THE

SCIENCE OF
PHOTOBIOLOGY
SECOND EDITION
 THE
SCIENCE OF
PHOTOBIOLOGY
SECOND EDITION

Edited by Kendric C. Smith

Stanford University School of Medicine
Stanford, California

PLENUM PRESS. NEW YORK AND LONDON

 Library of Congress Cataloging in Publication Data

The Science of photobiology / edited by Kendric C. Smith.-2nd ed.

p. cm.
Includes bibliographies and index.
ISBN-13: 978-0-306-43059-6 e-lSBN-13: 978-1-4615-8061-4
DOl: 10.1007/978-1-4615-8061-4
1. Photobiology. I. Smith, Kendric C., 1926-
QH515.S37 1989 88-31624
574.19'153-dc19 CIP

© 1989 Plenum Press, New York

A Division of Plenum Publishing Corporation
233 Spring Street, New York, N.Y. 10013

All rights reserved

No part of this book may be reproduced, stored in a retrieval system, or transmitted

in any form or by any means, electronic, mechanical, photocopying, microfilming,
recording, or otherwise, without written permission from the Publisher
Softcover reprint of the hardcover 2nd edition 1989
Preface

The first edition of The Science of Photobiology was published in 1977, and was the first
textbook to cover all of the major areas of photobiology. The science of photobiology is
currently divided into 14 subspecialty areas by the American Society for Photobiology. In
this edition, however, the topics of phototechnology and spectroscopy have been com-
bined in a new chapter entitled "Photophysics." The other subspecialty areas remain the
same, i.e., Photochemistry, Photosensitization, UV Radiation Effects, Environmental
Photobiology, Photomedicine, Circadian Rhythms, Extraretinal Photoreception, Vision,
Photomorphogenesis, Photomovement, Photosynthesis, and Bioluminescence.
This book has been written as a textbook to introduce the science of photobiology to
advanced undergraduate and graduate students. The chapters are written to provide a
broad overview of each topic. They are designed to contain the amount of information that
might be presented in a one- to two-hour general lecture. The references are not meant to
be exhaustive, but key references are included to give students an entry into the literature.
Frequently a more recent reference that reviews the literature will be cited rather than the
first paper by the author making the original discovery. The chapters are not meant to be a
repository of facts for research workers in the field, but rather are concerned with demon-
strating the importance of each specialty area of photobiology, and documenting its
relevance to current and/or future problems of man.
Although written as a basic text for introductory courses in photobiology and as a
vehicle for encouraging students to enter the field, this book will also be of interest to
scientists outside of the area of photobiology and to interested laypersons, since it is now
becoming more apparent, even to the general public, that light, both natural and artificial,
has important consequences to man other than just as an aid to vision. Photobiology has
come of age as a major scientific discipline.
The reader should be aware of the major sources of literature and information
relevant to the science of photobiology.
For review articles, there are two major sources: Photophysiology, Volumes 1-8,
(A. C. Giese, ed.), Academic Press, New York, covering the years 1964-1973; and
Photochemical and Photobiological Reviews, Volumes 1-7 (K. C. Smith, ed.), Plenum
Press, New York, covering the years 1976-1983.
For research papers, the major source is the international journal Photochemistry and
Photobiology (Pergamon Press, London). Short yearly reviews on many areas of pho-
tobiology appear in the June and December issues of Photochemistry and Photobiology.
This journal was inaugurated in 1962, and is now the official organ of the American
Society for Photobiology.
The American Society for Photobiology was founded in 1972 to (1) promote original
v
vi Preface

research in photobiology, (2) facilitate the integration of different disciplines in the study
of photobiology, (3) promote the dissemination of knowledge of photobiology, and (4)
provide information on the photobiological aspects of national and international problems.
Membership in the society is open to persons who share the stated purpose of the society
and who have educational, research, or practical experience in photobiology or an allied
scientific field.
The name of the society was chosen to encompass both North and South America,
but members from all parts of the world are welcome. The journal is included in the
membership dues. The American Society for Photobiology holds an annual scientific
meeting (usually in June) and publishes frequent newsletters of interest to photobiologists.
Further information may be obtained by writing to the Executive Secretary, American
Society for Photobiology, 8000 Westpark Drive, Suite 400, McLean, VA 22102. A
history of the American Society for Photobiology has been published. (1)
The European Society for Photobiology was founded in 1986, and holds meetings
every other year. The next meeting is scheduled for 1989. The journal for the European
Society is the Journal of Photochemistry and Photobiology (Elsevier, Lausanne). For
information about this society, write to Dr. R. M. Tyrrell, Treasurer ESP, Swiss Institute
for Cancer Research, 1066 Epalinges/Lausanne, Switzerland.
The Association Internationale de Photobiologie sponsors an international congress
on photobiology every four years. The most recent congress was held in Israel in 1988.
The proceedings of these congresses are published and serve as an excellent source of
review articles. At these congresses, the Finsen Medal is awarded for "outstanding
fundamental contributions to photobiology." Two histories of the Association Interna-
tionale de Photobiologie have been published. (2,3)

Kendric C. Smith

REFERENCES

1. K. C. Smith, History of the American Society for Photobiology (ASP) (The first 10 years, and before),
Photochem. Photobiol. 35, 597-614 (1982).
2. D. Vince·Prue and D. O. Hall, International cooperation in photobiology, Comite International de Pho·
tobiologie. A History of CIP, Photochem. Photobiol. 22, 77-82 (1975).
3. L. O. Bjorn, International cooperation in photobiology, Association Internationale de Photobiologie, part 2,
1975-1984, Photochem. Photobiol. 41,497-499 (1985).
Contents

Chapter 1 Photophysics ......................................... . 1

Leonard I. Grossweiner

Chapter 2 Photochemistry.... .................................... 47

Leonard I. Grossweiner and Kendric C. Smith

Chapter 3 Photosensitization 79
John D. Spikes

Chapter 4 UV Radiation Effects: DNA Repair and Mutagenesis ..... 111

Kendric C. Smith

Chapter 5 Environmental Photobiology ........................... 135

Ronald Robberecht

Chapter 6 Photomedicine 155

John H. Epstein

Chapter 7 Circadian Rhythms. . . . . .. .. .. .. . . ... . . . .. . . . . . . . . .. .... 193

Jerry F. Feldman

Chapter 8 Extraretinal Photoreception ............................ 215

Michael Menaker

Chapter 9 Vision................................................. 231

Edward A. Dratz

Chapter 10 Photomorphogenesis 273

Lee H. Pratt and Marie-Michele Cordonnier

Chapter 11 Photomovement. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 305

Pill-Soon Song and Kenneth L. Poff

Chapter 12 Photosynthesis 347

David C. Fork

vii
viii Contents

Chapter 13 Bioluminescence. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 391

John Lee

Index ................................................. 419

 1
Photophysics

1.1. Introduction ...................................................................... I

1.2. The Nature of Light .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.3. Phototechnology .......................................... . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.3.1. Light Sources ............................................................... 5
1.3.1.1. Incandescent Lamps .................................................. 6
1.3.1.2. Arc Lamps .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
I. 3.1. 3. Fluorescent Lamps ................................................... 8
1.3.1.4. Lasers.............................................................. 10
1.3.2. Optical Components. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
I. 3.2. I. Lenses and Mirrors .......................... ... ,..................... 13
1.3.2.2. Optical Fibers ....................................................... 15
1.3.2.3. Filters and Monochromators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
1.3.2.4. Polarizers ........................................................... 18
1.3.3. Photodetectors .............................................................. 20
1.3.4. Radiometry and Actinometry. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
1.3.5. Light-Tissue Interactions ..................................................... 22
1.4. Molecular Structure and Excited States ................................................ 25
1.4.1. Interactions of Atoms with Light ............................................... 25
1.4.2. Molecular Orbitals ........................................................... 26
1.4.3. Quantum Yield, Lifetime, and Relaxation of Excited States ......................... 31
1.4.4. Fluorescence Quenching and Electronic Energy Transfer ............................ 32
1.5. Spectroscopic Measurements. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . 34
I. 5.1. Absorption Spectroscopy ...................................................... 34
1.5.2. Luminescence Spectroscopy ................................................... 35
1.5.2.1. Measurement of Luminescence Spectra. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 36
1.5.2.2. Fluorescence Lifetime . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 38
1.5.2.3. Polarization of Fluorescence. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
1.5.3. Optical Activity ............................................................. 40
1.5.4. Photoacoustic Spectroscopy. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
1.5.5. Action Spectra .............................................................. 42
1.6. References ....................................................................... 44

1.1. INTRODUCTION

Photophysics is concerned with the physical interactions of light and matter. The fIrst part
of this chapter reviews the general nature of light, light sources, and the devices used to

Leonard I. Grossweiner • Physics Department, Illinois Institute of Technology, Chicago,

Illinois 60616.
2 Chapter 1

TABLE 1·1. 51 Unitsa

Base units
Length meter m
Time second s
Mass kilogram kg
Temperature kelvin K
Electric current ampere A
Amount of substance mole mol
Luminous intensity candela cd
Plane angleb radian rad
Solid angleb steradian sr

Some SI-derived units

Area square meter m2
Volume cubic meter m3
Density kilogram per cubic meter kg/m3
Frequency hertz Hz
Speed meters per second mls
Force newton N
Pressure pascal Pa
Workl energy joule J
Power watt W
Quantity of electricity coulomb C
Potential difference volt V
Electric field strength volt per meter VIm
Electric resistance ohm n
Capacitance farad F
Magnetic flux weber Wb
Inductance henry H
Magnetic flux density tesla T
Magnetic field strength ampere per meter AIm
Specific heat capacity joule per kilogram kelvin J/(kg-K)
Thermal conductivity watt per meter kelvin W/(m-K)

a~adiometricunits are given in Table 1-5.

bSupplementary unit.

modify and measure light properties. This is followed by a discussion of the interactions
of light with matter at the atomic and molecular levels. Spectroscopic measurements are
then considered, emphasizing the aspects relevant to photobiology. The international
system of units is used (abbreviated SI from the French "Le System International
d'Unites") (Table 1-1).

1.2. THE NATURE OF LIGHT

By strict definition, "light" is the form of radiant energy perceived by the human
eye. However, the spectrum of sunlight at the surface of the earth includes invisible rays,
adjacent to the visible spectrum, that are absorbed by and affect biological systems (Fig.
1-1). Light is a form of electromagnetic radiation. The nominal wavelengths of visible
light are 400-760 nm. This is a very small region of the overall electromagnetic spectrum,
Photophysics 3

>-
u
z
~IOO
u
it 90
III

~ 80
o
! 70
:::E
360
or
050

III
>
j:
<0:
..J
III 400 500 600 700 800 900 1000 1100
or
WAVELENGTH (nm)

Fig. 1-1. Spectrum of sunlight from the zenith at the surface of the earth. The dips in the red and infrared
regions are due to absorption by water vapor [300-365 nm from R. D. Cadle and E. R. Allen, Science 167, 243-
249 (1970); 650-1100 nm from R. B. Withrow and A. P. Withrow, Chap. 3 in Radiation Biology, Vol. 3 (A.
Hollaender, ed.), McGraw-Hill, New York (1956)]. Spectrum of a 250-W photoflood lamp (color temperature,
3475°K) [From R. B. Withrow and A. P. Withrow, Chap. 3 in Radiation Bio[ogy, Vol. III (A. Hollaender, ed.),
McGraw-Hill, New York (1956)]. Spectrum of average human photopic visual sensitivity. (Adapted from Ref.
1.)

which ranges from short-wavelength "I rays to long-wavelength radio waves (Table 1-2).
The ultraviolet (UV) radiations important in photobiology lie in the wavelength band from
approximately 160 nm, below which oxygen is strongly absorbing, to the threshold of
violet light at 400 nm. Infrared (IR) radiation begins at the long-wavelength limit of red
light at 760 nm. Wavelengths longer than 1 f.Lm are essentially "heat waves," whose
biological effects are initiated by temperature elevation.
Electromagnetic radiation is propagated in the form of waves. A perfectly monochro-

TABLE 1-2. Electromagnetic Spectrum

Spectral Photon
banda Wavelength range energy (eV)
Microwave 100-0.1 em 10- 6 _10- 3
IR-C 1000-3 fLm 0.001-0.4
IR-B 3-1.4 fLm 0.4-0.9
IR-A 1.4-0.76 fLm 0.9-1.6
Visible 760-400 nm 1.6-3.1
UV-A 400-320 nm 3.1-3.9
UV-B 320-280 nm 3.9-4.4
UV-C 280-100 nm 4.4-12.4
Vacuum UV 100-10 nm 12.4-124
X rays 10-0.01 nm 102 -105
'f rays 0.1-0.0001 nm IO L IO?
aIR = infrared radiation; UV = ultraviolet radiation.
4 Chapter 1

matic light wave is characterized by a single wavelength (A) and a single frequency (v).
The wave propagates at a wave velocity (V) that depends on the refractive index of the
medium (n) according to the fundamental relationship:

v = AV = cln (1-1)

where c is the speed of light in vacuum (3.00 x 108 m/s). Monochromatic light is an
idealized concept because a real electromagnetic wave must have a finite "bandwidth."
Light may be considered as essentially monochromatic if the bandwidth is sufficiently
small that it does not affect the interactions of the light with its environment. Poly-
chromatic light contains a distribution of wavelengths, e.g., the spectrum of sunlight in
Fig. 1-1. An electromagnetic wave carries with it coupled electric and magnetic fields,
which are responsible for its interactions with matter. According to the classical elec-
tromagnetic theory of Maxwell, the electric and magnetic fields lie in the plane transverse
to the direction of wave motion. (In optics, this direction is the "ray" and the perpen-
dicular plane is the "wavefront.") The electric and magnetic fields of electromagnetic
radiations are mutually perpendicular, and their amplitudes oscillate with the frequency of
the light. With the advent of quantum optics in the late 19th and early 20th centuries, it
was deduced that certain properties of electromagnetic radiation may be described more
accurately by considering the radiation to be composed of small energy packets or quanta.
The energy content of a light quantum, which is also known as a photon, is related to the
frequency by the equation:

E = hv = he/A (1-2)

where h is Planck's constant (6.63 x 10- 34 J-s). Equation (1-2) specifies the energy
content (J) of a photon at a given frequency (s -1) or wavelength (m). The quantum
energy associated with different wavelengths of electromagnetic radiations are given in
Table 1-2 in units of eV, where 1 eV = 1. 60 X 10 - 19 J. When electromagnetic radiation
interacts with matter, the microscopic transfer of energy is limited by the photon energy,
e.g., X rays can eject strongly bound inner-shell electrons from atoms, while visible light
interacts only with the less strongly bound valence electrons. The macroscopic properties
of electromagnetic radiation may be described by wavelike and quantum concepts. The
wave description is based on the radiant power or "flux" carried by the waves, and the
corresponding quantum description is based on the flow of discrete photons. In general,
the quantum energy distribution determines the types of microscopic processes that may
be initiated by a given electromagnetic radiation, and the radiant power limits the rate at
which a process actually does take place under given conditions.

