In
similar to ASIC2-mediated currents in both native cells and amphibia, apically localized K channels have been shown
heterologous expression systems. Interestingly, acetic acid to be inhibited directly by acids (i.e., protons) in a concen-
generates larger responses from MDEG-1 than does HCl at tration-dependent fashion, leading to taste cell depolariza-
an equivalent pH, which correlates with the perceived tion. Activation of cation channels by acids has also been
intensities of these two acids by humans. Though it is not reported and would be expected to depolarize taste cells
clear yet what roles these identified elements (ASIC, VR1, via cation influx at normal TRC resting potentials. Direct
MDEG-1) actually play in acid taste, the ability of all proton influx through the ENaC-like channels has been
members of this family to function as proton-gated cation reported in mammalian TRCs. Proton influx would
channels that are sensitive to amiloride is consistent with directly depolarize the TRC, though this process appar-
the hypothesized mechanisms for acid taste transduction ently requires a low concentration of mucosal Na ions in
(Fig. 2). More detailed experiments linking these molecu- order for sufficient proton entry to activate the TRC. The
lar approaches with the physiological properties of acid molecular identification of several acid-sensing channels
responses in TRCs are needed. in taste cells may prove promising to our understanding of
Other acid-gated conductances have also been impli- acid transduction, but there is little physiological evidence
cated as contributing to acid responses in mammalian to support a functional role for these candidate transduc-
TRCs. Two recent reports have independently demon- tion elements. The recent finding that NPPB-sensitive con-
strated that acid responses in mouse (Miyamoto et al., ductances in TRCs may contribute to acid taste implicates
1998) and rat (Lin et al., 2000) TRCs are sensitive to the the involvement of other ion channels in acid transduction.
compound NPPB (5-nitro-2-[3-phenylpropylamino]- Finally, similar to the contribution of the paracellular path-
benzoic acid). While Miyamoto et al., (1998) concluded way to sodium salt taste transduction, the proton perme-
the NPPB sensitivity was consistent with the activation of ability of the tight junctions in taste buds and the ability of
a Cl conductance, Lin et al. (2000) showed evidence TRCs to track these pH changes may contribute to acid
consistent with acids activating primarily a NPPB-sensi- sensing. It is clear that the ubiquity of proton effects on ion
tive cation current, which was Cl independent. Finally, channels and other acid-sensitive intracellular targets will
as mentioned above, the high permeability of protons result in the identification of numerous candidates for acid
through the paracellular pathways in the lingual epithe- transduction components.
lium has been hypothesized to play a role in acid taste
(DeSimone et al., 1995). The data suggest that protons
alter the permeability of the tight junctions between VI. AMINO ACID TRANSDUCTION
TRCs from cation-selective to anion-selective, changing MECHANISMS
the local ionic environment around the taste bud proper.
Clearly, the properties of the paracellular pathway have a Monosodium glutamate (MSG) has often been considered
profound influence on the performance of the TRCs to have a unique taste, separate from the other four gener-
themselves (Stewart et al., 1997). TRCs also have the ally agreed-upon basic taste qualities of sweet, bitter, salty,
ability to track changes in extracellular pH intracellularly and sour (Yamaguchi, 1991). Yet until very recently, little
(Stewart et al., 1998), and these changes in intracellular was known about how the amino acid glutamate is
pH have the potential to affect a variety of pH-sensitive detected in TRCs. Most of the early research on amino
intracellular targets (ion channels, enzymes). However, acid transduction mechanisms centered on the channel cat-
specific effects of intracellular changes in pH in TRCs fish, Ictalurus punctatus, which has an exquisitely sensi-
Molecular Physiology of Gustatory Transduction 723

tive system for the detection of amino acids. The catfish cific variant of mGluR4 has provided support for this the-
has at least three broad classes of amino acid receptors. ory. Chaudhari and colleagues (2000) successfully cloned
Individual receptors for L-arginine and L-proline activate and expressed this mGluR4 variant, called taste-mGluR4,
cation channels, while those for the third major class, that has all the hallmarks expected of the umami receptor.
L-alanine, activate metabotropic receptor(s) that ultimately Consistent with Lin and Kinnamon’s electrophysiological
lead to production of the second messengers, cAMP and study on native rat TRCs (1999), the heterologously
IP3 (for reviews on amino acid transduction in catfish, see expressed taste-mGluR4 bound glutamate in the milli-
Brand et al., 1991; Caprio et al., 1993). molar range was inhibited by L-AP4 and led to decreased
More recently, however, the emphasis on amino acid production of cAMP. In addition, taste-mGluR4 activation
transduction mechanisms has shifted to mammalian by L-glutamate was potentiated by 5’-ribonucleotides,
systems and the mechanism underlying umami taste. another property of umami responses in vivo (Yamamoto
Though L-glutamate activates both ionotropic and et al., 1991). Though not definitively proven, it appears
metabotropic receptors in mammalian TRCs, the latter is that the taste-mGluR4 is an excellent candidate for the
believed to represent the mechanism for the taste of umami receptor mediating umami taste.
(Chaudhari and Roper, 1998). Consistent with the ability of Two recent studies have shed more light on the mecha-
L-glutamate to activate two types of receptors, two types of nism(s) underlying the response to L-amino acids like L-
responses are recorded from rat TRCs during glutamate glutamate. The heterodimer T1r1/T1r3 is expressed in a
stimulation. In patch clamp recordings of rat TRCs, Lin and subset of taste receptors cells and responds to a variety of
Kinnamon (1999) showed that depolarizing responses to L- L-type amino acids (Li et al., 2002; Nelson et al., 2002).
glutamate were mimicked by N-methyl-D-aspartate Moreover, as predicted for an umami receptor, responses of
(NMDA), inhibited by NMDA antagonists like AP-5 (D-2- heterologously expressed T1r1/T1r3 to L-type amino acids
amino-5-phosphonobutyric acid) and potentiated by were potentiated by 5’-ribonucleotides and were elicited
glycine. This response was thought to be representative of by concentrations of amino acids that were physiologically
synaptic responses to L-glutamate. Hyperpolarizing relevant. Thus, the T1r1/T1r3 heterodimer appears to be a
responses to L-glutamate, on the other hand, were thought broadly tuned amino acid receptor that may contribute to
to represent the transduction mechanism for umami taste. umami taste. The relative contributions of mGluR4 and
Consistent with activity at metabotropic glutamate recep- T1r1/T1r3 to umami taste remains to be determined.
tors (mGluR), these hyperpolarizing responses were inhib-
ited by the mGluR antagonist CPPG ([RS]--cyclopropyl-
4-phosphophenyl-glycine) and mimicked by the mGluR VII. UNASSIGNED CANDIDATE
agonist L-2-amino-phosphonobutyric acid (L-AP4). TRANSDUCTION ELEMENTS
Evidence for a role of cAMP in the hyperpolarizing
response was demonstrated by the ability of a membrane In a number of cases, ion channels, G proteins, and GPCRs
permeant analog of cAMP (8-bromo-cAMP) to inhibit this have been molecularly cloned from TRCs or otherwise
response. Thus, one might propose a mechanism for umami shown to be expressed in TRCs but, lacking functional
taste involving mGluR activation, leading to activation of a data, could not be assigned even tentatively to any trans-
PDE and a subsequent decrease in intracellular cAMP con- duction pathway(s). In the absence of such corroborating
centration (Fig. 1). The ultimate target that is modulated by evidence, these potential taste transduction elements must
this decrease in cAMP is not known, but has been hypoth- be considered candidates only. Prior to the identification of
esized to be the disinhibition of a cNMP-suppressible chan- the T2R/TRB GPCRs taste receptors, two groups had
nel (Chaudhari et al., 1996). cloned seven transmembrane-helix receptors from taste
The hypothesis that the umami receptors may be related tissue (Abe et al., 1993a,b; Matsuoka et al., 1994): these
to glutamate receptors in the mammalian brain was cloned receptors were quite similar to the family of odor-
initially posed during experiments that demonstrated that ant receptors cloned by Buck and Axel (1991). However, it
ligands of brain glutamate receptors (ibotenate, aspartate, seems unlikely that these particular receptors play any
tricholomic acid, homocysteinic acid) could elicit umami prominent role in taste transduction: in one case it was
taste in humans (Maga, 1983) and induce responses in the shown that the receptor is not expressed in the taste buds,
chorda tympani nerve (Faurion, 1991). Because of their but rather in the surrounding nontaste epithelium (and in
ability to bind peptides that can elicit umami taste, type 4 this case only under conditions of reduced stringency)
metabotropic glutamate receptors (mGluR4) have been (Abe et al., 1993b). In the other case (Matsuoka et al.,
suggested to play a role in the taste response to MSG 1994), Northern blot analysis showed that the predominant
(Monastryskaia et al., 1999). Identification of a taste-spe- site of expression was in the olfactory epithelium (orders
724 Gilbertson and Margolskee

of magnitude higher than in tongue epithelium). Abe, K., Kusakabe, Y., Tanemura, K., Emori, Y., and Arai, S.
Furthermore, mRNA expression levels in tongue contain- (1993b). Primary structure and cell-type specific expression
ing CV papillae, foliate papillae, or no taste papillae were of a gustatory G protein-coupled receptor related to olfactory
approximately equal, suggesting that this receptor is receptors. J. Biol. Chem. 268:12033–12039.
Adler, E., Hoon, M. A., Mueller, K. L., Chandrashekar, J., Ryba,
expressed at only very low levels and in the nontaste
N. J. P., and Zuker, C. S. (2000). A novel family of mam-
epithelium of the tongue. To prove that a receptor protein
malian taste receptors. Cell 10:693–702.
plays a role in taste transduction requires the demonstra- Akabas, M. H., Dodd, J., and Al-Awqati, Q. (1988). A bitter sub-
tion of its presence in the TRCs themselves, followed by a stance induces a rise in intracellular calcium in a subpopula-
demonstration of the biological function of the protein in tion of rat taste cells. Science 242:1047–1050.
taste transduction. Asanuma, N., and Nomura, H. (1982). Histochemical localiza-
tion of adenylate cyclase and phosphodiesterase activities in
the foliate papillae of the rabbit. II. Electron microscopic
VIII. FUTURE PROSPECTS observations. Chem. Senses 7:1–9.
Asanuma, N., and Nomura, H. (1995). Cytochemical localization
During the past 10 years our knowledge of the molecular of guanylyl cyclase activity in rabbit taste bud cells. Chem.
components underlying taste transduction has increased Senses 20:231–237.
dramatically. The powerful combination of biochemistry, Avenet, P. (1992). Role of amiloride-sensitive sodium channels in
molecular biology, electrophysiology, genetics, and trans- taste. In Sensory Transduction, D. P. Corey and S. D. Roper
genic mouse models has provided several important new (Eds.). Rockefeller University Press, New York, pp. 271–280.
insights into the cellular and molecular mechanisms of taste Avenet, P., and Lindemann, B. (1987). Patch-clamp study of iso-
transduction. The identification and molecular cloning of lated taste receptor cells of the frog. J. Membr. Biol.
97:223–240.
gustducin’s  subunits, followed by the generation and
Avenet, P., and Lindemann, B. (1989). Chemoreception of salt
in vivo analysis of -gustducin knockout mice provided an
taste: the blockage of stationary sodium currents by amiloride
understanding of the physiology of taste transduction at the in isolated receptor cells and excised membrane patches. In
molecular level. The use of molecular cloning and Chemical Senses: Molecular Aspects of Taste and Odor
genomics to identify and clone taste receptors responsive to Reception, J. G. Brand, J. H. Teeter, R. H. Cagan and M. R.
bitter (T2R/TRB), sweet (T1r2, T1r3), and glutamate Kare (Eds.)., Marcel Dekker, New York, pp. 171–182.
(T1r1, T1r3, mGluR4) has opened up new avenues to Avenet, P., Hofmann, F., and Lindemann, B. (1988a).
approach these pathways. Behavioral and electrophysiolog- Transduction in taste receptor cells requires cAMP-dependent
ical analysis of transgenic and knockout mice altered in protein kinase. Nature 331:351–354
these (and other) taste transduction components will pro- Avenet, P., Hofmann, F., and Lindemann, B. (1988b). Signalling
vide new insights into the functions of these particular pro- in taste receptor cells: cAMP-dependent protein kinase causes
depolarization by closure of 44 pS K-channels. Comp.
teins and taste pathways in general and provide a means to
Biochem. Biophysiol. 90A:681–685.
test specific hypotheses regarding the involvement of par-
Bachmanov, A. A., Reed, D. R., Ninomiya, Y., Inoue, M.,
ticular components in specific taste transduction pathways. Tordoff, M. G., Price, R. A., and Beauchamp, G. K. (1997).
During the next 5 years we expect it to be possible to recon- Sucrose consumption in mice: major influence of two genetic
stitute entire taste transduction pathways in vitro using loci affecting peripheral sensory responses. Mammal Genome
cloned and/or biochemically purified reagents. 8:545–548.
Bachmanov, A. A., Li, X., Reed, D. R., Ohmen, J. D., Li, S.,
Chen, Z., Tordoff, M. G., de Jong, P. J., Wu, C., West, D. B.,
ACKNOWLEDGMENTS Chatterjee, A., Ross, D. A., and Beauchamp, G. K. (2001).
Positional cloning of the mouse saccharin preference (sac)
This research was supported by the following grants: NIH locus. Chem. Senses 26:925–933.
DC03055 and NIH DC03155 to RFM; NIH DC02507, Bartoshuk, L. M. (1979). Bitter taste of saccharin: related to the
NIH DC00353, and a Novartis Research Award to TAG. genetic ability to taste the bitter substance G-n-propylth-
RFM is an Associate Investigator of the Howard Hughes iouracil (PROP). Science 205:934–935.
Medical Institute. Bartoshuk, L. M., Dateo, G. P., Vanderbelt, D. J., Buttrick, R. L.,
and Long, L., Jr. (1969). Effects of gymnema sylvestre dulci-
ficum on taste in man. In Olfaction and Taste, Vol. 3, C.
REFERENCES Pfaffman (Ed.). Rockefeller University Press, New York, pp.
436–444.
Abe, K., Kusakabe, Y., Tanemura, K., Emori, Y., and Arai, S. Bartoshuk, L. M., Murphy, C., and Cleveland, C. T. (1978).
(1993a). Multiple genes for G protein-coupled receptors and Sweet taste of dilute NaCl: phsychophysical evidence for a
their expression in lingual epithelia. FEBS Lett. 316:253–256. sweet stimulus. Physiol. Behav. 21:609–613.
Molecular Physiology of Gustatory Transduction 725

Beets, M. G. J. (1978). Structure-activity relationships. In Human sensitivity to sucrose octaacetate in mice. Behav. Genet.
Chemoreception. Science Publishers, London, p. 354. 22:655–663.
Behe, P., DeSimone, J. A., Avenet, P., and Lindemann, B. (1990). Caprio, J., Brand J. G., Teeter, J. H., Valentincic, T., Kalinoski, D.
Membrane currents in taste cells of the rat fungiform papilla. L., Kohbara, J., Kumazawa, T., and Wegert, S. (1993). The
J. Gen. Physiol. 96:1061–1084. taste system of the channel catfish: from biophysics to behav-
Benos, D. J., and Stanton, B. A. (1999). Functional domains ior. Trends Neurosci. 16:192–197.
within the degenerin/epithelial sodium channel (Deg/ENaC) Chandrashekar, J., Mueller, K. L., Hoon, M. A., Adler, E., Feng,
superfamily of ion channels. J. Physiol. 520:631–644. L., Guo, W., Zuker, C. S., Ryba, N. J. P. (2000). T2Rs func-
Bernhardt, S. J., Naim, M., Zehavi, U., and Lindemann, B. tion as bitter taste receptors. Cell 100:703–711.
(1996). Changes in IP3 and cytosolic Ca2 in response to sug- Chaudhari, N., and Roper, S. D. (1998). Molecular and physio-
ars and non-sugar sweeteners in transduction of sweet taste in logical evidence for glutamate (umami) taste transduction via
the rat. J. Physiol. 490:325–336. a G protein-coupled receptor. Ann. NY Acad. Sci.
Bigiani, A., and Roper, S. D. (1994). Reduction of electrical cou- 855:398–406.
pling between Necturus taste receptor cells, a possible role in Chaudhari, N., Yang, H., Lamp, C., Delay, E., Cartford, C., Than,
acid taste. Neurosci. Lett. 176:212–216. T., and Roper, S. (1996). The taste of monosodium glutamate:
Blizzard, D. A., Kotlus, B., Frank, M. E. (1999). Quantitative trait membrane receptors in taste buds. J. Neurosci. 16:3817–3826.
loci associated with short-term intake of sucrose, saccharin Chaudhari, N., Landin, A. M., and Roper, S. D. (2000). A
and quinine solutions in laboratory mice. Chem. Senses metabotropic glutamate receptor variant functions as a taste
24:373–385. receptor. Nat. Neurosci. 3:113–119.
Boughter Jr., J. D., Pumplin, D. W., Yu, C., Christy, R. C., and Chen, C. C., England, S., Akopian, A. N., and Wood, J. N. (1998).
Smith, D. V. (1997). Differential expression of alpha-gust- A sensory neuron-specific, proton-gated ion channel. Proc.
ducin in taste bud populations of the rat and hamster. J. Natl. Acad. Sci. USA 95:10240–10245.
Neurosci. 17:2852–2858. Clapp, T. R., Stone, L. M., Margolskee, R. F., and Kinnamon, S.
Boughter, J. D. Jr., and Gilbertson T. A. (1999). From channels to C. (2001). Immunocytochemical evidence for co-expression
behavior: an integrative model of NaCl taste. Neuron of type III IP3 receptor with signaling components of bitter
22:213–215. taste transduction. BMC Neuroscience 6:2–10.
Brand, J. G., Teeter, J. H., and Silver, W. L. (1985). Inhibition by Cummings, T. A., and Kinnamon, S. C. (1992). Apical K chan-
amiloride of chorda tympani responses evoked by monovalent nels in Necturus taste cells. Modulation by intracellular fac-
salts. Brain Res. 34:207–214. tors and taste stimuli. J. Gen. Physiol. 99:591–613.
Brand, J. G., Teeter, J. H., Kumazawa, T., Huque, T., and Bayley, Cummings, T. A., Powell, J., and Kinnamon, S. C. (1993). Sweet
D. L. (1991). Transduction mechanisms for the taste of amino taste transduction in hamster cells: Evidence for the role of
acids. Physiol. Behav. 49:899–904. cyclic nucleotides. J. Neurophys. 70:2326–2336.
Brouwer, J. N., Hellekant, G., Kasahara, Y., van der Wel, H., and Cummings, T. A., Daniels C., and Kinnamon, S. C. (1996). Sweet
Zotterman, Y. (1973). Electrophysiological study on the gusta- taste transduction in hamster cells: Sweeteners and cyclic
tory effects of the sweet proteins thaumatin and monellin in nucleotides depolarize taste cells by reducing a K current. J.
monkey, guinea pig and rat. Acta Physiol. Scand. 89: 550–557. Neurophys. 75:1256–1263.
Brouwer, J. N., Glaser, D., Hard, C., Segerstad, A. T., Hellekant, DeSimone, J. A., Heck, H. F., and Bartoshuk, L. M. (1980).
G., Ninomiya, Y., and Van der Wel, H. (1983). The sweetness- Surface active taste modifiers: a comparison of the physical
inducing effect of miraculin: behavioral and neurophysiolog- and psychophysical properties of gynmemic acid and sodium
ical experiments in the rhesus monkey macaca mulatta. J. lauryl sulfate. Chem. Senses 5:317–330.
Physiol. 337:221–240. DeSimone, J. A., Heck, G. L., and DeSimone, S. K. (1981).
Buck, L., and, Axel, R. (1991). A novel multigene family may Active ion transport in dog tongue: A possible role in taste.
encode odorant receptors: A molecular basis for odor recog- Science 214:1039–1041.
nition. Cell 65:175–187. DeSimone, J. A., Callaham, E. M., and Heck, G. L. (1995).
Cagan, R. H. (1971). Biochemical studies of taste sensation. I. Chorda tympani taste response of rat to hydrochloric acid sub-
Binding of 14C-labeled sugars to bovine taste papillae. ject to voltage-clamped lingual receptive field. Am. J. Physiol.
Biochim. Biophys. Acta 252:199–206. 268:C1295–C1300.
Cagan, R. H., and Morris, R. W. (1979). Biochemical studies of Doolin R. E., and Gilbertson T. A. (1996). Distribution and
taste sensation: binding to taste tissue of 3H-labeled monellin, characterization of functional amiloride-sensitive sodium
a sweet tasting protein. Proc. Natl. Acad. Sci. USA channels in rat tongue. J. Gen. Physiol. 107:545–554.
76:1692–1696. DuBois, G. E., Walters, D. E., and Kellogg, M. S. (1993).
Canessa, C., Schild, L., Buell, G., Thorens, B., Gautschi, I., Mechanism of human sweet taste and implications for
Horisberger, J. D., and Rossier, B. C. (1994). Amiloride-sen- rational sweetener design. In Flavor Measurement, C-. T. Ho
sitive epithelial sodium channel is made of three homologous and C. H. Manley (Eds.). Marcel Dekker, New York,
subunits. Nature 367:463–467. pp. 239–266.
Capeless, C. G., Whitney, G., and Azen, E. A. (1992). Elliot, E. J., and Simon, S. A. (1990). The anion in salt taste: a
Chromosome mapping of Soa, a gene influencing gustatory possible role for paracellular pathways. Brain Res. 535:9–17.
726 Gilbertson and Margolskee

Farbman, A. I., and Hellekant, G. (1989). Evidence for a novel Hall, M. J., Bartoshuk, L. M., Cain, W. S., and Stevens, J. C.
mechanism of binding and release of stimuli in the primate (1975). PTC taste blindness and the taste of caffeine. Nature
taste bud. J. Neurosci. 9:3522–3528. 253:442–443.
Farbman, A. I., Ogden-Ogle, C. K., Hellekant, G., Simmons, S. Harada, S., and Smith, D. V. (1992). Gustatory responses of the
R., Albrecht, R. M., and Van der Wel, H. (1987). Labeling of hamster’s soft palate. Chem. Senses 17:37–51.
sweet taste-binding sites using a colloidal gold-labeled sweet Harder, D. B., Capeless, G. C., Madgio, J. C., Boughter, J. D., Jr.,
proteins, thaumatin. Scan. Microsc. 1:351–357. Gannon, K. S., Whitney, G., and Azen, E. A. (1992).
Faurion, A. (1991). Are umami taste receptor sites structurally Intermediate sucrose octa-acetate sensitivity suggests a third
related to glutamate CNS receptor sites? Physiol. Behav. allele at mouse bitter taste locus SOA and SOA-RUA identity.
49:905–912. Chem. Senses 17:391–401.
Faurion, A., and Vayssettes-Courchay, C. (1990). Taste as a high Hellekant, G. (1976). On the gustatory effects of monellin and
discriminative system: A hamster intrapapillar single unit thaumatin in dog, hamster, pig and rabbit. Chem. Senses Flav.
study with 18 compounds. Brain Res. 512:317–332. 2:97–105.
Formaker, B. K., and Frank, M. E. (1996). Responses of the ham- Hellekant, G., and Niromiya, Y. (1991). On the taste of umami in
ster chorda tympani nerve to binary component taste stimuli: chimpanzee. Physiol. Behav. 49:927–934.
evidence for peripheral gustatory mixture interactions. Brain Herness, M. S. (1987). Effect of amiloride on bulk flow and ino-
Res. 727:79–90. tophoretic taste stimuli in the hamster. J. Comp. Physiol.
Formaker, B. K., and Hill, D. L. (1991). Lack of amiloride sensi- 160:281–288.
tivity in SHR and WKY glossopharyngeal taste responses to Herness, M. S. (1992). Aldosterone increases the amiloride-
NaCl. Physiol. Behav. 50:765–769. sensitivity of the rat gustatory neural response to NaCl. Comp.
Fox, A. L. (1932). Relationship between chemical constitution Biochem. Physiol. A 103:269–273.
and taste. Proc. Natl. Acad. Sci. USA 18:115–120. Herness, M. S. (1993). Cyclic AMP increases a membrane con-
Fuchs, W., Larsen, E. H., and Lindemann, B. (1977). Current- ductance in rat taste cells. Soc. Neurosci. Abstr. 19:1428.
voltage curve of sodium channels and concentration depen- Herness, M. S., and Gilbertson, T. A. (1999). Cellular mecha-
dence of sodium permeability on frog skin. J. Physiol. (Lond.) nisms of taste transduction. Ann. Rev. Physiol. 61:873–900.
267:137–166. Hiji, T. (1975). Selective elimination of taste responses to sugars
Garty, H., and Palmer, L. G. (1997). Epithelial sodium channels: by proteolytic enzymes. Nature 256:427–429.
function, structure and regulation. Physiol. Rev. 77: 359–296. Hoon, M. A., Adler, E., Lindemeier, J., Battey, J. F., Ryba, N. J.
Gent, J. F., and Bartoshuk, L. M., (1983). Sweetness of sucrose, P., and Zuker, C. S. (1999). Putative mammalian taste recep-
neohesperidin dihydrochal cone and saccharin is related to tors: a class of taste-specific GPCRs with distinct topographic
genetic ability to taste the bitter substance 6-N-propy- selectivity. Cell 96:541–551.
thiouracil. Chem. Senses 7:265–272. Huang, L., Shanker, Y. G., Dubauskaite, J., Zheng, J. Z., Yan, W.,
Gilbertson, T. A., (2002). Hypoosmotic stimuli activate a chloride Rosenzweig, S., Spielman, A. I., Max, M., and Margolskee,
conductance in rat taste cells. Chem. Senses 27:383–394. R. F. (1999). Ggamma13 colocalizes with gustducin in taste
Gilbertson, D. M., and Gilbertson, T. A. (1994). Amiloride receptor cells and mediates IP3 responses to bitter denato-
reduces the aversiveness of acids in preference tests. Physiol. nium. Nat. Neurosci. 2:1055–1062.
Behav. 56:649–654. Hwang, P. M., Verma, A., Bredt, D.S., and Snyder, S. (1990).
Gilbertson, T. A., and Fontenot, D. T. (1998). Distribution of Localization of phosphatidylinositol signaling components in
amiloride-sensitive sodium channels in the oral cavity of the rat taste cells: role in bitter taste transduction. Proc. Natl.
hamster. Chem. Senses 23:495–499. Acad. Sci. USA 87:7395–7399.
Gilbertson, T. A., and Zhang, H. (1998a). Self-inhibition in Iwasaki, K., and Sato, M. (1984). Neural and behavioral effects
amiloride-sensitive sodium channels in taste receptors cells. of some heavy metal ions on taste nerve response in mice.
J. Gen. Physiol. 111:667–677. Jpn. J. Physiol. 34:907–918.
Gilbertson, T. A., and Zhang, H. (1998b). Characterization of Iyengar, B., Smith, P., Van der Ouderaa, F., Van der Wel, H., van
sodium transport in gustatory epithelia from the hamster and Brouweshave, J., Ravstein, P., Richters, G., and Van
rat. Chem. Senses 23:283–293. Wassenaar, P. D. (1979). The complete amino acid sequence
Gilbertson, T. A., Avenet, P., Kinnamon, S. C., and Roper, S. D. of the sweet protein thaumatin. Eur. J. Biochem. 96: 193–204.
(1992). Proton current through amiloride-sensitive Na chan- Jakinovich, W. (1985). Stimulation of gerbil’s gustatory receptors
nels in hamster taste cells: role in acid transduction. J. Gen. by methyl glycopyranosides. Chem. Senses 10:591–604.
Physiol. 100:803–824. Jakinovich, W., Jr., and Goldstein, I. J. (1976). Stimulation of the
Gilbertson, T. A., Roper, S. D., and Kinnamon, S. C. (1993). gerbil’s gustatory receptors by monosaccharides. Brain Res.
Proton currents through amiloride-sensitive Na channels in 110:491–504.
isolated hamster taste cells: enhancement by vasopressin and Jakinovich, W., Jr., and Sugarman, D. (1988). Sugar taste recep-
cAMP. Neuron 10:931–942. tion in mammals. Chem. Senses 13:13–31.
Gilbertson, T. A., Damak, S., and Margolskee, R. F. (2000). The Jones, D. T., and Reed, R. R. (1989). Golf: an olfactory neuron
molecular physiology of taste transduction. Curr. Opin. specific G-protein involved in odorant signal transduction.
Neurobiol. 10:519–527. Science 244:790–795.
Molecular Physiology of Gustatory Transduction 727

Katz, A., Wu, D., and Simon, M. I. (1992). Subunits beta gamma cells, with special reference to the occurrence of Gi2 as a
of heterotrimeric G protein activate beta 2 isoform of phos- major G species. Chem. Senses 25:525–531.
pholipase C. Nature 360:686–9. Law, J. S., and Henkin, R. I. (1982). Taste Bud Adenosine-3´-
Kier, L. B. (1972). A molecular theory of sweet taste. J. Pharm. Monophosphate phosphodiesterase: activity, subcellular dis-
Sci. 61:1394–1397. tribution and kinetic parameters. Res. Commun. Chem.
Kinnamon, S. C. (1993). Role of apical ion channels in sour taste. In Pathol. Pharm. 38:439–452.
The Molecular Basis of Smell and Taste Transduction, T. E. Finger Li, X., Reed, D. R., Huque, T., Puchalski, R. B., Tordoff, M. G.,
and W. L. Silver (Eds.). Wiley, Chichester, UK, pp. 201–217. Beauchamp, G. K., Bachmanov, A. A. Abstract 112, 22nd Mtg
Kinnamon, S. C., Dionne, V. E., and Beam, K. G. (1988). Apical Assoc Chemoreception Sci, April 25-30, 2000, Sarasota, FL.
localization of K channels in taste cells provides the basis Li, X., Staszewski, L., Xu, H., Durick, K., Zoller, M., and Adler,
for sour taste transduction. Proc. Natl. Acad. Sci. USA E. (2002). Human receptors for sweet and umani taste. Proc.
85:7023–7027. Natl. Acad. Sci. USA 99:4694–4696.
Kitagawa, M., Kusakabe, Y., Miura, H., Ninomiya, Y., and Hino, Lin, W., and Kinnamon, S. C. (1999). Physiological evidence for
A. (2001). Molecular genetic identification of a candidate ionotropic and metabotropic glutamate receptors in rat taste
receptor gene for sweet taste. Biochem. Biophys. Res. cells. J. Neurophysiol. 82:2061–2069.
Commun. 283:236–242. Lin, W., Finger, T. E., Rossier, B. C., and Kinnamon, S. C.
Kleyman, T. R., and Cragoe E. J. Jr. (1988). Amiloride and its (1999). Epithelial Na channel subunits in rat taste cells:
analogs as tools in the study of ion transport. J. Membr. Biol. localization and regulation by aldosterone. J. Comp. Neurol.
105:1–21.
405:406–420.
Koefoed-Johnsen, V., and Ussing, H. H. (1958). The nature of the
Lin, W., Ogura, T., and Kinnamon, S. C. (2000). Ionic depend-
frog skin potential. Acta Physiol. Scand. 42:298–308.
ence of the proton-activated current in rat vallate taste cells.
Kolesnikov, S. S., and Margolskee, R. F. (1995). A cyclic-
Chem. Senses. 25:685.
nucleotide-suppressible conductance activated by transducin
Lin, W., Ogura, T., and Kinnamon, S. C. (2002). Acid-activated
in taste cells. Nature 376:85–88.
cation currents in rat vallate taste receptor cells. J
Kozasa, T., Itoh, M., Tsukamoto, T., and Kaziro, Y. (1988).
Neurophysiol. 88:133–141.
Isolation and Characterization of the Human Gs  gene. Proc.
Lindemann, B. (1984). Fluctuation analysis of sodium channels
Natl. Acad. Sci. USA 85:2081–2085.
in epithelia. Annu. Rev. Physiol. 46:497–515.
Kretz, O., Bock, R., and Lindemann, B. (1998). Occurrence of
Lindemann, B. (1996). Taste reception. Physiol. Rev. 76:
nitric oxide synthase in taste buds of the rat vallate papilla.
719–766.
Histochem. J. 30:293–299.
Lindemann, B. (1999). Receptor seeks ligand: on the way to
Kretz, O., Barbry, P., Bock, R., and Lindemann, B. (1999).
Differential expression of RNA and protein of the three pore- cloning the molecular receptors for sweet and bitter taste. Nat.
forming subunits of the amiloride-sensitive epithelial sodium Med. 5:381–382.
channel in taste buds of the rat. J. Histochem. Cytochem. Lindemann, B., Gilbertson, T. A., and Kinnamon, S. C. (1999).
47:51–64. Amiloride-sensitive sodium channels in taste. Curr. Top.
Kunishima, N., Shimada, Y., Iougi, Y., Sato, T., Yamamoto, M., Membr. 47:315–336.
Kumasaka, T., Nakanish; S., Jingani, S., and Morikawa, K. Ling, B. N., and Eaton, D. C. (1989). Effects of luminal Na on
(2000). Structural basis of glutamate recognition by a dimeric single Na channels in A6 cells, a regulatory role for protein
metabotropic glutamate receptor. Nature 407:971–977. kinase C. Am. J. Physiol. 256:F1094–103.
Kurihara, K. (1972). Inhibition of cyclic 3´, 5´ nucleotide phos- Liu, L., and Simon, S. A. (1999). Responses of rat taste receptor
phodiesterase in bovine taste papillae by bitter taste stimuli. cells to decreases in acidity. Soc. Neurosci. Abstr. 25:2183.
FEBS Lett. 27:279–281. Liu, L., and Simon, S. A. (2001). Acidic stimuli activates two
Kurihara, K., and Koyama, N. (1972). High activity of adenyl distinct pathways in taste receptor cells from rat fungiform
cyclase in olfactory and gustatory organs. Biophys. Res. papillae. Brain Res. 923:58–70.
Commun. 48:30–34. Lush, I. E. (1981). The genetics of tasting in mice. I. Sucrose octa
Kurihara, K., Kurihara, Y., and Beidler, L. M. (1969). Isolation acetate. Genet. Res. 38:93–95.
and mechanism of taste modifiers; taste modifying protein and Lush, I. E. (1984). The genetics of tasting in Mice. III. Quinine.
gymnemic acids. In Olfaction and Taste, Vol. 3, pp 450–469. Genet. Res. 44:151–160.
C. Pfaffman (Ed.). Rockefeller University Press, New York. Lush, I. E. (1986). The genetics of tasting in mice IV. The
Kusakabe, Y., Yamaguchi, E., Tanemura, K., Kameyama, K., acetates of raffinose, galactose and -lactose. Genet. Res.
Chiba, N., Arai, S., Emori, Y., and Abe, K. (1998). 47:117–123.
Identification of two alpha-subunit species of GTP-binding Lush, I. E., and Holland, G. (1988). The genetics of tasting in
proteins, Galpha15 and Galphaq, expressed in rat taste buds. mice V. Glycine and cycloheximide. Genet. Res. 52:207–212.
Biochim. Biophys. Acta 1403:265–272. Lush, I. E., Hornigold, N., King, P., and Stoye, J. P. (1995). The
Kusakabe, Y. Yasuoka, A., Asano-Miyoshi, M., Iwabuchi, K., genetics of tasting in mice VII. Glycine revisited, and the
Matsumoto, I., Arai, S., Emori, Y., and Abe, K. (2000). chromosomal location of Sac and Soa. Genet Res.
Comprehensive study on G protein -subunits in taste bud 66:167–174.
728 Gilbertson and Margolskee

Maga, J. A. (1983). Flavor potentiators. Crit. Rev. Food Sci. Nutr. Miyamoto, T., Fujiyama, R., Okada, Y., and Sato, T. (1998a).
18:231–312. Salty and sour transduction. Multiple mechanisms and strain
Matsunami, H., Montmayeur, J. -P., and Buck, L. B. (2000). A differences. Ann. NY Acad. Sci. 855:128–133.
family of candidate taste receptors in human and mouse. Miyamoto, T., Fujiyama, R., Okada, Y., and Sato, T. (1998b).
Nature 404:601–604. Sour transduction involves activation of NPPB-sensitive con-
Matsuoka, I., Mori, T., Aoki, J., Sato, T., and Kurihara, K. (1993). ductance in mouse taste cells. J. Neurophysiol.
Identification of novel members of G-protein coupled recep- 80:1852–1859.
tor superfamily expressed in bovine taste tissue. Biochem. Miyamoto, T., Fujiyama, R., Okada, Y., and Sato, T. (1999).
Biophys. Res. Commun. 194:504–511. Strain differences in amiloride-sensitivity of salt-induced
Max, M.,Shanker, Y. G., Huang, L., Kong, M., Liu, Z., responses in mouse non-dissociated taste cells. Neurosci. Lett.
Campagne, F., Weinstein, H., Damak, S., and Margolskee, R. 277:13–16.
F., (2001). Tas1r3, encoding a new candidate taste recepter, is Miyoshi, M. A., Abe, K., and Emori, Y. (2001). IP3 receptor type
allelic to the sweet responsiveness locus Sac. Nat. Genet. 3 and PlCbeta2 are co-expresses with taste receptors T1R and
28:58–63. T2R in rat taste bud cells. Chem. Senses 26:259–265.
McLaughlin, S. K., McKinnon, P. J., and Margolskee, R. F. Monastryrskaia, K., Lundstrom, K., Plahl, D., Acuna, G.,
(1992a). Gustducin is a taste-cell-specific G protein closely Schweitzer, C., Malherbe, P., and Mutel, V. (1999). Effect of
related to the transducins. Nature 357:563–569. the umami peptides on the ligand binding and function of rat
McLaughlin, S. K., McKinnon, P. J., and Margolskee, R. F. mGlu4a receptor might implicate this receptor in the
(1992b).  Gustducin: a taste cell specific G protein subunit monosodium glutamate taste transduction. Br. J. Pharmacol.
closely related to the  transducins. In Chemical Signals in 128:1027–1034.
Vertebrates VI, R. L. Doty and D. Müller-Schwarze (Eds.). Montmayeur, J. P., Liberles, S. D., Matsunami, H. and Buck, L.
Plenum Press, New York, pp. 9–14. B. (2001). A candidate taste receptor gene near a sweet taste
McLaughlin, S. K., McKinnon, P., Spickofsky, N., Danho, W., locus. Nat. Neurosci. 4:492–498.
and Margolskee, R. F. (1994). Molecular cloning of G pro-
Naim, M., Ronen, T., Striem, B. J., Levenson, M., and Zehavi, U.
teins and phosphodiesterases from rat taste cells. Physiol.
(1991). Adenylate cyclase responses to sucrose stimulation in
Behav. 56:1157–1164.
membranes of pig circumvallate taste papillae. Comp.
Mierson, S., DeSimone, S. K., Heck, G. L., and DeSimone, J. A.
Biochem. Physiol. 100B:455–458.
(1988). Sugar-activated ion transport in canine lingual epithe-
Naim, M., Seifert, R., Nürnberg, B., Gruünbaum, L., and Shultz,
lium. J. Gen. Physiol. 92:87–111.
G. (1994). Some taste substances are direct activators of
Miller, I. J., Jr., and Bartoshuk, L. M. (1991). Taste perception, taste
G-proteins. Biochem. J. 297:451–459.
bud distribution, and spatial relationships. In Smell and Taste in
Nelson, G., Hoon, M. A., Chandrashekar, J., Zhang, Y., Ryba, N.
Health and Disease, T. V. Getchell, R. L. Doty, L. M. Bartoshuk,
J. P. and Zuker, C. S. (2001). Mammalian sweet taste recep-
J. B. Snow, Jr. (Eds.). Raven Press, New York, pp. 205–233.
tors. Cell 106:381–390.
Ming, D., Ruiz-Avila, L., Margolskee, R. F. (1998).
Characterization and solubilization of bitter-responsive recep- Nelson, G., Chandrashekar, J., Hoon, M. A., Feng, L., Zhao, G.,
tors that couple to gustducin. Proc. Natl. Acad. Sci. USA Ryba, N. J., and Zuker, C. S. (2002). An amino-acid taste
95:8933–8938. receptor. Nature 416:199–202.
Ming, D., Ninomiya, Y., and Margolskee, R. F. (1999). Blocking Ninomiya, Y., Sako, N., and Funakoshi, M. (1989). Strain differ-
taste receptor activation of gustducin inhibits gustatory ences in amiloride inhibition of NaCl responses in mice, Mus
responses to bitter compounds. Proc. Natl. Acad. Sci. USA musculus. J. Comp. Physiol. A 166:1–5.
96:9903–9908. Ninomiya, Y., Kajiura, H., Naito, Y., Mochizuki, K., Katsukawa,
Misaka, T., Kusakabe, Y., Emori, Y., Gonoi, T., Arai, S., and Abe, H., and Torii, K. (1994). Glossopharyngeal denervation alters
K. (1997). Taste buds have a cyclic nucleotide-activated chan- responses to nutrients and toxic substances. Physiol. Behav.
nel, CNGgust. J. Biol. Chem. 272:22623–22429. 56:1179–1184.
Misaka, T., Kusakabe, Y., Emori, Y., Arai, S., and Abe, K. (1998). Nomura, H. (1978). Histochemical Localization of adenylate
Molecular cloning and taste bud-specific expression of a cyclase and phosphodiesterase activities in the foliate papillae
novel cyclic nucleotide-gated channel. Ann. NY Acad. Sci. of the rabbit. I. Light microscopic observations. Chem. Senses
855:150–159. Flav. 3:319–324.
Misaka, T., Ishimaru, Y., Iwabuchi, K., Kusakabe, Y., Arai, S., Ogura, T., and Kinnamon, S. C., (1999). IP(3)-Independent
Emori, Y., and Abe, K. (1999). A gustatory cyclic nucleotide- release of Ca(2) from intracellular stores: A novel mechan-
gated channels CNGgust, is expressed in the retina. ism for transduction of bitter stimuli. J. Neurophysiol.
Neuroreport 10:743–746. 82:2657–2666.
Miyamoto, T., Okada, Y., and Sato, T. (1988). Ionic basis of Ogura, T., Mackay-Sim, A., and Kinnamon, S. C. (1997). Bitter
receptor potential of frog taste cells induced by acid stimuli. taste transduction of denatonium in the mudpuppy Necturus
J. Physiol. (Lond.) 405:699–711. maculosus. J. Neurosci. 17:3580–3587.
Molecular Physiology of Gustatory Transduction 729

Okada, Y., Miyamoto, T., and Sato, T. (1987). Cation dependence Schiffman, S. S., Cohn, H., and Lindley, M. G. (1981). Multiple
of frog gustatory neural responses to acid stimuli. Comp. receptor sites mediate sweetness: evidence from cross-adapta-
Biochem. Physiol. A 88:487–490. tion. Pharm. Biochem. Behav. 15:377–388.
Okada, Y., Miyamoto, T., and Sato, T. (1988). Ionic mechanism Schiffman, S. S., Lockhead, E., and Maes, F. W. (1983).
of generation of receptor potential in response to quinine in Amiloride reduces the taste intensity of Na and Li salts and
frog taste cell. Brain Res. 450:295–302. sweeteners. Proc. Natl. Acad. Sci. USA 80:6136–6140.
Okada, Y., Miyamoto, T., and Sato, T. (1990). Aldosterone Schiffman, S. S., Diaz, C., and Beeker, T. G. (1986). Caffeine
increases gustatory neural response to NaCl in frog. Comp. intensifies task of certain sweeteners: role of adenosine recep-
Biochem. Physiol. A 97:535–536. tor. Pharmacol. Biochem. Behav. 24:429–432.
Okada, Y., Miyamoto, T., and Sato, T. (1994). Activation of a Settle, R. G., Meehan, K., Williams, G. R., Doty, R. L., and
cation conductance by acetic acid in taste cells isolated from Sisley, A. C. (1986). Chemosensory properties of sour tas-
the bullfrog. J. Exp. Biol. 187:19–32. tants. Physiol. Behav. 36:619–623.
Orola, C. N. Yamashita, T., Harada, N., Amano, H., Ohtani, M., Shallenberger, R. S. (1978). Bev, Food Sci. Proc. Int. Congr. Food
and Kumazawa, T. (1992). Intracellular free calcium concen- Sci. Technol., pp. 360–366.
trations in single taste receptor cells in the guinea pig. Acta Shallenberger, R. S., and Acree, T. E. (1967). Molecular theory of
Otolaryngol. 112:120–127. sweet taste. Nature 216:480–482.
Price, S. (1973). Phosphodiesterase in tongue epithelium: activ- Shallenberger, R. S., Acree, T. E., and Lee, C. Y. (1969). Sweet
ation by bitter taste stimuli. Nature 241:54–55. taste of D- and L- sugars and amino acids and the steric nature
Reed, D. R., Nanthakumar, E., North, M., Bell, C., Bartoshuk, L. of their chemoreceptor site. Nature 221:555–556.
M., and Price, R. A. (1999). Localization of a gene for bitter- Shimazaki, K., Sato, M., and Nakao, M. (1986). Photoaffinity
taste perception to human chromosome 5p15. J. Hum. Genet. labeling of thaumatin-binding protein in monkey circumval-
64:1478–1480. late papillae. Biochim. Biophys. Acta 884:291–298.
Rehnberg, B. G., Hettinger, T. P., and Frank, M. E. (1992). Simon, S. A., Labarca, P., and Robb, R. (1989). Activation by
Salivary ions and neural taste responses in the hamster. Chem. saccharides of a cation-selective pathway on canine lingual
Senses 17:179–190. epithelium. Am. J. Physiol. 256:R394–402.
Rosenzweig, S., Yan, W., Dasso, M., and Spielman, A. I. (1999). Smith, D. V., and Margolis, F. L. (1999). Taste processing: whet-
Possible novel mechanism for bitter taste mediated through ting our appetites. Curr. Biol. 9:R453–R455.
cGMP. J. Neurophysiol. 81:1661–1665. Spielman, A. I. (1998). Gustducin and its role in taste. J. Dent.
Rossler, P., Kroner, C., Freitag, J., Noe, J., and Breer, H. (1998). Res. 77:539–544.
Identification of a phospholipase C beta subtype in rat taste Spielman, A. I., Mody, I., Brand, J. G., Whitney, G., MacDonald,
cells. Eur. J. Cell Biol. 77:253–261. J. F., and Salter, M. W. (1989). A method for isolating and
Rossler, P., Boekhoff, I., Tareilus, E., Beck, S., Breer, H., and patch-clamping single mammalian taste receptor cells. Brain
Freitag, J. (2000). G protein betagamma complexes in cir- Res. 503:326–329.
cumvallate taste cells involved in bitter transduction. Chem. Spielman, A. I., Huque, T., Nagai, H., Whitney, G., and Brand, J.
Senses 25:413–21. (1994). Generation of inositol phosphates in bitter taste trans-
Ruiz-Avila, L., McLaughlin, S. K., Wildman, D., McKinnon, P. duction. Physiol. Behav. 56:1149–1155.
J., Robichon, A., Spickofsky, N., and Margolskee, R. F. Stevens, D. R., Seifert, R., Bufe, B., Muller, F., Kremmer, E.,
(1995). Coupling of bitter receptor to phosphodiesterase Gauss, R., Meyerhoff, W., Kaupp, U. B., and Lindemann, B.
through transducin in taste receptor cells. Nature 376: (2001). Hyperpolarization-activated channels HCN1 and
80–85. HCN4 mediate responses to sour stimuli. Nature 413:631–635.
Ruiz-Avila, L., Wong, G. T., and Margolskee, R. F. (2001). Stewart, R. E., DeSimone, J. A., and Hill, D. L. (1997). New per-
Dominant loss of responsiveness to sweet and bitter com- spectives in gustatory physiology: transduction, development
pounds caused by a single mutation in gustducin. Proc. Natl. and plasticity. Am. J. Physiol. 272:C1–C26.
Acad. Sci. USA 98:8868–8873. Stewart, R. E., Lyall, V., Feldman, G. M., Heck, G. L., and
Sainz, E., Korley, J. N., Bottey, J. A., and Sullivan, S. L., (2001). DeSimone, J. A. (1998). Acid-induced responses in hamster
Identification of a novel member of the T1R family of puta- chorda tympani and intracellular pH tracking by taste recep-
tive taste receptors. J Neurochem. 77:896–903 tor cells. Am. J. Physiol. 275:C227–C238.
Saroli, A. (1984). Structure-activity relationship of a bitter com- Stocklin, W. (1968). Gymnestrogenin, ein neues
pound: denatonium chloride. Naturwissenschaften 71: 428–429. Pentahydroxytriterpen aus den blattern von Gymnema
Sato, T., and Beidler, L. M. (1982). The response characteristics sylyestre RBr. Helv. Chim. Acta 51:1235–1242.
of rat taste cells to four basic taste stimuli. Comp. Biochem. Striem, B. J., Pace, U., Zehavi, U., Naim, M., and Lancet, D. (1989).
Physiol. Acta 73:1–10. Sweet tastants stimulate adenylate cyclase coupled to GTP bind-
Saul, L. B., Kennedy, L. M., and Stevens, D. A. (1985). Selective ing protein in rat tongue membranes. Biochem. J. 260:121–126.
suppression of sweetness-suppressing principle from Hovenia Striem, B. J., Yamamoto, T., Naim, M., Lancet, D. Jakinovich,
dulcis leaves. In Proc. ISOT IX, Snowmass, Village, CO, p. 56. W., and Zehavi, U. (1990). The sweet taste inhibitor methyl
730 Gilbertson and Margolskee

4,5-dichloro-4,6 dideoxy--D-galactopyranside inhibits Varkevisser, B., and Kinnamon, S. C. (2000). Sweet taste trans-
sucrose stimulation of the chorda tympani nerve and of the duction in hamster: role of protein kinases. J. Neurophysiol.
adenylate cyclase in anterior lingual membranes of rats. 83:2526–2532.
Chem. Senses 15:529–536. Waldmann, R., Champigny, G., Bassilana, F., Heurteaux, C., and
Striem, B. J., Naim, M., and Lindemann, B. (1991). Generation Lazdunski, M. (1997). A proton-gated channel involved in
of cyclic AMP in taste buds of the rat circumvallate papillae acid-sensing. Nature 386:173–177.
in response to sucrose. Cell Physiol. Biochem. 1:46–54. Welch, J. M., Simon, S. A., and Reinhart, P. H. (2000). The activ-
Sugimoto, K. (1997). Abstract 120, 12th Int Symp Olfaction and ation mechanism of rat vallinoid receptor 1 by capsaicin
Taste, July 7–12, San Diego, CA, 17–20. involves the pore domain and differs from the activation by
Swaroop, A., Agarwal, N., Gruen, J. R., Bick, D., and Weissman, either acid or heat. Proc. Natl. Acad. Sci. USA 97:
S. M. (1991). Differential expression of novel Gs  signal 13889–13894.
transduction protein cDNA species. Nucl. Acids Res. Whitney, G., and Harder D. B. (1986). Single-locus control of
19:4725–4729. sucrose octa acetate tasting among mice. Behav. Gen.
Tinti, J. M., and Nofre, C. (1991). Why does a sweetener taste 16:559–574.
sweet? A new model in sweeteners: discovery, molecular design, Wong, G. T., Gannon, K. S., and Margolskee, R. F. (1996).
chemoreception. American Chemical Symposium Series 450, Transduction of bitter and sweet taste by gustducin. Nature
D. E. Walters, F. T. Orthoefer, and G. E. DuBois (Eds.). 381:796–800.
American Chemical Society, Washington DC, pp. 206–213. Yamada, H., Imoto, I., and Yoshioka, S. (1985). Suppression of
Tonosaki, K., and Funakoshi, M. (1984). The mouse taste cell sweet sensitivity by extracts of Ziziphus jujuba leaves in the
response to five sugar stimuli. Comp. Biochem. Biophysiol. rat and hamster. Chem. Senses 10:445.
79A:625–630. Yamaguchi, S. (1991). Basic properties of umami and effects on
Tonosaki, K., and Funakoshi, M. (1988a). Cyclic nucleotides humans. Physiol. Behav. 49:833–841.
may mediate taste transduction. Nature 331:354–356. Yamamoto, T., Matsuo, R., Fujimoto, Y., Fukunaga, I., Miyasaka,
Tonosaki, K., and Funakoshi, M. (1988b). Voltage- and current- A., and Imoto, T., (1991). Electrophysiological and behavioral
clamp recordings of the receptor potential in mouse taste cell. studies on the taste of umami substances in the rat. Physiol.
Brain Res. 445:363–366. Behav. 49:919–925.
Tsunenari, T., Kurahashi, T., and Kaneko, A. (1999). Activation Yamashita, H., Theerasilp, S., Aiuchi, T., Nakaya, K.,
by bitter substances of a cationic channel in membrane Nakamura, Y., and Kurihara, Y. (1990). Purification and com-
patches excised from the bullfrog taste receptor cell. plete amino acid sequence of a new type of sweet protein
J. Physiol. 519:397–404. with taste-modifying activity, curculin. J. Biol. Chem.
Uchida, Y., and Sato, T. (1997a). Intracellular calcium increase in 26:15770–15775.
gerbil taste cell by amino acid sweeteners. Chem. Senses Yan, W., Sunavala, G., Rosenzweig, S., Dasso, M., Brand,
22:83–91. J. G., and Spielman, A. I. (2001). Bitter taste transduced by
Uchida, Y., and Sato, T. (1997b). Changes in outward K currents PLCbeta2-dependent rise in IP3 and alpha-gustducin-
in response to two types of sweeteners in sweet taste trans- dependent fall in cyclic nucleotides. Am. J. Physiol.
duction of gerbil taste cells. Chem. Senses 22:163–169. 280:C742–C751.
Ugawa, S., Minami, Y., Guo, W., Saishin, Y., Takatsuji, K., Ye, Q., Heck, G. L., and DeSimone, J. A. (1991). The anion para-
Yamamoto, T., Tohyama, M., and Shimada, S. (1998). dox in sodium taste reception: resolution by voltage-clamp
Receptor that leaves a sour taste in the mouth. Nature 395: studies. Science 254:724–726.
555–556. Zhao, A. Z., Yan, C., Sonnenburg, W. K., and Beavo, J. A.
van der Wel, H., and Loeve, K. (1972). Isolation and character- (1997). Recent advances in the study of Ca 2/CaM-
ization of thaumatin I and II, the sweet tasting proteins from activated phosphodiesterases: expression and physiological
Thaumatococcus danieuii Benth. Eur. J. Biochem. functions. Adv. Second Messenger Phosphoprotein Res.
31:221–225. 31:237–251.
35

Gustatory Neural Coding

David V. Smith
University of Tennessee Health Science Center, Memphis, Tennessee, U.S.A.

Thomas R. Scott
San Diego State University, San Diego, California, U.S.A.

I. INTRODUCTION II. TASTE-GUIDED BEHAVIOR AND

INFORMATION EXTRACTION
By providing sensory information important for
A. The Role of Taste in Behavior:
decisions about ingestion and rejection, the sense of taste
Ingestive Decisions
serves as the gateway to the body’s internal milieu.
Gustatory neurobiologists generally agree on the
Before examining how information about taste stimuli is
existence of at least four basic taste qualities—sweet,
coded in the nervous system, it is instructive to consider
salty, sour, and bitter—although they disagree about the
the functions that taste serves in the life of an organism.
possible existence of others (e.g., the taste of amino
The gustatory system functions as the gatekeeper of the
acids). These qualities and the behaviors they elicit help
internal milieu, acting to guide ingestive and avoidance
to ensure the animal’s energy supply (carbohydrates,
behaviors. Because the anatomical organization of the cen-
amino acids), maintain the proper electrolyte and pH
tral taste pathway parallels that of other visceral sensory
balance (salts, acids), and avoid the ingestion of toxins
input, and because taste stimulation triggers autonomic as
(acids, alkaloids). Of considerable debate is the way in
well as perceptual responses, it is useful to consider taste
which taste quality is represented in the activity of gusta-
as a component of the visceral afferent system (Norgren,
tory neurons. Many investigators accept the idea that
1985). Taste serves both a discriminative and an affective
peripheral gustatory nerve fibers and central neurons can
function, allowing animals to make distinctions among
be classified into groups on the basis of similarities and
potential foods and providing information that guides deci-
differences in their response profiles and other biological
sions about the ingestion or rejection of chemical stimuli
properties. Taste neurons have typically been character-
(Scott, 1987b).
ized by analyses that employ a neuron classification
scheme in an attempt to impose some order on these data.
1. Perception Versus Visceral Function
However, the breadth of tuning and multimodal sensitiv-
ity of central gustatory neurons make it difficult to accept As humans, it is natural for us to think about the gustatory
a strict labeled-line code for taste quality, in which each system in terms of our own experience. Consequently,
neuron type signals a particular quality. Rather, it is research on the taste system has been directed primarily
likely that the various neuron types play a critical role in toward understanding the neural basis of the human sensa-
defining unique across-neuron patterns, which can tions of sweet, salty, sour, and bitter. These qualities are
unambiguously represent gustatory quality. associated with stimuli that are important for animals to
731
732 Smith and Scott

recognize as nutrients or toxins (Carpenter, 1956; acids, and sodium salts. These various experiments
Pfaffmann, 1964). Certainly these perceptual categories demonstrate that much of the neural distinction among
are important to our investigation of the taste system, but stimuli of different qualities, as evidenced by specific pat-
input from gustatory receptors also directly controls an terns of motor output, can occur within the hindbrain and
array of reflexive motor responses involved in food does not require the contribution of more rostral levels of
ingestion and a range of autonomic responses important the gustatory system.
for digestive and other homeostatic processes, such as
electrolyte balance. Salivary, gastric, and pancreatic secre-
3. Generalization and Discrimination
tions and cardiovascular and thermoregulatory responses
resulting from gustatory stimulation are referred to as In order to make decisions about ingestion or rejection,
cephalic phase reflexes, which prepare the body for the the ability to discriminate among different taste stimuli
arrival and utilization of nutrients (Friedman and Mattes, is essential. Behavioral studies have employed operant
1991). Because gustatory stimulation produces a variety of conditioning techniques or conditioned taste aversion
responses beyond taste perception, understanding the func- generalization to determine the discriminative capacity
tional neural organization of the gustatory system requires of the gustatory system of experimental animals. A taste
consideration of the many roles that taste plays in the phys- aversion can be conditioned to a particular tastant by
iology of the organism, from perception to autonomic pairing that stimulus with gastrointestinal malaise pro-
function. duced by drugs or radiation (Garcia et al., 1970).
Following such learning, the degree to which the aver-
sion generalizes to other stimuli serves as an indication
2. Ingestion and Rejection: Taste Reactivity
of the gustatory similarity between the conditioned stim-
Grill and Norgren (1978b) first described the reflexive ulus and any others (Nachman, 1963; Smith et al., 1979;
behaviors in the rat resulting from gustatory stimulation. Nowlis et al., 1980; Nowlis and Frank, 1981). The
They termed these ingestive and aversive response results of these kinds of studies show that hamsters and
sequences, which largely reflect stimulus palatability, rats generalize a learned sucrose aversion to several
“taste reactivity.” Sucrose and other preferred stimuli other stimuli described as sweet by humans, including
evoke a pattern of ingestive responses that begins fructose, glucose, and sodium-saccharin (Smith et al.,
with rhythmic mouth movements followed by midline 1979; Nowlis et al., 1980). These techniques also
tongue protrusions, lateral tongue flicks, and swallowing. demonstrate that a number of sodium and lithium salts
Aversive stimuli, such as quinine, produce a response are considered similar by rodents and distinct from non-
sequence that includes oral gapes and a number of somatic sodium salts such as KCl or NH4Cl, which are behav-
motor responses associated with rejection, such as chin iorally similar to acids. Finally, bitter-tasting substances,
rubbing, paw flailing, head shaking, and face washing. such as quinine hydrochloride (QHCl) and MgSO4, show
These response sequences occur in decerebrate rats (Grill cross generalization in these experiments. To a consider-
and Norgren, 1978c), suggesting that the neural substrate able extent, such generalization studies produce group-
for these reflexive behaviors is contained within the brain- ings of gustatory stimuli that are similar to those
stem. In intact rats, but not in those that are decerebrate, produced by analyzing the responses of peripheral or
these responses are modifiable by learning, so that a nor- central gustatory neurons (see Fig. 7 below) or the pat-
mally ingestive sequence becomes an aversive one after a terns of taste reactivity (Fig. 1).
conditioned taste aversion (Grill and Norgren, 1978a). In Discriminative operant conditioning has been used to
general, the results from taste reactivity tests correspond examine both generalization and discrimination among
with other short-term palatability measures such as lick- gustatory stimuli and the role of different cranial nerve
rate tests (Schwartz and Grill, 1984). The patterns of taste inputs in taste-guided behavior. Using shock avoidance,
reactivity responses are quite distinct for stimuli of Erickson (1963) demonstrated that rats conditioned to
different taste qualities (Brining et al., 1991). Closely avoid KCl would also avoid NH4Cl but not NaCl. A simi-
related stimuli produce strikingly similar patterns of taste lar pattern of generalization occurred with an appetitive
reactivity, as shown in Figure 1. Patterns of taste reactivity operant task, in which rats were trained to respond on one
to the four basic stimuli are shown in Figure 1A, and the of two levers, depending upon the taste of a previously
similarities among patterns for 12 taste stimuli are conditioned stimulus (Morrison, 1967). Using similar
depicted in the factor space of Figure 1B. Based on corre- operant techniques, Spector and colleagues have exam-
lations among the patterns of taste reactivity, these stimuli ined the role of the chorda tympani (CT) and glossopha-
fell into four groups: sugars, bitter-tasting substances, ryngeal (IXth) nerves in taste discrimination performance
Gustatory Neural Coding 733

Figure 1 Taste reactivity in the hamster. (A) Profiles of taste reactivity responses to the four basic stimuli, depicting the percentage of
animals showing each of eight different responses. Abbreviations: LTP, lateral tongue protrusions; TP, tongue protrusions; LO, locomo-
tion; CR, chin rubbing; FF, forelimb flailing; AP, aversive posturing; FR, fluid rejection, and G, gapes. (B) Two-dimensional representa-
tion of the correlations of hamster taste reactivity profiles to 12 different stimuli with two common factors (I and II). Each point in the
space represents the correlation of one stimulus profile with two factors. Each stimulus profile consisted of eight different measures of
taste reactivity, ranging from lateral tongue protrusions to oral gapes (see A). Zero is at the origin, where the axes cross, and  0.50 is
marked on each axis by crossing line segments. A value of  1.0 is indicated by the intersections of the circle with the axes. A point
would fall on the circumference of the circle if 100% of the variance of its profile were accounted for by these two factors. Stimuli with
similar taste quality (e.g., the sugars or the bitter-tasting stimuli) produce similar patterns of taste reactivity, as indicated by their prox-
imity within this factor space. The first factor (I) separated the highly preferred sugars from the highly aversive bitter-tasting substances,
which showed opposite patterns of taste reactivity (see A). (Data from Brining et al., 1991.)
734 Smith and Scott

in the rat. Transection of the CT nerve but not the IXth defining feature of this sense. Most researchers agree that
nerve severely impairs the ability of rats to discriminate human gustatory experience includes the sweet, salty, sour,
between NaCl and KCl (Spector and Grill, 1992; St. John and bitter qualities (McBurney and Gent, 1979). However,
et al., 1997). These and other data suggest that there has been considerable debate over whether some
information arriving via the CT nerve is important for other qualities (e.g., umami—the taste of monosodium
discriminating NaCl from nonsodium salts. Treatment of glutamate) should be included or whether any qualities
the tongue with amiloride, which blocks the response in warrant being designated as “basic” (McBurney, 1974;
NaCl-best fibers of the CT nerve to NaCl, completely Erickson, 1977; Yamaguchi, 1979; Schiffman and
eliminates the ability of rats to make a NaCl/KCl discrim- Erickson, 1980). Intensity is a dimension common to all
ination (Spector et al., 1996). In addition, when rats are sensory systems, reflecting the magnitude of the evoked
conditioned to avoid NaCl in the presence of amiloride, sensation. In the gustatory system, increasing stimulus
they generalize that aversion to KCl (Hill et al., 1990). concentration generally leads to greater action potential
Since rats do not have an amiloride-sensitive response to frequency in the response of single neurons, as shown in
NaCl in the IXth nerve (Formaker and Hill, 1991) nor the responses of four neurons in the nucleus of the solitary
even any N-best IXth nerve fibers (Frank, 1991), it is tract (NST) of the hamster in Figure 2. The responses of
undoubtedly the comparison between N-best and other individual neurons are not always monotonic across stimu-
nerve fibers in the CT and perhaps the greater superficial lus concentration, which results in some variation in each
petrosal (GSP) nerve, serving the palate, that allows this cell’s response profile. Nevertheless, increases in impulse
discrimination. Similarly, VIIth nerve transection virtually frequency and the recruitment of additional cells at higher
eliminates the ability of rats to discriminate quinine from concentrations are the mechanisms by which gustatory
KCl, whereas IXth nerve damage has no effect (St. John neurons likely code information about stimulus intensity.
and Spector, 1998). This latter result was surprising Hedonic value, the perceived pleasantness or unpleas-
because the IXth nerve is highly responsive to quinine and antness of a taste sensation, is a response based on
its individual fibers respond differentially to quinine and genetic, physiological, and experiential factors as well as
KCl (Frank, 1991). These data imply that the VIIth nerve the characteristics of the stimulus. Taste is an inherently
is more important for the discrimination of taste stimuli, hedonic sense, with close ties to motivated behavior
whereas the IXth nerve serves a protective function (St. (Pfaffmann, 1964). Although these three dimensions of
John and Spector, 1998). For example, the IXth nerve taste information (quality, intensity, and hedonic value)
clearly contributes to the avoidance of quinine and other can be assessed separately, they are not independent. For
stimuli (Spector and St John, 1998; Markison et al., 1999) example, the perceived qualities of some taste stimuli
and is important in the elicitation of aversive taste reactiv- (e.g., Na-saccharin) change with concentration. Similarly,
ity (Travers et al., 1987). the hedonic value of a stimulus is largely determined by
its quality and intensity. Many omnivorous mammals
share a concentration-dependent preference for sub-
B. What Information Is Coded by
stances that humans describe as tasting sweet or salty and
Gustatory Neurons?
an avoidance of substances humans term sour or bitter
(Richter and Campbell, 1940; Carpenter, 1956;
Most neurophysiological studies of the gustatory system
Pfaffmann, 1964). Presumably, these predispositions
have focused on the neural representation of gustatory
reflect evolutionary pressures related to the ingestional
quality. Single peripheral gustatory fibers and central
consequences (i.e., nutritional or toxic) of potential foods.
neurons are broadly sensitive across taste qualities and
Indeed, omnivores inherit a taste system that permits
stimulus concentrations and often respond to tactile and
accurate evaluation of stimulus nutritive value or toxicity
thermal stimuli as well. Understanding the functional
across a wide range of chemicals with diverse physical
role of a taste-responsive neuron requires an appreci-
characteristics (Scott and Mark, 1987; Scott and Giza,
ation of the kinds of information that must be coded by
2000). The neural code that underlies this behavior con-
these cells.
sists of afferent activity that parcels chemicals out accord-
ing to the degree to which they offer sources of nutrition
1. Quality, Intensity, and Hedonic Tone:
(carbohydrates, proteins) or disrupt physiological activity
A Nutritional Dimension
(extremes of pH, alkaloids, etc.). In retrospect, this is a
The gustatory system codes at least three types of inform- reasonable organization for the taste system, which was
ation about chemical stimuli: quality, intensity, and shaped by evolutionary pressure to select nutrients from
hedonic value. The unique perception of taste quality is the among the wide array of useless and toxic chemicals in
Gustatory Neural Coding 735

Figure 2 Concentration-response functions for four hamster PbN neurons. (A) Responses (impulses/s) to each stimulus at five concen-
tration levels. Zero is the level of spontaneous activity and the dashed line indicates 1 SD above spontaneous rate. (B) Response profiles
of each cell at three concentration levels (lowest, middle, and highest of each series). The excitatory breadth of tuning (H) at each con-
centration level is indicated above each profile. Both the breadth of tuning and the most effective stimulus varied with concentration.
(From Van Buskirk and Smith, 1981.)

the environment. Selection among foragers presumably configure the system to match the available chemicals with
favored those with a taste system that activated the appro- the animal’s nutritional needs. However, there are also
priate hedonic tone to match the consequences of ingest- mechanisms allowing experience and physiological factors
ion: attraction to nutrients and revulsion by toxins (Scott to influence gustatory neural processing and perception.
and Mark, 1987). Species-specific predispositions that impart hedonic value
This organization of the gustatory system is largely to a stimulus can be overcome via learned preferences or
determined by genetic and developmental factors that aversions (London et al., 1979; Grill, 1985; Bertino et al.,
736 Smith and Scott