1.3. PHOTOTECHNOLOGY

A photobiological system consists of a coupled light source and a receiver. Artificial

light sources are generally employed for research because sunlight is too variable for good
reproducibility, although the scaling of biological responses obtained with lamps to those
expected for sunlight can be a difficult problem. Whatever the size of the receiver, the
Photophysics 5

ultimate absorbers are molecular entities referred to as chromophores, which may range in
size from individual molecules (e.g., chlorophyll) to subunits of a macromolecule (e.g.,
aromatic amino acid residues in proteins). This section is primarily concerned with the
methods of light generation, control, and characterization. Much of the research in pho-
tobiology utilizes precision optical and electronic equipment, assembled so as to accom-
plish a specific task. In order to understand and carry out experiments in photobiology, it
is necessary to acquire a basic knowledge of optical instruments and measurements.
The First Law of Photochemistry states that photons must be absorbed before pho-
tochemistry can occur. Therefore, the wavelength distribution is a key determinant of
biological responses to light. An important objective in photobiology involves measuring
and understanding the responses of biological systems to light of different wavelengths.
(The ubiquitous term "system" is used to designate the object of interest, which may
range from a single molecule to the entire earth.) The key role of wavelength has led to the
designation of certain wavelength bands by the Committee on Photobiology of the CIE
(Commission Internationale de l'Eclairage). Ultraviolet radiation of biological importance
has been divided into three subbands (Table 1-2). UV-C is the band from 100 to 200 nm.
All cellular constituents absorb UV-c. However, sunlight at the earth's surface contains
virtually no UV -C because of strong filtering by the stratospheric ozone layer. The UV -B
band ranges from 280 to 320 nm. The nucleic acids and aromatic amino acids are the
major cellular absorbers of UV-B. UV-A ("black light") ranges from 320 to 400 nm.
Biological effects induced by UV -A may involve either direct energy absorption, e.g., in
the long-wavelength absorption tail of proteins and DNA, or photosensitization by endog-
enous or exogenous substances (Chapters 3 and 6). Visible light is the band from 400 to
760 nm, and is responsible for most natural phenomena in photobiology, including
photosynthesis (Chapter 12) and vision (Chapter 9). Notable exceptions are the effects of
UV radiation on human skin, leading to sunburn and vitamin D production (Chapter 6).
The designated bands ofIR radiation are IR-A (760-1400 nm), IR-B (1.4-3 /-Lm), and IR-
C (3-1000 /-Lm). IR-A is sometimes called "near-IR" radiation. Except for the short-
wavelength edge of IR-A, the photon energy of IR radiation is too low for the initiation of
photochemical reactions, and the dominant effect of IR absorption is heating . Water is the
major biological absorber of IR radiation above 900 nm, e.g., absorption by water vapor
accounts for the sharp decreases in the near-IR spectrum of sunlight (Fig. 1-1).

1.3.1. Light Sources(1-4)

The important properties of a light source are the spectral power output (watts of
radiant energy emitted in a given wavelength band), the dimensions of the emitting
surfaces, lifetime, stability, and cost. The spectral power output determines whether a
specific lamp is useful for the intended purposes. The source dimensions should be
matched to the light delivery system and the receiver. Light from a small source, e.g., a
laser, can be delivered to a small surface area with good efficiency. An extended source,
e.g., a fluorescent tube, is useful for illumination of large areas. The stability of a lamp
should be appropriate for the duration of the experiments. Cost evaluation should include
the replacement of lamps as well as the initial cost of the entire system. In general,
photochemical sources are "bright," because they are intended to accelerate processes
that are normally slow or inactive at ambient conditions. It is usually desired to irradiate in
6 Chapter 1

TABLE 1-3. Operating Characteristics of Typical Photochemical Lamps

Range
Output Initial Lamp
Lamp type UV Vis Near-IR power cost lifetime
Tungsten-halogen + + Medium Low Short
Deuterium + Medium High Long
Mercury arc + + High High Long
Xenon arc + + + High High Long
Fluorescent + + + Low Low Long
Lasers + + + High High Varies

a specific spectral region, which may vary from selected monochromatic wavelengths to
broad-band radiation. The radiation must be coupled efficiently to the receiver. The
general properties of light sources used as photochemical lamps are summarized in Table 1-3.

1.3 .1.1. Incandescent Lamps(l)

Incandescent lamps utilize light emission from an electrically heated filament. This
radiation approximates the blackbody spectrum of a hot object. According to the Stefan-
Boltzman law, the total radiant power output of an ideal blackbody increases with the
fourth power of the absolute temperature, e.g., an ideal blackbody emits 92 W /cm2 at
2000K and 466 W/cm2 at 3000K. According to Planck's radiation law, the spectral
distribution of blackbody radiation shifts toward the blue with increasing temperature.
Therefore, a hotter object emits more light at all wavelengths, with an increase in the
relative fraction of UV radiation. The blackbody spectra in Fig. 1-2 illustrate both laws.

1600 0 K

~
Fig. 1-2. Spectral irradiance of blackbody radiation. The ordi-
nate is the spectral power emitted (W/cm2) per unit wavelength
IOOOOK (cm). The total radiation emitted per unit surface area increases
~ with the fourth power of the absolute temperature (Stefan-
Boltzmann law). The product of the absolute temperature and
I04L--L-J__L--L-J__L-~~~
o 2 3 4 5 6 7 8 9 the wavelength of maximum emission is constant (Wien's dis-
WAVELENGTH (fLm) placement law). (Adapted from Ref. I.)
Photophysics 7

"Infrared thennometers" are highly sensitive, noninvasive devices for measuring surface
temperature, based on the intensity of the emitted blackbody radiation. The concept of
, 'color temperature" is used in the lighting industry to relate the spectral output of a lamp
(usually an incandescent lamp) to a blackbody radiator. It is defined as the temperature of
an ideal blackbody that emits approximately the same visible spectral distribution as the
lamp. The color temperature of an incandescent lamp is always lower than the physical
temperature, and it may be shifted with color filters. Tungsten filament lamps can operate
under nitrogen gas at 33600 K color temperature for approximately 10 h, e.g., a photoflood
lamp. The output of a photoflood lamp is "redder" than sunlight, with approximately
0.5% UV and 16% visible radiation (Fig. 1-1). The useful lifetime can be extended to
approximately 100 h by employing a quartz envelope, and by introducing a small quantity
of halogen vapor, which reacts with the black tungsten film deposited on the envelope and
redeposits it back on the filament. The spectral output of a 100-W quartz-tungsten-
halogen lamp is shown in Fig. 1-3. This type of lamp is a useful photochemical source for
the visible region, with the advantages of broad-band emission, and low initial and
replacement costs.

1.3.1.2. Arc Lamps

Arc lamps are probably the most frequently employed type of photochemical source.
A typical arc lamp has a quartz envelope containing a filling gas, and two main electrodes
that carry a high current. In some arc lamps the cathode is heated with a separate filament
("hot cathode" type), and in others the cathode operates at ambient temperature ("cold
cathode" type). The radiation emitted by an arc is characteristic of the filling gas and
consists of strong lines at specific wavelengths superimposed on a continuous radiation
"background" (Fig. 1-3). The mechanism of arc action is complex and has been the
subject of much investigation. (1) Arc lamps require a starting device because the voltage
required to initiate an arc is considerably higher than the nonnal operating voltage. The
arc lamps used as photochemical sources are usually filled with mercury vapor, an inert
gas such as argon, or both. The low-pressure mercury arc is the standard source for nearly

100

1
50 206 W MERCURY LAMP

~t
150 IW
w
I
~
r--J
r---
-1- - - XENON LAMP-I

--
10
~~ - lUI IIA
Fig. 1-3. Spectral irradiance of several -[-

lamps used as photochemical sources. The

ZVl
::!Ci
5
\J~ v f\ I!!l!
F'" ~ /--""'""
ordinate is the power in a I-urn bandwidth ~ ~ 1.0

incident on a l-cm 2 receiver located at 50 cm ~ 2 0.5

V- '"
/,,-100 W QUARTZ HArGEN L(MP
from the lamp. The spectra are for a 200-W ;:itt .L /
J
high-pressure mercury arc lamp (Osram :: E 0.1
It-:
J
'"
u c:
HBO 200Wf2), a 150-W high-pressure w ..' 50
Q. E
xenon arc lamp (Hanovia 901-CI), a 100-W
~ .......,
-DEUTERIUM LAMP
Vl~
0.0 I
A•
...
~
quartz-halogen-tungsten lamp (Oriel 6333),
.:!-0.005
and a 50-W deuterium lamp (Oriel 6316).
(Adapted from 1985 Oriel Catalog, VoL 25,
Light Sources, Monochromators, Detection
0.001
200 300 400 500
0'[\ 600
\
700 800 900
Systems.) WAVELENGTH (nm)
8 Chapter 1

monochromatic 254-nm radiation. It operates at low power, and 90% of the output is at
254 nm, with the remainder distributed among the other primary mercury emission lines at
313,365,405,436,546, and 577 nm. Germicidal lamps (e.g., General Electric Co.) are
hot cathode, low-pressure mercury arcs (Fig. 1-4). The most intense, general-purpose arc
lamps available for research are cold cathode, high-pressure xenon and xenon-mercury
arcs. They are made with closely spaced electrodes in a quartz envelope, and operate at
50-80 times atmospheric pressure to permit a high current density. The emission lines of
a high-pressure arc are quite broad compared to low-pressure arcs, and there is a strong
continuous background, providing useful intensities at all wavelengths transmitted by
quartz (Fig. 1-3). High-pressure arcs are relatively expensive, requiring a regulated, high-
current power supply, an air-cooled or water-cooled lamp housing, and an ignition device.
They are long-lived (typically 1000 h) with moderately high replacement cost. Arc lamps
with quartz envelopes generate significant amounts of toxic ozone, and must be vented or
operated in a well-ventilated area.

1.3 .1.3. Fluorescent Lamps( 1 ,2)

Fluorescent lamps are low-pressure mercury arcs, in which the 254-nm emission is
absorbed by a phosphor coating on the inside of the glass envelope, and is re-emitted as a
longer wavelength fluorescence (Section 1.4). The output consists of one or several
overlapping broad bands with superimposed "spikes" at the mercury emission lines. The
spectral range is determined by the chemical composition of the phosphor coating and the
glass envelope (Fig. 1-5). About 20% of the electrical energy is converted to useful visible
or near-UV radiations. Fluorescent lamps are inexpensive, low-power sources, useful for
long-term irradiation of large areas. Many special types of fluorescent lamps are avail-
able. The phosphor in a fluorescent "sunlamp" emits from 260 to 400 nm with the peak
at 320 nm. The envelope is made of a glass that limits the output to wavelengths greater
than 320 nm. "Black-light" fluorescent lamps emit from 325 to 440 nm, with the peak
near 360 nm (Fig. 1-6). "Daylight" fluorescent lamps give approximately the same
visual response as sunlight, without the UV and IR radiations. Fluorescent lamps from

~ 30
a. 275.9
2
~ 25
.....
E
<: Fig. 1-4. Spectrum of a "ger-
Q 20 -0.3
micidal" low-pressure mercury arc
,-0.15
15.5 (General Electric Co. G30T8) after
100 h of operation. The numbers ad-
jacent to each narrow line indicate
8.9
the radiant power output, in units
of mW per 10 nm bandwidth per W
2.0
of input power. Approximately 85%

350 400 450 500 550

n 600 650 700 750
of the total radiant power is emitted
in the 254-nm "resonance line."
WAVELENGTH (nm) (Courtesy of General Electric Co.)
Photophysics 9

80

>-
tl) GREEN
....II:z 60
....
\oJ
>
~ 40
....-'a::

Fig. 1-5. Spectral output of various colored fluores-

cent lamps of equal wattage. The mercury spectral
lines are omitted. (Courtesy of Westinghouse Electric
Corp.) WAVELENGTH (nmJ

different suppliers with similar designations are usually not identical, e.g., black light.
Furthermore, a specific lamp desigination does not guarantee that the phosphor or enve-
lope will not be changed by a manufacturer. Readers who use fluorescent lamps for
research are advised to measure the emission spectrum or obtain the current specifications
from the manufacturer.

1.5

~ 1.0
/ \ '"\- f--
~
" BLB

...'"z BL
... '/ \
...> '/ \
~ 1\
~ 0.5

\ \
..J \ \ r
y \h II
o
300
''; K
400
'- 500
I n 600 700

WAVelENGTH - (nm)

Fig. 1-6. Spectral output of 40-W Tl2 black-light (BL) and black-light-blue (BLB) fluorescent lamps after 100
h of operation. (Courtesy of General Electric Co.)
10 Chapter 1

1.3.1.4. Lasers(5,6)
A laser is a special type of light source that emits intense, monochromatic radiation
in a highly collimated beam. The fIrst practical laser was constructed in 1960 by Maiman,
utilizing a ruby rod (chromium ions in a matrix of crystalline aluminum oxide), which was
followed rapidly by a remarkable diversity of new laser designs. A laser is described by its
"active medium," i.e., the chemical substances and their physical state in which laser
action takes place. The more common types of lasers include solid state (e.g., ruby, Nd-
yttrium-aluminum garnet [Nd-YAG]) , gas (e.g., He-Ne), molecular (e.g., CO 2 ), gas ion
(e.g., Ar, Kr, Xe), metal vapor (e.g., He-Cd, He-Cu, He-Au), excimer (e.g., ArF, KrF,
XeCl), chemical (e.g., HF, HCI), dye (e.g., fluorescein, rhodamine 6G), and diode (e.g.,
GaAs, GaInAs). Laser action orginates in the fundamental principle of atomic physics
which states that the internal energies of atoms are quantized in discrete energy levels
(Section 1.4.1). When a substance is in thermal equilibrium with its surroundings at
ordinary temperatures, almost all atoms of the substance are in their lowest energy state.
Atoms can be elevated to higher energy states by heating, carrying out a chemical
reaction, establishing an electrical discharge, or irradiating with light at wavelengths
absorbed by the substance. In laser terminology, this process is known as "pumping."
(The reader should be warned that laser science is replete with jargon, some of which must
be introduced if this section is to be of practical use.) It is diffIcult to pump more atoms to
higher energy states than remain in lower energy states, e.g., heat alone cannot accom-
plish this population inversion. However, population inversion can be achieved in special
circumstances by proper selection of the active medium and the operating conditions,
which is the basis oflaser action. When population inversion occurs, the absorption of one
photon at a wavelength corresponding to the energy difference between an upper state and
a lower state can initiate the emission of more than one photon. This process is known as
stimulated emission, leading to the acronym LASER, for' 'light amplification by stimu- .
lated emission of radiation. " The photons emitted by stimulated emission have the same
wavelength, polarization (the direction ofthe electric fIeld, Section 1.3.2.4), and move in
the same direction as the initiating photons. Laser light is coherent, i.e., the light waves
move together in space and time. The properties of very narrow bandwidth, collimation,
and coherence make it possible to transmit laser light over long distances with very little
spreading, and to focus a laser beam to a very small, extremely intense spot. Although the
physical appearance of various types of lasers may be quite different, all lasers employ an
optical cavity to contain the active medium and promote amplifIcation of those light
waves propagating along the cavity axis. The helium-neon laser, depicted in Fig. 1-7, was
the first practical gas laser, constructed in 1961 by Javan, Bennett, and Harriott. The
optical cavity is a glass or metal cylinder, containing a mixture of helium and neon at low
pressure, with mirrors at each end. One mirror is highly reflecting at the emission
wavelength, and the other has a low transmittance to permit extraction of laser light after
many "round trips" in the cavity. The helium-neon laser is pumped by a high-voltage
discharge, and may be designed to emit a single line at 632.8 nm. Table 1-4 summarizes
the operating characteristics of some lasers used in photobiology.
The two most important properties of a laser are the emission wavelengths and the
time span over which laser action occurs. Most lasers emit strong lines at one or a few
wavelengths. A laser can be tuned to emit single wavelengths by locating a wavelength
Photophysics 11