1986; Rameriz, 1991) or in response to metabolic or include considerations of the numerous anatomical
pharmacological manipulations (Richter, 1956; Parker et interconnectons between visceral afferent and efferent
al., 1992). Responses to gustatory stimulation recorded systems and the brainstem gustatory relay nuclei.
from brainstem cells in rodents are subject to several mod- Gustation is represented in the rostral pole of a functional
ulatory influences, including the effects of gastric disten- column that also receives input related to respiration,
sion (Glenn and Erickson, 1976), blood glucose and blood pressure, blood pH, gastrointestional activity, and
insulin levels (Giza and Scott, 1983, 1987), intraduodenal pain. Anatomically, this functional column corresponds to
lipids (Hajnal et al., 1999), degree of sodium deprivation the NST, with the gustatory zone located in the rostral half
(Jacobs et al., 1988; McCaughey and Scott, 2000) and con- of the nucleus. This visceral afferent column is closely
ditioned taste aversion learning (Chang and Scott, 1984). associated with both somatic motor nuclei (i.e., the retro-
That is, the responses of rodent brainstem cells are modu- facial area, motor nucleus of V, nucleus ambiguus, and
lated not only by stimulus quality, intensity, and modality, hypoglossal nucleus) and visceromotor (salivatory and
but also by the animal’s physiological state and prior expe- vagal preganglionic parasympathetic neurons) nuclei.
rience. In primates, the responses of taste neurons in the Collectively, these cell columns contain motor neurons
orbitofrontal cortex (Rolls et al., 1988), but not in the NST that initiate chewing, tongue movement, salivation, swal-
(Yaxley et al., 1985) or primary taste cortex (Rolls et al., lowing, gastrointestional motility, acid secretion, and
1988), are modulated by the animal’s state of satiety. other reflexes associated with eating, drinking,
Although the circuitry underlying such modulatory effects and painful stimulation. The gustatory and other general
on taste processing is not well understood, in rodents there visceral afferent systems provide the requisite sensory
are descending connections to brainstem taste nuclei from information to control these assorted visceromotor
all of the forebrain areas that receive gustatory input, reflexes. Some reflexes are clearly involuntary, such as
including the cortex, amygdala, and lateral hypothal- salivation or baro- and chemoreceptor regulation of car-
amus. Descending influences on brainstem taste responses diac and respiratory rates. Other actions have a voluntary
can be either excitatory or inhibitory (Hayama et al., or somatic motor component that is generally considered
1985; DiLorenzo, 1988; Mark et al., 1988; DiLorenzo, visceromotor in function, such as the stereotypic tongue
1990; DiLorenzo and Monroe, 1995). Neurophysiological and swallowing movements that occur during ingestion.
studies employing both in vitro (King et al., 1993; Liu Activation of the gustatory system is necessary for an
et al., 1993; Wang and Bradley, 1993, 1995; Bradley et al., organism to assess the palatability of food or drink. Taste
1996) and in vivo (Smith et al., 1994, 1998; Davis and input generates activity in the ascending gustatory pathway
Smith, 1997; Li and Smith, 1997; Smith and Li, 1998, and presumably evokes a perceptual component that
2000) preparations have implicated a number of involves an awareness of taste quality. Simultaneously,
neurotransmitters and peptides in the modulation of NST taste stimulation activates the afferent limb of a variety of
neuronal activity. somatic and visceromotor acceptance or rejection reflexes
Thus gustatory afferent input provides at least three associated with eating and drinking (Grill and Norgren,
types of information, interrelated in complex ways. This 1978b). Electrophysiological studies have shown that
input guides the organism to make appropriate ingestive responses to sucrose and quinine generally (but not
decisions, directed by physiological and nutritional needs, always) occur in separate subsets of cells, suggesting par-
which provide modulatory feedback over brainstem gust- allel systems that control the ingestion of nutrients and the
atory neurons. How taste intensity, quality, and hedonic rejection of toxic substances (Travers and Smith, 1979;
value are represented in the nervous system is the problem Smith et al., 1994). The relatively high sensitivity of the
of gustatory neural coding. CT and GSP nerves to sugars could selectively activate
a set of second-order neurons in the gustatory NST that
ultimately drive brainstem circuits that initiate tongue
2. Taste and Visceromotor Integration
protrusions and other ingestive responses. In contrast, the
The processing of gustatory information is often IXth nerve shows a relatively greater response to
discussed solely in terms of sensory coding and taste bitter-tasting stimuli like quinine and could activate a
discrimination. However, it is becoming more difficult to separate population of gustatory NST neurons that project
discuss the gustatory system outside the context of to brainstem regions that initiate a sequence of protective
visceromotor integration. Norgren and colleagues responses. These are extreme examples and involve
(Norgren, 1985, 1995; Travers, 1993) have provided specific sensory inputs that generate opposite stereotypic
compelling arguments that a more contemporary view of behaviors (as indicated in Fig. 1). In the case of most NST
gustation and the autonomic nervous system should neurons, which have multiple sensitivities to a wide range
Gustatory Neural Coding 737

of sapid stimuli (as in Fig. 2), the scenario is more complex 1. Multiple Sensitivity of Rat Taste Receptor Cells
because the output of such neurons would have to activate
Intracellular recording experiments have suggested for
a mixture of appropriate motor repertoires reflecting the
several years that taste cells are broadly responsive to stim-
hedonic quality of the stimulus.
uli representing different taste qualities (Ozeki and Sato,
Although the descending projections of the gustatory
1972; Sato, 1972; Tonosaki and Funakoshi, 1984; Sato and
NST have been studied, the differential contributions of
Beidler, 1997). However, because of their relatively small
sucrose-or quinine-responsive NST neurons, for example,
membrane potentials and the possibility of leak currents
to this caudally directed pathway are not known. A multi-
associated with penetrating such small cells with sharp
synaptic pathway(s) supports these ingestive and rejection
electrodes, many investigators have viewed these intracel-
reflexes and probably involves intrinsic projections of the
lular experiments with skepticism (Kinnamon, 1988;
gustatory NST to other parts of the NST and to premotor
Avenet and Lindemann, 1989; Lindemann, 1996; Herness
brainstem areas involved in swallowing, salivation, and
and Gilbertson, 1999). More recent experiments have
chewing (Travers and Norgren, 1983; Travers, 1988;
employed patch-clamp recording methods on isolated taste
Beckman and Whitehead, 1991; Dinardo and Travers,
receptor cells (Akabas et al., 1988; Kinnamon et al., 1988;
1997). The taste-responsiveness of NST neurons must pro-
Gilbertson et al., 1993; Cummings et al., 1996), but the
vide a gating mechanism for differentiating palatable from
range of stimuli that can be applied to an isolated cell
unpalatable substances.
preparation is limited and recording is hindered by having
the apical and basolateral membranes in the same bathing
III. BASIC RESPONSE PROPERTIES medium. Recent experiments, however, have combined
OF GUSTATORY NEURONS patch-clamp recording with apically restricted stimulus
application (Gilbertson et al., 2001). Whole-cell record-
A universal characteristic of gustatory neurons is their ings were made from 103 cells in rat fungiform taste buds
responsiveness to stimuli representing more than one of the maintained in an intact lingual epithelium. Up to six taste
classic four taste qualities. The responses of these neurons stimuli were applied to the apical membrane of each cell
are also modulated by changes in stimulus concentration by perfusion through a closed mucosal chamber, which
and often by tactile and thermal stimulation. Therefore, any effectively separated the apical from the basolateral taste
theory of the role of these neurons in coding sensory infor- cell membranes. The data showed that individual taste
mation must take this breadth of responsiveness into receptor cells often exhibited a range of chemical sensitiv-
account. Even when gustatory neurons can be shown to be ities, confirming the earlier data obtained with intracellular
relatively narrowly tuned across taste qualities, as occurs in electrodes. Over two thirds of the cells responded with
cells of the primate orbitofrontal cortex (Rolls, 1995), these a reversible change in membrane current (5 pA) to more
neurons still can be modulated by stimulus concentration than one of the four basic stimuli (sucrose, NaCl, HCl, or
and are likely to be multimodal in their responsiveness. QHCl). However, the receptor cells showed greater stimu-
This multiple sensitivity raises a serious issue with regard lus specificity than typically seen in first-or second-order
to taste quality coding by labeled lines, as discussed below. afferent neurons, which become more broadly tuned
through convergence (see below). Thus, one source of the
multiple sensitivity of peripheral and central gustatory
A. Gustatory Sensitivities of Taste Receptor Cells
neurons arises at the initial step of stimulus recognition by
the taste receptor cells themselves.
Taste transduction involves a variety of mechanisms,
including direct permeation or block of ion channels and
2. Random Distribution of Sensitivities: Maximum
activation of metabotropic and ionotropic receptors (for
Information Transfer
recent reviews, see Lindemann, 1996; Herness and
Gilbertson, 1999) (see also chapter 34). Whereas most of Analysis of the distribution of responses to the four basic
the data that demonstrate these mechanisms have been stimuli across rat fungiform taste cells demonstrates that
obtained either through patch recording of isolated taste the sensitivities to sucrose, NaCl, HCl, and QHCl are ran-
receptor cells or from molecular or biochemical methods, domly distributed and independent of each other
there is little information about how these mechanisms are (Gilbertson et al., 2001). If the proportion of cells in a sam-
distributed within and across receptor cells. Recent whole- ple responding to each stimulus is known, it is possible,
cell recording experiments from taste cells in intact tongue using the laws of probability, to predict the frequency of
epithelium, however, confirm that taste receptor cells are occurrence of various combinations of responses to the
broadly tuned to gustatory stimuli (Gilbertson et al., 2001). stimuli if the sensitivities to each are independent and
738 Smith and Scott

randomly distributed. For example, in this experiment, 0.6 The maximum information content, which would occur if
of the cells responded to sucrose and 0.5 responded to all possible response combinations among these four stim-
QHCl. A random distribution predicts that 0.3 (i.e., 0.6  uli (n
16) occurred with equal frequency, would be
0.5) should respond to both sucrose and QHCl, which was 4 bits (i.e., log2 16
4). On the other hand, if responses to
seen in the data. Similarly, all other possible combinations the basic stimuli were restricted to only four kinds of cells,
of sensitivities (NaCl with HCl or sucrose with NaCl or each specific to one of the basic tastes and equally distrib-
sucrose with HCl and QHCl, etc.) were also predictable by uted, the maximum information that could be transferred
random combination. Similar analyses on the responses from the receptors to the afferent nerve would only be
of rat CT and IXth nerve fibers (Frank and Pfaffmann, 2 bits (i.e., log2 4
2).
1969) and on data obtained with intracellular electrodes on In general, the greater the amount of information trans-
rat taste receptor cells (Ozeki and Sato, 1972) have also fer, the finer the discriminations that can be made on the
demonstrated a random distribution of sensitivities. basis of the sensory input (see Pfaff, 1975). Even if one were
The broad responsiveness of taste receptor cells could to assume only four taste qualities, the many hundreds of
result from an overlap in the transduction mechanisms for potential gustatory stimuli would be composed of subtle
different classes of taste stimuli within single receptor combinations of these four. The ability of rats to make
cells, as reported in hamster taste cells for sodium salts and a behavioral discrimination between, for example, the taste
acids, which both use the amiloride-sensitive Na channel of sucrose and maltose (Spector et al., 1997) or QHCl and
(Gilbertson et al., 1992, 1993). Alternatively, multiple KCl (St. John and Spector, 1998), depends on a system with
receptors and transduction cascades could be present more subtle discrimination capabilities than could be
within a single cell (see Glendinning and Hills, 1979) or expected of one with specifically tuned cells. An inform-
there could be cell-to-cell communication within the taste ation content of 4 bits provides four times the discriminabil-
bud (see Roper, 1993) that allows the sharing of inform- ity of one of only 2 bits. Thus, the extensive, independent
ation. Recent data showing the co-expression of a large distribution of taste sensitivities across receptor cells and the
number of a family of putative bitter taste receptors in sin- resulting broadly tuned afferent fibers and central neurons
gle taste receptor cells (Adler et al., 2000; Chandrashekar (see below) provides the substrate for an across-neuron
et al., 2000) are not incompatible with the broad tuning of pattern code, which permits relatively subtle behavioral dis-
receptor cells, which leaves open the possibility that these crimination performance (see also Pfaff, 1975).
same cells could express other receptors as well. Further
molecular studies should be able to provide definitive B. Multiple Sensitivity of Peripheral
evidence for the origin of the multiple sensitivities of indi- Taste Fibers
vidual taste receptor cells.
At first glance, a random distribution of sensitivities The broad responsiveness of CT fibers in the rat, cat, and
seems to be counterintuitive. What possible advantage rabbit referred to above first led Pfaffmann (1941, 1955) to
could there be to such an arrangement? One possibility propose that taste quality is represented by the pattern of
may lie in the greater capacity of such a system for trans- activity across these afferent nerve fibers. Since gustatory
mitting information (see Pfaff, 1975). A basic tenet of nerve fibers are not specifically tuned to a single taste qual-
information theory is that multicomponent messages con- ity, many investigators have attempted to classify these
vey maximum information only when the individual com- fibers into functional groups in an effort to impose order
ponents are independent of each other (Shannon and on the organization of gustatory afferent information.
Weaver, 1959). This means that sensory systems that
encode information by the pattern of activity across
1. Responses to Basic Taste Stimuli: Response
broadly tuned neural elements are inherently capable of
Profiles and Fiber Types
transmitting greater amounts of information than systems
using specifically tuned cells (Pfaff, 1975). The amount of Although there have been several schemes for classifying
information that can be transmitted by a system can be taste fibers into meaningful groups, the most common are
calculated from the probabilities of events, such as the pro- those based on similarities and differences in the fibers’
portions of various response combinations in taste receptor response profiles. Of these, the most widely adopted has
cells. An informational analysis of the response distribu- been the “best-stimulus” classification, first used by Frank
tions across rat taste receptor cells indicates that the broad (1973, 1974) for grouping hamster and squirrel monkey
and independent distribution of sensitivities across these CT nerve fibers. When the anterior tongue was stimulated
cells permits almost as much information as possible in the by mid-range concentrations of 0.1 M sucrose, 0.03 M
responses to four stimuli (3.93 bits) (Smith et al., 2000). NaCl, 0.003 M HCl, and 0.001 M QHCl, fibers of the
Gustatory Neural Coding 739

hamster’s CT nerve could be categorized into one of three responsiveness of a taste neuron to other stimuli within the
groups: sucrose (S)-best, NaCl (N)-best, or HCl (H)-best same quality class. That is, if a neuron responded vigor-
(Frank, 1973). In Frank’s study, only one CT fiber out of ously to sucrose it almost always gave strong responses to
79 responded best to QHCl (Q-best). Interestingly, when other stimuli that are sweet to humans and that are
the stimuli were arranged in order of taste preference (i.e., classified as similar to sucrose by hamsters in behavioral
S, N, H, and Q), the response profile of each best-stimulus experiments. Similar observations followed for the
group peaked at a single point (the best stimulus), with the responses of N-best neurons to other sodium and lithium
responses to the “side-band” stimuli diminishing on either salts and of H-best neurons to other acids and nonsodium
side. The responses of S-, N-, and H-best CT fibers of the salts. Consequently, cluster analysis was applied to
hamster to 13 taste stimuli are shown in Figure 3A. These responses of taste fibers in the CT nerve (Frank et al.,
cells were relatively narrowly tuned to stimuli of a given 1988) and to neurons in the NST and PbN (Smith et al.,
class, but even at these moderate concentrations, they 1983b) of the hamster. The results gave strong support for
responded to stimuli of more than one quality. Similar the classification of these neurons into types on the basis
groups were seen at central gustatory nuclei of the hamster of similarities and differences in their response profiles.
(Fig. 3B, C) when the same stimuli were applied to the The response profiles of S-, N-, and H-best neurons were
anterior portion of the tongue (Travers and Smith, 1979; strikingly different from one another, even when the
Van Buskirk and Smith, 1981). responses to as many as 18 stimuli were used to define the
The homogeneity of the response profiles within a best- broad profile of each type. There was enough similarity
stimulus group made it relatively easy to predict the within each neuron type to warrant considering each to be

Figure 3 Mean neural response profiles for S-, N-, and H-best neurons in the CT nerve (A), the NST (B), and the PbN (C) of the ham-
ster to an array of 13 stimuli, all at midrange concentrations (see Smith et al., 1983b; Frank et al., 1988). The response of each class of
neurons to the four basic stimuli (sucrose, NaCl, HCl, and QHCl) are shaded. The mean breadth of tuning (H) is given above each pro-
file. Breadth of tuning increased systematically from the periphery to the pons, especially the tuning of S-best cells. (Data are from Frank
et al. (1988) for the CT and Smith et al. (1983b) for the NST and PbN.)
740 Smith and Scott

a single class of neurons. Hierarchical cluster analysis has hindbrain typically increase their firing rates as stimulus
been applied in many other species and at several levels of concentration is increased, but the slopes of these concen-
the gustatory system (Van Buskirk and Smith, 1981; Frank tration-response functions often vary across stimuli, even in
et al., 1988; Hanamori et al., 1988; Scott and Giza, 1990; the same neuron (see Fig. 2). Consequently, the shapes of
Giza and Scott, 1991; Travers, 1993; Rolls, 1995), and the response profiles may be considerably different across
most investigators agree on the usefulness of classifying the range of effective stimulus concentrations. As seen in
neurons into functional types. At the very least, describing auditory tuning curves (Kiang, 1965) and in other sensory
taste neurons by their best stimulus provides a convenient modalities, increasing stimulus intensity generally produces
method for imposing some order on the system. Beyond greater breadth of responsiveness (Smith and Travers,
that, there are a number of studies suggesting that these 1979). When stimulus concentrations are reduced, cells
neuron types are biologically relevant, e.g., studies show- become more narrowly tuned until they only respond to
ing that only N-and S-best neurons receive input from the a single (best) stimulus. Therefore, measures of the breadth
amiloride-sensitive Na transduction mechanism of tuning of gustatory neurons, regardless of the method
(Hettinger and Frank, 1990; Scott and Giza, 1990; Giza used to quantify this concept, can be greatly confounded by
and Scott, 1991; Boughter and Smith, 1998; Boughter the relative concentrations of the test stimuli. Making
et al., 1999; St.John and Smith, 2000). The role that these comparisons across neural levels, across experiments from
neuron types play in the coding of taste quality is different laboratories, and especially across species is a dif-
addressed below. ficult task. There are instances, however, where stimulus
concentrations and other factors have been held constant
2. Breadth of Responsiveness across experiments, allowing a direct comparison of the
breadth of tuning of taste neurons under different conditions,
Because taste neurons typically respond to stimuli as in Figure 3 (Van Buskirk and Smith, 1981; Rolls, 1995).
representing more than one taste quality, many investiga-
tors have attempted to describe their breadth of 3. Ambiguity in the Response of One Cell
responsiveness across these stimuli. Early attempts
generally employed idiosyncratic criteria for what consti- Because the responses of taste neurons are broadly tuned
tuted a response, making it difficult to compare cells in across both quality and intensity, the response of any one
different species or at different levels of the gustatory neuron is entirely ambiguous with respect to either para-
system. Smith and Travers (1979) introduced a measure of meter (Pfaffmann, 1955, 1959). In addition, gustatory
the breadth of tuning based on the entropy equation used in neurons are often responsive to thermal and tactile stimuli
information theory to measure information transmission. (Ogawa et al., 1968; Travers and Smith, 1984; Hanamori
In this procedure, the response of a neuron to each stimulus et al., 1987; Travers, 1993), as discussed below. This multi-
is expressed as a proportion (pi) of the total response to all modal sensitivity is not entirely unexpected because taste is
four. The breadth of tuning (H) is given by the equation: only one aspect of oral sensation and both temperature and
touch have been shown to influence taste perception
n
H
 冱 pilog pi (Moskowitz and Arabie, 1970; Bartoshuk et al., 1982; Green
i
1 and Frankmann, 1987; Cruz and Green, 2000).
Nevertheless, impulse traffic in a single neuron may be
where H is the breadth of responsiveness and K is a
related to several stimulus modalities, making the unam-
scaling constant (1.661 for four stimuli). The value of H
biguous interpretation of that signal impossible without
varies from 0.0, indicating a neuron that responds exclu-
comparing it to activity in other cells (Crick, 1979;
sively to one of the four stimuli, to 1.0, indicating a neuron
Erickson, 1982a, 1984).
that responds equally to all four. Therefore, larger values of
H reflect a greater breadth of tuning across the four basic
4. Neuron Types in Different Nerves and Different
stimuli. This measure has been used extensively to
Species
compare the breadth of tuning of gustatory neurons
across different levels of the nervous system (e.g., see The response profiles of gustatory nerve fibers have been
Fig. 3) (Van Buskirk and Smith, 1981; Rolls, 1995) and characterized in several species, but those in the hamster are
across species (Travers, 1993). the easiest to compare because the data were collected
Because gustatory neurons respond to different stimulus under similar conditions in each experiment. Fibers of the
classes and are also modulated by stimulus concentration, facial (VIIth) nerve innervate taste buds on the anterior
the response profile of a cell, and its apparent breadth of tongue via the CT nerve and on the soft palate and within
tuning, can vary with stimulus intensity. Taste neurons in the the nasoincisor ducts via the GSP nerve. Within the hamster
Gustatory Neural Coding 741

CT nerve, there are three distinct fiber types: S-, N-, and gustatory fibers of the VIIth and IXth nerves of the
H-best (Frank, 1973). Mean response profiles for 20 S-, 41 hamster, those of the SLN do not discriminate among taste
N-, and 17 H-best CT fibers are shown in Figure 3A. In the qualities (Smith and Hanamori, 1991). These fibers
CT nerve, the S-best fibers are clearly the most specifically respond well to water (W), NaCl, and HCl, but poorly to
tuned (see Fig. 3A), with a mean breadth of tuning (H) of sucrose and QHCl. On the basis of response magnitude, it
0.39, compared to 0.59 and 0.67 for the N- and H-best was possible to identify 26 W-best, 17 N-best, 20 H-best,
fibers, respectively (Van Buskirk and Smith, 1981). The and 2 Q-best fibers in the hamster SLN, but a hierarchical
S-best fibers are quite narrowly tuned to sucrose and other cluster analysis showed that these fiber types were not at
sweet-tasting stimuli, with some sensitivity to NaCl and all distinct, as are those in the CT and IXth nerves of this
very little response to HCl or QHCl. In comparison to the species. Instead, the sensitivities of SLN fibers were
rat, the CT nerve of the hamster is considerably more graded along a continuum, with no distinct differences
responsive to sweet stimuli. The greater breadth of tuning among fibers responding best to different stimuli.
of the N- and H-best fibers in the hamster CT arises Laryngeal nerve fibers do not discriminate well among the
predominantly from the fact that these fiber types respond four basic taste stimuli, but appear to serve a role in
to both NaCl and HCl, albeit to different degrees. protection of the airway from foreign substances.
There is considerably more sensitivity to sweet stimuli Taste fibers recorded from primates show sensitivities
in the GSP nerve of both the rat (Nejad, 1986) and hamster similar to those of the rodent species that have been inves-
(Harada and Smith, 1992) than in the CT nerve of either tigated, except that they appear to be somewhat more
species. This difference is more striking in the rat than the narrowly tuned across the four basic taste qualities.
hamster. In addition, single-neuron recordings have been Responses of 45 CT fibers of the squirrel monkey
made from cells in the rat NST that respond to stimulation were divisible into distinct S-, N-, or H-best categories,
of the nasoincisor ducts (Travers and Norgren, 1991), very similar to those observed in the hamster CT nerve
showing a considerable number of S-best neurons, many (Frank, 1974; Pfaffmann et al., 1976). Although the
of which receive converging input from both GSP and CT breadth of tuning was not quantified for these fibers, the
nerves. Behavioral experiments have also shown that taste squirrel monkey CT fibers were somewhat more narrowly
receptors innervated by the GSP contribute substantially to tuned to their best stimulus than those of the hamster.
the rat’s ability to respond to sweet-tasting stimuli (Krimm Fibers of the macaque CT nerve are divisible into four rel-
et al., 1987; Spector et al., 1993, 1997). atively distinct groups, responding best to sucrose, NaCl,
Bitter stimuli, such as QHCl, are relatively ineffective HCl, or QHCl (Sato et al., 1975). Like the squirrel monkey
in driving fibers of either the CT or the GSP, at least at CT fibers, those in the macaque appear to be more specif-
mid-range concentrations. Fibers of the glossopharyngeal ically tuned to their best stimulus than those in either rat or
(IXth) nerve, however, are much more responsive to QHCl hamster. Recent data from the chimpanzee also suggest a
in both the rat and hamster. In comparison to the hamster’s high degree of specificity of CT afferent fibers (Hellekant
CT nerve, where there are few Q-best and many S- and and Ninomiya, 1991). Thus, in comparison to rodents, the
N-best fibers, the distribution of these fiber types is peripheral gustatory afferent fibers of primates are more
reversed in the IXth nerve. The magnitudes of the narrowly tuned to the basic taste stimuli.
responses to these chemicals are also reversed between the
CT, where sucrose and NaCl produce much larger C. Convergence and Breadth of Tuning of
responses, and the IXth nerve, where the response to QHCl Central Neurons
is considerably greater than that to either sucrose or NaCl.
In the rat, many Q-best fibers are found in the IXth nerve Gustatory neurons in the CNS are typically more broadly
and no N-best fibers at all (Frank, 1991). Consequently, tuned than peripheral nerve fibers in the same species. This
the information projected to the NST in both of these increased breadth of tuning is likely due to convergence of
species by the VIIth nerve is dominated by preferred stim- peripheral fibers onto brainstem neurons. Neurons in the
uli (sucrose and NaCl), whereas that from the IXth nerve NST, for example, can often be driven by stimulation of
is more concerned with aversive stimuli (QHCl). As noted taste receptors in anatomically separate regions of the oral
above, however, behavioral discrimination among all types cavity (Sweazey and Smith, 1987; Travers and Norgren,
of stimuli appears to depend predominantly on input from 1991). In addition to greater breadth of tuning to gustatory
the VIIth nerve, rather than the IXth. stimuli, central taste neurons often respond to thermal and
Taste buds within the laryngeal mucosa are innervated tactile stimuli (Travers and Smith, 1984; Travers, 1993)
by fibers of the internal branch of the superior laryngeal and at cortical levels to visual and olfactory stimuli (Rolls
nerve (SLN), a branch of the vagus (Xth) nerve. Unlike and Baylis, 1995).
742 Smith and Scott

1. Increase in Breadth of Tuning in NST and PbN laboratories or species, these data suggest that the trend of
increasing breadth of responsiveness between periphery
Experiments in which CT fibers (Frank, 1973) and neurons
and brainstem appears to hold across several species. They
of the NST (Travers and Smith, 1979) and parabrachial
further suggest no additional increase in the breadth of tun-
nuclei (PbN; Van Buskirk and Smith, 1981) of the hamster
ing of taste cells between brainstem and cortex and per-
were stimulated by applying the same stimuli (e.g., 0.1 M
haps even a slight return toward the narrower tuning char-
sucrose, 0.03 M NaCl, 0.003 M HCl, and 0.001 M QHCl) to
acteristic of peripheral fibers. Data from several
the anterior tongue have revealed a systematic increase in
experiments on the macaque (Rolls, 1995) show that cen-
the breadth of responsiveness of these cells from the periph-
tral gustatory neurons in this primate decrease their
ery to the pons. A comparison of the mean response profiles
breadth of tuning between the NST and the cortex, with
of S-, N-, and H-best neurons of the hamster CT, NST, and
NST cells having a breadth of tuning value of 0.87 (simi-
PbN was shown in Fig. 3. The mean breadth of tuning (H)
lar to that shown in Fig. 5 for the rat NST), frontal opercu-
increased from 0.56 for CT fibers to 0.66 in the NST and
lum cells a value of 0.67, neurons of the insula a value of
0.65 in the PbN. In particular, the breadth of tuning of the S-
0.56, and the cells in the secondary cortical taste area of
best cells, which are the most specifically tuned fibers in the
the orbitofrontal cortex a breadth of tuning of 0.39. In
CT nerve, increased from 0.39 in the CT to 0.59 in the NST
comparison, CT fibers in the macaque (Sato, 1975) have
to 0.74 in the PbN (see Fig. 3), making these neurons more
a mean breadth of tuning of 0.57, showing that the effect
broadly tuned in the pons than the very broadly tuned H-best
of synaptic processing from the CT to the insula is to
cells (Van Buskirk and Smith, 1981). This reflects a change
increase the breadth of tuning and then return it to periph-
from a typical S-best CT fiber that responds predominantly
eral levels. Even cells with entropies between 0.5 and 0.6
to sucrose with only slight sideband sensitivity to one other
are relatively broadly tuned (see Fig. 3). Although cells
stimulus (usually NaCl) to an S-best pontine neuron that
in the orbitofrontal cortex are quite narrowly tuned to
responds well to at least three of the four basic stimuli. The
tastants, they are often multimodal, responding also to
breadth of tuning of hamster PbN neurons is quite evident in
olfactory and visual stimuli (Rolls and Baylis, 1995).
Figure 4A, which shows the mean responses of S-, N-, and
Therefore, neurons may become more specific in one
H-best PbN neurons to an array of 18 stimuli.
modality while extending their sensitivities across others
An increase in breadth of responsiveness from periph-
to serve an emergent property, in this case perhaps an
ery to brainstem has been noted in other species as well
appreciation of flavor.
(Travers, 1993), although the comparisons have been made
across data collected in different laboratories and not
2. Convergence onto Single NST Cells from Different
necessarily employing stimuli at the same concentrations.
Fields
Across five CT fiber experiments on rat, hamster, and
monkey, the mean breadth of responsiveness was 0.56 There is some convergence of input onto single CT fibers,
(range 0.52–0.61), whereas the mean breadth across 10 because each one branches to innervate several fungiform
studies of NST cells in rat, hamster and monkey was 0.74 papillae (Miller, 1971; Oakley, 1975). The sensitivities of
(range 0.62–0.87). The responses of 100 rat NST neurons each branch of a CT axon appear to be the same, suggest-
to sucrose, NaCl, HCl, and QHCl are shown in Figure 5. ing little consequence of this peripheral convergence on
The cells are arranged in order of their best stimuli, with taste information processing other than might accrue
S-best cells on the left (shaded bars), N-best cells in the through spatial summation (Oakley, 1975). Indeed, studies
center (open bars), and H-best cells on the right (shaded of the effects of the area of stimulation on human taste
bars); there were no Q-best cells in this sample. Neurons responses show that spatial summation does occur on the
are arranged within each group in order of the response to anterior tongue (Smith, 1971). On the other hand, there is
their best stimulus; the response of any one cell is read considerable convergence of peripheral fiber inputs onto
from top to bottom. The mean breadth of tuning (entropy) second-order gustatory cells of the NST, often occurring
of these cells was 0.81, reflecting their extremely broad between anatomically separate subpopulations of taste
tuning to the four stimuli. buds. The receptive fields of taste-responsive cells in the
Travers (1993) also compared data at peripheral and NST of the sheep are considerably larger than those of the
brainstem levels to those from cells in insular cortex of rats CT nerve, suggesting that CT fibers converge onto
and insular-opercular cortex of monkeys, where the mean secondary neurons in the NST (Vogt and Mistretta, 1990).
breadth of tuning across six experiments was an interme- About half of the taste-responsive cells of the rat NST
diate level of 0.63 (range 0.54–0.76). Given the caveat receive converging input from separate taste bud regions
that these studies are not completely comparable across (Travers, 1993). Travers and colleagues have demonstrated
Gustatory Neural Coding 743

Figure 4 (A) Mean responses (impulses/s) of each of three neuron classes in the hamster PbN to 18 stimuli. The sweet-tasting stimuli
are shaded in the profile for the S-best cells, the sodium salts are shaded in the profile for the N-best neurons, and the nonsodium salts
and acids are shaded in the response profile for the H-best neurons. (Modified from Smith et al., 1983a.) (B) Patterns of activity gener-
ated across 31 hamster PbN neurons by two sodium salts (filled symbols) and by sucrose (open symbols). The across-neuron patterns to
NaCl and NaNO3 correlated  0.94, whereas that to sucrose was not correlated with the pattern evoked by either sodium salt (e.g.,
r
0.09 with NaCl). (Data from Smith et al., 1983a.)

that the patterns of convergence of peripheral fibers onto (Travers et al., 1986). That is, taste buds from the anterior
NST neurons in the rat are predominantly among taste or from the posterior oral cavity converge onto NST cells,
buds of the anterior tongue, nasoincisor ducts, or sublin- but there is little evidence that anterior and posterior
gual organ on the one hand, or among those of the foliate receptive fields combine in the responsiveness of an NST
papillae, soft palate, or retromolar mucosa on the other neuron. These patterns of convergence reflect the orderly
744 Smith and Scott

anterior-to-posterior progression of food through the oral hamster PbN suggests that within a receptor population
cavity during ingestion, making the simultaneous stimul- (anterior tongue) there may be further convergence at
ation of the converging inputs likely (Travers, 1993). higher levels (Van Buskirk and Smith, 1981).
Although convergence has been examined mostly in the
NST, it has also been noted at other levels of the central
3. Multimodal Sensitivity: Touch and Temperature
gustatory pathway. The available evidence suggests no
greater convergence of separate taste bud subpopulations In addition to their extensive breadth of responsiveness
at pontine or thalamic levels (Travers, 1993), although the across stimuli of different taste quality, gustatory neurons
increased breadth of responding of S-best neurons in the at all levels often respond to other sensory modalities, most

Figure 5 Net responses (impulses/s above or below spontaneous rate) of 100 rat NST neurons to the four basic taste stimuli. Neurons
are arranged along the abscissa according to their best stimulus, with cells 1–15 being S-best (shaded bars), 16–64 N-best (open bars),
and 65–100 H-best (shaded bars); no cells in this sample were Q-best. The mean spontaneous activity of each cell is shown at the
bottom of the figure. These cells had a mean breadth of tuning of  0.81. (Data selected from Giza and Scott, 1991; Giza et al., 1991,
1997.)
Gustatory Neural Coding 745

often to intraoral somatosensory stimuli. Fibers of the CT neurons receive converging input from gustatory and
nerve of rats and hamsters respond to both gustatory and tactile receptors (Travers, 1993). The few studies that have
thermal stimulation (Ogawa et al., 1968). This sensitivity examined this question at the level of the PbN and thala-
to warming and cooling is maintained in central gustatory mus show an average amount of gustatory/mechanical
neurons of the NST (Ogawa et al., 1988), PbN (Travers convergence of 77% and 88%, respectively (Travers,
and Smith, 1984), thalamus (Verhagen et al., 1999), and 1993). In awake, freely ingesting rats, there is some
cortex (Yamamoto et al., 1981). There is a strong correla- convergence of both thermal and olfactory stimuli onto
tion at both peripheral and central levels between the cortical gustatory neurons (Yamamoto et al., 1989).
responses to warming the tongue and to sucrose and Multimodal convergence has been examined in primates
between the responses to cooling and to NaCl or HCl. The most extensively in the secondary cortical taste area in the
responses of a hamster PbN neuron to gustatory and ther- orbitofrontal cortex, where there are cells responding to
mal stimulation are shown in Figure 6. This S-best neuron combinations of taste and olfactory, visual, or somatosen-
responded also to NaCl and to 37 °C distilled water sory stimuli (Rolls, 1995; Rolls and Baylis, 1995). Thus it
(Travers and Smith, 1984). Recent human psychophysical is clear that taste-responsive cells in the central nervous
experiments show that warming and cooling small areas on system are driven often by stimuli differing in taste quality,
the anterior tongue can evoke sensations of sweet and salty intensity, and sensory modality. The contribution of the
or sour, respectively (Cruz and Green, 2000). In the rat responses of these multimodal cells to the coding of any
NST, one half of the taste-responsive cells with receptive particular attribute, such as taste quality, must be assessed
fields in the anterior oral cavity and all of those with recep- with this multiple sensitivity in mind.
tive fields in the posterior oral cavity are responsive also to
mechanical stimulation of their receptive fields (Travers D. Temporal and Spatial Aspects of
and Norgren, 1995). In fact, across a number of studies it Taste Processing
has been shown that about 64% of taste-responsive NST
The responses of taste neurons typically are analyzed by
examining the impulse frequency during some defined
period of time, such as the 1- or 5-second interval after the
arrival of the stimulus at the tongue. The responses to tas-
tants, however, vary over time, and to some extent the tem-
poral parameters of the responses to different stimuli are
relatively distinct. As a consequence, some investigators
have suggested that temporal characteristics may be
important in coding information about taste quality. In
addition, there is some evidence that cells responding to
different stimuli are differentially distributed in central
gustatory nuclei, suggesting the possibility of a chemo-
topic code for taste quality. Overall, however, the data sup-
porting either a strictly temporal or topographic spatial
code for gustatory quality are not compelling.

1. Time Course of Taste Responses: Differences

Among Stimuli
The temporal characteristics of gustatory neural
responses are often ignored in studies of taste physiology.
Figure 6 Oscillographic records for the responses of a hamster Where they have been examined, investigators have noted
PbN neuron to the four basic gustatory stimuli: sucrose (S), NaCl
that different classes of stimuli produce responses with
(N), HCl (H), and QHCl (Q) and to warming (37 °C) and cooling
different time courses. For example, responses to NaCl
(17 °C) of the tongue. The arrows indicate the approximate
onsets of the responses. The histograms in the lower right show are characterized by a brief phasic portion followed by a
the mean responses of this neuron to the six stimuli for three tests slowly declining tonic component (Smith et al., 1975).
of the taste and two tests of the thermal stimuli. This neuron The responses to sucrose, on the other hand, have a much
responded as well to warm distilled water as to sucrose. (From less pronounced, if not absent, phasic component. There
Travers and Smith, 1984.) are differences among NaCl, HCl, QHCl, and sucrose in
746 Smith and Scott

the time it takes the rat CT nerve to reach its maximum rate at which it flows over the tongue. Beyond this initial
firing rate, the effects of stimulus concentration on that rapid decline of the phasic response, there is a secondary
rise time, and the ratio of the phasic to the tonic compon- decline during the tonic component of the response, which
ent of the response (Harada et al., 1983). Responses of rat probably reflects adaptation at the level of the taste recep-
CT nerve fibers to different tastants are correlated with tor cell, where a similar slow decline is observed (Ozeki,
different temporal patterns, but this correlation is not pre- 1971). Recovery of the rat CT response to NaCl after
cise enough to serve as an unambiguous code for taste adaptation follows a time course similar to this slow adap-
quality (Nagai and Ueda, 1981). The time course of the tation process (Smith et al., 1978).
response is largely a function of the stimulus, but also
depends on which neurons are activated. It is likely, how-
3. Chemotopic Correlates of Taste in the CNS:
ever, that differences in time course can add to the dis-
Orotopic Representation
tinctions that occur in the activity across cells to enhance
the discrimination among stimuli, resulting in a kind of The idea that taste quality might be represented in the
spatiotemporal pattern that provides a unique signature nervous system by a topographic pattern of activity has
for each stimulus (Scott, 1987a). Although differences in been around for many years, although it does not have
the temporal patterns of activity elicited by different strong experimental support. There is some evidence that
stimuli may enhance the resolution among stimuli, an cells responding to different stimuli are differentially
unambiguous neural representation of taste quality is distributed in central gustatory nuclei. Topographic distri-
provided by the pattern of activity across neurons, as butions of responsiveness have been noted in the NST of
discussed below. the rat (Halpern and Nelson, 1965), hamster (Dickman and
Smith, 1989; McPheeters et al., 1990), and monkey (Scott
et al., 1986b). Generally, NaCl and sucrose responses are
2. Response to Constant Stimulation: Rate of
more prevalent in the most rostral portion of the NST,
Onset Versus Adaptation
and responses to HCl and QHCl are prominent at more
The temporal characteristics of the response to a taste caudal levels. To a large extent, these differences reflect
stimulus reflect a number of underlying factors, including the differential sensitivity of the CT and IXth nerves, since
the rate of stimulus application and the process of adapta- these nerves terminate in a rostral to caudal order, project-
tion. The response to NaCl in the rat CT nerve has two ing an orotopic map onto the NST. However, even with
distinct components: an initial phasic discharge followed only anterior tongue stimulation, there is some indication
by a tonic response component (Pfaffmann and Powers, of a differential distribution of best-stimulus categories of
1964). This phasic/tonic response pattern is common to cells in the hamster PbN, with 75% of N- and S-best cells
many stimuli and occurs in the gustatory responses of all located in the caudal half and 65% of H-best neurons in the
species that have been investigated. Analysis of the time rostral half of the pontine taste area (Van Buskirk and
course of the NaCl response in the rat CT nerve shows it Smith, 1981). A similar crude topographic organization of
to decrease from the initial phasic peak in two exponential chemosensitivity was reported in the macaque NST, but
stages. The initial phasic response declines with a time not in the insular-opercular cortex of that species (Smith-
constant of about 0.5 second, and this is followed by Swintosky et al., 1991). In the rat cortex, however, distinct
a tonic response component that decays much more spatial patterns of responsiveness are generated by taste
slowly, with a time constant of about 30 seconds (Smith stimuli, with sucrose responses most dominant in the
et al., 1975). The ratio of the phasic and tonic components anterodorsal region, quinine responses in the posterior
of the CT response to NaCl varies directly with the rate of region, NaCl responses in the central and ventral regions,
stimulus onset. Impulse frequency during the phasic and HCl responses evenly distributed throughout the
response is a power function of the rate of stimulus rise, cortical taste area (Yamamoto et al., 1985). These “across-
and if the stimulus is applied slowly enough the phasic region patterns” have been proposed as a code for the
response is completely eliminated (Smith and Bealer, representation of taste quality (Yamamoto, 1989). Even if
1975). These data show that the gustatory system is such patterns can be reliably related to taste quality, their
exquisitely sensitive to the rate of stimulus change. Taste generation is dependent to a large extent on the topo-
neurons respond appropriately to increasing stimulus con- graphic distribution of afferent inputs from different parts
centration, but the rate at which that concentration of the oral cavity, which serves to create the regional dif-
changes is a particularly salient feature of the stimulus. ferences in responsiveness seen at various levels of the
Therefore, the time course of the neural response to a tas- CNS. Since taste qualities can be reliably identified by
tant reflects not only the chemical stimulus, but also the humans following stimulation of a single fungiform
Gustatory Neural Coding 747

papilla (Bealer and Smith, 1975), it is hard to conceive 1963). This theory accommodated the multiple sensitivity
of how such crude topographic representations of the input of taste fibers and required neither specific fiber types nor
of the entire oral cavity could serve as a neural code for taste primaries.
gustatory quality. However, the prevailing view of the importance of pri-
mary tastes led to a modification of the labeled-line theory,
in which it was proposed that taste quality is coded by the
activity in a few “best-stimulus” channels, that is, by
IV. TASTE QUALITY PROCESSING: neurons that respond best, although not specifically, to one
GUSTATORY NEURAL CODING of the basic taste qualities (Pfaffmann, 1974; Pfaffmann
et al., 1976).
The principal debate over taste quality coding has Although each coding theory has its strengths, both
focused on two competing theories: the across-fiber pat- strain to encompass the full range of data. For example,
tern theory (Pfaffmann, 1959; Erickson et al., 1965; accumulating evidence suggests that there are indeed func-
Erickson, 1968, 1982a) and the labeled-line hypothesis tional classes of neurons that correspond in some way
(Pfaffmann, 1974; Pfaffmann et al., 1976). There has to primary taste qualities (Boudreau and Alev, 1973;
been considerable disagreement over the past two Frank, 1973; Contreras and Frank, 1979; Smith et al.,
decades about the merits of these two ways of represent- 1983b; Chang and Scott, 1984; Frank et al.,
ing taste quality. Adherents to the labeled-line hypothesis 1988; Hanamori et al., 1988; Jacobs et al., 1988; Ninomiya
propose that gustatory neuron types are specific coding and Funakoshi, 1988; Hettinger and Frank, 1990; Scott
channels for taste quality. Proponents of the across-fiber and Giza, 1990; Frank, 1991; Giza and Scott, 1991; Smith
pattern theory, on the other hand, propose that quality is and Frank, 1993). On the other hand, analysis of the rela-
coded by the pattern of activity across neurons. In the tive activity of these neuron groups shows that no single
following discussion, we examine the role of gustatory class of cells in isolation can reliably discriminate among
neuron types in the definition of the across-neuron pat- different taste qualities (Smith et al., 1983b; Smith, 1985;
terns. From the result of such an examination, we see that Smith and Frank, 1993). The neural coding problem essen-
the separate neuron types establish and define the dis- tially rests on whether the activity in a given taste neuron,
tinctions among the patterns that are necessary for taste or class of neurons, is an unambiguous representation of
discrimination. Therefore, the same neurons are critical the quality of the stimulus applied to its receptors (i.e., a
for coding taste quality, regardless of whether they are labeled line) or whether this activity is meaningful only in
considered a “labeled line” or a necessary part of the the context of activity in other afferent neurons (i.e., in
“across-fiber pattern.” Taste discrimination is impossible their relative patterns of response). Here we review the rel-
without the contribution of each neuron type. evant neurophysiological data that bear on this issue and
suggest that taste quality is coded in the relative rates of
A. Theories of Quality Coding activity across several neuron types (Smith, 1985; Smith
and Frank, 1993).
Prior to the development of neurophysiological recording
methods, a long tradition of human psychophysical
1. Breadth of Tuning: Across-Fiber Patterns
research had provided considerable support for the notion
that taste experience could be reduced to a few basic It was the multiple sensitivity of fibers in the CT nerve that
qualities, although not necessarily the traditional four first led Pfaffmann (1941, 1955, 1959) to propose that taste
(Bartoshuk, 1971). This idea, combined with Müller’s quality is coded by the pattern of activity across taste
doctrine of specific nerve energies, led to the expectation fibers. According to this hypothesis, taste quality of the
that the perception of taste quality would arise from the basic stimuli remains invariant through the mid-range of
activation of a few specific neuron types, each coding a intensity, even though any single neuron may increase its
single taste quality. This strict “labeled-line” theory was breadth of responsiveness. The pattern of activity gener-
discounted by early neurophysiological recordings ated across the entire array of taste neurons rises with
showing that peripheral taste fibers in several species were increasing concentration, but retains its shape (Erickson
responsive to stimuli representing more than one taste et al., 1965; Ganchrow and Erickson, 1970; Erickson,
quality (Pfaffmann, 1941, 1955). In response, an “across- 1982a, 1984). Stimuli with similar tastes, such as sodium
fiber pattern” theory was proposed, which held that taste salts, generate highly similar patterns of activity across
quality is coded by the relative activity across the popula- afferent taste neurons, as depicted by the across-neuron
tion of responsive neurons (Pfaffmann, 1959; Erickson, patterns for NaCl and NaNO3 across hamster PbN neurons
748 Smith and Scott

shown in Figure 4B. The pattern of response to sucrose, 2. Response Profiles: Labeled Lines
on the other hand (Fig. 4B), is strikingly different from
Although mammalian taste neurons are broadly tuned,
those evoked by the sodium salts. The similarities
many investigators have attempted to group them into
among these patterns are typically measured by calculating
functionally meaningful categories (Pfaffmann, 1941,
the across-fiber correlation between pairs of stimuli
1955; Boudreau and Alev, 1973; Frank, 1973; Travers
(Erickson et al., 1965), although other indices have been
and Smith, 1979; Van Buskirk and Smith, 1981; Smith
proposed (Gill and Erickson, 1985; DiLorenzo, 1989).
et al., 1983b; Frank et al., 1988; Hanamori et al., 1988).
Several behavioral investigations have shown that
It is obvious from knowledge about the organization of
stimuli that evoke highly correlated neural patterns are
taste sensitivities that it is possible to group taste neu-
judged by experimental animals to have similar tastes
rons into classes based on their best responses to the four
(Erickson, 1963; Morrison, 1967; Smith et al., 1979;
prototypical stimuli (see Fig. 3). Indeed, hierarchical
Nowlis et al., 1980; Nowlis and Frank, 1981). This across-
cluster analyses of the similarities and differences in the
fiber pattern view of quality coding incorporates the
shapes of their response profiles results in distinct
multiple sensitivity of gustatory neurons as an essential
groups of neurons that correspond to these “best-stimu-
part of the neural code for taste quality. This theoretical
lus” classes (Smith et al., 1983b; Chang and Scott, 1984;
view stresses that the code for quality is given in the
Scott et al., 1986a, 1986b; Frank et al., 1988; Hanamori
response of the entire population of cells (Pfaffmann,
et al., 1988; Rolls et al., 1988; Frank, 1991). The impli-
1959; Erickson, 1968, 1982a, 1984), placing little empha-
cation of distinct fiber types in the coding of taste qual-
sis on the role of an individual neuron. Erickson (1968)
ity began with the categorization of hamster CT fibers
has argued that pattern coding is used by many sensory
into best-stimulus groups (Frank, 1973). This type of
systems.
categorization became the focus of an ensuing contro-
When the across-neuron correlations among the
versy over the neural representation of taste quality
responses to an array of gustatory stimuli are calculated
when it was proposed that these fiber types code taste
across either peripheral or central neurons, stimuli with
quality in a labeled-line fashion (Pfaffmann, 1974;
similar tastes correlate highly and those with different
Pfaffmann et al., 1976). This hypothesis suggests that
tastes correlate poorly. Almost every neurophysiological
“sweetness” is coded by activity in S-best neurons,
study that has taken this approach to analyzing the
“saltiness” by activity in N-best neurons, etc. Pfaffmann
responses of gustatory neurons has shown that the across-
(1974) first proposed a labeled-line code for sweetness
neuron patterns reflect the qualitative similarities among
because activity in S-best fibers of the squirrel monkey
taste stimuli (Erickson et al., 1965; Erickson, 1968,
CT nerve correlated better with the animal’s preference
1982a, 1984; Pfaffmann et al., 1976; Smith et al., 1979,
behavior toward a number of sugars than did activity in
1983a; Travers and Smith, 1979; Van Buskirk and Smith,
the whole nerve. Thus, in contrast to a “population”
1981; Smith, 1985; Scott et al., 1986a, 1986b, 1991; Frank
approach to taste coding (across-neuron patterns), this
et al., 1988; Scott and Giza, 1990; Giza and Scott, 1991).
labeled-line position advocates a “feature extraction”
Often these across-neuron correlations serve as input to a
approach, in which particular neurons (or groups of neu-
multivariate statistical procedure in order to generate a
rons) play specific roles in the representation of taste
“taste space,” which represents the neural similarities and
quality. A labeled-line code, by definition, implies that
differences among the stimuli. A taste space for 18 stimuli
activity in that “line” carries a specific message, without
is shown in Figure 7; this three-dimensional space was
any reliance on the activity in other “lines” (Perkel and
derived from correlations among the responses of 31 ham-
Bullock, 1968).
ster PbN neurons and was generated using multidimen-
sional scaling (Smith et al., 1983a). Within this space,
there is a clear separation between the sweet-tasting B. The Role of Neuron Types in Quality Coding
stimuli, the sodium salts, the nonsodium salts and acids,
and the two bitter-tasting stimuli. This arrangement of The labeled-line hypothesis requires the existence of
stimuli, based on similarities among their across-neuron neuron types for the coding of taste quality, whereas the
patterns, suggests that there is sufficient information across-neuron pattern theory does not. The number of
within these patterns to discriminate among these four labeled lines would equal the number of discrete taste
groups of stimuli, even though any one cell in the hamster qualities, which would each be signaled by activity in sep-
PbN is likely to respond to stimuli of more than one group arate afferent channels. Consequently, the existence of
(see Fig. 4A) (Van Buskirk and Smith, 1981; Smith et al., gustatory neuron types has been sharply contested
1983a, 1983b). (Woolston and Erickson, 1979; Erickson, 1985), based on
Gustatory Neural Coding 749

Figure 7 Three-dimensional “taste space”

showing the locations of 18 stimuli,
obtained from multidimensional scaling of
the correlations in the responses evoked
across 31 hamster PbN neurons. Four
groups of stimuli are indicated by different
symbols: sweet-tasting (open circles),
sodium salts (), nonsodium salts and acids
(solid circles), and bitter-tasting (open trian-
gles). Although these cells were activated
by applying stimuli only to the anterior
tongue, there is enough information in the
across-neuron patterns to clearly differenti-
ate these four classes of gustatory stimuli,
which are clearly behaviorally distinct to
the hamster. (Modified from Smith et al.,
1983a.)

the assumption that their existence somehow implicates Responses in each neuron type dominate and essentially
them as labeled lines. However, the mere existence of fiber define the similarities in the across-neuron patterns evoked
types (defined by their best stimulus, similarities in their by their “best” stimulus class, as seen in the responses to
response profiles, or by other criteria such as their several acids in hamster PbN neurons (Fig. 8). Without the
amiloride sensitivity) does not imply that these classes of contribution of the H-best neurons (shaded), these acids
cells compose labeled lines. A classic example where would not produce similar across-neuron patterns.
receptor types are evident but where there is general agree- Examination of the contribution of each neuron type to the
ment about the existence of a pattern code is in vertebrate across-neuron patterns showed that each neuron type was
color vision (Crick, 1979; Erickson, 1984; Smith, 1985; necessary for determining the similarities among stimuli
Smith and Frank, 1993). for which it was “best,” but that the reliable discrimination
among stimuli of different classes depended upon the
activity in more than one neuron type. For example, within
1. Neural Discrimination of Taste Inputs
a multivariate taste space based on the across-neuron pat-
Because peripheral taste fibers and central neurons are terns generated from the responses of neurons in the ham-
broadly tuned, it is difficult for the activity in a single ster PbN, sodium salts, and nonsodium salts were readily
neuron or even in a single neuron type to discriminate distinguishable (see Fig. 7). However, if either the N-best
among stimuli that differ in taste quality. Several years or the H-best class of cells, which are preferentially sensi-
ago, Smith and colleagues (Smith et al., 1983a, 1983b; tive to sodium salts and nonsodium salts, respectively,
Smith, 1985) examined the roles played by neuron types in were missing from the data matrix, the patterns across the
the hamster brainstem in the definitions of the across- remaining cells did not permit distinctions between these
neuron patterns and the ability of these neuron types two classes of salts. Thus, the discrimination among stim-
alone or in combination to discriminate among stimuli. uli with different tastes depended upon the relative activity
750 Smith and Scott

in different neuron types; one neuron type alone was insuf- can be encoded accurately by considering the relative
ficient to discriminate among stimuli with different taste activity in these three photoreceptors, that is, by a pattern
qualities (Smith et al., 1983a). (Erickson, 1968; Boynton, 1971). Deficiencies in one or
This “across-neuron type” code is simply a restatement more of the photoreceptor pigments result in various forms
of the across-fiber pattern theory first proposed by of visual chromatic deficiency, or “color blindness”
Pfaffmann (1959), except that it puts an emphasis on the (Boynton, 1971). For example, the lack of the long-wave-
activity in recognizable neuron types. This coding mecha- length photopigment results in a confusion in color
nism is similar to the coding of stimulus wavelength by the discrimination (protanopia), where individuals perceive
vertebrate visual system, where three types of broadly long-wavelength stimuli as one color (yellow) and short
sensitive photoreceptor pigments are involved (Marks wavelength stimuli another (blue), with a neutral (gray)
et al., 1964). The wavelength of light falling on the retina point at 494 nm (Graham et al., 1961; Boynton, 1971).

Figure 8 The responses of 31 hamster PbN neu-

rons to four acids, arranged into across-neuron
patterns according to the magnitude of their
response to 0.003 M tartaric acid. The H-neuron
group is shaded in each pattern, and the correla-
tion of each pattern with that for tartaric acid is
shown. Without the contribution of the H-best
neurons, the responses to these behaviorally simi-
lar stimuli would not be well correlated. (From
Smith et al., 1983a.)
Gustatory Neural Coding 751

Data on color-blind individuals show that the absence of neuron types that defines the unique patterns that represent
any one of the three photoreceptor types results in the taste quality. In that sense, whether these neuron types are
inability to discriminate among particular sets of wave- viewed as a “labeled line” or the critical part of an “across-
lengths (Smith, 1985; Smith and Frank, 1993). Similarly, fiber pattern,” they are crucial to the neural code for taste
the neural discrimination among different classes of gusta- quality. Taste discrimination depends on the differential
tory stimuli depends upon the relative activity in different responses in these separate neuron classes, the activity of
neuron types. which defines the across-neuron patterns.

2. Comparison of Activity Across Neuron Types 3. Synthetic Versus Analytic Processing

Taste offers no known analogy to the pathology of color Drawing parallels between color vision and taste pro-
blindness, but experiments on the rat NST provide an vides an interesting way to understand the broad tuning
experimental demonstration of an analogous phenomenon of gustatory neurons, but is not without its limitations. In
(Scott and Giza, 1990; Giza and Scott, 1991; St. John and his earliest formulations of the across-fiber pattern the-
Smith, 2000). Blocking the NaCl response of the N-best ory, Erickson (1968, 1973) based many of his ideas on
neurons with amiloride resulted in the inability of the vertebrate photoreceptors, which are broadly responsive
remaining cells to discriminate between sodium and non- across stimulus wavelength and which code color infor-
sodium salts, that is, their across-neuron patterns were not mation by their relative patterns of response. Inherent in
distinct without the input from the N-best cells. Without these early arguments was the idea that if broadly tuned
considering the differential response of two neuron types neurons serve to code stimulus information by patterns,
(N-best and H-best) to these two classes of stimuli, they then the system must be synthetic rather than analytic
were coexistent within the taste space of the amiloride- (Erickson, 1977). That is, in color vision, a mixture of
treated rat (Fig. 9). These results are compatible with the two colored lights (e.g., red and yellow) results in the
interpretation that taste quality discrimination depends on synthesis of a new sensation (e.g., orange), whereas the
a comparison of activity across broadly tuned neuron components of a mixture in an analytic modality main-
types, comparable to the coding of color vision by tain their individuality (as in a musical chord). Erickson
broadly tuned photoreceptors (Smith et al., 1983a; Smith, and colleagues have published several papers suggesting
1985; Smith and Frank, 1993). In both taste and color that taste is not entirely analytic, as is pitch perception,
vision, elimination of any one cell type results in a lack of and may be more akin to the synthetic senses (Erickson,
separation among stimuli of different qualities within the 1977, 1982b; Erickson and Covey, 1980). Bartoshuk and
multidimensional space (Smith, 1985; Scott and Giza, colleagues have countered that human subjects clearly
1990; St. John and Smith, 2000) and a loss of behavioral recognize the individual components in taste mixtures,
discrimination among the same stimuli (Graham et al., arguing that the gustatory system is analytic in nature
1961; Boynton, 1971; Hill et al., 1990; Spector et al., (Bartoshuk, 1978; Bartoshuk and Gent, 1985). Still oth-
1996). ers suggest that taste is neither purely synthetic nor
The data cited above show that no one gustatory neuron purely analytic but that mixtures exhibit fusion, creating
type is capable of providing information that can reliably new sensations, as occurs in food flavors, but in which
distinguish the across-neuron patterns evoked by dissimilar- the individual components can often be recognized, espe-
tasting compounds. Multiple neuron types must contribute cially by trained observers (McBurney, 1986; Halpern,
to the pattern for dissimilar stimuli to be coded as being 1997). That separate taste sensations fuse into a new, but
distinct. Thus, in that sense, the various neuron types (S-, ultimately analyzable, sensation may be most compatible
N-, H-, or Q-best) are critically important for the with a neural system that codes information by patterns,
discrimination of taste quality. The N-best cells can define which themselves are dominated by activity in clearly
the similarities among sodium salts, but activity in both defined neuronal cell types. That is, both the neurophys-
neuron types is required to distinguish sodium salts from iology of the gustatory system and a good deal of psy-
nonsodium salts and acids. Behavioral and neural data in chophysical data support the idea that taste is not entirely
rats support this requirement, but there is no evidence to synthetic or analytic in nature (see Halpern, 1997, for a
date that N-best cells are labeled lines signaling “salti- thorough discussion of this issue). A challenge to the fur-
ness.” On the contrary, N-best cells provide the critical part ther understanding of neural coding in this system is to
of a pattern across neuron types that codes saltiness, yet determine precisely the relationship between the activity
these cells by themselves cannot distinguish sodium salts of these broadly tuned afferent neurons and the
from other stimuli. It is the activity in these particular psychophysical responses to taste mixtures.
752 Smith and Scott

it difficult to accept a pure labeled-line code for gustatory

quality. Whereas a labeled-line code might be more com-
patible with an analytic sensory system than a pattern code
would be (Bartoshuk and Gent, 1985), no other aspect of
gustatory neurobiology favors a simple labeled-line expla-
nation. The most troubling problem facing the labeled-line
approach is the multiple sensitivity of gustatory afferent
neurons, especially those in the central nervous system,
which are also multimodal in many instances (see above).
It was this multiple sensitivity in peripheral afferent fibers
that led to Pfaffmann’s (1955, 1959) original formulation
of the across-fiber pattern idea and this finding has only
been reinforced by subsequent research. Even taste recep-
tor cells are broadly tuned across stimuli representing
different taste qualities (e.g., Gilbertson et al., 2001; Ozeki
and Sato, 1972).
In broadly tuned taste cells, the activity elicited by stim-
uli with different taste qualities can be interpreted by the
labeled-line theory in one of two ways, which are depicted
in Fig. 10. The “sideband” activity in a neuron type (e.g., the
response to HCl by N-best cells) represents either noise
(Fig. 10A) or signal (Fig. 10B). If it represents noise, then
there is a very poor signal-to-noise ratio in gustatory
neurons (see also Travers and Smith, 1979; Scott and Giza,
2000), making it difficult to conceive of how a signal could
ever be reliably detected (Fig. 10A). This problem is espe-
cially apparent as the concentration of nonbest stimuli is
raised because, as noted by Pfaffmann (1955, 1959), a con-
centration can usually be found where qualitatively distinct
stimuli produce equivalent responses in gustatory neurons
(see Fig. 2, above). If, on the other hand, such sideband
responses represent signal, then the responses of broadly
tuned neurons in the rat NST, for example, would suggest
Figure 9 Three-dimensional taste spaces showing the relative that the “basic” taste stimuli, such as NaCl or HCl, have
similarity of the across-neuron patterns among 15 stimuli before multiple taste qualities to the rat (Fig. 10B). Behavioral
(A) and after (B) treatment of the tongue and palate with experiments, however, flatly refute this implication. Rats
amiloride. Stimulus locations before amiloride are typical of trained to avoid one of the basic taste stimuli do not gener-
those of the normally functional rat taste system, with sweet-tast- alize that aversion to any of the others (Nowlis and Frank,
ing stimuli separated from sodium and lithium salts, which are in 1981), even though rats will avoid mixtures of the condi-
turn separated from acids and nonsodium salts and quinine. After tioned stimulus (CS) with other stimuli in proportion to the
amiloride application, all the Na-bearing stimuli, except MSG, concentration of the CS in the mixture (Smith and Theodore,
migrate to the positions of the acids, nonsodium salts, and qui-
1984). Thus it is difficult to conceive of how activity in
nine, indicating similar across-neuron patterns to these stimuli
after amiloride. Abbreviations: N1–N4, 0.01–0.3 M NaCl; Li,
different neuron types can represent a specific taste quality.
LiCl; S, sucrose; G, glucose; Sa, sodium saccharin; H, HCl; Ci, Another advantage of pattern coding arises out of the
citric acid; Q, qunine hydrochloride; K, KCl; Ca, CaCl2; M, efficiency of stimulus representation by multiple cells.
MSG; P, polycose. (Adapted from Giza and Scott, 1991.) Detection or recognition of a gustatory stimulus requires
the nervous system to select a neural signal out of back-
ground noise. Each potential signal has some probability
4. Advantages of a Pattern Code: Detecting Signals
distribution associated with it, as does background noise.
in Noise
The task of the system is to detect the signal in the noise
The broad tuning of gustatory cells at every level of the and the ability of a system to do this depends upon the size
nervous system, from the receptor cell to the cortex, makes of the signal and the variability of both the noise and sig-
Gustatory Neural Coding 753

Figure 10 Schematic diagrams of the implications of a

“labeled-line” coding mechanism for the interpretation of the
activity in broadly tuned neurons of the rat NST. (A) The
mean response profiles of N-best and H-best neurons in the rat
NST. If the activity in a given cell type (N-best or H-best) that
is elicited by stimuli other than the “best” stimulus is “noise”
(shaded area), then there is very little “signal” (solid area)
being provided by these broadly tuned neurons. Indeed, there
are more impulses generated in these cells by the “sideband”
stimuli together than by the best stimulus. (B) The mean
responses of each of three neuron types (S, N, and H-best) in
the rat NST to 0.1 M NaCl and 0.01 M HCl. If the response to
NaCl in H-best neurons or the response to HCl in N-best neu-
rons is “signal” (gray bars), then NaCl should have a large
“acid” taste to rats and HCl should have a large “sodium”
taste. Behavioral experiments (Nowlis et al., 1980) clearly do
not support such a conclusion, even though it has been shown
that rats are capable of responding to the presence of a condi-
tioned stimulus in a taste mixture (Smith and Theodore, 1984).
(Data from St. John and Smith, 2000.)

nal distributions (Pfaff, 1975). Neural networks composed stimuli. These and other considerations, as delineated
of multiply sensitive cells are more efficient at detecting above, suggest strongly that the gustatory system repre-
signals than systems composed of specific cells (see also sents taste quality by the patterns of activity across broadly
Winograd, 1963). As more cells (n) are used in combin- tuned neurons.
ation to represent a signal, the signal becomes easier to
detect because the mean firing rate increases as a function
REFERENCES
of n, whereas the variability associated with the probabil-
ity distribution of the signal across the combination of cells
Adler, E., Hoon, M. A., Mueller, K. L., Chandrasheskar, J., Ryba,
increases as a function of the square root of n (see Pfaff,
N. J. P., and Zuker, C. S. (2000). A novel family of mam-
1975, for a detailed demonstration of this idea). Therefore, malian taste receptors. Cell 100:693–702.
the reliability of detection increases with the number of Akabas, M. H., Dodd, J., and Al-Awqati, Q. (1988). A bitter sub-
cells involved in the representation of the stimulus. For a stance induces a rise in intracellular calcium in a subpopula-
given system, the more broadly tuned the cells, the greater tion of rat taste cells. Science 242:1047–1050.
the number of cells available to represent any given Avenet, P., and Lindemann, B. (1989). Perspectives of taste
stimulus and the finer the discriminations possible among reception. J. Membrane Biol. 112:1–8.
754 Smith and Scott

Bartoshuk, L. M. (1971). The chemical senses. In Woodworth Cummings, T. A., Daniels, C., and Kinnamon, S. C. (1996).
and Schlosberg’s Experimental Psychology, J. W. Kling and Sweet taste transduction in hamster: sweeteners and cyclic
L. A. Riggs (Eds.). Holt, Rinehart and Winston, New York, nucleotides depolarize taste cells by reducing a K current.
pp. 169–191. J. Neurophysiol. 75:1256–1263.
Bartoshuk, L. M. (1978). Gustatory system. In Handbook of Davis, B. J., and Smith, D. V. (1997). Substance P modulates
Behavioral Neurobiology. I. Sensory Integration, R. B. Masterton taste responses in the nucleus of the solitary tract of the ham-
(Eds.). Plenum, New York, pp. 503–567. ster. NeuroReport 8:1723–1727.
Bartoshuk, L. M., and Gent, J. F. (1985). Taste mixtures: an Dickman, J. D., and Smith, D. V. (1989). Topographic distribu-
analysis of synthesis. In Taste, Olfaction, and the Central tion of taste responsiveness in the hamster medulla. Chem.
Nervous System, D. W. Pfaff (Ed.). The Rockefeller Senses 14:231–247.
University Press, New York, pp. 210–232. DiLorenzo, P. M. (1988). Taste responses in the parabrachial
Bartoshuk, L. M., Rennert, K., Rodin, J., and Stevens, J. C. pons of decerebrate rats. J. Neurophysiol. 59:1871–1887.
(1982). Effects of temperature on the perceived sweetness of DiLorenzo, P. M. (1989). Across unit patterns in the neural
sucrose. Physiol. Behav. 28:905–910. response to taste: vector space analysis. J. Neurophysiol.
Bealer, S. L., and Smith, D. V. (1975). Multiple sensitivity to 62:823–833.
chemical stimuli in single human taste papillae. Physiol. DiLorenzo, P. M. (1990). Corticofugal influence on taste responses
Behav. 14:795–799. in the parabrachial pons of the rat. Brain Res. 530:73–84.
Beckman, M. E., and Whitehead, M. C. (1991). Intramedullary DiLorenzo, P. M., and Monroe, S. (1995). Corticofugal influence
connections of the rostral nucleus of the solitary tract in the on taste responses in the nucleus of the solitary tract in the rat.
hamster. Brain Res. 557:265–279. J. Neurophysiol. 74:258–272.
Bertino, M., Beauchamp, G. K., and Engelman, K. (1986). DiNardo, L. A., and Travers, J. B. (1997). Distribution of Fos-like
Increasing dietary salt alters salt taste preference. Physiol. immunoreactivity in the medullary reticular formation of the
Behav. 38:203–213. rat after gustatory elicited ingestion and rejection behaviors.
Boudreau, J. C., and Alev, N. (1973). Classification of chemore- J. Neurosci. 17:3826–3839.
sponsive tongue units of the cat geniculate ganglion. Brain Erickson, R. P. (1963). Sensory neural patterns and gustation. In
Res. 54:157–175. Olfaction and Taste, Y. Zotterman (Ed.). Pergamon, Oxford,
Boughter, J. D., Jr., and Smith, D. V. (1998). Amiloride blocks pp. 205–213.
acid responses in NaCl-best gustatory neurons of the hamster Erickson, R. P. (1968). Stimulus coding in topographic and non-
solitary nucleus. J. Neurophysiol. 80:1362–1372. topographic afferent modalities: on the significance of the
Boughter, J. D., Jr., St. John, S. J., and Smith, D. V. (1999). activity of individual sensory neurons. Psychol. Rev. 75:
Neural representation of the taste of NaCl and KCl in gusta- 447–465.
tory neurons of the hamster solitary nucleus. J. Neurophysiol. Erickson, R. P. (1973). Parallel “population” neural coding in fea-
81:2636–2646. ture extraction. In The Neurosciences: Third Study Program,
Boynton, R. M. (1971). Color vision. In Woodworth and F. O. Schmitt and F. G. Worden (Eds.). MIT Press,
Schlosberg’s Experimental Psychology, J. W. Kling and L. A. Cambridge, MA, pp. 155–169.
Riggs (Eds.). Holt, Rinehart and Winston, New York, Erickson, R. P. (1977). The role of “primaries” in taste research.
pp. 315–368. In Olfaction and Taste VI, J. LeMagnen and P. MacLeod
Bradley, R. M., King, M. S., Wang, L., and Shu, X. (1996). (Eds.). Information Retrieval, London, pp. 369–376.
Neurotransmitter and neuromodulator activity in the gustatory Erickson, R. P. (1982a). The “across-fiber pattern” theory: an
zone of the nucleus tractus solitarius. Chem. Senses 21:377–385. organizing principle for molar neural function. In
Brining, S. K., Belecky, T. L., and Smith, D. V. (1991). Taste Contributions to Sensory Physiology, Vol. 6, W. D. Neff (Ed.).
reactivity in the hamster. Physiol. Behav. 49:1265–1272. Academic Press, New York, pp. 79–110.
Carpenter, J. A. (1956). Species differences in taste preference. Erickson, R. P. (1982b). Studies on the perception of taste: Do
J. Comp. Physiol. Psychol. 49:139–144. primaries exist? Physiol. Behav. 28:57–62.
Chang, F. -C. T., and Scott, T. R. (1984). Conditioned taste aver- Erickson, R. P. (1984). On the neural basis of behavior. Am. Sci.
sions modify neural responses in the rat nucleus tractus soli- 72:233–241.
tarius. J. Neurosci. 4:1850–1862. Erickson, R. P. (1985). Grouping in the chemical senses. Chem.
Chandrasheskar, J., Mueller, K. L., Hoon M. A., Adler, E., Feng, Senses 10:333–340.
L., Guo, W., Zuker, C. S., and Ryba, J. P. (2000). T2Rs func- Erickson, R. P., and Covey, E. (1980). On the singularity of taste
tion as bitter taste receptors. Cell 100:703–711. sensations: What is a taste primary? Physiol. Behav.
Contreras, R., and Frank, M. (1979). Sodium deprivation alters 25:527–533.
neural responses to gustatory stimuli. J. Gen. Physiol. Erickson, R. P., Doetsch, G. S., and Marshall, D. A. (1965). The
73:569–594. gustatory neural response function. J. Gen. Physiol.
Crick, F. H. C. (1979). Thinking about the brain. Sci. Am. 49:247–263.
241:219–232. Formaker, B. K., and Hill, D. L. (1991). Lack of amiloride sensi-
Cruz, A., and Green, B. G. (2000). Thermal stimulation of taste. tivity in SHR and WKY glossopharyngeal taste responses to
Nature 403:889–892. NaCl. Physiol. Behav. 50:765–769.
Gustatory Neural Coding 755

Frank, M. (1974). The classification of mammalian afferent taste Graham, C. H., Sperling, H. G., Hsia, Y., and Coulson, A. H.
nerve fibers. Chem. Sens. Flav. 1:53–60. (1961). The determination of some visual functions of a uni-
Frank, M., and Pfaffmann, C. (1969). Taste nerve fibers: a ran- laterally color-blind subject: methods and results. J. Psychol.
dom distribution of sensitivities to four tastes. Science 51:3–32.
164:1183–1185. Green, B. G., and Frankmann, S. P. (1987). The effect of cooling
Frank, M. E. (1973). An analysis of hamster afferent taste nerve the tongue on the perceived intensity of taste. Chem. Senses
response functions. J. Gen. Physiol. 61:588–618. 12:609–619.
Frank, M. E. (1991). Taste-responsive neurons of the glossopha- Grill, H. J. (1985). Introduction: physiological mechanisms in
ryngeal nerve of the rat. J. Neurophysiol. 65:1452–1463. conditioned taste aversion. Ann. NY Acad. Sci. 443:67–88.
Frank, M. E., Bieber, S. L., and Smith, D. V. (1988). The organi- Grill, H. J., and Norgren, R. (1978a). Chronically decerebrate
zation of taste sensibilities in hamster chorda tympani nerve rats demonstrate satiation but not bait shyness. Science
fibers. J. Gen. Physiol. 91:861–896. 201:267–269.
Friedman, M. I., and Mattes, R. D. (1991). Chemical senses Grill, H. J., and Norgren, R. (1978b). The taste reactivity test. I.
and nutrition. In Smell and Taste in Health and Disease, Mimetic responses to gustatory stimuli in neurologically nor-
T. V. Getchell, R. L. Doty, L. M. Bartoshuk and J. B. Snow, Jr. mal rats. Brain Res. 143:263–280.
(Eds.). Raven Press, New York, pp. 391–404. Grill, H. J., and Norgren, R. (1978c). The taste reactivity test. II.
Ganchrow, J. R., and Erickson, R. P. (1970). Neural correlates Mimetic responses to gustatory stimuli in chronic thalamic
of gustatory intensity and quality. J. Neurophysiol. and decerebrate rats. Brain Res. 143:281–297.
33:768–783. Hajnal, A., Takenouchi, K., and Norgren, R. (1999). Effect of
Garcia, J. R., Kovner, R., and Green, K. F. (1970). Cue properties intraduodenal lipid on parabrachial gustatory coding in awake
vs. palatability of flavors in avoidance learning. Psychon. Sci. rats. J. Neurosci. 19:7182–7190.
20:313–314. Halpern, B. P. (1997). Psychophysics of taste. In Handbook of
Gilbertson, T. A., Avenet, P., Kinnamon, S. C., and Roper, S. D. Perception and Cognition: Tasting and Smelling, G. K.
(1992). Proton currents through amiloride-sensitive Na Beauchamp and L. Bartoshuk (Eds.). Academic Press, New
channels in hamster taste cells: role in acid transduction. York, pp. 77–123.
J. Gen. Physiol. 100:803–824. Halpern, B. P., and Nelson, L. M. (1965). Bulbar gustatory
Gilbertson, T. A., Roper, S. D., and Kinnamon, S. C. (1993). responses to anterior and to posterior tongue stimulation in
Proton currents through amiloride-sensitive Na channels in the rat. Am. J. Physiol. 209:105–110.
isolated hamster taste cells: enhancement by vasopressin and Hanamori, T., Ishiko, N., and Smith, D. V. (1987). Multimodal
cAMP. Neuron 10:931–942. responses of taste neurons in the frog nucleus tractus solitar-
Gilbertson, T. A., Booghter, J. D., Jr., Zhang, H., and Smith, D. V. ius. Brain Res. Bull. 18:87–97.
(2001). Distribution of gustatory sensitivities in rat taste cells: Hanamori, T., Miller, I. J., Jr., and Smith, D. V. (1988). Gustatory
whole-cell responses to apical chemical stimulation. J. responsiveness of fibers in the hamster glossopharyngeal
Neurosci. 21:4931–4941. nerve. J. Neurophysiol. 60:478–498.
Gill, J. M., and Erickson, R. P. (1985). Neural mass differences in Harada, S., Marui, T., and Kasahara, Y. (1983). Analysis of the
gustation. Chem. Senses 10:531–548. initial taste responses from rat chorda tympani nerve. Jpn. J.
Giza, B. K., and Scott, T. R. (1983). Blood glucose selectively Oral Biol. 25:566–570.
affects taste-evoked activity in rat nucleus tractus solitarius. Harada, S., and Smith, D. V. (1992). Gustatory sensitivities of the
Physiol. Behav. 31:643–650. hamster’s soft palate. Chem. Senses 17:37–51.
Giza, B. K., and Scott, T. R. (1987). Intravenous insulin infusions Hayama, T., Ito, S., and Ogawa, H. (1985). Responses of solitary
in rats decrease gustatory-evoked responses to sugars. Am. J. tract nucleus neurons to taste and mechanical stimulation
Physiol. 252:R994–R1002. of the oral cavity in decerebrate rats. Exp. Brain Res.
Giza, B. K., and Scott, T. R. (1991). The effect of amiloride on 60:235–242.
taste-evoked activity in the nucleus tractus solitarius of the rat. Hellekant, G., and Ninomiya, Y. (1991). On the taste of umami in
Brain Res. 550:247–256. chimpanzee. Physiol. Behav. 49:927–934.
Giza, B. K., Scott, T. R., Sclafani, A., and Antonucci, R. F. Herness, M. S., and Gilbertson, T. A. (1999). Cellular mecha-
(1991). Polysaccharides as taste stimuli: their effect in the nisms of taste transduction. Ann. Rev. Physiol. 61:873–900.
nucleus tractus solitarius of the rat. Brain Res. 555:1–9. Hettinger, T. P., and Frank, M. E. (1990). Specificity of amiloride
Giza, B. K., Ackroff, K., McCaughey, S. A., Sclafani, A., and inhibition of hamster taste responses. Brain Res. 513:24–34.
Scott, T. R. (1997). Preference conditioning alters taste Hill, D. L., Formaker, B. K., and White, K. S. (1990). Perceptual
responses in the nucleus of the solitary tract of the rat. Am. J. characteristics of the amiloride-suppressed sodium chloride
Physiol. 273:R1230–R1240. taste response in the rat. Behav. Neurosci. 104:734–741.
Glendinning, J. I., and Hills, T. T. (1997) Electrophysiological Jacobs, K. M., Mark, G. P., and Scott, T. R. (1988). Taste
evidence for two transduction pathways within a bitter-sensi- responses in the nucleus tractus solitarius of sodium-deprived
tive taste receptor. J. Neurophysiol. 78:734–745. rats. J. Physiol. (Lond.) 406:393–410.
Glenn, J. F., and Erickson, R. P. (1976). Gastric modulation of Kiang, N. Y.-S. (1965). Stimulus coding in the auditory nerve and
gustatory afferent activity. Physiol. Behav. 16:561–568. cochlear nucleus. Acta Otolaryngol. 59:186–200.
756 Smith and Scott

King, M. S., Wang, L., and Bradley, R. M. (1993). Substance P Nachman, M. (1963). Learned aversion to the taste of lithium
excites neurons in the gustatory zone of the rat nucleus trac- chloride and generalization to other salts. J. Comp. Physiol.
tus solitarius. Brain Res. 619:120–130. Psychol. 56:343–349.
Kinnamon, S. (1988). Taste transduction: A diversity of mecha- Nagai, T., and Ueda, K. (1981). Stochastic properties of gustatory
nisms. Trends Neurosci. 11:491–496. impulse discharges in rat chorda tympani fibers. J.
Kinnamon, S. C., Dionne, V. E., and Beam, K. G. (1988). Apical Neurophysiol. 45:574–592.
localization of K channels in taste cells provides the basis Nejad, M. S. (1986). The neural activities of the greater superfi-
for sour transduction. Proc. Natl. Acad. Sci. USA cial petrosal nerve of the rat in response to chemical stimula-
85:7023–7027. tion of the palate. Chem. Senses 11:283–293.
Krimm, R. F., Nejad, M. S., Smith, J. C., Miller, I. J., and Beidler, Ninomiya, Y., and Funakoshi, M. (1988). Amiloride inhibition of
L. M. (1987). The effect of bilateral sectioning of the chorda responses of rat single chorda tympani fibers to chemical and
tympani and the greater superficial petrosal nerves on the electrical tongue stimulations. Brain Res. 451:319–325.
sweet taste in the rat. Physiol. Behav. 41:495–501. Norgren, R. (1985). Taste and the autonomic nervous system.
Li, C.-S., and Smith, D. V. (1997). Glutamate receptor antago- Chem. Senses 10:143–161.
nists block gustatory afferent input to the nucleus of the soli- Norgren, R. (1995). The gustatory system. In The Human
tary tract. J. Neurophysiol. 77:1514–1525. Nervous System, G. Paxinos (Ed.). Academic Press, New
Lindemann, B. (1996). Taste reception. Physiol. Rev. 76: York, pp. 845–861.
719–766. Nowlis, G. H., and Frank, M. E. (1981). Quality coding in gusta-
Liu, H., Behbehani, M. M., and Smith, D. V. (1993). The influ- tory systems of rats and hamsters. In Perception of Behavioral
ence of GABA on cells in the gustatory region of the hamster Chemicals, D. M. Norris (Ed.). Elsevier/North Holland,
solitary nucleus. Chem. Senses 18:285–305. Amsterdam, pp. 59–80.
London, R. M., Snowdon, C. T., and Smithana, J. M. (1979). Nowlis, G. H., Frank, M. E., and Pfaffmann, C. (1980).
Early experience with sour and bitter solutions increases sub- Specificity of acquired aversion to taste qualities in hamsters
sequent ingestion. Physiol. Behav. 22:1149–1155. and rats. J. Comp. Physiol. Psychol. 94:932–942.
Mark, G. P., Scott, T. R., Chang, F.-C. T., and Grill, H. J. (1988). Oakley, B. (1975). Receptive fields of cat taste fibers. Chem.
Taste responses in the nucleus tractus solitarius of the chronic Sens. Flav. 1:431–442.
decerebrate rat. Brain Res. 443:137–148. Ogawa, H., Hayama, T., and Yamashita, Y. (1988). Thermal sen-
Markison, S., St. John, S. J., and Spector, A. C. (1999). sitivity of neurons in a rostral part of the rat solitary tract
Glossopharyngeal nerve transection reduces quinine avoid- nucleus. Brain Res. 454:321–331.
ance in rats not given presurgical stimulus exposure. Physiol. Ogawa, H., Sato, M., and Yamashita, S. (1968). Multiple sensi-
Behav. 65:773–778. tivity of chorda tympani fibres of the rat and hamster to gus-
Marks, W. B., Dobelle, W. H., and MacNichol, E. F., Jr. (1964). tatory and thermal stimuli. J. Physiol. (Lond.) 199:223–240.
Visual pigments of single primate cones. Science Ozeki, M. (1971). Conductance change associated with receptor
143:1181–1183. potentials of gustatory cells in rat. J. Gen. Physiol. 58:688–699.
McBurney, D. H. (1974). Are there primary tastes for man? Ozeki, M., and Sato, M. (1972). Responses of gustatory cells in
Chem. Sens. Flav. 1:17–28. the tongue of rat to stimuli representing four taste qualities.
McBurney, D. H. (1986). Taste, smell, and flavor terminology: Comp. Biochem. Physiol A 41:391–407.
taking the confusion out of fusion. In Clinical Measurement Parker, L. A., Maier, S., Rennie, M., and Crebolder, J. (1992).
of Taste and Smell, H. L. Meiselman and R. S. Rivlin (Eds.). Morphine- and naltrexone-induced modification of palatabil-
Macmillan Publishing Co., New York, pp. 117–125. ity: analysis by the taste reactivity test. Behav. Neurosci.
McBurney, D. H., and Gent, J. F. (1979). On the nature of taste 106:999–1010.
qualities. Psychol. Bull. 86:151–167. Perkel, D. H., and Bullock, T. H. (1968). Neural coding.
McCaughey, S. A., and Scott, T. R. (2000). Rapid induction of Neurosci. Res. Program Bull. 6(3):221–348.
sodium appetite modifies taste activity in the rat nucleus of Pfaff, D. W. (1975). Theoretical consideration of cross-fiber pat-
the solitary tract. Am. J. Physiol., 279:R1121–R1131. tern coding in the neural signalling of pheromones and other
McPheeters, M., Hettinger, T. P., Nuding, S. C., Savoy, L. D., chemical stimuli. Psychoneuroendocrinology 1:79–93.
Whitehead, M. C., and Frank, M. E. (1990). Taste-responsive Pfaffmann, C. (1941). Gustatory afferent impulses. J. Cell.
neurons and their locations in the solitary nucleus of the ham- Comp. Physiol. 17:243–258.
ster. Neuroscience 34:745–758. Pfaffmann, C. (1955). Gustatory nerve impulses in rat, cat and
Miller, I. J., Jr. (1971). Peripheral interactions among single papilla rabbit. J. Neurophysiol. 18:429–440.
inputs to gustatory nerve fibers. J. Gen. Physiol. 57: 1–25. Pfaffmann, C. (1959). The afferent code for sensory quality. Am.
Morrison, G. R. (1967). Behavioural response patterns to salt Psychol. 14:226–232.
stimuli in the rat. Cand. J. Psychol. 21:141–152. Pfaffmann, C. (1964). Taste, its sensory and motivating proper-
Moskowitz, H. R., and Arabie, P. (1970). Taste intensity as a ties. Am. Sci. 52:187–206.
function of stimulus concentration and solvent viscosity. J. Pfaffmann, C. (1974). Specificity of the sweet receptors of the
Texture Stud. 1:502–510. squirrel monkey. Chem. Sens. Flav. 1:61–67.
Gustatory Neural Coding 757

Pfaffmann, C., and Powers, J. B. (1964). Partial adaptation of Scott, T. R., and Giza, B. K. (2000). Issues of gustatory neural
taste. Psychon. Sci. 1:41–42. coding: where they stand today. Physiol. Behav., 69:65–76.
Pfaffmann, C., Frank, M., Bartoshuk, L. M., and Snell, T. C. Scott, T. R., and Mark, G. P. (1987). The taste system encodes
(1976). Coding gustatory information in the squirrel monkey stimulus toxicity. Brain Res. 414:197–203.
chorda tympani. In Progress in Psychobiology and Scott, T. R., Yaxley, S., Sienkiewicz, Z. J., and Rolls, E. T.
Physiological Psychology, Vol. 6, J. M. Sprague and A. N. (1986a). Gustatory responses in the anterior operculum of the
Epstein (Eds.). Academic Press, New York, pp. 1–27. alert cynomolgus monkey. J. Neurophysiol. 56:876–890.
Rameriz, I. (1991). Influence of experience on response to bitter Scott, T. R., Yaxley, S., Sienkiewicz, Z. J., and Rolls, E. T. (1986b).
taste. Physiol. Behav. 49:387–391. Gustatory responses in the nucleus tractus solitarius of the alert
Richter, C. P. (1956). Salt appetite in mammals: its dependence cynomolgus monkey. J. Neurophysiol. 55:182–200.
on instinct and metabolism. In L’Instinct dans de Scott, T. R., Plata-Salaman, C. R., Smith, V. L., and Giza, B. K.
Comportement des Animaux et de L’Homme, M. Autuori (1991). Gustatory neural coding in the monkey cortex: stimu-
(Ed.). Masson et Cie, Paris, pp. 171–176. lus intensity. J. Neurophysiol. 65:76–86.
Richter, C. P., and Campbell, K. (1940). Taste thresholds and taste Shannon, C. E., and Weaver, W. (1959). The Mathematical
preferences of rats for five common sugars. J. Nutr. 20:31–46. Theory of Communication. University of Illinois Press,
Rolls, E. T. (1995). Central taste anatomy and neurophysiology. Urbana.
In Handbook of Olfaction and Gustation, R. L. Doty (Eds.). Smith, D. V. (1971). Taste intensity as a function of area and con-
Marcel Dekker, New York, pp. 549–573. centration: differentiation between compounds. J. Exp.
Rolls, E. T., and Baylis, L.L. (1995). Gustatory, olfactory and Psychol. 87:163–171.
visual convergence within the primate orbitofrontal cortex. Smith, D. V. (1985). The neural representation of gustatory qual-
J. Neurosci. 14: 5437–5452. ity. In Progress in Clinical and Biological Research,
Rolls, E. T., Scott, T. R., Sienkiewicz, Z. J., and Yaxley, S. (1988). Contemporary Sensory Neurobiology, Vol. 176, M. J. Correia
The responsiveness of neurones in the frontal opercular gus- and A. Perrachio (Eds.). Alan R. Liss, New York, pp. 75–97.
tatory cortex of the macaque monkey is independent of Smith, D. V., and Bealer, S. L. (1975). Sensitivity of the rat gus-
hunger. J. Physiol. (Lond.) 397:79–86. tatory system to the rate of stimulus onset. Physiol. Behav.
Rolls, E. T., Sienkiewicz, J., and Yaxley, S. (1989). Hunger mod- 15:303–314.
ulates the responses to gustatory stimuli of single neurons in Smith, D. V., and Frank, M. E. (1993). Sensory coding by periph-
the orbitofrontal cortex. Eur. J. Neurosci. 1:53–60. eral taste fibers. In Mechanisms of Taste Transduction, S. A.
Roper, S. D. (1993). Synaptic interactions in taste buds. In Simon and S. D. Roper (Eds.). CRC Press, Boca Raton, Fl,
Mechanisms of Taste Transduction, S. A. Simon and S. D. pp. 295–338.
Roper (Eds.). CRC Press, Boca Raton, FL, pp. 275–293. Smith, D. V., and Hanamori, T. (1991). Organization of gustatory
Sato, M. (1975). Response characteristics of taste nerve fibers in sensitivities in hamster superior laryngeal nerve fibers.
macaque monkeys: comparison with those in rats and ham- J. Neurophysiol. 65:1098–1114.
sters. In Olfaction and Taste V, D. A. Denton and J. P. Coghlan Smith, D. V., and Li, C. -S. (1998). Tonic GABAergic inhibition
(Eds.). Academic Press, New York, pp. 23–26. of taste-responsive neurons in the nucleus of the solitary tract.
Sato, M., Ogawa, H., and Yamashita, S. (1975). Response prop- Chem. Senses 23:159–169.
erties of macaque monkey chorda tympani fibers. J. Gen. Smith, D. V., and Li, C.-S. (2000). GABA-mediated corticofugal
Physiol. 66:781–810. inhibition of taste-responsive neurons in the nucleus of the
Sato, T. (1972). Multiple sensitivity of single taste cells of the frog solitary tract. Brain Res. 858:408–415.
tongue to four basic stimuli. J. Cell. Physiol. 80:207–218. Smith, D. V., and Theodore, R. M. (1984). Conditioned taste
Sato, T., and Beidler, L. M. (1997). Broad tuning of rat taste cells aversions: generalization to taste mixtures. Physiol. Behav.
to four basic taste stimuli. Chem. Senses 22:287–293. 32:983–989.
Schiffman, S. S., and Erickson, R. P. (1980). The issue of primary Smith, D. V., and Travers, J. B. (1979). A metric for the breadth
tastes versus a taste continuum. Neurosci. Biobehav. Rev. of tuning of gustatory neurons. Chem. Sens. Flav. 4:215–229.
4:109–117. Smith, D. V., Bealer, S. L., and Van Buskirk, R. L. (1978).
Schwartz, G. J., and Grill, H. J. (1984). Relationships between Adaptation and recovery of the rat chorda tympani response to
taste reactivity and intake in the neurologically intact rat. NaCl. Physiol. Behav. 20:629–636.
Chem. Senses 9:249–272. Smith, D. V., Steadman, J. W., and Rhodine, C. N. (1975). An
Scott, T. R. (1987a). Coding in the gustatory system. In analysis of the time course of gustatory neural adaptation in
Neurobiology of Taste and Smell, T. E. Finger and W. L. Silver the rat. Am. J. Physiol. 229:1134–1140.
(Eds.). John Wiley, New York, pp. 355–378. Smith, D. V., Travers, J. B., and Van Buskirk, R. L. (1979).
Scott, T. R. (1987b). The Janus head of taste. In Olfaction and Brainstem correlates of gustatory similarity in the hamster.
Taste IX, S. D. Roper and J. Atema (Eds.). New York Brain Res. Bull. 4:359–372.
Academy of Sciences, New York, pp. 600–601. Smith, D. V., Van Buskirk, R. L., Travers, J. B., and Bieber, S. L.
Scott, T. R., and Giza, B. K. (1990). Coding channels in the taste (1983a). Coding of taste stimuli by hamster brainstem neu-
system of the rat. Science 249:1585–1587. rons. J. Neurophysiol. 50:541–558.
758 Smith and Scott

Smith, D. V., Van Buskirk, R. L., Travers, J. B., and Bieber, S. L. tion and rejection of sapid stimuli: an electromyographic
(1983b). Gustatory neuron types in the hamster brainstem. analysis in the rat. Behav. Brain Res. 25:233–246.
J. Neurophysiol. 50:522–540. Travers, S. P. (1993). Orosensory processing in neural systems of
Smith, D. V., Liu, H., and Vogt, M. B. (1994). Neural coding of the nucleus of the solitary tract. In Mechanisms of Taste
aversive and appetitive gustatory stimuli: interactions in the Transduction, S. A. Simon and S. D. Roper (Eds.). CRC Press,
hamster brain stem. Physiol. Behav. 56:1189–1196. Boca Raton, FL, pp. 339–394.
Smith, D. V., Li, C.-S., and Davis, B. J. (1998). Excitatory and Travers, S. P., and Norgren, R. (1991). Coding the sweet taste in
inhibitory modulation of taste responses in the hamster brain- the nucleus of the solitary tract: differential roles for anterior
stem. In Olfaction and Taste XII: An International tongue and nasoincisor duct gustatory receptors in the rat.
Symposium, C. Murphy (Ed.). New York Academy of J. Neurophysiol. 65:1372–1380.
Sciences, New York, pp. 450–456. Travers, S. P., and Norgren, R. (1995). Organization of orosen-
Smith-Swintosky, V. L., Plata-Salaman, C. R., and Scott, T. R. sory responses in the nucleus of the solitary tract of the rat.
(1991). Gustatory neural coding in the monkey cortex: stimu- J. Neurophysiol. 73:2144–2162.
lus quality. J. Neurophysiol. 66:1156–1165. Travers, S. P., and Smith, D. V. (1984). Responsiveness of
Spector, A. C., and Grill, H. J. (1992). Salt taste discrimination neurons in the hamster parabrachial nuclei to taste mixtures.
after bilateral section of the chorda tympani or glossopharyn- J. Gen. Physiol. 84:221–250.
geal nerves. Am. J. Physiol. 263:R169–R176. Travers, S. P., Pfaffmann, C., and Norgren, R. (1986).
Spector, A. C., and St John, S. J. (1998). Role of taste in the Convergence of lingual and palatal gustatory neural activity in
microstructure of quinine ingestion by rats. Am. J. Physiol. the nucleus of the solitary tract. Brain Res. 365:305–320.
274:R1687–1703. Van Buskirk, R. L., and Smith, D. V. (1981). Taste sensitivity of
Spector, A. C., Travers, S. P., and Norgren, R. (1993). Taste hamster parabrachial pontine neurons. J. Neurophysiol.
receptors on the anterior tongue and nasoincisor ducts of rats 45:144–171.
contribute synergistically to behavioral responses to sucrose. Verhagan, J., Giza, B. K., and Scott, T. R. (1999). Taste in the rat
Behav. Neurosci. 107:694–702. thalamus. Neurosci. Abstr. 25:2184.
Spector, A. C., Guagliardo, N. A., and St. John, S. J. (1996). Vogt, M. B., and Mistretta, C. M. (1990). Convergence in mam-
Amiloride disrupts NaCl versus KCl discrimination perfor- malian nucleus of the solitary tract during development and
mance: implications for salt taste coding in rats. J. Neurosci. functional differentiation of salt taste circuits. J. Neurosci.
16:8115–8122. 10:3148–3157.
Spector, A. C., Markison, S., St. John, S. J., and Garcea, M. Wang, L., and Bradley, R. M. (1993). Influence of GABA on neu-
(1997). Sucrose vs. maltose taste discrimination by rats rons of the gustatory zone of the rat nucleus of the solitary
depends on the input of the seventh cranial nerve. Am. tract. Brain Res. 616:144–153.
J. Physiol. 272:R1210–R1218. Wang, L., and Bradley, R. M. (1995). In vitro study of afferent
St. John, S. J., and Smith, D. V. (2000). Neural representation of synaptic transmission in the rostral gustatory zone of the rat
salts in the rat solitary nucleus: brainstem correlates of taste nucleus of the solitary tract. Brain Res. 702:188–198.
discrimination. J. Neurophysiol., 84:628–638. Winograd, S. (1963). Redundancy and complexity of logical ele-
St. John, S. J., and Spector, A. C. (1998). Behavioral discrimina- ments. Inform. Control. 5:177–194.
tion between quinine and KCl is dependent on input from the Woolston, D. C., and Erickson, R. P. (1979). Concept of neuron
seventh cranial nerve: implications for the functional roles of types in gustation in the rat. J. Neurophysiol. 42:1390–1409.
the gustatory nerves in rats. J. Neurosci. 18:4553–4362. Yamaguchi, S. (1979). The umami taste. In Food Taste Chemistry,
St. John, S. J., Markison, S., Guagliardo, N., Hackenberg, T. D., and J. C. Boudreau (Ed.). American Chemical Society,
Spector, A. C. (1997). Chorda tympani nerve transection and Washington, DC, pp. 33–51.
selective desalivation differentially disrupt two-lever salt dis- Yamamoto, T. (1989). Role of the cortical gustatory area in taste
crimination performance in rats. Behav. Neurosci. 111:450–459. discrimination. In Neural Mechanisms in Taste, R. H. Cagan
Sweazey, R. D., and Smith, D. V. (1987). Convergence onto ham- (Ed.). CRC Press, Boca Raton, FL, pp. 197–219.
ster medullary taste neurons. Brain Res. 408:173–184. Yamamoto, T., Yuyama, N., and Kawamura, Y. (1981). Cortical
Tonosaki, K., and Funakoshi, M. (1984). Intracellular taste cell neurons responding to tactile, thermal and taste stimulation of
responses of mouse. Comp. Biochem. Physiol. A 78:651–656. the rat’s tongue. Brain Res. 221:202–206.
Travers, J. B. (1988). Efferent projections from the anterior Yamamoto, T., Yayuma, N., Kato, T., and Kawamura, Y. (1985).
nucleus of the solitary tract of the hamster. Brain Res. Gustatory responses of cortical neurons in rats. III. Behavioral
455:283–294. vs. neural correlates of taste quality. J. Neurophysiol.
Travers, J. B., and Norgren, R. (1983). Afferent projections to the 53:1370–1386.
oral motor nuclei in the rat. J. Comp. Neurol. 220:280–298. Yamamoto, T., Matsuo, R., Kiyomitsu, Y., and Kitamura, R.
Travers, J. B., and Smith, D. V. (1979). Gustatory sensitivities in (1989). Taste responses of cortical neurons in freely ingesting
neurons of the hamster nucleus tractus solitarius. Sens. Proc. rats. J. Neurophysiol. 61:1244–1258.
3:1–26. Yaxley, S., Rolls, E. T., Sienkiewicz, Z. J., and Scott, T. R. (1985).
Travers, J. B., Grill, H. J., and Norgren, R. (1987). The effects of Satiety does not affect gustatory activity in the nucleus of the
glossopharyngeal and chorda tympani nerve cuts on the inges- solitary tract of the alert monkey. Brain Res. 347:85–93.
36

Development of the Taste System: Basic Neurobiology

Charlotte M. Mistretta
University of Michigan, Ann Arbor, Michigan, U.S.A.

David L. Hill
University of Virginia, Charlottesville, Virginia, U.S.A.

I. INTRODUCTION fungiform papilla and surrounding lingual epithelium, but

not taste buds; (3) the petrosal, innervating taste buds in
Taste receptor cells reside in epithelial specializations, the posterior folds of the foliate papillae, the circumvallate
taste buds, which are located primarily in the tongue, soft papillae and taste buds, and surrounding posterior lingual
palate, and epiglottis in mammals (Mistretta, 1991; epithelium; and (4) the nodose, innervating epiglottal taste
Mistretta and Hill, 1995) (see Chapter 32). On the tongue buds and epithelium (Mistretta and Hill, 1995). Within the
taste buds reside solely in specialized sensory structures, the brainstem, central projections of the taste and taste-related
gustatory papillae, and on the soft palate and epiglottis taste nerves from the sensory ganglia reflect the orderly
buds occur in high density in specialized regions. Thus, arrangement of taste buds in oral-pharyngeal regions.
mammalian taste buds are not randomly scattered across the The emphasis on specialized distributions, or patterns, of
mucosae of the mouth and pharynx, but rather are concen- receptors and neural projections suggests a high degree of
trated in specific locations. This provides a basis for func- order in the taste system, and we use this concept of ordered
tional specialization that contributes to optimal processing arrangements as a broad focus in this revision of our chap-
in this life-essential sensory system, the gustatory sense. ter from the first edition of this book (Doty, 1995).
Seated between the peripheral receptors and central ner- Reflecting the models and approaches used in recent stud-
vous system are the cranial, sensory ganglia composed of ies, discussion is limited primarily to data from the rodent
neurons that extend fibers to synapse with taste bud cells taste system, rat and mouse, and to taste buds and papilla
within peripheral oral/facial tissues (Fig. 1). To establish organs on the tongue. Other comprehensive reviews of the
taste pathways during development, neurites extend from developmental biology of the sense of taste, with discussion
neurons within the ganglia to gustatory organs in the across species, can be found in Mistretta (1991), Hill (2001),
tongue in the peripheral nervous system and to the brain- and Mistretta and Hill (1995). There is also an extensive
stem in the central nervous system. Reflecting the central body of experimental results demonstrating that peripheral
role of the ganglia in the establishment of taste sensation, and central gustatory function and morphology is especially
this chapter is organized to discuss the ganglia first, then susceptible to environmental manipulations during early
the peripheral taste organs, and lastly the central neural development. The reader is referred to recent reviews of
taste regions. Relevant ganglia are (1) the geniculate, these topics by Stewart et al. (1997) and Hill (2001).
innervating taste buds on the soft palate and in fungiform For referring to embryonic and postnatal ages, through-
papillae on the tongue; (2) the trigeminal, innervating out the chapter we use the day that the rodent dam is sperm

759
760 Mistretta and Hill

Figure 1 Diagram illustrating cells of the trigeminal, geniculate, and petrosal ganglia that extend neurites during development to
innervate papilla, taste bud, and lingual epithelial target organs in the tongue and to synapse centrally with neurons in taste nuclei in the
brainstem. The sensory ganglia thus have a principal role in establishing taste circuits from peripheral target organs to central neural
structures.

positive as gestational day 0 (E0). The day of birth is pharyngeal endoderm through signaling actions of BMP-7
referenced as postnatal day 0 (P0). (Begbie, 1998). In contrast, the trigeminal placodes are
induced by local tissue in the midbrain to rostral hindbrain
region. Basic helix-loop-helix neurogenic factors (bHLH)
are expressed in precursors of placode-derived sensory
II. THE SENSORY GANGLIA INNERVATING
neurons. Specifically, neurogenin 2 is essential for devel-
LINGUAL GUSTATORY PAPILLAE AND
opment of sensory ganglia derived from epibranchial
TASTE BUDS AND CENTRAL GUSTATORY
placodes, including the geniculate and petrosal (Fode et
STRUCTURES
al., 1998). Developing trigeminal neurons depend on neu-
rogenin 1 function (Ma et al., 1998). In each ganglion,
A. When and How Do the Ganglia Develop, and other bHLH factors are subsequently expressed including
What Are the Known Regulatory Elements? Math3, NeuroD, and NSCL1 (Fode et al., 1998; Ma et al.,
1998). Thus, the ganglia that provide sensory innervation
The geniculate, petrosal, and trigeminal ganglia are collec- to the tongue have different placodal derivations and dif-
tions of neurons that extend fibers to innervate specific ferent early molecular regulators. These differences serve
lingual targets peripherally and localized groups of to highlight a distinction in sensory functions—the
neurons in the solitary nucleus centrally. These ganglia chemosensory nature of geniculate and petrosal ganglia
arise from placodes, local thickenings of cells that will and the somatosensory nature of the trigeminal.
delaminate from the original placode location, migrate, The geniculate, petrosal, and trigeminal ganglia have
and aggregate to form the presumptive ganglia (Baker and begun to develop on E8.5 in mouse (Fode et al., 1998; Ma
Bronner-Fraser, 2001). The geniculate and petrosal ganglia et al., 1998) and E9.5–10 in rat (Altman and Bayer, 1982;
arise from epibranchial placodes at pharyngeal pouches 1 White et al., 1995), and the earliest axons emerge from the
and 2, whereas the trigeminal ganglion derives from two mouse trigeminal ganglion at E9.5 (Davies, 1997).
more rostrally located placodes, the opthalmic and the Ganglion development thus is well underway for a few
maxillo-mandibular, near the midbrain-hindbrain junction days in advance of emergence of the tissue swellings that
(Graham and Begbie, 2000). Studies with chick embryo subsequently will coalesce to form the tongue (lingual
indicate that the epibranchial placodes are induced by swellings are apparent on gestational day 11.5 in mouse;
Taste System Development 761

day 13 in rat) (Kaufman, 1992; Mistretta, 1972). various neurotrophins are ganglion-specific and tempo-
Therefore, the initial formation of ganglion cells is not rally restricted (Snider, 1994).
directed by factors in the tongue (Fig. 2). Only by gesta- An important influence of neurotrophins in sustaining
tional day 14–15 in the rat have neurites extended into geniculate and nodose/petrosal ganglion cells was demon-
early taste papillae to potentially benefit from target- strated in transgenic mice that had a deletion of the gene
derived molecules that could support, maintain, and regu- encoding BDNF, with an approximate 60% loss of neurons
late differentiation of ganglion cells (Mistretta, 1998). in these ganglia (Ernfors et al., 1994; Jones et al., 1994). In
Once neurites from the ganglia have reached the contrast, only about 20% of trigeminal ganglion cells were
periphery, what molecules in lingual targets might interact lost in the null mutants. A synthesis of published literature on
with the growing nerves to influence development of mutant mice with deletion of genes encoding neurotrophin
ganglion cells? Among those families receiving much molecules makes clear the distinctive dependence of the
attention in current literature are the neurotrophins. These geniculate ganglion, a gustatory ganglion, versus the trigem-
soluble proteins include nerve growth factor (NGF), brain- inal ganglion, which is nongustatory, subserving somatosen-
derived growth factor (BDNF), neurotrophin-3 (NT3), sation and pain (Table 1). Unfortunately, data for study of the
and neurotrophin-4/5 (NT4/5) (Davies, 1997; Farinas and petrosal ganglion alone are not available; rather the
Reichardt, 1996). The neurotrophins have demonstrated petrosal/nodose ganglia have been evaluated as a complex.
roles in ganglion cell survival, maintenance, and morpho- However, the primary dependence of the petrosal/nodose on
logical and functional differentiation, and effects of the BDNF is clear from the literature (Table 1).

Figure 2 Photomicrographs of sagittal sections through embryonic rat heads at gestational days 12 (E12, top two images) and 13 (E13,
bottom two images). Trigeminal (tg), geniculate (gg), and petrosal (pg) ganglia already have formed by E12 (top, left), and neurites extend
into the maxillary (mx) and mandibular (mn) target zones (top, right). The embryonic tongue has not yet formed at E12. By E13, ganglia
have more extensive neurite branches (bottom, left), and some neurites have reached the base of the embryonic tongue (bottom, right;
hatched line demarcates base of tongue). (Data from Mbiene and Mistretta, 1997.)
762 Mistretta and Hill

Table 1 Ganglion Reduction in Neurotrophin Null Mutant Mice Presented as Estimate of Remaining Ganglion Cell Numbers or Volume
(% In Mutant Relative to Wild-Type Mice)
% Ganglion remaining in null mutantsa
BDNF NT4 NT3 NGF BDNF/NT4 BDNF/NT3 BDNF/NT4/NT3

Geniculate 55 50 65 na 5 35 0
Petrosal/Nodoseb 45 40 60 na 15 35 5
Trigeminal 70 98 35 30 70 25 10

na
not available.
a Percentages have been averaged across published studies and then rounded to the nearest 5% for ease in comparing across ganglia and

neurotrophins.
b Studies of petrosal ganglion as a separate entity in neurotrophin mutant mice are not published; data are for combined petrosal/nodose

ganglia.
Sources: Conover et al., 1995; Ernfors et al. 1994; Farinas and Reichardt, 1996; Jones et al., 1994; Liebl et al., 1997; Liu and Jaenisch,
2000; Liu et al., 1995; Snider, 1994.

There is substantial understanding about the role of discussued in later sections of the chapter. The expression
neurotrophins in development of the rodent trigeminal of the neurotrophins themselves in sensory ganglia, for
ganglion. Based on observations in mouse trigeminal, it is example, BDNF in rat geniculate at E14.5 (Schecterson
proposed that ganglion neurons survive and extend axons and Bothwell, 1992) and NT3 in mouse at E18 (Ernfors et
independently of neurotrophins during initial development al., 1992), raises questions about autocrine support within
(Davies, 1997). Later, ganglion neurons are dependent on the ganglia and anterograde transport to target tissues.
trophic support from tissues through which their axons
extend and then from target organs. For neuron survival,
B. Neurites Growing to the Periphery from Ganglia:
trigeminal dependence on neurotrophins shifts from a
How Does Early Innervation of the Tongue
requirement for BDNF or NT3 at E10.5–11.5 in mouse, to
Proceed?
NGF after E12.5 (Buchman and Davies, 1993; Davies,
1997, Paul and Davies, 1995). Whereas all three of the
A direct demonstration for neurotrophins in guiding nerve
high-affinity Trk receptors for neurotrophins, Trks A, B,
outgrowth into the embryonic tongue has not been made
and C, are expressed within the embryonic trigeminal gan-
with in vivo or in vitro systems. However, from detailed
glion, neurons with specific Trks develop in two waves
studies of early lingual innervation in embryonic rat
(Huang et al., 1999). Trk B–and Trk C–expressing neurons
tongue, there is substantial understanding of basic regula-
are generated first, with peak numbers at E11.5, and Trk
tory principles in development of tongue pathways. To test
A–expressing neurons later, with peak numbers at E13.5.
the hypothesis that axons growing into the rat tongue
These investigators have also demonstrated that expression
distribute homogeneously and then redirect to densely
of the Trk receptors is confined to trigeminal neurons and
innervate gustatory papillae, the lingual/chorda tympani,
is not in neuronal precursors. Also, by E13.5 there is no co-
glossopharyngeal, and the hypoglossal motor nerve fibers
expression of the Trk receptors in individual trigeminal
were systematically traced in serial sections at different
ganglion neurons.
embryonic ages (Mbiene and Mistretta, 1997). With this
Data on the geniculate ganglion and neurotrophin
study, a new set of working principles for development of
dependence are not extensive. In the rat, most or all genic-
tongue innervation was defined:
ulate cells express Trk B mRNA at E13 through E18
(Ernfors et al., 1992). Most cells also express TrkC at E13, 1. Tongue nerves do not first grow in widely distributed
but fewer cells express this trk at E16–E18. Neurite territories and then retract or redirect to particular
outgrowth is supported to some extent by multiple neu- tongue areas and to taste papillae. Points of entry and
rotrophins at E13–E16; however, outgrowth is more dense initial tongue pathways are restricted and precise
in response to BDNF and NT4, compared to NT3 or NGF from time of tongue formation through morphogene-
(Rao et al., 1997; Rochlin et al., 2000). Neurotrophins have sis, so there are distinctive lingual territories for each
a clear and specific role in functional differentiation of nerve (Fig. 3).
embryonic rat geniculate (Al-Hadlaq et al., 2001, 2002) 2. Although sensory fibers of the chorda/lingual nerve
and trigeminal ganglia (Grigaliunas et al., 2002), as are within the tongue earlier than the hypoglossal
Taste System Development 763

papillae during the periods just preceding, and during,

motor fibers, sensory and motor nerves distribute nerve ingrowth (Nosrat and Olson, 1995).
independently of each other, neither following a “pio-
neering” path prepared by the other. Recent experiments in mice that express a green fluo-
3. Anterior and posterior tongue zones are discrete for rescent protein in growing nerves have demonstrated that
nerve entry into and distribution within the lingual neurotrophins in vivo guide sensory nerves from dorsal
tissues. root ganglia into the limb bud (Tucker et al., 2001).
4. Once fungiform and circumvallate papillae have Applying such models to studies of embryonic tongue will
formed, through nerve-independent processes eventually define precise roles of neurotrophins in direct-
(Mbiene et al., 1997), they become densely inner- ing lingual innervation.
vated; in contrast, surrounding nonpapilla epithelium Numerous other molecules must play a role in the
does not have periodic clusters of dense innervation. stereotypic growth of nerves into the embryonic tongue.
This latter principle provides indirect evidence for Chemorepellants, including those in the semaphorin fam-
the potential role of papilla-derived molecules in ily, have an apparent role in directing the early growing
attracting sensory nerves once they are within the axons away from midline facial structures, from midline
tongue. Indeed it has been shown that neurotrophin tongue, and reportedly away from early tongue epithelium
expression is most intense in developing rat gustatory (Rochlin and Farbman, 1998; Rochlin et al., 2000).

Figure 3 Photomicrographs of sagittal sections through the embryonic rat tongue at gestational day 15 (E15). Figures A through D cor-
respond to sections taken from more lateral (A) to progressively more medial (D) tongue regions. Anterior tongue faces left. Points of
entry and lingual pathways of the chorda/lingual (ch/l), hypoglossal (hy), and glossopharyngeal (gl) nerves are illustrated. Within the
body of the tongue, separate distributions of all nerves are maintained. Near the midline (D) the tongue is essentially devoid of nerves
with the exception of a dense plexus of fibers under the developing circumvallate papilla in posterior tongue. (Data from Mbiene and
Mistretta, 1997.)
764 Mistretta and Hill

The presence of C-Kit in taste bud cells of embryonic between E12 and E15 with a peak at E13 (Altman and
circumvallate and foliate papillae and in nerve fibers Bayer, 1982). This suggests that neurons are produced
within the core of these papillae suggests interactions that within the nucleus well in advance of initial entry of taste
might include tropic or tropic roles (McLaughlin, 2000). fibers into the tract (Mistretta and Hill, 1995).
C-Kit is one of the protein tyrosine kinase receptors, mol- Trigeminal axons grow into the brainstem as early as
ecules with known involvement in organogenesis in vari- E12 in rat (Mbiene and Mistretta, 1997). Using a slice
ous systems. preparation with ganglion and central and peripheral tar-
gets, it was shown that NGF promotes axon elongation
without branching in the brainstem, whereas NT3 pro-
C. Neurites Growing Centrally from Ganglia: Initial motes formation of short axon collaterals (Ulupinar et al.,
Innervation and Establishment of the Nucleus of 2000). However, levels of NGF and NT3 mRNA are appar-
the Solitary Tract ently very low in developing rodent hindbrain (Davies,
1997), so in vivo neurotrophin effects on ganglion neurite
In the rat, the peak production of geniculate ganglion neu- growth centrally is far from clear.
rons that will send central projections into the solitary tract Interestingly, in DRG neurons studied in NGF/BAX
is on day E12 (Altman and Bayer, 1982), somewhat in double null mice, a requirement for NGF signaling in
advance of the petrosal and nodose ganglia, with a peak peripheral sensory neuron growth and differentiation was
production at E13 (Fig. 4). Fibers within the solitary tract established, whereas NGF was apparently not required for
are identified first at E17, whereas postsynaptic, second- extension of central axon projections (Patel et al., 2000). If
order taste neurons within the solitary nucleus are produced this result generalizes to other sensory populations, it indi-
cates that peripheral and central ganglion extensions have
different molecular dependencies for extension and growth.

III. PERIPHERAL TARGETS OF GANGLIA:

TONGUE, PAPILLAE, AND TASTE BUDS

Just as sensory ganglion neurons develop and begin to dif-

ferentiate in advance of target development, and therefore,
independently of target-derived support, so too the tongue
and lingual papillae develop without initial neural depen-
dence (Farbman and Mbiene, 1991; Mbiene et al., 1997).
With continued papilla morphogenesis, and eventual dif-
ferentiation of taste buds within papillae, nerve-target
interactions ensue, and the nature and extent of these inter-
actions is an area of active current research. Because much
of the work on papillae and taste bud development derives
from studies of rodents, we will summarize these
processes for these species. However, there have been
major advances in understanding how the peripheral gus-
tatory structures develop in other species, especially
human (e.g., see Bradley, 1972; Bradley and Stern, 1967;
Ganchrow and Ganchrow, 1985, 1987; Witt and Reutter,
1997, 1998). There are few detailed reports of the mor-
phological development of foliate papillae (Harada et al.,
2000; Oakley, 1988); thus, morphogenesis of fungiform
Figure 4 Diagram, based on rat embryo, of times of peak pro-
and circumvallate papillae only will be discussed.
duction of neurons in sensory ganglia that send processes into
brainstem nuclei, including the nucleus of the solitary tract (SL).
Peak production in the geniculate ganglion is at 12 days of gesta- A. Formation of the Tongue and Timing of Lingual
tion (E12), somewhat in advance of that in the petrosal ganglion Papilla Development
at E13. VG, trigeminal ganglion; VIII Gs, spiral ganglion; VIII Gv,
vestibular ganglion; IX-X Gs, superior or proximal, nongustatory The rodent tongue emerges from a series of three swellings
ganglia of nerves IX and X (From Mistretta and Hill, 1995.) of mesenchyme tissue in the floor of the mandible (Mbiene
Taste System Development 765

et al., 1997; Mistretta, 1972). By E14 in rat and about E13 or 14, and temporal information is retained because
E12.5 in mouse, the three swellings have merged to form papillae form at an equivalent of E15 in vivo, after 2 days
the spatulate tongue, followed by extensive growth prena- in culture from E13, or one day in culture from E14. The
tally (Kaufman, 1992; Mistretta, 1972). Fungiform papilla patterned rows, characteristic of tongues obtained from
formation in rat begins at E14 (gestation
21 days), with older fetuses, are also clearly seen on cultured tongues. In
appearance of small protuberances on the dorsal tongue addition to histological integrity, papillae that form on
surface (Mbiene et al., 1997; Mistretta, 1972). By E15, the tongues in organ culture retain expression of neurotrophins
emerging papillae are clearly evident in that they are ele- similar to in vivo stages (Nosrat et al., 2001). Since the cul-
vated from surrounding epithelia, and by E16, fungiform tured tongue is dissected without sensory ganglia or intact
papillae are distinct in rostral to caudal rows. The patterned trigeminal or gustatory nerves, it is clear that papillae
orientation of rows is maintained at adulthood, although (Farbman and Mbiene, 1991; Mbiene et al., 1997) and
the pattern is much less apparent amid the background of their signature neurotrophins (Nosrat et al., 2001) form ini-
dense filiform papillae that appear on the tongue at about tially without these nerves.
the time of birth. Because of the patterned arrangement, Whereas it can be concluded that nerves do not direct
fungiform papillae are ideal targets to study developmen- papilla induction and patterning, there are only beginning
tally related cellular and molecular processes in organo- indications of the gene products that might be central play-
genesis. ers (Mistretta, 1998). Several morphogens have been iden-
Histologically, fungiform papillae are a cluster of tified in the early mouse tongue and forming papillae
epithelial cells formed from stratification and elongation of (Bitgood and McMahon, 1995; Dassule and McMahon,
cells in more ventral layers of the early embryonic tongue 1998; Hall et al., 1999a; Jung et al., 1999) (Table 2).
epithelium (Bradley, 1972; Farbman, 1965; Farbman and Preliminary data provide similar results for the sonic
Mbiene, 1991). The characteristic fungiform papilla mor- hedgehog protein (Shh) in papilla development in rat
phology is formed by the development of an epithelial tongue (Hall et al., 1999b). Just as the Shh, BMP, and FGF
thickening, with subsequent epithelial-mesenchymal inter- families are key morphogens in tooth, hair, and feather for-
actions to form the connective tissue core surrounded by mation, these molecules also will undoubtedly be impor-
an annular downgrowth of epithelum (Bradley, 1972). As tant in papilla formation (Mistretta, 1998).
the surrounding epithelium continues its ventral growth, Direct demonstration for a role of the Shh signal trans-
the papilla becomes further raised on the tongue surface. duction pathway in papilla formation and patterning has
At its mature morphological stage, layers of epithelial cells recently emerged (Mistretta et al., 2000, 2002). Using a
cover the papilla. specific blocker of the Shh pathway, cyclopamine, in organ
Circumvallate papillae located on the posterior tongue cultures of embryonic rat tongue, a doubling of fungiform
have distinct and unique patterning that, like fungiform papillae on the tongue was observed, compared to control
papillae, originate early in development (Mistretta, 1991). cultures (Fig. 5). Because fungiform papillae were distrib-
In the adult rat, the single circumvallate located on the uted even on posterior tongue surrounding the circumval-
midline is relatively large, is located most posteriorly of all late papilla, the results also demonstrate a broad compe-
lingual papillae, and is surrounded by a trench that invagi- tence for lingual epithelium to generate fungiform papilla.
nates into the tongue. Initially, the circumvallate papilla This novel finding presents new modes of thinking about
lacks a surrounding trench, which emerges later as the the tongue epithelium and its developmental competencies
epithelium grows down along the connective tissue core and restrictions.
(Bradley, 1972; Mistretta, 1991). As noted for fungiform Other molecules that play a role in nerve/target interac-
papillae, the rat circumvallate papilla forms between E14 tions are expressed in specific papilla regions at mid-
and E15. Therefore, morphogenesis of fungiform and cir-
cumvallate papillae occurs at about the same time, Table 2 Expression of “Patterning” Molecules in Embryonic
although adult papilla morphology is very different Mouse Tongue Based on In Situ Hybridization Data
between the two papilla types.
Shh Ptc Gli Bmp2 Bmp4 Fgf8
B. Fungiform Papilla Induction and Differentiation E11–12      
E13   
Organ culture studies have verified that innervation is not E14     
necessary for papilla formation and patterned distribution E15–16     
(Mbiene et al., 1997). Fungiform papillae grow and differ- Source: Bitgood and McMahon, 1996; Dassule and McMahon,
entiate when entire rat tongues are cultured beginning at 1998; Hall et al., 1999; Jung et al., 1999.
766 Mistretta and Hill

Figure 5 Scanning electron micrographs of embryonic whole tongue cultures, begun at gestational day 14 and maintained in organ
culture for 2 days. Fungiform papillae develop in the usual pattern in tongues cultured with standard medium (A). However, in tongues
cultured in medium with addition of the steroidal alkaloid cyclopamine, which disrupts the sonic hedgehog signaling pathway, the num-
ber of fungiform papillae doubles (B). Also, fungiform papillae develop on posterior tongue in front of the circumvallate papilla. In A
and B the circumvallate papilla is demarcated with an arrow. (Data from Mistretta et al., 2000, 2002.)

gestational stages of development. In rat fungiform and to be permanent in that papillae do not regenerate at adult-
circumvallate papilla from E14-E16, BDNF is localized to hood (Sollars and Bernstein, 2000); however, papillae do
papilla regions that will later form taste buds (Nosrat and regenerate when chorda tympani sectioning occurs at adult-
Olson, 1995). In contrast, NT-3 is found primarily in the hood (St. John et al., 1995). The early developmental effect
mesenchymal core and in nongustatory epithelium of the is most likely attributed to an induced loss of geniculate
papillae (Nosrat et al., 1996). ganglion cells in young rats due to nerve section (S. Sollars,
personal communication); subsequently, the limited num-
C. Papilla Maintenance ber of surviving neurons may not support the postnatal
papillae. In contrast, only sectioning the lingual branch of
Although the gustatory papillae develop in organ culture the trigeminal nerve results in a transient degeneration of
without intact sensory innervation (Farbman and Mbiene, 44% of fungiform papillae in postnatal day 10 animals, fol-
1991; Mbiene et al., 1997), after 6 days in whole tongue lowed by a reappearance of papillae within 50 days after
cultures initiated on E13 or E14, large numbers of fungi- nerve cut (Guagliardo et al., 1999). These findings collec-
form papillae are no longer sustained (Mbiene et al., tively show that maintenance of the postnatal fungiform
1997). It appears, therefore, that whereas the early process papilla is dependent on innervation by both the chorda tym-
of papilla induction is not nerve dependent, long-term pani and trigeminal nerves and that there is likely an inter-
papilla maintenance is. play between these nerves to maintain papilla structure.
Further evidence for the role of innervation in papillae
maintenance comes from nerve cut studies in early postna- D. Taste Bud Induction
tal rats. Sectioning the combined chorda tympani nerve and
lingual branch of the trigeminal nerve on postnatal day 1 is While papilla induction clearly does not require innerva-
followed by degeneration of papillae over the following 21 tion by gustatory and trigeminal nerves, there is consider-
days (Nagato et al., 1995). Indeed, sectioning only the able controversy concerning the role of these nerves in
chorda tympani in postnatal day 10 rats causes 65% of early taste bud formation. A long-standing view held that
fungiform papillae to degenerate within 30 days, even lingual taste buds required innervation for formation
though the lingual nerve is left intact (Sollars and because nerves reached the dorsal papilla epithelium just
Bernstein, 2000; Sollars et al., 1996, 2002). Of the fungi- before the appearance of rudimentary taste buds (e.g.,
form papillae that remain, nearly 80% have unusual mor- Farbman, 1965). Thus, due principally to descriptions of
phologies (Sollars and Bernstein, 2000). The morphologi- timing, nerves have been considered essential to induce
cal changes after early chorda tympani nerve section appear taste buds.
Taste System Development 767

Recent studies on both nonmammalian and mammalian axolotl studies would be expected to yield results that speak
taste bud development, however, have reassessed mecha- to different mechanisms from those in mammals (Brockes,
nisms underlying taste bud induction. In classical embry- 1997). Furthermore, taste bud development in amphibia is
ological experiments in salamander, Stone (1933, 1940) complex and highly specialized. The early barrel-shaped,
reported that taste buds differentiate independently from axolotl taste buds from the larval stage are transformed
nerve fibers. In extensions of these experiments, grafts of and/or replaced during metamorphosis into taste buds with
presumptive oropharyngeal tissue in axolotl that were a disc-like, wider cell mass (Takeuchi et al., 1997). In fact,
placed ectopically on the trunk of host embryos prior to a separate, more rostral tongue structure, with wide disc-
innervation and prior to appearance of taste buds formed shaped taste buds on specialized papillae, grows to
well-differentiated taste buds, even in the absence of neu- predominate over and replace the larval tongue with its
rites (Barlow et al., 1996). Tissue culture studies in which population of smaller, oval taste buds that eventually disap-
the presumptive axolotl oropharyngeal region was cultured pear. Even the innervation of the two tongues and taste bud
prior to innervation further supported the idea that taste populations is not clearly known between nerves VII and
bud development occurs independently of innervation IX. The transformation and rearrangement of taste organs
(Barlow et al., 1996). Experiments with trkB / knock- on the salamander tongue regions suggest that a totally dif-
out mice, which lack the tyrosine kinase receptor for brain- ferent set of taste organ structures could be addressed in the
derived neurotrophic factor, provided support for the idea studies of Stone (1933, 1940) and Barlow et al. (1996)
that taste bud induction does not require innervation. In the compared to the postmetamorphosis organism.
absence of trkB, knockout mice lose the geniculate gan- Whereas chemosensory scientists generally think that
glion neurons that usually innervate taste buds in fungi- mammalian taste buds in all oropharyngeal locations will
form papillae. Yet taste buds apparently develop, although have identical basic mechanisms for induction, there may
it is important to note that cytology is atypical, and size indeed be variations between, for example, the epiglottal
and numbers of these taste buds have not yet been speci- taste buds and those on the tongue. Mistretta (1991, 1998)
fied (Fritzsch et al., 1997). has emphasized that on the mammalian tongue, the gusta-
Other studies support the more long-standing neural tory papillae are limiting elements for taste bud formation.
dependence view. For example, if the glossopharyngeal This makes interpretation of results from knockout mice,
nerve is avulsed in newborn rat long before all taste buds where papillae are morphologically disrupted or not
have begun to form in the circumvallate papilla, the full present at all, difficult to interpret. There also has been
number of taste buds is never acquired in the denervated a tendency to use terminology somewhat loosely in some
papilla (Hosley et al., 1987). In addition, fungiform and cir- recent papers, so that the process of taste bud “induction”
cumvallate taste bud numbers in knockout mice missing has sometimes been generalized to all of taste bud
genes for BDNF are severely reduced, and this is highly “development.”
correlated with reduced numbers of innervating neurons Key experiments and cautious data interpretation will
(Jones et al., 1994; Liu, et al., 1995; Mistretta et al., 1999). be necessary to determine mechanisms of taste bud induc-
This result is consistent with a neural dependence view of tion in mammals. This will be a tall order for a sensory
taste bud induction in which decreased numbers of inner- receptor organ as complex as the taste bud, in which there
vating neurons should correspond directly with decreased is as yet no identified stem cell population or markers.
numbers of taste buds. However, since mutant mice were
examined only postnatally, after taste buds were induced, it E. Taste Bud Formation, Numbers, and Size
is not clear whether the reduced number of taste buds
reflects an initial failure of formation or whether taste buds Taste buds are added to the oropharynx both pre- and post-
develop in a rudimentary form and then quickly degenerate. natally in some mammalian species, such as sheep and
Using intra-amniotic injection in pregnant mice of the human (Mistretta, 1991). In contrast, the full complement
neurotoxin -bungarotoxin, another demonstration of of taste bud number in rat and hamster is acquired gradu-
neural dependence in the taste system was observed ally during postnatal development (Belecky and Smith,
(Morris-Wiman et al., 1999). Taste buds were not found in 1990; Harada et al., 2000; Hosley and Oakley, 1987;
fungiform and circumvallate papillae in neurotoxin- Mistretta, 1972), in each lingual papilla type (Fig. 6). For
injected animals, compared to controls, and indeed num- each species there are differences in the timing of when the
bers of fungiform papillae were also much reduced. “mature” numbers of taste buds are formed. Often
Although some of the current studies challenge the researchers use the formation of a taste “pore” to denote
dogma that taste bud induction is dependent upon innerva- when the taste bud is morphologically mature. For exam-
tion, induction of taste bud cells is far from understood. The ple, in the rat, about 50% of taste buds in the soft palate
768 Mistretta and Hill

across the tongue. In rat there is a systematic increase in

papilla size, from the anterior quarter of the tongue
where diameter averages 135 m, to the posterior quar-
ter with average diameter of 180 m (Mistretta and
Baum, 1984). Although the difference in papilla size is
substantial, only one taste bud is observed per fungiform
papilla in rat, suggesting particular controls on taste bud
number.
In contrast to rodent, in the human, sheep, and primate
tongue there are multiple taste buds per fungiform papilla,
and a direct relation between size of papilla and taste bud
numbers has been demonstrated from fetal through adult
stages in sheep (Mistretta et al., 1988). Furthermore, the
numbers of taste buds per papilla increase prenatally and
then decrease after birth in sheep, in direct association with
an increase and subsequent decrease in number of fibers in
the chorda tympani nerve (Mistretta et al., 1988). The
important question of factors regulating taste bud number
is not understood and may well relate to different cell and
molecular factors in lingual versus extralingual taste bud
populations; there is experimental evidence for epithelial,
neural, and dermal matrix controls (Mistretta and Hill,
1995).
Whereas numbers of taste buds vary widely within
lingual papillae and in soft palate and epiglottis, taste
bud size throughout the oropharynx is more homoge-
neous than taste bud number within a species. Even
without papilla size constraints, extremely large taste
Figure 6 Number of taste buds (A) and percentage of taste buds
buds are not found on the adult epiglottis or soft palate
with pores (B) on the soft palate and in lingual fungiform, cir-
cumvallate, and foliate papillae, during postnatal development in
(discussion in Mistretta, 1991). In fact, during develop-
rat. Development of mature taste bud numbers and pore acquisi- ment large taste buds apparently “split” to yield more
tion have different temporal sequences in different oral regions. numerous smaller taste buds (Bradley, 1972; Bradley
(Data redrawn from Harada et al., 2000.) et al., 1980). It has been proposed that very large taste
buds may form in fungiform papillae during develop-
have taste pores at birth (Harada et al., 2000) (Fig. 6), ment under the influence of excess innervation and then
whereas the majority of fungiform taste buds do not have divide to yield more numerous, smaller taste buds as
taste pores until after P12 (Mistretta, 1972), and about neural competition yields appropriate receptive fields
50% of circumvallate papilla taste buds have taste pores by (Mistretta, 1998). Furthermore, it will be clear from dis-
approximately P30 (Harada et al., 2000; Hosley and cussion in a later section that taste bud size in rat fungi-
Oakley, 1987). While using the appearance of a taste pore form papilla varies directly in relation to innervating ele-
is useful in denoting the presence of taste buds, it should ments, with an intriguing developmental progression
not be the sole marker of taste bud maturity. Functionally, (Krimm and Hill, 1998).
Mbiene and Farbman (1993) suggest that the absence of a The circumvallate papilla contains a few hundred taste
pore during early development may not prevent access of buds in rat, and as a taste organ it contrasts in many ways
stimuli to taste buds. Instead, the epithelium covering the with the fungiform papilla. Recent work incorporates use
pore is permeable to stimuli. Therefore, the presence of a of bax null mutant mice to disrupt programmed cell death
pore may not be necessary for the peripheral gustatory sys- during development and thereby increase gustatory inner-
tem to function at some, albeit immature, level. vation (Zeng et al., 2000). The excessively innervated
Various factors have been proposed to regulate final circumvallate papilla in bax / mice was larger than in
numbers of mammalian taste buds (Mistretta and Hill, wild-type animals and had larger taste buds, demonstrat-
1995), including papilla size constraints for lingual taste ing a direct link among innervation, papilla size, and taste
buds. The fungiform papillae are not uniform in size bud size.
Taste System Development 769

F. Taste Bud Cells: Cycle and Life Span

Aside from the specific factors that regulate taste bud num-
bers and size, the final site of action is the control of taste
receptor cell cycle and life span. Since taste receptors turn
over in adult mammals approximately every 10 days, there
must be a continual replacement of dying cells by way of
new cell divisions (Beidler and Smallman, 1965).
Therefore, by altering the rates of cell division, cell life
span, and cell death, the size and numbers of taste cells
within each taste bud will be affected. Surprisingly, very
little attention has been devoted to study of such processes.
There now are emerging studies, however, that address
these issues.
Indeed, Zeng et al. (2000) have recently identified some
cell death pathways in circumvallate taste cells by using
bax null mutant and wild-type mice. They provide evi-
dence for p53, Bax, and Caspase-2 in taste cell death. This
is unusual in that these multiple factors are generally not
evident in keratinocytes, cells that also are sloughed from
lingual epithelia, and they indicate tight regulation of taste
cell death.
Perhaps the most interesting, but most difficult to study,
aspect of taste cell life span is in the proliferative phase.
Specifically, it is not clear what cells give rise to the taste
receptor cells, which themselves are almost entirely
postmitotic (Beidler and Smallman, 1965). Stone and Figure 7 Photomicrograhs of fungiform papilla and taste bud in
colleagues (1995), using transgenic X chromosome–inacti- a postnatal rat aged 10 days (A) and an adult rat (B). The nuclei
vated mosaic mice, reported that taste cells and epithelial of all cells are labeled green, whereas taste bud cells only are
cells come from a common progenitor and that taste recep- labeled red with an antibody directed against cytokeratin 19. The
tor cells may originate from both ectoderm and endoderm. postnatal day 10 taste bud is smaller and has fewer cells than the
In part because of this complexity, the location and cell adult bud. (Data from Hendricks and Hill, 1999.) (See color
types that are taste bud progenitors have yet to be identified. insert.)
While there have been some advances in understanding
taste cell life span in adult taste buds, virtually nothing is G. Development of the Ganglion Cell Peripheral
known about the earliest formation of taste buds and cell Receptive Field: Numbers of Fungiform Papillae
cycle. Recently, however, Hendricks and Hill (1999) pro- and Taste Buds Per Ganglion Cell and Numbers
vided preliminary data suggesting that the proliferation of Ganglion Cells Per Papilla and Taste Bud
rates, the rates of dividing cells that are entering the taste
bud, and the life span of taste cells are much longer in rats The number and location of taste buds innervated by a sin-
aged 10 days postnatal compared to adults (Fig. 7). gle fiber or ganglion cell defines the receptive field for that
Therefore, at least some of the cell cycle parameters of the neuron. Therefore, the functional unit for a primary affer-
developing taste system in rat fungiform taste buds seem to ent neuron relates to its receptive field. The development
be much different than in adults. This is consistent with pre- and organization of receptive fields were first examined in
vious results suggesting that fungiform papillae have very fetal and early postnatal sheep (Mistretta et al., 1988), in
low proliferation rates in late fetal rat that later increase which an inverted “U” function described the relationship
(Farbman and Mbiene, 1991). Similarly, Ganchrow and between age and the number of taste buds innervated by
colleagues (1995) provide evidence that cell proliferation is single chorda tympani neurons (discussion in Mistretta and
absent during early periods of chick taste primordium for- Hill, 1995). The change from relatively small, to large, to
mation. The new data and continuing studies on taste bud small receptive fields indicates a highly dynamic system of
cell proliferation will be crucial for discerning core biolog- neuron/target interactions. One of the implications of
ical principles of taste bud formation. this morphological remodeling with age is that there are
770 Mistretta and Hill

corresponding functional changes, addressed at length in a buds is similar from postnatal day 10 to postnatal day 40 in
later section. rat; only taste bud size increases with age (Krimm and
By examining the innervation of single taste buds in rat, Hill, 1999). Moreover, the majority of neurons that inner-
findings complementary to the receptive field results cited vate a taste bud at 10 days postnatal continues to innervate
above have been provided. The approach in rat is different it through 40 days. Surprisingly, the size of the taste bud in
from that of quantifying the number of taste receptor ele- 40-day-old rats is predicted by the number of geniculate
ments per innervating fiber, because it determines the ganglion cells labeled at 10 days. These findings indicate
number of ganglion neurons that innervate a single taste that the neural “template” for the receptive field is deter-
bud (Fig. 8). In the latter approach, the location of the taste mined early and becomes matched with taste bud size later
bud on the tongue surface does not predict how many in development. This match may involve interactions
geniculate ganglion cells innervate a fungiform taste bud between taste buds and innervating neurons, most likely by
(Krimm and Hill, 1998). However, there is a highly way of diffusible factors. At least some of the interaction
ordered and predictable relationship in mature rats, when likely includes the neurotrophin family of molecules, dis-
the number of ganglion cells is related to the size of the cussed in a later section.
taste bud. Beginning at postnatal day 40 and continuing These collective findings reveal the type and nature of
through adulthood, the number of neurons that innervate nerve/target interactions during developmental regulation
each taste bud is directly related to taste bud size; the larger and maintenance of taste organ morphology, receptive
the taste bud, the greater the number of innervating gan- fields, and taste function. For example, in rat, it is from late
glion cells (Fig. 9). By contrast, there is no relationship gestation through the second postnatal week that individ-
between the size of the taste bud and the number of inner- ual taste buds receive their mature complement of
vating neurons in rats aged postnatal day 10 through post- innervating neurons, which in turn establishes the size that
natal day 30 (Fig. 10). the taste bud acquires. These processes must include inter-
Further studies on this age-related relationship demon- actions with molecules produced by target tissues (attrac-
strated that the number of ganglion cells innervating a taste tant and repellent) that act on innervating neurons and,
bud is established early in development, long before taste reciprocally, are related to molecules that are produced by
buds reach their mature size (Krimm and Hill, 1999). The neurons that innervate taste receptor cells. In short, this
mean number of ganglion cells that innervate single taste reflects proper matching between neurons and targets and

Figure 8 (A) Photomicrograph of a coronal section through a fungiform papilla and taste bud in which Fluoro-Gold had been ion-
tophoretically injected. (B) Photomicrograph of a section through the geniculate ganglion, illustrating a ganglion cell that was labeled
with Fluoro-Gold after the papilla injection in A. The ganglion cell in B, therefore, innervated the taste bud illustrated in A. (From Krimm
and Hill, 1998.) (See color insert.)
Taste System Development 771

(Mistretta et al., 1999; Nosrat et al., 1997). Interestingly,

fungiform papillae in BDNF null mutants were dispropor-
tionately spared on the anterior tip of the tongue (Mistretta
et al., 1999), demonstrating that not all papillae are
affected similarly through the loss of this neurotrophin.
Deletion of the gene for BDNF also affected development
of the circumvallate papilla (Mistretta et al., 1999; Nosrat
et al., 1997; Oakley et al., 1998), which was reduced in
size by about 40% compared to the wild-type papilla (Fig.
11). Liebl et al. (1999) extended these findings by demon-
strating that mice with targeted deletions of the gene for
neurotrophin NT4/5, which also uses the trkB receptor,
sustain a loss of fungiform papillae at birth.
As may be predicted from the results listed above, dele-
tions in the genes for trkB (Fritzsch et al., 1997), BDNF
Figure 9 Graph of the number of ganglion cells that innervate (Mistretta et al., 1999; Nosrat et al., 1997; Oakley et al.,
a single taste bud in fungiform papillae as a function of the
1998), and NT4/5 (Liebl et al., 1999) all have led to sig-
respective taste bud volume in adult rat. Open dots denote data
nificantly decreased taste bud numbers and size of remain-
from taste buds on the tongue tip, and solid dots denote data from
taste buds on the mid-region of the tongue. The regression line for ing taste buds. In some cases, such as in circumvallate
the mid-tongue data is shown as a solid line, and dotted lines on papilla in BDNF knockout mice, the magnitude of taste
either side show the 95% confidence limits. The correlation coef- bud loss exceeds the magnitude of change in papilla size
ficient and regression equation are given in the lower right region (Mistretta et al., 1999), indicating that these gene deletions
of the figure. (From Krimm and Hill, 1998.) may have an even greater effect on taste bud sustenance
than on papilla development (Fig. 11). Regardless of the
specific gene affected (i.e., trkB vs. BDNF) or the target
actions of an ensemble of molecules that have varying (i.e., fungiform vs. circumvallate papillae and taste buds),
functions during well-defined periods of development. the common effect is that there is an accompanying, sig-
One family of molecules that plays multiple roles and has nificant loss of the respective ganglion cells (geniculate
had much recent experimental attention is the neu- and petrosal). It is likely, therefore, that the morphological
rotrophins. changes seen in the target organs are related to the loss of
innervating neurons.
H. Neurotrophin Molecules and Roles in Peripheral Because there has been no study of taste papilla forma-
Taste Organ Development tion in embryonic BDNF or NT4/5/ mice, conclusions
about early neurotrophin support from tongue papillae for
In addition to the role of neurotrophins in ganglion devel- geniculate or petrosal ganglia cannot be derived with these
opment and neuron survival (see Sec. II. A), these mole- animals. Ganglia are dependent on neurotrophins in
cules also are critical in processes related to taste receptor advance of taste papilla formation, so decreased numbers
organ development. These processes include early taste of ganglion neurons from early stages could then lead to
organ formation, growth, and maintenance and refinement loss of neural support for sustaining the early taste papil-
of receptive field size (i.e., neuron/target matching). lae, which form without intact innervation (Mistretta,
Furthermore, functional differentiation of ganglia that 1998). What is clear from use of such models, however, is
innervate the tongue is altered by neurotrophins (discussed a direct association between ganglion cell loss and gusta-
in a later section). Thus, neurotrophins do not simply per- tory organ loss.
mit gustatory neurons to survive, but they play various An important conclusion emphasized by Mistretta et al.
roles in the complex processes of peripheral taste system (1999) is that with loss of about 60% of the fungiform
development. papillae (and thus resident taste buds) and of about 50% of
As noted earlier, targeted gene deletion of the neu- neurons in the geniculate ganglion in BDNF/ mice, the
rotrophin receptor, trkB, results in severe alterations in remaining geniculate neurons do not extend neurites into
gustatory papillae in mice (Fritzsch et al., 1997). the tongue to “rescue” papillae (and resident taste buds)
Furthermore, deletion of the gene for the trkB ligand, that are lost due to reduced lingual innervation (Mistretta
brain-derived neurotrophic factor (BDNF), results in as et al., 1999). This indicates that there is specificity in the
much as a 60% decrease in numbers of fungiform papillae nature of fungiform papilla innervation: all fungiform
772 Mistretta and Hill

Figure 10 Relationship between the number of geniculate ganglion cells that innervate a single taste bud and the respective taste bud
volume, shown for fungiform papillae in rats aged 10, 20, 30, and 40 days postnatal. Open symbols in A and B denote data from taste
buds on the tongue tip, and solid dots in all graphs denote data from taste buds in the mid-tongue region. The dashed regression line in
each graph is from data obtained from taste buds in the mid-tongue of adult rats. Dotted lines on either side of the regression line in D
show the 95% confidence intervals corresponding to the adult data. The correlation coefficient and regression equation for postnatal day
40 rats are in the lower right region of panel D. (From Krimm and Hill, 1998.)