HV

["'-N' -
w

'--------W
'~
T
C M2

Fig. 1-7. Schematic diagram of a helium-neon laser. The cylindrical cavity (C) is filled with a mixture of
helium and neon at low pressure. A gaseous discharge established by the high-voltage power supply (HV)
excites helium atoms to metastable (long-lived) states, which transfer their energy to neon atoms by collisions.
This selective popUlation of neon-excited states leads to population inversion. Laser action is initiated by
radiative transitions of excited neon atoms to lower energy excited states. The photons emitted parallel to the
longitudinal axis are trapped in the optical cavity formed by a totally reflecting mirror (M 1) and a partially
reflecting mirror (M 2). The longitudinal light waves corresponding to the neon emission wavelengths are
amplified by stimulated emission. The strongest visible line is at 632.8 nm. The glass windows (W) of the cavity
are tilted at the Brewster angle of approximately 57°, which tunes the laser to a single transverse mode parallel to
the plane of incidence, and polarizes the emitted laser radiation.

control element in the optical cavity, e.g., a prism. The dye laser is a notable exception, in
which the emission wavelength can be tuned over the fluorescence band of a dye, typ-
ically 100 nm (Section 1.5.2). A wider range of wavelengths is available with inter-
changeable dyes. The wavelength (or frequency) generated in normal laser operation is
known as the "fundamental." Wavelengths shorter than the fundamental can be obtained
by "frequency doubling." This is accomplished with useful efficiency (typically 5 to
30%) in high-power lasers by passing the laser light through a crystal with "nonlinear"

TABLE 1-4. Properties of Lasers Useful in Photobiology

Typical operation
Principal Average
Type wavelengths (run) CW Pulsed power"
Helium-neon 633, 1153 + Very low
Krypton ion 458, 568, 647 + Medium
Argon ion 458, 488, SIS + High
Nitrogen 337 + Low
Copper vapor 510,578 + High
Gold vapor 628 + High
Excimer (rare gas 193, 248, 308, 351 + High
halide)
Ruby 694 b + Low
Nd:YAG lO64e + High
Dye 260-900 + + High
Carbon dioxide 10,600 + High
Semiconductor Red, near-IR + Very low
"Power levels vary considerably with laser construction and applications. The ranges are based on <0.1 W (very low), 0.1-1 W
(low), 1-10 W (medium), 10-100 W (high).
b347 nm with frequency doubling.
c532 nm with frequency doubling, 355 nm with frequency tripling, 266 nm with frequency quadrupling.
12 Chapter 1

dielectric properties, e.g., potassium dihydrogen phosphate (KDP). Halving the wave-
length is referred to as "doubling," dividing the wavelength by three is "tripling," etc.
The helium-neon laser operates in the "continuous wave" (CW) mode, i.e., light is
emitted continuously during pumping. (The output of a CW laser has a substructure of
very short pulses, induced by standing waves generated in the optical cavity, which do not
affect its usual applications as a continuous light source.) The power output of a CW laser
is not remarkable, e.g., a lO-W laser corresponds to intermediate power (Table 1-4).
However, good collimation makes it possible to focus a laser beam to a very small area,
providing high irradiance (Table 1-5), e.g., the irradiance of a l-cm-diameter, lO-W laser
beam focused to 1 mm is 1.3 kW/cm2 . By comparison, the total visible light irradiance in
the collimated beam of a 150-W high-pressure xenon arc is approximately 0.2 mW/cm 2 .
Since laser light is highly monochromatic, the spectral irradiance (WI m2 -nm) exceeds
that of strong arc lamps by many orders of magnitude. Pulsed operation is the other
general type of laser action. It is initiated by pulsed pumping, e.g., by pumping a Nd-
YAG laser with a gas-filled flash lamp. If pulsed pumping is carried out repetitively, the
output consists of a sequence of relatively short pulses separated by longer time intervals.
Some lasers can operate only in CW mode, others only in pulsed mode, and some in either
mode, depending on the construction and type of pumping (Table 1-4).
There has been a continuing effort to develop light sources with the shortest possible

TABLE 1-5. Radiometric and Photometric Units

Radiometric
quantitya Concept SI unit Photometric equivalentb
Energy (Q) Quantity of light J lm-sc
Flux (F) Power W 1m
Energy density Energy content per unit volume J/m 3 Im/m 3
(W)
Irradiance (E) Flux per unit area incident on a smaIl- W/m 2 Illuminanced lm/m 2
plane surface
Excitance (M) Flux per unit area leaving a small-plane W/m 2 Im/m 2
surface
Intensity (I) Flux emitted into a unit solid angie Wlsr Candlepower" cd (lm/sr)
Radiance (L) Flux in a given direction per unit solid W/m2-sr Luminancei' cd/m 2
angle and per unit area normal to the
direction of propagation
Exposure (Q) Energy per unit area incident on one side J/m 2 lx-s
of a small-plane area
aThese terms should be prefixed by "radiant" to avoid confusion with the photometric units, e.g., radiant energy.
bThese terms should be prefIXed by "luminous" to avoid confusion with the radiometric units, e.g., luminous flux.
eThe lumen (1m) is the basic unit of luminous flux. It is the luminous flux emitted into a unit solid angle by a uniform point
source whose luminous intensity is I candela (cd). The radiant flux in the lumen is weighted according to its efficiency in
producing a visual response. One lumen is equivalent to approximately 682 W at 555 nm and to fewer watts at other
wavelengths, where the phototopic response is lower.
dIlluminance is the basic unit of illumination. An illuminance of I lumen/m2 is also called I lux (Ix). In the British system, I
lumen/ft2 is I foot-candle.
eThe candela is the SI unit of luminous intensity. It is based on the luminous intensity of a blackbody at the melting temperature
of platinum. One lumen of luminous power per steradian corresponds to I cd. The luminous of a point source intensity in
candela equals the candlepower.
.!Luminance is luminous intensity per unit projected area. One candela/m2 square meter is also known as I "nit," and I ed/em2 is
I "stilb."
Photo physics 13

pulse durations in order to study fast processes in photochemistry and photobiology, e. g.,
for flash photolysis (Section 1.5.1) and fluorescence lifetime measurements (Section
1.5.2.2). Shortly after the invention of lasers, it was discovered that the pulse duration of
ruby and Nd-YAG lasers is reduced from the order of 10- 3 to 10- 8 S by blocking the
optical path until the pumping light flash is nearly complete. This technique is known as
"Q switching" and can be accomplished with a fast mechanical or electro-optical shutter.
"Mode-locking" was a subsequent development, in which the output consists of a pulse
"train" of approximately 10 ns overall duration, with a substructure of pulses ranging
from less than 1 ps to several ns duration. Single mode-locked pulses of about 10 ps
duration have been obtained by amplifying one pulse of the train. (A laser "amplifier" is
another laser of the same type, pumped by the emission from the laser "oscillator.") The
peak power is the energy in a single pulse divided by the average pulse duration, and the
average power is the energy in a single pulse divided by the time between the pulses. The
peak power is important for interactions of the laser light with matter that occur in time
durations comparable to that of the laser pulse. The average power is applicable for
relatively slow processes that do not respond to the individual pulses. Some applications
of lasers in medicine will be described in Section 1.3.5.

1.3.2. Optical Components

Equation (1-1) indicates that the velocity of light in a medium equals c/ n, where
"empty space" has a refractive index of unity. When propagating monochromatic light
encounters a change of refractive index, e.g., on going from air to polished glass, the
incident beam is partially reflected at the interface (specular reflection) and partially
refracted (bent) on entering the new medium. (Diffuse reflection at a rough surface is
considered in Section 1.3.5.) Polychromatic light introduces the additional complication
that the different wavelengths are propagated at different speeds in a uniform material.
This effect is referred to as dispersion, and is responsible for the ability of a prism to
separate "white" light into its component wavelengths. Diffraction phenomena occur
when light passes through apertures with dimensions comparable to the wavelength, e.g.,
diffraction limits the ability of a light microscope to image objects smaller than about 0.2
f.lm. (However, an electron microscope has a much smaller effective wavelength and can
image objects of atomic dimensions.) Inteiference is a related phenomenon, in which light
waves combine in a regular pattern of intensity variations or "fringes." The propagation
of light through a material with directional properties can affect its polarization, i.e., the
orientation of the electric field in the plane of the wave front. Polarizing devices are used
to polarize ordinary light and alter the properties of polarized light. With this brief
introduction into the language of classical optics, we proceed to consider the optical
elements most widely used in photobiology. Additional information about "physical"
optics and the optical properties of materials is available in standard texts. (7-9)

1.3.2.1. Lenses and Mirrors

Lenses are used to relay light between optical elements. The most important proper-
ties of a lens or lens system are the focal length, aperture, and wavelength transmission.
The focal length is the distance from the lens at which collimated rays are brought to
14 Chapter 1

t~
:0
,
I
,r-- P
F
~..
{oj
I
I
I

q
I
I
'
I
' Fig. 1-8. Ray diagrams for spherical thin
lenses. In ray tracing, a ray parallel to the lens
axis is refracted through a focal point (F and F');
a ray diverging from a focal point is refracted as
paraIlellight; a ray passing through the lens cen-
ter is not deflected. The convergence of rays
from a point on the object (0) produces an
equivalent point on the real image. (a) An object
located outside the front focal plane of a convex
lens (F) gives a real, inverted image. (b) An
I'I F' object located inside the front focal plane of a
I convex lens gives a virtual, upright image. (An
I image is virtual if the rays appear to diverge
I
from image points not actually reached by the
I---q---I
( b)
light.) (c) An object located outside the front
focal plane of a concave lens gives an upright,
virtual image. The distance of the image plane
from the lens (q) is related to the distance of the
object plane to the lens (p) and the focal length
of the lens (F) by: lip + l/q = lIF, and the
10 F: ' I
I I I linear magnification is q/p. The distances q and
i : p are taken as positive for real objects and im-
: ~q---l ages and negative for virtual objects and images.
I--- P -----l The linear magnification is negative for an in-
(c) verted image and positive for an upright image.

convergence. The fundamentals of image formation by thin lenses are described in Fig.
1-8. A "converging" lens can be used to collect light from a small source (a "condens-
er"), to match a collimated light beam to a receiver ("relay" lens), and to focus a light
beam. Diverging lenses are used as intermediate elements in compound lenses, e.g., a
telephoto lens, and to correct the human eye for near-sightedness. The light-gathering
ability of a lens is limited by its aperture. The relative aperture or ''f number" is the ratio
of the focal length to the lens diameter. Light-gathering effectiveness is inversely propor-
tional to the square oftheJnumber. Thus, anJ/ 1.0 lens collects four times more light than
JI2.0. The resolving power of a lens is the ability to separate two closely spaced objects.
For a magnifying lens (e.g., a microscope objective), the resolving power equals
A/2no sin a, where A is the wavelength, no is the refractive index of the medium from
which the light enters the lens, and a is the half-angle of the acceptance cone. (The
acceptance cone of an optical fiber is shown in Fig. 1-9.) The wavelength transmissivity
of the lens material should be given special attention because ordinary optical glass
transmits only from about 370 nm to 2.5 J.Lm, and cannot be used in the UV and far-IR
regions. The short-wavelength limit is lowered to 330 nm with Pyrex glass, and below
220 nm with fused silica or crystal quartz. Lenses for use in the IR region can be ground
from ionic crystals, e.g., NaCI (0.25-16 J.Lm). The IR "heat waves" generated by light
sources are usually undesirable, and can be eliminated with heat absorbing glass and a
water filter. Although a multielement lens system has superior imaging properties com-
pared to a simple lens, reflections at the air-glass interfaces lead to significant losses of
light, which are the order of 25-30% overall. Front-surfaced mirrors provide an alter-
Photophysics 15

Fig. 1-9. Propagation of light by a cylindrical optical fiber. A ray striking the input face within the maximum
acceptance angle 0. is totally reflected at the interfaces between the fiber (refractive index n) and the cladding
(refractive index n') because the angle of incidence exceeds the critical angle 8 c ' where sin 8 c = n'ln. This
process is repeated until the ray exits at the opposite face. The numerical aperture (N.A.) is defined as no sin 0.,
where no is the refractive index of the external medium (unity for air). It is related to the fiber properties by:
N.A. = v'(n2 - n'2)lno' A typical optical fiber with a fused silica core and plastic cladding in air has a N.A. of
0.22, corresponding to a lens '1 number" of approximately 2.

native method for collimating and focusing light with lower losses than obtained with
glass lenses. The reflectivity of an evaporated aluminum mirror exceeds 90% from 240 nm
to the far-IR. Unlike lenses, mirrors are not subject to chromatic aberration (dependence of
focal length on wavelength) because specular reflection does not depend on the refractive
index of the surface. Focusing mirrors are fabricated with spherical, elliptical, parabolic,
and aspheric surfaces, depending on the application.

1.3.2.2. Optical Fibers(lO)

Optical fibers are efficient devices for transporting light to awkward-to-reach loca-
tions and over long distances. An optical fiber consists of a glass or fused-silica core,
covered by a lower refractive index cladding. Light rays incident at one end are transmit-
ted down the fiber by total internal reflection at the interfaces between the core and
cladding (Fig. 1-9). The light-gathering power of a flat-ended fiber depends on the
numerical aperture and compares quite favorably with a fast (Le., low f number) lens.
Glass fibers transmit from 340 to 1300 nm. Fused-silica fibers transmit down to 220 nm
but are about four times more costly. Fiber bundles or "light guides" can transmit higher
optical powers than single fibers and have greater mechanical strength. Useful optical
devices can be fabricated by utilizing fibers to transmit several optical signals. A bifur-
cated cable or "Y guide" can act as an optoelectronic sensor of surface reflection or
motion. The incident light is transmitted down one arm of the Y and out of the common
leg, and the reflected light is collected by the common leg and measured at the other arm
of the Y. This principle is used in the "oximeter," a device that measures the oxygen
content of blood in vivo, by detecting the diffusely scattered light from hemoglobin at
several wavelengths.(lJ) An ordinary light guide cannot transmit an image because the
spatial arrangement of the fibers at the input is not retained at the exit. Image transmission
has been accomplished with "coherent bundles," where the relative positions of the
individual fibers at the two ends are in a simple relationship. Medical endoscopy is a
major biological application of coherent fiber bundles.