papillae are not uniformly dependent on a generalized Nosrat et al., 2000), illustrating the complexity of interac-
innervation from any of a set of homogeneous ganglion tions bewteen neurotrophins and lingual targets. In a
cells. Furthermore, in the BDNF/ mice from the somewhat unexpected result, in mice that overexpress
Reichardt laboratory used by Mistretta et al. (1999), only BDNF there is a decrease in size and number of fungiform
about 16% of trigeminal ganglion cells were lost, yet and circumvallate papillae and a decrease in number of
remaining trigeminal neurons did not “rescue” fungiform taste buds (Krimm et al., 2000; Ringstedt et al., 1999). In
papillae from the tongue. In short, neither remaining one of the overexpressor models, the promoter sequence
geniculate nor trigeminal neurons substitute to rescue from the nestin gene resulted in a bolus of BDNF in the
fungiform papillae and/or their resident taste buds. core of the tongue muscle and innervating fibers “stalled”
Some investigators have proposed that NT-3 sustains at that location (Ringstedt et al., 1999). However, in an
remaining papillae and taste buds in BDNF/ mice important and detailed study, Krimm et al. (2001) used a
(Nosrat et al., 1997). Clearly multiple neurotrophins will keratin 14 promoter to drive overexpression of either
participate in development of taste organs, and the nature BDNF or NT4 in basal epithelial cells, so that fibers
and specifics of this participation are still emerging. reached and innervated the lingual epithelium. Geniculate
Interesting data derive from examination of transgenic ganglion cells were increased in number by 93% (K14-
mice that are BDNF overexpressors (Krimm et al., 2001, BDNF mice) and 140% (K14-NT4 mice). Neuron number
Taste System Development 773

bud development, after induction and initial differentiation

stages, may be regulated in relation to neurotrophin func-
tions (Fig. 12).
In postnatal rat, the size of individual fungiform papil-
lae and taste buds varies considerably, but systematically,
across the tongue surface (e.g., Krimm and Hill, 1998).
Once taste buds are induced, BDNF, which is expressed at
the apex of developing fungiform papillae (Nosrat and
Olson, 1995; Nosrat et al., 1996), could be made in dis-
parate amounts in different papillae and taste buds, thereby
determining the number of neurons that are sustained in
synpatic relations with taste buds cells. The number of
neurons sustained to innervate the taste bud would be
proportional to the amount of BDNF produced by each
taste bud. Therefore, the number of innervating gustatory
neurons could be determined very early in development. In
fact from the work by Krimm and Hill (2000) presented
earlier (see Sec. III.E), the mature number of innervating
neurons for each rat fungiform taste bud is determined
before postnatal day 10. As seen earlier, the number of
innervating neurons may then help determine the mature
size of the taste bud achieved about 30 days later, when
taste bud size is directly related to the number of innervat-
ing neurons (e.g., Krimm and Hill, 1998). That is, the max-
Figure 11 Photomicrographs of histological sections of the cir-
imal number of taste bud cells supported in each taste bud
cumvallate papilla and taste buds from the tongue of wild-type
(/) and BDNF null mutant (/) mice at postnatal day 25. may be determined by the number of innervating neurons.
Sections were made parallel to the surface of the tongue and Once the maximal number of cells is achieved, the rate of
stained with hematoxylin and eosin. Compared with wild-type cell loss is matched by the rate of cell replacement. While
tongues, the circumvallate papilla in null mutants was smaller in the mechanism by which neurons direct taste bud growth
diameter and length and lacked morphological integrity. Taste is unknown, there is strong evidence that a factor(s)
buds remained in / circumvallate papilla but were much released from gustatory nerves support taste bud cells. This
reduced in number. (From Mistretta et al., 1999.) is best illustrated by the series of experiments in adult
rodents showing that sectioning of gustatory nerves (Cheal
in the trigeminal ganglion was unchanged (LeMaster et al., and Oakley, 1977; Fujimoto and Murray, 1970; Guth,
1999). However, fungiform papillae and taste buds were 1957; Von Vintschgau and Honigschmied, 1876) or block-
reduced in number and size (Krimm et al., 2001). Papillae age of axoplasmic transport by colchicine (Sloan et al.,
formed in normal numbers embryonically but were gradu- 1983) results in loss of taste buds. Neurotrophins could be
ally lost postnatally. Careful analysis of innervation one factor that participates in various aspects of taste bud
patterns in these tongues demonstrated that lingual inner- cell development because neural transport of neu-
vation was directed to filiform papillae in preference to rotrophins can proceed anterogradely as well as retro-
fungiform papillae. Although the lingual epithelium had gradely (Tonra et al., 1998).
generalized dense innervation from geniculate neurons,
taste buds did not form in these extrapapilla areas, demon-
IV. CENTRAL TARGETS OF GANGLIA:
strating again the necessity of papilla epithelium for lin-
BRAINSTEM TRACTS AND NEURONS AND
gual taste bud development (Mistretta, 1991, 1998). The
HIGHER ORDER RELAYS
“misdirection” of taste afferents into filiform papillae
suggests that a specific localized distribution of BDNF and
NT-4 is essential for target-nerve interactions that sustain A. Projections into the Nucleus of the Solitary Tract
gustatory organs (Krimm et al., 2001).
Data on effects from loss of neurotrophin and associ- In the rat, chorda tympani fibers begin synapse formation in
ated receptors provide insight into normal developmental the rostral pole of the nucleus of the solitary tract (NTS) as
processes. The following is a working scheme of how taste early as postnatal day 1 (Lasiter et al., 1989), even though
774 Mistretta and Hill

Figure 12 Model of possible interactions between young, developing taste buds (after induction and initial stages of differentiation and
innervation) and innervating neurons. Once taste buds are induced, BDNF could be made in disparate amounts in different papillae and
taste buds, influencing the number of neurons sustained in synaptic relations with taste bud cells. See text for discussion of model.

some nerve fibers may arrive centrally much earlier (Altman organization of gustatory afferents, similar to that seen in
and Bayer, 1982; Scott and Atkinson, 1998). However, the other sensory systems (see, e.g., Coleman, 1990).
terminal field of the chorda tympani does not reach its full
size until approximately postnatal day 25 (Lasiter, 1992). B. Neurons in the Nucleus of the Solitary Tract
From their initial projections into the rostral pole of the
NTS, chorda tympani axons migrate caudally and surround Neurons postsynaptic to chorda tympani fibers in the rat
local neurons and the output neurons of the NTS. NTS also show dramatic changes in morphology, with
In contrast to terminations of the chorda tympani nerve dendritic lengths and dendritic branching increasing
in the rostral NTS, fibers of the glossopharyngeal nerve, approximately threefold between P8 and P25 (Lasiter,
which innervate taste receptors on the posterior tongue, do 1992; Lasiter et al., 1989), a time window that approxi-
not enter the intermediate zone of the NTS until postnatal mates the developmental period when the terminal field of
day 9–10 in rat (Lasiter, 1992). Moreover, the rostral-cau- the chorda tympani nerve matures. Indeed, the collective
dal expansion of this field is not complete until approxi- increase in pre- and postsynaptic neurites in the NTS with
mately postnatal day 45 (Lasiter, 1992). age indicates that an increasingly large afferent message
Preliminary findings (Sollars and Hill, unpublished may influence circuit formation (e.g., Lasiter, 1992;
data) indicate that the terminal field of the greater superfi- Mistretta and Hill, 1995). These results are reinforced
cial petrosal nerve (GSP) is mature in size and topography through findings in developing sheep, which, as noted
long before the terminal fields of the chorda tympani and earlier, have a significant prenatal gustatory development.
glossopharygeal nerves. This is especially noteworthy Mistretta and Labyak (1994) demonstrated that the
because the GSP and chorda tympani terminal fields over- dendritic lengths and dendritic spine numbers of putative
lap considerably at adulthood (Hamilton and Norgren, relay neurons in the NTS increased during the time when
1984). It is interesting to note that the sequence of primary functional convergence occurred and when taste circuits
afferent projection into the NTS follows the pattern of matured. Therefore, in both sheep and rat, there appears to
morphological and functional maturation noted for the be a correspondence of maturation of function and struc-
respective receptor populations (see Sec. V). It is possible, ture in neurons that receive and then transmit gustatory
therefore, that activity may play a role in terminal field information to more central structures.
Taste System Development 775

In contrast to the timing delay between terminal field sheep, in which there is a significant prenatal taste devel-
and dendritic development in the rostral NTS, the period opment (Bradley and Mistretta, 1973), low responses to
that separates these events in more caudally located NTS NaCl occur from the last trimester and increase until adult-
neurons is significantly shorter. However, maturation of hood (Mistretta and Bradley, 1983). In rats, in which taste
pre- and postsynaptic elements occurs later in more cau- development occurs almost entirely postnatally (Farbman,
dally located NTS zones. Specifically, in rat only one week 1965; Mistretta, 1972), responses to sodium salts soon
separates the morphological maturation of the pre- and after birth are very low and then increase in magnitude
postsynaptic elements in the NTS where the glosopharyn- through weaning (Ferrell et al., 1981; Hill and Almli,
geal nerve projects (at bout P45) (Lasiter, 1992). 1980; Yamada, 1980). In both species, sodium salts elicit
the largest magnitude response of all stimuli in the chorda
C. Higher-Order Gustatory Relays tympani at adulthood. By comparison, the developing gus-
tatory system is highly responsive to some nonsodium salt
The primary gustatory projection from the NTS is to the stimuli as soon as it begins functioning. Citric acid and
parabrachial nuclei in the pons (PBN). The development of ammonium chloride (NH4Cl) produce vigorous responses
this projection does not parallel that of the chorda tympani in the chorda tympani nerve in rats aged 2 days postnatal
nerve to the NTS (Lasiter and Kachele, 1988, 1989). (Hill and Almli, 1980). Therefore, the taste system
Specifically, second-order neuron projections from the responds with “mature” magnitudes to some stimuli early
NTS to the PBN do not begin until postnatal day 7 (about in development (Hill et al., 1982), unlike other sensory
1 week later than the chorda tympani to NTS projection) systems where peripheral neurons and receptor cells are
and the terminal field maturation is slower, attaining com- “sluggish” to all stimuli (see Coleman, 1990).
pletion at about postnatal day 60 (Lasiter, 1992). There is Developmentally, there is also a functional/morpholog-
an apparent serial development of these central gustatory ical reorganization of receptive fields, as demonstrated in
projections; therefore, significant development of the PBN sheep (Nagai et al., 1988). Namely, during the period when
does not occur until some level of maturation is achieved the average receptive field size decreases in sheep (see Sec.
in the NTS. III.G), there is an increase in the proportion of fibers that
Little information is available concerning the morpho- respond maximally to NaCl. Furthermore, the neurons
logical development of the PBN and its more rostral relays. with small receptive fields throughout development
The delayed development of presynaptic elements (i.e., the respond with higher responses to NaCl and fibers with
terminal field) in the PBN is reflected in postsynaptic cells. larger receptive fields respond maximally to NH4Cl.
Dendrites of PBN neurons arborize significantly between
postnatal days 16 and 35 in rat (Lasiter and Kachele, 1988), B. Epithelial Sodium Channels and Salt Taste
the period during which projections to the next relay in the Development
thalamic taste area occur and metabolic activity of PBN
neurons increases (Lasiter and Kachele, 1988). These later- With the advent of amiloride, a drug used to block epithe-
developing morphological changes suggest that functional lial sodium channels, significant advances have been real-
response maturation of the PBN may differ from those in ized in identifying sodium taste transduction pathways and
the NTS and, as noted for terminal field development, may the development of these pathways. In adult rats, lingual
be serially dependent on NTS development. application of amiloride attenuates responses to NaCl by
70–80% and eliminates responses to sodium acetate
(Formaker and Hill, 1988). The difference in the amount of
V. FUNCTIONAL DEVELOPMENT OF THE suppression relates to multiple or single transduction path-
GUSTATORY SYSTEM ways for NaCl and sodium acetate, respectively, and
appears to be dependent on the size of the anion (see
A. Taste Buds and Peripheral Nerve Function Stewart et al., 1997). Developmental studies that have used
the epithelial sodium channel blocker are summarized in
Large increases in responses to some, but not all, taste the following paragraphs.
stimuli characterize the functional development of the Increased sensitivity to sodium occurs along with
peripheral gustatory system. The most notable change is in increased sensitivity to amiloride. The initial small taste
response to sodium salts. For example, the chorda tympani responses to NaCl from the chorda tympani nerve in rats
nerve in sheep (Mistretta and Bradley, 1983) and rat younger than 13 days are suppressed little by amiloride.
(Ferrell et al., 1981; Hill and Almli, 1980; Yamada, 1980) However, the NaCl taste response becomes larger through
is poorly responsive to NaCl during early development. In development, as does the amount of response suppression
776 Mistretta and Hill

by amiloride (Hill and Bour, 1985; Sollars and Bernstein, neurophysiological taste responses from the chorda tym-
1994). Thus, increased sensitivity to sodium appears to pani nerve (Hendricks et al., 2000). With the use of bio-
reflect an increase in functional amiloride-sensitive physical modeling techniques, parameters such as channel
sodium channels (Hill and Bour, 1985). densities (i.e., numbers/unit volume) and channel affinities
This relatively specific developmental change that may (i.e., function) (Ye et al., 1993) were derived in developing
involve a single transduction pathway makes these rats rats. The number of functional sodium channels on the
interesting models for immunohistochemical study. With anterior tongue increased monotonically with age (Fig.
the use of a polyclonal antibody directed at the amiloride 13), and furthermore, the channels also became more effi-
channel, immunopositive taste buds are unexpectedly seen cient (i.e., have higher affinities for sodium salts). Thus, an
on the tongue of rats aged one day postnatal (Stewart et al., increase in channel densities and in channel function
1995). This is unexpected because chorda tympani accounts for the dramatic increase in sodium salt responses
responses from early postnatal rats have small (or no) during development.
responses to sodium salts and little sensitivity to amiloride The apparently conflicting results between the whole
(Hill and Bour, 1985; Sollars and Bernstein, 1994). Since cell patch recordings (Kossel et al., 1997) and results from
the antibody was polyclonal, it was proposed that at least the in vitro epithelia (Settles and Mierson, 1993) and in
some form of the channel is present in early developing vivo neural recording (Hill and Bour, 1983) may be
taste buds and that the channels become functional later in explained by translocation of the channel from the basolat-
development (Stewart et al., 1995). Alternatively, the chan- eral to the apical domain, where they are able to “sense”
nels may be functional in taste receptor cells of young rats, sodium stimuli. The translocation of functional amiloride
but not available to sodium ions that stimulate the apical channels in polarized epithelial cells has been demon-
domain (Stewart et al., 1995). That is, the channels may strated in a variety of tissues (see Garty and Benos, 1988;
not be inserted into the membrane in the apical domain of Garty and Palmer, 1997) and is not unique to gustatory
taste receptor cells. tissue. Identification of factors and processes involved in
Support for this alternative is found in whole cell patch channel trafficking and/or channel regulation during devel-
clamp recordings of stimulus-induced conductances in opment would be a major advance in our understanding of
postnatal rat (Kossel et al., 1997). Receptor cells from salt taste development. Moreover, identification of the
fungiform taste buds in P2 rats were as sensitive to cellular and molecular events of channel production,
amiloride as cells from P30 rats. Sodium-related conduc- including epithelial sodium channel subunit complementa-
tances were suppressed in taste bud cells similarly tion, would serve to provide the underlying cellular
throughout development. An important aspect of these mechanisms that are involved in this process.
studies is that apical and basolateral domains are not func-
tionally separated in whole cell patch clamp recordings.
Therefore, all stimulus-induced conductances are
recorded, regardless of the location of the transduction site.
Indeed, the relative lack of amiloride sensitivity in neural
recordings (Hill and Bour, 1983; Sollars and Bernstein,
1995) suggests that whole cell recordings detect functional
amiloride-sensitive channels that are not normally reached
by chemical stimuli in intact tissue because of their loca-
tion in the basolateral domain of taste receptor cells
(Kossel et al., 1997).
Further definition of the initial availability of functional
sodium channels on the taste receptor membrane in early
postnatal rats comes from biophysical analyses. Stimulus-
induced ionic currents were recorded in vitro from dorsal
lingual epithelia from P14 rats to adult, and sodium con-
ductances reportedly increased during development with a
corresponding increase in amiloride sensitivity (Settles Figure 13 Chorda tympani nerve response ratios versus NaCl
and Mierson, 1993). Recently, these results were extended electrochemical concentration for postnatal rats in four age
with an in vivo voltage clamp procedure. Briefly, voltage groups. The data demonstrate a developmental increase in num-
was applied across lingual epithelia in anesthetized rats in ber and efficiency of sodium channels in taste bud cells. (From
order to control the driving force of ions while recording Hendricks et al., 2000.)
Taste System Development 777

In summary, these findings provide critical information chorda tympani nerve was responsive (Bradley and
regarding sodium response ontogeny. First, they point to Mistretta, 1980; Mistretta and Bradley, 1978a). This
the cellular locus in the development of sodium responses. absence of NaCl responses was not seen in NTS in rats as
Second, they establish a developing transduction pathway young as postnatal day 14, which was the youngest age
that is relatively well characterized in gustatory and examined (Hill et al., 1983); however, it may be found in
nongustatory tissue (e.g., Avenet and Lindemann, 1988; younger rats. Furthermore, in both rat (Hill et al., 1983)
DeSimone and Ferrell, 1985; Garty and Benos, 1988). and sheep (Bradley and Mistretta, 1980; Mistretta and
Collectively, these results can be exploited to better under- Bradley, 1978), mature NTS responses were achieved later
stand the underlying mechanisms involved in sodium taste in development compared to the chorda tympani (Bradley
development. and Mistretta, 1973; Ferrell et al., 1981; Hill and Almli,
1980; Mistretta and Bradley, 1983; Yamada, 1980). For
C. Gustatory Ganglia

As seen above, there have been considerable advances in

characterizing the developing peripheral gustatory system
through neurophysiological studies of gustatory nerves
and, more recently, taste receptor cells. However, virtually
nothing is known about the functional development of cell
bodies of the taste nerves in gustatory ganglia. It is likely,
because of the recent attention devoted to the role that such
ganglia have in the morphological induction and mainte-
nance of taste buds, that more emphasis will be given to
the functional characteristics of these cells. Indeed, in
recent reports, investigators make use of ganglion explants
from embryonic rat and whole cell recordings from the
cultured ganglia. Numerous passive membrane and action
potential properties are different in neurons from E16
trigeminal compared to geniculate ganglion (Fig. 14), indi-
cating that at early stages in ganglion development, neu-
rons have begun to differentiate functionally (Al-Hadlaq
et al., 1999; Grigaliunas et al., 2002). Furthermore,
explanted ganglion cells have altered membrane and action
potential properties after exposure in culture to different
neurotrophins (Al-Hadlaq et al., 2001, 2002). Because the
neurotrophins used in these experiments are encountered
by growing neurites in the tongue in vivo, the results
indicate that neurotrophins could modulate ganglion cell
function during outgrowth to gustatory target organs in the
embryo.

D. Central Gustatory Nuclei

Recordings of taste responses from NTS neurons in rat Figure 14 Whole cell patch recordings in response to series of
(Hill et al., 1983) and sheep (Bradley and Mistretta, 1980; hyperpolarizing and depolarizing current pulses, shown at bot-
Mistretta and Bradley, 1978) demonstrated that central tom, in neurons from trigeminal and geniculate ganglia. Ganglia
were dissected and maintained in culture from embryonic rats at
response development was similar in general sequence to
gestational day 16. Numerous passive membrane and action
chorda tympani nerve response development. Response
potential properties are different between trigeminal and genicu-
magnitudes to sodium salts increased profoundly in NTS late cells even at this early embryonic stage. For example, as
neurons during development. In contrast, NTS neuron illustrated, at threshold currents trigeminal neurons generally
responses to other stimuli, such as to NH4Cl, did not produce a single action potential, whereas a substantial propor-
change with age. There was a complete lack of sensitivity tion of geniculate neurons produce mutiple action potentials.
in NTS neurons to sodium salts in fetal sheep, whereas the (Data from Grigaliunas et al., 2002.)
778 Mistretta and Hill

example, in rats, mature taste responses to sodium salts in REFERENCES

the chorda tympani occur at postnatal weeks 3–4. In the
NTS, mature responses appear about one week later. Al-Hadlaq, S. M., Bradley, R. M., MacCallum, D. K., and
Therefore, central taste response development is related Mistretta, C. M. (1999). Neurotrophin effects on neurophysi-
not only to peripheral changes, but also to central synaptic ology of cultured geniculate and trigeminal ganglion cells
changes (Bradley and Mistretta, 1980; Hill et al., 1983; from embryonic rat. Neurosci. Abst. 25:1011.
Al-Hadlaq, S. M., Bradley, R. M., MacCallum, D. K., and
Mistretta and Bradley, 1978).
Mistretta, C. M. (2001). Survival and neurophysiology of
Synaptic organization or reorganization may occur dur- embryonic rat geniculate ganglion neurons are altered by neu-
ing development in the NTS, which further delays devel- rotrophins. Chem. Senses 26:1075.
opment of taste responses. This probably results in a Al-Hadlaq, S. M., Bradley, R. M., MacCallum, D. K., and
degraded neural signal across the first gustatory relay dur- Mistretta, C. M. (2002). Geniculate neurons in culture
ing the immature NTS period. Indeed, taste response have neurotrophin-specific electrophysiological properties.
development in sheep NTS neurons follows a structural Neuroscience in press.
refinement of receptive fields that can best be attributed to Altman, J., and Bayer, S. (1982). Development of the cranial
a remodeling of synapses from chorda tympani fibers onto nerve ganglia and related nuclei in the rat. Adv. Anat.
NTS neurons (Vogt and Mistretta, 1990). Specifically, Embryol. Cell Biol. 74:1–90.
there is an increase in receptive field sizes during develop- Avenet, P., and Lindemann, B. (1988). Amiloride-blockable
sodium currents in isolated taste receptor cells. J. Memb. Biol.
ment in sheep NTS neurons that most likely reflects an
105:245–255.
increased convergence of chorda tympani fibers onto the Baker, C. V. H., and Bronner-Fraser, M. (2001). Vertebrate cra-
central neurons with age. These findings, as well as nial placodes. I. Embryonic induction. Dev. Biol. 232:1–61.
increases in spontaneous rates and stimulus-elicited Bao, H., Bradley, R. M., and Mistretta, C. M. (1995).
response frequencies (Bradley and Mistretta, 1980; Hill et Development of intrinsic electrophysiological properties in
al., 1983; Mistretta and Bradley, 1978), suggest that neural neurons from the gustatory region of rat nucleus of solitary
activity plays a role in the way that primary afferents tract. Dev. Brain Res. 86:143–154.
synapse onto their central target neurons. Barlow, L. A., Chien, C-B., and Northcutt, R. G. (1996).
The delay cannot be attributed to poorly developed NTS Embryonic taste buds develop in the absence of innervation.
neurons because functional membrane characteristics are Development 122:1103–1111.
mature before the taste-elicited responses. In fact, the Begbie, J., Brunet, J-F., Rubenstein, J. L. R., and Graham, A.
(1999). Induction of the epibranchial placodes. Development
membrane parameters of resting membrane potential,
126:895–902.
action potential, and discharge properties change most Beidler, L. M., and Smallman, R. L. (1965). Renewal of cells
between P5 and P15 in rats, with mature values reached by within taste buds. J. Cell. Biol. 27:263–272.
P20 (Bao et al., 1995). Spine density on these neurons Belecky, T. L., and Smith, D. V. (1990). Postnatal development of
increases from P5 to P10 and then decreases to adult stages. palatal and laryngeal taste buds in the hamster. J. Comp.
This provides further evidence that changes in the synapses Neurol. 293:646–654.
and not the NTS neurons per se are what account for the Bitgood, M. J., and McMahon, A. P. (1995). Hedgehog and Bmp
prolonged functional changes postnatally. genes are coexpressed at many diverse sites of cell-cell inter-
Similar developmental patterns are observed in the next action in the mouse embryo. Dev. Biol. 172:126–138.
synaptic relay, the PBN. Responses to taste stimuli that Bradley, R. M. (1972). Development of the taste bud and gusta-
change developmentally at more peripheral levels also tory papillae in human fetuses. In Third Symposium on Oral
Sensation and Perception: The Mouth of the Infant, J. F.
occur in the PBN (Hill, 1987). However, PBN neurons
Bosma (Ed.). Charles C Thomas, Springfield, IL, pp.
have increased response magnitudes to all stimuli from 137–162.
weaning to early adulthood, including responses to stimuli Bradley, R. M. (1973). Investigations of intravascular taste using
that are very effective in eliciting NTS responses during perfused rat tongue. Am. J. Physiol. 225:300–304.
early development. Therefore, PBN neurons appear to be Bradley, R. M., and Mistretta, C. M. (1973). The gustatory sense
“sluggish” to all stimuli early in development, and then in foetal sheep during the last third of gestation. J. Physiol.
sensitivity gradually increases with age. The delayed func- (London) 231:271–282.
tional maturation at successively higher synaptic levels Bradley, R. M., and Mistretta, C. M. (1980). Developmental
suggests that central anatomical and/or neurochemical changes in neurophysiological taste response from the
events are responsible for the unique developmental medulla in sheep. Brain Res. 191:21–34.
patterns at each neural relay. Bradley, R. M., and Stern, I. B. (1967). The development of the
human taste bud during the foetal period. J. Anat. 101:743–752.
Taste System Development 779

Bradley, R. M., Cheal, M. L., and Kim, Y. H. (1980). Quantitative Nuerogenin2 is a determination factor for epibranchial
analysis of developing epiglottal taste buds in sheep. J. Anat. placode-derived sensory neurons. Neuron 20:483–494.
130:25–32. Formaker, B. K., and Hill, D. L. (1988). An analysis of residual
Brockes, J. P. (1997). Amphibian limb regeneration: Rebuilding a NaCl taste response after amiloride. Am. J. Physiol.
complex structure. Science 276:81–87. 255:R1002–R1007.
Buchman, V. L., and Davies, A. M. (1993). Different neu- Fritzsch, B., Sarai, P. A., Barbacid, M., and Silos-Santiago, I.
rotrophins are expressed and act in a developmental Sequence (1997). Mice with a targeted disruption of the neurotrophin
to promote the survival of embryonic sensory neurons. receptor trkB lose their gustatory ganglion cells early but do
Develop. Suppl. 118:989–1001. develop taste buds. Internat. J. De. Neurosci. 15: 563–576.
Cheal, M., and Oakley, B. (1977). Regeneration of fungiform Fujimoto, S., and Murray, R. G. (1970). Fine structure of degen-
taste buds: temporal and spatial characteristics. J. Comp. eration and regeneration in denervated rabbit vallate taste
Neurol. 200:609–626. buds. Anat. Rec. 168:398–413.
Coleman, J. R. (1990). Development of Sensory Systems in Ganchrow, D., and Ganchrow, J. R. (1985). Number and distrib-
Mammals. John Wiley & Sons, New York. ution of taste buds in the oral cavity of hatchling chicks.
Conover, J. C., Erickson, J. T., Katz, D. M., Bianchi, L. M., Physiol. Behav. 34:889–894.
Poueymirou, W. T., McClain, J., Pan, L., Helgren, M., Ip, N. Ganchrow, J. R., and Ganchrow, D. (1987). Taste bud develop-
Y., Boland, P., Friedman, B., Wiegand, S., Vejsada, R., Kata, ment in chickens (Gallus gallus domesticus). Anat. Rec.
A. C., DeChiara, T. M., and Yancopoulos, G. D. (1995). 218:88–93.
Neuronal deficit, not involving motor neurons, in mice lack- Ganchrow, D., Ganchrow, J. R., Gross-Isseroff, R., and
ing BDNF and/or NT4. Nature 375:235–238. Kinnamon, J. C. (1995). Taste bud cell generation in the peri-
Dassule, H. R., and McMahon, A. P. (1998). Analysis of epithe- hatching chick. Chem. Senses 20:19–28.
lial-mesenchymal interactions in the initial morphogenesis of Garty, H., and Benos, D. J. (1988). Characteristics and regulatory
the mammalian tooth. Dev. Biol. 202:215–227. mechanisms of the amiloride-blockable Nachannel. Physiol.
Davies, A. M. (1997). Studies of neurotrophin biology in the Rev. 68:309–337.
developing trigeminal system. J. Anat. 191:483–491. Garty, H., and Palmer, L. G. (1997). Epithelial sodium channels:
Davies, A., and Lumsden, A. (1984). Relation of target encounter Function, structure, and regulation. Physiol. Rev. 77:359–396.
and neuronal death to nerve growth factor responsiveness in Graham, A., and Begbie, J. (2000). Neurogenic placodes: a com-
the developing mouse trigeminal ganglion. J. Comp. Neurol. mon front. Trends Neurosci. 23:313–316.
223:124–137. Grigaliunas, A., Bradley, R. M., MacCallum, D. K., and
DeSimone, J. A., and Ferrell, F. (1985). Analysis of amiloride Mistretta, C. M. (2002). Distinctive neurophysiological prop-
inhibition of chorda tympani taste response of rat to NaCl. erties of embryonic trigeminal and geniculate neurons in
Am. J. Physiol. 249:R52–R61. culture. J. Neurophysiol. 88:2058–2074.
Doty, R. L., Ed. (1995). Handbook of Olfaction and Gustation. Guth, L. (1971). Degeneration and regeneration of taste buds:
Marcel Dekker, New York. In: Handbook of Sensory Physiology, Vol IV, Chemical
Ernfors, P., Merlio, J-P., and Persson, H. (1992). Cells Senses, Beidler, L. M. (Ed.). Springer, New York, pp. 63–74.
expressing mRNA for neurotrophins and their receptors dur- Hall, J. M., Hooper, J. E., and Finger, T. E. (1999a). Expression
ing embryonic rat development. Eur. J. Neurosci. of Sonic Hedgehog, Patched and Glil in developing taste
4:1140–1158. papillae of the mouse. J. Comp. Neurol. 406:143–155.
Ernfors, P., Lee, K-F., and Jaenisch, R. (1994). Mice lacking Hall, J. M., Finger, T. E., MacCallum, D. K., and Mistretta, C. M.
brain-derived neurotrophic factor develop with sensory (1999b). Sonic hedgehog signaling in rodent tongue cultures.
deficits. Nature 368:147–150. Chem. Senses 24:572–573.
Farbman, A. I. (1965). Electron microscope study of the develop- Hamilton, R. B., and Norgren, R. (2000). Projections of the para-
ing taste bud in the rat fungiform papilla. Dev. Biol. brachial nucleus in the old world monkey. Exp. Neurol.
11:110–135. 165:101–117.
Farbman, A. I., and Mbiene, J-P. (1991). Early development and Harada, S., Yamaguchi, K., Kanemaru, N., and Kasahara, Y.
innervation of taste bud - bearing papillae on the rat tongue. J. (2000). Maturation of taste buds on the soft palate of the post-
Comp. Neurol. 304:172–186. natal rat. Physiol. Behav. 68:333–339.
Farinas, I., and Reichardt, L. F. (1996). Neurotrophic factors and Hendricks, S. J., and Hill, D. L. (1999). Lower proliferation of
their receptors:implications of genetic studies. Neurosciences presumptive taste receptor cells during early postnatal devel-
8:133–143. opment. Neurosci. Abstr. 25:753.
Ferrell, M. F., Mistretta, C. M., and Bradley, R. M. (1981). Hendricks, S. J., Stewart, R. E., Heck, G. L., DeSimone, J. A.,
Development of chorda tympani taste responses in rat. and Hill, D. L. (2000). Development of rat chorda tympani
J. Comp. Neurol. 198:37–44. sodium responses: Evidence for age-dependent changes in
Fode, C., Gradwohl, G., Morin, X., Dierich, A., LeMeur, M., global amiloride-sensitive Na channel kinetics. J.
Goridis, C., and Guillemot, F. (1998). The bHLH protein Neurophysiol. 84:1531–1544.
780 Mistretta and Hill

Hill, D. L. (1987). Development of taste responses in the rat Lasiter, P. S. (1992). Postnatal development of gustatory recipi-
parabrachial nucleus. J. Neurophysiol. 2:481–495. ent zones within the nucleus of the solitary tract. Brain Res.
Hill, D. L. (2001). Taste development. In Handbook of Bull. 28:667–677.
Behavioral Neurobiology, 13, Developmental Psychobiology Lasiter, P. S., and Kachele, D. L. (1988). Postnatal development
E. Blass (Ed.), Kluwer Academic/Plenum Pub., NY, pp. of the parabrachial gustatory zone in rat: Dendritic morphol-
517–549. ogy and mitochondrial enzyme activity. Brain Res. Bull.
Hill, D. L., and Almli, C. R. (1980). Ontogeny of chorda tympani 21:79–94.
nerve responses to gustatory stimuli in the rat. Brain Res. Lasiter, P. S., and Kachele, D. L. (1989). Postnatal development
20:310–313. of protein P-38 (‘Synaptophysin’) immunoreactivity in pon-
Hill, D. L., and Bour, T. C. (1985). Addition of functional tine and medullary gustatory zones of rat. Dev. Brain Res.
amiloride-sensitive components to the receptor membrane: A 48:27–33.
possible mechanism for altered taste responses during devel- Lasiter, P. S., Wong, D. M., and Kachele, D. L. (1989) Postnatal
opment. Dev. Brain Res. 20:310–313. development of the rostral solitary nucleus in rat: dendritic
Hill, D. L., Mistretta, C. M., and Bradley, R. M. (1982). morphology and mitochondrial enzyme activity. Brain Res.
Developmental changes in taste response characteristics of rat Bull. 22:313–321.
single chorda tympani fibers. J. Neurosci. 2:782–790. LeMaster, A. M., Krimm, R. F., Davis, B. M., Noel, T., Forbes,
Hill, D. L., Bradley, R. M., and Mistretta, C. M. (1983). M. E., Johnson, J. E., and Albers, K. M. (1999).
Development of taste responses in rat nucleus of solitary tract. Overexpression of brain-derived neurotrophic factor enhances
J. Neurophysiol. 50:879–895. sensory innervation and selectively increases neuron number.
Hosley, M. A., and Oakley, B. (1987). Postnatal development of J. Neurosci. 18:5919–5931.
the vallate papilla and taste buds in rats. Anat. Record Liebl, D. J., Tessarollo, L., Palko, M. E., and Parada, L. F. (1997).
218:216–222. Absence of sensory neurons before target innervation in
Hosley, M. A., Hughes, S. E., and Oakley, B. (1987). Neural brain-derived neurotrophic factor, neurotrophin-3, and TrkC-
induction of taste buds. J. Comp. Neurol. 260:224–232. deficient emrbyonic mice. J. Neurosci. 17:9113–9121.
Huang, E. J., Wilkinson, G. A., Farinas, I., Backus, C., Zang, K., Liebl, D. J., Mbiene, J-P., and Parada, L. F. (1999). NT4/5 mutant
Wong, S. L., and Reichardt, L. F. (1999). Expression of Trk mice have deficiency in gustatory papillae and taste bud for-
receptors in the developing mouse trigeminal ganglion: in mation. Dev. Biol. 213:378–389.
vivo evidence for Nt-3 activation of TrkA and TrkB in addi- Liu, X., and Jaenisch, R. (2000). Severe peripheral sensory
tion to TrkC. Development 126:2191–2203. neuron loss and modest motor neuron reduction in mice
Jones, K. R., Farinas, I., Backus, C. and Reichardt, L. F. (1994). with combined deficiency of brain-derived neurotrophic
Targeted disruption of the BDNF gene perturbs brain and sen- factor, neurotrophin 3 and neurotrophin 4/5. Dev. Dyn. 218:
sory neuron development but not motor neuron development. 94–101.
Cell 76:989–999. Liu, X., Ernfors, H., Wu, H., and Jaenisch, R. (1995). Sensory but
Jung, H-S., Oropeza, V., and Thesleff, I. (1999). Shh, Bmp-2, not motor deficits in mice lacking NT4 and BDNF. Nature
Bmp-4 and Fgf-8 are associated with initiation and patterning 375:248–241.
of mouse tongue papillae. Mech. Devel. 81:179–182. Ma, Q., Chen, Z., Barrantes, I. B., de la Pompa, J. L., and
Kaufman, M. H. (1992). The Atlas of Mouse Development. Anderson, D. J. (1998). neurogeninl is essential for the deter-
Academic Press, San Diego, pp. 421–424. mination of neuronal precursors for proximal cranial sensory
Kossel, A. H., McPheeters, M., Lin, W., and Kinnamon, S. C. ganglia. Neuron 20:469–482.
(1997). Development of membrane properties in taste cells of Mbiene, J.P., and Farbman, A. I. (1993). Evidence for stimulus
fungiform papillae: Functional evidence for early presence of access to taste cells and nerves during development: an elec-
amiloride-sensitive sodium channels. J. Neurosci. tron microscopic study. Micros. Res. Tech. 26:94–105.
17:9634–9641. Mbiene, J.-P., and Mistretta, C. M. (1997). Initial innervation of
Krimm, R. F., and Hill, D. L. (1998). Innervation of single fungi- embryonic rat tongue and developing taste papillae: nerves
form taste buds during development in rat. J. Comp. Neurol. follow distinctive and spatially restricted pathways. Acta
398:13–24. Anat. 160:139–158.
Krimm, R. F., and Hill, D. L. (1999). Early dietary sodium Mbiene, J.-P., Macallum, D. K., and Mistretta, C. M. (1997).
restriction disrupts the peripheral anatomical development of Organ cultures of embryonic rat tongue support tongue and
the gustatory system. J. Neurobiol. 39:218–226. gustatory papilla morphogenesis in vitro without intact sen-
Krimm, R. F., and Hill, D. L. (2000). Neuron/target matching sory ganglia. J. Comp. Neurol. 377:324–340.
between chorda tympani neurons and taste buds during post- McLaughlin, S. K. (2000). Erb and c-Kit receptors have distinc-
natal rat development. J. Neurobiol. 43:98–106. tive patterns of expression in adult and developing taste papil-
Krimm, R. F., Miller, K. K., Davis, B. M., and Albers, K. M. lae and taste buds. J. Neurosci. 20:5679–5688.
(2001). Epithelial overexpression of BDNF or NT4 disrupts Mistretta, C. M. (1972). Topographical and histological study of
targeting of taste neurons that innervate the anterior tongue. the developing rat tongue, palate and taste buds. In Third
Dev. Biol. 232:508–521. Symposium on Oral Sensation and Perception: The Mouth of
Taste System Development 781

the Infant, J. F. Bosma (Ed.). Charles C Thomas, Springfield, Nosrat, C. A., and Olson, L. (1995). Brain-derived neurotrophic
IL, pp. 163–187. factor mRNA is expressed in the developing taste bud-bearing
Mistretta, C. M. (1991). Developmental neurobiology of the taste tongue papillae of rat. J. Comp. Neurol. 360:698–704.
system. In Smell and Taste in Health and Disease, T. V. Nosrat, C. A., Ebendal, T., and Olson, L. (1996). Differential
Getchell, R. L. Doty, L. M. Bartoshuk, and J. B. Snow (Eds.). expression of brain-derived neurotrophic factor and neu-
Raven Press, New York, pp. 35–64. rotrophin 3 mRNA in lingual papillae and taste buds indicates
Mistretta, C. M. (1998). The role of innervation in induction and roles in gustatory and somatosensory innervation. J. Comp.
differentiation of taste organs: introduction and background. Neurol. 376:587–602.
Ann. NY Acad. Sci. 855:1–13. Nosrat, C. A., Blomlof, J., ElShamy, W. M., Ernfors, P., and
Mistretta, C. M., and Baum, B. J. (1984). Quantitative study of Olson, L. (1997). Lingual deficits in BDNF and NT3 mutant
taste buds in fungiform and circumvallate papillae of young mice leadung to gustatory and somatosensory disturbances,
and aged rats. J. Anat. 138:323–332. respectively. Development 124:1333–1342.
Mistretta, C. M., and Bradley, R. M. (1978). Taste responses in Nosrat, C. A., MacCallum, D. K., and Mistretta, C. M. (2001).
sheep medulla: Changes during development. Science Distinctive spatiotemporal expression patterns for neu-
202:535–537. rotrophins develop in gustatory papillae and lingual tissues in
Mistretta, C. M., and Bradley, R. M. (1983). Neural basis of embryonic tongue organ cultures. Cell Tissue Res. 303:35–45.
developing salt taste sensation: Response changes in fetal, Oakley, B. (1988). Taste bud development in rat vallate and foliate
postnatal, and adult sheep. J. Comp. Neurol. 215:199–210. papillae. In Mechanoreceptors, Hnik, P., Soukup, T., Vejsada,
Mistretta, C. M., and Hill, D. L. (1995). Development of the taste R., and Zelena, J. (Eds.). Plenum Press, New York, pp. 17–22.
system. Basic neurobiology. In Handbook of Olfaction and Oakley, B., Brandemihl, A., Cooper, D., Lau, D, Lawton, A., and
Gustation, R. L. Doty (Ed.). Marcel Dekker, New York, pp. Zhang, C. (1998). The morphogenesis of mouse vallate gusta-
635–668. tory epithelium and taste buds require BDNF-dependent taste
Mistreta, C. M., and Labyak, S. (1994). Maturation of neuron neurons. Brain Res. 105:85–96.
types in nucleus of solitary tract associated with functional Paul, G., and Davies, A. (1995). Trigeminal sensory neurons require
convergence during development of taste circuits. J. Comp. extrinsic signals to switch neurotrophin dependence during the
Neurol. 345:359–376. early stages of target field innervation. Dev. Biol. 171:590–605.
Mistretta, C. M., Gurkan, S., and Bradley, R. M. (1988). Patel, T. D., Jackman, A., Rice, F. L., Kucera, J., and Snider, W.
Morphology of chorda tympani fiber receptive fields and pro- D. (2000). Development of sensory neurons in the absence of
posed neural rearrangements during development. J. NGF/TrkA signaling in vivo. Neuron 25:345–357.
Neurosci. 8:73–78. Rao, H., Xu, Z., MacCallum, D. K., and Mistretta, C. M. (1997).
Mistretta, C. M., Goosens, K. A., Farinas, I., and Reichardt, L. F. BDNF and NGF differ in promoting neurite outgrowth from
(1999). Alterations in size, number, and morphology of gusta- cultured embryonic rat geniculate and trigeminal ganglia.
tory papillae and taste buds in BDNF null mutant mice Proc. of the Internatl. Symposium on Olfaction and Taste XII
demonstrate neural dependence of developing taste organs. J. and AChemS XIX, San Diego, CA. Chem. Senses 22:775.
Comp. Neurol. 409:13–24. Ringstedt, T., Ibanez, C. F., and Nosrat, C. A. (1999). Role of
Mistretta, C. M., Gaffield, W., Grabauskiene, S., and MacCallum, brain-derived neurotrophic factor in target invasion in the gus-
D. K. (2000). Fungiform papillae develop in increased num- tatory system. J. Neurosci. 19:3507–3518.
bers and atypical locations in cyclopamine-treated rat tongue Rochlin, M. W., and Farbman, A. I. (1998). Trigeminal ganglion
cultures. Chem. Senses 25:628. axons are repelled by their presumptive targets. J. Neurosci.
Mistretta, C. M., Gaffield, W., Liu, H-X., and MacCallum, D. K. 18:6840–6852.
(2002). Cyclopamine and jervine in embryonic rat tongue cul- Rochlin, M. W., O’Connor, R., Giger, R. J., Verhaagen, J., and
tures demonstrate a role for Shh signaling in taste papilla Farbman, A. I. (2000). Comparison of neurotrophin and repel-
developemnt and patterning: fungiform papillae double in lant sensitivities of early embryonic geniculate and trigeminal
number and form in novel locations in dorsal lingual epithe- axons. J. Comp. Neurol. 422:579–593.
lium. Dev. Biol. in press. Schecterson, L. C., and Bothwell, M. (1992). New roles for neu-
Morris-Wiman, J., Basco, E., and Du, Y. (1999). The effects of rotrophins are suggested by BDNF and NT-3 mRNA expres-
-bungarotoxin on the morphogenesis of taste papillae and sion in developing neurons. Neuron 9:449–463.
taste buds in the mouse. Chem. Senses 24:7–17. Scott, L., and Atkinson, M. E. (1998). Target pioneering and early
Nagai, T., Mistretta, C. M., and Bradley, R. M. (1988). morphology of the murine chorda tympani. J. Anat. 192:91–98.
Developmental decrease in size of peripheral receptive fields Settles, A. M., and Mierson, S. (1993). Ion transport in rat tongue
of single chorda tympani nerve fibers in relation to increasing epithelium in vitro: A developmental study. Pharmacol.
NaCl taste sensitivity. J. Neurosci. 8:64–72. Biochem. Behav. 46:83–88.
Nagato, T., Matsumoto, K., Tanioka, H., Kodama, J., and Toh, H. Sloan, H. E., Hughes, S. E., and Oakley, B. (1983). Chronic
(1995). Effect of denervation on morphogenesis of the rat impairment of axonal transport eliminates taste responses and
fungiform papilla. Acta Anatom. 153:301–309. taste buds. J. Neurosci. 3:117–123.
782 Mistretta and Hill

Snider, W. D. (1994). Functions of the neurotrophins during ner- Timmes, A., Nguyen, T., Lindsay, R. M., Acheson, A., and
vous system development: what the knockouts are teaching DiStefano, P. S. (1998). Axotomy upregultaes the anterograde
us. Cell 77:627–638. transport and expression of brain-derived neurotrophic factor
Sollars, S. I., and Bernstein, I. L. (1994). Amiloride sensitivity in by sensory neurons. J. Neurosci. 18: 4374–4383.
the neonatal rat. Behav. Neurosci. 108:981–987. Tucker, K. L., Meyer, M., and Barde, Y. A. (2001). Neurotrophins
Sollars, S. I., and Bernstein, I. L. (1996). Neonatal chorda tym- are required for nerve growth during development. Nat.
pani transection alters adult preference for ammonium chlo- Neurosci. 4:29–37.
ride in the rat. Behav. Neurosci. 110:551–558. Ulupinar, E., Jacquin, M., and Erzurumlu, R. S. (2000).
Sollars, S. I., Shuler, M. G., and Hill, D. L. (1996). Disappearance Differential effects of NGF and NT-3 on embryonic trigemi-
of fungiform papillae and taste pores in rats with unilateral nal axon growth patterns. J. Comp. Neurol. 425:202–218.
chorda tympani section at 10 days postnatal. Chem. Senses Von Vintschgau, M., and Honigschmied, J. (1877). Nervus glos-
21:673. sopharyngeus and Schmeckbecher. Pflugers Arch. 14:443–448.
Sollars, S. I., Smith, P. C., and Hill, D. L. (2002). Time course of Vogt, M. B., and Mistretta, C. M. (1990). Convergence in mam-
morphological alterations of fungiform papillae and taste buds malian nucleus of solitary tract during development and func-
following chorda tympani transection in neonatal rats. J. tional differentiation of salt taste circuits. J. Neurosci. 10:
Neurobiol. 51:223–236. 3148–3157.
Stewart, R. E., Lasiter, P. S., Benos, D. J., and Hill, D. L. (1995). White, F. A., Chiaia, N. L., MacDonald, G. J., and Rhoades, R. W.
Immunohistochemical correlates of peripheral gustatory sen- (1995). Birth dates and survval after axotomy of neurochemi-
sitivity to sodium and amiloride. Acta Anatom. 153:310–319. cally defined subsets of trigeminal ganglion cells. J. Comp
Stewart, R. E., DeSimone, J. A., and Hill, D. L. (1997). New per- Neurol. 352:308–320.
spectives in gustatory physiology: Transduction, develop- Witt, M., and Reutter, K. (1997). Scanning electron microscopi-
ment, and plasticity. Am. J. Physiol. 272:C1–C26. cal studies of developing gustatory papillae in humans. Chem.
St. John, S. J., Markison, S., and Spector, A. C. (1995). Salt dis- Senses 22:601–612.
criminability is related to number of regenerated taste buds Witt, M., and Reutter, K. (1998). Innervation of developing
after chorda tympani nerve section in rats. Am. J. Physiol. human taste buds. An immunohistochemical study.
269:R141–R153. Histochem. Cell Biol. 109:281–291.
Stone, L. S. (1933). Independence of taste organs with respect to Yamada, T. (1980). Chorda tympani responses to gustatory stim-
their nerve fibers demonstrated in living salamanders. Proc. uli in developing rats. Jpn. J. Physiol. 30:631–643.
Soc. Exp. Biol. Med. 30:1256–1257. Ye, Q., Heck, G. L. and DeSimone, J. A. (1993). Voltage depen-
Stone, L. S. (1940). The origin and development of taste organs dence of the rat chorda tympani response to Na salts:
in salamanders observed in the living condition. J. Exp. Zool. Implications for the functional organization of taste receptor
83:481–506. cells. J. Neurophysiol. 70:167–178.
Stone, L. M., Finger, T. E., Tam, P. P. L., and Tan, S.-S. (1995). Zeng, Q., Kwan, A., and Oakley, B. (2000). Gustatory innerva-
Taste receptor cells arise from local epithelium, not neuro- tion and bax-dependent caspase-2: participants in the life and
genic ectoderm. Proc. Natl. Acad. Sci. 92:1916–1920. death pathways of mouse taste receptor cells. J. Comp.
Takeuchi, H., Ido, S., Kaigawa, Y., and Nagai, T. (1997). Taste Neurol. 424:640–650.
disks are induced in the lingual epithelium of salamanders Zhang, C., Brandemihl, A., Lau, D., Lawton, A., and Oakley, B.
during metamorphosis. Chem. Senses 22:535–545. (1997). BDNF is required for the normal development of taste
Tonra, J. R., Curtis, R., Wong, V., Cliffer, K. D., Park, J. S., neurons in vivo. NeuroReport 8:1013–1017.
37

Contemporary Measurement of Human Gustatory Function

Marion E. Frank, Thomas P. Hettinger, Michael A. Barry

University of Connecticut Health Center, Farmington, Connecticut, U.S.A.

Janneane F. Gent
Yale University, New Haven, Connecticut, U.S.A.

Richard L. Doty
University of Pennsylvania, Philadelphia, Pennsylvania, U.S.A.

I. INTRODUCTION discrimination from human observers. Magnitude esti-

mates and absolute detection thresholds for sweet, salty,
The chemosensory systems of taste, smell, and chemical sour, and bitter stimuli are measures of stimulus strength
irritation all discriminate among strengths and qualities of and are by far the most common taste data. Although taste-
chemical stimuli. However, the three chemosensory sys- quality categories are well defined, tests of taste-quality
tems utilize distinct neural pathways, have receptors that identification and discrimination are less well developed.
are distributed differently, and provide distinct information The rationale for focusing on measures of taste intensity of
about chemical stimuli (Frank and Rabin, 1989). Strategies prototypical stimuli representing four taste qualities is
for measuring taste function reflect its special characteris- based on the concept that taste is truly a set of independent
tics. The sense of taste, a special visceral chemical sense sensory systems (McBurney and Gent, 1979) that may be
with receptors located within the oral cavity, discriminates addressed separately (Bartoshuk, 1989a,b; Frank and
chemical features that indicate nutrient or poison Smith, 1991). In contrast, tests of odor identification (Cain,
(Bartoshuk, 1991; Frank et al., 1992; Scott and Mark, 1989; Doty et al., 1984) and discrimination (Wright, 1987)
1987). It is generally accepted that most taste perceptions use a variety of odors without purporting to assess the
are sweet, salty, sour, or bitter. “Nontraditional” tastes gamut of odor qualities (Dravnieks, 1985). It is interesting
(Hettinger et al., 1990) often are flavor sensations derived that measures of threshold and identification for qualita-
from retronasal stimulation of olfactory receptors (see tively distinct odors are highly correlated, suggesting that
Chapters 10, 22, and 44). The savory umami quality may be olfaction is a synthetic sensory system. Such is not the case
an exception (Hettinger et al., 1996; Yamaguchi, 1991). In for taste; correlations among measures of sweet, salty,
contrast, smells are more diverse, defying consistent per- sour, and bitter function are low (Cowart et al., 1997).
ceptual categorization (Beets, 1971), and irritations may Assessments of gustatory function that bypass human
form one perceptual category (Green and Lawless, 1991). judgments include evoked potentials (electroencephalogra-
Taste testing most frequently measures functional com- phy) and magnetoencephalography (Kobayakawa et al.,
petence based on judgments of human subjects, an 1999; Plattig, 1991), as well as a number of brain imaging
approach with origins in the late nineteenth century techniques (see below). Complicated computerized chem-
(Bartoshuk, 1978). Psychophysical techniques are ical stimulus–delivery devices are used because reliable
designed to systematically obtain ratings of sensory inten- timing of multiple stimulus repetitions is required.
sity, sensory thresholds, quality identification, and quality Activating a chemosensory system electrically greatly

783
784 Frank et al.

simplifies these biological approaches; electric-taste stim- example, one may wish to evaluate preferences among
uli are promising in this regard but do not separately acti- product formulations with test panels made up of individ-
vate taste subsystems (Frank and Smith, 1991). uals without chemosensory dysfunction (Heymann and
This chapter focuses on contemporary assessments of Lawless, 1997).
taste function commonly employed in applied settings and Determination of a detection (absolute) threshold
suggests possible future assessments. Other basic psy- requires reliable identification of the presence of some-
chophysical paradigms, described for olfaction in thing, not a qualitatively discernable sensation, such as a
Chapter 10, may also be applied to gustation. The reader is sweet or salty taste. Classically, detection thresholds were
also referred to related chapters on the psychophysics of determined directly by procedures such as the methods of
taste mixtures (Chapter 38) and the genetics of human constant stimuli and limits or indirectly by the method of
taste perception (Chapter 40). adjustment (Fechner, 1860). (See below and Chapter 10
for more on classical psychophysical methods.) The
method of adjustment was used to directly measure how
II. WHOLE-MOUTH AND REGIONAL TASTE perceived intensities of taste stimuli are adjusted with the
TESTING passage of time or when presented as components of mix-
tures (e.g., Pangborn, 1984; Vanne et al., 1998; Woskow,
There are two general approaches to taste testing with dis- 1967). However, because of the large numbers of trials
tinct objectives: whole-mouth testing and regional testing. required, the possibility of sensory adaptation, and/or sub-
As noted elsewhere in this volume (see Chapters 32, 35, ject fatigue, other direct methods have been developed that
and 44), taste is composed of several subsystems with dis- minimize trials and yet provide a reliable threshold esti-
tinct taste bud fields that are associated with distinct mate. In Sec. II.A and II.B popular whole-mouth and
papillary structures [long known to contain taste-receptive regional techniques for determining taste detection thresh-
elements (Bell, 1803; Haller, 1763; Horn, 1825; Müller, olds formulated in the mid-twentieth century are illus-
1838)] and branches of cranial nerves (CN) (Frank et al., trated. They differ from each other in method of stimulus
1992). Branches of the facial nerve (CN VII), the chorda delivery and stringency of criterion performance. The
tympani and greater superficial petrosal nerves, innervate whole-mouth techniques use 3 drops (Henkin et al., 1963)
anterior tongue and palatal fields, respectively; in these or 8 cups (Harris and Kalmus, 1949) to present taste solu-
fields taste buds are dispersed among many papillae tions, a difference that affects the reliability at criterion
(Imfeld and Schroeder, 1992; Miller and Bartoshuk, 1991). performance as well as threshold values. The regional
Lingual branches of the glossopharyngeal nerve (CN IX) technique, electrogustometry, presents weak electrical
innervate posterior tongue fields; in these fields many currents via localized electrodes to stimulate “metallic”
hundreds of taste buds are concentrated in several large electric taste in the separate taste bud fields (Krarup, 1958;
circumvallate (Arey et al., 1935) and foliate papillae (Hou- Mackenzie, 1955).
Jensen, 1933). Taste buds located in the caudal pharynx
and larynx are innervated by the superior laryngeal branch A. Three Drops or Eight Cups
of the vagus nerve (CN X) (Lalonde and Eglitis, 1961);
they are sensitive to tonicity or pH and not thought to be The 3-drop and 8-cup techniques are two popular thresh-
involved in gustatory discrimination (Bradley, 2000). old procedures in the clinical taste literature (Gent et al.,
Functional differences of taste bud fields (Collings, 1974; 1997; Hertz et al., 1975). The two techniques
Dunér-Engström et al., 1986; Sandick and Cardello, 1981) (Weiffenbach et al., 1983) define threshold by criterion
(see also below) and knowing that disease and injury may performances that have quite different probabilities for
affect individual fields are rationales for regional taste test- false positives (Table 1). The probability of guessing
ing. With regional testing, function in different taste fields correctly (PG) at the threshold concentration is 0.259 for
is assessed separately by limiting stimuli to specific oral the 3-drop and 0.014 for the 8-cup technique, but additional
sites. Losses restricted to single taste bud fields due to a requirements at lower and higher concentrations
single malfunctioning nerve may go unnoticed (Bartoshuk, (associated events) reduce probabilities for false-positive
1989b), perhaps because of mutual inhibition among the thresholds.
separate fields (Catalanotto et al., 1993; Lehman et al., In the 3-drop test (1/3), subjects are presented, on
1995). In the more common whole-mouth testing, taste each trial, with three drops of liquid, one of which contains
stimuli are not restricted spatially and may reach all taste a taste stimulus; the other two contain pure water.
bud fields, as in most real-life experiences. Whole-mouth Threshold is the concentration at which the subject
testing is the norm in the food industry, where, for chooses the drop containing the stimulus either three trials
Contemporary Measurement of Human Gustatory Function 785

Table 1 Probabilities for False Positives in Threshold Tasks influence on repeated or sustained testing of taste when
Events intervening water rinses are not possible (see below).
When a stimulus is applied constantly to the same lingual
Task Trials correct Threshold Associated Combined
location, its perceived intensity will decrease markedly; in
1/2 5 in row 0.0313 0.500 0.0156 fact, under certain circumstances it disappears within a few
1/2 4 reversals 0.0156 0.250 0.00391 minutes (Bujas et al., 1991a; Ganzevles and Kroeze,
1/3 At least 2/3 0.259 0.192 0.0498 1987a, b; Gent and McBurney, 1978; McBurney, 1976).
4/8 1 errorless 0.0143 0.757 0.0108 Mouth movements tend to prevent complete adaptation
Each threshold test is identified by task (e.g., one of two stimuli contains (von Békésy, 1965). The spread of stimulus solution to
taste: 1/2) and trials correct for threshold event (e.g., five in row); see tongue recesses untouched by the original tastant applica-
text for additional information. The threshold event is the trial or series of tion and an increase in stimulus intensity at those sites may
trials on which the subject succeeds, and that stimulus is considered the
threshold stimulus. Associated events are requirements on trials before
be at play. Recovery from adaptation occurs rapidly after
and/or after the threshold trial. replacing the adapting stimulus with water (Bujas et al.,
1991a; Hahn, 1934). Both adaptation time and recovery
in a row or in two of three trials (PG
1/33  6/33). The time after rinsing vary with stimulus intensity and quality
subject must also perform as well at the next higher con- (Meiselman, 1968).
centration (probability
7/27) and make three incorrect Interestingly, a moderate degree of adaptation to 100
choices or two of three incorrect choices on three trials at mM NaCl may actually enhance differential sensitivity
the next lower concentration (PG
8/27  12/27
(McBurney et al., 1967). The Weber fraction ( I/I), where
20/27). The entire threshold sequence occurs by chance  I reflects the differential threshold from the standard (I),
with a probability of (7/27) (7/27) (20/27), or about 0.05 was 0.20 for subjects with unadapted tongues and 0.75 for
(Table 1). subjects with adapted tongues. McBurney (1984) argues
In the 8-cup test (4/8), subjects are presented on each that adaptation may serve to adjust the taste system (as
trial with 8 cups; 4 contain a taste stimulus and 4 contain well as other sensory systems) to be maximally sensitive to
pure water. The threshold event requires errorless perfor- changes in stimuli near a prevailing intensity.
mance; the subjects must divide the 8 cups into two
groups, one containing the taste stimulus (PG
(4/8) (3/7) B. Localized Electric Current
(2/6) (1/5)
1/70) (Table 1). The subjects must also make
two or more errors at the next lower concentration (PG
In electrogustometry, an electrical stimulus is applied to
53/70). The entire threshold sequence occurs by chance oral sites via a stimulus electrode, typically a small
with a probability of about 0.01 (Table 1). Thus, five metal disk (for review, see Frank and Smith, 1991).
thresholds might be false positives with the 3-drop tech- Electrogustometers have the benefit of being light and
nique to each one false positive with the 8-cup technique. portable and provide a well-defined stimulus to specific
In a direct comparison, the three-drop technique yielded regions of the oral cavity. There is no need for chemical
higher thresholds than the eight-cup procedure for both solutions, and the stimuli can be applied discretely in space
sucrose and NaCl (Weiffenbach et al., 1983) (Table 2), and time. Detection thresholds for anodal stimuli are estab-
which is no surprise. The volume of stimulus sampled is lished, which—for 12.5 mm2 stainless steel electrodes—
either a drop (0.1 mL) or a sip (1 mL), yielding a big average 8 A for the anterior tongue, 13 A for the
difference in numbers of taste buds contacted (Doty et al., posterior tongue, and 20 A for the palate in normal
2001; Smith, 1971). Also, taste receptors are saliva- Japanese subjects (Tomita et al., 1986). However, taste
adapted for the 3-drop (no water rinses) but water-adapted
for the 8-cup (includes water rinses). Both stimulus vol-
ume and rinsing affect taste measurements (Lawless,
1987), and they may both affect NaCl detection more than Table 2 3-Drop and 8-Cup Detection Thresholds for 14 People
sucrose detection (Weiffenbach et al., 1983). Choice of one Threshold in mM ( SEM)
of these two threshold techniques relies more on whether
3-drop 8-cup
one wishes to test taste bud receptors in a saliva-adapted or
water-adapted state (Rehnberg et al., 1992) than on differ- NaCl 38 (  7.2) 5.8 (  2.1)
ences in PG (Table 1). Sucrose 32 (  5.6) 12.9 (  1.9)
Temporal parameters of stimulation are important to NaCl ( p , 0.001) and sucrose ( p , 0.01) mean thresholds (6 SEM)
consider in taste testing (Halpern, 1997). For example, differ for the two methods.
sensory adaptation over time can have a profound Source: Data from Weiffenbach et al., 1983.
786 Frank et al.

thresholds are sensitive to the procedure by which values III. CONTEMPORARY TASTE TESTING
are obtained (Weiffenbach et al., 1983), including the spe-
A. Assessment of Gustatory Function
cific task and stringency of criterion performance. For
example, average electrogustometric thresholds for the
Specialty clinics focusing on disorders of the chemical
anterior tongue were 20 A when European subjects
senses have in the last two decades devised and evaluated
detected a current presented simultaneously with a tactile
psychophysical protocols designed to effectively and
stimulus (Krarup, 1958) and 3 A when subjects
efficiently assess taste function (Gent et al., 1997). Two
detected a current added to an ongoing tactile stimulus
measures of gustatory intensity processing (detection
(Fons, 1970).
thresholds and intensity ratings) and one measure of gus-
Standard psychophysical approaches are used to
tatory quality processing (quality identification) are
establish electric-taste thresholds. In a two-alternative
typically taken. Subjects sample chemical solutions from
forced-choice procedure (1/2) (Cain, 1989), paired
cups (“sip and spit”) and rinse with water between stimuli
current and blank are randomly and sequentially deliv-
for whole-mouth testing. Electrical or chemical stimuli are
ered through electrodes already positioned on the
restricted to the separate taste bud fields for regional test-
tongue until the subject chooses the current side five
ing. Throughout this section and Section IV, examples of
times in a row (Murphy et al., 1995; Sampson et al.,
measuring effects of age and taste disorders of common
1993). In one study, threshold values were, on average,
etiology on taste function are noted (see also Chapter 44).
4 A for 15 young subjects but 10 A for 17 seniors
(Murphy et al., 1995). A correct two-alternative forced
choice series (PG
1/25) must be preceded by an incor- 1. Whole-Mouth Taste Intensity
rect choice (PG
1/2), giving the threshold sequence a
Contemporary testing of intensity processing by the gusta-
chance probability of 1/64 (Table 1). In staircase proce-
tory system typically involves use of a single chemical
dures (see below) peri-threshold stimuli are repeatedly
compound to represent each of the taste qualities: sweet
sampled to increase reliability of threshold values (e.g.,
(e.g., sucrose), salty (e.g., NaCl), sour (e.g., citric acid),
Miller et al., 2001).
and bitter (e.g., quinine•HCl). Whole-mouth testing, in
Electrogustometry efficiently verifies suspected
chorda tympani nerve injury. Anodal currents of less which taste solutions are sipped and moved throughout the
than 40 A were undetected on the front of the tongue oral cavity, assesses functioning of all taste bud fields
after nerve section (Grant et al., 1989), and a greater simultaneously, either at threshold, which is “traversed” at
than 100-A side-to-side threshold difference is the onset of any stimulation (Halpern, 1997), or across a
an accurate predictor of a unilateral denervation range of effective stimulus concentrations.
from Bell’s palsy (Groves and Gibson, 1974).
Electrogustometry may be more powerful in a clinical a. Taste Detection Thresholds. Detection thresholds
setting (Ovesen et al., 1991; Stillman et al., 2000) than have been defined by a two-alternative forced-choice
testing with chemicals (chemogustometry), which is (1/2) tracking (staircase) procedure (see above) as the
more tedious and taxes patients’ already strained average of concentrations at which the last four of five
resources. In lung, breast, and ovarian cancer patients, reversals occurred. A “reversal” is either an increase in
for whom no differences were detected with chemogus- presented concentration if one choice is incorrect, or
tometry, the average anterior-tongue anodal threshold decrease in concentration if two sequential choices are cor-
was 30 A compared to 10 A in noncancer rect. A four-reversal threshold event (1wrong—2 right—1
patients (Ovesen et al., 1991). Recently, an automated wrong—2 right) has a PG of (1/2)(1/4)(1/2)(1/4)
1/64
procedure for determining electrogustometric thresh- and must be preceded by 2 correct responses; thus a “false
olds was developed and evaluated for test-retest positive” occurs with a probability of 1/256. Tracking is an
reliability (Lobb et al., 2000; Stillman et al., 2000). order of magnitude more stringent than the 3-drop test
Extending electrogustometry to assess supra-threshold described above (Table 1). Median “tracked” detection
taste function (Salata et al., 1991) with currents less thresholds for 32 normal controls less than 36 years of age
than 100 A could be valuable for future psychophysi- approximated 0.63 M for quinine sulfate and 1.6 mM for
cal and biological clinical evaluation (evoked potential, NaCl (Cowart, 1989). In comparison, average 8-cup NaCl
functional brain imaging) (see sec. IV). In the next threshold was 5.8 mM and average 3-drop NaCl threshold
section, contemporary testing as practiced in clinics was 38 mM (Table 2). The rigorous tracking identified no
specializing in chemosensory disorders is described age-related differences in whole-mouth detection thresh-
and evaluated. olds for either citric acid or quinine sulfate (Cowart, 1989),
Contemporary Measurement of Human Gustatory Function 787

yet citric acid is rated 22% and quinine is rated 40% lower magnitude matching protocol are used to normalize indi-
in intensity by seniors at high suprathreshold levels (Frank vidualistic numerical ranges (Bartoshuk, 1989a, b). The
et al., 1992; Weiffenbach et al., 1986). Whole-mouth ratio relations among intensities of the different stimuli
detection thresholds may not be the best “clinical are defined with magnitude matching, and responses are
indicators” (Bartoshuk, 1989a), perhaps in part because of not confined to categories or limited by a visual analog
compensatory mechanisms among the several taste-bud scale, as they may be with category-scale procedures. In
fields (Bartoshuk, 2000; Catalanotto et al., 1993; Lehman order to evaluate a patient’s level of function (normal or
et al., 1995; Matsuda and Doty, 1995). abnormal), comparisons are made between taste intensity
ratings for a particular patient and those of appropriate
normal controls.
b. Taste Intensity Rating Scales. Rating scales have Taste intensity rating using either procedure has identi-
been used to quantify responses to taste stimuli at fied effects of age and clinical conditions that are associ-
suprathreshold intensity levels, which represent much of ated with taste disorders. For example, to seniors, 18 mM
taste perceptual experience (Halpern, 1997). By the mid- citric acid and 180 M quinine sulfate had modest cate-
twentieth century, direct methods (e.g., Stevens, 1961), in gorical taste intensities, whereas to young subjects they
which subjects make direct judgments of “subjective” were strong (Table 3), (Cowart, 1989). Also, patients who
magnitude, had replaced classical indirect methods (e.g., had experienced head trauma rated citric acid concentra-
Fechner, 1860) (in which scales are derived from “objec- tions numerically lower than other patients (Deems et al.,
tive” measures) as scaling methods of choice. With direct 1991). A possible problem with these ratings is that indi-
magnitude estimation, subjects may simply rate the inten- viduals with a tendency for hyperbole/underestimation
sity of subjective perceptions relative to a self-determined may use high/low values to describe perceptual experi-
standard to produce a scale, whereas indirect scales are ences of many modalities (Cowart, 1989). The normaliza-
constructed from “just noticable difference” (JND) steps tion to a modality incidental to taste with magnitude
measured over the range of stimulus intensity. The multi- matching corrects for this problem; however, to be effec-
ple JNDs required may be measured directly or indirectly tive it requires subjects have normal hearing.
using the method of adjustment (see sec. II), also known as Taste intensity ratings for sucrose, NaCl, citric acid, and
the method of average error. The standard deviation of the quinine.HCl, normalized to loudness in a magnitude
mean intensity match of a comparison stimulus to a stan- matching protocol, increased sixfold in young subjects but
dard is considered the JND. Ratings scales are often fourfold in seniors with a 10-fold increase in stimulus
employed in industrial applications to establish the degree concentration (Bartoshuk, 1989b; Frank et al., 1992). This
to which the taste of food is liked. For example, Pangborn difference is consistent with seniors being unable to
and Giovanni (1984) had adults, whose intake of sweet discriminate between two binary mixtures of sucrose and
foods was documented with a questionnaire, directly rate tartaric acid with mixture components differing slightly in
both the degree of liking and the perceived intensity of intensity (Kaneda et al., 2000). Stimulus mixtures, which
sweetness of lemonade containing different amounts of are the preponderance of stimuli encountered in natural sit-
sucrose. In general, perceived taste intensity increased uations, may prove effective in assessing taste dysfunction
with sucrose concentration; however, for most individuals (Stevens, 1996; Stevens and Cain, 1993). Taste mixtures
lemonade with intermediate levels of sucrose was liked
most, yet there were some individuals who found either
high or low sucrose levels more likeable. The subjects’ Table 3 Mean Taste Intensities, 0 (No Taste) to 12 (Extreme
intake of sweet foods and their hedonic responses to the Taste), of Citric Acid and Quinine Sulfate by Two Age Groups
sweetness of lemonade were correlated.
Intensity ratings
Numerical category and magnitude-matching proce-
Citric acid Quinine sulfate
dures are the scaling methods (see Chapter 10 for more on
methodological detail) used commonly in contemporary Age group 1.8 m 18 m 18 m 180 m
clinical applications. In the former procedure, patients rate
19–35 years 5.0 9.2 3.5 9.4
taste intensities of concentration series within defined 65–79 years 3.2 7.5 2.0 7.3
numerical categories (e.g., 0–12, labeled as “no taste” to
“extremely strong taste”) (Cowart, 1989; Deems et al., These data were a portion of the measurements made on 137 normal
subjects from 10 to 79 years old. Statistical analysis of the entire data set
1991); in the latter, they concurrently rate the taste indicated a significant effect of age on citric acid and quinine sulfate
intensities and the loudness of a series of 1000 Hz tones on ratings.
self-consistent ratio scales. The loudness ratings in the Source: Data from Cowart, 1989.
788 Frank et al.

have been used primarily to address mechanisms 2. Whole-Mouth Taste Quality

employed by the gustatory system to evaluate components
A straightforward test of taste is to have subjects report
of complex stimuli (e.g., Bartoshuk and Gent, 1985;
whether a tastant produces a sweet, salty, sour, or bitter
Breslin and Beauchamp, 1995; Prescott et al., 2001).
sensation, although there is a bothersome semantic uncer-
Chapter 38 presents information on how the gustatory sys-
tainty in the use of taste quality names (Cowart et al.,
tem deals with stimulus mixtures, and the reader is referred
1997; Hettinger et al., 1999; Meiselman and Dzendolet,
to Chapter 10 for information on how the olfactory system
1967). Fractionated judgments, in which a subject rates the
deals with mixtures.
relative magnitude of the sweet, salty, sour, and bitter
An example of measurement of dysfunction with scal-
components, can also be made (e.g., Smith and McBurney,
ing procedures in those experiencing taste disorders comes
1969; van der Klaauw and Smith, 1995), but such
from studies of taste function in women with burning
techniques may require selection of subjects (Kuznicki
mouth syndrome (BMS), a disorder discussed in Chapter
et al., 1983; Ossebaard et al., 1997) and/or training in order
44. Magnitude-matched taste intensities of women with
to increase reliability (e.g., Settle et al., 1986).
BMS were 12% lower than controls overall, with higher
In contemporary chemosensory clinics, patients are
levels in NaCl and sucrose affected most (Formaker and
simply expected to identify the predominant taste quality
Frank, 2000). Also, it is well known that many medications
of members of concentration series of prototypal sweet,
reversibly alter taste function (Frank et al., 1992; Mott
salty, sour, and bitter solutions (Gent et al., 1997; Seiden
et al., 1993; Schiffman et al., 1999, 2000). (The reader is
et al., 1992). This involves identifying the taste quality of
also referred to Chapter 44 for more on this issue.) The
15–20 solutions that are suprathreshold to normal controls
resulting dysfunction of the gustatory system may be mea-
(Cowart, 1989; Formaker and Frank, 2000). Such identifi-
sured with scaling procedures. For example, treatment
cation tasks are “objective” in the sense scoring is correct-
with 0.12% chlorhexidine gluconate, a bitter drug used to
incorrect (Campbell, 1957; Weiffenbach, 1987). Although
control accumulations of oral bacteria, results in frequent
highly educated people perform at 90% correct—better
taste complaints and reduced patient compliance (Lang
than less educated people, who perform at 83% correct
et al., 1988). Magnitude-matched intensities of midrange
(Cowart, 1989)—age and medical conditions reported to
levels of NaCl and quinine •HCl were reduced 50% (Helms
affect taste (see Chapter 44) are reliably reflected in iden-
et al., 1995) following treatment, and similar effects were
tification performance. Seniors more than 65 years of age
observed with a fixed-interval, 10-point category scale
correctly identified the taste quality of 87% of the
(Frank et al., 2001). Thus, direct ratings of subjective
solutions presented, whereas people 35 years of age or
intensity permit across-subject comparisons and separate
younger identified 95% (Cowart, 1989). Head trauma
normal from abnormal function.
patients correctly identified the taste quality of 80% of the
Labeled magnitude scales (LMS) (Green et al., 1996)
solutions, other patients identified 84% (Deems et al.,
for taste have emerged recently as potential alternatives to
1991), and women with BMS identified 81% compared to
direct magnitude estimation scales that require normaliza-
92% for controls (Formaker and Frank, 2000). Stimuli
tion (Bartoshuk, 1989a) or to labeled linear category scales
rated low in intensity are associated with more mistakes
(Cowart, 1989; Deems et al., 1991) that can suffer from
(Formaker and Frank, 2000; Frank et al., 2001), suggesting
ceiling effects (Bartoshuk, 2000; Lucchina et al., 1998).
a redundancy in rating and identification data. Consistent
On one LMS, verbal labels of intensity are spaced quasi-
misidentification patterns, however, may identify specific
logarithmically rather than linearly. When subjects are
taste weaknesses (Gent et al., 1999) or distortions (Frank
instructed to treat the upper bound as “strongest imagin-
et al., 2001; Helms et al., 1995).
able oral sensation” or “strongest imaginable taste,” the
LMS produces psychophysical functions equivalent to
3. Regional Taste Function
those produced by magnitude estimation (i.e., more likely
to reflect an expansive continuum) (Green et al., 1996; Measurement of taste function on the separate regions of
Lawless et al., 2000). This particular LMS has proven to be the tongue may use either electrical (electrogustometry) or
a useful psychophysical tool to categorize a subject’s chemical (chemogustometry) stimuli (Frank and Smith,
genetically determined taster status with intensity ratings 1991). Electrogustometry, which is by nature a regional
of a concentration series of the bitter 6-n-propylthiouracil technique, was described above in the context of defining
(PROP) (Lucchina et al., 1998). PROP “supertasters,” regional detection thresholds. In contrast to electrogustom-
whose ratings accumulate at the ceiling of a fixed-interval etry, regional chemogustometry can separately address
category scale, are accommodated by the LMS tastes within each taste bud region, with measurements
(Bartoshuk, 2000). similar to those described above for whole-mouth testing.
Contemporary Measurement of Human Gustatory Function 789

Contemporary specialty chemosensory clinics deliver restricted to small lingual sites easily detected by younger
sweet, salty, sour, and bitter chemical solutions to the individuals.
separate taste bud fields via 5-mm-diameter filter paper Regional intensity ratings can identify dysfunction
discs (Tomita et al., 1986), cotton swabs (Bartoshuk, localized to specific taste bud fields not detected with
1989a), or pipettes (Deems et al., 1991). It is possible to whole-mouth ratings (Mott et al., 1994; Zuniga et al.,
thicken stimulus solutions with cellulose to minimize stim- 1994). For example, patients suffering from bulimia rated
ulus spread (Kroger et al., 1997) or present chemical stim- palatal taste stimuli less intense than controls, but lingual
ulants to circumscribed oral regions via gelatin cubes and whole-mouth ratings were normal (Rodin et al., 1990).
(Delwiche et al., 2000) or flow chambers attached to the Yet, following extraction of the four third molars, intensity
lingual surface (e.g., Bujas et al., 1991a, b; Kelling and ratings for whole mouth, palate and tongue were lower
Halpern, 1988). A microprocessor-controlled gustometer than controls (Shafer et al., 1999). Perhaps as the number
that is capable of delivering chemical stimuli to small of fields affected increases, whole-mouth function fails.
regions of the anterior tongue isolated by a vacuum seal Taste loss in specific oral regions, which may be
(Hebhardt et al., 1999) may rival electrogustometry in accompanied by reports of taste phantoms and distortions
stimulus control and sensitivity (Doty et al., 2001; (dysgeusia), has most often been associated with damage
Matsuda and Doty, 1995; Tunsuriyawong et al., 2000) in a to taste nerves. (See also Chapter 44 for the kinds of nerve
clinical setting. injury that may affect taste function and the effectiveness
Contemporary clinical chemogustometry compares of nerve repair.) Regional taste deficits have been mea-
either an individual patient’s “quality recognition” sured following neural damage stemming from middle-ear
thresholds or quality identification and intensity ratings for surgery/disease (Arnold, 1974; Bull, 1965; Chilla et al.,
suprathreshold stimuli to control values. Effects of aging 1982; Wiberg, 1971), facial/glossopharyngeal nerve dis-
on taste thresholds have been measured with the “filter- ease/compression (Groves and Gibson, 1974; May and
paper-disc” method (Tomita et al., 1986), in which Schlaepfer, 1975; Taillibert et al., 1998; Todrank and
concentration series are used to establish a threshold for a Bartoshuk, 1991), oral surgery (Deems et al., 1991; Jacks
“clear taste quality,” which is named by the subject. et al., 1998; Mott et al., 1994; Zuniga et al., 1994, 1997),
A stimulus is presented once via a filter paper disc, and if and head trauma (Bartoshuk et al., 1996; Costanzo and
no clear sensation (“no taste,” “undefined taste”) is indi- Zasler, 1991). Brain infarcts may also result in localized
cated, a stronger stimulus is presented. “Recognition” taste dysfunction (Fujikane et al., 1999).
thresholds determined with the filter paper discs showed
seniors (60–92 years) generally had “low to moderate B. Evaluation of Gustatory Assessments
hypogeusia” for taste bud fields on the anterior tongue,
posterior tongue, and palate. This conclusion was based on Many of the tests devised by contemporary specialty clin-
a doubling of average threshold concentration for sucrose, ics primarily address the processing of gustatory intensity
NaCl, and tartaric acid, and a quintupling of threshold con- perceptions with an expectation of establishing taste blind-
centration for quinine•HCl in the seniors compared to 11 to ness (ageusia) or weakness (Gent et al., 1997). But subjec-
29-year-olds (Tomita et al., 1986). This general increase in tive intensity, a quantitative perceptual variable, can be
taste thresholds contrasts with specific effects of age on influenced adventitiously. Distinguishing between a gen-
regional supra-threshold intensity ratings. Young people of uine taste weakness and “normal” taste function requires
22 years average age rated citric acid sourness on the front data from specific controls precisely matched to patient
of the tongue five times stronger than seniors over 74 years characteristics, including their age, which had not, until
of age; however, the two age groups rated sucrose similarly recently, been considered so critical a variable (see above;
(Bartoshuk, 1989b). The differences between regional rat- see also Chapter 44) (Frank, 1994; Gilmore and Murphy,
ings and quality thresholds cannot be explained by com- 1989; Murphy and Gilmore, 1989; Stevens et al., 1995).
pensatory mechanisms among several taste-bud fields The range of normal taste function at any age is wide, per-
(Bartoshuk, 2000) alone. But recruitment mechanisms haps reflecting in part varied numbers of taste buds present
(Matsuda and Doty, 1995) may be more prominent for on individuals’ tongues (Doty et al., 2001; Miller and
sweet taste than sour taste. If so, regional thresholds may Bartoshuk, 1991; Zuniga et al., 1993) and genetic taster
be more sensitive “clinical indicators” than regional status (see below and also Chapter 40). Controls based on
ratings. Results with signal-detection measures of taste genetic status should account, in part, for the measured
sensitivity derived from judgments of certainty (Matsuda variability. Also, the disordered gustatory processing of
and Doty, 1995) are consistent with this suggestion. Many taste quality manifest in those who experience dysgeusia
seniors were unable to detect high concentrations of NaCl has not been measured. Dysgeusia is considered the most
790 Frank et al.

crippling of taste disorders (Cowart et al., 1997; Deems identification of sweet, salty, sour, and bitter stimuli is
et al., 1996; Mott et al., 1993; Seiden et al., 1992) routinely assessed (Gent et al., 1997), and recognition
thresholds may be roughly determined (Tomita et al.,
1. Aging and Taste Function 1986); however, precise taste-quality recognition thresh-
olds (Collings, 1974), taste-quality discrimination, and
Diagnosis of hypogeusia, alterations in perceptions of taste
taste-quality profiles (Hettinger et al., 1990; Schiffman,
intensity, addressed with detection thresholds and magni-
1979) are not determined. Patients are asked to identify
tude ratings, requires comparison to age-matched controls.
taste quality, but clinical data are reported simply as total
“correct” (Cowart, 1989; Deems et al., 1991); even though
a. Thresholds. Detection-threshold concentrations,
patterns of errors could provide diagnostic information, as
stimulus levels at which subjects reliably detect the pres-
they do for smell (Cain, 1989). Their potential clinical
ence of a stimulus, are “objective” assessments. Adult age
value is suggested by frequent misidentifications of the
extremes (averages: 24 and 83 year) show a 10-fold differ-
salty prototype NaCl as bitter/sour following chlorhexidine
ence in average threshold concentration for sucrose, NaCl,
treatment (Frank et al., 2001; Helms et al., 1995) (see also
citric acid, and quinine•HCl (Bartoshuk et al., 1986). For
above). Taste-quality misidentification patterns could
example, the median threshold for citric acid was 15 M
provide objective information to supplement self-reported
for the younger subjects and 150 M for the older sub-
quality distortions by patients with dysgeusia (Cowart
jects. In fact, acid thresholds for only 3 of 18 young people
et al., 1997; Gent et al., 1987, 1997; Mott et al., 1993). See
overlapped values for the 18 seniors (Table 4). Age-
the next section for promising future evaluations that
specific distributions of normal thresholds are needed for
address this problem.
specific protocols to diagnose hypogeusia.