1.3.2.3. Filters and M onochromators(7-9)

Filters and monochromators are used to restrict radiation to selected wavelengths.
Glass filters are prepared from polished optical glass, by incorporating additives that limit
16 Chapter 1

Fig. 1-10. Percent transmission of some

80 high-transmission combinations of glass
filters. (1) Corning C.S. No.7-54; (2)
Corning C.S. No.7-54 + C.S. No. 0-54;
60 (3) Wratten 39 + Corning C.S. No.
0-52; (4) Corning C.S. No.5-56 + C.S.
"!oT No. 3-74; (5) Schott BG-14 + Corning
C.S. No. 4-97 + C.S. No. 3-72; (6)
CorningC.S. No. 4-97 + C.S. No. 3-71;
(7) Schott BG-38 + Kodak infrared cut-
off, No. 301 + Corning C.S. No. 3-69;
(8) Kodak 301 + Corning C.S. No.
20

3-66; (9) Schott RG630; (10) Schott

RG695; (11) Kodak 39 + Schott RG695;
(12) Corning C.S. No.5-56 + Schott
RG695. [Adapted from W. D. Herk-
stroeter, A series of sharp-cut and band-
pass glass filters for the range 252-800 nanometers. Mol. Photochem. 4, 551-557 (1972).] The Corning glass
numbers in the original reference have been replaced by the currently used "color specification" (C.S.)
numbers.

the transmission to certain bands (bandpass fIlters) or above certain wavelengths (short-
wavelength cutoff filters). Hundreds of glass filter types are available from the major
suppliers. Filter combinations can be employed to obtain high transmission in selected
wavelength bands (Fig. 1-10). Glass fIlters are relatively inexpensive and are available in
large sizes. Interference filters are the other major type of filter. The simplest type of
interference fIlter consists of two parallel, partially reflecting mirrors (usually on a glass
base) separated by a layer of dielectric materiaL This arrangement preferentially transmits
light of wavelength: A = 2dn, where n is the refractive index of the dielectric layer and d
is its thickness. (The incident ray and a ray that reflects twice before emerging add
constructively, and other wavelengths are attenuated by destructive interference.) Inter-
ference filters are more expensive than glass filters, but they can be made to much more
exact specifications. However, a given design is exact only for perpendicular collimated
rays, and off-axis rays are transmitted with a wavelength shift, which may be employed to
tune the fIlter over a small bandwidth.
A monochromator is a precision optical device that separates broad-band light into its
composite wavelengths. The standard dispersing elements in monochromators are glass or
quartz prisms and diffraction gratings. A prism disperses polychromatic light because the
velocity of light in a transparent material depends on the wavelength. A longer wavelenth
ray moves more rapidly than a shorter wavelength ray, and therefore it takes a shorter path
through the prism, and exits at a smaller angle to the prism face. The wavelengths are
selected by rotating the prism (or a mirror) to pass a narrow bandwidth through an exit slit.
A diffraction grating disperses polychromatic light by multiple interference of the light
waves reflected from closely spaced grooves in a metal coating (usually aluminum) on a
glass base. Figure 1-11 shows the arrangement in a small diffraction grating mono-
chromator designed for high optical throughput (light gathering power). The dispersion of
a monochromator indicates its ability to spread the wavelengths of light. The angular
Photo physics 17

Fig. 1-11. Optical paths in a small monochromator with symmetrical, in-line optics. The incident light is
focused on the entrance slit (SI)' deflected by a 45° mirror (M I ), collected by a collimating mirror (M 2 ), and
directed to the diffraction grating (G). The different wavelengths are reflected at different angles by the grating,
which are imaged by mirror (M 3 ) and deflected to the exit slit (S2) by a 45° mirror (~). The exiting bandwidth
equals the product of the inverse linear dispersion of the monochromator (nm/mm) and the exit slit width (mm).
The grating is rotated to select different wavelengths.

dispersion ((3) is defined as the angular difference (in radians) between two rays leaving
the prism or diffraction grating that differ in wavelength by 1 nm. The linear dispersion
(m) is the physical width at the plane of the exit slit occupied by a i-nm bandwidth. They
are related by m = (3F, where F is the focal length of the lens or mirror (M3 in Fig. 1-11)
that collects the light from the dispersing element and focuses it on the exit slit. The
"inverse linear dispersion" is usually specified in units of nm/mm at the exit slit. Prism
monochromators were used in most laboratory instruments until the 1960s because high-
quality diffraction gratings must be ruled with a grating engine and are quite expensive.
Prism monochromators have the disadvantage that the dispersion is nonuniform and must
be calculated at each wavelength from data provided by the manufacturer. Grating mono-
chromators are designed to have linear dispersion, i.e., the bandwidth for a given exit slit
width is the same at all wavelengths. The development of high-quality plastic copies of
ruled gratings (replica gratings) has led to the dominance of gratings in optical instru-
ments. Holographic gratings are a recent advance, in which interference fringes from two
laser beams are recorded on a photosensitive surface, which is then selectively etched.
Holographic gratings are more expensive than replica gratings, but they have the advan-
tages of higher efficiency (the ratio of the diffracted energy to the incident energy at a
given wavelength), less "stray light" (the background of unwanted wavelengths in the
exit beam), and complete absence of "ghosts" (spurious spectral lines). In addition to
linear dispersion, grating monochromators have higher light-gathering power than prism
instruments of comparable cost, and the convenience of interchangeable gratings to cover
a wide wavelength range.
The key determinants in monochromator selection are the wavelength range, the
18 Chapter 1

dispersion, the throughput, and the scattered light. Wavelength selection is the most
frequent use of monochromators in optical instruments, e.g., in absorption and lumines-
cence spectrophotometers (Section 1.5), and irradiation systems. The desirable charac-
teristics for such applications are high throughput, small size, and moderate cost, which
are obtained at the expense of high dispersion and high resolving power. Diffraction-
grating theory shows that multiple "orders" of spectra are diffracted into the same
directions, such that the first-order wavelength)" coincides with the second-order wave-
length ),,/2, the third-order wavelength ),,/3, etc. (The complication of multiple orders does
not occur with prism monochromators.) A high fraction of the output energy can be
localized in a given order (usually the first order) by ruling the grooves at an angle with
respect to the grating surface or "blazing." A diffraction grating is most efficient at the
blaze wavelength (typically 50-70% for small gratings), which decreases by a factor of
0.7 at 2/3 and 3/2 times the blaze wavelength. The overlapping spectral orders at the same
angle can be eliminated with short-wavelength cutoff "blocking" filters. Most mono-
chromator applications in photobiology can be handled within the wavelength range of
200-1000 nm, which requires at least two gratings-one blazed in the UV (e.g., 250 nm)
and one blazed in the visible region (e.g., 500 nm). The dispersion and the cost of a
diffraction grating increase with the number of linesl mm. A replica or holographic grating
with 1200 lines/mm provides a reverse linear dispersion in the order of 5 nm/mm in a
typical small, high-aperture instrument, which corresponds to useful bandwidths from
0.05 to 20 nm with variable slits.
The efficient use of monochromators is described in standard texts. (12) The blaze of
the grating should be optimized for the working wavelength range. Ideally, the mono-
chromator and light source should have the same f number for the most efficient coupling.
Energy is lost and stray light increases if the grating is overfilled (i.e., the fnumber of the
source is lower than that of the monochromator), and the resolving power is lower if the
grating is underfilled. Similarly, the exit beam should underfill the detector area to avoid
loss of light. A relay lens or mirror can be used with very small detectors, to focus the
diverging exit cone from the monochromator on the detector. Appropriate blocking filters
should be used if the detector is sensitive to higher order spectra. For symmetrical optical
systems, the entrance and exit slit widths should be the same. For a broad-band source,
the throughput increases with the square of the entrance slit width, although a larger slit
width also increases the bandwidth. The full entrance slit height should be illuminated for
maximum throughput. The signal-to-noise ratio is limited by stray light. The stray light in
a typical small, high-aperture-grating monochromator is 0.1%. The use of a "double
monochromator" (two monochromators in tandem) is the most effective way of reducing
stray light, and may be required for spectral resolution of broad-band, low-intensity
signals.

1.3.2.4. Poiarizers(7-9)
Spectroscopic measurements with polarized light provide information about the di-
rectional properties of molecules, e.g., optical activity, dichroism, and fluorescence
polarization (Section 1.5). The electric field of electromagnetic radiation is located in a
plane perpendicular to the ray and oscillates with the frequency of the light. The magni-
tude and direction of the electric field are conveniently represented by a vector, depicted
Photophysics 19

as an "arrow" oflength proportional to the strength of the electric field and pointing in
the direction of the field. (Since the electric field is moving through space and oscillating
in time, a two-headed vector is used, whose length indicates the amplitude of the oscilla-
tion.) The magnetic vector is perpendicular to the electric vector in the same plane, and
oscillates with the same phase. (The phase of an oscillation of frequency f is 2-rrjt, where t
is the time measured from an arbitrary zero. Two oscillations are "in phase" if they reach
their minimum and maximum amplitudes at the same time.) In "ordinary" light, the
electric vector has a random orientation and the light wave has no inherent directionality
perpendicular to the ray. (Alternatively, the electric vector can be resolved into two
perpendicular vibrations, of the same amplitudes and without phase coherence.) This
picture of ordinary light (0) is shown in Fig. 1-12a. In linearly polarized light the
direction of the electric vector is fixed. One way of producing linearly polarized light is to
pass ordinary light through a material with a different absorption for the two polarization
components of ordinary light. This property, referred to as dichroism, is exhibited by

o P LP A LP'

(0)

100%5

50% P
\
\
\

(b)

Fig. 1-12. (a) Polarization of light by a thin-film polarizer. The electric vectors in ordinary light (0) have a
random orientation, as depicted by their equal projections in two perpendicular directions. The polarizer (P) is a
dichroic material, with its maximum transmission for the polarization depicted by the parallel lines. The
transmitted light is linearly polarized (LP). If the linearly polarized light is passed through another polarizer (the
"analyzer"), the transmission is maximum if the two polarizers are aligned and minimum if they are
"crossed." In the case shown, the analyzer is set at an intermediate angle, and the resultant linearly polarized
light is attenuated (LP'). (b) Polarization by reflection and refraction. The electric vectors of the incident light
are resolved into equal parallel (P) and perpendicular (S) components, relative to the plane containing the
incoming ray and the normal direction to the glass plate (dashed line), which is referred to as the "plane of
incidence." For incidence at Brewster's angle (Op), the reflected ray is 100% polarized in the S direction, and
the transmitted light is partially linearly polarized. Brewster's angle is given by tan Op = no/n, where no is the
refractive index of the external medium and n is the refractive index of the glass. It is approximately 57° for
optical glass in air.
20 Chapter 1

some materials, e.g., tourmaline, and by plastic films containing aligned organic mole-
cules, e.g., a Polaroid film. The polarization of ordinary light by a dichroic type of
polarizer is shown in Fig. 1-12a. The linearly polarized light is detected by passing it
through a second polarizer, designated by "analyzer." The intensity of the light transmit-
ted by the analyzer varies from zero, when the two polarizers are "crossed," to a
maximum when they are aligned. Linearly polarized light is also produced by passing
ordinary light through an optically anisotropic crystal with the property of birefringence or
"double refraction," e.g., crystal quartz or calcite. "Nicol" and "Glan Taylor" polar-
izers are calcite crystals cut in such a way that beams of each polarization are separated.
The passage of linearly polarized light through a thin plate cut from a birefringent crystal
leads to interference effects that alter the polarization properties. A "half-wave" plate
rotates the direction of linearly polarized light by an angle that depends on the relative
directions of the electric vector and the crystal axes. Passing linearly polarized light
through a "quarter-wave" plate at the correct angle generates circuJarly polarized light,
in which the electric vector rotates rapidly in its plane. The polarization of ordinary light
after reflection by a glass plate was discovered by Malus in 1808. At a specific angle of
incidence, "Brewster's angle," the reflected light is linearly polarized and the transmitted
component is partially polarized (Fig. 1-12b). Brewster's angle is approximately 57° for
optical glass in air. The windows on a laser cavity are usually inclined at Brewster's
angle, to selectively amplify a single transverse mode and obtain a smooth beam cross-
section (Fig. 1-7).

1.3.3. Photodetectors
The two basic types of photodetectors are thermal detectors and quantum detectors.
Thermal detectors respond to radiant power and are independent of the wavelength.
Thermopiles and pyroelectric detectors are of this type. An illuminated thermopile gener-
ates a low voltage, induced by the absorption of radiant energy in a series array of
dissimilar metal junctions (thermocouples). Thermopiles are very accurate in all spectral
regions, but have low sensitivity and a slow time response. A "pyroelectric" detector is
fabricated from a crystal in which a surface voltage is generated while its temperature is
changing. (Pyroelectricity is a property of certain "ferroelectric" crystals, e.g., lithium
niobate, with an inherent dielectric polarization or charge separation). Pyroelectric detec-
tors have a faster time response than thermopiles, with equally wide spectral response and
comparable sensitivity. However, they respond only to pulsed or chopped radiation,
necessitating an AC measurement system. Since thermal detectors respond to average
optical power, the readings can be converted to photons conveniently only for monochro-
matic radiation.
Quantum detectors respond to the average photon absorption rate, usually with a
different sensitivity at each wavelength. (A dye fluorescence "quantum counter" is an
exception; Section 1.5.2.2.) The standard types of quantum detectors are vacuum pho-
todiodes, photomultipliers, and solid state photodetectors. A vacuum photodiode consists
of an evacuated glass bulb, containing a photocathode that emits electrons when exposed
to light and an anode that collects the emitted photoelectrons. In a photomultiplier, a set of
auxiliary "dynodes" are located between the photocathode and anode. Dynodes have the
property of "secondary electron emission," i.e., several electrons are emitted by a
dynode for each electron absorbed. The effect is cumulative, and the output current can be
Photophysics 21

TABLE 1-6. Properties of Photodetectors

Spectral Time response D*
Type range (11m) (l1s) Typical gaina Cv'Hz-cm/W)b
Vacuum photodiode 0.2-0.7 < 0.01 0.04 A/W 1011
Photomultiplier 0.2-0.7 < 0.Ql 2500 A/W 10 14
0.4-1.1 < 0.01 40A/W
Thennopile 0.2-50 105 lOV/W 108
Pyroelectric 0.2-100 1 900 V/W 108
Silicon photodiode 0.3-1.1 0.01 0.5 A/W 1012
aThe "gain" of a photodetector is defined as the output electrical signal in amperes (A) or volts (V) per unit incident radiant flux
in watts rvrJ.
bD is a standard figure of merit for solid state photodetectors. It is based on the ratio of the output signal to the random noise
signal per unit bandwidth. scaled to a uniform detector area. The values of D* are estimates at the spectral sensitivity maxima.
The higher the value of D*. the more sensitive the detector.

orders of magnitude higher than in a vacuum photodiode. Vacuum photodiodes and

photomultipliers have excellent stability and fast time response. However, the spectral
sensitivity is nonuniform and depends on the type of photocathode. Solid state pho-
todetectors are a less expensive alternative to photomultipliers, useful in the visible and
near-IR regions. A "photovoltaic" cell operates at zero bias (no applied voltage) and
generates a photovoltage when exposed to light. A photoconductive cell is connected to an
external voltage and generates a photocurrent when exposed to light. The useful sen-
sitivity of a photodetector is limited by the random noise power generated in the dark. The
"noise equivalent power" (NEP) is defined as the radiant flux required to give an output
signal (in volts) equal to the detector noise. NEP has the unusual units of W (VHz,
because electrical noise power is proportional to the bandwidth and noise voltage is
proportional to the square root of the noise power. The reciprocal of NEP is the "detec-
tivity" (D). A standard figure of merit for comparing solid state photodectors is referred
to as D*; it equals D multiplied by the square root of the detector area (Table 1-6).

1.3.4. Radiometry and Actinometry(3.13.14)

Radiometry refers to the measurement of radiant power emitted by a source or
incident on a receiver. The total radiometric power can be measured with a thermal
detector. More frequently, it is desired to measure the radiant power in a given wave-
length band. This may be done with inexpensive devices, consisting of band-pass filters, a
solid state photodetector, and a readout meter. Accurate measurements of the spectral
power distribution require a spectroradiometer, utilizing a monochromator instead of
filters. The instrument must be calibrated at each wavelength, e.g., with a standard lamp
of known spectral irradiance. The applications of radiometry require a working knowl-
edge of radiometric units, a subject that has been complicated by a diversity of definitions
and concepts. Radiometric units are based on the following concepts: radiant flux emitted
by a small area of a source in a given direction (radiance), radiant flux emitted by a point
source in a given direction (intensity), radiant flux received by a small area on a surface
(irradiance) , and radiant flux leaving a small area of an emitting or reflecting surface
(exitance). Each radiometric unit has a photometric equivalent, which is scaled to the
22 Chapter 1

phototopic (black and white) response of the human eye. The SI radiometric units and
their photometric equivalents are summarized in Table 1-5. Photometric units are used
extensively in the lighting and photographic industries. However, they are not applicable
in the UV and IR regions, and should be avoided whenever possible.
Actinometry is based on the incidence and absorption of photons, usually in connec-
tion with photochemistry or photobiology. The fluence rate is defined as the number of
photons per second incident in a perpendicular direction on a unit area of the receiver. The
fluence is the corresponding total number of photons incident on a unit area. In pho-
tobiology, fluence rate and fluence are frequently referred to as "dose rate" and "dose,"
respectively. However, this usage is ambiguous because "dose" is also employed for the
number of photons absorbed per unit volume of the medium. In order to avoid confusion,
"dose" and "dose rate" should always be defined when these terms are used. There is no
simple method for measuring absolute fluence rates with polychromatic light because
processes initiated by photon absorption usually depend on the wavelength of the radia-
tion. The quantum efficiency is the microscopic efficiency of a photoprocess, defined as
the probability that the absorption of one photon at a given wavelength will lead to single
observed event, e.g., photolysis of one molecule, emission of one photon, etc. (Quantum
efficiency is often called quantum "yield.") A useful actinometer should have a constant
quantum efficiency over the wavelength range of application, or at least known accurate
values of the dependence of the quantum efficiency on wavelength. Since only absorbed
radiation can initiate a photoprocess, the fraction of incident light absorbed by the ac-
tinometer must be known at each wavelength. This complication is avoided if the ac-
tinometer is completely absorbing over the desired wavelength range. An actinometer
should have the properties of "linearity" (e.g., doubling the fluence doubles the re-
sponse) and "reciprocity" (e.g., doubling the fluence rate and halving the time gives the
same response). Finally, the measurements must be reproducible, and preferably conve-
nient. In view of these requirements, it is not surprising that relatively few practical
actinometers have been devised. One of the most widely used chemical actinometers is
based on the photochemical decomposition of aqueous potassium ferrioxalate.(12,15) This
system has a constant quantum yield of 1.2 ± 0.1 for ferrous ion production from 210 to
440 nm, and can be "pushed" to about 550 nm with a correction for the decreasing
quantum yield above 440 nm. The measurements are made in darkness or under weak red
light, and must be corrected for the fractional absorption by the actinometer solution and
reflection losses at the windows. The calculation of absorbed dose rates from incident
tluence rates in an actual photochemical system may be quite difficult, especially for
polychromatic radiation.

1.3.5. Light-Tissue Interactions

The interactions of light and tissue are involved in many aspects of photobiology,
e.g., plant growth and optical dosimetry in photomedicine. When collimated light im-
pinges on a highly turbid material, the independent concepts of reflection, refraction, and
diffraction are no longer distinguishable, and the interactions of the light with the material
are described as scattering. The "cloudy" appearance of biological tissue is caused by
multiple scattering at diverse inhomogeneities, including macromolecules, cell organ-
elles, microscopic "pools" of water, etc. Scattering leads to the spreading of a collimated
Photophysics 23

light beam, and loss of its initial directionality. The fraction of incident radiation that exits
at the back face of a scattering layer is the diffuse transmittance, and the fraction that exits
at the front face is the remittance or "diffuse reflectance." In addition to scattering
centers, biological tissues contain chromophores with significant absorptions in certain
wavelength regions. The attenuation of light within tissue depends on both scattering and
absorption. The optical penetration depth (8) provides a measure of how far light can
penetrate into tissue. For a nonscattering medium, 8 equals the depth at which the
transmissivity is lie or 0.37. The initial irradiance is reduced by approximately 90% at
28. The situation is more complicated for a scattering medium because the radiant flux
inside the material is no longer collimated and the irradiance does not provide an adequate
measure of the power incident on a molecule. The concept of space irradiance (X) has
been introduced for noncollimated light. It is defined as the radiant flux incident from all
directions on a small sphere inside the material divided by the cross-sectional area of the
sphere. For completely isotropic radiation, X = 4E. In highly scattering materials, the
flow of photons parallels particle diffusion in a fluid medium, where X is the analog of
particle concentration. The application of the "diffusion approximation" to a thick,
uniform tissue layer, exposed to a collimated beam of monochromatic light, leads to a
relatively simple expression for the space irradiance at depth x from the illuminated face:

x(x) = 2EoO + R) Exp - (x/8) (1-3)

where Eo is the incident irradiance, R is the remittance, and 8 is the optical penetration
depth. (16) The radiant flux entering the tissue layer is augmented by back-scattered radia-
tion, and may be up to fourfold higher near the front surface compared to nonscattering
media. The value of 8 depends on the nature of the tissue and the wavelength. Important
absorbers in tissues include proteins, hemoglobin, melanin, and water.(l7) In general, 8
decreases with the vascularity (blood content) of the tissue and is lower for blue light than
red light, e.g., for adult brain tissue, 8 is 0.45 mm at 488 nm and 1.5 mm at 710 nm.(l6)
An earlier approach to tissue optics is based on the theory of Kubelka and Munk, which
assumes that the flux in a layer of turbid material can be divided into a forward beam and a
backward beam, each of which undergoes back-scattering and absorption.(I7·18) The
transmitted light is taken as the forward beam that exits at the rear face, and the remitted
light is taken as the backward beam that exits at the front face. The Kubelka-Munk theory
was developed for layers of finite thickness and has been used extensively in the pigment
and paper industries. Butler pointed out that the optical path for light absorption may be
lengthened significantly in highly scattering tissues because the photons are scattered
repeatedly in different directions until a chance encounter occurs with a chromophore. He
exploited this phenomenon to obtain remarkably good absorption spectra of very weak
chromophores in seeds and tissue slices, using a special spectrophotometer in which the
photodetector measured the light transmitted over a wide angular distribution. (19,20) An-
other method for measuring absorption and reflection spectra of turbid materials is based
on an "integrating sphere." The sample and photodetector are located inside a hollow
sphere, coated with a white, highly reflecting powder. When the sample is illuminated
with diffuse radiation, most of the transmitted or remitted light exiting the sample at
different angles eventually reaches the photodetector.
The use of lasers in medicine is one of fastest growing applications of photobiology.
24 Chapter 1

Lasers are used in surgery primarily as a heat source. The beam is usually focused to an
image less than 2 mm in diameter, which provides a high power density and good control.
Although the detailed analysis of laser tissue heating is quite complicated, considerations
based on the wavelength dependence of 3 provide insight into the basic principles. For
CW laser operation, the heating effect originates in tissues with the higher values of 3.
Oxyhemoglobin in blood is the dominant tissue absorber for the blue-green emission of
the argon laser, which passes readily through water and colorless tissue. The argon laser is
well suited for inducing coagulation of hypervascularized tissue, which requires heating to
approximately 60°C. There are major applications in ophthalmology because the radiation
can pass through the cornea, the lens, and vitreous media, and induce heating in the highly
pigmented retina, e.g., for treatment of diseases of the retina in which abnormal blood
vessels are formed. The 1064-nm emission of the Nd-YAG laser lies in a "window"
between strongly absorbing blood at shorter wavelengths and strongly absorbing water at
longer wavelengths. Consequently, Nd-YAG laser radiation is highly penetrating and may
be employed for heating relatively large tissue regions. A major clinical application is
tissue coagulation with good hemostatic properties (stopping of blood flow), e.g., for
endoscopic surgery of the gastrointestinal tract. The 1O.. 6-f1,m emission of the carbon
dioxide laser is strongly absorbed by water. The high density of energy absorption leads to
vaporization of tissue water. The heat is carried away by the steam and cannot damage the
adjacent tissue. The carbon dioxide laser acts as a "laser scalpel," and it has been used
for a wide variety of surgical procedures, especially for tissue of high water content. The
temperature reached in laser heating of tissue depends on a balance between the rate at
which energy is introduced by the laser and the rate at which heat is carried away by blood
flow. However, light absorption and heat transfer are not the only factors when laser
energy is delivered in short, high-power pulses, e.g., with a Q-switched Nd-YAG laser.
In this case, the high electrical fields associated with the electromagnetic radiation induce
a sequence of very fast electromechanical events, leading to strongly localized disruption
of the tissue. A recent clinical application involves ophthalmic treatments in the anterior
regions of the eye with good protection against retinal damage. Some types of lasers
induce photochemical reactions in tissues with potential clinical applications. Excimer
lasers emit high-intensity pulses in the UV region that are strongly absorbed by tissue
proteins (Table 1-4). This radiation produces a very sharp, well-defined cut with minimal
thermal damage in the neighboring tissue, which is referred to as "photoablation."
Laser-tissue interactions may be initiated by "photosensitization," in which the energy
absorbed by a low concentration of a colored exogeneous agent is responsible for the
effects (Chapter 3). "Photodynamic therapy" of malignant tumors is a recent develop-
ment, in which the photosensitizer is a mixture of porphyrins referred to as "hema-
toporphyrin derivative. "(21) This drug has the property of localizing preferentially in
tumor tissue after intravenous administration. A pink fluorescent light emission induced
by illumination of the tissue surface with uv -A makes it possible to identify a tumor and
visualize its dimensions. Exposure of the tumor to strong red light, e.g., from a dye laser,
initiates the sequence of events leading to eradication of the tumor in favorable cases.
Clinical trials on photodynamic therapy of superficial and internal tumors have been in
progress since the late 1970s, and an evaluation of the efficacy of this treatment for certain
types of cancer should be available within the next few years (Chapter 6).
Photophysics 25

1.4. MOLECULAR STRUCTURE AND EXCITED STATES

1.4.1. Interactions of Atoms with Light

The exchange of energy between light and matter is involved in all phenomena of
photobiology. Electromagnetic radiation carries electric and magnetic fields, which can
interact with the electrons and nuclei of atoms and molecules. The strongest interaction is
between the electric field of the radiation and the electric charge of the electrons. An
electric field exerts a force on a free electron, causing a change in its kinetic energy, e. g. ,
the electrons released by light at the photocathode of a vacuum photodiode are acceler-
ated to the anode by the applied electric field. The situation is more complicated for the
interactions of light with atoms, where the electrons move in stable orbits corresponding
to the discrete energy levels of the atom. The transfer of one photon from electromagnetic
radiation to an atom raises one electron to a new stable orbit, subject to certain "selection
rules," involving the properties ofthe initial and final states ofthe atom and the polariza-
tion of the light. Similarly, an electronically excited atom loses energy by emitting one
photon, either by "spontaneous" emission in which light is not required, or by "stimu-
lated" emission, in which the atom interacts with radiation. The most important conse-
quence of the interactions between atoms and radiation is that radiative transitions are
efficient only when the photon energy of the radiation is close to the energy difference of
the atom in the initial and final states. Therefore, an absorption or emission spectrum
provides direct information about the atomic energy levels because each strong line
corresponds to the energy difference between two allowed energy states. The photon
energies in the UV, visible, and IR bands correspond to radiative transitions of the valence
electrons, typically 0.1-10 eV (Table 1-2). The inner-shell electrons are more strongly
bound to the nuclei, and transitions leading to the occupation or depletion of these levels
correspond to the lines in X-ray spectra.
In addition to its electric charge, an electron has an intrinsic magnetic field, identi-
fied with electron "spin." This term originated in early quantum theory, when it was
postulated that an electron is similar to a small rotating sphere with an electric charge on
the surface, and the magnetic field is generated by the equivalent electric current. (In
classical physics, a charged particle possessing angular momentum, e.g., in circular or
elliptical motion, is equivalent to a current loop, and therefore it generates a magnetic
field.) The present view is that spin is an intrinsic angular momentum of the electron, in
addition to any angular momentum it acquires from its orbital motion, which is the source
of its magnetic field. Quantum theory shows that the electron spin of a free electron can be
oriented in only two directions relative to an external magnetic field, which are referred to
as "spin up" ( i ) and "spin down" ( t). Quantization of electron spin direction also
applies to electrons in atoms. Each electronic state in an atom is identified with a set of
four "quantum numbers," three of which specify its orbital motion and the fourth
specifies its spin orientation. The spins of electrons in a complete shell are "paired"
(i t), which maximizes the number of electrons that can occupy an electron shell. (The
Pauli Exclusion Principle states that no two electrons in an atom can have the same set of
four quantum numbers.) The valence shell in free atoms is unfilled, except for the rare
gases, and therefore a free atom can have one or more unpaired electrons. The spin state
26 Chapter 1

of an atom is designated by the quantity (2S + 1), where S is the total spin quantum
number. [The laws of quantum physics show that an atom with total spin S can orient itself
in (2S + 1) directions relative to a magnetic field. This "spin multiplicity" can be
determined experimentally from atomic spectra.] For a single unpaired electron, S = liz
and (2S + 1) = 2. This spin state is referred to as "doublet," which indicates that a single
electron can orient parallel or antiparallel to an applied magnetic field. A hydrogen atom
is always in a doublet state because it has only one electron. If all spins in an atom are
paired, S = 0 and the mUltiplicity is 1; this state is referred to "singlet." The rare gas
"ground" (lowest energy) states are always singlet. If two spins are unpaired, S = 1 and
therefore (2S + 1) = 3; this is a "triplet state." A helium atom, with two electrons, can
exist in singlet states and triplet states. The ground state of the helium atom is a singlet,
and the strong absorption lines show only transitions to higher energy singlet states. This
is an example of a selection rule, limiting allowed optical transitions to states of the same
spin mUltiplicity. The spin selection rule does not apply to states populated by collisions
of an atom with electrons, e.g., the helium arc spectrum has emission lines corresponding
to transitions between singlet states and between triplet states. (Neon atoms in a helium-
neon laser are pumped by collisions with relatively long-lived (metastable) excited helium
atoms.) The spin selection rule may be "relaxed" in heavy atoms, where interactions
between the spin and orbital motions of an electron, induced by the electric field of the
nucleus ("spin-orbit" interaction), leads to a "mixing" of singlet character into triplet
states and vice versa. The 254-nm emission line of the low-pressure mercury arc (Fig.
1-4) is a transition from an excited triplet state to the singlet ground state. This line has
high intensity because the same transition occurs in absorption, leading to a "resonance"
between emission and absorption. The repetitive emission and reabsorption of 254-nm
radiation, prior to escape of the photons through the lamp envelope, directs a high fraction
of the energy into this line. The strong yellow emission lines of the sodium vapor arc lamp
(589.0 and 589.6 nm) are also resonance lines.