3. Suitability for Clinical Setting

b. Ratings. Intensity ratings address suprathreshold The dimensions of comprehensive taste testing could be
function, not function at threshold (Bartoshuk, 1989a). For daunting, given multiple taste substances and multiple
example, subjects rate intensities of 1.8–18 mM citric acid bilateral taste fields. If possible, taste evaluations should be
(Cowart, 1989) when threshold concentrations fall appropriate, concise, portable, faithfully administrable,
between 0.015 and 0.150 mM (Bartoshuk et al., 1986). and readily interpretable.
Intensity ratings are “subjective,” but control distributions
of ratings (Bartoshuk, 1989a,b) allow a patient to be
a. Appropriateness. In existing specialty chemosen-
described as falling at some percentile within normal func-
sory clinics, triage limits comprehensive taste testing to
tion or outside of normal function. Because “normal”
selected subpopulations of patients (Cowart et al., 1997;
seniors rate suprathreshold taste stimuli less strong than do
Deems et al., 1991; Mott et al., 1993; Seiden et al., 1992).
young subjects (Table 3) (Cowart, 1989), control distribu-
Justification for triage includes low frequencies of verified
tions of ratings used to diagnose taste weakness should be
taste complaints, which relates to patient confusions
age-matched to patients.
between taste and retronasal smell (see also Chapter 10).
Of 750 chemosensory patients evaluated at the University
2. Assessment of Taste-Quality Processing
of Pennsylvania Smell and Taste Center (Deems et al.,
Taste-quality discrimination is sparsely addressed 1991), two thirds complained of taste loss, but 87% of
currently in specialty chemosensory clinics. Taste-quality these “taste complainers” also complained of smell loss.
Test results verified taste loss in about 4% of all patients,
Table 4 Distribution of Detection Thresholds for Citric Acid about 6% of “taste complainers.” Verification of a com-
for 18 Young Subjects and 18 Seniors plaint, however, depends upon the test procedure. (See sec.
III for additional comments on the sensitivity of “clinical
Molarity (mM)
indicators.”)
0.01 0.011–0.05 0.051–0.1 0.11–0.5 0.5
To dispel confusion between chemosenses (taste, smell,
Young 4 11 3 0 0 common chemical), patients may need to be asked to focus
subjects on sweet, salty, sour, or bitter tastes in an initial interview
Seniors 0 0 6 9 3 (Gent et al., 1987; Mott et al., 1993). Of 493 patients of the
There is a statistically significant difference in average threshold for the
University of Connecticut (UConn) Taste and Smell
two age groups (Mann-Whitney U
13, p  0.001). Clinic, 42% complained about one, two, three, or four of
Source: Data from Bartoshuk 1986. the taste qualities. Distributions of “total” taste intensity
Contemporary Measurement of Human Gustatory Function 791

ratings for thusly defined “taste complainers” and the 58% (Stillman et al., 2000; Tomita et al., 1986). However,
“noncomplainer controls” are shown in Table 5. “Total” anodal current, which is used in electrogustometry, elicits
taste is the sum of intensity ratings for sucrose, NaCl, cit- “metallic,” “sour-salty” (Tomita et al., 1986), or “sour”
ric acid, and quinine•HCl concentration series. There is a (Bujas, 1971; Bujas et al., 1986) perceptions, usually asso-
highly significant difference in the distributions (p  ciated with salts and acids. In fact, anodal current may
0.001; -square), suggesting that many patients do, as they stimulate taste by driving cations (Na, H) through ion
report, perceive taste stimuli as less intense. Four percent channels (Herness, 1985a; Ninomiya and Funakoshi,
of “noncomplainer” controls rated taste intensities at 200 1989). Electrogustometry does not use cathodal current;
or less, but 28% of taste complainers rated taste intensities “cathode-on” (100A) elicits indistinct sweet-bitter
this low. Thus, about one quarter of taste complainers tastes (Bujas, 1971) accompanied by somatosensory sen-
had verified taste loss, and they may be the patients who sations and, at “cathode-off,” “metallic” sensations (Bujas,
would most benefit from comprehensive taste testing (see 1971; Herness, 1985b). Because regional weaknesses
also below). specific to any single taste quality—“dissociated” taste
problems—are uncommon (Tomita and Horikawa, 1986),
b. Management. The amount of time that the testing electrogustometry may be sufficient for evaluating most
takes (conciseness), its portability, and the likelihood it complaints of taste blindness or weakness in receptive
will be faithfully administered in various clinical settings fields for each of the taste nerves (e.g., Grant et al., 1989;
are important applied test-management considerations. Groves and Gibson, 1974) (see also above).

i. Conciseness. Most taste-testing protocols used in ii. Portability. Portable, self-administered taste tests
contemporary specialty chemosensory clinics require at would facilitate longitudinal monitoring of taste function.
least one hour of patient-staff contact (Deems et al., 1991). Highly portable, self-administered smell tests utilizing
The tests are primarily whole-mouth procedures that include microencapsulated “scratch-and-sniff” odor strips have
presentations of multiple concentrations of compounds rep- been developed (Doty et al., 1984), and their results corre-
resenting four taste qualities in order to determine detection late quite well with clinically administered tests of smell
thresholds and/or scale the subjective experience of sensory identification and threshold (Cain and Rabin, 1989).
intensity (Gent et al., 1997) for each of the qualities. Similar reliable taste tests are not yet available. Although
Sufficient data must have accumulated during the 20 years electrogustometers designed for clinic use are portable
of specialty chemosensory clinic operation in the United (Frank and Smith, 1991), test self-administration, at home,
States to support an evaluation of results aimed at establish- would not be advisable with current designs. A practical
ing efficient tests of whole-mouth taste function. means for easily assessing whole-mouth function is to
Comprehensive regional chemogustometry (six fields, have subjects chew pieces of filter paper impregnated
four qualities) is composed of 24 subtests in a single repli- with taste stimuli, a technique used to demonstrate
cate (Tomita et al., 1986); with 2 replicates this yields 48 the genetically based bimodal human sensitivity to
trials with rinses after each. By collapsing the quality phenylthiocarbamide (PTC) (Kalmus, 1971) and for PTC
domain to one, electrogustometry (Frank and Smith, 1991) taste phenotyping in genetic studies (Reed et al., 1999).
requires 6 subtests per replicate; and, because it does not Similar techniques using paper sticks (Hummel et al.,
require solutions and rinsing, it is a much faster procedure 2001), wafer, or tablet substrates (Hummel et al., 1997;
Ahne et al., 2000) could be self-administered and would be
sensitive to whole-mouth function, including saliva pro-
Table 5 Taste Intensity Ratings for “Taste Complainers” and
duction. Unfortunately, clinical utility of such approaches
“Controls”
is limited because regional testing may be required to
“Total” taste intensity ratings establish the specific nerves that are compromised in some
0–200 201–400 401–600 600 clinical situations.
Complainers 58 101 40 6
(28%) (49%) (20%) (04%) iii. Administration. Faithful administration of taste
Controls 11 160 99 18 tests is essential for consistent and comparable outcomes.
(04%) (56%) (34%) (06%) Many protocols used in specialty clinics require oversight
by a psychophysicist, precluding easy technology transfer
Entries are number of patients and, for each row, percentages in paren-
theses below. Frequency distributions for complainers and controls differ to clinical staff. Detection threshold protocols vary in com-
(-square
64.24, p  0.001). plexity and acuteness (Cowart, 1989; Weiffenbach et al.,
Source: Data from the UConn Taste and Smell Center. 1983) (also see above), although acuteness may not corre-
792 Frank et al.

late with sensitivity to patient reports of taste disorders. Table 6 Trial 1–2 Correlation for Whole-Mouth Taste Test
The sensitivity of lengthy, complex taste testing and short, Concentration series r n
simpler testing is yet to be directly compared (see below).
Sucrose 0.72 843
c. Interpretation. Divergent findings present in the NaCl 0.56 843
Citric acid 0.54 839
taste literature such as those reported on the effects of
Quinine•HCl 0.66 690
aging (see above and Chapter 44) likely reflect differences
n-Propylthiouracil 0.74 838
in reliability and sensitivity of the various measurements. “Total taste” 0.64 678
Unlike taste testing in studies designed to distinguish
among groups, clinical test interpretation is for individual Based on scatter plots, 0–2 outliers were identified and eliminated from
patients. Repeatability (or stability) of a measurement on each data set. Pearson’s correlation coefficients (r) are significant for each
concentration series and “total taste” (see above) ( p  0.01); n is the
retest (reliability) and validity of measurements for identi-
number of patients.
fying taste disorders are key here. Source: Data from the UConn Taste and Smell Center.

i. Reliability. Reliability for clinical tests is best

studied for the range of function seen clinically to assess taste disorder who test “negative” defines specificity. The
inherent measurement consistency. The correlation coeffi- proportion of all persons who test positive who actually
cients (Pearson’s r) between replicate measurements have a taste disorder defines positive predictive value.
calculated to quantify reliability (see also Chapter 10) are Ideally, each indicator would equal 1.00. These measures
very sensitive to the range of function represented. For of validity are yes-or-no proportions of specified popula-
example, replicate detection thresholds for n-butanol for tions; they do not address how well a test may document
32 normal people showing a narrow range of smell func- the severity of a disorder.
tion were more moderately correlated (r
0.68) (Cain and Validity of a clinical taste test is most easily
Gent, 1991) than replicate odor-identification scores addressed if it is known that a disorder is present. Thus,
(r
0.95) for 35 people with a wide range of smell func- we “created” a reversible dissociated taste disorder (bit-
tion (Doty et al., 1985). Few studies of the reliability of ter weakness) in 15 normal people by treatment with
taste measures have been conducted. chlorhexidine (Frank et al., 2001; Gent et al., 1999;
Correlation coefficients were calculated for two repli- Helms et al., 1995). Quinine•HCl and citric acid (con-
cate intensity ratings of sucrose, NaCl, citric acid, qui- trol) intensity ratings obtained from these people for
nine•HCl, and n-propylthiouracil (PROP) concentration concentration series were compared to control-popula-
series obtained from patients of the UConn Taste and tion percentiles used for diagnosis by the UConn Taste
Smell Clinic (Table 6). The subjects included taste com- and Smell Clinic. With weakness defined by a rating
plainers and noncomplainers (see above). Measurement below the 15th percentile (Gent et al., 1997), 10 of the
“reliability” for the intensity ratings ranged from r
0.54 15 people were positive for bitter weakness—a test sen-
for citric acid to r
0.74 for PROP. Sensitivity to this sitivity (“hit rate”) of 0.67. Thirteen of the 15 people
phenylthiourea-like compound is known to vary much were negative (greater than 15th percentile rating) for
among “tasters” and “nontasters” (see below). The 0.72 sour-weakness—a “correct rejection rate” of 0.87, about
coefficient for sucrose ratings compares to a 0.86 value for as expected for the criterion. When the criterion was set
detection thresholds in 12 normal subjects, retested on the at the 5th percentile, the hit rate was 0.60, but the speci-
same day (Mattes, 1988). The moderate coefficients for ficity was 0.93. This illustrates the trade-off between
replicate ratings on a single occasion justify including true and false positives.
duplicate ratings in “subjective” tests of suprathreshold To calculate the “positive predictive value” for a clin-
taste function. There is a need for additional evaluation of ical taste test, we assumed that UConn Taste and Smell
measurement reliability for all clinical taste testing. Clinic patients who presented with a taste complaint
(defined as above) (Table 5) had a taste disorder. For
ii. Validity. A test is valid if it accurately measures 288 “noncomplaining” controls, a “total” taste intensity
what it purports to measure. A clinical test has validity if it rating (see above) of 270 falls at the 15th percentile,
identifies genuine taste dysfunction. Consistent with yielding 43 false positives. Of 205 complainers, 86 (42%)
clinical test validity are sensitivity, specificity, and positive rated taste intensities at least this low: a test sensitivity of
predictive value for taste disorders (Doty, 1992). The pro- 0.42 for hypogeusia (total taste weakness). The propor-
portion of persons with a taste disorder who test “positive” tion of complainers testing positive of all patients testing
defines sensitivity. The proportion of persons without a positive (complainers and noncomplainers) was 86/129, a
Contemporary Measurement of Human Gustatory Function 793

0.67 positive predictive value. Thus, two thirds of 129 tions.” Chlorhexidine treatment resulted in a greater reduc-
people who tested positive had a taste disorder, and one tion in NaCl intensity ratings for low than for high concen-
third did not have a taste disorder. If a 5th percentile cri- trations (Lang et al., 1988) and an increased slope. In con-
terion were adopted, test scores of less than 210 would trast, a slope flattening resulted from greater age-related
indicate taste weakness, resulting in a hit rate of 0.30 and decrements in taste intensity ratings for high than for low
a positive-predictive value of 0.82. This illustrates the stimulus concentrations (Bartoshuk et al., 1986; Frank et
trade-off between positive predictive value and hit rate al., 1992; Schiffman and Clark, 1980; Schiffman et al.,
when test values for afflicted and normal populations 1981, 1981a), an effect also seen in women suffering from
overlap extensively. By identifying and segregating oral burning (Formaker and Frank, 2000). Just as distinct
sources of taste score variability, it may be possible to slope changes were observed in chorda-tympani nerve
identify more taste disorders without drastically reducing responses for NaCl concentration series in desalivated
the positive predictive value. (decrease) and sodium-deprived (increase) rats (Catalanotto
Because of the wide and overlapping range of taste func- et al., 1986), distinct psychophysical function signatures
tion for young people and seniors or complainers and non- may designate distinct biological conditions. If slope dif-
complainers (e.g., Tables 4 and 5), it may be necessary to ferences reflect differences in how well intensities are
evaluate taste-test validity using different cutoffs for specific discriminated, as they may (Kaneda et al. 2000), intensity
subpopulations. For example, for the measure of total taste discrimination tests for pairs of low and high stimulus
weakness just addressed, a criterion of 5th percentile may be levels could provide “objective” diagnostics for specific
appropriate for noncomplainers but 15th percentile for com- taste disorders.
plainers. This “good-faith” weighting of scores based on
complainer status results in a sensitivity of 0.42, a specificity
2. Recognition Thresholds
of 0.95, and a positive-predictive value of 0.87 for the dis-
tributions in Table 5. The measures of validity might Recognition thresholds are the lowest concentrations for
improve if subpopulations were further divided on the basis which stimulus quality can be identified. Unfortunately,
of age or genetic taster status (see below). In the next sec- taste stimuli may have several qualities, especially at
tion, promising future tests of taste function are described. weaker intensities (Bartoshuk et al., 1978; Halpern, 1997).
Nonetheless, quality recognition thresholds can be mea-
sured “objectively” if a “correct” quality is defined
IV. POTENTIAL FUTURE TESTING (Weiffenbach, 1983). For example, subjects may be asked
to select which drop (two water, one sucrose) has a taste
Psychophysical measures of taste stimulus discrimination, and to name its taste quality (Henkin and Christiansen,
biological assessments of gustatory CNS structure/ 1966). If they choose the sucrose, they are correct in
function, and the identification of genetic taster status hold choosing the taste (detection threshold); if they label it
promise for future clinical use. sweet, they have “correctly” recognized its taste quality.
Choosing sweet, salty, sour, and bitter decreases “false
positives” for the 3-drop task (Table 1) by one quarter to
A. Taste Stimulus Discrimination
0.0125. If the number of labels from which subjects
choose is not specified, probabilities for “false positives”
“Objective” intensity discrimination, quality recognition
(Schiffman et al., 1990) cannot be calculated. Two distrib-
threshold, and taste stimulus identification paradigms may
utions of detection and recognition thresholds for sucrose
have value in the diagnosis of cases of hypoguesia and
(Henkin and Christiansen, 1967) are presented in Table 7.
dysgeusia.
Before the tongue was anesthetized, median recognition
threshold (30 mM) was higher than detection threshold
1. Intensity Discrimination
(6 mM); with the tongue anesthetized, recognition and
Intensity ratings from low to high stimulus levels may indi- detection thresholds were at about the same high value
rectly address intensity discrimination. The slope of the (150 mM). Apparently, 150 mM sucrose could be detected
psychophysical function illustrates how sharply “subjec- with little lingual sensory input.
tive” intensity grows with concentration and may correlate A triad threshold procedure, used to compare thresholds
with how well subjects “objectively” discriminate between of many chemical compounds for young and senior sub-
concentration levels. Magnitude estimate data on chlorhex- jects, defines the detection threshold event as correct selec-
idine treatment, aging, and burning mouth suggest that the tion of the stimulus (presented in one of three cups) on
“slopes” may be affected distinctly by different “condi- three consecutive trials (Schiffman et al., 1979). A variant
794 Frank et al.

Table 7 Distribution of Detection and Recognition Thresholds acute for the foliate papillae, with mean threshold falling at
for Sucrose for 11 Subjects Before and After Lingual Anesthesia 0.7 mM, and quinine•HCl recognition was most acute on
Molarity (mM) the soft palate, with mean threshold falling at 13 M.
Adapting this Sw-Sa-So-Bi recognition threshold pro-
3 6 12 30 60 150 300
tocol to the clinical setting, either in a whole-mouth or a
Normal regional form, would be a worthwhile achievement. It
Detect 2 7 2 0 0 0 0 should be possible to correct for unequal individualistic
Recognize 0 1 3 7 0 0 0 guessing rates (Weiffenbach, 1983) by having a computer
Anesthetized keep frequency counts and adjusting reversal require-
Detect 0 0 0 2 3 4 2
ments. Recognition thresholds, like detection thresholds,
Recognize 0 0 0 0 0 8 3
are “objective,” but quality recognition may assess more
There is a significant difference in the detection and recognition thresh- closely the “real-world” taste domain and become a reli-
old medians in normal (p  0.01) but not anesthetized states (Wilcoxon able and valid addition to a test battery.
tests). The numbers of subjects at the median values are italicized.
Source: Data from Henkin and Christiansen, 1967.
3. Quality or Stimulus Identification
of this procedure has been employed for years in industry As noted in detail in Chapter 38, intensity ratings may be
and is known as the “triangle test.” On each trial the con- obtained for a “predominant quality” of a stimulus
centration is increased. “False-positive” probability is (Bartoshuk, 1989b; Tennissen and McCutcheon, 1996)
(1/3) (1/3) (1/3) (2/3)
2/81 (0.0247) (compare to Table or alternately for its Sw-Sa-So-Bi components
1): three correct choices preceded by an incorrect choice. (McBurney and Bartoshuk, 1973; Sandick and Cardello,
A similar three-correct criterion was used for recognition 1981; Smith and McBurney, 1969; Smith and van der
threshold but no quality-label set was specified. Under Klaauw, 1995). The result for the latter procedure is a
these “open-label” conditions, about half the subjects had Sw-Sa-So-Bi profile of perceptual intensity for each
not recognized NaCl as “salty” even at 64 times the detec- subject, which may accurately identify effects of disor-
tion threshold concentration. However, with a closed set of ders or experimental manipulations (Ossebaard et al.,
four labels (sweet, salty, sour, bitter), salty recognition for 1997) on taste quality processing. Another way to profile
NaCl occurred, on average, at twice the detection concen- taste quality is to have a group of subjects choose qual-
tration, with an individual maximum of 10 times (Henkin ity labels for stimuli and calculate frequencies across
and Christiansen, 1967). Thus, it is practical and efficient subjects. Like Sw-Sa-So-Bi intensity profiles, Sw-Sa-
to restrict label choice in quality-recognition testing. So-Bi frequency profiles for NaCl differ for the front and
Recognition thresholds for stimuli of four distinct taste back of the tongue. NaCl is most frequently “salty” for
qualities (sweet, salty, sour, bitter: Sw-Sa-So-Bi) have the front and “sour” for the back; congruently, its salty
been measured simultaneously, with random presentation intensity is rated highest on the front and its sour inten-
of four solutions (Collings, 1974; Collings et al., 1976; sity rated highest on the back (Sandick and Cardello,
McBurney and Moskat, 1975). If the stimulus was sucrose, 1981). Profiling is employed in industrial settings to
the correct response was “sweet,” if NaCl, “salty,” if acid, determine the subcomponents of an overall perceptual
“sour”; and if quinine, “bitter”. Solution-soaked, 4-mm- effect (Heymann and Lawless, 1997).
diameter filter paper disks were used to measure regional Subjects given additional quality labels, such as
recognition thresholds (Collings, 1974) with a forced- “soapy,” “sulfurous,” or “metallic” (Hettinger et al., 1990),
choice, six-reversal tracking procedure. “False-positive” generate more detailed frequency profiles, especially if
probability for a six-reversal threshold sequence (one stimuli have nongustatory aspects. Table 8 presents profile
wrong—one right—one wrong—one right—one wrong— frequencies for Polycose® (Ross Laboratories, Columbus,
one right) combined with an initial reversal is (3/4) (1/4) OH), a dietary supplement composed of a mixture of glu-
(3/4) (1/4) (3/4) (1/4)
27/4096 (1/4)
0.00164. cose-containing oligosaccharides (Hettinger et al., 1996)
Recognition of sweet sucrose and salty NaCl were most that may have a unique polysaccharide taste to rats (Feigin
acute for fungiform papillae, mean thresholds approximat- et al., 1987; Sclafani, 1987). At 3.2% (w/v) Polycose had a
ing 20 mM. Although few taste receptors were stimulated prominent “sweet” quality with the nose open but not with
and intensity ratings increase with numbers of receptors the nose clamped. Vanillin, benzaldehyde, isoamyl acetate,
stimulated (Smith, 1971), the value is similar to 3-drop, and cinnamaldehyde also have “sweet” odors (Dravnieks,
whole-mouth recognition thresholds (Henkin and 1985). At higher concentrations, Polycose had a sweet
Christiansen, 1966). Sour citric acid recognition was most taste and some other olfactory quality.
Contemporary Measurement of Human Gustatory Function 795

The above work emphasizes the need to control olfaction et al. (1991). Performance measures calculated from the
in taste-profiling studies. It is wise to profile taste stimuli matrix formed from the correct and incorrect responses
with the nose occluded to minimize olfactory influences. (Table 9) include percent correct and two measures taken
Taste stimuli may also be irritants at high concentrations from information theory (calculated in bits of information
(Green and Lawless, 1991), but proper selection of taste transferred) (Attneave, 1959) that quantify response
stimulus concentration may minimize such confounds consistency (T10, for a 10-stimulus test) and pairwise stim-
(Frank et al., 2001; Gilmore and Green, 1993; Portmann et ulus discrimination (T2).
al., 2001). Clearly, subjects integrate nongustatory sensations Confusion matrix methodology detected effects of
into ratings for concordant taste stimuli (Clark and Lawless, taste-altering compounds such as gymnemic acid (Gent
1994; Frank and Archambo, 1986; Frank et al., 1989; 1993; et al., 1999) and chlorhexidine gluconate (Gent et al.,
Hettinger et al, 1990; Schiffman et al., 1981, 1981b). 2002), demonstrating its potential utility as an “objective”
Taste-quality frequency profiles might be useful for clinical test of taste function. Gymnemic acid treatment
clinical tests if patients were presented with selected stim- imparts a specific deficit for the detection of sweet stimuli
uli numerous times. Patient profiles (or “taste spaces” (Frank et al., 1992); chlorhexidine treatment imparts taste
derived from multivariate analyses) could be compared distortions along with deficits in the detection of salty
with profiles (or spaces) for control subjects (e.g., and bitter stimuli, but sweet stimuli are unaffected (Frank
Schiffman, 1979; Schiffman and Dackis, 1975). Because et al., 2001).
of the serious impact of dysgeusia on patients’ lives Comparison of average TCMs for gymnemic acid-
(Deems et al., 1996), such comprehensive evaluation of treated to water-treated subjects is analogous to compar-
taste function, including hedonic ratings (Frank and ing performance of a dichromat to a trichromat on a
Archambo, 1986) for taste stimuli, may be warranted. test of color deficiency (e.g., Ishihara plates) and can
Measuring the confusion among taste stimuli in a task likewise provide an objective result. Comparison of Ts
in which stimuli are presented for identification multiple for particular pairs of stimuli may distinguish among dif-
times is a promising “objective” procedure for document- ferent kinds of dysgeusia, analogous to testing for
ing sensory dysfunction. Such an approach in the form of protanopia and deuteranopia with color matching.
a confusion matrix for odors (Köster, 1975; Wright, 1987) Although taste weakness may be more common among
was used as part of a clinical olfactory test battery (Wright dysgeusic patients (18%) than controls (7%), the need for
et al., 1991) (see also Chapter 10). More recently, the con- a test that clearly differentiates them is well recognized
fusion matrix paradigm was applied to taste (Gent et al., (Cowart et al., 1997).
1999, 2002; Hettinger et al., 1999). For taste confusion
matrix (TCM) studies, a number of taste stimuli (e.g., 10) B. Biological Correlates of Taste Function
are presented repeatedly (10 times each) for identification
from a list of stimulus labels. Stimulus names, not quality In previous sections we focused on psychophysical mea-
names, are used to avoid limitations imposed by the typi- surements that have a direct association to psychological
cal use of four qualities and to directly parallel Wright sensations and perceptions. They accurately and quantita-

Table 8 Chemosensory Quality Profiles for Polycose®

Quality names

Sweet SA/SO/BI SP/SU/ME Other None

Smell          
3.2% 5 0 1 2 3 0 3 0 0 8
10% 8 8 0 0 1 0 4 1 0 2
32% 9 10 0 0 0 0 3 0 0 0
Total 22 18 1 2 4 0 10 1 0 10

The category SA/SO/BI includes salty, sour, and bitter; SP/SU/ME includes soapy, sulfurous, and metallic. The 3 concentrations of Polycose elicit differ-
ent patterns of response (p  0.001), the pattern for 3.2% Polycose differs for nose-open (smell ) vs. nose-clamped (smell ) ( p  0.001), and total fre-
quencies across categories differ for nose-open vs. nose-clamped ( p  0.001). Data are for 10 subjects with one or two responses each.
Source: Data from Hettinger et al., 1996.
796 Frank et al.

Table 9 Calculation of Information Transferred, T10, for an Experimental Subject from the subject’s Taste Confusion Matrix
Response label
Salt Artificial Salt- Acid Quinine-
Stimulus Salt substitute MSG Quinine Acid Sugar sweetener sugar sugar sugar Row sum (X)
A. Matrix of Response Frequencies
NaCl 8 0 0 0 0 0 0 2 0 0 10
KCl 2 7 0 1 0 0 0 0 0 0 10
MSG 0 6 0 0 0 0 1 2 1 0 10
Quinine HCl 0 0 2 8 0 0 0 0 0 0 10
Citric Acid 0 0 0 0 8 0 0 0 1 1 10
Sucrose 0 1 1 1 1 0 4 0 1 1 10
Aspartame 0 0 0 1 2 1 4 0 2 0 10
NaCl-sucrose 6 0 0 0 0 0 0 4 0 0 10
Acid-sucrose 0 0 0 0 7 0 0 0 3 0 10
Quinine-sucrose 0 0 2 4 0 0 1 1 0 2 10
Column sums(Y) 16 14 5 15 18 1 10 9 8 4 100

Salt Artificial Salt- Acid Quinine-

Stimulus Salt substitute MSG Quinine Acid Sugar sweetener sugar sugar sugar px•log2(1/px)
B. Matrix of Information [pxy•log2(1/pxy)], where pxy
Cell Frequency/100
NaCl 0.29 0.00 0.00 0.00 0.00 0.00 0.00 0.11 0.00 0.00 0.33
KCl 0.11 0.27 0.00 0.07 0.00 0.00 0.00 0.00 0.00 0.00 0.33
MSG 0.00 0.24 0.00 0.00 0.00 0.00 0.07 0.11 0.07 0.00 0.33
Quinine HCl 0.00 0.00 0.11 0.29 0.00 0.00 0.00 0.00 0.00 0.00 0.33
Citric Acid 0.00 0.00 0.00 0.00 0.29 0.00 0.00 0.00 0.07 0.07 0.33
Sucrose 0.00 0.07 0.07 0.07 0.07 0.00 0.19 0.00 0.07 0.07 0.33
Aspartame 0.00 0.00 0.00 0.07 0.11 0.07 0.19 0.00 0.11 0.00 0.33
NaCl-sucrose 0.24 0.00 0.00 0.00 0.00 0.00 0.00 0.19 0.00 0.00 0.33
Acid-sucrose 0.00 0.00 0.00 0.00 0.27 0.00 0.00 0.00 0.15 0.00 0.33
Quinine-sucrose 0.00 0.00 0.11 0.19 0.00 0.00 0.07 0.07 0.00 0.11 0.33
py•log2(1/py) 0.42 0.40 0.22 0.41 0.45 0.07 0.33 0.31 0.29 0.19

C. Calculation of T
T
Hx  Hy  Hxy
1.71, where
Hx
[px•log2(1/px)]
3.32; Hy
[py•log2(1/py)]
3.08; and Hxy
[pxy•log2(1/pxy)]
4.69
Source: Data from Hettinger et al., 1999.

tively assess the functioning of the gustatory system and excellent spatial but poor temporal resolution. Of these,
have provided the basis for much of our current under- fMRI is preferred because of resolution, cost, availability,
standing of human gustatory function. Combined with and noninvasiveness. Use of evoked potentials (electroen-
stimulus specification from psychophysical measures, cephalography: EEG) and magnetoencephalography
biological correlates of taste function such as evoked (MEG) can provide excellent temporal but poor spatial res-
potentials and brain images reveal patterns of central olution of taste-related events, but they are in the early
nervous system activation resulting from gustatory stimuli stages of development. The use of multichannel systems,
in normal subjects and, more recently, in patients (see particularly for MEG, improves spatial resolution, but
Chapters 11 and 12 for olfactory applications). EEG is preferred due to cost and availability.
Functional magnetic resonance imaging (fMRI) and Taste has been at a disadvantage compared to hearing,
positron emission tomography (PET) reveal primary and sight, or touch with regard to the detection of sensory-
secondary cortices involved in gustatory processing with evoked potentials or functional images such as PET, MEG,
Contemporary Measurement of Human Gustatory Function 797

and fMRI. This is in part due to the relatively deep position distinct trigeminal or general visceral systems (Gilmore
and mixed modality of the primary gustatory cortex, which and Green, 1993). Water itself may activate CNS regions
makes signals difficult to detect and localize. The location that respond to taste stimuli (Zald and Prado, 2000), and
of the human primary gustatory cortex, the largest poten- important activation may be obscured if the response to
tial source of pure gustatory, cerebral cortical-evoked water is subtracted from the response (Frey and Petrides,
responses, is the superior part of the insula in humans 1999). The use of artificial saliva may reduce such “water”
(Small et al., 1999). The insula is a deeper structure than effects (O’Doherty et al., 2001).
primary visual, auditory, or somatosensory cerebral cor- Remarkable recent progress has been made in imaging
tices, and, compared to other primary sensory cortices, human gustatory areas. Cortical activation in response to
more nongustatory responses occur and, conversely, chemical and electric-taste stimuli has been localized to
a smaller percentage of neurons are taste responsive (Plata- the insula and overlying operculi with PET, MEG, and
Salaman et al., 1992). Furthermore, it is difficult to present fMRI (Barry et al., 2001; Faurion et al., 1999; Frey and
precisely timed taste stimuli at high frequency, and taste Petrides, 1999; Kobayakawa et al., 1999; O’Doherty et al.,
stimuli are usually accompanied by somatosensory cues, 2001; Small et al., 1999). Multiple insular areas have
which complicate determination of the taste component. been found to be activated. Co-activation of the inferior
Nonetheless, advances in stimulus presentation techniques parietal cortex may reflect somatosensory stimulus com-
and in the resolution of signal versus noise now make taste ponents, although this is not known for sure. Interestingly,
imaging studies feasible and useful. co-activation occurs in the amygdala and orbitofrontal
Taste stimuli can be delivered with semi-precise timing cortex, two secondary gustatory cortical areas that in non-
to restricted regions of the tongue either in aqueous solu- human primates respond to motivational state (Rolls,
tion (Kobayakawa et al., 1999; Plattig, 1991) or as gases 2000; Yan and Scott, 1996).
(e.g., acetic acid) (Genow et al., 1998; Kobal, 1985). Although there is little evidence for spatial separation
Gaseous stimuli can be applied to a broader area of the of cortical sweet, salty, sour, and bitter representations,
tongue compared to regionally applied aqueous solutions. aversive and pleasant tastes may be separately represented
Although lower concentrations are effective with a larger within the orbitofrontal cortex (O’Doherty et al., 2001),
stimululated area (Smith, 1971), few taste stimuli are and aversive tastes may differentially activate the
gaseous. Electric taste stimuli can be delivered with pre- pregenual cingulate gyrus (Zald et al., 1998). These corti-
cise timing (see above), and regardless of mechanism, cal representations may have relevance to brain imaging of
electric taste sensitivity accurately parallels spatial chemi- the dysgeusic patient who perceives taste phantoms as
cal taste sensitivity in patients with taste loss (Frank and unpleasant. Hedonic ratings (Frank and Archambo, 1986)
Smith, 1991). Thus, in these patients electric taste is of taste stimuli and phantoms might also be undertaken in
a promising tool for measurement of evoked cortical these cases to help interpret results. Based on taste-quality
responses. Sensory adaptation (see above) limits the pre- recognition deficits in subjects with temporal lobectomy
sentation frequency of electric taste and many chemical and preferential activation of the right anterior temporal
taste stimuli to about 4 Hz, a frequency sufficient for mea- lobe (including the amygdala) by taste stimuli in normal
suring cortical responses. However, bitter or irritant chem- subjects, the right anterior temporal lobe has been impli-
icals such as capsaicin are more problemmatic because cated in taste-quality recognition (Small et al., 1997).
their effects may linger for minutes (Breslin and Tharp, Furthermore, taste stimuli restricted to either the left or
2001; Green, 1991; McBurney et al., 1972, 1997). right side of the tongue preferentially activate the right
Delivery of taste stimuli without nontaste (somatosen- insular cortex (Barry et al., 2001), and a small ventral insu-
sory) cues requires careful attention. Chemical taste lar region is primarily activated in the dominant hemi-
stimuli can be delivered in a constant flow of carrier (water sphere (Faurion et al., 1999). These findings suggest that
or artificial saliva) or in a gaseous form with minimal patterns of hemispheric dominance influence gustatory
somatosensory cues. Electrial stimuli can be presented processing. Central representations of taste phantoms
with few accompanying somatosensory cues if the (dysgeusias that occur without taste stimulation) have not
electrode is held in position during control and stimulus been identified. However, CNS activation of many of the
periods (Barry et al., 2001). Nonetheless, it is still possible cortical areas noted above occurred when patients recalled
for a subject to associate the touch of stimulator probes a taste phantom, and effective treatment of the phantom
with taste stimuli, and the resulting memory may affect resulted in less evoked activity (Henkin et al., 2000).
CNS responses. The high levels of chemical and electrical In summary, new imaging technologies show promise
stimuli required when using small stimulus probes with for revealing human gustatory system function. With
patients may stimulate common chemical receptors of attention to the limitations of techniques and stimulation
798 Frank et al.

paradigms, these imaging technologies will become PROP tasting status has been determined in a number of
increasingly useful in the clinical setting. ways, including measuring detection thresholds, which
have been very important in establishing genetic models
C. Genetic Taster Status (Guo and Reed, 2001; Reed et al., 1995), and perceptual
scaling. Magnitude estimates of a series of PROP solutions
As noted above, identification of a patient’s taster status from 0.056 to 1.8 mM [concentrations surrounding the
may help account for normal variability in taste measure- antimode of the bimodal PROP threshold distribution
ments and thus increase the positive predictive value (Reed et al., 1995)], normalized to tones, have also been
of a clinical test. The idea is to have controls specific used to phenotype individuals (Bartoshuk, 1989a). A
for PTC/PROP taster status subpopulations. Also, associa- simpler test uses filter paper that has been impregnated
tions of PROP tasting phenotypes with disease states, with saturated PROP and dried (Luchina et al., 1998;
food choice, and taste perception (Bartoshuk, 2000; Reed et al., 1999). Thresholds and ratings may be differ-
Bartoshuk et al., 1996; Drewnowski et al., 2001; Guo entially associated with two genetic loci (Guo and
and Reed, 2001) (see also Chapter 40), make it appropriate Reed, 2001).
to ascertain genetic taster status in patients with taste
complaints.
The ability to taste thioureas such as phenylthiocar- V. SUMMARY AND CONCLUSIONS
bamide and related compounds such as n-propylthiouracil
is genetically determined; “tasters” can detect and recog- Current taste testing in specialty clinics comprehensively
nize as bitter PROP solutions of 0.1 mM or less, whereas addresses alterations in perceptual intensity. Evaluation of
“nontasters” cannot. One interpretation of the phenotype is the extensive patient and control data on taste weakness
that PROP tasting is a dominant trait due to a single and taste-quality misidentification should suggest stream-
autosomal gene located on 5p15 (Reed et al., 1999) that lined testing that is both reliable and valid. Additional
may be associated with a member of a family of taste attention to objective measures of taste discrimination and
receptors (Adler et al., 2000; Matsunami et al., 2000). This CNS processing will promote appreciation of conse-
has important implications for mechanisms of bitter quences and causes of all taste disorders, including dys-
perception. geusia.
PROP is preferred for phenotyping because PTC has a
sulfurous odor that may be detected retronasally. The pro-
portion of nontasters in Caucasian Europeans, about 30%, ACKNOWLEDGMENTS
is higher than for most other racial groups; some
American Indian populations may be exclusively tasters This work was supported by the National Institute of
of PTC (Kalmus, 1971). A consequence of the existence Deafness and other Communicative Disorders at NIH (P50
of PTC-PROP tasters and nontasters in the preponderance DC00168) and the University of Connecticut Health
of population studies (Guo and Reed, 2001) is the occur- Center (UCHC). We gratefully acknowledge the help of
rence of bimodal distributions of thresholds and intensity Dr. Jonathan M. Clive (UCHC Office of Biostatistical
ratings in most populations. Compared to nontasters, Consultation) with probability calculations and statistical
PROP tasters perceive other substances, especially other comparisons.
bitter stimuli, to be more intense compared to nontasters
(Bartoshuk, 2000; Delwiche et al., 2001). Although even
sweet stimuli (Gent and Bartoshuk, 1983), qualitatively REFERENCES
similar to sucrose in tasters and nontasters alike (Rankin
Adler, E., Hoon, M. A., Mueller, K. L., Chandrashekar, J., Ryba,
and Marks, 1992), were rated less intense by nontasters
N. J., and Zuker, C. S. (2000). A novel family of mammalian
(Bartoshuk, 1979), taster-nontaster intensity differences taste receptors. Cell 100:693–702.
for stimuli not structurally related to phenylthiourea Ahne, G., Erras, A., Hummel, T., and Kobal, G. (2000).
are quantitatively much smaller than PTC or PROP Assessment of gustatory function by means of tasting tablets.
differences. It is possible that specific phenotyping for Laryngoscope 110:1396–1401.
PROP/PTC tasting is obscured by a general taste Arey, L. B., Tremaine, M. J., and Monzingo, F. L. (1935). The
sensibility (Frank and Korchmar, 1985; Guo and Reed, numerical and topographical relations of taste buds in human
2001), which could be corrected by appropriate normal- circumvallate papillae throughout the life span. Anat. Rec.
ization (Gent and Bartoshuk, 1983). 64:9–25.
Contemporary Measurement of Human Gustatory Function 799

Arnold, S. M. (1974). The vulnerability of the chorda Bujas, Z., Szabo, S., Ajduković, D., and Mayer, D. (1991a).
tympani nerve to middle ear disease. J. Laryngol. Otol. Magnitude estimation of adaptation to salt using a flow cham-
88:457–466. ber for stimulus delivery. Physiol. Behav. 49:735–737.
Attneave, F. (1959). Applications of Information Theory to Bujas, Z., Szabo, S., Ajduković, D., and Mayer, D. (1991b). Time
Psychology, Holt, Rinehart, & Winston, New York. course of recovery from gustatory adaptation to NaCl.
Barry, M. A., Gatenby, J. C., Zeiger, J. D. and Gore, J. C. (2001). Percept. Psychophys. 49:517–521.
Hemispheric dominance of cortical activity evoked by focal Bull, T. R. (1965). Taste and the chorda tympani. J. Laryngol.
electrogustatory stimuli. Chem. Senses 26:471–482. Otol. 79:479–493.
Bartoshuk, L. M. (1978). History of taste research. In Handbook Cain, W. S. (1989). Testing olfaction in a clinical setting. Ear
of Perception, Vol. VIA, E. C. Carterette, and M. P. Friedman Nose Throat J. 68:316–328.
(Eds.). Academic Press, New York, pp. 3–18. Cain, W. S., and Gent, J. F. (1991). Olfactory sensitivity:
Bartoshuk, L. M. (1979). Bitter taste of saccharin related to the Reliability, generality, and association with aging. J. Exp.
genetic ability to taste the bitter substance 6-n-propylthio- Psychol. Human Percept. Perform. 17:382–391.
uracil. Science 205:934–935. Cain, W. S., and Rabin, M. D. (1989). Comparability of two tests
Bartoshuk, L. M. (1989a). Clinical evaluation of the sense of of olfactory functioning. Chem. Senses 14:479–485.
taste. Ear Nose Throat J 68:331–337. Campbell, D. T. (1957). A typology of tests, projective and oth-
Bartoshuk, L. M. (1989b). Taste. Robust across the age span? N erwise. J. Consult. Psychol. 21:207–210.
Y Acad. Sci., 561:65–75. Catalanotto, F. A., Frank, M. E., and Contreras, R. J. (1986).
Bartoshuk, L. M. (1991). Sweetness: History, preference, and Animal models of taste alteration. In Clinical Measurement of
genetic variability. Food Technol. 45:108–113. Taste and Smell, H. L. Meiselman and R. S. Rivlin (Eds.).
Bartoshuk, L. M. (2000). Comparing sensory experiences across MacMillan Publishing, New York, pp. 429–442.
individuals: Recent psychophysical advances illuminate Catalanotto, F. A., Bartoshuk, L. M., Östrom, K. M., Gent, J. F.,
genetic variation in taste perception. Chem. Senses 25: and Fast, K. (1993). Effects of anesthesia of the facial nerve
447–460. on taste. Chem. Senses 18:461–470.
Bartoshuk, L. M., and Gent, J. F. (1985). Taste mixtures: an Chilla, R., Nicklatsch, J., and Arglebe, C. (1982). Late sequelae
analysis of synthesis. In Taste, Olfaction, and the Central of iatrogenic damage to chorda tympani nerve. Acta
Nervous System, D. W. Pfaff (Ed.). The Rockefeller Otolaryngol. (Stockh.) 94:461–465.
University Press, New York, pp. 210–232. Clark, C. C., and Lawless, H. T. (1994). Limiting response alter-
Bartoshuk, L. M., Murphy, C., and Cleveland, C. T. (1978). natives in time-intensity scaling: An examination of the halo-
Sweet taste of dilute NaCl: psychophysical evidence for a dumping effect. Chem. Senses 19:583–594.
sweet stimulus. Physiol. Behav. 21:609–613. Collings, V. B. (1974). Human taste response as a function of
Bartoshuk, L. M., Rifkin, B., Marks, L. E., and Bars, P. (1986). locus of stimulation on the tongue and soft palate. Percept.
Taste and aging. J. Gerontol. 41:51–57. Psychophys. 16:169–174.
Bartoshuk, L. M., Duffy, V. B., Reed, D., and Williams, A. (1996). Collings, V. B., Lindberg, L., and McBurney, D. H. (1976).
Supertasting, earaches and head injury: Genetics and patho- Spatial interactions of taste stimuli on the human tongue.
logy alter our taste worlds. Neurosci. Biobehav. Rev. 20:79–87. Percept. Psychophys. 19:69–71.
Bradley, R. M. (2000). Sensory receptors of the larynx. Am. J. Costanzo, R. M., and Zasler, N. D. (1991). Head truama. In Smell
Med. 108 (Suppl. 4a): 47s–50s. and Taste in Health and Disease, T. V. Getchell, R. L. Doty,
Breslin, P. A., and Beauchamp, G. K. (1995). Suppression of bit- L. M. Bartoshuk, and J. B. Snow (Eds.). Raven Press, New
terness by sodium: variation among bitter taste stimuli. Chem. York, pp. 711–730.
Senses 20:609–623. Cowart, B. J. (1989). Relationships between taste and smell
Breslin, P. A., and Tharp, C. D. (2001). Reduction of saltiness across the adult life span. N Y Acad. Sci., 561:39–55.
and bitterness after chlorhexidine rinse. Chem. Senses 26: Cowart, B. J., Young, I. M., Feldman, R. S., and Lowry, L. D.
105–116. (1997). Clinical disorders of smell and taste. Occup. Med.
Beets, M. G. J. (1971). Olfactory response and molecular struc- 12:465–483.
ture. In Handbook of Sensory Physiology, Vol. IV, Chemical Deems, D. A., Doty, R. L., Settle, R. G., Moore-Gillon, V., Shaman,
Senses Part 1: Olfaction, L. M. Beidler (Ed.). Springer- P., Mester, A. F., Kimmelman, C. P., Brightman, V. J., and Snow,
Verlag, Berlin, pp. 257–321. J. B., Jr. (1991). Smell and taste disorders, a study of 750
Bell, C. (1803). On the sense of tasting. In The Anatomy of the patients from the University of Pennsylvania Smell and Taste
Human Body. A. Strahan, London. Center. Arch. Otolaryngol. Head Neck Surg. 117:508–528.
Bujas, Z. (1971). Electrical taste. In Handbook of Sensory Deems, D. A., Yen, D. M., Kreshak, A., and Doty, R. L. (1996).
Physiology, Vol. IV, Chemical Senses Part 2: Taste, L. M. Spontaneous resolution of dysgeusia. Arch. Otolarvngol.
Beidler (Ed.). Springer-Verlag, Berlin, pp. 180–199. Head Neck Surg. 122:961–963.
Bujas, Z., Ajduković, D., Szabo, S., Mayer, D., and Vodanović, Delwiche, J. F., Lera, M. F., and Breslin, P. A. (2000). Selective
M. (1986). Some observations on the mechanism of electric removal of a target stimulus localized by taste in humans.
taste. Acta Biol. JAZU 12:1–13. Chem. Senses 25:181–187.
800 Frank et al.

Delwiche, J. F., Buletic, Z., and Breslin, P. A. (2001). Covariation Frank, R. A., and Archambo, G. (1986). Intensity and hedonic
in individuals’ sensitivities to bitter compounds: evidence judgments of taste mixtures: an information integration analy-
supporting multiple receptor/transduction mechanisms. sis. Chem. Senses 11:427–438.
Percept. Psychophys. 63:761–776. Frank, R. A., and Korchmar, D. L. (1985). Gustatory processing
Doty, R. L. (1992). Diagnostic tests and assessment. J. Head differences in PTC tasters and non-tasters: a reaction time
Trauma Rehabil. 7:47–65. analysis. Physiol. Behav. 35:239–242.
Doty, R. L., Shaman, P., and Dann, M. (1984). Development of Frank, R. A., Ducheny, K., and Mize, S. J. S. (1989). Strawberry
the University of Pennsylvania Smell Identification Test: odor, but not red color, enhances the sweetness of sucrose
A standardized microencapsulated test of olfactory function. solutions. Chem. Senses 14:371–377.
Physiol. Behav. 32:489–502. Frank, R. A., Mize, S. J. S., Kennedy, L. M., de los Santos, H. C.,
Doty, R. L., Newhouse, M. G., and Azzalina, J. D. (1985). and Green, S. J. (1992). The effect of Gymnema sylvestre
Internal consistency and short-term test-retest reliability of extracts on the sweetness of eight sweeteners. Chem. Senses
the University of Pennsylvania Smell Identification Test. 17:461–479.
Chem. Senses 10:297–300. Frank, R. A., van der Klaauw, N. J., and Schifferstein, H. N. (1993).
Doty, R. L., Bagla, R., Morgenson, M., and Mirza, N. (2001). Both perceptual and conceptual factors influence taste-odor and
NaCl thresholds: Relationship to anterior tongue locus, area taste-taste interactions. Percept. Psychophys. 54:343–354.
of stimulation, and number of fungiform papillae. Physiol. Frey, S., and Petrides, M. (1999). Re-examination of the human
Behav. 72:373–378. taste region: a positron emission tomography study. Eur.
Dravnieks, A. (1985). Atlas of Odor Character Profiles. ASTM J. Neurosci. 11:2985–2988.
Data Series, DS 61 (PCN 05–061000–36) Philadelphia. Fujikane, M., Itoh, M., Nakazawa, M., Yamaguchi, Y. Hirata, K.,
Drewnowski, A., Henderson, S. A., and Barratt-Fornell, and Tsudo, N. (1999). Cerebral infarction accompanied by
A. (2001). Genetic taste markers and food preferences. Drug dysgeusia: A clinical study on the gustatory pathway in the
Metab. Dispos. 29:535–538. CNS. Rinsho Shinkeigaku 39:771–774 Japanese.
Dunér-Engström, M., Fredholm, B. B., Larsson, O., Lundberg, Ganzevles, P. G. J., and Kroeze, J. H. A. (1987a). Cross adapta-
J. M., and Saria, A. (1986). Autonomic mechanisms underly- tion in taste measured with a filter-paper method. Chem.
ing capsaicin induced oral sensations and salivation in man. Senses 12:341–353.
J. Physiol. (London) 373:87–96. Ganzevles, P. G. J., and Kroeze, J. H. A. (1987b). The sour taste
Faurion, A., Cerf, B., Van De Moortele, P. F., Lobel, E., of acids. The hydrogen ion and the undissociated acid as sour
MacLeod, P., and Le Bihan, D. (1999). Human taste cortical agents. Chem. Senses 12:563–576.
areas studied with functional magnetic resonance imaging: Genow, A., Hummel, T., Kroger, H., Bagla, R. and Bigelow, D. C.
Evidence of functional lateralization related to handedness. (1998). Gustatory event-related potentials in healthy controls
Neurosci. Lett. 277:189–192. and patients with hypoguesia or aguesia (abstr), Chem. Senses
Fechner, G. T. (1860). Elemente der Psychophysik. Breitkopf and 23:606–607.
Hätel, Leipzig. Gent, J. F., and Bartoshuk, L. M. (1983). Sweetness of sucrose,
Feigin, M. B., Sclafani, A., and Sunday, S. R. (1987). Species dif- neohesperidine dihydrochalcone, and saccharin is related to
ferences in polysaccharide and sugar taste preferences. genetic ability to taste the bitter substance 6-n-propylthio-
Neurosci. Biobehav. Rev. 11:231–240. uracil. Chem. Senses 7:265–272.
Fons, M. (1970). Electrogustometry: Spatial threshold variations. Gent, J. F., and McBurney, D. H. (1978). Time course of gusta-
Acta Otolaryngol. (Stockh.) 69:445–449. tory adaptation. Percept. Psychophys. 23:171–175.
Formaker, B. K., and Frank, M. E. (2000). Taste function in Gent, J. F., Goodspeed, R. B., Zagraniski, R. T., and Catalanotto,
patients with oral burning. Chem. Senses 25:575–581. F. A. (1987). Taste and smell problems: Validation of ques-
Frank, M. E. (1994). Effects of aging on structure and function of tions for the clincal history. Yale J. Biol. Med. 60:27–35.
taste buds. In The Effect of Aging in Oral Mucosa and Skin, Gent, J. F., Frank, M. E., and Mott, A. E. (1997). Taste testing in
C. A. Squier and M. W. Hill (Eds.). CRC Press, Boca Raton, clinical practice. In Taste and Smell Disorders, A. M. Seiden
pp. 143–149. (Ed.). Thieme Medical, New York, pp. 146–158.
Frank, M. E., and Rabin, M. D. (1989). Chemosensory neu- Gent, J. F., Hettinger, T. P., Frank, M. E., and Marks, L. E. (1999)
roanatomy and physiology. Ear Nose Throat J. 68:292–296. Taste confusions following gymnemic acid rinse. Chem.
Frank, M. E., and Smith, D. V. (1991). Electrogustometry: a sim- Senses 24:393–403.
ple way to test taste. In Smell and Taste in Health and Gent, J. F., Frank, M. E., and Hettinger, T. P. (2002). Taste con-
Disease, T. V. Getchell, R. L. Doty, L. M. Bartoshuk, and J. B. fusions following chlorhexidine treatment. Chem. Senses
Snow (Eds.). Raven Press, New York, pp. 503–514. 27:73–80.
Frank, M. E., Hettinger, T. P., and Mott, A. E. (1992). The sense Gilmore, M. M., and Green, B. G. (1993). Sensory irritation and
of taste: Neurobiology, aging and medication effects. Crit. taste produced by NaCl and citric acid: effects of capsaicin
Rev. Oral Biol. Med. 3:371–393. desensitization. Chem. Senses 18:257–272.
Frank, M. E., Gent, J. F., and Hettinger, T. P. (2001). Effects of Gilmore, M. M., and Murphy, C. (1989). Aging is associated with
chlorhexidine on human taste perception. Physiol. Behav. increased Weber ratios for caffeine, but not for sucrose.
74:85–99. Percept. Psychophys. 46:555–559.
Contemporary Measurement of Human Gustatory Function 801

Grant, R., Miller, S., Simpson, D., Lamey, P. J., and Bone, I. Hettinger, T. P., Myers, W. E., and Frank, M. E. (1990). Role of
(1989). The effect of chorda tympani section on ipsilateral and olfaction in perception of non-traditional “taste” stimuli.
contralateral salivary secretion and taste in man. J. Neurol. Chem. Senses 15:755–760.
Neuorosurg. Psychiatry 52:1058–1962. Hettinger, T. P., Frank, M. E., and Myers, W. E. (1996). Are the
Green, B. G. (1991). Temporal characteristics of capsaicin sensi- tastes of Polycose and monosodium glutamate unique? Chem.
tization and desensitization on the tongue. Physiol. Behav. Senses 21:341–347.
49:501–505. Hettinger, T. P., Gent, J. F., Marks, L. E., and Frank, M. E. (1999).
Green, B. G., and Lawless, H. T. (1991). The psychophysics of A confusion matrix for the study of taste perception. Percept.
somatosensory chemoreception in the nose and mouth. In Psychophys. 61:1510–1521.
Smell and Taste in Health and Disease, T. V. Getchell, R. L. Heymann, H., and Lawless, H. T. (1997). Sensory Evaluation of
Doty, L. M. Bartoshuk, and J. B. Snow (Eds.). Raven Press, Food: Principles and Practices. Aspen Publications,
New York, pp. 235–253. Gaithersburg MD,
Green, B. G., Dalton, P., Cowart, B., Shaffer, G., Rankin, K., and Horn, W. (1825). Über den Geschmackssinn des Menschen. Karl
Higgins, J. (1996). Evaluating the “Labeled Magnitude Scale” Groos, Heidelberg.
for measuring sansations of taste and smell. Chem. Senses Hou-Jensen, H. M. (1933). Die Papillae foliatae des Menschen.
21:323–334. Z. Anat. Entwicklungsges. 102:348–388.
Groves, J., and Gibson, W. P. R. (1974). Significance of taste and Hummel, T., Erras, A., and Kobal, G. (1997). A test for screening
electrogustometry in assessing the prognosis of Bell’s (idio- of taste function. Rhinology 35:146–148.
pathic) facial palsy. J. Laryngol. Otol. 88:855–861. Hummel, T., Nesztler, C., Kallert, S., Kobal, G., Bende, M., and
Guo, S. W., and Reed, D. R. (2001). The genetics of phenylthio- Nordin, S. (2001). Gustatory sensitivity in patients with anos-
carbamide perception. Ann. Hum. Biol. 28:111–142. mia (abstr). Chem. Senses: 1118.
Hahn, H. (1934). Die Adaptation des Geschmackssinnes. Imfeld, T. N., and Schroeder, H. E. (1992). Palatal taste buds in
Z. Sinnephysiol. 65:105–145. man: Topographical arrangement in islands of keratinized
Haller, A. von (1763). Gustus. In Elementa Physiologiae, epithelium. Anat. Embryol. 185:259–269.
F. Grasset, Lausanne, pp 99–124. Jacks, S. C., Zuniga, J. R., Turvey, T. A., and Schalit, C. (1998).
Halpern, B. P. (1997). Psychophysics of taste. In Tasting and A retrospective analysis of lingual nerve sensory changes
Smelling, Handbook of Perception and Cognition, 2nd ed., after mandibular bilateral sagittal split osteotomy. J. Oral
G. K. Beauchamp and L. M. Bartoshuk (Eds.). Academic Maxillofac. Surg. 56:700–704.
Press, San Diego, CA, pp. 77–123. Kalmus, H. (1971). Genetics of taste. In Handbook of Sensory
Harris, H., and Kalmus, H. (1949). The measurement of taste sen- Physiology, Vol. IV, Chemical Senses Part 2: Taste, L. M.
sitivity to phenylthiourea (P.T.C.). Ann. Eugen. 15:24–31. Beidler (Ed.). Springer-Verlag, Berlin, pp. 165–179.
Hebhardt, P., Bagla, R., and Doty, R. L. (1999). An automated Kaneda, H., Maeshima, K., Goto, N., Kobayakawa, T., Ayabe-
regional taste-testing system. Behav. Res. Meth. Instrum. Kanamura, S. and Saito, S. (2000). Decline in taste and
Comput. 31: 464–469. odor discrimination abilities with age, and relationship
Helms, J. A., Della-Fera, M. A., Mott, A. E., and Frank, M. E. between gustation and olfaction. Chem. Senses 25:
(1995). Effects of chlorhexidine on human taste perception. 331–337.
Arch. Oral Biol. 40:913–920. Kelling, S. T., and Halpern, B. P. (1988). Taste judgements and
Henkin, R. I., and Christiansen, R. L. (1967). Taste localization gustatory stimulus duration: taste quality, taste intensity, and
on the tongue, palate and pharynx of normal man. J. Appl. reaction time. Chem. Senses 13:559–586.
Physiol. 22:316–320. Kobal, G. (1985). Gustatory evoked potentials in man.
Henkin, R. I., Gill, J. R. Jr., and Bartter, F. C. (1963). Studies on Electroenceph. Clin. Neurophysiol. 62:449–454.
taste thresholds in normal man and in patients with adrenal Kobayakawa, T., Ogawa, H., Kaneda, H., Ayabe-Kanamura, S.,
cortical insufficiency: The role of adrenal cortical steroids Endo, H. and Saito, S. (1999). Spatiotemporal analysis of cor-
and of serum sodium concentration. J. Clin. Invest. tical activity evoked by gustatory stimulation in humans.
42:727–735. Chem. Senses 24:201–209.
Henkin, R. I., Levy, L. M., and Lin, C. S. (2000). Taste and smell Köster, E. P. (1975). Human psychophysics in olfaction. In
phantoms revealed by brain functional MRI (fMRI). Methods in Olfactory Research, D. G. Moulton, A. Turk, and
J. Comput. Assist. Tomogr. 24:106–123. J. W. Johnston, Jr. (Eds.). Academic Press, New York,
Herness, M. S. (1985a). Neurophysiological and biophysical evi- pp. 345–374.
dence on the mechanism of electric taste. J. Gen. Physiol. Krarup, B. (1958). Electro-gustometry; a method for clinical taste
86:59–87. examinations. Acta Otolaryngol. (Stockh.) 49:294–305.
Herness, M. S. (1985b). The cathodal OFF response of electric Kroger, H., Doty, R. L. and Hoffman, K. (1997). Development of
taste in rats. Exp. Brain Res. 60:318–322. a regional taste sensitivity test using isointense and isoviscid
Hertz, J., Cain, W. S., Bartoshuk, L. M., and Dolan, T. F. Jr. stimuli (abstr.). Chem. Senses 22:725–726.
(1975). Olfactory and taste sensitivity in children with cystic Kuznicki, J. T., Hayward, M., and Schultz, J. (1983). Perceptual
fibrosis. Physiol. Behav. 14:89–94. processing of taste quality. Chem. Senses 7:273–292.
802 Frank et al.