1.4.2. Molecular Orbitals(22,23)

The total energy of a molecule in its ground state includes contributions from the
motions ofthe electrons and the atomic nuclei. According to "semiclassical" physics, the
nuclei have translational, vibrational, and rotational motions in space, and the electrons
move in discrete orbits about one or more nuclei. This general picture is retained in wave
mechanics, except the discrete electron orbits are replaced by "smeared out" electron
density distributions. Each set of electron and nuclear motions corresponds to a possible
energy state of the entire molecule. The transitions between energy states induced by the
absorption or emission of light may be identified by optical spectroscopy. Molecular
spectra in the UV and visible regions involve transitions between different electronic
energy levels. The sharp absorption lines in Fig. 1-13a correspond to discrete electronic
transitions. Electronic spectra usually have a vibronic substructure, resulting from the
vibrational motions associated with each electronic level, and leading to broadening of the
electronic transitions into bands (Fig. 1-13b). In an "allowed" (high-probability) absorp-
tion process, a molecule absorbs one photon and is raised to a higher energy state.
Similarly, in an allowed emission process, the molecule falls to a lower energy state and
emits one photon. As with atoms, these transitions are efficient only when the quantum
Photophysics 27

Fig. 1-13. (a) Atomic-like electronic energy

levels. The optical transitions are very nar-
row because only light of wavelengths corre-
sponding to the electronic energy differences
between the sharp levels can be absorbed.
The lower absorption line results from the
transition So ~ S], and the higher absorp-
tion line results from the transition So ~ S2.
The strength of each transition is indicated
by the extinction coefficient (E), and the line
width is detennined by the resolution of the
spectrometer. (b) Molecular-like energy lev-
els. The lighter horizontal lines are
the vibrational energy states associated with
the excited electronic states S 1 and S2' The
vibrational levels of the ground state So are
not shown. The interactions of the electronic (0 ) (b)
and vibrational energy levels broadens the
electronic transitions into bands. The structure in the absorption spectrum So ~ SI results from the vibronic
transitions 0-0', 0-1', 0-2', etc. Only the envelope of the vibronic transitions is resolved in the So ~ S2
spectrum.

energy of the photon is close to the difference between the energy of the molecule in the
initial and final states. Therefore, the character of absorption and emission spectra pro-
vides information about the structure and properties of molecules.
Quantum mechanics leads to a powerful description of the electron energies and
motions in molecules. The inner-shell electrons of atoms larger than hydrogen are tightly
bound to their own nuclei, and they can be considered as localized in atomic orbitals
(AO), which do not participate in optical transitions, except at vacuum UV wavelengths.
The valence electrons are less strongly bound, and can be excited by absorbing radiation
from UV-C through the visible and near-IR spectral regions. The orbits of the valence
electrons usually extend over two or more atoms. They can be described by molecular
orbitals (MO), formed by combining the AO. The combination of atomic orbitals to form
molecular orbitals is described in standard texts on molecular physics. (22,23) There are
several basic types of MO in common organic molecules, as illustrated for ethylene and
formaldehyde in Fig. 1-14. A bonding sigma orbital (a) is strongly localized around the
line between the two atoms. A covalent single bond consists of one a orbital occupied by
two electrons. Each a orbital has a counterpart (a*), in which the electron density extends
away from the two carbon atoms along the molecular axis. This MO is called "antibond-
ing" because a molecule is less stable when the a* orbital is occupied, compared to the
corresponding a orbital. The bonding pi MO (11") is localized in two parallel lobes, above
and below the axis, with a node (no electron density) on the axis. The antibonding pi MO
(11"*) has an additional node halfway between the atoms and perpendicular to the axis. A
covalent double bond contains four electrons, two electrons in one a MO and two
electrons in one 11" MO. Atoms such as oxygen and nitrogen have "lone pair" 11" electrons
that may not participate in chemical bonding, e.g., the lone pair electrons on the oxygen
atom in formaldehyde (Fig. 1-14b) and on the nitrogen atom in pyridine. The correspond-
28 Chapter 1

0-*

_-
>--8 ,
€--(
( \
\ I
-~ '- / n
I I 7r 7r*
..... _-----"".
b

Fig. 1-14. (a) Schematic representation of some molecular orbitals of ethylene. Ethylene is a planar molecule;
all six atoms are located in the plane of the page. Each carbon atom has four valence electrons. The IT molecular
orbital (MO) connecting the two carbon atoms has cylindrical symmetry with to the C-C axis. The IT* MO is the
counterpart of the IT MO, with a node halfway between the two carbon atoms. It is referred to as "antibonding",
because the molecule is less stable if this MO is occupied by an electron, compared to the IT MO. A carbon-
carbon 1T MO has two lobes; one lobe is located above the plane (solid line) and the other lobe is located below
the plane (dashed line). There is a node in the molecular plane. An equivalent 1T MO is oriented at 90° to the one
shown (perpendicular to the page). In the C-C double bond, one electron from each carbon atom, of opposite
spins, occupies the IT MO, and one electron from each carbon atom, of opposite spins, occupies a 1T MO. The
other two valence electrons on each carbon atom form IT orbitals with the hydrogen atoms, accounting for four
single bonds (not shown). (b) Formaldehyde has carbon-oxygen orbitals similar to the carbon-carbon MOs of
ethylene, labeled as IT, IT* , 1T, and 1T* . Two of the four valence electrons on the carbon atom and the two valence
electrons on the oxygen atom form the carbon-oxygen double bond. The other two valence electrons on the
carbon atom form single bonds with the two hydrogen atoms (not shown). Oxygen has two paired 1T electrons in
the outer shell that are not used for bonding in formaldehyde, referred to as "lone-pair" electrons. They are
localized in a nonbonding MO (n), oriented perpendicular to the C-O axis.

ing MOs are referred to as nonbonding (n). However, the nitrogen lone pair electrons in
aniline and pyrrole are conjugated with the 1T orbitals in the rings, and they are not
nonbonding.
The "configuration" of a given molecule is described by specifying the number of
electrons and their spin directions in each occupied MO. In almost all stable molecules,
each MO is occupied by two electrons with paired spins. Since the inner shells are fully
occupied by electrons with paired spins, S = 0 for the molecule, corresponding to a
singlet state. A few stable molecules exist with two unpaired electrons, i.e., S = 1 and the
molecule is a triplet state. Ordinary molecular oxygen is one of these unusual stable
molecules with unpaired electron spins. An allowed optical absorption elevates one elec-
tron in an occupied MO to a vacant MO, without a change of electron spin direction. One
way of describing this process is to specify the MOs involved, e.g., 1T ~ 1T* indicates that
an electron is a 1T MO is excited to a 1T* MO. Alternatively, the effect of the transition on
the spectroscopic state of the entire molecule may be specified, e.g., So ~ Sl indicates
that a molecule in its singlet ground state makes a transition to the first excited singlet
Photophysics 29

Fig. 1-15. (a) The occupation of mo-

a
----0"*
lecular orbitals in the lower electronic
states of ethylene. Each arrow repre-
sents the spin direction of the electron. ----71'*
In the ground state (So), two electrons
with paired spins occupy the lowest <T
and 1T orbitals. In the first excited sin- --11+-1-71'
glet state (51)' one electron in the 1T
orbital is elevated to the empty 1T* or- 11 11 ---H1I,....--0"
bita
, without a change of spin direc-
tion. In the lowest triplet state, one
electron in the 1T orbital is elevated to
11"_71"*
the empty 1T* orbital, with a reversal of
spin direction. The electrons in the four b
carbon-hydrogen bonds and the inner - - - cr*
shell atomic core electrons are not de-
picted. (b) The occupation of mo-
-+- -+- --t- --+- --71'*
--I+-- -+- --I+-- -+- --tt-- n
lecular orbitals in the lower electronic
states of formaldehyde. The two lone-
-+- --it- -+- -tt- --tt-- 71'
pair electrons of the oxygen atom oc-
-tt- -It-- -tt- -tt- --tt-- 0"
cupy the n orbital in the ground state T2 T1 S2 S1 So
(So), In the lowest excited singlet state 71'-71'* n -7r* 71'-71'* n-71'*
(51), one electron in the n orbital is
C
elevated to the empty 1T* orbital, with-
out a change of spin direction. In the 160
second excited singlet state (Szl, one
electron in the 1T orbital is elevated to
the empty 1T* orbital, without a change
of spin direction. A reversal of spin
direction takes place in the correspond-
ing excitations to the lowest triplet E(kcal/mol)
state (T 1) and second triplet state (T2)' 40
(c) State energy diagrams for ethylene
(left) and formaldehyde (right). The So- o 50--
energy scale shows the energy of the
electronically excited molecules rela-
tive to the corresponding ground states. The one-electron transitions are indicated in parentheses, e.g., S 1(1T, 1T*)
designates the excited singlet state of ethylene formed when one electron in the highest filled 1T orbital is elevated
to the lowest empty 1T* orbital, without a change of spin direction. The singlet and triplet states are displaced
horizontally to clarify the representation. The energy scale of kcal/mol is used frequently in molecular pho-
tophysics; I kcal = 4184 J.

state. The occupation of individual MOs are shown in Figs. 1-15a and b for ethylene and
formaldehyde, respectively. Figure 1-15c is a "state energy" diagram, showing the
energy of the excited states relative to the ground state of each molecule.
The concept of transition moment is used in molecular photophysics to describe the
interactions between light and molecules. The transition moment can be viewed as the
shift of electric charge in a molecule when it makes a transition between two electronic
states. This charge separation is induced by the electric field of the light, and it oscillates
with the light frequency. In classical physics, an entity consisting of two separated
charges of equal magnitude and opposite sign is known as a dipole. The dipole moment is
30 Chapter 1

Fig. 1-16. Jablonski diagram for an

organic molecule, showing the transi-
tions between the singlet (So, S l> S2,
... ) and triplet (T I> T 2, T 3, ... ) elec-
tronic levels (heavy horizontal lines).
The energy levels of only one vibrational
I~ mode are portrayed for each electronic
~A-t--:lt-Sj isc -I- state (light horizontal lines). Radiative
):=:.. RlJOR~ 5r i((~~ transitions are shown as straight arrows.

So v=o
~~ivr~~~~==~iSC~=;~~~~~~_
4
2
~ 11
PHOS
t shown as wavyrelaxation
Radiationless arrows: vr, vibrational
processes
relaxation; ic, internal conversion; isc,
So intersystem crossing. The radiative de-
are

cay processes depicted are fluorescence

(FLUOR) and phosphorescence (PHOS). The diagram exemplifies three spin-allowed absorption processes: the
transition from the thennally equilibrated, ground state, So(v = 0), to a vibrationally excited level of the second
excited singlet state, S2(V = 4'); the transition from the thennally equilibrated, first excited singlet state, S\ (v =
0), to a vibrationally excited level of the third excited singlet state, S3(V = 3'); the transition from the thennally
equilibrated, lowest triplet state, T\(v = 0), to a vibrationally'excited level of the second excited triplet state,
T3(v = 3'). Fluorescence is a spin-allowed, radiative transition from the thennally equilibrated, first excited
singlet state, S\(v = 0), to different vibronic levels ofthe ground state, S\(v = 0, 1',2', 3', ... ). Phosphores-
cence is a spin-forbidden, radiative transition from the thennally equilibrated, lowest triplet state, T\(v = 0), to
different vibronic levels of the ground state, S\(v = 0', 1',2',3', . , .).

defined as the product of the charge magnitude and the separation distance. This termi-
nology has been carried over into radiation physics, and the transition moment connecting
two molecular energy states is also called the "transition dipole. " In an allowed optical
transition, the magnitude of the transition moment is comparable to the displacement of
one electronic charge (1.6 X 10- 19 C) by a distance the order of atomic dimensions
(10- 10 m). The strength of an optical transition depends on the square of the transition
dipole connecting the two states, and the radiative lifetime of an excited state varies
inversely with the square of the transition dipole connecting it to the lower state. Large
transition dipoles are associated with strong absorption, strong emission, and short-lived
excited states, and vice versa. The concept of the transition dipole as a measure of light
interactions with the electrons in a molecule will be used to help explain the general
properties of excited states.
The electronic transitions in typical organic molecules can be represented by a
"Jablonski diagram" (Fig. 1-16), which is especially useful for relating the photophysical
properties to optical spectra. The heavy horizontal lines are the lowest vibrational levels of
the singlet states (So, Sl> S2' ... ) and the triplet states (T I' T2, ... ). The vibronic levels
associated with one vibrational mode of each electronic level are indicated by the sets of
lighter horizontal lines. The straight vertical arrows indicate allowed radiative transitions,
and the wavy vertical arrows indicate the very fast loss of excess vibrational energy by
vibrational relaxation (vr). The horizontal wavy lines indicate two important types of
energy-conserving, internal processes: In internal conversion (ic), the system shifts from
the lowest vibronic level of one electronic state to an upper vibronic level of another
electronic state. In intersystem crossing (isc), the system shifts from the lowest vibronic
level of the SI state to an upper vibronic level of the T 1 state. The isc process is promoted
Photophysics 31

by the spin-orbit interaction, which gives some triplet character to singlet states and vice
versa. Mter ic or isc, the system relaxes (via vr) to the lowest vibronic state of the
electronic level within 10- 11 s. Normal light absorption involves the transition So ~ Sn'
where n = I, 2, .... The radiative lifetime of an S 1 state is typically 10 - 9 s, and the
light emitted in the radiative process S1 ~ So is fluorescence. The diagram shows a
competing, nonradiative transition to So' which is relatively inefficient in many aromatic
and heterocyclic molecules. Intersystem crossing is another competing process, leading to
population of the lowest triplet state (T 1)' The T 1 state is long-lived because of the small
transition dipole connecting it with the ground state So. Light emission by a triplet state is
phosphorescence. Thermal excitation can induce the transition T 1 ~ S 1 if the levels are
close in energy, typic all y less than 0.1 eV. This process leads to a luminescence with the
lifetime of phosphorescence and the emission spectrum of fluorescence; it is referred to as
"delayed fluorescence." Triplet states are important photochemical intermediates be-
cause they are energy-rich, long-lived species that can participate in many types of
chemical reactions with unexcited molecules.
The processes shown in Fig. 1-16 are applicable to (1T,1T*) and (n,1T*) transitions.
However, there are significant differences between the two types of transitions, because
(n,1T*) transition dipoles are much smaller than (1T,1T*). In general, n ~ 1T* absorptions
are more than lO-fold weaker than 1T ~ 1T* absorptions in the same molecule and show
more vibrational structure. Solvents of higher polarity shift n ~ 1T* absorptions to shorter
wavelengths and 1T ~ 1T* absorptions to longer wavelengths. (The interactions of an
electric charge with a dielectric medium tends to stablize the charge by lowering the
energy. The extended electron distribution in a 1T ~ 1T* excited state interacts more
strongly with a polar medium than the ground state, leading to a smaller energy difference
and a "red shift" of the absorption band. On the other hand, the n ~ 1T* ground state
interacts with a polar medium more strongly than the excited state, leading to a higher
transition energy and a "blue shift.") The fluorescence of (n,1T*) excited singlet states is
less efficient and longer lived than (1T,1T*) singlet states. An important difference in the
photochemistry of (n,1T*) and (1T,1T*) excited states is the high efficiency of hydrogen
abstraction by (n,1T*) states. Consequently, photoreduction may be much more important
for carbonyl compounds than similar aromatic hydrocarbons (Section 2.1.2).