Lalonde, E. R., and Eglitis, J. A. (1961). Number and distribution McBurney, D. H., Smith, D. V., and Shick, T. R. (1972).
of taste buds on the epiglottis, pharynx, larynx, soft palate and Gustatory cross-adaptation: sourness and bitterness. Percept.
uvula in a human newborn. Anat. Rec. 140:91–95. Psychophys. 11:228–232.
Lang, N. P., Catalanotto, F. A., Knöpfli, R. U., and Antczak, A. A. A. McBurney, D. H., Balaban, C. D., Christopher, D. E., and
(1988). Quality-specific taste impairment following the applica- Harvey, C. (1997). Adaptation to capsaicin within and across
tion of chlorhexidine digluconate mouthrinses. J. Clin. days. Physiol. Behav. 61:181–190.
Periodontol. 15:43–48. Meiselman H. L. (1968). Magnitude estimations of the course of
Lawless, H. T. (1987). Gustatory psychophysics. In Neurobioloqy gustatory adaptation. Percept. Psychophys. 4:193–196.
of Taste and Smell, T. E. Finger and W. L. Silver (Eds.). Wiley, Meiselman H. L. and Dzendolet, E. (1967). Variability in gusta-
New York, pp. 401–420. tory quality identification. Percept. Psychophys. 2:496–498.
Lawless, H. T., Horne, J., and Spiers, W. (2000). Contrast and Miller, I. J., Jr., and Bartoshuk, L. M. (1991). Taste perception,
range effects for category, magnitude and labeled magnitude taste bud distribution, and spatial relationships. In Smell and
scales in judgements of sweetness intensity. Chem. Senses Taste in Health and Disease, T. V. Getchell, R. L. Doty, L. M.
25:85–92. Bartoshuk, and J. B. Snow (Eds.). Raven Press, New York,
Lehman, C. D., Bartoshuk, L. M., Catalanotto, F. C., Kveton, pp. 205–233.
J. F., and Lowlicht, R. A. (1995). Effect of anesthesia of the Miller, S. L., Mirza, N. and Doty, R. L. (2002).
chorda tympani nerve on taste perception in humans. Physiol. Electrogustometric thresholds: Relationship to anterior
Behav. 57:943–951. tongue locus, area of stimulation, and number of fungiform
Lobb, B, Elliffe, D. M., and Stillman, J. A. (2000). Reliability of papillae. Physiol. Behav. 75:753–757.
electrogustometry for the estimation of taste thresholds. Clin. Mott, A. E., Grushka, M., and Sessle, B. J. (1993). Diagnosis and
Otolaryngol. 25:531–534. management of taste disorders and burning mouth syndrome.
Lucchina, L. A., Curtis, O. F., 5th Putnam, P., Drewnowski, A., Dent. Clin. North Am. 37:33–71.
Prutkin, J. M., and Bartoshuk, L. M. (1998). Psychophysical Mott A. E., Shafer D., Miller, D., Sanger. D., Banki, M., and
measurement of 6-n-propylthiouracil (PROP) taste percep- Norton, L. (1994). Gustatory function after oral maxillofacial
tion. Ann. NY Acad. Sci. 855:816–819. surgery (abstr). Chem. Senses 19:526–527.
Mackenzie, I. C. K. (1955). A simple method of testing taste. Müller, J. (1838). Vom Geschmackssinn. In Handbuch der
Lancet 2:377–378. Physiologie des Menschen. Hölscher, Koblenz.
Matsuda, T., and Doty, R. L. (1995). Regional taste sensitivity to Murphy, C., and Gilmore, M. M. (1989). Quality-specific effects
NaCl: Relationship to subject age, tongue locus and area of of aging on the human taste system. Percept. Psychophys.
stimulation. Chem. Senses 20:283–290. 45:121–128.
Matsunami, H., Montmayeur, J. P., and Buck, L. B. (2000). Murphy, C., Quiñonez, C., and Nordin, S. (1995). Reliability and
A family of candidate taste receptors in human and mouse. validity of electrogustometry and its application to young and
Nature 404:601–604. elderly persons. Chem. Senses 20:499–503.
Mattes, R. D. (1988). Reliability of psychophysical measures of Ninomiya, Y., and Funakoshi, M. (1989). Selective procaine inhi-
gustatory function. Percept. Psychophys. 43:107–114. bition of rat chorda tympani responses to electric taste stimu-
May, M., and Schlaepfer, W. M. (1975). Bell’s palsy and the lation. Comp. Biochem. Physiol. 92A:185–188.
chorda tympani nerve: A clinical and electron microscopic O’Doherty, J., Rolls, E. T., Francis, S., Bowtell, R., and
study. Laryngoscope 85:1957–1975. McGlone, F. (2001). Representation of pleasant and aversive
McBurney, D. H. (1976). Temporal properties of the human taste taste in the human brain. J. Neurophysiol. 85:1315–1321.
system. Sens. Processes. 1:150–162. Ossebaard, C. A., Polet, I. A. and Smith, D. V. (1997). Amiloride
McBurney, D. H. (1984). Taste and olfaction: sensory discrimi- effects on taste quality: comparison of single and multiple
nation. In Handbook of Physiology. Sect. 1: The Nervous response category procedures. Chem. Senses 22:267–275.
System Processes, Vol. III. Sensory Processes, I. Darian- Ovesen, L., Sorensen, M., Hannibal, J., and Allingstrup, L.
Smith (Ed.). Am. Physiol. Soc., Washington, DC, pp. (1991). Electrical taste detection thresholds and chemical
1067–1086. smell detection thresholds in patients with cancer. Cancer
McBurney, D. H., and Bartoshuk, L. M. (1973). Interactions 68:2260–2265.
between stimuli with different taste qualities. Physiol. Behav. Pangborn, R. M. (1984). Sensory analysis as an analytical labo-
10:1101–1106. ratory tool in food research. In Modern Methods of Food
McBurney, D. H., and Gent, J. F. (1979). On the nature of taste Analysis, K. K. Stewart and J. R. Whitaker (Eds.), AVI,
qualities. Psychol. Bull. 86:151–167. Westport, CT, pp. 265–292.
McBurney, D. H., and Moskat, L. J. (1975). Taste thresholds in Pangborn, R. M., and Giovanni, M. E. (1984). Dietary intake of
college-age smokers and non-smokers. Percept. Psychophys. sweet foods and of dairy fats and resulant gustatory responses
18:71–73. to sugar in lemonade and to fat in milk. Appetite 5:317–327.
McBurney, D. H., Kasschau, R. A. and Bogart, L. M. (1967). The Plata-Salaman, C. R., Scott, T. R., and Smith-Swintosky, V. L.
effect of adaptation on taste jnd’s. Percept. Psychophys. (1992). Gustatory neural coding in the monkey cortex:
2:175–178. L-amino acids. J. Neurophysiol. 67:1552–1561.
Contemporary Measurement of Human Gustatory Function 803

Plattig, K.-H. (1991). Gustatory evoked brain potentials in Schiffman, S. S., Sennewald, K., and Gagnon, J. (1981b).
humans. In Smell and Taste in Health and Disease, T. V. Comparison of taste qualities and thresholds of D- and
Getchell, R. L. Doty, L. M. Bartoshuk, and J. B. Snow (Eds.). L-amino acids. Physiol. Behav. 27:51–59.
Raven Press, New York, pp. 277–286. Schiffman, S. S., Crumbliss, A. L., Warwick, Z. S., and Graham,
Portmann, M. O., McConville, P., Alexander, S., Breslin, P., and B. G. (1990). Thresholds for sodium salts in young and
Beauchamp, G. (2001). Amelioration of bitterness and oral elderly human subjects: correlation with molar conductivity
sensations of chlrohexidine digluconate by sodium glu- of anion. Chem. Senses 15:671–678.
conate and sodium propionate salts. (abstr) Chem. Senses Schiffman, S. S., Zervakis, J., Heffron, S., and Heald, A. E.
26: 1036–1037. (1999). Effect of protease inhibitors on the sense of taste.
Prescott, J., Ripandelli, N., and Wakeling, I. (2001). Binary taste Nutrition 15:767–772.
mixture interactions in PROP non-tasters, medium-tasters and Schiffman, S. S., Zervakis, J., Suggs, M. S., Budd, K. C., and
super-tasters. Chem. Senses 26:993–1003. Iuga, L. (2000). Effect of tricyclic antidepressants on taste
Rankin, K. M., and Marks, L. M. (1992). Effects of context on responses in humans and gerbils. Pharmacol. Biochem.
sweet and bitter tastes: Unrelated to sensitivity to PROP (6-n- Behav. 65:599–609.
propylthiouracil). Percept. Psychophys. 52:479–486. Sclafani, A. (1987). Carbohydrate taste, appetite, and obesity: an
Reed, D. R., Bartoshuk, L. E., Duffy, V., Marino, S., and Price, R. overview. Neurosci. Biobehav. Rev. 11:131–153.
A. (1995). Propylthiouracil tasting: determination of underly- Scott, T. R., and Mark, G. P. (1987). The taste system encodes
ing threshold distributions using maximum likelihood. Chem. stimulus toxicity. Brain Res. 414:197–203.
Senses 20:529–533. Seiden, A. M., Duncan, H. J., and Smith, D. V. (1992). Office
Reed, D. R., Nanthakumar, E. North, M., Bell, C., Bartoshuk, management of taste and smell disorders. Otolaryngol. Clin.
L. M., and Price, R. A. (1999). Location of a gene for bitter- North Am. 25:817–835.
taste perception to human chromosome 5p15. Am. J. Hum. Settle, R. G., Meehan, K., Williams, G. R., Doty, R. L. and Sisley,
Genet. 64:1478–1480. A. C. (1986). Chemosensory properties of sour tastants.
Rehnberg, B. G., Hettinger, T. P., and Frank, M. E. (1992). Physiol. Behav. 36:619–623.
Salivary ions and neural taste responses in the hamster. Chem. Shafer, D. M., Frank, M. E., Gent, J. F., and Fischer, M. E. (1999)
Senses 17:179–190. Gustatory function after third molar extraction. Oral Surg.
Rodin, J., Bartoshuk, L., Peterson, C., and Schank, D. (1990). Oral Med. Oral Path. Oral Radiol. Endod. 87:419–428.
Bulimia and taste: possible interactions. J. Abnorm. Psychol. Small, D. M., Jones-Gotman, M., Zatorre, R. J., Petrides, M., and
99:32–39. Evans, A. C. (1997). A role for the right anterior temporal lobe
Rolls, E. T. (2000). The orbitofrontal cortex and reward. Cereb. in taste quality recognition. J. Neurosci. 17:5136–5142.
Cortex 10:284–294. Small, D. M., Zald, D. H., Jones-Gotman, M., Zatorre, R. J.,
Salata, J. A., Raj, J. M., and Doty, R. L. (1991). Differential sen- Pardo, J. V., Frey, S., and Petrides, M. (1999). Human cortical
sitivity of tongue areas and palate to electrical stimulation: gustatory areas: a review of functional neuroimaging data.
a suprathreshold cross-modal matching study. Chem. Senses NeuroReport 10:7–14.
16:483–489. Smith, D. V. (1971). Taste intensity as a function of area and con-
Sampson, C. T., Frank, M. E., and Mott, A. E. (1993). Clinical centration: differentiation between compounds. J. Exp.
use of electrogustometry (abstr). Chem. Senses 18:622. Psychol. 87:163–171.
Sandick, B., and Cardello, A. V. (1981). Taste profiles from sin- Smith, D. V., and McBurney, D. H. (1969). Gustatory cross-adap-
gle circumvallate papillae: Comparison with fungiform pro- tation: does a single mechanism code the salty taste? J. Exp.
files. Chem. Senses 6:197–214. Psychol. 80:101–105.
Schiffman, S. (1979). Changes in taste and smell with age: psy- Smith, D. V., and van der Klaauw, N. J. (1995). The perception of
chophysical aspects. In Sensory Systems and Communication in saltiness is eliminated by NaCl adaptation: implications for
the Elderly, J. M. Ordy and K. Brizzee (Eds.). Raven Press, New gustatory transduction and coding. Chem. Senses 20:545–557.
York, pp. 227–246. Stevens, J. C. (1996). Detection of taste in mixture with other
Schiffman, S. S., and Clark, T. B. III. (1980). Magnitude esti- tastes: issues of masking and aging. Chem. Senses 21: 211–221.
mates of amino acids for young and elderly subjects. Stevens, J. C., and Cain, W. C (1993). Changes in taste and fla-
Neurobiol. Aging 1:81–91. vor in aging. Crit. Rev. Food Sci. Nutr. 33:27–37.
Schiffman, S. S., and Dackis, C. (1975). Taste of nutrients: Stevens, J. C., Cruz, L. A., Hoffman, J. M., and Patterson, M. Q.
Amino acids, vitamins, and fatty acids. Percept. Psychophys. (1995). Taste sensitivity and aging: high incidence of decline
17:140–146. revealed by repeated threshold measures. Chem. Senses
Schiffman, S. S., Hornack, K., and Reilly, D. (1979). Increased 20:451–459.
taste thresholds of amino acids with age. Am. J. Clin. Nutr. Stevens, S. S. (1961). The psychophysics of sensory function. In
32:1622–1627. Sensory Communication, W. A. Rosenblith, (Ed.). MIT Press,
Schiffman, S. S., Lindley, M. G., Clark, T. B., and Makino, C. Cambridge, MA pp. 1–33.
(1981a). Molecular mechanism of sweet taste: relationship of Stillman, J. A., Morton, R. P., and Goldsmith, D. (2000). Automated
hydrogen bonding to taste sensitivity for both young and electrogustometry: a new paradigm for the estimation of taste
elderly. Neurobiol. Aging 2:173–175. detection thresholds. Clin. Otolaryngol. 25:120–125.
804 Frank et al.

Taillibert, S., Bazin, B., and Pierrot-Deseilligny, C. (1998). Weiffenbach, J. M., Cowart, B. J., and Baum, B. J. (1986). Taste
Dysgeusia resulting from internal carotid dissection. A lim- intensity perception in aging. J. Gerontol. 41:460–468.
ited glossopharyngeal nerve palsy. J. Neurol. Neurosurg. Wiberg, A. (1971). Function of the chorda tympani before and
Psychiatry 64:691–692. after operation for clinical otosclerosis. Dissertation,
Tennissen, A. M., and McCutcheon, N. B. (1996). Anterior University of Umea, Sweden.
tongue stimulation with amiloride suppresses NaCl Saltiness, Woskow, M. H. (1967). Some new methods of flavor evaluation.
but not citric acid sourness in humans. Chem. Senses MBAA Tech. Q. 4:68–72.
21:113–120. Wright, H. N. (1987). Characterization of olfactory dysfunction.
Todrank, J., and Bartoshuk, L. M. (1991). A taste illusion: Taste Arch. Otolaryngol. Head Neck Surg. 113:163–168.
sensation localized by touch. Physiol. Behav. 50:1027–1031. Wright, H. N., Sheehe, P. R., and Leopold, D. A. (1991). The
Tomita, H., and Horikawa, Y. (1986). Dissociated taste disorder. odorant confusion matrix as an aid to diagnosis (abstr). Chem.
Auris-Nasus-Larynx (Tokyo) 13 (Suppl. I): S17–S23. Senses 16: 601.
Tomita, H., Ikeda, M., and Okuda, Y. (1986). Basis and practice Yamaguchi, S. (1991). Basic properties of umami and effects on
of clinical taste examinations. Auris-Nasus-Larynx (Tokyo) 13 humans. Physiol. Behav. 49:833–841.
(Suppl. I):S1–S15. Yan, J., and Scott, T. R. (1996). The effect of satiety on responses
Tunsuriyawong, P., Bagla, R., & Doty, R. L. (2000). Effects of of gustatory neurons in the amygdala of alert cynomolgus
a pressure surround on the regional taste detection threshold macaques. Brain Res. 740:193–200.
for sodium chloride. Acta Otolaryngol. 120:81–86. Zald, D. H., and Pardo, J. V. (2000). Cortical activation induced
van der Klaauw, N. J., and Smith, D. V. (1995). Taste quality profiles by intraoral stimulation with water in humans, Chem. Senses
of fifteen organic and inorganic salts. Physiol. Behav. 58:295–306. 25:267–275.
Vanne, M., Laurinen, P., and Tuorila, H. (1998). Ad libitum mix- Zald, D. H., Lee, J. T., Fluegel, K. W., and Pardo, J. V. (1998).
ing in a taste memory task: methodological issues. Chem. Aversive gustatory stimulation activates limbic circuits in
Senses 23:379–384. humans. Brain 121:1143–1154.
von Békésy, G. (1965). The effect of adaptation on the taste Zuniga, J. R., Davis, S. H., Englehardt, R. A., Miller, I. J. Jr.,
threshold observed with a semi-automatic gustometer. J. Gen. Schiffman, S. S., and Phillips, C. (1993). Taste performance
Physiol. 48:481–488. on the anterior human tongue varies with fungiform taste bud
Weiffenbach, J. M. (1983). Taste quality recognition and forced- density. Chem. Senses 18:449–460.
choice response. Percept. Psychophys. 33:251–254. Zuniga, J. R., Chen, N., and Miller, I. J., Jr. (1994). Effects of
Weiffenbach, J. M. (1987). Taste perception mechanisms. Front. chorda-lingual nerve injury and repair on human taste. Chem.
Oral Physiol. 6:151–167. Senses 19:657–665.
Weiffenbach, J. M., Wolf, R. O., Benheim, A. E., and Folio, C. J. Zuniga, J. R., Chen, N., and Phillips, C. L. (1997). Chemosensory
(1983). Taste threshold assessment: A note on quality specific and somatosensory regeneration after lingual nerve repair in
differences between methods. Chem. Senses 8: 151–159. humans.. J. Oral Maxillofac. Surg. 55:2–13.
38

Human Perception of Taste Mixtures

Hendrik N. J. Schifferstein
Delft University of Technology, Delft, The Netherlands

I. INTRODUCTION Despite differences in approach, taste research by both

groups has mainly focused on quality and intensity.
During the consumption of a food product, the human gus- Because of this, the current chapter is restricted to these
tatory system is stimulated by a large number of different two dimensions of taste sensations.
chemicals. Perceiving such a chemically complex stimulus A number of pioneering studies performed in psycho-
requires a reception mechanism, a transduction logical laboratories in Germany in the late nineteenth
mechanism, and an integration mechanism. After specific century focused specifically on the role of taste quality
chemicals have bound to receptor sites, the receptor cells in the perception of taste mixtures. Indeed, a classic
activate the afferent neural system. The perceptual system debate developed between Kiesow (1894, 1896) and
then integrates the signals released by the taste receptor Öhrwall (1891) with regard to the nature of the percept
cells. During the perceptual process many substances, or elicited by taste mixtures. Kiesow argued that sweet,
the signals they elicit, interact. The sensation elicited by an sour, salty, and bitter were four qualities within one
unmixed component, therefore, usually differs from the modality, whereas Öhrwall argued that they constituted
sensation elicited by that same component as part of a four separate modalities. Von Skramlik (1926) supported
complex stimulus. The present chapter focuses on the ways Kiesow’s perspective by suggesting that mixtures of sub-
in which mixture components interact and the methods by stances representing the four basic tastes could duplicate
which these taste interactions can be determined. the complex taste experience elicited by any inorganic
Research on taste interactions has evolved from two salt. He developed a set of “mixture equations” consist-
streams of research: experimental psychology and food ing of the concentration levels of quinine hydrochloride,
technology.* Experimental psychologists try to unravel the sodium chloride, tartaric acid, and glucose that produced
mechanisms by which humans perceive the world sur- a mixture with taste properties similar to a particular
rounding them. In perceptual research they have found inorganic salt concentration. As we shall see later in this
taste sensations to be characterized by the dimensions of chapter, the analytic versus synthetic debate continues
quality, intensity, duration, and location (Külpe, 1893; until the present day.
Boring, 1942). Food technologists, on the other hand, In taste intensity research, early workers initially
strive to develop foods that are optimally palatable. employed simple stimuli (e.g., solutions of sucrose in
water) to determine the lowest concentration that could be
reliably discerned, i.e., the detection threshold (e.g., Von
Skramlik, 1926) (see also Chapter 37). After having stud-
*Formore extensive historical overviews, the reader is referred to ied the taste properties of single substances, researchers
Bartoshuk et al. (1985) and De Graaf (1989). subsequently began to investigate how taste sensitivity

805
806 Schifferstein

changes when a second substance is added (e.g., Hahn and sucrose concentrations. When sweetness intensity was
Ulbrich, 1948). Unfortunately, detection threshold meas- expressed in glucose units, however, mixture intensity
urements only inform about whether a subject can detect was approximately equal to the sum of the intensities of
a stimulus concentration or not. In and of itself, a thresh- the unmixed components.
old gives no information about the dynamic range of More recently, investigators have tried to assess per-
sensory functioning. Hence, research attention shifted ceived intensity by the use of category rating or magnitude
from determining absolute sensitivity to assessing the estimation, the so-called direct scaling methods. In direct
relationship between physical concentration and supra- scaling, the subject is instructed to give a direct estimate of
threshold intensity, namely, the psychophysical function. perceived intensity by choosing a response category,
For taste mixture research, this implied that researchers putting a mark on a line scale, or giving a numerical
investigated how the perceived intensity of a substance response. Direct scaling experiments can usually be
was affected when a second substance was added. Most of performed much faster than discrimination or matching
this research was done after the Second World War, experiments. However, the investigator is unsure whether
largely in response to complaints of American soldiers the direct rating is a good representation of the perceived
about the sensory properties of the food rations they had intensity. According to an S-O-R (stimulus-organism-
received. This increased the need to understand the deter- response) view of single stimulus intensity judgment, three
minants of flavor and food acceptance and stimulated different stages can be distinguished in direct judgment of
research on taste perception in the United States (Peryam, a single stimulus (Fig. 1). First, a psychophysical function
1990). As foods can be regarded as highly complex mix- transforms the physical stimulus into a subjective experi-
tures of tastants, food technologists became particularly ence, a percept. The percept, temporarily stored in the
interested in taste and smell interactions. sensory buffer, is subsequently encoded in order to enable
The first psychophysical functions relating physical storage in memory and judgmental processing. During
concentration to perceived taste intensity for a single sub- encoding, attentive processes enable the subject to focus
stance were probably derived from measures of absolute upon the sensation attribute required by the experimental
and differential sensitivity using Fechner’s (1860) law. task. The subject’s conception of what is meant by the
However, this indirect scaling approach has rarely been requested attribute (e.g., “sweetness intensity”) affects
used in taste research (Meiselman, 1972), largely because stimulus processing at this stage. After encoding, the
it is very laborious. In mixture investigations, a different stimulus can be represented on an internal, subjective
indirect scaling procedure became popular: taste inter- continuum. The coded sensation is no longer a sensation,
actions were determined by relating the intensities of but a cognition. Any sensory scale derived from rated
mixed and unmixed stimuli to equi-intense concentrations intensities necessarily represents coded sensations. After
of a reference substance (e.g., Fabian and Blum, 1943; encoding, the third processing stage transfers the values of
Yamaguchi et al., 1970). For example, in sweetener mix- the coded sensations on the subjective continuum into
ture research each stimulus was matched in sweetness responses on the response scale provided. This process
intensity to a sucrose concentration. In such a “scaling- reflects the judgment function or response output function
by-matching” procedure, the form of the psychophysical (Frijters, 1993). For direct scaling methods, the perceived
function for the reference substance partly determines (coded) intensity of a single stimulus can only be derived
whether the mixture components interact or not. For from the obtained responses by assuming that the judg-
example, Cameron (1947) found that sugar mixture inten- ment function implies a linear transformation (interval
sity often exceeded the sum of the intensities of its scale values) or a multiplication with a constant (ratio scale
components when sweetness intensity was expressed in values). It is possible to test the validity of this assumption

Figure 1 Conceptual outline of a direct scaling procedure using single stimulus presentation. The relationship between physical stim-
ulus and perceived sensation is given by the psychophysical function. The sensation is subsequently encoded and transformed into a
response. (Adapted from Schifferstein, 1994.)
Human Perception of Taste Mixtures 807

in appropriately designed experiments (Klitzner, 1975; mixture type, dissimilar tasting substances are mixed,
De Graaf et al., 1987). leading to the formation of a heterogeneous percept in
Direct scaling in taste mixture research started with which several taste qualities can be identified. For
descriptions of taste interactions in mixtures with similar example, a sucrose/citric acid mixture elicits a sweet and
or dissimilar tastes (e.g., Kamen et al., 1961). From a sour taste. Subjects may be requested to judge the sweet-
Pangborn’s investigations on taste interactions in aqueous ness, sourness, or total taste intensity of such a mixture.
solutions and foods (Pangborn, 1960, 1961, 1962, 1965; Total taste intensity then refers to the overall strength of
Pangborn and Chrisp, 1964; Pangborn and Trabue, 1964), the percept, irrespective of taste quality. In the present
the idea appeared that dissimilar tasting substances gener- chapter, the discussion of mixture models is kept to a mini-
ally suppress each other’s specific taste intensity. mum. The interested reader is referred to reviews by
After the taste interactions were described, mathemat- Frijters (1987) and Schifferstein and Frijters (1993).
ical models were developed (e.g., Moskowitz, 1974; The next sections of this chapter focus on the percep-
Bartoshuk and Cleveland, 1977; Frijters and Oude Ophuis, tion of mixtures of similar or dissimilar tasting substances.
1983) to predict and understand taste mixture interactions. In each section, a methodological section is followed by an
The diversity of these models is large. This is not overview of interactions that occur at near-threshold and
surprising if we realize that the locus of the mechanism supra-threshold concentration levels. In addition, the sec-
responsible for the interaction between two substances A tion on mixtures of dissimilar tasting substances discusses
and B may, in principle, reside anywhere in the pathway qualitative changes, perceived complexity, and the role of
from aqueous solution to the overt behavior (response). cognitive factors in mixture perception.
The substances may react chemically, forming new
compounds, or physicochemically, forming complex struc-
tures. Furthermore, they may interact biophysically in the
II. MIXTURES OF SIMILAR TASTING
periphery of the sensory system, on their way to the recep-
SUBSTANCES
tor or in competing for receptor sites (peri-receptor events)
(e.g., Birch, 1980; Carr et al., 1989). In addition, they may A. Determination of Mixture Interactions
interact at the level of the receptor cell, in the afferent
nerve bundles, in the central nervous system or at the level In binary mixtures of similar tasting substances (often
of conscious experience, i.e., the percept (Kroeze, 1978, referred to as homogeneous mixtures), both compounds
1979; Lawless, 1979; De Graaf and Frijters, 1989). For the contribute to the intensity of the sensation perceived, and
development of mixture models, assumptions have to be interactions between substances can be described by terms
made concerning the locus of the mixture interactions and like hypoadditivity, additivity, and hyperadditivity
the mechanisms responsible for those interactions. Thus, (Berglund et al., 1976). The absence of interaction (addi-
there are mixture models that describe interaction on the tivity) is best defined as the case where the intensity of a
basis of competition for receptor sites or transducer mech- mixture equals the intensity that can be expected on the
anisms (e.g., Beidler, 1971; Frijters and Oude Ophuis, basis of the psychophysical functions for the individual
1983; Ennis, 1989, 1991). Another set of models focuses components (Berenbaum, 1989; Sühnel, 1993). Since a
on interaction in the afferent neural system (e.g., psychophysical function is usually nonlinear, doubling the
Moskowitz, 1974; Carr et al., 1989; McBride, 1989), intensity elicited by concentration x differs from the
whereas others focus upon the interactions between per- intensity elicited by concentration 2x. Therefore, the inten-
ceived sensations at the central perceptual level (e.g., sities elicited by unmixed components cannot be added,
Berglund et al., 1973; McBride, 1989; Schifferstein and but must be corrected for the nonlinearity in the form of
Frijters, 1993). the psychophysical functions of the mixture components
In this chapter I discuss taste interactions at both thresh- (Bartoshuk, 1975; De Graaf and Frijters, 1988). If a
old and supra-threshold intensity levels. With respect to scaling-by-matching procedure is used in which each
gustatory quality, I distinguish between mixtures of similar mixed and unmixed stimulus is matched in intensity to a
and dissimilar tasting substances. In the first mixture type, reference substance (see sec. I), the degree of mixture
all components elicit similar taste qualities. If such a mix- interaction is corrected for nonlinearity in the psychophys-
ture is tasted, it leads to the formation of a homogeneous ical function of the reference substance. It is preferable,
percept, consisting of only one taste sensation. For however, to derive what is meant by “additivity” from the
example, if a subject tastes a mixture of sucrose and aspar- psychophysical functions of the two mixture components
tame, only one sweet sensation is perceived (assuming that and not from the psychophysical function of an arbitrary
both substances do not elicit side tastes). In the second reference substance.
808 Schifferstein

1. Comparison Rules Starting from a Fixed below the additivity line (  1), the mixture components
Intensity Level exhibit hyperadditivity. In this case, the concentrations in
Starting with concentration x of substance A, a concentra- an equi-intense mixture are lower than expected under
tion y of a similar tasting substance B can be found equal additivity. If the iso-intensity curves lie above the additiv-
in intensity to x. If A and B do not interact, the definition ity-line (  1), A and B behave hypoadditively. In formu-
of additivity given above implies that a linear combination las, hyperadditivity is found when:
of x and y must result in a mixture equal in intensity to x R(pxqy)
Rx
Ry with p  q
1 and   1 (4)
and y. If the psychophysical response is used as a direct
measure of intensity, additivity thus implies that: In other words, hyperadditivity is demonstrated if:
Rpxqy
Rx
Ry if p  q
1 (1) Rpxqy  Rx with Rx
Ry and p  q
1 (5)
where Rpxqy is the response to the mixture, Rx is the Figure 2 shows the results De Graaf and Frijters (1986)
response to concentration x of A, Ry is the response to con- obtained when they used the iso-intensity comparison to
centration y of B, and p and q are the proportions of x and y investigate the taste interaction between glucose and fruc-
in the mixture. According to De Graaf and Frijters (1986), tose. The binary mixtures equisweet to a fixed glucose level
this iso-intensity comparison can be theoretically deduced tend to lie below the lines connecting the glucose levels to the
from Beidler’s (1971) mixture model under the assumption equisweet fructose levels. Figure 2 thus leads to the conclu-
that a linear relationship exists between the electrophysio- sion that glucose and fructose exhibit hyperadditivity: lower
logical response to a mixture and the intensity of the sen- concentrations of glucose and fructose are needed to obtain a
sation elicited by that mixture. The Beidler model assumes certain sweetness intensity than expected under additivity.
that the mixture components compete for adsorption at the
same receptor sites. Additivity could thus be regarded as 2. Comparison Rules Starting from Fixed
equivalent to competition for receptor sites. Stimulus Levels
Under additivity, the relationship between a concentra-
If fixed concentration levels are used in the construction of
tion of A and a concentration of B in a binary mixture is
series of mixtures and the intensities of these mixtures are
given by (De Graaf and Frijters, 1986):
determined, three different comparison rules can be used
A
x  xy B if Rx
Ry (2)
This equation implies that if the concentration of B is
plotted as a function of the concentration of A, under addi-
tivity all equi-intense mixtures should lie on a straight line
connecting concentration x on the x-axis with y on the
y-axis. For multicomponent mixtures, the previous equa-
tion can be extended.* For example, if a third component
C is added to the mixture, the iso-intensity comparison
predicts that:
A
x  xy B  xz C if Rx
Ry
Rz (3)
where z is the concentration of C that is equi-intense to x
and y (Nahon et al., 1996).
Binary mixture series can be constructed with constant
A/B ratios that vary in solute concentrations, the so-called
equiratio mixtures (Frijters and Oude Ophuis, 1983). For
each equiratio mixture series, a concentration level
[(pxqy)] can be determined that elicits a sensation,
equi-intense to the sensations elicited by x and y. Using
these data, an iso-intensity curve can be constructed by Figure 2 Iso-intensity comparison for glucose/fructose mix-
connecting all equi-intense mixtures. If the curve lies tures. Dotted lines connect the concentration levels of
glucose/fructose mixtures predicted to be equisweet under addi-
tivity. The drawn lines connect concentration levels that were
*The symbols used here are different from those used by De found to be equisweet. (Data from De Graaf and Frijters, 1986;
Graaf and Frijters (1986) and Nahon et al. (1996). figure adapted from Ennis, 1989).
Human Perception of Taste Mixtures 809

to determine whether taste interaction occurs. These com- glucose/fructose (GluFru) mixtures (De Graaf et al.,
parison rules assume that the perceived intensities of all 1987). In the construction of this figure, psychophysical
mixed and unmixed stimuli are known. According to functions for fructose, glucose, and the GluFru 50/50
Figure 1, however, the response to a stimulus is related to equiratio mixture type were approximated using second-
its perceived intensity by a response output function. order polynomials in which the natural logarithm of the
Therefore, in order to use these comparison rules, assump- perceived sweetness intensity was used as the dependent
tions have to be made concerning the form of the response variable, and the natural logarithm of the concentration
output function. Some comparison rules require intensities and its squared value were used as the independent
assessed on an interval level (linear response output func- variables (R2  0.999). These functions were used to
tion), whereas others assume ratio scale values (linear interpolate the sweetness intensities that were not deter-
response output function with fixed zero point). mined experimentally. For simplicity, the curves for
unmixed glucose and unmixed fructose were constructed
a. Summated Response Comparison. In the sum- under the assumption that they were based on 50/50 self-
mated response comparison, the response to a mixture mixtures.
(Rxy) is compared to the sum of the responses to the No rules exist as to whether the mixture curve should be
unmixed components (Rx  Ry). Since the summated compared to the fructose curve, the glucose curve or to
response comparison implies an addition of responses, the some compromise between these two. Therefore, an addi-
taste intensities must be assessed at the ratio level. If the tivity curve was constructed based on the predictions of the
intensity of a mixture is plotted against the sum of the iso-intensity comparison. Figure 3 shows that all curves
intensities of the unmixed components, usually a curvilin- are negatively accelerating. The GluFru 50/50 mixture
ear relationship is found. Since additivity according to the curve lies above the fructose curve and the additivity
summated response comparison implies Rxy
Rx  Ry, curve. It crosses the glucose curve at low intensity levels.
the part of the curve lying above the diagonal suggests These results indicate that glucose/fructose mixtures
hyperadditivity, whereas the part below the diagonal sug- exhibit hyperadditivity.
gests hypoadditivity.
As already noted in sec. II.A, however, each compari- b. Factorial Plot Comparison. In the factorial plot
son rule has to incorporate the fact that psychophysical comparison, mixture intensity is plotted as a function of
functions are usually nonlinear. If concentrations x of A
and y of B are equi-intense, the intensity of a mixture con-
taining x  y should be compared to the intensity of 2x or
2y and not to the sum of the intensity of x plus the
intensity of y. The latter comparison is exactly what the
summated response comparison does. Therefore, the inter-
actions found with the summated response comparison
could better be referred to as apparent taste interactions.
The only way in which the summated response com-
parison can make a useful contribution to taste interaction
research is by comparing the observed mixture interaction
to the apparent within-substance interactions of the
unmixed substances. For each unmixed component, a
curve can be constructed for a set of hypothetical mixtures
of the unmixed substance with itself. Constructing such a
curve is possible if a solution containing concentration 2x
is conceived of as a mixture of x with x, or as a mixture of
1/2 x with 11/2 x, and so on. Subsequently, the intensity of

the hypothetical mixture (S2x) can be plotted as a function

of the sum of the intensities of the unmixed components
Figure 3 Summated response comparison for glucose/fructose,
(SxSx or S1/2 xS11/2 x). In this way, a curve can be con- glucose/glucose, and fructose/fructose 50/50 mixtures. The
structed for each of the unmixed substances, and these two intensity of a mixture is plotted as a function of the sum of the
curves can be compared to the mixture curve. intensities of its unmixed components. The additivity curve
Figure 3 shows the results of the summated response shows the predictions of the iso-intensity comparison. (Data from
comparison for the perceived intensities of 50/50 De Graaf et al., 1987.)
810 Schifferstein

one mixture component’s intensity. For each concentration By determining psychophysical functions for equiratio
of the second component, a separate curve is drawn. mixture series, the position of each mixture curve relative
According to this comparison rule, apparent additivity to the curves for the unmixed substances can be estab-
implies that the effect of adding a certain concentration y lished (Fig. 5). However, there are no strict definitions of
of component B to a solution of component A is constant hyper- and hypoadditivity within the equimolar compari-
for all levels of A. The factorial plot then shows a set of son. Only if the mixture curves lie above or below both
parallel curves. In the case of apparent hypoadditivity, the component curves is taste interaction evident.
curves converge; in the case of apparent hyperadditivity
they diverge.
Similar to the summated response comparison, the ori- B. Describing and Predicting Mixture
ginal factorial plot comparison rule does not take into Interactions
account that psychophysical functions are not linear.
Therefore, mainly apparent taste interactions are revealed In the previous sections, methods were described that
by this rule. In order to reveal the “real” degree of taste determine the nature and the degree of interaction between
interaction, again the apparent mixture interactions have to two substances. If we want to use the outcomes of this type
be compared to the apparent within-substance interactions. of research to develop a general model that describes and
Since the psychophysical functions for tastants are usually predicts mixture interactions, the comparison rules that
negatively accelerating, the apparent within-substance start out from fixed intensity levels or from fixed stimulus
interactions mostly exhibit a converging pattern (De Graaf concentrations both have their limitations. If we determine
and Frijters, 1988; Schifferstein and Frijters, 1990). If the the degrees of interaction for several intensity levels, we
degree of convergence in the mixture plot is greater than still need a model relating physical concentrations to per-
the degrees of convergence in the plots for the unmixed ceived intensity levels. On the other hand, if we start out
substances, A and B behave hypoadditively. If the degree from fixed stimulus concentrations, we need to calculate
of convergence is smaller for the mixtures, the components
behave hyperadditively. No method is yet available to test
whether the degrees of convergence differ significantly.
Using the definition of additivity from the iso-intensity
comparison, a factorial plot can be constructed showing
the pattern of apparent interactions under conditions of
additivity. Figure 4 shows the expected mixture intensity
as a function of the sweetness of unmixed fructose if glu-
cose and fructose behave additively. Psychophysical
functions for the unmixed substances were estimated using
second-order polynomials (see above). Similar second-
order polynomials were estimated relating glucose
concentration to the sweetness intensity of glucose/fruc-
tose mixtures for each fixed level of fructose.
The experimentally determined sweetness intensities of
the four GluFru 50/50 mixtures in Figure 4 are higher than
the predicted intensities. Apparently the experimental data
indicate less convergence than the additive case. These
results, therefore, point at hyperadditivity between fruc-
tose and glucose.
Figure 4 Factorial plot comparison for glucose/fructose mix-
c. Equimolar Comparison. In the equimolar compar- tures. The intensity of each mixture is plotted as a function of the
intensity of each concentration of fructose with a separate curve
ison, the perceived intensities of the unmixed components
for each level of glucose. The sweetness of unmixed fructose was
are compared to the mixture intensity at one total molarity.
assessed for solutions containing 0, 0.125, 0.25, 0.5, and 1.0 M
For example, the intensity of a mixture of 0.5 M fructose fructose. The factorial plot was obtained by calculating the
and 0.5 M glucose is compared to the intensities elicited by sweetness of these fructose levels mixed with each of five levels
1.0 M fructose and by 1.0 M glucose. The equimolar com- of glucose expected under additivity. The filled circles show the
parison is the only comparison rule that has incorporated perceived intensities of GluFru 50/50 mixtures obtained experi-
the form of the psychophysical function in its comparison. mentally. (Data from De Graaf et al., 1987.)
Human Perception of Taste Mixtures 811

mixture model. This model starts out from the psycho-

physical functions of the two unmixed substances. These
functions are approximated using the formula of S.
S. Stevens’s (1975) power law:
R
kC n (6)
where R is the group average intensity response, C is the
concentration level in molarity, and k and n are constants.
Subsequently, Schifferstein (1996) determined a concen-
tration level at which the two substances yield the same
average response. This concentration level is called the
intensity unit (IU) and the response corresponding to one
IU is called Req. The revised mixture model uses equally
intense concentration levels as the unit of physical meas-
urement, instead of molarity. If concentration C is
expressed in IU, the psychophysical functions for sub-
stances A and B read:
Figure 5 Equimolar comparison for glucose, fructose, and nA nB
RAi
ReqCAi and RBj
ReqCBj (7)
three equiratio mixture types. (Adapted from De Graaf et al.,
1987.)
respectively. For a set of mixtures in which the relative
proportion of IUs for substance A is p and for substance B
is q
1  p, the revised equiratio mixture model predicts
the intensity levels expected under additivity to determine a mixture function by:
whether the mixture components interact.
pn qnB
In an attempt to construct a psychophysical mixture RABpqij
ReqCABpqij
A
(8)
model that can be universally applied to predict the
reported intensity for all binary combinations of similar In this equation, CABpqij represents the total mixture
tasting substances, Schifferstein (1996) proposed a revised concentration level in IU. Analogous to the iso-intensity
version of Frijters and Oude Ophuis’s (1983) equiratio comparison, a mixture with CABpqij
1 is predicted to

Figure 6 The sweetness of aspartame, acesulfame-K, and aspartame/acesulfame-K equiratio mixtures in 3.3 mM citric acid. (A)
Geometric mean of the observed sweetness responses as a function of the total sweetener concentration of the stimulus. (B) Predictions
of the equiratio mixture model presented in Schifferstein (1995). (C) Predictions of the equiratio mixture model as revised by
Schifferstein (1996). (Parts of this figure were adapted from Schifferstein, 1996.)
812 Schifferstein

yield the same response as the unmixed components (Req) the threshold concentration of A with 1-p times the thresh-
(Schifferstein, 1995). old concentration of B [0  p  1]. This line of reasoning
Figure 6 shows the predictions of this revised equiratio is similar to the iso-intensity comparison with the
mixture model for the sweetness of aspartame/acesulfame-K detection threshold as the fixed intensity level. For about
mixtures in 3.3 mM citric acid. Figure 6A shows the observed 200 combinations of bitter-, salty-, sweet-, and sour-tasting
magnitude estimates for the sweetness of unmixed aspartame, substances, the mixture components were found to behave
unmixed acesulfame-K, and three AspAcK equiratio mixture additively. Analogously, J. C. Stevens (1997) found that
series. The concentration ratios in molarity in these three the detection threshold concentrations in a mixture of three
series are 75/25, 50/50, and 25/75, respectively. Figure 6B sweeteners did not differ significantly from one third times
gives the approximations of the psychophysical functions for the threshold concentrations for the unmixed substances.
the two unmixed substances obtained by regression analysis In a mixture of six sweeteners, the threshold concentra-
using Eq. (7) after log-transformation of responses and tions did not differ significantly from one sixth times the
concentration levels. Furthermore, Figure 6B shows the pre- concentrations for the unmixed substances.
dictions of the equiratio mixture model using concentration
levels expressed in IU [Eq. (8)]. D. Suprathreshold Interactions
The predictions of the sweetness responses to the mix-
tures in Figure 6B are low compared to the observed Many studies investigating suprathreshold interactions
responses. These deviations are mainly due to mixture between similar tasting substances have suffered from
interactions. To account for the hyperadditivity and to confusion concerning the way in which mixture interactions
increase the predictive validity of the equiratio mixture should be assessed. Most of the interactions reported as mix-
model, Schifferstein (1996) suggested replacing the con- ture interactions were contaminated with apparent within-
stant Req by an empirical estimate: substance interactions. For the present overview, studies
were selected that used the iso-intensity comparison or that
Req
Req  I 兹苶
pq (9) separated apparent interactions from mixture interactions.
Most research on mixtures of similar tasting substances
where I is an interaction index. Furthermore, predictive has been carried out using sweet-tasting substances.
validity can be increased by correcting departures from the Hyperadditivity was found in mixtures of fructose/glucose
power law by a nonlinear response output transformation. (De Graaf and Frijters, 1986), glucose/sucrose (Frijters
These deviations are apparent when the observed et al., 1990), fructose/sucrose (De Graaf and Frijters,
psychophysical functions for aspartame and acesulfame-K 1988), and aspartame/acesulfame (Schifferstein, 1996).
(Fig. 6A) are compared to their estimated counterparts Frank et al. (1989) reported that out of 31 binary combina-
(Fig. 6B). If different output transformation functions are tions of sugars, sugar alcohols, and artificial sweeteners,
used for different scaling methods, the equiratio mixture 18 mixture types showed hyperadditivity, 2 showed
model can be applied to psychophysical data gathered hypoadditivity, and 11 behaved approximately additively.
with any scaling method. Figure 6C shows the improve- A striking example of hyperadditivity outside sweetener
ment in predictions after including the interaction index mixture research is found for the so-called taste
(I
14.85) and correcting for deviations from the power enhancers. For this type of taste interaction to occur,
law (see Schifferstein, 1996). For reference, the regression monosodium glutamate (MSG) or glutamic acid is mixed
functions of the unmixed substances (Fig. 6B) have also with a 5-ribonucleotide (Yamaguchi, 1967; Rifkin and
been transformed using the nonlinear transformation (Fig. Bartoshuk, 1980). For these substances, combining two
6C). The revised equiratio mixture model has been shown virtually tasteless solutions results in a mixture with an
to be very successful in predicting the intensity response to easily perceptible umami taste (Fig. 7).
mixtures of sweeteners and mixtures of acids Apparently, mixtures of similar tasting substances
(Schifferstein, 1996). mostly show additivity or hyperadditivity. De Graaf and
Frijters (1986) noted that the degree of hyperadditivity
C. Near-Threshold Interactions increases with increasing concentration levels (see Fig. 2).
Additivity corresponds with the predictions of the Beidler
Hahn and Ulbrich (1948) determined taste thresholds for (1971) mixture equation, which is based on competition
many combinations of similar tasting substances. Two for one set of receptors. Therefore, De Graaf and Frijters
substances A and B were defined to behave additively if the (1986) argued that glucose and fructose show complete
threshold concentration for AB mixtures deviated less than competition at low sweetness levels. With increasing
5% from the concentration predicted by combining p times sweetness levels, however, the degree of competition
Human Perception of Taste Mixtures 813

tures of artificial sweeteners (aspartame, saccharin, acesul-

fame, and cyclamate mixtures). This observation suggests
that substances with similar chemical structures (sugars,
sugar alcohols) show less hyperadditivity than substances
with more variation in chemical structures (artificial
sweeteners). Such an inference makes sense if multiple
sweetener receptors exist (Faurion et al., 1980) (see
Chapter 35). Alternatively, the observed hyperadditivity
can be described by a model stating that different numbers
of molecules react with one receptor or by a model stating
that independent substance-receptor complexes compete
for separate transducers (Ennis, 1989, 1991).

Figure 7 The taste interaction between monosodium glutamate III. MIXTURES OF DISSIMILAR TASTING
(MSG) and disodium 5-inosinate (IMP). The total taste SUBSTANCES
intensity of a solution is plotted as a function of the proportion of
A. Determination of Mixture Interactions
IMP in the mixture. The total amount of substance was kept at
0.05 g/dL. (Adapted from Yamaguchi, 1967.)
In mixtures of dissimilar tasting substances (often called
heterogeneous mixtures), one component does not con-
decreases because one or both sweeteners have additional tribute to the intensity of the sensation elicited by the other
binding sites, next to a large number of common sites. component. The phenomenon that the intensities of the
These results are in line with the finding of asymmetrical component sensations within and outside the mixture are
cross-adaptation between substances (Schiffman et al., equal is called independence. If the intensity in the mixture
1981; Lawless and Stevens, 1983). Frank et al. (1989) percept is higher than the intensity outside the mixture, this
observed the largest degrees of hyperadditivity for mix- is called synergism or mixture enhancement. The converse

Figure 8 (A) The sweetness intensity of sucrose, citric acid, and citric acid/sucrose mixtures as a function of the sweetness of unmixed
sucrose with a separate curve for each citric acid concentration. (B) The sourness of citric acid, sucrose, and citric acid/sucrose mixtures
as a function of the sourness of unmixed citric acid with a separate curve for each sucrose concentration. (Adapted from Schifferstein and
Frijters, 1990.)
814 Schifferstein

is called antagonism or mixture suppression (Berglund the research methodology is similar to that used for study-
et al., 1976; Frijters, 1987). ing mixtures of similar tasting substances (see sec. II.A).
Investigating taste interactions in mixtures of dissimilar De Graaf and Frijters (1989) developed a conceptual
tasting substances by studying the specific taste intensities framework describing the interrelationships among the
(sourness, sweetness, saltiness, and bitterness) of the physical and psychological intensity variables that play a
components inside and outside the mixture is fairly simple role in the perception of mixtures of dissimilar tasting
and straightforward. If the substances under investigation substances at supra-threshold concentration levels (Fig. 9).
elicit only one taste quality, the effect of substance A on the The physical concentration of an unmixed stimulus is
taste intensity of substance B can be assessed by comparing denoted by  and the physical concentration of a compon-
the intensity of unmixed B to the intensity of B in an AB mix- ent in a mixture by . The taste intensities of single
ture. For this type of investigation, only ordinal information is substances outside the mixture are denoted by  and the
needed to conclude whether suppression or enhancement taste intensities of the mixture or its components within the
occurs. Figure 8 shows the results of a study on taste inter- mixture are denoted by !. The Roman subscripts A and B
action in sucrose/citric acid mixtures (Schifferstein and refer to two dissimilar tasting chemicals, while the Greek
Frijters, 1990). In Figure 8A, the sweetness of the mixtures is subscripts  and  refer to the qualities of the specific taste
given as a function of the sweetness of unmixed sucrose. The sensations elicited by these two substances. The Greek
sweetness intensity of all mixtures lies below the diagonal, subscript  refers to total taste intensity. The subscripts i
implying mixture suppression. Figure 8B shows the sourness and j represent particular concentrations of the chemicals
of the mixtures as a function of the sourness of unmixed citric A and B in mol/L.
acid. Similar to Figure 8A, all mixtures lie below the diago- In Figure 9, relation 1 describes the physical mixing of i
nal, implying sourness suppression by sucrose. In addition, mol A (Ai) and j mol B (Bj) to obtain the mixture ABij.
Figure 8B clearly shows that the degree of suppression The lines connecting Ai and i (2A) and Bj and j (2B)
increases with increasing sucrose concentration. represent the psychophysical functions relating the physical
If taste interaction is studied using total taste intensity concentrations of the unmixed substances to their corre-
estimates of mixed and unmixed stimuli, both mixture sponding specific taste intensities (e.g., sucrose
components contribute to the intensity studied. Since taste concentration and sweetness intensity). The psychophysical
quality has now become an irrelevant stimulus attribute, functions relating stimulus concentrations (Ai and Bj) to

Figure 9 Outline of interrelationships among concentration levels, perceived specific intensities, and total taste intensities when two
qualitatively dissimilar tasting substances are mixed. (Adapted from De Graaf and Frijters, 1989.)
Human Perception of Taste Mixtures 815

total taste intensities (i and j) are given by relation 3.

Relation 4 relates the specific taste intensity elicited by
a stimulus to its total taste intensity. If the substances under
investigation elicit no side tastes, this relationship can be
described by an identity function (j
j).
Each physical mixture ABij evokes a total taste
intensity !ij (5). In addition, the mixture elicits two spe-
cific taste sensations, !i and !j (6). Relationship 7 gives
the connection between the specific taste sensations inside
(!j) and outside (j) the mixture. This is the relationship
where mixture interactions (suppression/enhancement) are
most evident. The total taste intensity of the mixture per-
cept may be related to the total taste intensities (8: i and
j) or specific taste intensities (9: i and j) of the
unmixed components, or to the specific taste intensities of
Figure 10 Relative ratio of the intensities in the mixture RM

the component sensations within the mixture percept
!sweet/!sweet  !sour as a function of the relative intensities of
(10: !i and !j).
the unmixed components RU
sweet/sweet  sour for
When the conceptual framework specified in Figure 9 is sucrose/citric acid mixtures. (Data from Schifferstein and
used, the intensity of each sensation evoked by the mixture Frijters, 1990. Adapted from Schifferstein and Kleykers, 1996.)
(!i, !j, and !ij) is measured as an attribute of the mix-
ture percept. Relation 10 investigates how the overall
intensity of the mixture is related to the intensities of its
component sensations. The specification of this relation- more citric acid should be added (RU
0.39).
ship implies that the specific taste sensations are regarded Asymmetrical suppression is frequently observed for
as the elements of a larger construct, the percept. The per- mixture components eliciting different taste qualities
cepts are considered sensory in nature, and a structuralistic (Schifferstein and Kleykers, 1996).
view is adopted in which specific taste sensations and total
taste intensity are all regarded as attributes of the mixture B. Qualitative Changes
percept.
Taste mixture interaction patterns depend mainly on the For two dissimilar tasting substances A and B that are
relative ratios of the perceived component intensities and mixed in successively different concentration ratios,
not on the absolute intensities (Schifferstein and Kleykers, Hambloch and Püschel (1928) described five stages of the
1996). Consequently, the interaction pattern can be char- perceptual experience. Beginning with a mixture where the
acterized by plotting the relative ratio of the intensities in quality elicited by substance A has a high intensity, while
the mixture the concentration of B is hardly perceptible, the quality
elicited by B is totally suppressed. This can be the case
RM
!i / (!i  !j) (10)
even though the concentration of B is detected when pre-
as a function of the relative intensities of the unmixed com- sented unmixed. When, subsequently, the concentration of
ponents A is decreased while that of B is increased, the quality of B
may not yet be recognizable but the presence of B changes
RU
i /(i  j) (11)
the character of the mixture percept. The mixture percept
(Olsson, 1993). The latter index was originally proposed seems to remain homogeneous in this second stage.
under the name  by Patte and Laffort (1979). Figure 10 However, the mixture can be discriminated from a solution
shows the results of these calculations for the sucrose/citric of unmixed A on the basis of taste quality. In the third
acid mixtures presented in Figure 8 after correcting stage, both components can be discriminated and can be
these data for differences in the size of scale units on the attended to. When both mixture components are easily
sweetness and the sourness scale (Schifferstein and perceptible, the sensations in the mixture percept differ
Frijters, 1990). This analysis shows that mixture suppres- qualitatively from the sensations elicited by the unmixed
sion in citric acid/sucrose mixtures is asymmetrical: when components. Kuznicki and Ashbaugh (1979) have shown
equi-intense components (RU
0.50) are mixed, the mix- that the size of the quality shift is directly related to the
ture tastes primarily sweet (RM
0.70). For a mixture in concentration of the added component. For example, if
which both components taste equally strong (RM
0.50), sucrose is added to NaCl, the qualitative difference in salti-
816 Schifferstein

ness quality between the mixture and unmixed NaCl sensation !ij, it should be impossible to scale the specific
increases with increasing sucrose level. After a further taste intensities of the component sensations within the
decrease of the A/B ratio, the quality elicited by A is no mixture percept (!i and !j), since no separate sensations
longer recognizable, but the quality of the mixture differs can be distinguished within a homogeneous percept. As
from that of a solution of unmixed B. Finally, in the fifth noted earlier in this chapter, the controversy between those
stage, the taste of A is fully suppressed by the presence of who argue that the taste modality functions analytically and
substance B. The extensions of the five stages discussed those who state that it is a synthetic sense goes back to the
above depend upon the two substances that are investi- nineteenth century (Öhrwall, 1891; Kiesow, 1894). More
gated (Hambloch and Püschel, 1928). recently, McBurney and colleagues (McBurney, 1974;
Some of these qualitative changes in the mixture per- McBurney and Gent, 1979) have defended the analytical
cept become apparent when subjects are asked to report position, whereas Schiffman and Erickson (1971, 1980)
whether a mixed stimulus consists of component A, com- have argued in favor of the synthetic view (see Chapter 37).
ponent B, or an AB mixture. Figure 11 gives the propor- According to Kubovy’s (1981) theory of indispensable
tion of subjects who thought a sucrose/citric acid mixture attributes, the perceived numerosity (heterogeneity) of a
consisted of only sweetener, only acid, or a mixture of discrete stimulus depends on whether the stimulus
these two substances as a function of the relative intensity elements vary on an indispensable attribute or not. Without
of the unmixed components (RU). The probability of detectable variation on such an indispensable attribute, the
identifying the mixture as one of the components reported perceived numerosity deviates from the physical
increases as the relative taste intensity of that component numerosity. Similar to all other senses, event time is an
increases. The probability that the stimulus is identified as indispensable attribute for taste: if two taste sensations are
a mixture is largest when the probability that it is identi- perceived one after the other, subjects are likely to con-
fied as a sweetener equals the probability that it is identi- clude that the overall percept consists of two elements.
fied as an acid. These data confirm Hambloch and Taste quality could be the second indispensable attribute
Püschel’s (1928) idea that intensity and quality are two for taste. However, taste quality has no corresponding
attributes that are heavily interdependent, and that both entity in the physical world. Chemically entirely different
attributes depend upon both solute concentrations: the substances may elicit similar taste sensations (e.g., sucrose
probability that a stimulus is identified as a mixture is and aspartame), whereas chemically more similar sub-
highest when the component intensities in the mixture are stances may elicit entirely different taste sensations (e.g.,
about equal (RM
0.50 in Fig. 10). HCl and NaCl). Although the complexity of a mixture per-
cept varies with the number of physical components and
C. Complexity of Mixture Perception with the ratio and the absolute concentration levels of the
components (O’Mahony et al., 1983), unmixed stimuli are
The sensations  and  elicited by two dissimilar tasting not consistently judged as singular (homogeneous percept)
substances may be perceived separately (analysis) or may and mixtures are not consistently perceived as more-than-
combine into one new mixture sensation  (synthesis). If one (heterogeneous percept) (Erickson and Covey, 1980;
the taste sensations synthesize into a new, homogeneous Erickson, 1982) (Fig. 11).

Figure 11 The proportion of cases in which

a sucrose/citric acid mixture was identified as
unmixed sweetener, unmixed acid, a mixture,
or water as a function of the relative intensities
of the unmixed components RU
sweet/sweet
 sour. (Adapted from Schifferstein and
Kleykers, 1996.)
Human Perception of Taste Mixtures 817

An important issue that has been overlooked in the ana- be good, they tend to oversimplify the perceptual
lytic-synthetic discussion is that even though taste stimuli process. As a result, they are unable to account for
may be automatically processed holistically, they can be several empirical phenomena, such as asymmetrical
analyzed if time and mental resources are available. When mixture interaction (Schifferstein and Frijters, 1993).
performing tasks under time constraints, subjects are Unlike the first type of models, models that predict mix-
incapable of selectively attending to one individual taste ture total intensity from the specific taste intensities in the
sensation in a mixture percept without noticing the other mixture do not have to account for mixture interactions,
sensation(s) in the percept. In addition, the presence of non- because they use the specific sensations elicited by the
target tastes results in a quality shift in the target sensation mixture as input. Implicit in the latter models is a serial
(Kuznicki and Ashbaugh, 1979, 1982). These findings show processing view: first, the mixture components interact,
that the sensory system primarily integrates information in and, subsequently, the resulting specific taste sensations
order to include all stimulus information into the percept yield the total taste intensity of the mixture. This type of
(Kroeze, 1990). The degree of dissimilarity between the model can only be applied to heterogeneous mixture
sensations is likely to affect the perceived complexity of the percepts, since no specific taste sensations can be
mixture percept. If component sensations are easily con- distinguished within a homogeneous percept. De Graaf
fused, the percept is likely to be more homogeneous than if and Frijters (1989) suggested that the total taste intensity
they are not. For example, sweetness is easily discriminated of a mixture percept follows from a weighted sum of the
from other sensations (Schiffman and Erickson, 1971), specific taste sensations elicited by that mixture. Since
whereas subjects often confuse sourness and bitterness empirical studies yielded weights near unity (De Graaf and
(Gregson and Baker, 1973). Therefore, sweet-sour percepts Frijters, 1989; Schifferstein and Frijters, 1990;
are probably more heterogeneous than sour-bitter percepts. Schifferstein and Frijters, 1992), this model was simplified
When subjects are not forced to make judgments under time to the (unweighted) sum of sensations model:
constraints, cognitive processes gain importance. Stimulus
!ij
!i  !j (12)
dimensions that are originally perceived as integral can be
analyzed into their component sensations, as more process- This model yields accurate predictions, and it can handle
ing time becomes available (Lockhead, 1966; Garner, all empirical mixture phenomena (Schifferstein and
1981). During the analysis of a pattern of taste sensations, Frijters, 1993).
attention shifts between the different parts of the percept. In
analyzing the mixture percept, subjects select the sensation E. Cognitive Factors
they are expected to judge without considering the other
sensations. The effort required in analysis depends heavily Mixture interactions are mostly regarded as the outcome of
upon the outcomes of the primary perceptual processes: the psychophysical and psychosensory processes. However,
complexity of the percept, the degree of organization, and mixture interactions as observed at the level of the
the locus of the integration process (Kroeze, 1990). obtained responses are affected by changes in stimulus set
and task requirements. Different tasks require different
D. Total Taste Intensity types of decision making and will, therefore, yield differ-
ent results. Consequently, every result has to be interpreted
Schifferstein and Frijters (1993) investigated what in relation to the method by which it was obtained.
happens when subjects are instructed to integrate all In an experiment on quinine•HCl/citric acid mixtures,
stimulus information to judge the total taste intensity of Frank et al. (1993) found the interaction pattern to vary
a mixture. In past research, the total intensity of a mix- with response task. If subjects were instructed to judge
ture (!ij) has been related either to the total taste inten- only the bitterness of the samples, the responses to the
sities of the unmixed components (i and j) (relation mixtures exhibited enhancement. If they judged the total
8 in Fig. 9) or to the specific taste intensities of the intensity of the stimulus and subsequently broke this rating
unmixed components within the mixture percept (!i up into eight component ratings, the bitterness responses
and !j) (relation 10 in Fig. 9). The first type of models exhibited suppression. Mixtures of quinine with sucrose
relates the mixture total intensity to the total intensities exhibited mixture suppression under all task instructions
of the unmixed components. As a consequence, these (Fig. 12). Frank et al. (1993) argued that the discrepancy
models need to account for mixture interactions. In addi- between the two conditions stems from a difference in the
tion they have to predict how total intensity is derived conceptual definitions used under the different task
from the characteristics of the mixture percept. Although instructions. Due to confusion of sourness and bitterness,
the predictive validity of some of these models may sourness intensity is (partly) included in the bitterness
818 Schifferstein

Figure 12 The effects of 0.25 M sucrose and 2.5 mM citric acid on the bitterness of quinine hydrochloride (QHCl). (A) Bitterness rat-
ings when subjects were instructed to judge only the bitterness intensity of each solution. (B) Bitterness ratings when subjects were
instructed to judge total taste intensity and break this rating up into eight component ratings (sweet, salty, sour, bitter, almond, lemon,
fruity, other). (Adapted from Frank et al., 1993.)

judgments during the first task, but not during the second are strikingly similar. Analogous to expectations derived
task. The sensations of sweetness and bitterness are from stimulus contexts, expectations based on precon-
qualitatively dissimilar and thus unlikely to be confused. ceived ideas about stimulus interactions could affect sub-
Consequently, they show a similar interaction pattern jects’ responses to taste mixtures.
under both task instructions.
Some experimental results suggest that characteristics of
the stimulus set, such as the proportion of mixtures in the F. Near-Threshold Interactions
stimulus set (Kroeze, 1982a; Schifferstein, 1994) and the
taste qualities perceived (Rankin and Marks, 1991), differ- J. C. Stevens (1995) investigated whether dissimilar tasting
entially affect ratings for mixed and unmixed stimuli. In components behave additively at detection threshold level.
these cases, the investigator is inclined to conclude that the Following his line of reasoning, additivity implies that the
degree of mixture interaction depends on stimulus set com- threshold concentration of each component in a mixture of
position. This effect of stimulus set composition is probably n components should equal 1/n times its threshold concen-
mediated by the expected properties for a target stimulus as tration measured separately. J. C. Stevens found no signifi-
evoked by the preceding stimuli (Cardello et al., 1993). cant deviations from the additivity rule for mixtures of
Human subjects can produce a fairly good estimate of sucrose, NaCl, citric acid, and quinine • HCl. In more
the intensity they expect for an imaginary mixture, of complex mixtures with up to 24 components representing
which they have only tasted the unmixed components the four taste qualities, however, the detection concentration
(Schifferstein, 1997; Stevenson and Prescott, 1997). levels tended to be higher than those expected under
Figure 13 shows a graph in which the perceived sourness conditions of additivity (Stevens, 1997). These findings sug-
of sucrose/citric acid mixtures (panel A) can be compared gest that some hypoadditivity occurs, even when all compo-
to the expected sourness of an imaginary sucrose/citric nents are present below their detection threshold level.
acid mixture (panel B). In the latter case, the subjects did Hahn and Ulbrich (1948) performed a series of experi-
not actually taste the mixture—they only tasted the ments in which the threshold concentration for substance
components separately. Despite differences, the two plots A was determined in the presence of a subthreshold
Human Perception of Taste Mixtures 819

Figure 13 Factorial plot of the mean sourness intensity ratings for perceived (A) and imaginary (B) mixtures of sucrose and citric acid.
The citric acid concentrations were 0.0, 1.25, 2.5, 5.0, and 10.0 mM. (Adapted from Schifferstein, 1997.)

concentration of a dissimilar tasting substance B. For 93 G. Suprathreshold Interactions

out of 123 combinations tested, they found thresholds
equal to the thresholds for unmixed A. For the other 30 As noted in previous sections, the experimental results of
binary combinations, the threshold concentration of A in a mixture experiment depend upon the way in which the
the mixture was lower than the concentration for unmixed experiment is carried out. Mixture components may elicit
A. This decrease in threshold concentration was partly taste sensations that are, to some degree, similar or con-
accounted for by the fact that component B produced a side fusable. In this case, instructing subjects to judge only one
taste equal in quality to the taste elicited by A. specific taste intensity may result in finding mixture
Threshold concentrations generally increase if a enhancement (Frank et al., 1993). The degree of enhance-
suprathreshold concentration of a dissimilar tasting com- ment is positively correlated with the reported intensity of
ponent is added (Heymans, 1899). The size of the the components’ side tastes (Kroeze, 1982b). For example,
increase in threshold concentration is highly substance the enhancement of the sweet taste of sucrose by NaCl
specific. J. C. Stevens (1996) reported that a high con- depends upon the intensity of the sweetness of NaCl.
centration of NaCl raised the threshold concentration of However, if subjects are requested to judge multiple
citric acid nearly 13-fold, whereas those for sucrose and stimulus attributes or if they are instructed to judge the dif-
quinine•HCl increased only 3-fold. The effects of a ference in intensity between a mixed and an unmixed stim-
masking component on the detection threshold of a ulus, mixture suppression is found (Pangborn, 1961; Frank
dissimilar tasting substance seem to be symmetrical: for et al., 1993). Since mixture suppression can occur under all
all combinations of sucrose, NaCl, citric acid, and different types of instructions, but enhancement occurs
quinine•HCl, Stevens (1996) found that the effect of a only for some of them, mixture suppression seems to be
proportional increase in the concentration of masker A on sensory in nature, whereas enhancement seems primarily
the threshold concentration for B was similar to the effect cognitive.
of a proportional increase in the level of masker B on the Mixture suppression has been demonstrated for many
threshold concentration of A. When two supra-threshold different substance combinations (Pangborn, 1960). The
maskers are combined, their effects on the detection degree of suppression depends upon the method used, the
threshold of a third substance seem to add up (Stevens taste qualities involved, the substances used, and their con-
and Traverzo, 1997). centration levels. It would be confusing and tedious to
820 Schifferstein

summarize all mixture interaction patterns reported in the of interaction for similar tasting substances is related to
existing literature. Instead, a number of phenomena have differences in the components’ chemical structures.
been observed in mixture research that give an impression In interactions between dissimilar tasting substances, not
of the complexity of this field (see below). the chemical properties but the relative intensities of the
First, mixture interaction patterns depend mainly on the unmixed components appear to be a primary determinant
relative ratios of the perceived component intensities and of the degree of interaction. Unfortunately, the current state
not on the absolute intensities (Schifferstein and Kleykers, of knowledge does not permit predictions of the degree of
1996). In addition, the suppressive effect of one taste qual- mixture suppression and the direction and size of its asym-
ity on another quality may be similar for various sub- metry from knowledge about the physical and chemical
stances. For example, Schifferstein and Frijters (1991) characteristics of the mixture components only.
found that equisweet sweetener concentrations suppressed
the sourness of citric acid to the same degree. Further
research is needed to demonstrate whether this result is a ACKNOWLEDGMENTS
general finding for taste interactions or whether it is qual-
ity specific. However, both these principles indicate that The author thanks R. A. Frank, J. E. R. Frijters, and M. J.
perceived intensity is more important in determining the M. Theunissen for their useful comments and suggestions
degree of mixture suppression than the chemical structures on an earlier draft of this chapter.
or the concentration levels of the tastants involved.
Second, mixture suppression is often asymmetrical when REFERENCES
both mixture components are present at supra-threshold
levels (Schifferstein and Kleykers, 1996). At equi-intense Bartoshuk, L. M. (1975). Taste mixtures: Is mixture suppression
levels of the unmixed components, one sensation may be related to compression? Physiol. Behav. 14:643–649.
suppressed more than the other sensation (Fig. 10). In Bartoshuk, L. M., and Cleveland, C. T. (1977). Mixtures of sub-
quinine/NaCl mixtures, for example, the bitterness of qui- stances with similar tastes: A test of a psychophysical model
nine is suppressed to a large degree when NaCl is added to of taste mixture interactions. Sensory Proc. 1:177–186.
quinine, whereas the saltiness elicited by NaCl remains Beidler, L. M. (1971). Taste receptor stimulation with salts and
almost unaffected (Schifferstein and Frijters, 1993). Third, acids. In Handbook of Sensory Physiology, Vol. 4, Chemical
the total taste intensity of a binary mixture is well pre- Senses, Part 2, Taste, L. M. Beidler (Ed.). Springer-Verlag,
New York, pp. 200–220.
dicted by the sum of the specific taste intensities in the
Berenbaum, M. C. (1989). What is synergy? Pharmacol. Rev.
mixture percept (De Graaf and Frijters, 1989; Schifferstein 41:93–141.
and Frijters, 1993). Consequently, the total intensity of a Berglund, B., Berglund, U., Lindvall, T., and Svensson, L. T.
mixture can be higher or lower than the intensity of one of (1973). A quantitative principle of perceived intensity sum-
its unmixed components, depending on the degree of sup- mation in odor mixtures. J. Exp. Psychol. 100:29–38.
pression in the mixture. Berglund, B., Berglund, U., and Lindvall, T. (1976).
Psychological processing of odor mixtures. Psychol. Rev.
83:432–441.
IV. SUMMARY Birch, G. G. (1980). Theory of sweetness. In Carbohydrate
Sweeteners in Food and Nutrition, P. Koivistoinen, and
Most food products can be conceptualized as extremely L. Hyvönen (Eds.). Academic Press, London, pp. 61–75.
complex mixtures of substances that can stimulate one or Boring, E. G. (1942). Sensation and Perception in the History of
more of the senses. Taste mixture research teaches us that Experimental Psychology. Appleton-Century-Crofts, New York.
even the interaction between two components that elicit Cameron, A. T. (1947). The Taste Sense and the Relative
gustatory sensations only is already quite complex. At Sweetness of Sugars and Other Sweet Substances. Scientific
detection threshold level, many substances behave addi- Report Series No. 9, Sugar Research Foundation, New
tively. However, threshold concentrations increase in the York.
Cardello, A. V., Melnick, S. M., and Rowan, P. A. (1993).
presence of a dissimilar tasting supra-threshold component.
Expectation as a mediating variable in context effects. In
If two components are present at supra-threshold level, Proceedings of the Food Preservation 2000 Conference,
mixtures of similar tasting substances behave additively or Natick, MA.
hyperadditively, whereas dissimilar tasting substances Carr, W. E. S., Trapido-Rosenthal, H. G. and Gleeson, R. A.
exhibit suppression. Reports of mixture enhancement in (1989). Stimulants of feeding behavior in marine organ-
heterogeneous mixture types are probably due to side tastes isms: receptor and perireceptor events provide insight into
of the substances used or to method-dependent changes mechanisms of mixture interactions. In Perception of
in the subjects’ conceptual definitions. Possibly, the degree Complex Smells and Tastes, D. G. Laing, W. S. Cain, R. L.
Human Perception of Taste Mixtures 821

McBride, and B. W. Ache (Eds.). Academic Press, Sydney, Garner, W. R. (1981). The analysis of unanalyzed perceptions. In
pp. 27–45. Perceptual Organization, M. Kubovy, and J. R. Pomerantz
De Graaf, C. (1989). Psychophysical Studies of Mixtures of (Eds.). Erlbaum, Hillsdale, NJ, pp. 119–139.
Tastants. Doctoral dissertation, Wageningen Agricultural Gregson, R. A. M., and Baker, A. F. H. (1973). Sourness and bit-
University, Wageningen, the Netherlands. terness: confusions over sequences of taste judgments. Br. J.
De Graaf, C., and Frijters, J. E. R. (1986). A psychophysical investi- Psychol. 64:71–76.
gation of Beidler’s mixture equation. Chem. Senses 11:295–314. Hahn, H., and Ulbrich, L. (1948). Eine systematische
De Graaf, C., and Frijters, J. E. R. (1988). Assessment of the taste Untersuchung der Geschmacksschwellen. III. Abschliessende
interaction between two qualitatively similar-tasting sub- Mitteilung. Pflügers Arch. Gesammte Physiol. Menschen
stances: a comparison between comparison rules. J. Exp Tiere 250:357–384.
Psychol. Human Percept. Perform. 14:526–538. Hambloch, H., and Püschel, J. (1928). Über die sinnlichen
De Graaf, C., and Frijters, J. E. R. (1989) Interrelationships Erfolge bei Darbietung von Geschmacksmischungen.
among sweetness, saltiness and total taste intensity of Zeitschrift Sinnesphysiol. 58:136–150.
sucrose, NaCl and sucrose/NaCl mixtures. Chem. Senses Heymans, G. (1899) Untersuchungen über psychische
14:81–102. Hemmung. Zentralbl. Physiol. 21:321–359.
De Graaf, C., Frijters, J. E. R., and van Trijp, H. C. M. (1987). Kamen, J. M, Pilgrim, F. J., Gutman, N. J., and Kroll, B. J.
Taste interaction between glucose and fructose assessed by (1961). Interactions of suprathreshold taste stimuli. J. Exp.
functional measurement. Percept. Psychophys. 41:383–392. Psychol. 62:348–356.
Ennis, D. M. (1989). A receptor model for binary mixtures Kiesow, F. (1894). Beiträge zur physiologischen Psychologie des
applied to the sweetness of fructose and glucose: De Graaf Geschmackssinnes. Philosoph. Studien 10:523–561.
and Frijters revisited. Chem. Senses 14:597–604. Kiesow, F. (1896). Beiträge zur physiologischen Psychologie des
Ennis, D. M. (1991). Molecular mixture models based on com- Geschmackssinnes. Philosoph. Studien 12:255–278.
petitive and non-competitive agonism. Chem. Senses 16:1–17. Klitzner, M. D. (1975). Hedonic integration: test of a linear
Erickson, R. P. (1982). Studies on the perception of taste: Do pri- model. Percept. Psychophys. 18:49–54.
maries exist? Physiol. Behav. 28:57–62. Kroeze, J. H. A. (1978). The taste of sodium chloride: Masking
Erickson, R. P., and Covey, E. (1980). On the singularity of taste and adaptation. Chem. Senses Flavour 3:443–449.
sensations: What is a taste primary? Physiol. Behav. 25: Kroeze, J. H. A. (1979). Masking and adaptation of sugar sweet-
527–533. ness intensity. Physiol. Behav. 22:347–351.
Fabian, F. W., and Blum, H. B. (1943). Relative taste potency of Kroeze, J. H. A. (1982a). The influence of relative frequencies of
some basic food constituents and their competitive and com- pure and mixed stimuli on mixture suppression in taste.
pensatory action. Food Res. 8:179–183. Percept. Psychophys. 31:276–278.
Faurion, A., Saito, S., and MacLeod, P. (1980). Sweet taste Kroeze, J. H. A. (1982b). The relationship between the side tastes
involves several distinct receptor mechanisms. Chem. Senses of masking stimuli and masking in binary mixtures. Chem.
5:107–121. Senses 7:23–37.
Fechner, G. T. (1860). Elemente der Psychophysik. Leipzig: Kroeze, J. H. A. (1990). The perception of complex taste stim-
Breitkopf und Härtel. uli. In Psychological Basis of Sensory Evaluation, R. L.
Frank, R. A., Mize, S. J. S., and Carter, R (1989). An assessment McBride, and H. J. H. MacFie (Eds.). Elsevier, London, pp.
of binary mixture interactions for nine sweeteners. Chem. 41–68.
Senses 14:621–632. Kubovy, M. (1981). Concurrent-pitch segregation and the theory
Frank, R. A., van der Klaauw, N. J., and Schifferstein, H. N. J. of indispensable attributes. In Perceptual Organization, M.
(1993). Both perceptual and conceptual factors influence Kubovy, and J. R. Pomerantz (Eds.). Erlbaum, Hillsdale, NJ,
taste-odor and taste-taste interactions. Percept. Psychophys. pp. 55–98.
54:343–354. Külpe, O. (1893). Grundriss der Psychologie. Engelmann,
Frijters, J. E. R. (1987). Psychophysical models for mixtures of Leipzig, pp. 30–33.
tastants and mixtures of odorants. In Olfaction and Taste IX, Kuznicki, J. T., and Ashbaugh, N. (1979). Taste quality differ-
S. D. Roper, and J. Atema (Eds.). The New York Academy of ences within the sweet and salty taste categories. Sensory
Sciences, New York, pp. 67–78. Proc. 3:157–182.
Frijters, J. E. R. (1993). Functional measurement in the study of Kuznicki, J. T., and Ashbaugh, N. (1982). Space and time
mixture percepts. Chem. Senses 18:93–100. separation of taste mixture components. Chem. Senses
Frijters, J. E. R., and Oude Ophuis, P. A. M. (1983). The con- 7:39–62.
struction and prediction of psychophysical power functions Lawless, H. T. (1979). Evidence for neural inhibition in bittersweet
for the sweetness of equiratio sugar mixtures. Perception taste mixtures. J. Compar. Physiol. Psychol. 93:538–547.
12:753–767. Lawless, H. T., and Stevens, D. A. (1983). Cross adaptation
Frijters, J. E. R., van der Klaauw, N. J. B. M., and Kranen, B. of sucrose and intensive sweeteners, Chem. Senses
(1990). The use of molal concentrations in a psychophysical 7:309–315.
investigation of Beidler’s mixture equation. Chem. Senses Lockhead, G. R. (1966). Effects of dimensional redundancy on
15:659–669. visual discrimination. J. Exp. Psychol. 72:95–104.
822 Schifferstein

McBride, R. L. (1989). Three models for taste mixtures. In Schifferstein, H. N. J. (1995). Prediction of sweetness intensity
Perception of Complex Smells and Tastes, D. G. Laing, W. S. for equiratio aspartame/sucrose mixtures. Chem. Senses 20:
Cain, R. L. McBride, and B. W. Ache (Eds.). Academic Press, 211–219.
Sydney, pp. 265–282. Schifferstein, H. N. J. (1996). An equiratio mixture model for
McBurney, D. H. (1974). Are there primary tastes for man? non-additive components: A case study for aspartame/acesul-
Chem. Senses Flavour 1:17–28. fame-K mixtures. Chem. Senses 21:1–11.
McBurney, D. H., and Gent, J. F. (1979). On the nature of taste Schifferstein, H. N. J. (1997). Perceptual and imaginary mixtures
qualities. Psychol. Bull. 86:151–167. in chemosensation. J. Exp. Psychol. Human Percept. Perform.
Meiselman, H. L. (1972). Human taste perception. CRC Crit. 23:278–288.
Rev. Food Technol. 3:89–119. Schifferstein, H. N. J., and Frijters, J. E. R. (1990). Sensory inte-
Moskowitz, H. R. (1974). Models of additivity for sugar sweet- gration in citric acid/sucrose mixtures. Chem. Senses 15:
ness. In Sensation and Measurement, H. R. Moskowitz, B. 87–109.
Scharf, and J. C. Stevens (Eds.) Reidel, Dordrecht, The Schifferstein, H. N. J., and Frijters, J. E. R. (1991). The effec-
Netherlands, pp. 379–388. tiveness of different sweeteners in suppressing citric acid
Nahon, D. F., Roozen, J. P., and De Graaf, C. (1996). Sweetness sourness. Percept. Psychophys. 49:1–9.
flavour interactions in soft drinks. Food Chem. Schifferstein, H. N. J., and Frijters, J. E. R. (1992). Two-stimulus
56(3):283–289. versus one-stimulus procedure in the framework of functional
Öhrwall, H. (1891). Untersuchungen über den Geschmackssinn. measurement: a comparative investigation using
Skand. Arch. Physiol. 2:1–69. quinineHCl/NaCl mixtures. Chem. Senses 17:127–150.
Olsson, M. J. (1993). The perception of odors in interaction. Schifferstein, H. N. J., and Frijters, J. E. R. (1993). Perceptual
Doctoral dissertation, department of psychology, University integration in heterogeneous taste percepts. J. Exp. Psychol.
of Stockholm, Sweden. Human Percept. Perform. 19:661–675.
O’Mahony, M., Atassi-Sheldon, S., Rothman, L., and Murphy- Schifferstein, H. N. J. and Kleykers, R. W. G. (1996). An empir-
Ellison, T. (1983). Relative singularity/mixedness judgements ical test of Olsson’s interaction model using mixtures of tas-
for selected taste stimuli. Physiol. Behav. 31:749–755. tants. Chem. Senses 21:283–291
Pangborn, R. M. (1960). Taste interrelationships. Food Res. Schiffman, S. S., Cahn, H., and Lindley, M. G. (1981). Multiple
25:245–256. receptor sites mediate sweetness: evidence from cross adapta-
Pangborn, R. M. (1961). Taste interrelationships. II. tion. Pharmacol. Biochem. Behav. 15:377–388.
Suprathreshold solutions of sucrose and citric acid. J. Food Schiffman, S. S., and Erickson, R. P. (1971). A theoretical review:
Sci. 26:648–655. a psychophysical model for gustatory quality. Physiol. Behav.
Pangborn, R. M. (1962). Taste interrelationships. III. 7:617–633.
Suprathreshold solutions of sucrose and sodium chloride. Schiffman, S. S., and Erickson, R. P. (1980). The issue of primary
J. Food Sci. 27:495–500. tastes versus a taste continuum. Neurosci. Biobehav. Rev.
Pangborn, R. M. (1965). Taste interrelationships of organic acids 4:109–117.
and selected sugars. In Food Science and Technology. Vol. III. Stevens, J. C. (1995). Detection of heteroquality taste mixtures.
Quality Analysis and Composition of Foods, J. M. Leitsch Percept. Psychophys. 57:18–26.
(Ed.). Gordon and Breach, London, pp. 291–305. Stevens, J. C. (1996). Detection of tastes in mixture with other
Pangborn, R. M., and Chrisp, R. B. (1964). Taste interrelation- tastes: issues of masking and aging. Chem. Senses 21: 211–221.
ships. VI. Sucrose, sodium chloride, and citric acid in canned Stevens, J. C. (1997). Detection of very complex taste mixtures:
tomato juice. J. Food Sci. 29:490–498. generous integration across constituent compounds. Physiol.
Pangborn, R. M., and Trabue, I. M. (1964). Taste interrelation- Behav. 62:1137–1143.
ships. V. Sucrose, sodium chloride, and citric acid in lima Stevens, J. C., and Traverzo, A. (1997). Detection of a target taste
bean purée. J. Food Sci. 29:233–240. in a complex masker. Chem. Senses 22:529–534.
Patte, F., and Laffort, P. (1979). An alternative model of olfactory Stevens, S. S. (1975). Psychophysics. Wiley, New York.
quantitative interaction in binary mixtures. Chem. Senses Stevenson, R. J., and Prescott, J. (1997). Judgments of chemosen-
Flavour. 4:267–274. sory mixtures in memory. Acta Psychol. 95:195–214.
Peryam, D. R. (1990). Sensory evaluation—the early days. Food Sühnel, J. (1993). Evaluation of interaction in olfactory and taste
Technol. 44(1):86–91. mixtures. Chem. Senses 18:131–149.
Rankin, K. M., and Marks, L. E. (1991). Differential context Von Skramlik, E. (1926). Handbuch der Physiologie der
effects in taste perception. Chem. Senses 16:617–629. Niedrigen Sinne. Vol. 1. Die Physiologie des Geruchs- und
Rifkin, B., and Bartoshuk, L. M. (1980). Taste synergism Geschmackssinnes. Georg Thieme, Leipzig.
between monosodium glutamate and disodium 5-guanylate. Yamaguchi, S. (1967). The synergistic taste effect of mono sodium
Physiol. Behav. 24:1169–1172. glutamate and disodium 5-inosinate. J. Food Sci. 32: 473–478.
Schifferstein, H. N. J. (1994). Sweetness suppression in fruc- Yamaguchi, S., Yoshikawa, T., Ikeda, S., and Ninomiya, T. (1970).
tose/citric acid mixtures: a study of contextual effects. Studies on the taste of some sweet substances. Part II.
Percept. Psychophys. 56:227–237. Interrelationships among them. Agric. Biol. Chem. 34:187–197.
39

The Ontogeny of Human Flavor Perception

Judith R. Ganchrow
The Hebrew University–Hadassah School of Dental Medicine Founded by the Alpha Omega Fraternity, Jerusalem, Israel
Julie A. Mennella
Monell Chemical Senses Center, Philadelphia, Pennsylvania, U.S.A.