1.4.3. Quantum Yield, lifetime, and Relaxation of Excited States(23)

The absorption of light at ordinary intensities leads to the population of thermally
equilibrated S1 states at a rate proportional to the incident fluence rate. The relaxation
(energy loss) processes obey first-order kinetics, i.e., the loss rate at any time is propor-
tional to the number of surviving S1 molecules. (The radioactive decay of atoms follows
the same rate law.) A first-order process is characterized by its rate constant (k) or,
equivalently, the mean lifetime ('T), where 'T = 11k. The first-order rate constants for
transitions out of S1 are kj' k ic ' and k isc ; the corresponding transitions are indicated in Fig.
1-16. The fraction of fluorescence events equals kJk, where k = (kf + kic + k isc )' This
fraction is the quantum efficiency (or quantum yield) of fluorescence (<Pj ). The experi-
mental fluorescence lifetime ('T) equals 11k because the fluorescence rate monitors the
overall decay of S1 excited states. The theoretical radiative lifetime of the S1 state ('Tf) is
32 Chapter 1

defmed as I! kf . The ratio of the experimental fluorescence lifetime to the radiative lifetime
is the fluorescence efficiency:

(1-4)

Therefore, when an excited state decays by both radiative and nonradiative processes, the
lifetime of the excited state is reduced and the fluorescence efficiency is less than unity. A
similar kinetics analysis can be carried out for metastable triplet states. (23)

1.4.4. Fluorescence Quenching and Electronic Energy Transfer

Many substances "quench" (inhibit) the fluorescence of organic molecules. Fluores-

cence quenching in liquid media can take place by either of two general mechanisms. In
"dynamic quenching" the molecules of the fluorescent species lfluorophore) interact
with the molecules of the quencher via intermolecular collisions. This process provides an
additional pathway for the relaxation of Sl states and leads to a reduction in the experi-
mental fluorescence decay lifetime. Alternatively, the presence of a quencher molecule
close to a fluorophore may induce the immediate deactivation of Sl states. This process is
"static quenching." Some quenchers exhibit dynamic quenching at low concentrations
and static quenching at high concentrations. Dynamic quenching follows the well-known
Stern-Volmer equation:

(1-5)

where <pJ and <Pf are the fluorescence efficiencies in the absence and presence of the
quencher, respectively, T is the fluorescence lifetime, kQ is the bimolecular rate constant
for reactions of the quencher with the fluorophore, and CQ is the quencher concentration.
When this model is applicable, a plot of <PJI<Pf vs. CQ gives a straight line with its slope
equal to TkQ' The rate constant KQ may be measured, or estimated from molecular theory.
Electronic energy transfer is a special type of quenching in which the electronic
excitation energy of a donor molecule is transferred to an acceptor molecule via a
radiationless process. It is characterized by a shift of the fluorescence band from that of
the donor to that ofthe acceptor, and is called "sensitized fluorescence." Several mecha-
nisms of energy transfer have been identified in liquid and rigid media. One of the most
important is resonance excitation transfer, which is also referred to as "dipole-dipole"
or "long-range" energy transfer. The theory of resonant transfer, as developed by For-
ster, is based on two coupled radiationless transitions, leading to deactivation of the donor
and simultaneous excitation of the acceptor (Fig. 1-17). The strength of the interaction
depends on the product of the emission dipole of the donor and the absorption dipole of
the acceptor. The detailed calculation of the energy transfer rate shows that it depends on
the overlap between the fluorescence band of the donor and the absorption band of the
acceptor, and varies with the inverse sixth power of their separation. (3,23 ,24) The range of
Forster-type energy transfer attains 1-10 nm in favorable cases, which is far greater than
the collision diameters of small molecules. Energy transfer by this mechanism has been
Photo physics 33

0*

A*
t
llE ELECTRONIC
+ +I
+, II II
I I I
II
, I I
I
, ,i I
I I A
D
I I I
I I I

L - -_ _ _ _ _--l' II
COUPLED OR RESONANT TRANSITIONS

········OONOR EMISSION
- - DONOR ABSORPTION
/ ......\/............. - - - ACCEPTOR ABSffiPTlON

/:(tt®~. . . . . . . . . . ..
-E REGION OF
SPECTRAL OVERLAP

Fig. 1-17. Electronic energy transfer via coupled transitions for a donor-acceptor pair. The downward transi-
tions of the thermally equilibrated. excited donor (D*) to different vibronic levels of the donor ground state (D)
match the upward transitions of the acceptor ground state (A) to different vibronic levels of the acceptor excited
state (A*). The electronic energy difference ~E is tbe excess energy of D* relative to A*. in their thermally
equilibrated, excited singlet states. The overlap of many broadened vibrational levels in a condensed medium
ensures resonance, if the fluorescence band of the donor overlaps the absorption band of tbe acceptor (lower
diagram).

observed in fluid and rigid media, and between chromophores in a macromolecule. An

important example of the latter takes place in proteins, where the direction of excitation
energy transfer is phenylalanine ~ tyrosine ~ tryptophan. In human serum albumin, with
one tryptophan and 18 tyrosine residues, approximately 50% of the potential tyrosine
fluorescence excited at 275 nm is transferred to the tryptophan residue. Resonance energy
transfer between triplet states is less efficient than between singlet states. The long
lifetime of triplet states favors an alternative short-range energy transfer process, known
as "triplet-triplet" energy transfer. The triplet energy is the key parameter, defined as
the energy difference between the T[ and So states (Fig. 1-16). Triplet-triplet energy
transfer is always "downhill", i.e., the triplet energy of the donor must be higher than
that of the acceptor. The formation of singlet oxygen in photosensitized chemical reac-
tions takes place by triplet-triplet energy transfer (Section 3.2.3).
34 Chapter 1

1.5. SPECTROSCOPIC MEASUREMENTS

A major objective of molecular spectroscopy is to deduce fundamental information

about the structure and properties of molecules. Spectroscopy is used also as a tool in
many laboratory procedures, including chemical analysis, biochemical assays, enzyme
kinetics, etc. A basic understanding ofthe underlying photophysics is required for all but
the most routine spectroscopic measurements in order to utilize the instrumentation prop-
erly and interpret the results.

1.5.1. Absorption Spectroscopy

The measurement of spectra with photoelectric detectors is referred to as spec-
trophotometry. In absorption spectrophotometry, the fraction of incident light transmitted
by a sample is measured over a selected range of wavelengths. Modem instruments
measure the transmittance CD and the optical density (O.D.), which are related by

O.D. = 10glO (lID (l-6)

An O.D. range of 0-4 corresponds to a change of T from 1 to 0.0001. Measurements of

O.D. must be corrected for the spectral intensity distribution of the source, and the
dependence of the photodetector sensitivity on wavelength. This correction is made by
subtracting the O.D. of a "reference" from that of the "sample" at each wavelength.
Any process that alters the light signal reaching the detector contributes to O.D., e.g.,
absorption, light scattering, and luminescence. The absorbance (A) is used to designate
the contribution of absorption alone to optical density. The ideal absorbance of a dilute
solute in a transparent solvent follows the Lambert-Bouguer:"'Beer law:

A = Ecd (1-7)

where E (liters/mole-cm) is the molar extinction coefficient of the solute at the wavelength
of measurement, c (moles/liter) is the molarity of the solute, and d (cm) is the optical
pathlength. The concentration of a solute can be calculated by measuring A, when E is
known at the measurement wavelength. The linear dependence of A on c in Eq. (1-7) may
not hold if there are concentration-dependent effects, e.g., aggregation or acid-base
dissociation ofthe solute. The applicability of "Beer's law" is usually tested by measur-
ing the dependence of O.D. on solute concentration. A linear dependence confirms that
Beer's law is valid for the given conditions. This "standard curve" can then be used to
relate O.D. to solute concentration for analytical purposes. A typical two-beam, recording
spectrophotometer is shown in Fig. 1-18. In this arrangement, the light transmitted by the
sample and the reference are measured alternately by means of a rotating sector, the O.D.
difference is calculated electronically, and displayed at each wavelength. The more recent
instruments utilize microprocessors to store and analyze the data. The spectra can be
displayed in many formats, including plots of the first and higher derivatives of the
absorption bands vs. wavelength, and the sum and difference of several spectra. Flash
photolysis is a major application of absorption spectroscopy. In this technique, the system
of interest is irradiated with a short, intense light flash, and the short-lived intermediates
Photophysics 35

ROTATING SECTOR

Fig. 1-18. A schematic representation of a

dual-beam recording spectrophotometer. Light
from the source (S) is focused on the entrance
slit of the monochromator (Me). The exiting
monochromatic radiation is collimated by a
lens, and passed alternately through the sam-
S
ple cuvette and the reference cuvette by a par-
tially silvered, rotating sector. The amplifier
processes the photomultiplier signals, and out-
puts either the transmission or the optical
density to the y axis of the recorder. The RECORDER

wavelength (A) drive on the monochromator

generates an electrical signal that drives the x axis of the recorder. More than one source may be used to cover a
wide wavelength range, e.g., a deuterium lamp for the UV region and a tungsten-halogen lamp for the visible-IR
regions.

are identified and followed in real time by making fast optical absorption measure-
ments.(25) A pulsed laser flash photolysis apparatus is shown schematically in Fig. 1-19.
This type of apparatus has been used to identify short-lived free radicals and triplet states
and to investigate their fast chemical reactions.

1.5.2. Luminescence Spectroscopy(12,24,26)

Luminescence spectra lead to information about optical transitions from higher to
lower energy states, which augments the information about transitions from lower to

Fig. 1-19. Schematic diagram of a

laser flash photolysis apparatus. The
PM
sample in the cuvette (C) is irradiated
with a photolytic flash from a laser (L).
The photochemical intermediates are
detected by passing a monitoring light
beam (ML) through the cuvette, and to
a monochromator (MC) with a pho- Me
tomultiplier (PM) at the exit slit. The PO
monitoring source may be a continuous
lamp, e.g., an arc lamp, or a flash
lamp of relatively long duration on the o
time scale of the intermediates. The L
photomultiplier output (S) is displayed
on an oscilloscope (0), which is trig-
gered (T) by a signal from an auxiliary
5 T
photodiode (PO), activated by the laser
pulse with a beam splitter (B). The oscilloscope trace records the changes in transmission induced by the laser
pulse at the wavelength setting of the monochromator. Absorption spectra of the transient intermediates can be
determined by repeating the measurements at different wavelengths. The time resolution in flash photolysis
ranges from microseconds with gas-filled photolytic flash lamps, to nanoseconds with Q-switched lasers, and
picoseconds with mode-locked lasers. The effective spectral range depends on the choice of monitoring lamp,
monochromator, and photodector, typically 200-900 nm. In an alternative arrangement, the photolysis beam
and the monitoring beam are passed through the sample in the same direction.
36 Chapter 1

,,,
, Fig. 1-20. Absorption and emission
'-,\ '
I
spectra of carbazole in ethanol. The
intensities of the various spectra are
\ I
t
oJ
t not to scale. The spectra are por-
I e trayed on a linear energy scale (in-
creasing from left to right), in which
case the wavelength scale is nonuni-
form. This display best demonstrates
the mirror image relationship be-
tween the vibronic structure of the
absorption band (So - SI) and that
of the fluorescence band (S I - So).
The phosphorescence band (T 1 -
550 500 450 400 350 300
So) is located at higher wavelengths
WAVELENGTH (nm)
(lower energy) than the fluorescence
band, because nonradiative relaxation processes are involved in the intersystem crossing from the fluorescent
state to the triplet state. The weak So - TI absorption (not shown) is expected to commence near 400 nm.

higher energy states provided by absorption spectra. The relationship between absorption
and emission spectra is exemplified in Fig. 1-20, showing the fIrst absorption band (So ~
Sl) and fluorescence band (Sl ~ So) of carbazole in ethanol. The fluorescence band is
always centered at longer wavelengths than the absorption band, which is referred to as
the Stokes' shift. The Stokes' shift occurs because some energy is lost via nonradiative
relaxation processes prior to the population of the thermally equilibrated Sl state from
which fluorescence takes place (Fig. 1-16). Excitation into the higher excited states S2'
S3' ... with shorter wavelength light almost always leads to the Sl ~ So fluorescence
because the higher excited states are depopulated by fast nonradiative relaxation pro-
cesses. (The S2 ~ So fluorescence can be detected with low quantum efficiency in many
molecules.) Luminescence measurements are not routine. They require careful attention
to the experimental conditions, and may involve sophisticated instrumentation and consid-
erable skill in data analysis. However, the wealth of information provided by lumines-
cence justifies this effort, and every researcher in photobiology should be familiar with the
basic aspects of the subject and its potential applications.

1.5.2.1. Measurement of Luminescence Spectra

Steady-state fluorescence spectra can be measured with a luminescence spec-
trophotometer. In a typical arrangement (Fig. 1-21), light from the source is passed
through an excitation monochromator onto the sample; the fluorescence is collected at 90°
to the excitation beam, analyzed by the emission monochromator, and measured with a
photodetector. Two types of fluorescence spectra may be obtained. In a fluorescence
emission spectrum. the excitation wavelength is fixed and the emission monochromator is
scanned over the fluorescence band. This mode is frequently employed for the identifica-
tion of fluorescent substances, energy transfer studies, and concentration measurements.
In afluorescence excitation spectrum. the emission wavelength is fixed and the excitation
Photophvsics 31

o
Fig. 1-21. Schematic representation of a lumi- x-v RECORDER
nescence spectrophotometer. Light from the
source (S) is focused on the entrance slit of the
excitation monochromator, and the exiting
monochromatic radiation is collimated by a
lens and relayed to the sample cuvette. The
fluorescence is collected at 90° to the excitation
beam and focused on the entrance slit of the
emission monochromator. A photomultipler
(PM) located behind the exit slit provides an
electrical signal proportional to the emission
intensity, that drives the y axis of the x-y re-
corder. For fluorescence emission spectra, the
wavelength (A) drive on the excitation mono-
chromator is fixed, and the A drive on the emis-
sion monochromator is scanned over the fluo- PM
rescence band, and drives the x axis of the
recorder; for fluorescence excitation spectra, the A drive on the emission monochromator is fixed, and the A drive
on the excitation monochromator is scanned over the absorption spectrum and drives the x axis of the recorder. The
fluorescence spectra as recorded are "uncorrected," i.e., the excitation bands are not a true representation of the ab-
sorption bands responsible for the fluorescence, and the emission bands are not a true representation of the
fluorescence. Procedures for measuring corrected spectra are described in the literature. (26,27)

monochromator is scanned over the absorption band of the sample, The excitation spec-
trum selects the absorption of fluorophores from the overall absorption spectrum.
Fluorescence measurements are subject to instrumental errors that must be eliminated
or corrected. The "inner filtering" of fluorescence by the absorption of the sample is
minimized by keeping the absorbance low in the region of the spectrum where fluores-
cence and absorption overlap. Another source of error is distortion of excitation spectra
when the extent of light absorption is not proportional to the absorbance of the sample.
According to Eq. (1-6), T = 10-O.D., and the fractional absorption (l - T) is an
exponential function of a.D. The error is minimized if a.D. < 0.02 at all excitation
wavelengths, in which case the absorption by the sample will be almost proportional to its
a.D. Fluorescence excitation spectra must be corrected for variations in the spectral
intensity distribution of the light that reaches the sample. This can be done by measuring
the output of the lamp and excitation monochromator with a calibrated power meter as a
function of wavelength and dividing the measured excitation spectrum by this calibration
curve to obtain the corrected spectrum. Some instruments make this correction internally
with a "quantum counter," which is a device that measures the relative photon flux
incident on the sample at each excitation wavelength. A useful quantum counter is a
concentrated dye solution, e.g., rhodamine B in ethylene glycol, that absorbs all incident
light from 220 to 600 nm and emits a fluorescence band centered at 630 nm that is
proportional to the incident fluence rate, independent of the excitation wavelength. (I 2)
Emission spectra must be corrected for the spectral response of the photomultiplier, and
the efficiency and bandwidth of the emission monochromator. A simple but reliable
procedure for correcting fluorescence emission spectra involves calibrating the instrument
with standard fluorophores for which corrected spectra are available, e.g., quinine
bisulfite in 0.1 N sulfuric acid and anthracene in ethanoI.<12) The fluorescence of a
38 Chapter 1

standard is measured at exactly the same conditions as the sample (including the same
solvent), and the ratio of the measured emission spectrum to the corrected spectrum
provides correction factors at wavelengths where both substances emit. A similar method
can be used to determine the relative fluorescence efficiency, by comparing the emission
spectrum of a given sample to that of a standard fluorophore with a correction for the
relative absorbance by the two fluorophores at the excitation wavelength. Since the two
emission bands are different, it is necessary to integrate over the corrected emission
spectra of the sample and standard in order to compare the relative fluorescence inten-
sities. Some expensive spectrophotometers make an automatic correction of emission
spectra with microprocessors or by interfacing to a microcomputer.
Although phosphorescence is weaker than fluorescence, it persists longer and can be
measured by chopping the incident radiation and synchronizing the detector to read only
during the dark periods between the light pulses. This was first accomplished with a
mechanical device by Becquerel in 1871. Two circular disks with holes cut around the
circumference were mounted on the same axle, the holes in the first disk being offset from
those in the second. The specimen was placed between the disks, and when the axle was
rotated, the specimen was illuminated intermittently by a beam of exciting light passing
through the holes in the first disk. The phosphorescence was viewed by looking at the
sample through the holes in the second disk, and the lifetime could be estimated by
varying the relative position of the two sets of holes. Modern instruments use high-speed
choppers or repetitively pulsed flash lamps to screen the photodetector from fluorescence.
Phosphorescence may be detectable in an ordinary spectrofluorimeter in rigid media at
low temperature, which suppress delayed fluorescence, and quenching of triplet states by
their diffusional reactions with unexcited fluorophore molecules and impurities. Much of
the available information about triplet energy levels has been obtained by phosphores-
cence measurements.