I. INTRODUCTION Retronasal olfaction contributes significantly to the

complexity of flavor. This is clearly noted by head cold
Flavor, a powerful determinant of human consummatory sufferers who lose the ability to discriminate common
behavior throughout the life span, is a product of several foods when olfactory receptors are blocked. Similarly,
sensory systems, most notably those of taste and smell. foods often “taste” better after a person quits smoking,
The perceptions arising from these two senses are often perhaps because their sense of smell has improved (Frye
confused and misappropriated, with such sensations as et al., 1990), allowing them to detect more subtleties of
vanilla, meat sauce, fish, chocolate, and coffee being erro- flavor. The role of smell in flavor is critical in distin-
neously attributed to the taste system per se. In fact, taste guishing the flavor of strawberry from cherry and in
sensations, mediated by taste buds distributed throughout enjoying foods like licorice, vanilla, and citrus, which are
the oral cavity (see Chapter 32), are largely those of sweet, experienced through the sense of smell. It should be noted
sour, bitter, salty, and perhaps metallic and savory (e.g., the that other properties of food (e.g., texture, temperature,
“umami” taste of monosodium glutamate). Smell sensa- irritation) are also very important to its perceived flavor.
tions, on the other hand, encompass thousands of diverse Because little experimental work has been done in this
qualities, some of which are noted above. The receptors for area of infant flavor perception, this article focuses on
the olfactory system, located high in the nasal chambers, human taste and olfactory development, alone and in uni-
are stimulated not only during inhalation (orthonasal son, as it relates to flavor. Further discussion of the
route), but during suckling in infants and deglutition in ontogeny of olfactory perception as it relates to social
adults, when molecules reach the receptors by passing behaviors can be found in Chapter 15.
from the oral cavity through the nasal pharynx (retronasal Contemporary research on the development of taste and
route) (Fig. 1). It is this retronasal stimulation arising from smell is based on more than a century-long legacy examin-
the molecules of foodstuffs that leads to the predominant ing receptor morphogenesis and determining functional
flavor sensations. As noted in detail in several chapters of maturity. Early studies attempted to determine the onset of
this volume (e.g., Chapters 2–6), olfactory sensations function and how it related to mature perception. However,
result from the activation of 1000 or more distinct types of research on preverbal infants obviated the use of traditional
chemical receptors located on millions of receptor cells threshold and suprathreshold psychophysical measures,
lining the upper recesses of the nasal cavity. and when such methods are applied in early childhood,
823
824 Ganchrow and Mennella

These difficulties notwithstanding, dramatic strides

have been taken in the past few years, especially in the
measurement of outcomes of early experience, in poten-
tially modifying the hedonic value of oral stimuli during
development. Both preference measures of ingestion and
accompanying motor displays in infants reflect the hedonic
aspect of chemosensory stimuli (e.g., Berridge, 2000) and
have provided a key for understanding developmental pref-
erences and aversions. Available data support the thesis
that behaviors associated with gustatory and olfactory
function are robust during early development and con-
tribute significantly to adaptation, if not survival, in the
developing infant’s world. At the same time, flavor
perception and the accompanying expression of consum-
Figure 1 Sagittal section of infant’s head demonstrating matory-related behaviors are influenced by postnatal
retronasal and orthonasal routes of olfaction. maturational and both physiological and environmental
events and are exquisitely vulnerable to modification
throughout the life span.
The focus of this chapter is on the ontogeny of human
cognitive factors, such as attention span, often make testing utilization of the senses of taste and smell, often in the
difficult. One solution to such problems was to focus on context of flavor. A brief description of taste and olfactory
reflex-like responses (e.g., salivation, patterned facial receptor development pre- and perinatally will serve as a
motor reactions including sucking). However, because foundation for assessing the behavioral findings.
touch by itself can initiate sucking behavior, it could be
mistakenly concluded that an infant doesn’t respond to a
liquid tastant if its concentration is not sufficient to override
II. ONTOGENY OF TASTE AND OLFACTORY
touch responses. This may explain why several early stud-
RECEPTORS
ies failed to demonstrate taste discrimination in infants (see
Lichtenstein, 1893/94). Furthermore, gustatory modulation A. Taste
of some of these response patterns appears to have a very
limited developmental time span, so the same measures Early studies, often based on very limited observations
cannot be performed at all ages (see below). (e.g., Hellman, 1922; Hoffmann, 1875; Tuckerman, 1890),
As an alternative to the study of reflex motor coordin- suggested that taste buds and their positions in taste papil-
ation, consummatory behavior has also been used for lae appear in an adult-like form prenatally. More detailed
cross-age comparisons. This, too, can be problematic recent studies indicate that fungiform, foliate, and circum-
because such behavior assumes uniformity in the ability to vallate papillae are all present by about week 10 of gesta-
control and modulate ingestion as a function of taste or fla- tional age (Bradley and Stern, 1967; Hersch and Ganchrow,
vor feedback irrespective of age and potentially confounds 1980; Piras and Mazzarello, 1985; Witt and Reutter, 1997).
sensory measurement with preference measurement (indif- Originally, the distribution of fungiform papillae was
ference does not imply not tasting). In addition, fluctu- believed to shift from a uniform scattering across the ante-
ations in factors mediating thirst and/or postingestional rior two thirds of the tongue in infants to more lateral and
responses may override the effect of flavor components in distal concentrations in children and adults. Goldschmidt
controlling intake volume. Moreover, not all chemicals (1927) refuted this concept by mapping the tongues of 80
utilized in adult taste perception studies can be offered to living infants and children ranging in age from 1 week to
infants for ingested volume measures (e.g., see Crook, 13 years. He found both types of papillar distributions evi-
1981). Lastly, consummatory behavioral studies may be dent across this age range, with no progressive tendency
highly variable if basic stimulus (e.g., concentration, tem- favoring either. Indeed, the number and distribution of late
perature, volume, and type of diluent) and methodological gestational fetal papillae appear to be roughly similar to
(e.g., degree of satiation and experience) parameters are those throughout childhood and adulthood.
not controlled. However, many of these issues can be Serial section analysis of fetal tongues by Goldschmidt
resolved by using a within-subject design that controls for (1927) revealed that early fungiform papillae (about
these variables. 4 months) contained only single apical taste buds. More
Ontogeny of Human Flavor Perception 825

recent studies have identified taste pores in fetal fungiform in discrimination, or taste perception in general, might be
papillae before the end of the 4th month (e.g., Bradley, expected to depend on central rather than peripheral pro-
1972; Hersch and Ganchrow, 1980; Witt and Reutter, cessing changes during development, although some taste
1996) (see also Chapter 32). Taste pores provide access for cell receptor membrane features could be dependent upon
gustatory stimuli to interact with taste bud receptor cells experience or ontogeny.
and are generally considered to mark functional maturity
(Mistretta, 1972), although taste bud cells continue differ-
entiating after the pores open (Witt and Reutter, 1997). B. Olfaction
Between weeks 8 and 13, taste cell synaptogenesis is
increasingly apparent (Witt and Reutter, 1996). As noted in detail in Chapters 2, 6, and 15, the human
Clearly the newborn infant is endowed with a rich olfactory system and the formation of the nasal structures
population of gustatory receptors. For example, extralin- are also well developed prior to birth. Primary olfactory
gually, 2583 taste buds were reported to be dispersed over receptors are present by the 8th week of gestation (see
the soft palate, epiglottis, pharynx, and larynx (Lalonde Schaal, 1988; Schaal et al., 1995b, for review).
and Eglitis, 1961). In addition, each circumvallate papilla Furthermore, histochemical analyses of human fetal
(of which there are usually 9–12) is estimated to already olfactory tissue have noted the presence of olfactory
contain around 250 taste buds between the ages of 0 and 11 marker protein, a biochemical correlate of olfactory recep-
months (Heiderich, 1906); each foliate papilla around tor function, as early as the 24th week of gestation (Chuah
1500 buds (Tuckerman, 1888); and each fungiform papil- and Zheng, 1987; Johnson et al., 1995).
lae 0–12 taste buds (Arvidson, 1979). In general, there is By the 6th month of gestation, the epithelial plugs that
no linkage between age and number of taste buds per obstruct the external nares have resolved, and these open air-
fungiform papilla (e.g., Arvidson, 1979). In children, the way passages then become continuously bathed in amniotic
number of taste buds per circumvallate papilla was fluid (Schaffer, 1910). During the later stages of gestation,
reported by Heiderich (1906) to remain constant from 1–3 the fetus swallows significant amounts of amniotic fluid,
years (242  18, N
20) through 4–20 years (252  74, inhaling more than twice the volume it swallows (Pritchard,
N
9). Other work suggests there may be a slight increase 1965; Schaffer, 1910). As will be discussed below, increasing
in number of foliate taste buds from birth to 60 years (see evidence suggests that the fetus may be exposed to a unique
Cowart, 1981, for review). Due to large interindividual chemosensory environment prior to birth. In other words,
variability in adult papillary taste bud numbers (e.g., amniotic fluid may represent the first exposure to flavors
Miller, 1988), longitudinal developmental studies would (both tastes and retronasal odors) that will subsequently be
be required to assess whether there is an intraindividual provided by mother’s milk and then the foods of the table.
progression in bud numbers. Unfortunately, this is a tech-
nically difficult, if not impossible, task. However, it is
doubtful that taste recognition would be significantly
III. ONTOGENY OF BEHAVIORAL
affected, even if there was an increase in bud numbers dur-
RESPONSES: TASTE
ing childhood development, since (1) a single fungiform
papilla contains the taste receptor population sufficient to A. Fetus
discriminate between different taste stimuli (e.g., Harper et
al., 1966) and (2) adults can recognize salt, sweet, sour, Although the response of human fetuses to tastants has
and bitter using only one taste bud (Arvidson and Friberg, never been directly investigated, indirect evidence suggests
1980). Using videomicroscopy, it has recently been that prenatal monitoring of taste stimuli dissolved in amni-
reported that while the number of taste pores/fungiform otic fluid is possible in late gestation. Thus, chemosensory
papilla is similar in young adults and 8-year-olds, the lat- stimuli may modulate the intrauterine tendency to swallow
ter have a higher density of papillae (Hutchinson et al., (see Beauchamp et al., 1991; Mistretta and Bradley, 1975,
2000). This could give children an advantage for both for comment and review). In this context, increased swal-
threshold performance and suprathreshold magnitude esti- lowing following sweet, and decreased swallowing follow-
mates if only a small area of tongue is stimulated. This ing bitter, could be interpreted as an early preference and
advantage should disappear with tongue growth increasing aversion, respectively. In addition, consequences of physi-
the space between papillae (see Temple et al., 2002). ological challenges (e.g., electrolyte imbalances) to the
Overall, the gustatory system is remarkably redundant fetus during pregnancy, in terms of postnatal preferences
in its receptor endowment. Assuming the taste bud (see below), could be partially mediated by the accompa-
complement is complete at birth, any major improvement nying fetal taste experience.
826 Ganchrow and Mennella

B. Premature Infant Thus, on the basis of these studies it may be concluded

that taste buds are capable of conveying gustatory infor-
Taste-induced behavioral responses have been reported in mation to the central nervous system by the 6th gestational
premature infants (6–9 months gestational age) using a month and that information is available to systems orga-
variety of chemicals, stimulation techniques, and record- nizing changes in salivation, sucking, facially expressive,
ing methodologies. From early in the 6th gestational and other affective behaviors.
month, sweet stimuli (represented by lactose, glucose,
sucrose, and/or saccharin solutions) were found to consis-
tently modulate the sucking reflex whether via an artificial C. Newborn Infant
nipple for electrophysiological measurement or via stimu-
lus-soaked absorbent cotton for direct observation of 1. Skeletal and Autonomic Reactions to Taste Stimuli
motor responses (Eckstein, 1927; Maone et al., 1990; Facial expression has long been recognized as a potential
Martin du Pan 1955; Stirnimann, 1935; Tatzer et al., 1985). tool for studying gustatory sensation in newborn infants,
While some of these studies were not done blinded, the although late nineteenth- and early twentieth-century
authors usually interpreted the response as a “strong pos- research yielded inconsistent results (see Büssem, 1895;
itive” or “acceptance” behavior. These data suggest that Kussmaul, 1894; Peiper, 1928; Preyer, 1912; Stirnimann,
the gustatory system is functioning and interacting with 1935). More recently, J. E. Steiner pioneered the use of
systems controlling affect. Taste reactivity, as measured by videoanalysis in the study of taste-mediated facial expres-
orofacial expressive motor patterns at any age, may be sub- sions in infants, and results have been cross-culturally
ject to hierarchical brain modulation such that the final confirmed, indicating neonates are capable of robust and
product carries more of a hedonic or palatability message, differentiable taste experiences in the first postnatal hours
rather than straight stimulus identification (see Berridge, or days (e.g., Bergamasco and Berlado, 1990; Ganchrow
2000, for review). et al., 1983; Kalmus, 1976; Rosenstein and Oster, 1988;
Infants born around 8 weeks earlier than term (1.1–2.8 Steiner, 1973, 1974, 1977, 1979a, b, 1987). Taste stimuli
kg) cease crying transiently when repeatedly receiving 0.1 eliciting these behaviors have included chemical solutions
ml of 12% sucrose or 10% glucose intraorally, but the labeled by adults as sweet, sour, bitter, and umami (“deli-
response is much less durable than is seen for preterm (born cious”). As in premature infants, the definition of the
36–37 weeks) or term-born infants (Smith and Blass, response to “salty” (NaCl) remains elusive, but generally
1996). Water under the same conditions is ineffective. In tends towards the indifferent-to-aversive (grimace, crying)
addition, palatable (e.g., sweet) taste stimuli promote end of the spectrum (e.g., Lichtenstein, 1893/1894;
“calming,” analgesic-like reactions in preterm as well as Maekawa et al., 1991; Rosenstein and Oster, 1988;
more mature infants during heel lance and other invasive Stirnimann, 1935).
procedures (see below), suggesting possible early Figure 2 presents examples of facial features elicited by
functional connections with limbic forebrain and periaque- taste stimuli. In Steiner’s laboratory, hundreds of full-term
ductal gray matter (e.g., Pomonis et al., 2000; see Berridge, infants (with Apgar scores ranging from 8 to 10 and weigh-
2000, for review of brain substrates for palatability). ing at least 2.9 kg) have been examined before their first
In addition, a drop of pure lemon juice is commonly hospital postnatal feeding experience. In brief, the face of
found to increase reflex salivation in premature infants the relaxed and reclining (tilted up to about 35°) newborn
(weighing 1.15–1.83 kg), accompanied by apparent infant is videotaped for at least 60 seconds (unstimulated
increases in sucking vigor and sometimes retching condition), and then 0.2–0.5 ml of sterile taste solution or
(Eckstein, 1927). The bitter stimulus quinine (0.0005, distilled water are inserted into the mouth via a disposable
0.005, and 0.05 M) typically retards sucking in such graduated pipette (stimulated condition). Water rinses or
infants. 3- to 5-minute time intervals intercede between stimulus
Individual differences in facial expressions have been presentations. Later, specific facial movements are tabu-
noted in premature infants and appear to be most marked to lated by trained and/or untrained observers in a blind
“salty” stimuli. For example, in one study of 20 premature procedure. The type, degree, frequency, and duration of
infants (1.2–2.9 kg) presented with 0.9% NaCl, more than complexes of facial motor movements are related to the
50% responded with a rejecting grimace, although 4 read- different taste stimuli and their intensity (e.g., Ganchrow et
ily accepted this solution (Martin du Pan, 1955). In another al., 1983; Steiner, 1987).
study (Eckstein, 1927), NaCl (1% solution) produced indif- In brief, sweet and umami stimuli [e.g., 0.1–0.5%
ference in two thirds of the infants, and rejection in the monosodium- or potassium-glutamate (MSG or PG) dis-
other third, during the final trimester of gestation. solved in clear vegetable soup] are frequently associated
Ontogeny of Human Flavor Perception 827

Figure 2 Examples of facial reactions to “sweet” (0.04 M sucrose), “sour” (0.24 M citric acid), “bitter” (0.00007 M quinine sulfate),
and “umami” (0.5% monosodium or potassium glutamate) as compared to “neutral” (distilled water) expressed in term-born neonates
during their first postnatal hours. Single-frame pictures photographed from previously videotaped materials. (Courtesy of J. E. Steiner,
The Hebrew University, Jerusalem.)

with lip licking, (e.g., sweet: Fig. 2, rows 2 and 3; umami: pursing (Fig. 2, rows 2–4) and nose wrinkling. Water, on
Fig. 2, rows 2 and 3), accompanied by rhythmic sucking the other hand, usually elicits only transient alerting and
(often with a distinct smacking sound) (e.g., sweet: Fig. 2, a minimum of mouth movements associated with swal-
rows 1 and 4; umami: Fig. 2, row 5), followed by relaxed lowing, while responses to the unseasoned clear soup
face/smiling (e.g., sweet: Fig. 2, row 5; umami: row 4). diluent are described as slightly to moderately aversive
The constellation of these patterned lip and tongue move- (Steiner, 1987).
ments are sometimes referred to as “mouthing” and used Observers unaware of stimulus parameters regularly
as a reliable measure of sweet or “palatable” in infant stud- interpret these videotaped responses to sweet and umami as
ies (e.g., Barr et al., 1999). hedonically positive (acceptance) and to bitter as hedo-
Bitter stimuli elicit mouth corner depression (Fig. 2, nically negative (rejection) (e.g., Ganchrow et al., 1983;
rows 1 and 2), gaping (Fig. 2, rows 3–5), tight eye closure Steiner, 1987). While sour is generally considered to be a
(Fig. 2, rows 2–5), flat tongue protrusion (Fig. 2, rows 3 somewhat aversive taste, in one study (Rosenstein and
and 4), head turning/shaking, and drooling. Similar but Oster, 1988), 0.12 M citric acid has been categorized as
diminished responses are produced by sour stimuli (e.g., eliciting neutral-to-hedonically positive reactions mirroring
mouth corners down: Fig. 2, row 1; gaping: Fig. 2, row results obtained in newborn rats and rabbits (Ganchrow
5), with the addition of sustained and/or rhythmic lip et al., 1979, 1986). This could be an interpretation error due
828 Ganchrow and Mennella

to an apparent overlap of some sour- and sweet-induced conditions also produce a very transient calming (Graillon
features (e.g., lip pursing/sucking), or an immaturity of et al., 1997), possibly due to attentional factors, and indica-
sour-taste receptor mechanisms, or a lack of learned hedo- tive again of a broadly functional taste system at birth.
nic response associated with these stimuli. This ambiguity A functioning affective component to the newborn’s
was not found in precocial hatchling chicks under similar response to taste stimuli is also suggested by taste-induced
experimental conditions and using the same citric acid autonomic responses (e.g., Ashmead et al., 1980; Blass and
concentrations: Observers classified the chicks’ responses Watt, 1999; Crook and Lipsitt, 1976; Haouari et al., 1995;
to citric acid as hedonically negative (Ganchrow et al., 1990). Herschel et al., 1998; Lipsitt et al., 1976; Lipsitt, 1977; Rao
That facial response is reflex-like is supported both by et al., 1997). Experimental conditions seem to dictate the
its appearance in anencephalics (Steiner, 1973, 1979b) and direction of change. For example, heart rate increased pro-
by the fact that facial motor features elicited by sucrose portionally with the introduction of 5% and 15% sucrose
can be classically conditioned to a tactile stimulus in new- onto ongoing sucking from an artificial nipple in a situation
borns 2–48 hours of age (Blass et al., 1984). In addition, where increased sucking amplitude did not yield larger vol-
single response components such as tongue movements umes. Conversely, heart rate decreased when sweet was
can be reliably elicited in newborns by tastants in a con- introduced during agitation, producing an overall calm (e.g.,
centration-dependent manner (e.g., Nowlis and Kessen Blass and Watt, 1999). Furthermore, Fox and Davidson
1977; Weiffenbach, 1977). Further support for its root (1986) reported a taste-induced asymmetry of brain elec-
reflexive nature comes from studies demonstrating similar trical activity, which is usually associated with hedonically
taste-induced oral motor behaviors in decerebrate and positive emotional reactions, although the response had
intact rats (Grill and Norgren, 1978). not yet achieved a mature form. Specifically, they observed
As in premature infants, newborn infants show individ- left-sided frontal activation (a correlative sign of affectively
ual differences in taste-induced orofacial reactivity positive or approach behavior) in response to sucrose as
(e.g., Bergamasco and Beraldo, 1990; Ganchrow et al., compared with water in 2- to 3-day-old newborns.
1983; Maekawa et al., 1991; Stirnimann, 1935), possibly
reflecting either genetic differences in the predisposition
2. Sucking and Intake
for facial emotional expression or in taste receptor endow-
ment. For example, human taste bud density can vary as Direct measures of intake in the first postnatal days also
much as 100-fold between individuals irrespective of age suggest the neonate has a strong, positive hedonic response
(Miller, 1988), affecting the perceived intensity of a tastant to sweet-tasting stimuli. Sucrose, as well as MSG solutions
and the spectrum of expressive movements that is elicited. (the latter dissolved in clear soup), are consistently
Thus, one would expect that the weaker the stimulus, the ingested in larger volumes compared to bitter, sour, salty,
more the individual variation in response emission. and/or neutral stimuli (e.g., Beauchamp and Moran, 1982;
Prenatal chemosensory experience might also contribute to Beauchamp and Pearson, 1991; Desor et al., 1973; Maller
this variability (see below). and Desor, 1973). In addition, heavier infants tend to con-
Sweet taste solutions of sucrose and aspartame (0.12%) sume greater absolute volumes of sweetened solutions
also reliably elicit hand-mouth contacts perinatally (e.g., than those with less weight (e.g., Nisbett and Gurwitz,
Barr et al., 1994, 1999; Blass et al., 1989; Smith and Blass, 1970; Desor et al., 1973; Kajiura et al., 1992). For umami,
1996; Zeifman et al., 1996), and these could be interpreted the possibility exists that glutamates (e.g., MSG) in clear
as precursors to later feeding-related behaviors. This taste- broths may interact with odor and/or viscosity (Maga and
elicited behavior is stimulus bound to the presence of the Lorenz, 1982), affecting, in turn, intake and/or flavor-
sweet taste and declines during the first 2 weeks postpar- induced facial expressions. To date this issue has not been
tum. behaviorally addressed in infants.
It is noteworthy that ongoing crying is diminished and Newborns can also discriminate among different con-
eliminated for some time after the gentle insertion of centrations of sweet stimuli as determined by sucking rates
droplets (as little as 0.1 ml) of sucrose or aspartame into (e.g., Engen et al., 1978) and measures of the temporal
the oral cavity. The ability of the sweet-perceptive system organization of sucking (e.g., duration of sucking bursts,
to regulate ongoing crying is most apparent perinatally and interburst intervals, and within-burst pace of sucking)
appears to decrease over the first 6 weeks (e.g., Barr et al., (e.g., Crook, 1977; Crook and Lipsitt, 1976; Eckstein,
1994; Smith and Blass, 1996; Zeifman et al., 1996), often 1927). For instance, increasing the sucrose concentration
requiring larger volumes of stimulus to obtain the effect at (0.0612–0.5 M) produces corresponding increases in the
all. Bitter tastants (0.25% quinine) presented under similar number of sucks regardless of birthweight (Grinker et al.,
Ontogeny of Human Flavor Perception 829

1986). However, these measures are not always sensitive Although more detailed investigation is needed, it
to differences among nonsweet substances, in contrast to would appear that, within limits, variations in sweet mole-
facial reaction measures. For example, in one study infants cule type, suprathreshold stimulus concentration, solution
from birth to 6 days of age did not express rejection of volume, and presentation as a flow or on a dipped pacifier
0.12–0.24 M urea compared to the 0.07 M sucrose diluent do not differentially affect the calming- or pain-reducing
on several measures of intake and sucking behavior effects of sweet-tasting stimuli (e.g., Barr et al., 1999;
(Kajiura et al., 1992). This could imply immaturity of cel- Blass and Shah, 1995; Blass and Watt, 1999; Carbajal
lular receptor processing specific to urea. However, the et al., 1999; Ramenghi et al., 1996; see also Haouari et al.,
impact of the sweet diluent at this young age may also 1995). This suggests the possibility of an “on-off” rather
have been important since earlier studies using quinine dis- than a graded relation between the gustatory afferent and
solved in water did report a clear disruption of newborn the presumed opioid-mediated efferent response (Blass
sucking behavior (e.g., Desor et al., 1975b; Eckstein 1927) and Shah, 1995; see also Lieblich et al., 1984; Pomonis et
as well as a tendency for a damped response to sour and al., 2000). Exceptionally, lactose failed to reduce distress
salt stimuli (Crook, 1978; Desor et al., 1975b; Eckstein in the concentration range of 0.17–0.51 M in both humans
1927). Indeed, a significant depression of sucking for (Blass, 1997; Blass and Smith, 1992) and rats (Blass and
0.2–0.4 M salt solutions has been reported (Beauchamp Shide, 1994). However, isocontour matching of sweetness
et al., 1994), consistent with previous data (Maller and across various sugars in adults (Cameron, 1947) would
Desor, 1973) revealing a depressed intake of 0.2 M NaCl require a 0.8 M lactose concentration to match the optimal
compared to water. 0.33 M (12%) sucrose, suggesting the concentration of lac-
In summary, by all measures newborns perceive and tose used during testing was probably below threshold for
respond positively to sweet stimuli. For other tastes, test-
producing the effect.
ing conditions seem especially important. Although under
The analgesic effect appears to be due to taste, per se,
certain circumstances relatively concentrated bitter, sour,
since the same sweet solution is ineffective in producing
and salt stimuli elicit definitive behavioral responses, the
analgesia when administered intragastrically in preterm
diluent, the stimulus and its concentration, and the depen-
(32–36 weeks) infants (Ramenghi et al., 1999). Preterm
dent measure may influence the results. Also, “negative”
infants as young as 25 weeks postconception are able to
tastes experienced may not always be able to override and
express the pain-relieving sweet response (Johnston et
inhibit an ongoing sucking response. Preference (“want-
al., 1997). The calming effect is induced within
ing”), measured as amount ingested, and orofacial hedonic
reactivity (“liking”) are sometimes independent and prob- seconds of sucrose delivery, well in advance of
ably mediated by separate neural mechanisms (see stomach clearance or absorption. Overall, in these peri-
Berridge, 2000, for review), whose emerging expression natal ages, gender, gestational age, and immediate
could be developmentally dependent. postnatal age are noninfluencial factors (e.g., Overgaard
and Knudsen, 1999).
Other flavor and orosensory components are also suffi-
3. Palatability and Pain Measures cient to produce the above effects and may act synergisti-
In addition to initiating neonatal responses supporting con- cally with sucrose. These include milk flavor and some of
summatory behavior, sweet taste sensibility has recently its constituents (fat, protein) (e.g., Barr et al., 1996; Blass,
been shown to diminish pain reactivity, with an accompa- 1997), but not colostrum (E. M. Blass and L. B. Watt-
nying calming effect and decreased heart rate (see above) Miller, personal communication). Orotactile components,
in preterm and term-born infants, extending comparable such as a pacifier (Blass and Hoffmeyer, 1991; Blass and
findings in rats (see Blass and Watt, 1999, for review). Watt, 1999; Carbajal et al., 1999), also produce the anal-
Behavioral measures have included facial expression, gesic effect but are transient compared to sweet stimula-
visual analog scale estimates of distress, incidence and tion (Smith et al., 1990) and are not accompanied by con-
duration of crying, or some combination of these. The sonant significant drops in heart rate (e.g., Blass and Watt,
effect is generally sustained for at least 4 minutes after ces- 1999). It is interesting to note that besides oro-tactile stim-
sation of the gustatory stimulus. Two ml of 12% (0.33 M) ulation (e.g., Rolls et al., 1999), fatty acids may depolarize
sucrose was usually sufficient to consistently produce a taste cells (see Gilbertson, 1998). Taste “salience” has
significant analgesic response (see Stevens et al., 1997, for been suggested to play a role in human infants since
systematic review), although 30–75% sucrose or glucose another lipid, corn oil, did not elicit similar calming effects
are more commonly used in clinical trials. (Graillon et al., 1997).
830 Ganchrow and Mennella

D. The First Year Responses to sodium chloride appear to be substantially

modified during development. Unlike newborns, who
1. Facial Reactions and Sucking mostly appear to be indifferent or averse to the taste of salt,
4-month-old infants begin to prefer saline solutions and
Behavioral responses to taste stimuli shift towards produc-
6-month- to 2-year-old infants have developed a clear pref-
ing more voluntary consummatory consequences in the pre-
erence for salted solutions (0.1–0.2 M) and foods
verbal first year. Spontaneous facial expressions to taste
(Beauchamp et al., 1986, 1994). This could be related to
stimuli remain fairly stable over the first postnatal month
maturation of salt receptor function as reported in other
and then gradually decrease, being replaced by more notice-
mammals (for review, see Hill and Mistretta, 1990) (see
able intentional behaviors such as refusal to open the mouth
also Chapter 36) or central nervous system developmental
or pushing the spoon away with the hands (Lichtenstein,
factors that have not yet been elucidated.
1893/94). Still, comparing infants in the first 6 postnatal
Human infants may be born “hard-wired” to discrim-
days to those 0.5–6 months of age, Kajiura et al. (1992)
inate tastes with survival value (e.g., potentially harmful
demonstrated that the balance of facial expressions and
bitter and sour-tasting stimuli from potentially nutritious
body movements indicative of rejection of urea (0.12–0.24
sweet and umami stimuli) (see Scott, 1992) but
M) shifted away from indifference and towards rejection.
“soft-wired” so that a particular culinary environment can
This was confirmed by a more even decrease in intake in
influence taste (and flavor) preferences essential for
older infants, suggesting a change in urea bitter sensitivity.
culture-specific survival (see, e.g., Beauchamp and
Changes in sucking parameters are less evident early
Mennella, 1998; Gerber, 1991; Hudson and Distel, 1999;
on, and until 1.5 months sucking responses are similar to
Mennella, 1997a). Considering taste alone, Beauchamp
those of younger infants (Eckstein, 1927). Still, Maekawa
and Moran (1982) reported that infants fed sweetened
et al. (1991) argue that a discriminative sucking response
water any time during the first 6 months maintained their
to taste stimuli has a decremental tendency, especially
high sweet preference established at birth, while those not
around 3–5 months, such that 0.25% NaCl and 0.1% tar-
so fed diminished their sweet preference during this time
trate are progressively less likely to induce a sucking reac-
interval. This effect was still apparent 1.5 years later
tion of rejection. Therefore, it becomes useful to discuss
(Beauchamp and Moran, 1984).
facial expression and sucking in conjunction with intake
Likewise, salt intake may be affected by early experi-
under the heading of “preference,” i.e., willingness to con-
ence. Salt preference is sometimes unexpectedly high at
sume a taste substance.
around 16–17 weeks for infants exclusively breastfed until
testing (Harris and Booth, 1987). However, this study did
2. Preference
not take into account possible fetal exposure to electrolyte
Using several different acceptance measures, it has been imbalances and accompanying dehydration. Such events
found that responses to sweet and umami stimuli (the lat- may up regulate later tendencies to prefer higher salt
ter in soup or salt solutions) throughout the first postnatal concentrations. For example, 16-week-old infants of moth-
year ranges from preferred to indifferent, with no age ers who had experienced morning sickness, or vomiting
trends evident after the first week (e.g., Beauchamp and during pregnancy, were more willing to consume 0.1 and
Pearson, 1991; Beauchamp et al., 1984, 1998; Büssem, 0.2 M NaCl (i.e., showed higher preference ratios com-
1895; Lichtenstein, 1893/1894; Vasquez et al., 1982). pared to water) and to ingest larger volumes of 0.2 M
In general, bitter and sour stimuli are rejected at con- NaCl, as compared to control infants whose mothers did
centrations well above threshold (e.g., Lichtenstein not express this symptom (Crystal and Bernstein, 1998).
1893/1894; Neuman, 1896; Vasquez et al., 1982; see also Furthermore, infants of the sick mothers were less likely to
Desor et al., 1975b). Futhermore, when the bitter tastant express aversive facial reactivity patterns, and more likely
was presented in one study as urea dissolved in 0.07 M to exhibit hedonically positive responses, to the salty solu-
sucrose, older (2-week- to 6-month-old) infants tended to tions, especially to 0.2 M NaCl. These data might also sup-
reject all urea concentrations (0.12–0.24 M) (Kajiura et al., port the proposition that preterm infants taste NaCl but
1992). This is in contrast to newborns (see above), whose under most normal testing conditions are indifferent to it.
sucking responses were more controlled by order of stim- This altered salt preference apparently carries through to
ulus presentation than by stimulus taste. These results sug- adulthood, as reflected by (1) greater self-reported use of
gest that for some bitter substances, developmental factors, salt, (2) greater intake of salt in the laboratory, and (3)
presumably at the receptor level, may govern the ability to preferences for salty snack food compared to those whose
actively reject “bitter” presented within a mildly “sweet” mothers had mild or no morning sickness symptoms
context (see also Beauchamp et al., 1991). (Crystal and Bernstein, 1995). The mechanism is not
Ontogeny of Human Flavor Perception 831

understood, but maternal renin-angiotensin involvement in an experiment wherein 4- to 5-year-olds were exposed
and consequential elevated salt intake are no doubt impor- to either salty or sweet or plain tofu tasted many times over
tant elements (e.g., Argüelles et al., 2000; Crystal and several weeks (Sullivan and Birch, 1990). When tested
Bernstein, 1998). later, they preferred the tofu with the taste they were accus-
tomed to, but the taste preference did not generalize to
other foods of similar color/texture (e.g., ricotta) that were
3. Threshold made sweet or salty.
Almost no research has been directed towards establishing
taste thresholds in infants. Osepian (1958) used classical 2. Thresholds and Intensity Scaling
conditioning techniques to establish discrimination thresh-
olds to sweet, salt, and sour in infants 2–9.5 months of age. Individual food aversions could be, in part, determined by
The resulting values were within the normal range of variability in the perception of bitterness by different chil-
detection and recognition thresholds reported for young dren. Five- to 7-year-old children have been tested for
adults (Cowart, 1981), although the study’s methodology their sensitivity to the bitter substance 6-n-propylth-
has been questioned (Beauchamp and Moran, 1986). iouracil (PROP). Unlike most other bitter substances (e.g.,
quinine), the ability to perceive PROP is a genetically
determined trait, with the general population recessive for
4. Palatability and Pain Measures nontasting. “Tasters” among the children had lower
In comparison to perinatal ages, “sucrose analgesia,” thresholds and provided higher intensity ratings across the
measured by reduced crying tendencies, was modest for four PROP concentrations tested. However, there were
2-and 4-month-old infants receiving diptheria-tetanus- proportionally fewer “nontasters” than predicted by adult
pertussis immunizations (Barr et al., 1995). A slight data, suggesting that PROP thresholds rise with age and
improvement (~40% crying reduction) occurred with an may partially account for greater food finickiness often
increased concentration (from 50% to 75% sucrose), an exhibited by younger children (Anliker et al., 1991).
increased volume (from 0.75 to 2 ml), and the addition of
cuddling and pacifier availability (Lewindon et al., 1998). 3. Palatability and Pain
Neither study attempted to stratify results by age. Critical
stimulus parameters need to be better defined for this age In contrast to 2-week-old infants, it appears that experi-
group and older. mental conditions become increasingly important in
obtaining the analgesic effect as maturation progresses.
Toddlers (to 18 months) only showed sucrose (2 ml 12%)
E. Early Childhood analgesia if they were administered one rather than 2 injec-
tions (Allen et al., 1996). This reinforces findings in 2- and
1. Intake and Preference
4- months-old infants (see above).
No age effects were found when solutions of urea (0–0.48
M) and citric acid (0–0.012 M) were tested: 2-year-olds
F. Late Childhood/Adolescence
decrease intake with increasing concentrations, as expected
(Vasquez et al., 1982). Similarly, 2- to 3-year-old children
1. Thresholds
increase the amount of sucrose solution they consume lin-
early with increasing concentration (0–0.4 M); however, Taste thresholds bear little relation to measures of dietary
unlike younger children, they decrease NaCl consumption preference (e.g., Adams and Butterfield, 1979; Nilsson and
(0–0.2 M) relative to water (Beauchamp et al., 1986; Holm, 1983) and, thus, may provide insight into develop-
Vasquez et al., 1982). From 2.5 to 5 years of age, rejection ment of the gustatory system per se. Procedural difficulties
of NaCl dissolved in water is quite pronounced have precluded extensive research on this topic. Some
(Beauchamp et al., 1986). On the other hand, when food studies support a general tendency for lower thresholds in
(soup or cereal) is the vehicle for salt, 2-year-old children 6- to 15-year-olds relative to 16- to 25-year-olds (Adams
prefer salted to unsalted foods (e.g., Beauchamp and and Butterfield, 1979; Catalanotto et al., 1979; Nilsson and
Moran, 1984). Experience and context may play a role Holm, 1983). However, other studies have reported the
here, since salty water would be a rather unusual beverage opposite effect (see Cowart, 1981, for review). Attention
for 2-year-olds (see Beauchamp and Engleman, 1991). span difficulties in younger children and environmental or
The important role experience and context play in ethnicity factors including within-group variability in taste
determining salt and sweet preference is also demonstrated receptor endowment were thought to contribute to these
832 Ganchrow and Mennella

discrepancies (e.g., Okoro et al., 2000). However, a recent 3. Preference

well-controlled study of 8- to 9-year-olds and young adults
Preference for salty foods established by the second year
revealed the importance of sex and age: James et al. (1997)
is maintained throughout early childhood and early ado-
noted that the gustatory system of the prepubertal male
lescence before stabilizing at adult levels (Beauchamp and
subgroup had not reached maturity. This was expressed by
Cowart, 1987). Thus, a comparison of NaCl preferences
significantly higher detection thresholds for sucrose and
among 3- to 6-year-old children, 7- to 10-year-old chil-
sodium chloride than those observed in men and women as
dren, and 18- to -26-year-old young adults (0.0–1.0 M
well as prepubertal girls. Interestingly, the citric acid
concentrations dissolved in vegetable soup) found the
thresholds of the boys were seven times higher, on average,
adults to have diminished preference for the highest
than that of the adults, although caffeine thresholds were
concentration (65, 78, and 13% of these respective groups
similar across subgroups (see Table 1). A longitudinal
maximally preferred the highest concentration), with chil-
study following up on these remarkable results is needed.
dren preferring on average around 0.4 M NaCl, compared
to adults at around 0.2 M (Beauchamp and Cowart, 1987,
2. Intensity and Discrimination Scaling
1990). Similar results have also been obtained with
Early studies of taste intensity perception met with mixed chicken soup as the diluent (Nijjar et al., 1991). Hispanic
results, some suggesting that children are more sensitive preschoolers expressed higher sodium preference than
and some that they are less sensitive to taste intensity their North American non-Hispanic white counterparts,
changes (e.g., Enns et al., 1979; Shapera et al., 1986). Age again suggesting possible cultural/local environmental
is a factor in obtaining reliable results of intensity differ- influences.
ences, as those from children 2–3 years of age are practi- The most comprehensive pioneering study examined
cally random, unlike those of children 6 years and older taste preferences for sucrose (0.075–0.6 M), lactose
(Kimmel et al., 1994). Recently, successful magnitude (0.1–0.4 M), and sodium chloride (0.05–0.40 M) in 618
scaling results have been obtained from children 8 years of 9-to 15-year-olds and 140 adults. This study found that,
age. Direct magnitude scaling (James et al., 1999) of relative to adults, a greater percentage of children preferred
sucrose sweet intensity produced similar functions for 8- the higher concentrations of each stimulus (Desor et al.,
to 9-year-olds and adults if the diluent was water. When 1975a). Within the younger group there were also sex and
the diluent was orange juice, the children’s intensity slopes race differences, namely, males and African Americans
flattened compared to adults, making context an important tended, on average, to prefer higher concentrations. The
issue. Using a 5-point category scale to assess sweet inten- age-related decrease in most-preferred sucrose concentra-
sity, comparisons between older children, adolescents, and tion is a consistent finding and has been recently suggested
young adults yielded borderline (p
0.08) significant to be related to the higher energy requirements of younger
differences for the age factor. When orangeade was the children (De Graaf and Zandstra, 1999), providing them
diluent, no significant age effect was observed, but the age- with a sensory signal for energy content. A follow-up lon-
concentration interaction was significant, with younger gitudinal study of a subsample of the younger group tested
children providing lower intensity estimates (DeGraaf and at 19–25 years of age in the Desor et al. (1975a) study con-
Zandstra, 1999). The impression is that sweet taste inten- firmed that preference declines within individuals during
sity perception in children is quite stabilized, but testing maturation and that the results are not simply due to a gen-
parameters may affect how they it is expressed. Other taste eration-related change in preference behaviors (Desor and
stimuli need to be examined. Beauchamp, 1987).

Table 1 Taste Detection Thresholds of Adults and Children

Adult Children
Tastant Female Male Female Male
Sucrose 0.0062a 0.0017b 0.0068  0.0010 0.0072  0.0026 0.0170  0.0033
Sodium chloride 0.0023  0.0005 0.0034  0.0005 0.0027  0.0005 0.0061  0.0008
Citric acid 0.0002  0.0001 0.0002  0.0003 0.0005  0.0001 0.0014  0.0005
Caffeine 0.0008  0.0002 0.0018  0.0004 0.0011  0.0003 0.0020  0.0008
aExpressed as molar concentration.
bStandard error.
Source: From James et al., 1997.
Ontogeny of Human Flavor Perception 833

Developmental differences in taste preferences may be IV. ONTOGENY OF OLFACTORY

colored by several influences, including (1) experience, BEHAVIORAL RESPONSES AS
(2) hormonal effects, (3) differences in physiological need RELATED TO FEEDING
within an environment, (4) expression of genetic
programs, or some combination of these factors. For exam- A. Fetus
ple, hormonal influences may contribute to the wide
variations in salt preferences expressed across individuals Taste notwithstanding, the human fetus can respond to
over the life span such that, for instance, pregnant a wide variety of other sensory stimuli that occur naturally
teenagers express increased avidity for salty snack foods in its environment. For example, the fetus responds to
during, but not after, pregnancy (Skinner et al., 1998). extrauterine auditory stimuli, as indicated by changes in
Sugar content did not differentially affect preference heart rate (Fifer and Moon, 1988), and will blink when
during those conditiions. a bright light source is applied to the mother’s lower
Long-term effects of physiological challenges experi- abdomen (Birnholtz, 1988). Furthermore, prenatal sensory
enced early in development have been reported. Prenatal experiences can subsequently affect the behavior of the
mineral or fluid loss resulting from maternal vomiting, newborn. This is best illustrated by the finding that human
and/or electrolyte imbalances associated with excessive newborns exhibit a preference for a specific voice,
infantile diarrhea and vomiting, increased adolescent avid- melody, or passage experienced prenatally (De Casper and
ity for salt, but not for sugar, corroborating and extending Spence, 1986; Hepper, 1998a) and the fact that they can
similar previous findings (Crystal and Bernstein, 1995, discriminate between human voices (De Casper and Fifer,
1998; Leshem, 1998; see also Leshem et al., 1998). 1980). Such findings can be extended to the sense of
Furthermore, adolescents who were exposed during smell—newborns can discriminate and will prefer their
infancy to a chloride-deficient feeding formula favored own amniotic fluid for at least 2 days after birth (Marlier
highly salted foods and expressed a relative dislike for et al., 1998; Schaal et al., 1995a, 1998). This preference for
foods classified as being lower in saltiness as compared to amniotic fluid odors shifts to olfactory cues emanating
their nonexposed siblings. Sugar preferences were similar from mother’s milk shortly thereafter (Marlier et al., 1998).
in the two groups (Stein et al., 1996). Still, preterm infants Learning about the dietary choices of the mother may
undergoing neonatal furosemide (diuretic, natriuretic, and also be occurring prenatally since the environment in
chloriuretic) therapy did not express unusual salt prefer- which the fetus lives, the amnion, can indeed be odorous.
ences in a laboratory setting, although increased dietary Its odor not only can indicate certain disease states (Mace
salt intake was implied by higher sodium excretion in a et al., 1976), but can reflect the flavors of foods eaten by
subsample (Leshem et al., 1998). It has been pointed out the pregnant mother (Hauser et al., 1985; Mennella et al.,
that a combination of sensory measures is probably 1995). This has been demonstrated experimentally in a
required to properly assess salt preference, since different study in which amniotic fluid samples were obtained from
individuals may have different ways of expressing this 10 pregnant women undergoing routine amniocentesis
preference (Stein et al., 1996), not to mention other envi- who ingested either garlic or placebo capsules approxi-
ronmental factors that may influence an individual’s salt mately 45 minutes before the procedure (Mennella et al.,
habits. Still, taken together, these findings suggest that 1995). The odor of the amniotic fluid obtained from the
perinatal experience may have long-lasting effects on salt women who ingested the garlic, as determined by adult
appetite. It is not known how much taste receptor level human evaluators, was judged to be stronger or more like
alterations or organizational central nervous changes dur- garlic than amniotic fluid from the control women who did
ing this period of plastic maturational events may have not consume garlic.
contributed to these outcomes. Animal studies have revealed that flavor experiences in
utero can result in preferences postnatally (for reviews, see
Bilkó et al., 1994; Schaal and Orgeur, 1992; Smotherman
and Robinson, 1988) (see also Chapter 15). For example,
4. Palatability and Pain Measures rat pups born by Caesarean section preferred their
Sweet stimuli may continue to suppress pain in prepuber- mother’s fluid to that of an unrelated rat when tested
tal children (Miller, 1994), as previously described for immediately after birth, indicating a prenatal acquisition of
infants. Children in the 8- to 11-year-old range expressed this odor preference (Hepper, 1987). Moreover, the fetus
prolongation of tolerance threshold times (cold pressor can acquire information about the dietary choices of the
test) while holding 24% sucrose in their mouths as com- mother. For example, offspring of mother rabbits who had
pared to water. eaten garlic or juniper during pregnancy exhibited a
834 Ganchrow and Mennella

preference for these flavors when compared to offspring of B. Infancy

mothers who were not so exposed (Hepper, 1988b; Bilkó
et al., 1994). Such findings have recently been extended to Shortly after birth, human infants can detect a wide variety
human infants (Mennella et al., 2001). Weanling-aged of volatile chemosensory stimuli, and they appear to be as
infants, who had several weeks of exposure to the flavor of sensitive to odors as are adults, if not more so. Newborns
carrots in either amniotic fluid or mothers’ milk, consumed can detect and discriminate among qualitatively distinct
significantly more carrot-flavored cereal and were per- odorants as evidenced by changes in their facial responses
ceived by their mothers as enjoying the carrot-flavored (see Fig. 3), body movements, and heart and respiratory
cereal more when compared to plain cereal. In contrast, rates (for review see Engen, 1982; Mennella and
control infants, whose mothers drank water during preg- Beauchamp, 1993a, 1997a, 1999; Rovee, 1972;
nancy and lactation, exhibited no such preference. These Soussignan et al., 1997; Steiner, 1977, 1979a,b). Perhaps
findings are the first experimental demonstration that pre- the most salient odors for the newborn are those emanating
natal and early postnatal exposure to a flavor enhances the from the mother. Within hours after birth, mothers and
acceptance and enjoyment of that flavor during weaning. infants can recognize each other through the sense of smell
As will be discussed in the next section, these very early alone (Cernoch and Porter, 1985; Macfarlane, 1975;
flavor experiences may provide the foundation for cultural Schaal, 1986, 1988) (for further discussion, see Chapter 15).
and ethnic differences in cuisine. Such recognition, with little postpartum contact, may be

Figure 3 Examples of facial reac-

tions to food-related odors (BA.,
banana; VA., vanilla; FI., fish; BU.,
butter; R. E., rotten egg) as compared
to the odorless control (C.) expressed
in term-born infants in their first post-
natal hours. (From Steiner, 1977.)
Ontogeny of Human Flavor Perception 835

due, in part, to mothers learning about the olfactory iden- More recent studies have systematically explored
tity of their infants via changes in their own body odor dur- whether flavors from the woman’s diet are transmitted to
ing pregnancy, a change that reflects the odor type of the her milk and what effects, if any, this has on the behavior
child (Beauchamp et al., 1995) (see also Chapter 17). of her breast-fed infant. Two approaches have been used to
Maternal odors have both a calming and a feeding determine whether the odor, and consequently the flavor,
preparatory effect on the infant. This is demonstrated by of human milk is distinctively altered following the inges-
the findings that newborns spend more time orienting tion of different foods and beverages. First, it was deter-
toward a breast pad previously worn by their lactating mined whether a sensory panel of adults could detect
mothers than toward one worn by an unfamiliar lactating a change in the odor of human milk as a function of mater-
woman (Schaal, 1986, 1988), and they move their head nal ingestion of flavors. Milk samples obtained from each
and arms less, mouth more, and cry less during exposure to woman at fixed intervals before and after the ingestion of
their mothers’ body odors (Schaal, 1986; Sullivan and the flavor or placebo were evaluated by a sensory panel of
Toubas, 1998). Moreover, shortly after birth infants prefer adults within hours of expression (Mennella and
their mothers’ breast unwashed as compared to when it has Beauchamp, 1991a,b, 1993b, 1996b, 1998a, 1999).
been thoroughly washed and is therefore less odorous Second, mothers “tasted” their own expressed milk sam-
(Varendi et al., 1994). Thus, like other mammalian young, ples to determine whether they could detect a difference in
the recognition of and preference for maternal odors may the flavor of their milk following ingestion of the flavor
play a role in guiding the infant to the nipple area and facil- (Mennella, 1999b; Mennella and Beauchamp, 1999). In
itating early nipple attachment and breast-feeding. each case mothers were asked to maintain a bland diet and
avoid eating foods that were flavored with the flavor of
interest during the 3 days preceding each milk collection
1. Retronasal Perception of Flavors
session to ensure that the milk was devoid of these
a. Milk Feeding. Before the development of infant for- volatiles.
mulas, postpartum flavor experiences until the time of These methods revealed that human milk is rich in
weaning were largely restricted to those obtained through flavors and directly reflect the foods and spices eaten
mothers’ milk or the milk of another woman. (e.g., garlic, mint, alcohol, carrot, vanilla) or substances
Considerable research indicates that the milk of mam- inhaled (e.g., cigarette) by the mother (Mennella and
mals, including humans, is rich in flavors that directly Beauchamp, 1991a, 1993b, 1996, 1998, 1999). For
reflect the foods and spices eaten by the mother (see example, maternal alcohol consumption significantly and
Mennella, 1995, for review). The retronasal route of olfac- consistently increased the perceived intensity of the milk
tion may also be particularly salient for infants, allowing odor; this increase in odor intensity peaked in strength
them to experience the many odors present in their 0.5–1 hour after ingestion and decreased thereafter, par-
mother’s milk. The following section focuses on human alleling the changing concentrations of ethanol in
milk as a medium for early sensory experiences for the the milk (see Fig. 4). There was no perceived change in
infant and how these sensory experiences impact on later the odor of the milk on the days the mothers ingested the
preferences. control beverage.
In a study of the flavor properties of human milk As seen in Table 2, infants can detect these flavors in
(Barker, 1980; McDaniel, 1980), milk samples were milk as evidenced by changes in their patterning and dura-
collected from 24 lactating women during the morning tion of feeding and suckling (for review see Mennella and
hours on 3 consecutive days. Within 3 hours of expression, Beauchamp, 1997a). For the majority of flavors, they will
a trained sensory panel evaluated the milk samples for respond by suckling more at the breast or bottle, especially
taste quality and for textural properties such as viscosity if they have not experienced that flavor in the recent past.
and mouth-coating. Each of these sensory attributes varied One exception to this rule is the breast-feeding response
from mother to mother, with the primary taste quality of following maternal ingestion of alcohol (Mennella and
the milk being its sweetness. The sensory panelists also Beauchamp, 1991b, 1993b). Infants breast-fed less during
reported that off-flavors were present in some of the sam- the initial hours when the milk was flavored with alcohol.
ples. Of particular interest was the finding that panelists However, further studies revealed that this diminished
used the verbal descriptors hot, spicy, and peppery to intake is due, in part, to a direct effect of alcohol on milk
describe the milk of one woman who had consumed a production by the mother, indicating that the infants are
“spicy” meal during the test period. Thus, the types and not rejecting the flavor of alcohol in their mothers’ milk
intensity of flavors experienced by each infant in their (Mennella, 1998). When the infants were observed drink-
mother’s milk may be unique. ing their mothers’ milk from a bottle, no such decrease was
836 Ganchrow and Mennella

observed for other flavors, more experience with alcohol,

as assessed by maternal frequency of drinking, was associ-
ated with a diminished suckling response to alcohol-
flavored milk when feeding from a bottle. These findings
indicate that infants can readily detect the flavor of alcohol
in mother’s milk and experience will modify how they
respond to it in subsequent feedings.
That experience with a flavor in mothers’ milk will
modify the infants’ response to that particular flavor has
also been demonstrated (Mennella and Beauchamp,
1993c). For example, the infants of mothers who had
repeatedly consumed garlic capsules during the week
before testing, breast-fed for similar periods of time during
the 4-hour test session in which their mothers’ consumed
garlic, as compared with the session in which their moth-
ers ingested the placebo capsules. In contrast, the infants
Figure 4 Ethanol content of milk samples obtained at baseline who had no or minimal exposure to garlic volatiles in their
and at 30 minutes and 1, 2, and 3 hours after the ingestion of an mother’s milk during the previous week spent more time
alcoholic or nonalcoholic beverage (open circles) and percentage breast-feeding when their mothers ingested garlic com-
of time panelists chose milk samples as smelling “more like alco- pared to when their mothers ingested the placebo
hol” or “stronger” (closed circles). According to a forced-choice
(Mennella and Beauchamp, 1991a). In other words, the
paradigm, panelists were presented individually with pairs of
infants’ suckling response to a particular flavor in mothers’
milk samples and asked to indicate which of the pair smelled
“more like alcohol” or “stronger.” A value of 50% would be milk or formula depends not only on the flavor but the
expected if there were no difference in the odor of the samples, recency of exposure (Mennella and Beauchamp, 1993c,
and hence the panelists responded at random. (Data represent 1996b). Perhaps the garlic flavor became monotonous to
mean  SEM). (From Mennella and Beauchamp, 1991b.) infants who were repeatedly exposed to it in mother’s
milk. Over the short term, children (Birch and Deysher,
1986) and adults (Rolls et al., 1982) report that the palata-
observed (Mennella, 1997b). Rather, the infants consumed bility of a food, and the amount of it consumed, declines
significantly more and sucked more frequently when following repeated consumption of that food, whereas less
drinking the alcohol-flavored milk as compared with the recently consumed foods are considered more palatable
unaltered milk. That experience with the flavor of alcohol and stimulate food intake. Moreover, the garlic-flavored
in mother’s milk modified the infants’ responses to alcohol milk may have aroused the infants who were exposed to a
flavor is suggested by the relationship between the diet of mother’s milk relatively low in flavor, garlic-like
reported frequency of mothers’ drinking during lactation compounds, or both.
and the infants’ rhythm and frequency of sucking when When studying how a change in the flavor of mother’s
feeding the alcohol-flavored milk. Consistent with that milk affects the behavior of the infant at the breast, it is dif-

Table 2 Infants’ Suckling and Feeding Response to Flavored Breast Milk or Formula as a Function of Previous Sensory Experiencesa
Condition prior Length of feeding/
Flavor Medium to testing suckling response Milk intake Ref.
Garlic Human milk Bland diet Increase Increase? Mennella and
Beauchamp, 1991a
Garlic diet No effect No effect Mennella and
Beauchamp, 1993c
Alcohol Human milk Bland diet Increase Increase Mennella, 1997b
Vanilla Human milk Bland diet Increase Increase Mennella and
Beauchamp, 1996b
Formula Bland diet Increase No effect
Formula Vanilla formula No effect No effect
aSee text for further details.
Ontogeny of Human Flavor Perception 837

ficult to separate the direct effects on the infant from other have been shown to be reinforcers for early learning.
possible influences the consumed flavors could have on the Moreover, a substantial body of animal research indicates
mother (e.g., changes in the odor of the mother’s breath or that weanling animals actively seek and prefer the flavor of
sweat) (see Mennella and Beauchamp, 1991a, for further the diet eaten by the mother during nursing (for review, see
discussion). Consequently, one cannot unequivocally Mennella, 1995). In summary, exposure to flavors in
conclude that the flavor change in the mother’s milk was mother’s milk may be one of several ways in which the
responsible for the alteration of the infants’ suckling mother teaches her young what foods are “safe” (Rozin,
behavior. To examine the effects of flavoring directly, 1976). Consequently, young animals may tend to choose a
vanilla was added to the infants’ formula. Consistent with diet similar to that of the mother’s when faced with their
what was found for the breast-fed infant, the bottle-fed first solid meal.
infants’ response to the vanilla-flavored formula was
altered relative to their response to the unflavored formula. b. Solid Foods. For centuries the custom of many
In the first, short-term preference test, the infants sucked cultures has been to introduce grain products, such as pre-
more vigorously when feeding the vanilla-flavored for- cooked cereal, as the infant’s first solid food (Hervada and
mula, and in the second test, which encompassed an entire Newman, 1992; Raphael, 1982). The cereal was some-
feeding, they spent more time feeding on the first bottle times prepared with mother’s milk because many believed
when it contained the vanilla flavor (Mennella and that it was essential that the food be simple, smooth in tex-
Beauchamp, 1996). Following 2 days of exposure to this ture, and consistent in flavor (Fildes, 1986; Wachenheim,
flavor, however, their suckling response to the flavor 1915). In present times, baby food manufacturers and child
diminished. care manuals advise parents to prepare the cereal with
Why do infants respond to flavored milk by enhanced water or either mother’s milk or formula, depending on the
suckling? As mentioned previously, in the studies demon- feeding regimen of the infant. Still, little was known about
strating enhanced suckling response to a flavor in breast- the infants’ acceptance of differently flavored cereals.
fed infants (e.g., vanilla, garlic), the nursing mothers were Thus, a study was designed to determine (1) whether
asked to eat bland diets devoid of these flavors during the breast-fed infants’ acceptance of their first solid food—
3 days preceding each testing day (Mennella and cereal—is enhanced when it is flavored with mothers’ milk
Beauchamp, 1993c, 1996b). In the study with formula-fed and (2) whether the infants’ willingness to accept such fla-
infants, their mothers clearly were providing a very monot- vored cereal is correlated with their mothers’ reported will-
onous diet since formula is virtually unchanging ingness to try novel foods and flavors (Mennella and
(Mennella and Beauchamp, 1996b). Perhaps the novelty Beauchamp, 1997b).
was sufficient to induce increased suckling because these The study revealed that breast-fed infants who had been
flavors are inherently positive or arousing either as a result fed cereal for approximately 2 weeks but had only experi-
of prior exposure (an unlikely explanation for the formula- enced cereal prepared with water readily accepted and pre-
fed infant, although in utero experience is possible) or, like ferred the cereal when it was prepared with mothers’ milk
sweet taste, a hard-wired (innate) response. When aroused, (Mennella and Beauchamp, 1997). They not only con-
mammalian infants will suck more (Bridger, 1962) and sumed more of the cereal flavored with mothers’ milk, they
exhibit a variety of other oral behaviors (Korner et al., displayed a series of behaviors signaling their preferences.
1968; Terry and Johanson, 1987). In addition, the more varied the mother’s diet, the greater
These data, along with those reported previously the acceptance of the flavored cereal. Because an infant’s
(Mennella and Beauchamp, 1991a, b, 1993b, c, 1996b), first flavor experience may occur even before birth in
demonstrate that infants can detect flavors either added to amniotic fluid, breast milk may bridge the experiences of
formula or transmitted to human milk from the mother’s flavors in utero to those in solid foods. It is also possible
diet. The experience with a flavor in either medium altered that experience with a variety of flavors (e.g., in amniotic
how they responded to that flavor in subsequent feedings. fluid, the mother’s milk) may predispose the infant to be
The retronasal perception of odors in mothers’ milk may more willing to accept novel flavors during the weaning
provide the infant with the potential for a rich source of process. As discussed earlier, animal studies support this
varying chemosensory experiences and a possible route for hypothesis, and an additional study found that the more
the development of preference for a diet similar to that of varied the mother’s diet, the more likely the infant would
the mother’s. This is due in part to the fact that the context accept a novel-flavored formula (Mennella and
in which the flavor is experienced, with the mother and Beauchamp, 1996c).
during feeding, consists of a variety of elements (e.g., tac- Because human milk is rich in flavors that directly
tile stimulation, warmth, milk, mother’s voice), which reflect the dietary choices of the mother, the transition
838 Ganchrow and Mennella

from a diet exclusively of human milk to a mixed diet may C. Young Children
be facilitated by providing the infant with bridges of famil-
iarity such that the infant experiences a commonality of Food preference predispositions are susceptible to condi-
flavors in the two feeding situations. In this case, human tioning influences throughout childhood (for review, see
milk would be familiar and would facilitate the introduc- Birch, 1999). For example, conditioned flavor aversions
tion of a novel food, cereal. Animal model studies have have been demonstrated in 73% of children ranging in age
documented this phenomenon (Bilkó et al. 1994; Galef and from 2 to 15 years undergoing chemotherapy (Bernstein,
Clark, 1972), and, in fact, it has been exploited in the 1978), and comparable shifts in hedonic assessment of
development of weaning foods for pigs (Campbell, 1976). nausea-associated flavors have been observed throughout
Because breast-fed infants’ diet is more rich and varied adulthood (e.g., Schwartz et al., 1996). Human taste
than that of infants fed a monotony of flavors in standard conditioning alone has not been investigated in the devel-
formulas, Mennella and Beauchamp (1991a) hypothesized opmental context, although research suggests that both
that breast-fed infants should be more accepting of novel taste and smell may independently contribute to flavor
foods. This was supported by a recent study of 4- to 6- aversion (e.g., Kiefer et al., 1982; see Bernstein, 1999, for
month-old infants’ acceptance of novel food (e.g., green review). The taste-smell combination is sufficient to
beans, peas) both before and after 10 daily feedings of the support conditioned aversions even when pairing with an
food (Sullivan and Birch, 1994). The infants’ reaction to aversive stimulus occurs in utero in rats who are then
and intake of the food increased significantly after the tested more than 2 weeks postnatally (Stickrod et al., 1982).
exposure period; similar findings have been reported for The few published studies that have focused on olfac-
preschool children (Sullivan and Birch, 1990). Of particu- tory preferences in the verbal child indicate that they
lar interest, however, was the finding that the breast-fed indeed have likes and dislikes relating to a range of odors
infants’ acceptance of the vegetable was greater after the and flavors, but that their hedonic experience may be dif-
exposure period when compared to those who were ferent from that of adults. For example, children less that 5
formula fed. The breast-fed infants’ enhanced acceptance years of age appear to respond positively to some odors,
of the vegetables could be due to the breast-fed infants’ such as the odor of synthetic sweat and feces. This led
familiarity with specific flavors associated with the veg- investigators to conclude that young children do not have
etables in mothers’ milk, because they had much more aversions to certain odors that adults and older children
experience with flavor variety than did the formula-fed find offensive and are generally more tolerant of odors
infants, or both. than are adults (e.g., Engen, 1982).
As discussed earlier, amniotic fluid is a potential flavor In contrast to those studies, androstenone, a component
carrier. Because amniotic fluid and mother’s milk are of sweat, is rated as a “bad” odor by 92% of children
under the common priming of maternal diets, their aro- (Schmidt and Beauachamp, 1988) and 59% of adults.
matic profiles may overlap and serve as a “thread of chem- Although androstenone is not a flavor, it is of special inter-
ical continuity between the pre- and postnatal niches” est because it can be perceived as smelling urinous,
(Schaal and Orgeur, 1992). Indeed, recent findings clearly sweaty, musky, like sandalwood, or having no smell; twin
show that experience with the flavor of carrots, either in studies revealed that genetic differences between individu-
amniotic fluid or mothers’ milk, results in preference for als contribute to these differences in the perception of
the flavor of carrots during the time of weaning (Mennella androstenone (Wysocki and Beauchamp, 1984).
et al., 2001). Whether the apparent redundancy in exposure Approximately half of the adult population has a specific
to flavors in amniotic fluid and mother’s milk ensures that anosmia to it. (Specific anosmia is defined as the inability
the young animal can acquire preferences for a variety of of an individual, who otherwise has normal olfactory func-
foods eaten by the mother during different stages of devel- tioning, to smell a specific molecule or related class of
opment is clearly an area in which further research is molecules.) The fact that such a high percentage of chil-
needed in humans (see Bilkó et al., 1994; Capretta et al., dren rated this odor as bad suggested that most or perhaps
1975). These questions have ramifications not only for all 3-year-old children can detect it. A developmental shift
understanding the development of food habits, but also for in the perception of androstenone occurs at or near adoles-
understanding such nutritional concerns as excessive salt cence, when approximately 50% of individuals lose the
and fat consumption, which may lead to hypertension and ability to smell this odor. The mechanisms underlying this
obesity, respectively, and the effects of early exposure to change in perception remain unknown. Recent evidence
drugs such as alcohol and tobacco on later preferences. suggests that children’s thresholds are comparable to
Ontogeny of Human Flavor Perception 839

young adults for a number of other odorants, although such as nausea. Indeed, initial attempts to reverse sweet
children aged 4–11 were less accurate at odor identifica- preference in rats following taste deprivation during the
tion, and their semantic memory for odors was less well suckling period met with failure (Bernstein et al., 1986).
developed (Lehrner et al., 1999). This procedure is known to profoundly alter the normal
Some of the discrepancies as to when olfactory prefer- presynaptic development in the gustatory nucleus of the
ences and aversions arise can be traced to methodological solitary tract in this same species (Lasiter and Diaz, 1992).
and technical difficulties in testing young children. For Further research is required to sort out the critically impor-
example, children younger than 6 years of age tend to tant central nervous system factors acting to control flavor
answer a positively phrased question in the affirmative, perception and behavior during development.
and they have a shorter attention span than older children What qualifies as palatable is susceptible to change
(Engen, 1974; see also Richman et al., 1995). Thus, their with experience. For example, it has been demonstrated
responses may be biased and not reflect their actual reac- that salt preference in infants may change with age and that
tion to the odor. By using methodologies sensitive to these it can be influenced by prenatal maternal fluid/salt imbal-
behavioral limitations and embedding the task in the con- ances. By adulthood, the impact of early experience and
text of a game, research has demonstrated that olfactory childhood food experiences, as well as current physiologi-
preferences and aversions are evidenced in children as cal state, may impact on salt preferences. This is even more
young as 3 years (Schmidt and Beauchamp, 1988). apparent when flavor is considered. Although much
Like adults, children preferred the odor of C-16 aldehyde research is still needed to fully understand the impact of
(strawberry), phenylethyl-methylethyl carbinol (floral), early flavor experiences of the human infant, it is clear that
L-carvone (spearmint) and methyl salicylate (wintergreen), they are not passive recipients of flavored foods. Rather,
but disliked the odor of butyric acid (strong cheese/vomit) based on genetic predisposition and experience, they will
and pyridine (spoiled milk). avidly accept some flavors while decidedly rejecting
others. It is intriguing to begin to unravel how exposure to
flavors, most often experienced in amniotic fluid, mothers’
V. CONCLUSIONS milk, or formula, might affect later preferences, the devel-
opment of food habits, and the willingness to accept new
The sensory world of a fetus or infant is, no doubt, different foods at weaning or thereafter. It appears that early sensory
from that of a child or an adult. Moreover, the sensory experiences may be particularly important in human devel-
world of the newborn infant is different than that of a opment and the advent of formula feeding may not only
6-month-old infant. In adults, input from taste, retronasal deprive infants of important immunological and perhaps
olfaction, and oral irritation appears to fuse into one sensory psychological benefits (see Goldman et al., 1987, for
gestalt—that of flavor. It is not known whether the same is review), but may also limit their exposure to an important
true for the infant. Certainly the balance of these three sys- source of information and education about the flavor world
tems may be very different with relative insensitivity to of their mother, family, and culture (Mennella, 1995;
some taste components (e.g., salt, some bitter compounds) Mennella and Beauchamp, 1991, 1993a).
combined with adult-like or perhaps even heightened sensi- Many questions remain to be answered. For example,
tivity to others (e.g., odors, perhaps irritants). are there sensitive periods during early development when
Early gustatory and olfactory receptor development experiences with flavors produce particularly enduring
prepares even premature infants to respond appropriately effects? How do these intersect with possible influences of
to elementary chemosensory properties associated with the mothers’ physiological state? The scientific study of
nutrients essential for survival. Some spontaneous behav- sensitive periods during the development of behavior,
iors and preferences associated with taste or smell drop out made popular by the ethologist Konrad Lorenz (1965),
during development, while others endure. For example, the implies that there is a period during early development
pain-protective function of palatable (e.g., sweet) tastes, when the organism is primed to receive and perhaps per-
apparent in early infancy, endures into adulthood under manently encode important environmental information. An
certain conditions (Mercer and Holder, 1997; Zmarzty early knowledge of what are safe, appropriate, and nutri-
et al., 1997), while some reflex-like motor patterns dimin- tious foods would intuitively be important information for
ish or vanish. Sweet may be a uniquely potent stimulus in the fetus, infant, and young child. This is not to say that
its general durability as a preferred tastant unless repeat- later learning is not important, but it highlights the possible
edly associated with a negative consequence of intake, significance of these very early experiences.
840 Ganchrow and Mennella

ACKNOWLEDGMENT Beauchamp, G. K., and Mennella, J. A. (1998). Sensitive periods

in the development of human flavor perception and prefer-
This work was supported in part by Grant HD 37119 from ence. Annales Nestlé. Raven Press, New York, pp. 19–31.
the National Institute of Child Health and Development. Beauchamp, G. K., and Moran, M. (1982). Dietary experience and
sweet taste preference in human infants. Appetite 3:139–152.
Beauchamp, G. K., and Moran, M. (1984). Acceptance of sweet
REFERENCES and salty tastes in 2-year-old children. Appetite 5:291–305.
Beauchamp, G. K., and Moran, M. (1986). Taste in young chil-
Adams, D., and Butterfield, N. J. (1979). Taste thresholds and dren. In Clinical Measurement of Taste and Smell, H. L.
caries experience. J. Dent. 7:208–211. Meiselman and R. S. Rivlin (Eds.). Macmillan Press, New
Allen, K. D., White, D. D., and Walburn, J. N. (1996). Sucrose as York, pp. 305–315.
an analgesic agent for infants during immunization injections. Beauchamp, G. K., and Pearson, P. (1991). Human development
Arch. Pediatr. Adolesc. Med. 150:270–274. and umami taste. Physiol. Behav. 49:1009–1012.
Altbäcker, V., Hudson, R., and Bilkó, A. (1995). Rabbit mothers’ Beauchamp, G. K., Cowart, B. J., and Moran, M. (1986).
diet influences pups’ later food choice. Ethology 99:107–116. Developmental changes in salt acceptability in human infants.
Anliker, J. A., Bartoshuk, L., Ferris, A. M., and Hooks, L. D. (1991). Dev. Psychobiol. 19:17–25.
Children’s food preferences and genetic sensitivity to the bitter Beauchamp, G. K., Cowart, B. J., and Schmidt, H. J. (1991).
taste of 6-n-propylthiouracil (PROP). Am. J. Clin. Nutr. Development of chemosensory sensitivity. In Smell and Taste
54:316–320. in Health and Disease, T. V. Getchell, R. L. Doty, L. M.
Argüelles, J., Brime, J. I., López-Sela, P., Perillán, C., and Bartoshuk, and J. B. Snow (Eds.). Raven Press, New York, pp.
Vijande, M. (2000). Adult offspring long-term effects of high 405–416.
salt and water intake during pregnancy. Hormones Behav. Beauchamp, G. K., Cowart, B. J., Mennella, J. A., and Marsh, R.
37:145–162. R. (1994). Infant salt taste: Developmental, methodological
Arvidson, K. (1979). Location and variation in number of taste and contextual factors. Dev. Psychobiol. 27:353–365.
buds in human fungiform papillae. Scand. J. Dent. Res. Beauchamp, G. K., Katahira, K., Yamazaki, K., Mennella, J. A.,
87:435–442. Bard, J., and Boyse, E. A. (1995). Evidence suggesting
Arvidson, K., and Friberg, U. (1980). Human taste: response and that the odortypes of pregnant women are a compound of
taste bud number in fungiform papillae. Science 209: 807–808. maternal and fetal odortypes. Proc. Nat. Acad. Sci. USA
Ashmead, D. H., Reilly, B. M., and Lipsitt, L. P. (1980). Neonates 92:2617–2621.
sucking rhythm and sucking amplitude as a function of sweet Beauchamp, G. K., Bachmanov, A., and Stein, L. J. (1998).
taste. J. Exp. Psychol. 29:264–281. Development and genetics of glutamate taste preference. Ann.
Barker, E. (1980). Sensory evaluation of human milk. Master’s NY Acad. Sci. 855:412–416.
dissertation, University of Manitoba, Winnipeg, Canada. Bergamasco, N. H. P., and Beraldo, K. E. A. (1990). Facial
Barr, R. G., Quek, V., Cousineau, D., Oberlander, T. F., Brian, J. A., expressions of neonate infants in response to gustatory stim-
and Young, S. N. (1994). Effects of intraoral sucrose on crying, uli. Braz. J. Med. Biol. Res. 23:245–249.
mouthing and hand-mouth contact in newborn and six-week Bernstein, I. L. (1978). Learned taste aversions in children
old infants. Dev. Med. Child Neurol. 36:608–618. receiving chemotherapy. Science 200:1302–1313.
Barr, R. G., Young, S. N., Wright, J. H., Cassidy, K. L., Bernstein, I. L. (1999). Taste aversion learning: a contemporary
Hendricks, L., Bedard, Y., Yaremko, J., Leduc, D., and perspective. Nutrition 15:229–234.
Treherne, S. (1995). “Sucrose analgesia” and diptheria- Bernstein, I. L., Fenner, D. P., and Diaz, J. (1986). Influence of
tetanus-pertussis immunizations at 2 and 4 months. J. Dev. taste stimulation during the suckling period on adult taste
Behav. Pediatr. 16:220–225. preference in rats. Physiol. Behav. 36:913–919.
Barr, R. G., Young, S. N., Alkawaf, R., and Wertheim, L. (1996). Berridge, K. C. (2000). Measuring hedonic impact in animals and
Does mature hindmilk calm crying infants? Pediatr. Res. infants: microstructure of affective taste reactivity patterns.
39:16A. Neurosci. Biobehav. Rev. 24:173–198.
Barr, R. G., Pantel, M. S., Young, S. N., Wright, J. H., Hendricks, L. Bilkó, A., Altbäcker, V., and Hudson, R. (1994). Transmission of
A., and Gravel, R. (1999). The response of crying newborns to food preference in the rabbit: the means of information transfer.
sucrose: Is it a “sweetness” effect? Physiol. Behav. 66:409–417. Physiol. Behav. 56:907–912.
Beauchamp, G. K., and Cowart, B. J. (1987). Preference for Birch, L. L. (1980). The relationship between children’s food
extremely high levels of salt among young children. An. NY preferences and those of their parents. J. Nutr. Educ.
Acad. Sci. 510:171–172 12:14–18.
Beauchamp, G. K., and Cowart, B. (1990). Preference for high Birch, L. L. (1999). Development of food preferences. Annu. Rev.
salt concentrations among children. Dev. Psychol. 26:539–545. Nutr. 19:41–62.
Beauchamp, G. K., and Engleman, K. (1991). High salt intake: Birch, L. L., and Deysher, M. (1986). Caloric compensation and
sensory and behavioral factors. Hypertension 17 (Suppl. sensory specific satiety: Evidence for self-regulation of food
I):I176–I181. intake by young children. Appetite 7:323–331.
Ontogeny of Human Flavor Perception 841

Birnholtz, J. C. (1988). On observing the human fetus. In Cowart, B. J. (1981). Development of taste perception in humans:
Behavior of the Fetus, W. P. Smotherman and S. R. Robinson Sensitivity and preference throughout the life span. Psychol.
(Eds.). Telford Press Caldwell, NJ, pp. 47–60. Bull. 90:43–73.
Blass, E. M. (1997). Milk-induced hypoalgesia in human new- Crook, C. K. (1977). Taste stimulation and the temporal organi-
borns. Pediatrics 99:825–829. zation of neonatal sucking. In Taste and Development: The
Blass, E. M., and Hoffmeyer, L. B. (1991). Sucrose as an anal- Genesis of Sweet Preference, J. M. Weiffenbach (Ed.). U.S.
gesic for newborn infants. Pediatrics 87:215–218. Dept. H.E.W. Publications, Bethesda, MD, pp. 146–158.
Blass, E. M., and Shah, A. (1995). Pain-reducing properties of Crook, C. K. (1978). Taste perception in the newborn infant.
sucrose in human newborns. Chem. Senses 20:29–35. Infant Behav. Dev. 1:52–69.
Blass, E. M., and Shide, D. J. (1994). Some comparisons among Crook, C. K. (1981). Functional aspects of the chemical senses in
the calming and pain-relieving effects of sucrose, glucose, the newborn period. Dev. Med. Child Neurol. 23:247–250.
fructose and lactose in infant rats. Chem. Senses 19: Crook, C. K., and Lipsitt, L. P. (1976). Neonatal nutritive suck-
239–249. ing: Effect of taste stimulation upon sucking and heart rate.
Blass, E. M., and Smith, B. A. (1992). Differential effects of Child Dev. 47:518–522.
sucrose, fructose, glucose and lactose on crying in 1-3-day old Crystal, S. R., and Bernstein, I. L. (1995). Morning sickness:
human infants. Dev. Psychol. 28:804–810. impact on offspring salt preference. Appetite 25:231–240.
Blass, E. M., and Watt, L. B. (1999). Suckling- and sucrose- Crystal, S. R., and Bernstein, I. L. (1998). Infant salt preference
induced analgesia in human newborns. Pain 83:611–623. and mother’s morning sickness. Appetite 30:297–307.
Blass, E. M., Ganchrow, J. R., and Steiner, J. E. (1984). Classical DeCasper, A. J., and Fifer, W. P. (1980). Of human bonding:
conditioning in newborn humans 2-48 hours of age. Infant Newborns prefer their mother’s voices. Science 208:
Behav. Dev. 7:223–235. 1174–1176.
Blass, E. M., Fillion, T. J., Rochat, P., Hoffmeyer, L. B., and DeCasper, A. J., and Spence, M. J. (1986). Prenatal maternal
Metzher, M. A. (1989). Sensorimotor and motivational deter- speech influences newborns’ perception of speech sound.
minants of hand-mouth coordination in 1-3-day-old human Infant Behav. Dev. 9:133–150.
infants. Dev. Psychol. 25:963–975. De Graaf, C., and Zandstra, E. H. (1999). Sweetness intensity and
Bradley, R. M. (1972). Development of the taste bud and gusta- pleasantness in children, adolescents and adults. Physiol.
tory papillae in human fetuses. In Oral Sensation and Behav. 67:513–520.
Perception, J. F. Bosma (Ed.). Charles C Thomas, Springfield, Desor, J. A., and Beauchamp, G. K. (1987). Longitudinal changes
IL., pp. 137–162. in sweet preferences in humans. Physiol. Behav. 39:639–641.
Bradley, R. M., and Stern, I. B. (1967). The development of the Desor, J. A., Maller, O., and Turner, R. E. (1973). Taste in accep-
human taste bud during the foetal period. J. Anat. tance of sugars by human infants. J. Comp. Physiol. Psychol.
101:743–752. 84:496–501.
Bridger, W. H. (1962). Ethological concepts and human develop- Desor, J. A., Greene, L. S., and Maller, O. (1975a). Preferences
ment. Rec. Adv. Biol. Psychiatry 4:95–107. for sweet and salty in 9- to 15-year old and adult humans.
Büssem, H. (1895). Über Geschmacksempfindungen rachitischer Science 190:686–687.
and nichtrachitischer Kinder. Jahrb. Kinderheilk. 39:166–183. Desor, J. A., Maller, O., and Andrews, K. (1975b). Ingestive
Cameron, A. T. (1947). The Taste Sense and the Relative responses of human newborns to salty, sour and bitter stimuli.
Sweetness of Sugars and Other Sweet Substances. Sugar J. Comp. Physiol. Psychol. 89:966–970.
Research Foundation, Inc., New York (Sci. Rep. Ser. No. 9). Eckstein, A. (1927). Zur Physiologie der Geschmacks-
Campbell, R. G. (1976). A note on the use of feed flavour to stim- empfindung und des Saugreflexes bei Säuglingen. Z.
ulate the feed intake of weaner pigs. Anim. Prod. 23:417–419. Kinderheilk. 45:1–18.
Capretta, P. J., Petersik, J. T., and Steward, D. J. (1975). Engen, T. (1974). Method and theory in the study of odor prefer-
Acceptance of novel flavours is increased after early experi- ences. In Human Responses to Environmental Odors, A. Turk,
ence of diverse taste. Nature 254:689–691. J. W. Johnston and D. G. Moulton (Eds.). Academic Press,
Carbajal, R., Chauvet, X., Couderc, S., Olivier-Martin, M. New York, p. 121.
(1999). Randomised trial of analgesic effects of sucrose, glu- Engen, T. (1982). The Perception of Odors. Academic Press, New
cose, and pacifiers in term neonates. Br. Med. J. York.
319:1393–1397. Engen, T., Lipsitt, L. P., and Robinson, D. O. (1978). The human
Catalanotto, F. A., Gaulin-Kremer, E., and Shaw, J. L. (1979). newborn’s sucking behavior for sweet fluids as a function of
Sucrose taste functions and dental caries in children. J. Dent. birthweight. Infant Behav. Dev. 1:118–121.
Res. 58:1327–1332. Enns, M. P., Van Itallie, T. B., and Grinker, J. A. (1979).
Cernoch, J. M., and Porter, R. H. (1985). Recognition of mater- Contributions of age, sex and degree of fatness on preferences
nal axillary odors by infants. Child Dev. 56:1593–1598. and magnitude estimations for sucrose in humans. Physiol.
Chuah, M. I., and Zheng, D. R. (1987). Olfactory marker protein Behav. 22:999–1003.
is present in olfactory receptor cells of human fetuses. Fifer, W. P., and Moon, C. (1988). Auditory experience in
Neuroscience 23:363–370. the fetus. In Behavior of the Fetus, W. P. Smotherman
842 Ganchrow and Mennella

and S. R. Robinson (Eds.). Telford Press, Caldwell, NJ, pp. Hauser, G. J., Chitayat, D., Berbs, L., Braver, D., and Mulbauer, B.
175–188. (1985). Peculiar odors in newborns and maternal pre-natal
Fildes, V. (1986). Breasts, Bottles and Babies: A History of Infant ingestion of spicy foods. Eur. J. Pediatr. 44:403.
Feeding. Edinburgh University Press, Edinburgh. Heiderich, F. (1906). Die Zahl und die Dimension der
Fox, N. A., and Davidson, R. J. (1986). Taste-elicited changes Geschmacksknospen der Papilla vallata des Menschen in den
in facial signs of emotion and the asymmetry of brain electri- verschiedenen Lebensaltern. Nachr. koen. Gesell. Wissensch.
cal activity in human newborns. Neuropsychology 24: Goettingen math.-physik. Kl. 1:54–64.
417–422. Hellman, T. J. (1922). Die Genese der Zungenpapillen beim
Frye, R. E., Schwartz, B. S., and Doty, R. L. (1990). Dose-related Menschen. Uppsala Lak. For. Forh. 26/27:1–72.
effects of cigarette smoking on olfactory function. JAMA Hepper, P. G. (1987). The amniotic fluid: an important priming
263:1233–1236. role in kin recognition. Anim. Behav. 35:1343–1346.
Galef, B. G., and Clark, M. M. (1972). Mother’s milk and adult Hepper, P. G. (1988a). Adaptive fetal learning: Prenatal exposure to
presence: two factors determining initial dietary selection by garlic affects postnatal preferences. Anim. Behav. 36:935–936.
weanling rats. J. Comp. Physiol. Psych. 78:220–225. Hepper, P. G. (1988b) Foetal ‘soap’ addiction. Lancet
Ganchrow, J. R., Oppenheimer, M., and Steiner, J. E. (1979). 1(8598):1247–1248.
Behavioural displays to gustatory stimuli in newborn rabbit Hersch, M., and Ganchrow, D. (1980). Scanning electron
pups. Chem. Senses Flav. 4:49–61. microscopy of developing papillae on the tongue of human
Ganchrow, J. R., Steiner, J. E., and Daher, M. (1983). Neonatal embryos and fetuses. Chem. Senses 5:331–341.
facial expressions in response to different qualities and inten- Herschel, M., Khoshnood, B., Ellman, C., Maydew, N., and
sities of gustatory stimuli. Infant Behav. Dev. 6: 189–200. Mittendorf, R. (1998). Neonatal circumcision: Randomized
Ganchrow, J. R., Steiner, J. E., and Canetto, S. (1986). Behavioral trial of a sucrose pacifier for pain control. Arch. Pediatr. Adol.
displays to gustatory stimuli in newborn rat pups. Dev. Med. 152:279–284.
Psychobiol. 19:163–174. Hervada, A. R., and Newman, D. R. (1992). Weaning: historical
perspectives, practical recommendations, and current contro-
Ganchrow, J. R., Steiner, J. E., and Bartana, A. (1990).
versies. Curr. Prob. Pediatr. 22:223–240.
Behavioral reactions to gustatory stimuli in young chicks
Hill, D. L., and Mistretta, C. M. (1990). Developmental neurobi-
(Gallus gallus domesticus). Dev. Psychobiol. 23:103–118.
ology of salt taste sensation. Trends Neurosci. 13:188–195.
Gerber, S. (1991). Le gôut et les jeunes à table. J. Med.
Hoffmann, A. (1875). Über die Verbreitung der
Strasbourg 22:570–572.
Geschmacksknospen beim Menschen. Virchow’s Arch. Path.
Gilbertson, T. A. (1998). Gustatory mechanisms for the detection
Anat. Physiol. 62:516–530.
of fat. Curr. Opin. Neurobiol. 8:447–452.
Hudson, R., and Distel, H. (1999). The flavor of life: Perinatal
Goldman, A. S., Atkinson, S. A., and Hanson, L. (Eds.) (1987).
development of odor and taste preferences. Schweiz Med.
Human Lactation 3: The Effects of Human Milk on the
Wochenschr. 129:176–181.
Recipient Infant. Plenum Press, New York.
Hutchinson, I., Segovia, C., Laing, D. G., and Jinks, A. (2001).
Goldschmidt, H. (1927). Zur Physiologie der Geschmacks- Fungiform papillae — differences between children and
empfindung und des Saugreflexes bei Säuglingen. Z. adults. Chem. Senses 26:721.
Kinderheilk. 45:28–35. James, C. E., Laing, D. G., and Oram, N. (1997). A comparison
Graillon, A., Barr, R. G., Young, S. N., Wright, J. H., and of the ability of 8-9-year-old children and adults to detect taste
Hendricks, L. A. (1997). Differential response to oral sucrose, stimuli. Physiol. Behav. 62:193–197.
quinine and corn oil in crying human newborns. Physiol. James, C. E., Laing, D. G., Oram, N., and Hutchinson, I. (1999).
Behav. 62:317–325. Perception of sweetness in simple and complex taste stimuli
Grill, H. J., and Norgren, R. (1978). The taste reactivity test. II. by adults and children. Chem. Senses 24:281–287.
Mimetic responses to gustatory stimuli in chronic thalamic Johnson, E. W., Eller, O. M., and Jafek, B. W. (1995).
and chronic decerebrate rats. Brain Res. 143:281–297. Distribution of OMP-, PCP 9.5- and CaBP-like immunoreac-
Grinker, J. A., Gropman-Rubin, J., and Bose, K. (1986). Sweet tive chemoreceptor neurons in the developing human olfac-
preference and body fatness: neonatal data. Nutr. Behav. 3: tory epithelium. Anat. Embryol. (Berl.) 181:311–317.
197–209. Johnston, C. C., Stremler, R. L., Stevens, B. J., and Horton, L. J.
Haouari, N., Wood, C., Griffiths, G., and Levene, M. (1995). The (1997). Effectiveness of oral sucrose and simulated rocking
analgesic effect of sucrose in full term infants: a randomised on pain response in preterm neonates. Pain 72: 193–199.
controlled trial. Br. Med. J. 310:1498–1500. Kajiura, H., Cowart, B. J., and Beauchamp, G. K. (1992). Early
Harper, H. W., Jay, J. R., and Erickson, R. P. (1966). Chemically developmental change in bitter taste responses in human
evoked sensations from single human taste papillae. Physiol. infants. Dev. Psychobiol. 25:375–386.
Behav. 1:319–325. Kalmus, H. (1976). PTC testing of infants. Ann. Hum. Genet.
Harris, G., and Booth, D. A. (1987). Infants’ preference in food Lond. 40:139–140.
and previous dietary exposure in 6-month-old infants. IRCS Kiefer, S. W., Rusiniak, K. W., and Garcia, J. (1982). Flavor-
Med. Sci. 13:1178–1179. illness aversions: Gustatory neocortex ablations disrupt taste
Ontogeny of Human Flavor Perception 843

but not taste-potentiated odor cues. J. Comp. Physiol. Psychol. Maekawa, K., Sano, M., and Nakae, Y. (1991). Developmental
96:540–548. change of sucking response to taste in infants. Biol. Neonate
Kimmel, S. A., Sigman-Grant, M., and Guinard, J. X. (1994). 60 (Suppl. 1):62–74.
Sensory testing with young children. Food Technol. 94:92–99. Maga, J. A., and Lorenz, K. (1972). The effect of flavor
Korner, A. F., Chuck, B., and Dontchos, S. (1968). Organismic enhancers on direct headspace gas and liquid chromotography
determinants of spontaneous oral behavior in neonates. Child profiles of beef broth. J. Food Sci. 37:963–964.
Dev. 39:1145–1157. Maller, O., and Desor, J. A. (1973). Effect of taste on ingestion by
Kussmaul, A. (1894). Untersuchungen über das Seelenleben des human newborns. In Fourth Symposium on Oral Sensation
neugeborenen Menschen, 3rd ed., Verlag Moser, Tübingen, and Perception, J. F. Bosma (Ed.). U.S. Dept. H.E.W.
pp. 1–32. Publications, Bethesda, MD, pp. 279–291.
Lasiter, P. S., and Diaz, J. (1992). Artificial rearing alters devel- Maone, T. R., Mattes, R., Bernbaum, J., and Beauchamp, G. K.
opment of the nucleus of the solitary tract. Brain Res. Bull. (1990). A new method of delivering a taste without fluids to
29:407–410. preterm infants. Dev. Psychobiol. 23:179–191.
Lalonde, E. R., and Eglitis, J. A. (1961). Number and distribution Marlier, L., Schaal, B., and Soussignan, R. (1998). Neonatal
of taste buds on the epiglottis, pharynx, larynx, soft palate and responsiveness to the odor of amniotic and lacteal fluids:
uvula in a human newborn. Anat. Rec. 140:91–93. A test of perinatal chemosensory continuity. Child Dev. 69:
Lehrner, J. P., Glück, J., and Laska, M. (1999). Odor identifica- 611–623.
tion, consistency of label use, olfactory threshold and their Martin du Pan, R. (1955). Le role du gôut et de l’odorat dans
relationships to odor memory over the human lifespan. Chem. l’alimentation du nourrisson. Pediatrie 10:169–176.
Senses 24:337–346. Mennella, J. A. (1995). Mother’s milk: a medium for early flavor
Leshem, M. (1998). Salt preference in adolescence is predicted experiences. J. Human Lactation 11:39–45.
by common prenatal and infantile mineralofluid loss. Physiol. Mennella, J. A. (1997a). The flavor world of infants: a cross-
Behav. 63:699–704. cultural perspective. Nutr. Today 32:142–151.
Leshem, M., Maroun, M., and Weintraub, Z. (1998). Neonatal Mennella, J. A. (1997b). The human infants’ suckling responses
diuretic therapy may not alter children’s preference for salt to the flavor of alcohol in mother’s milk. Alcoholism: Clin.
taste. Appetite 30:53–64. Exp. Res. 21:581–585.
Lewindon, P. J., Harkness, L., and Lewindon, N. (1998). Mennella, J. A. (1998). Short-term effects of maternal alcohol
Randomised controlled trial of sucrose by mouth for the relief consumption on lactational performance. Alcoholism: Clin.
of infant crying after immunisation. Arch. Dis. Child Exp. Res. 22:1389–1392.
78:453–456. Mennella, J. A. (1999). The transfer of alcohol to human milk:
Lichtenstein, A. (1893/1894). Ueber die Geschmacksempfindung Sensory implications and effects on mother-infant interaction.
gesunder und rachitischer Kinder. Jahrb. Kinderheilk. In Alcohol and Alcoholism: Brain and Development, J. H.
37:76–90. Hannigan, N. Spear, L. Spear, and C. R. Goodlett (Eds.).
Lieblich, I., Cohen, E., Ganchrow, J. R., Blass, E. M., and Lawrence Erlbaum Associates, Inc., Mahwah, NJ, pp.
Bergmann, F. (1983). Morphine tolerance in genetically 177–198.
selected rats induced by chronically elevated saccharine Mennella, J. A., and Beauchamp, G. K. (1991a). Maternal diet
intake. Science 221:871–873. alters the sensory qualities of human milk and the nursling’s
Lipsitt, L. P. (1977). Taste in human neonates: Its effect on suck- behavior. Pediatrics 88:737–744.
ing and heart rate. In Taste and Development: The Genesis of Mennella, J. A., and Beauchamp, G. K. (1991b). The transfer of
Sweet Preference, J. M. Weiffenbach (Ed.). U.S. Dept. H.E.W. alcohol to human milk: Effects on flavor and the infant’s
Publications, Bethesda, MD, pp.125–140. behavior. N. Engl. J. Med. 325:981–985.
Lipsitt, L. P., Reilly, B. M., Butcher, M. J., and Greenwood, M. Mennella, J. A., and Beauchamp, G. K. (1993a). Early flavor
M. (1976). The stability and interrelationships of newborn experiences: When do they start? Pediatr. Basics 65:2–7.
sucking and heartrate. Dev. Psychobiol. 9:305–310. Mennella, J. A., and Beauchamp, G. K. (1993b). Beer, breast
Lorenz, K. (1965). Evolution and Modification of Behavior. The feeding and folklore. Dev. Psychobiol. 26:459–466.
University of Chicago Press, Chicago. Mennella, J. A., and Beauchamp, G. K. (1993c). The effects of
McDaniel, M. R. (1980). Off-flavors in human milk. In The repeated exposure to garlic-flavored milk on the nursling’s
Analysis and Control of Less Desirable Flavors in Foods and behavior. Pediatr. Res. 34:805–808.
Beverages, G. Charalambous (Ed.). Academic Press, New Mennella, J. A., and Beauchamp, G. K. (1996a). The early
York, pp. 267–291. development of human flavor preferences. In Why We Eat
Mace, J. W., Goodman, S. I., Centerwall, W. R., and Chinnock, What We Eat: The Psychology of Eating, E. D. Capaldi (Ed.).
R. F. (1976). The child with an unusual odor: a clinical American Psychological Association, Washington, DC, pp.
resumè. Clin. Pediatr. 15:57–62. 83–112.
MacFarlane, A. J. (1975). Olfaction in the development of social Mennella, J. A., and Beauchamp, G. K. (1996b). The human
preferences in the human neonate. Ciba Found. Sympo. 33: infant’s response to vanilla flavors in mother’s milk and for-
103–117. mula. Infant Behav. Dev. 9:13–19.
844 Ganchrow and Mennella

Mennella, J. A., and Beauchamp, G. K. (1997a). The ontogeny of Overgaard, C., and Knudsen, A. (1999). Pain-relieving effect of
human flavor perception. In Handbook of Perception and sucrose in newborns during heel prick. Biol. Neonate
Cognition: Tasting and Smelling, G. K. Beauchamp and L. 75:279–284.
Bartoshuk (Eds.). Academic Press, San Diego, pp. 199–221. Peiper, A. (1928). Die Hirntätigkeit des Säuglings. In Ergebnisse
Mennella, J. A., and Beauchamp, G. K. (1997b). Mother’s milk der inneren Medizin und Kinderheilkunde, Sonderabdruck:
enhances the acceptance of cereal during weaning. Pediatr. 33, Springer-Verlag, Berlin, pp. 505–605.
Res. 41:188–192. Piras, A. T., and Mazzarello, V. (1985). Caratteristiche della
Mennella, J. A., and Beauchamp, G. K. (1998). Smoking and the superficie dorsale della lingua fetale umana osservata al
flavor of breast milk—letter. N. Engl. J. Med. 339:1559–1560. microscopio a scansione. Arch. Ital. Anat. Embriol. XC:
Mennella, J. A., and Beauchamp, G. K. (1999). Experience with 17–30.
a flavor in mother’s milk modifies the infant’s acceptance of Pomonis, J. D., Jewett, D. C., Kotz, C. M., Briggs, J. E.,
flavored cereal. Dev. Psychobiol. 35:197–203. Billington, C. J., and Levine, A. S. (2000). Sucrose consump-
Mennella, J. A., Jagnow, C. P., Beauchamp, G. K. (2001). tion increases naloxone-induced c-Fos immunoreactivity in
Prenatal and postnatal flavor learning by human infants. limbic forebrain. Am. J. Physiol. Reg. Integr. Comp. Physiol.
Pediatrics 107:E88. 278:R712–R719.
Mennella, J. A., Johnson, A., and Beauchamp, G. K. (1995). Preyer, W. (1912). Die Seele des Kindes, 9th ed. Th. Griebens
Garlic ingestion by pregnant women alters the odor of amni- Verlag, Leipzig, pp. 65–69.
otic fluid. Chem. Senses 20:207–209. Pritchard, J. A. (1965). Deglutition by normal and anencephalic
Mercer, M. E., and Holder, M. D. (1997). Antinociceptive effects fetuses. Obstet. Gynecol. 5:289–297.
of palatable sweet ingesta on human responsivity to pressure Raphael, D. (1982). Weaning is forever. Lactation Rev. 6:8–34.
pain. Physiol. Behav. 61:311–318. Ramenghi, L. A., Griffith, G. C., Wood, C. M., and Levene, M. I.
Miller, A., Barr, R. G., and Young, S. N. (1994). The cold pressor (1996). Effect of non-sucrose sweet tasting solution on neona-
test in children: Methodological aspects and the analgesic tal heel prick responses. Arch. Dis. Child Fetal Neonatal Ed.
74:F129–F131.
effect of intraoral sucrose. Pain 56:175–183.
Ramenghi, L. A., Evans, D. J., and Levene, M. I. (1999).
Miller, I. J. (1988). Human taste bud density across adult age
“Sucrose analgesis”: Absorptive mechanism or taste percep-
groups. J. Gerontol. 43:B26–B30.
tion? Arch. Dis. Child Fetal Neonatal Ed. 80:F146–F147.
Mistretta, C. M. (1972). Topographical and histological study of
Richman, R. A., Sheehe, P. R., Wallace, K., Hyde, J. M., and
the developing rat tongue, palate and taste buds. In Third
Coplan, J. (1995). Olfactory performance during childhood.
Symposium on Oral Sensation and Perception. The Mouth of
II. Developing a discrimination task for children. J. Pediatr.
the Infant, J. F. Bosma (Ed.). Charles Thomas, Springfield, IL.
127: 421–426.
pp. 163–186.
Rao, M., Blass, E. M., Brignol, M. J., Marino, L., and Glass, L.
Mistretta, C. M., and Bradley, R. M. (1975). Taste and swallow-
(1997). Reduced heat loss following sucrose ingestion in pre-
ing in utero. Br. Med. Bull. 31:80–84. mature and normal human newborns. Early Hum. Dev.
Neumann, H. (1896). Bemerkung über die Geschmacksempfindung 48:109–116.
bei kleinen Kindern. Jahrb. Kinderheilk. 41: 155–159. Rolls, B. J., Rowe, E. S., and Rolls, E. T. (1982). How sensory
Nijjar, R., Straits, K., Sniffen, J., Gilmore, M., and Murphy, C. properties of foods affect human feeding behavior. Physiol.
(1991). Salt preference in anglo and hispanic preschoolers. Behav. 29:409–417.
Chem. Senses 16:563. Rolls, E. T., Critchley, H. D., Browning, A. S., Hernadi, I., and
Nilsson, B., and Holm, A. K. (1983). Taste thresholds, taste pref- Lenard, L. (1999). Responses to the sensory properties of fat
erences and dental caries in 15-year-olds. J. Dent. Res. of neurons in the primate orbitofrontal cortex. J. Neurosci.
62:1069–1072. 19:1532–1540.
Nisbett, R. E., and Gurwitz, S. B. (1970). Weight, sex and the eat- Rosenstein, D., and Oster, H. (1988). Differential facial
ing behavior of human newborns. J. Comp. Physiol. Psych. responses to four basic tastes in newborns. Child Dev.
73:245–253. 59:1555–1568.
Nowlis, G. H., and Kessen, W. (1977). From reflex to representa- Rovee, C. K. (1972). Olfactory cross-adaptation and facilitation
tion: Taste-elicited tongue movements in the human newborn. in human neonates. J. Exp. Child Psychol. 13:368–381.
In Taste and Development: The Genesis of Sweet Preference, Rozin, P. (1976). The selection of food by rats, humans and other
J. M. Weiffenbach (Ed.). U.S. Dept. H.E.W. Publications, animals. In Advances in the Study of Behaviors, Vol. 6, J.
Bethesda, MD, pp. 190–202. Rosenblatt, R. A. Hinde, C. Beer, and E. Shaw (Eds.).
Okoro, E. O., Brisibe, F., Jolayemi, E. T., and Hadizath Academic Press, New York, pp. 21–76.
Taimagari, G. (2000). Taste sensitivity to sodium chloride and Rozin, P. (1982). “Taste-smell confusions” and the duality of the
sucrose in a group of adolescent children in northern Nigeria. olfactory sense. Percept. Psychophy. 31:397–401.
Ethn. Dis. 10:53–59. Schaal, B. (1986). Presumed olfactory exchanges between
Osepian, V. A. (1958). Development of the function of the taste mother and neonate in humans. In Ethology and Psychology,
analyzer in the first year of life. Pavlov. J. Higher Nerv. Activ. J. Le Camus and J. Conier (Eds.). Private IEC, Toulouse, pp.
8:766–772.
101–110.
Ontogeny of Human Flavor Perception 845

Schaal, B. (1988). Olfaction in infants and children: development Steiner, J. E. (1977). Facial expressions of the neonate infant
and functional perspectives. Chem. Senses 13:145–190. indicating the hedonics of food-related chemical stimuli. In
Schaal, B., and Orgeur, P. (1992). Olfaction in utero: can the rodent Taste and Development: The Genesis of Sweet Preference,
model be generalized? Quart. J. Exp. Psychol. 44:245–278. J. M. Weiffenbach (Ed.). U. S. Dept. H. E. W. Publications,
Schaal, B., Marlier, L., and Soussignan, R. (1995a). Bethesda, MD, pp. 173–188.
Responsiveness to the odour of amniotic fluid in the human Steiner, J. E. (1979a). Oral and facial innate motor responses to
neonate. Biol. Neonate 67:397–406. gustatory and to some olfactory stimuli. In Preference
Schaal, B., Orgeur, P., and Rognon, C. (1995b). Odor sensing in Behaviour and Chemoreception, J. H. A. Kroeze (Ed.).
the human fetus: anatomical, functional and chemo-ecologi- Information Retrieval, Ltd., London, pp. 247–258.
cal bases. In Fetal Development: A Psychobiological Steiner, J. E. (1979b). Human facial expression in response to
Perspective, J. P. Lecanuet, W. P. Fifer, N. A. Krasnegor, and taste and smell stimulation. In Advances in Child
W. P. Smotherman (Eds.). Lawrence Erlbaum Associates, Development and Behavior, Vol. 13, H. W. Reese and L. P.
Mahwah, NJ, pp. 205–238. Lipsitt (Eds.). Academic Press, New York, pp. 257–293.
Schaal, B., Marlier, L., and Soussignan R. (1998). Olfactory Steiner, J. E. (1987). What the neonate can tell us about umami.
function in the human fetus: evidence from selective neonatal In Umami: A Basic Taste, Y. Kawamura and M. R. Kare
responsiveness to the odor of amniotic fluid. Behav. Neurosci. (Eds.). Marcel Dekker, New York, pp. 97–123.
112:1438–1439. Stevens, B., Taddio, A., Ohlsson, A., and Einarson, T. (1997). The
Schaffer, J. P. (1910). The lateral wall of the cavum nasi in man efficacy of sucrose for relieving procedural pain in neonates—
with special reference to the various developmental stages. A systematic review and meta-analysis. Acta Paediatr. Int.
J. Morphol. 21:613–617. J. Paediatr. 86:837–842.
Schmidt, H. J., and Beauchamp, G. K. (1988). Adult-like odor Stickrod, G., Kimble, D. P., and Smotherman, W. P. (1982). In
preferences and aversions in three-year-old children. Child utero taste/odor aversion conditioning in the rat. Physiol.
Dev. 59:1136–1143. Behav. 28:5–7.
Schwartz, M. D., Jacobsen, P. B., and Bovbjerg, D. H. (1996). Stirnimann, F. (1935). Versuche über Geschmack und Geruch am
Role of nausea in the development of aversions to a beverage ersten Lebenstag. Jahrb. Kinderheilk. 146:211–229.
paired with chemotherapy treatment in cancer patients. Sullivan, S., and Birch, L. L. (1990). Pass the sugar; pass the
Physiol. Behav. 59:659–663. salt: Experience dictates preference. Dev. Psychol.
Scott, T. R. (1992). The neural basis of body wisdom. World Rev. 26:546–551.
Nutr. Diet. 67:1–39. Sullivan, S., and Birch, L. L. (1994). Infant dietary experience
Shapera, M. R., Moel, D. I., Kamateh, D. K., Olsen, R., and and acceptance of solid foods. Pediatrics 93:271–277.
Beauchamp, G. K. (1986). Taste perception of children with Sullivan, R. M., and Toubas, P. (1998). Clinical usefulness of
chronic renal failure. J. Am. Dietet. Assoc. 86:1359–1365. maternal odor in newborns: soothing and feeding preparatory
Skinner, J. D., Pope, J. F., and Carruth, B. R. (1998). Alterations responses. Biol. Neonate 74:402–408.
in adolescents’ sensory taste preferences during and after Tatzer, E., Schubert, M. T., Timischl, W., and Simbruner, G.
pregnancy. J. Adolesc. Health 22:43–49. (1985). Discrimination of taste and preference for sweet in
Smith, B. A., and Blass, E. M. (1996). Taste-mediated calming in premature babies. Early Human Dev. 12:23–30.
premature, preterm and full-term human infants. Dev. Temple, E. C., Hutchinson, I., Laing, D. G., and Jinks A. L.
Psychol. 32:1084–1089. (2002) Taste development: differential growth rates of
Smith, B. A., Fillion, T. J., and Blass, E. M. (1990). Orally-medi- tongue regions in humans. Brain Res Dev. Brain Res.
ated sources of calming in one to three day-old human infants. 135:65–70.
Dev. Psychol. 26:731–737. Terry, L. M., and Johanson, I. B. (1987). Olfactory influences on
Soussignan, R., Schaal, B., Marlier, L., and Jiang, T. (1997). the ingestive behavior of infant rats. Dev. Psychobiol.
Facial and autonomic response to biological and artificial 20:313–332.
olfactory stimuli in human neonates: Re-examining early Tuckerman, F. (1888). The anatomy of the papilla foliata of the
hedonic discrimination of odors. Physiol. Behav. 62: 745–758. human infant. J. Anat. Physiol. 22:499–501.
Stein, L. J., Cowart, B. J., Epstein, A. N., Pilot, L. J., Laskin, Tuckerman, F. (1890). Further observations on the develop-
C. R., and Beauchamp, G. K. (1996). Increased liking for ment of the taste organs of man. J. Anat. Physiol.
salty foods in adolescents exposed during infancy to chloride- 24:130–131.
deficient feeding formula. Appetite 27:65–77. Vasquez, M., Pearson, P. B., and Beauchamp, G. K. (1982).
Steiner, J. E. (1973). The gustofacial response: Observation on Flavor preferences in malnourished Mexican infants. Physiol.
normal and anencephalic newborn infants. In Oral Sensation Behav. 28:513–519.
and Perception, J. F. Bosma (Ed.). U. S. Dept. H. E. W. Varendi, H., Porter, R. H., and Winberg, J. (1994). Does the
Publications, Bethesda, MD, pp. 254–278. newborn baby find the nipple by smell? Lancet 334:
Steiner, J. E. (1974). Innate discriminative human facial expres- 989–990.
sions to taste and smell stimulation. Ann. NY Acad. Sci. Wachenheim, F. L. (1915). Infant-Feeding: Its Principles and
237:229–233. Practice. Lea and Febiger, New York, pp. 308–320.
846 Ganchrow and Mennella

Weiffenbach, J. M. (1977). Sensory mechanisms of the new- Wysocki, C. J., and Beauchamp, G.K. (1984). Ability to smell
born’s tongue. In Taste and Development: The Genesis of androstenone is genetically determined. Proc. Natl. Acad. Sci.
Sweet Preference, J. M. Weiffenbach (Ed.). U. S. Dept. H. E. USA 81:4899–4902.
W. Publications, Bethesda, MD, pp. 205–212. Zeifman, D., Delaney, S., and Blass, E. M. (1996). Sweet taste,
Witt, M., and Reutter, K. (1996). Embryonic and early fetal looking, and calm in 2- and 4-week-old infants: The eyes have
development of human taste buds: A transmission electron it. Dev. Psychol. 32:1090–1099.
microscopical study. Anat. Rec. 246:503–523. Zmarzty, S. A., Wells, A. S., and Read, N. W. (1997). The influ-
Witt, M., and Reutter, K. (1997). Scanning electron microscopi- ence of food on pain perception in healthy human volunteers.
cal studies of developing gustatory papillae in humans. Chem. Physiol. Behav. 62:185–191.
Senses 22:601–612.
40

Genetics of Human Taste Perception

Adam Drewnowski
University of Washington, Seattle, Washington, U.S.A.

I. INTRODUCTION and other poisons generally have an unpleasant bitter taste

(Rouseff, 1990). Abnormal or extreme bitterness usually
The study of human taste genetics is largely the study of denotes toxicity and is the main reason for food rejection
bitter taste. Research on the genetics of human taste per- (Drewnowski and Gomez, 2000).
ception began in 1931 with the accidental finding that crys- Being able to identify and reject bitter poisons offers a
tals of phenylthiocarbamide (PTC) tasted bitter to some major evolutionary advantage. Bitter compounds, includ-
people but were tasteless to others (Blakeslee, 1931; Fox, ing extremely toxic ones, are detected by humans in micro-
1932). One in three people showed a heritable “taste blind- molar amounts. While there seems to be no linear relation
ness” to PTC (Blakeslee, 1931; Snyder, 1931) and to a between bitter taste thresholds and compound toxicity
related compound, 6-n-propylthiouracil (PROP). Since that (Hladik and Simmen, 1996), quinine is detected at 25
time, the phenotypic taste responses to PTC/PROP have mol/L concentrations and bitter plant toxins at levels far
been well described in the literature (Guo and Reed, 2001). below that. In contrast, taste detection thresholds for
However, the genes responsible for this trait have not been sucrose are in the order of 10,000 mol/L. Some
identified, and their exact location is still unknown (Reed, researchers have suggested that PROP responsiveness
2000). among women is hormone-driven, being enhanced during
The ability to taste PTC/PROP is a unique example of pregnancy and diminished following menopause
genetic taste polymorphism in humans. Though initially (Bartoshuk, 2000), although empirical data on this point
viewed as an all-or-none phenomenon, the ability to taste are sparse. Nonetheless, infants, young children, and preg-
PTC/PROP is a continuously distributed variable. nant women are most likely to avoid bitter taste and to
Sensitivity to PTC/PROP is influenced, moreover, by reject bitter-tasting foods (Drewnowski and Gomez, 2000).
race/ethnicity, sex, and age. How researchers characterize This taste-mediated rejection of bitter compounds is pre-
taste responses to PROP among individuals or groups, and sent at birth (Drewnowski, 1997).
what that tells us about genetic influences on human taste Although humans accept a low level of bitterness in
preferences, is the main focus of this review. some foods, intensely bitter tastes are aversive to most
people (Drewnowski, 2001; Drewnowski and Gomez,
II. THE ROLE OF BITTER TASTE 2000). Bitter taste sensation tends to be more prolonged as
compared to sweet, salty, or sour (McBurney, 1978). As a
While sweet taste is associated with dietary energy, result, food industry has taken steps to modulate the
bitterness tends to be equated with dietary danger (Hladik amount of bitterness present in the food supply
and Simmen, 1996; Rouseff, 1990) (see Chapter 35). (Drewnowski, 2001; Drewnowski and Gomez, 2000).
Hydrolyzed proteins, rancid fats, plant-derived alkaloids, Furthermore, individual sensitivity to bitter taste can vary

847
848 Drewnowski

widely. There is also a strong and inverse relationship allele models (Morton et al., 1981; Olson et al., 1989).
between perceived bitterness and acceptance ratings. Recent studies in humans have linked the ability to taste
While there is no consistent relationship between sweet- PROP to at least two genetic loci—at chromosome 5p15
ness intensity perception and liking for sweet taste, and at chromosome 7 (Guo and Reed 2001). Such studies
increased bitterness always predicts a greater dislike of bit- critically depend on an accurate classification of respon-
ter compounds and bitter-tasting foods. dents by PROP taster status. However, the definition of
what constitutes the ability to taste PROP has also evolved
with time.
III. BITTER TASTE GENETICS

As noted above, biology of bitter taste perception is poorly IV. TASTE DETECTION THRESHOLDS
understood. Among bitter-tasting compounds are amino
acids and peptides, esters and lactones, phenols and In early studies, taste responses to PTC crystals or PTC-
polyphenols, flavonoids and terpenes, methylxanthines impregnated filter papers served to establish genetic taste
(caffeine), sulfimides (saccharin), ureas and thioureas “blindness” (Fox, 1932; Snyder, 1931). PTC filter papers
(PTC, PROP), and organic and inorganic salts (Rousseff, were distributed by the American Genetic Association as
1990). These structurally unrelated compounds all give early as 1931. Later studies used PTC solutions or solu-
rise to a uniform sensation of bitter taste. That would tions of 6-n-propylthiouracil or PROP (Blakeslee and
suggest that bitterness is perceived through a variety of Salmon, 1935; Fischer and Griffin, 1964; Harris and
receptors and multiple transduction mechanisms. At least Kalmus, 1949). PROP lacks the faint sulfur smell of con-
three different types of bitter taste receptors were once centrated PTC solutions and is reportedly safer and easier
thought to exist, sensitive to quinine, urea, and to to use (Kalmus, 1971; Lawless, 1980).
PTC/PROP, respectively (McBurney, 1978). Since that Traditionally, tasters and nontasters of PTC or PROP
time, the estimated number of bitter taste receptors has were identified using threshold detection procedures
been revised upward. Studies have placed the number of (Harris and Kalmus, 1949). In contrast to most other
gustducin-linked bitter taste receptors in humans at substances, detection thresholds for PTC and PROP show
between 40 and 80 (Adler et al., 2000). a bimodal distribution, allowing for an easy segregation by
According to recent studies (Adler et al., 2000), these taster groups (Fischer, 1967; Fischer and Griffin, 1964;
candidate taste receptors (T2Rs) are organized in clusters Kalmus, 1971). The classic studies used a series of 15
and are genetically linked to loci that govern the perception PROP solutions that ranged in concentration from 0.001 to
of bitter taste. Candidate taste receptors were expressed in 3.2 mmol/L PROP, increasing in quarter-log steps
all taste buds of circumvallate, foliate, and palate papillae (Bartoshuk, 1979; Fischer, 1984; Harris and Kalmus,
(Adler et al., 2000). A follow-up study linked different 1949). The most concentrated solution (number 15 in the
receptors to specific compounds. Human bitter taste recep- PROP series) contained 0.54 g/L PROP, the next solution
tor (hT2R-4) responded only to denatonium and PROP, contained 0.30 g/L, and so on (Fischer, 1967; Kalmus,
while a mouse receptor MT2R-5 responded only to cyclo- 1971). Each subject was first presented with the least
heximide (Chandrashekar et al., 2000). The recognition of concentrated solution of PROP (solution 1) and then with
multiple bitter tastants may occur at the level of the increasingly higher solutions until he or she reported
individual cell. These studies are consistent with the notion detecting a taste distinct from that of water. The subject
that humans are capable of recognizing bitter substances, was then presented with two identical cups, one containing
but not always of distinguishing among them. the detected concentration of PROP and the other
PTC tasting in humans was initially thought to follow containing deionized water. The water was at the same
a simple Mendelian model, with taste deficiency inherited temperature and was stored in the same location as the
through a single genetic locus as an autosomal recessive PROP solution. The subject was asked to judge which of
trait (Blakeslee 1931; Snyder, 1931). PTC tasters were the two samples had the bitter taste (Bartoshuk et al., 1988;
thought to be homozygous (TT) or heterozygous (Tt or Drewnowski et al., 1997a,b,c; Fischer and Griffin, 1964).
tT) for the dominant allele (T), while nontasters were Wrong answers led to the presentation of the more
thought to carry two recessive alleles (tt) (Fischer, 1967; concentrated PROP solution, while correct answers led to
Kalmus, 1958, 1971). However, linkage studies later a second presentation of the same solution. Two consecu-
showed that the traditional single locus two-allele model tive correct answers at the same concentration led to the
of PTC sensitivity did not provide the best fit to the data, presentation of a less concentrated PROP solution.
and there have been suggestions of multilocus and multi- Reversal points were defined as the concentration at which
Genetics of Human Taste Perception 849

a series of correct responses turned to an incorrect (Kalmus, 1958). Furthermore, tasters who had taster
response or vice versa (Drewnowski et al., 1997a,b,c). siblings had lower PTC thresholds than tasters with at
Generally, thresholds were based on a mean of at least five least one nontaster sibling, suggesting that PTC tasting
reversals. Participants rinsed thoroughly with deionized was an incomplete dominant (Kalmus, 1958). A two-com-
water after tasting each PROP stimulus. ponent model would support a single locus dominant
PROP thresholds for a total of 541 women, ranging in model of inheritance, whereas a three-component model
age from 18 to 80 years, are shown in Figure 1. Consistent would support a single locus additive model of inheri-
with past studies, the data show an antimode around 0.1 tance. Reed et al. (1995) tested these hypotheses by
mmol/L PROP. PROP tasters are commonly defined as analyzing PROP taste thresholds collected from 1015
having thresholds below 0.1 mmol/L (solution 9), while unrelated subjects, using maximum likelihood estimation
nontasters are defined as having thresholds in excess of 0.2 and likelihood ratio tests. The results were inconclusive.
mmol/L (equivalent to solution 10). Bartoshuk (1979, Assuming equal variances, a three-component model was
1993). Respondents with PROP threshold between solu- more likely, but the two models were equally likely if the
tions 9 and 10 are often rejected as unclassifiable, since the variances could be unequal. However, unequal variances
taster and nontaster distributions are said to overlap in that lead to statistical problems, since maximum likelihood
range (Bartoshuk et al., 1994). is unbound and the likelihood ratio test is no longer
Since 1949 (Harris and Kalmus, 1949), taste detection appropriate.
thresholds have come to dominate laboratory research on Applying a Bayesian Finite Mixture Linear Model to
PTC/PROP. Some studies explored links between PROP PROP thresholds for a sample of 359 adult women, includ-
taster status and taste perception of other bitter or sweet ing 180 breast cancer cases, we found that threshold distri-
compounds (Bartoshuk, 1980). Other studies explored how bution, adjusted for age and cancer status, was consistent
the proportion of tasters and nontasters varied with sex, only with the two-component model. No three-component
age, ethnicity, maturation, or disease status (Bhalla, 1972; model fit the data. In other words, thresholds can distin-
Fischer and Griffin, 1964; Parr, 1934; Whissell-Buechy guish between tasters and nontasters, but do not support a
and Wills, 1989). In virtually all cases, the distinction multicomponent model. In particular, PROP thresholds
between tasters and nontasters was treated as a dichoto- cannot be used to determine whether individuals are
mous variable. homozygous or heterozygous for the dominant allele
This led to questions whether the distribution of (Bartoshuk et al., 1994).
PTC/PROP thresholds was truly bimodal or whether the
data could support more than two subgroups. The broad
range of PTC thresholds within the taster group suggested V. THE FILTER PAPER METHOD
that the ability to taste PTC was continuously distributed.
For the most part, threshold-based studies are limited to
small groups of subjects tested in the laboratory. Studies
conducted in the field, or with larger groups, have made
use of the filter paper method. Those studies presented
subjects with a piece of filter paper, which had been
impregnated with a saturated solution of PTC or PROP
(Parr, 1934). In early anthropological studies, responses to
PTC or PROP filter paper were used as a measure of
genetic inheritance. Those studies focused on the propor-
tion of tasters and nontasters in different populations and
ethnic groups (Das, 1958; Parr, 1934).
Filter papers are prepared by soaking laboratory filter
paper in a supersaturated solution of PROP, heated to close
to boiling point (Bartoshuk, 1993, Kaminski et al., 2000).
Some studies systematically varied the concentration of
PTC solution to produce papers of different degree of
bitterness. The paper was placed on the tongue and the
subject was asked about its taste. In some cases, subjects
Figure 1 Distribution of PROP detection thresholds for 541 were simply asked if the paper was bitter or not, in others
women. they were asked to estimate the intensity of bitter taste.
850 Drewnowski

Blakeslee and Salmon (1931) were the first to note that The correlation between PROP detection thresholds and
women were more sensitive than men to the bitter taste of bitterness intensity ratings was high, r
0.84.
PTC filter paper, an effect later confirmed by Bartoshuk Intensity scaling of PROP solutions is a much simpler
et al. (1994). procedure that the traditional detection threshold method.
The filter papers provide a crude but rapid way of assign- However, threshold methods have come to dominate the
ing taster status. However, different recipes for producing field to such an extent that Lawless’s approach was not fol-
PTC/PROP papers could lead to different classifications for lowed up until recently. In our studies, subjects tasted and
the same individual. The time of contact with the tongue and rated 5 PROP solutions at concentrations of 0.032, 0.1,
the achieved degree of hydration could also make a differ- 0.32, 1.0, and 3.2 mmol/L PROP. Following Lawless’s
ence. In most studies, subjects were simply instructed to (1980) approach, the solutions were arranged around the
place the paper flat on the tongue. In other studies, subjects antimode. We used standard 9-point category scales, where
who tasted nothing were asked to chew the filter paper, a 1
“not at all bitter” and 9
“extremely bitter.”
practice that might lead to false-positive responses (Lawless, Respondents also ranked each stimulus along a 9-point
1980). Pooled group data collected from attendees in the hedonic preference scale, initially developed by the U.S.
course of different lectures and presentations have also Army Quartermaster Corps (Peryam and Pilgrim, 1957).
found their way into the literature (Bartoshuk et al., 1994). That scale ranges from 1
“dislike extremely” to 9

These practices have produced a diversity of results. “like extremely,” with a neutral point at 5 (“neither like nor
The filter paper method has been criticized both for pro- dislike”). Allowing subjects to taste and rate several solu-
ducing false negatives among tasters and for producing tions across a perceptual range provides further informa-
false positives among nontasters (Lawless, 1980). Its tion about their taste response profile.
agreement with detection threshold data was reported as Figure 2 shows the distribution of summed intensity rat-
being only moderate (Lawless, 1980). Even so, the filter ings for the five PROP solutions obtained with 541
paper method has been the screening procedure of choice women. The distribution was a mirror image of the distri-
in many field studies. Under optimal conditions, the test- bution of thresholds, with a suggestion of bimodality. So
retest reliability of PROP filter papers from the same far there are no studies as to whether the distribution of
batch, tasted following uniform instructions can be very bitterness intensities supports a two- or a three-component
high. In our studies, subjects were instructed that the paper model, since all of the existing genetic linkage studies have
should remain in the mouth for a minimum of 3 seconds been conducted using threshold data (Olson et al., 1989;
(Ly and Drewnowski, 2000). Using three consecutive tests Reed et al., 1995; Whissell-Buechy and Wills, 1989).
some days apart, we obtained reliability coefficients of The distribution if hedonic ratings was found less useful
0.90 (Ly and Drewnowski, 2000). in this study. Since nobody likes PROP solutions much,
hedonic responses tended to occupy the bottom part of the
scale, from 1 through 5. However, as shown in Figure 3,
VI. BITTERNESS INTENSITY SCALING
OF SOLUTIONS

Detection thresholds do not always predict taste experi-

ence at above-threshold levels. Sensory acuity for very
dilute tastants provides a limited picture of everyday taste
function, especially in relation to eating habits (Bartoshuk,
2000; Drewnowski, 2001). Looking for a middle ground
between threshold procedures and the filter paper method,
some researchers have used a single solution of PTC or
PROP to determine taster status (Fischer and Griffin, 1961;
Lawless, 1979).
A near-antimode concentration was presented to subjects
in the expectation that nontasters would find it tasteless,
whereas tasters would find it very bitter. Lawless (1980)
tested samples of 0.1 mmol/L PTC and 0.56 mmol/L PROP
using an 8-point category scale that ranged from 0 (no taste)
to 7 (very strong taste). There was good agreement among Figure 2 Distribution of PROP bitterness ratings, summed over
threshold data and category-scale ratings of single stimuli. 5 solutions, for 541 women.
Genetics of Human Taste Perception 851

Figure 4 Bitterness intensity ratings as a function of PROP

concentrations. The data are for 541 women, classified into quin-
tiles by PROP detection thresholds. Highest threshold quintile
equals lowest responsiveness.
Figure 3 Distribution of PROP hedonic ratings, summed over 5
solutions, for 541 women. distinguish well between medium and high responders to
PROP.
some respondents gave neutral ratings to PROP solutions. The same respondents were then sorted by quintiles of
Those respondents were typically nontasters. As might be summed bitterness ratings for the 5 solutions of PROP.
inferred from the two figures, greater perceived bitterness Mean bitterness intensity functions for each group as a
was linked to a progressively greater dislike of PROP at an function of PROP concentration are shown in Figure 5.
individual level. Summed bitterness and summed hedonic Whereas a better separation between the curves was only
ratings for the 5 PROP solutions were strongly and to be expected, the shape of the response profiles was very
inversely linked (r
0.80; p  0.01). This relationship different. Respondents in the bottom quintile by bitterness
held for both tasters and nontasters. Assessing the dislike ratings showed a flat curve that only left the floor at 1
for PROP, as opposed to bitterness intensity scaling, may mmol/L PROP (solution 13). That is consistent with the
be a viable way of determining PROP taster status among notion that respondents in that group were nontasters, with
children (Drewnowski et al., 1997). a detection threshold above 0.2 mmol/L PROP. In contrast,
respondents in the top quintile perceived even the 0.03
VII. INTENSITY SCALING AND PROP TASTER mmol/L PROP solutions as intensely bitter, consistent with
STATUS the notion that they had lower thresholds and elevated

The use of bitterness intensity scaling of multiple PROP

solutions, instead of the traditional detection threshold pro-
cedure, is a relatively novel approach to the determination
of PROP taster status (Ly and Drewnowski, 2000). The
data of the study under discussion are based on 541 women
for whom both threshold data and intensity ratings were
available. Initially, we rank-ordered the threshold data into
5 equal-size groups (quintiles). Mean bitterness intensity
profiles were then established for each group as a function
of PROP concentration, as shown in Figure 4. PROP-
insensitive women or nontasters (top 20% of thresholds)
showed a profile that was different from the other four
groups. For PROP tasters, thresholds were a poor predictor Figure 5 Bitterness intensity ratings as a function of PROP
of bitterness intensity, consistent with past observations of concentrations. The data are for 541 women, classified into quin-
Bartoshuk et al. (1994). Threshold data allowed for a clear tiles by PROP bitterness ratings, summed over 5 solutions.
segregation of tasters and nontasters but did not Highest intensity quintile equals highest responsiveness. Mean
thresholds for each group are also indicated.
852 Drewnowski

sensitivity to PROP. Those respondents also reached intensity. Respondents were therefore ranked according to
the ceiling of the scale much sooner than the nontaster the ratio of bitterness intensity ratings for two PROP
group. solutions (1.0 and 3.2 mmol/L) relative to the perceived
Technically, the elbow of the curve should correspond saltiness of two solutions of sodium chloride (0.32 and 1
roughly to the mean detection threshold for that subject mol/L). Respondents in the top 25% of the values were
group. Figure 5 also shows the relationship between identified as supertasters. In other words, the cutpoint was
bitterness response profiles and the mean detection wholly arbitrary and was simply devised to produce 25%
thresholds for each group. Whereas mean threshold for the supertasters.
bottom quintile (nontasters) was above solution 10 (0.2 Respondents whose data did not fit that particular
mmol/L), consistent with the traditional definition of non- model were eliminated as unclassifiable. Figure 6
tasters, the mean threshold for the top quintile was around (Bartoshuk et al., 1994) shows a scatterplot of log PROP
solution 5 (0.01 mmol/L PROP), in the middle of the taster threshold versus PROP/NaCl ratio for 269 subjects. First,
range. respondents with thresholds between 0.1 and 0.2 mmol/L
Evidence was present for both floor and ceiling effects, PROP were eliminated since the taster and nontaster dis-
as shown in Figure 5. Bartoshuk (2000) has argued that tributions overlap in this range. Second, respondents with
rating single stimuli along a 9-point scale does not allow high PROP/NaCl ratios and low thresholds were
for a good separation among different categories of eliminated, as were respondents with low thresholds and
tasters. Indeed, in our study bitterness ratings of the most high ratios. Third, the arbitrary cutoff point was used to
concentrated 3.2 mmol/L solution of PROP clustered separate medium tasters from supertasters. In reducing the
toward the top of the scale and did not distinguish total number of classifiable subjects to 220, these
between medium and high responders. Conversely, ratings procedures yielded 16% nontasters, 56% medium tasters,
of the most dilute 0.03 mmol/L solution did not discrimi- and 28% supertasters.
nate between low and medium responders. On the other Other studies on PROP supertasters adopted the proce-
hand, there were no ceiling effects when subjects used the dures of Bartoshuk et al. (1994) with some modifications.
same 9-point scales to rate PROP solutions closer to the Rather than using magnitude estimation and the two
antimode, 0.1–0.3 mmol/L PROP. That is, of course, the highest concentrations of PROP and NaCl, we used cate-
reason why previous researchers used concentrations gory scales and all five solutions of PROP and NaCl.
around the antimode to distinguish between tasters and PROP solutions had concentrations as described above.
nontasters of PROP. Sodium chloride solutions had concentrations of 0.01,

VIII. TASTERS AND “SUPERTASTERS”

OF PROP

Being able to separate medium tasters from intense tasters

of PROP has some important implications. The wide vari-
ability of PTC detection thresholds among PTC tasters
suggested to Kalmus (1958) and later to Bartoshuk (1993)
that tasters could be subdivided into two groups.
Bartoshuk et al. (1994) speculated that nontasters had two
recessive alleles (tt); medium tasters were heterozygotes
with one dominant allele (Tt); while the most sensitive
“supertasters” had two dominant alleles (TT). The propor-
tion of “supertasters” expected on the basis of this single-
locus, two-allele model was 25%.
Since supertasters could not be identified using
threshold procedures, Bartoshuk used the ratio of per- Figure 6 A plot of mean PROP/NaCl ratios against PROP
ceived bitterness of PROP to the perceived saltiness of detection thresholds. The figure illustrates the classification of
NaCl to distinguish between the two groups. Studies had nontasters, tasters, and supertasters of PROP using a combination
shown that 3.2 mmol/L PROP was much more bitter to of detection threshold and intensity scaling procedures.
intense PROP tasters than 1.0 mol/L NaCl was salty. (Reprinted from Physiol. Behav. 56:1165–1171, 1994 with per-
Medium tasters found the two stimuli to be of comparable mission from Elsevier Science.)
Genetics of Human Taste Perception 853

hedonic response profiles were mirror images of each

other.
The question is whether a similar distribution can be
obtained using bitterness intensity ratings for 3 solutions,
as opposed to 5. Using the three middle solutions only
(0.01, 0.3, and 1.0 mmol/L), we obtained essentially the
same classification by PROP taster status. Mean thresholds
associated with each distribution are shown in Table 1. It
would appear that bitterness intensity scaling of three
PROP solutions around the antimode can provide an
acceptable screening tool for PROP taster status.

Figure 7 Bitterness intensity (left panel) and hedonic ratings

(right panel) as a function of PROP concentration for 541 IX. SCALING METHODOLOGIES
women. Respondents are classed by summed bitterness intensity
ratings for 5 solutions of PROP: bottom 25% (nontasters), There has been some disagreement as to which method for
25–75% (tasters), and top 25% (supertasters). intensity scaling best captures different taste responses to
PROP. Among the more popular choices are those between
0.032, 0.1, 0.32, and 1.0 mol/L (Drewnowski et al., 1997c, 9-point category scales, magnitude estimation, and labeled
1998). The PROP/NaCl ratio was based on the mean of magnitude estimation scales (Drewnowski et al. 1997,
five solutions of PROP and NaCl to salt. In one study with 1998; Green et al., 1993, 1996; Marks et al., 1991) (see also
young women, such procedures yielded 33% nontasters, Chapters 10 and 37). Some researchers have argued that
41% medium tasters, and 26% supertasters, although the category scale anchors, such as “very bitter,” may mean dif-
exact percentages varied somewhat from one study to ferent things to PROP tasters than to nontasters, given that
another. the experience of bitterness can be so different between the
However, there may be no need to use NaCl solutions at two groups (Bartoshuk, 2000; Prutkin et al., 2000).
all. Mean PROP/NaCl ratios and summed PROP intensity The anchored 9-point category scale remains the con-
ratings were highly correlated (r
0.72). Using summed ventional approach in laboratory studies, sensory evalua-
bitterness intensity ratings, we have assigned respondents tion tests, and in much of consumer research
to three taster groups by using a distribution based on four (Drewnowski, 1997a,b,c; Meiselman et al., 1974). Such
equal groups or quartiles. Respondents in the bottom 25% scales are anchored at either end with adjective ratings,
were classified as nontasters, respondents in the 25–75% such as “not at all bitter” and “very” or “extremely bitter.”
group were medium tasters, and respondents in the top In early studies (Kalmus 1958), subjects who indicated
25% were supertasters. Bitterness intensity and hedonic that a PTC filter paper was “very” or “extremely” bitter
response profiles obtained using this classification are were classified as tasters, whereas those who rated the
shown in Figure 7. As expected, bitterness intensity and paper as “not at all” or “slightly” bitter were classified as
nontasters. Later on, respondents were divided into three
Table 1 Percentage Distribution of Respondents (n
541) by taster groups: those who rated bitterness intensity as 1 or 2,
Summed Bitterness Intensity Ratings of 5 Versus 3 PROP those who rated it between 3 and 7, and those who rated it
Solutions 8 or 9 (Bartoshuk et al., 1994; Drewnowski, 2001). That
Summed bitterness intensity classification corresponds almost exactly to the 25:50:25
for 3 solutions percentile split described above.
The second approach, magnitude estimation, relies on
Summed Bottom 25–75% Top Mean ratio scaling. At the beginning of the session, each subject
bitterness 25% 25% threshold
tastes a standard stimulus and is asked to give it a specific
intensity for (solution #)
5 solutions
number, say 15. Perceived intensities of subsequent stim-