1.5.2.2. Fluorescence Lifetime

Fluorescence lifetimes provide information about the excited state properties of
fluorophores and their interactions with the environment. Several different techniques are
used for measurements of fluorescence lifetimes. In a single-pulse method, the sample is
excited with a strong light flash of nanosecond or shorter duration, and the fluorescence is
recorded in real time on a fast oscillocope or a transient recorder. The method is applica-
ble for relatively strong emissions, that can be recorded with adequate signal-to-noise in a
single pulse. When the fluorescence decay of a given sample is highly reproducible, the
signal-to-noise ratio can be improved by using repetitive pulses and an integrating mea-
surement system (stroboscopic method) because the random noise tends to average out.
One approach utilizes "pulse sampling," which measures the signal in short-time "win-
dows" at various delay times after the excitation pulse. A "sampling~' oscilloscope can
be employed, or the decay signals can be measured with a "boxcar" integrator and
printed out on a chart recorder. Time-correlated single-photon counting is an alternative
technique in which the time delay between the excitation pulse and the arrival of the first
photon emitted by the fluorophore at the photodetector is measured for a large number of
very weak pulses. A histogram showing the number of counts in narrow time delay
channels reproduces the fluorescence decay because the probability of detecting a single
Photo physics 39

photon at a certain time after excitation is directly proportional to the intensity of the
sample emission at the same time. The light source may be a high repetition rate (typically
2-50 kHz) gas-filled lamp of nanosecond duration or a repetitively pulsed laser. Single-
photon counting requires sophisticated and expensive electronics equipment. The decay
profIles measured with any pulse method must be corrected for the instrument response,
which includes the [mite width of the excitation pulses and the bandwidth of the detector
electronics. This process of "deconvolution" is usually carried out with a computer.
Fluorescence decay profIles are usually a single exponential function of time or, less
frequently, the sum of two exponential functions. The computer analysis leads to the
decay lifetime of each component and its relative contributions to the observed decay
profIle. A different approach to the measurement of fluorescence decay is based on
excitation with a very high-frequency, sinusoidally modulated lamp (phase modulation
method). (A lamp is "modulated" by superimposing a repetitive, time-varying pattern on
the intensity, e.g., the light emitted by an arc lamp operated from a poorly filtered power
supply is partially modulated at twice the power line frequency.) The basis of this method
is that the time delay between excitation and light emission leads to a phase shift corre-
sponding to the decay lifetime, and the extent of sinusoidal modulation is reduced by
mixing with the decay profile. These parameters can be measured with a phase-sensitive
detector (a "lock-in" amplifier), in which the phase and amplitude of a signal generated
by the modulated source are compared with the fluorescence signal. Lock-in amplifiers
can recover weak signals several orders of magnitude below the input noise level. The
phase modulation method can measure subnanosecond fluorescence lifetimes at modula-
tion frequencies in the order of 30 MHz. However, the method is based on the assumption
that the decay profile is either exponential or the sum of two exponentials, which is not
always the case. Time-resolved fluorescence spectroscopy is a field of rapidly growing
importance in photobiology. A time-resolved spectrum shows the changes of the fluores-
cence spectrum over a time span comparable to several decay lifetimes, providing infor-
mation about the interactions of the excited state, e.g., proton dissociation, the relaxation
of the medium immediately after the excited fluorophore reverts to the ground state, and
the decay of molecular complexes formed in the excited state (exciplexes).

1.5.2.3. Polarization of Fluorescence

Fluorescence spectra are inherently polarized because the absorption and emission
dipoles have specific directions relative to the geometry of the molecule. When a group of
rigid molecules in random orientations are exposed to linearly polarized light, the proba-
bility of light absorption by a given molecule increases with the alignment of the absorp-
tion dipoles with the electric vector of the light. Therefore, the polarization of the light
"photoselects" an absorbing population of molecules. Since the emission dipole has a
fixed orientation relative to an absorption dipole, the fluorescence will be partially polar-
ized in a direction that depends on the molecular structure and the specific absorption
transition. Polarization is measured by placing a polarizer between the excitation mono-
chromator and the sample and an analyzer in front of the emission monochromator. The
polarization is defined as

P = (111 - (J/(/ II + l-"J (1-8)

40 Chapter 1

where III and I-L. are the fluorescence signals with the analyzer parallel and perpendicular
to the analyzer, respectively. The maximum value of P for a random distribution of
molecular directions depends on the angle between the absorption and transition dipoles in
the molecule, e.g., P = +(1;2) when the absorption and emission moments are parallel,
and P = -(Y:J) when they are perpendicular.(24,26) The polarization of fluorescence
excitation spectra (in rigid media) provides information about the orientation of the
absorption dipoles of different excited states relative to the emission dipole. However,
fluorescence polarization cannot be observed if the molecule is rotating so rapidly that all
directionality is lost prior to emission. This effect is fluorescence depolarization. The
rotation rate of a molecule depends on its size and shape, the temperature, and the
viscosity of the medium. The rate of rotation is characterized by a "rotational relaxation
time," which can be calculated from hydrodynamic theory. The dependence of P on
temperature and viscosity leads to values of the rotational relaxation time and information
about the molecular dimensions. Measurements on small molecules (e.g., lO-lO_m radi-
us) require a viscous solvent, such as glycerol. Larger molecules (e.g., > 1O-9-m radius)
show fluorescence depolarization in fluid water and other low-viscosity solvents. Dynam-
ic information about the interactions of fluorophores with the medium can be obtained by
measuring the polarization of fluorescence decay. In the "fluorescence probe" technique,
a small fluorophore is attached to a larger entity, such as a macromolecule or a bio-
membrane. The probe fluorescence provides information about the "microenvironment"
in the vicinity of the probe, e.g., depolarization measurements lead to values of the
motional relaxation times for the probe and the larger entity. Certain probes have the ad-
ditional convenient property that their fluorescence efficiency depends markedly on the
polarity of the microenvironment, e.g., the fluorescence of the dye ethidium bromide is
weak in water and very intense when it intercalates into a double-helical region of a
nucleic acid.

1.5.3. Optical Activity(27)

Nearly all molecules synthesized by living organisms are optically active, i.e., they
alter the properties of polarized light. Optical activity results from asymmetry of small
molecules, especially asymmetric carbon atoms, and from the conformation of macromol-
ecules, e.g., helical structures. The rotation of the plane of linearly polarized light
exemplifies optical activity. The dependence of optical rotation on wavelength is referred
to as optical rotatory dispersion (ORD) and is usually measured in a spectral region where
the material is nonabsorbing. The method is equivalent to Fig. 1-12a, where the first
polarizer is the optically active substance. Another consequence of optical activity is a
difference in the absorption of right-handed and left-handed circularly polarized light,
where the "handedness" corresponds to the rotation direction of the electric vector. (In
circularly polarized light, the electric vector rotates with the frequency of the electromag-
netic wave.) This phenomenon is called circular dichroism (CD). A CD spectrum is based
on the differential absorption of a sample for right-handed and left-handed circularly
polarized light, e.g., as obtained by passing linearly polarized light through a quarter-
wave plate. ORD and CD are different manifestations of the same physical property, and
an ORD spectrum can be transformed into a CD spectrum by a mathematical operation,
and vice versa. CD measurements on proteins from 190 to 230 nm have led to estimates of
Photophysics 41

the relative extent of a-helical, J3-sheet, and random coil conformations in their secondary
structures. CD measurements on polynucleotides from 200 to 300 nm provide information
about the extent of single-stranded, double-stranded, and random coil regions.

1.5.4. Photoacoustic Spectroscopy(28,29)

Photoacoustic spectroscopy (PAS) is based on the generation of sound waves by the
absorption of modulated or chopped light. Although this phenomenon was discovered by
Alexander Graham Bell in 1880, there has been a great resurgence of interest in recent
years, with the development of highly sensitive sound detectors and intense light sources,
especially lasers. Heat generation in a medium initiates a pressure wave that travels with
the velocity of sound. The heat sources are the radiation less relaxation processes initiated
by light absorption. Since PAS responds only to absorbed light, it permits measurements
in highly scattering or opaque materials, including living tissues. In a typical PAS spec-
trometer, the sample is illuminated with modulated or chopped monochromatic light and
the detector is a microphone or a transducer attached to the walls of the sample holder.
The acoustic signal is measured with a phase-sensitive detector tuned to the chopping
frequency. A plot of the signal strength vs. wavelength gives a photoacoustic spectrum.
The phase shift between the incident light and the acoustic signal depends on the transit
time required for the acoustic pulse to arrive at the detector, which provides information
about the physical location of the absorbing chromophores relative to the position of the
detector. Photoacoustic spectroscopy has been used to separate the absorption spectra of
stratified chromophores by exploiting the effect of depth on the phase shift. Figure 1-22
shows the PA spectrum of an oat seedling. Similar measurements have been made on
animal tissues, including intact human skin. A major application of PAS involves pho-
tochemical energy conversion in biological systems. Since photochemical reactions com-
pete with the decay of metastable states, a decrease of heat production can be related to the
energy stored in chemical products. Photoacoustic spectroscopy has been. used to study
photosynthetic energy storage in chloroplasts and intact leaves. Very weak absorptions
have been measured with PAS, including the visible absorption spectra of water, absorbed
substances on various substrates, and aerosols.
Other techniques, related to PAS, are based on alternative detection methods. "Pho-

Fig. 1·22. Photoacoustic spectra of an oat seedling(28).

The sample was illuminated with modulated light (l2 Hz) at
different wavelengths (lO-nm bandwidth), and the rate of
energy absorption was measured with a detector of acoustic
waves. The phase shift between the photoacoustic signal ...J

and the exciting light depends on the distance the sound ~

wave must travel from the chromophore to the detector. The \!)
(f)
solid line was obtained by adjusting the phase shift of the
lock-in amplifier to null the absorption of chloroplasts from
500 to 650 nm; the strong absorption below 400 nm corre-
sponds to the cuticular coat. The dashed line was obtained -_/ '-"------
with the phase shift adjusted to emphasize the chlorophyll 800 700 600 500 400 300
absorption in the chloroplasts. WAVELENGTH (nm)
42 Chapter 1

tothennal radiometry" is the measurement of the IR radiation generated by optical absorp-

tion, with the advantage that there is no physical contact between the sample and the IR
detector. "Photorefractive" methods are based on the changes in the refractive index of
the medium induced by local heating. In a variation of this method, the defocusing of an
auxiliary CW laser beam by the refractive index gradient in the medium is monitored with
a photodetector located behind a pinhole; it is referred to as "thennallensing." A new
generation of PAS instrumentation is under development, utilizing the analytical tech-
niques currently employed in Fourier transfonn IR spectrometers. The sample is illumi-
nated with polychromatic light and the resultant intensity of the photoacoustic signal is
measured as a function of the path differences in a Michelson interferometer.(8) The
"spectrum" of intensity vs. path difference is converted to the desired spectrum of
intensity vs. frequency by mathematical operation, the Fourier transfonnation.

1.5.5. Action Spectra

The concept of action spectra dates to the 19th century, when biologists observed
that the rate of plant growth depends on the spectral region of the ambient light. The
quantitative aspects of action spectroscopy have been developed during the past 30 years,
leading to a better understanding of the acceptable conditions and practices required to
draw reliable conclusions from the results. An action spectrum is obtained by measuring
the light dose required to evoke the same biological response at different wavelengths.
Assuming that a higher light dose implies a weaker absorption by the responsible chro-
mophores and vice versa, a plot of the reciprocal of the light dose against wavelength
should parallel the absorption spectrum of the chromophores. Jagger(4) elucidated the
conditions whereby action spectra may be expected to parallel absorption spectra. These
include

1. The action mechanism must be the same at all wavelengths.

2. The quantum yield must be the same at all wavelengths.
3. The absorption spectrum of the chromophore must be the same in vitro and in
vivo.
4. Absorption of light by inactive chromophores and light scattering must be
negligible.
5. Only a small fraction of the incident light must be absorbed by the sample in the
wavelength range of interest.
6. Reciprocity of the exposure time and fluence rate must hold.

Action spectra are obtained by comparing the number of photons required for the
same biological effect at different wavelengths. It is convenient to borrow the concept of
cross-section from atomic physics as a measure of the response of a single microscopic
object to particulate radiation. The absorption cross section (<Tabs) of a molecule is propor-
tional to the probability of light absorption at a given wavelength. The absorption cross-
section and molar extinction coefficient are two different ways of specifying the same
molecular property. At a given wavelength they are related by

<Tabs = 3.82 X 10- 21 E cm2 (1-9)

Photophysics 43

The absorption cross-section of a strongly absorbing molecule is comparable to its phys-

ical cross-section. According to Eq. (1-9), the extinction coefficient of a molecule of
approximately O.I-nm radius is the order of 105 liters/mol-cm, which is typical of a
strongly absorbing pigment molecule. The product of the absorption cross-section and the
photon fluence is the probability that one photon will be absorbed by a molecule exposed
to that fluence. For a thin slab of thickness d and unit cross-sectional area, the number of
absorbed photons N after exposure to fluence D at a given wavelength is

(1-10)

where Cabs is the average concentration of chromophores in the material. If the observed
response is the same at two wavelengths, the incident fluence levels required for this
response are related by

(1-11)

where <P is the quantum yield of the biological effect. The left side of Eq. (1-11) is the
relative efficiency of the photoeffect at two wavelengths. A plot of the relative efficiency
vs. wavelength is the action spectrum. This plot parallels the absorption spectrum of the
chromophores involved in the biological effect, if <PI = <P z , and the system meets the
other criteria specified above.
The inactivation of cells by UV radiation represents an important application of
action spectra. The major UV chromophores in cells are nucleic acids and proteins, with
absorption maxima near 265 and 280 nm, respectively (Chapter 2). Because of spectral
overlap, it is virtually impossible to irradiate a cell with light absorbed only by nucleic
acids or proteins. However, when the action spectrum for cell inactivation, or another

2,-,-----,----,----,----,

I
8
-4 6
If) 10
If)
w 4
z
w
>
f= 2
u
W
lJ..
lJ..
W

w 8
>
I- 6
<:( -5
d 10
Fig. 1-23. Action spectrum for bacteria killing. The points a::
4

are the reciprocal of the incident energy required for 50%

killing of E. coli at each wavelength. The 260-nm band 2
corresponds to absorption by nucleic acids. [Adapted from F.
L. Gates, A study of the bacteriacidal action of ultraviolet
light. III. The absorption of ultraviolet light by bacteria, J.
Gen. Physiol. 14,31-41 (1930).]
44 Chapter 1

100r-------r:~

Ul 80
Ul
w
Z
W
>
j:::: 60
U
W
lL.
lL.
W

~ 40
I-
<t
--1
W
a::
20 Fig. 1-24. Action spectra for retardation of cleavage
in the sea urchin for irradiated spenn (e) and irradiated
eggs (0). The 280-run band corresponds to protein
OL-_ _- L_ _ _L -_ _-L_ _- - J absorption and the 260-nm band corresponds to nucleic
240 260 280 300 320 acid absorption. [Adapted from A. C. Giese, Radiation
WAVELENGTH (nm) and cell division, Q. Rev. Bioi. 22, 253-282 (1947)].

photoprocess, matches that of either absorber, it can be concluded that damage to the
corresponding chromophore was a significant factor in the photoprocess under observa-
tion. An action spectrum for killing of bacteria, reported more than 50 years ago, is
reproduced in Fig. 1-23. The peak near 260 nm indicates that the nucleic acids are a major
lethal target. The results in Fig. 1-24, also based on early work, show that nucleic acid
damage is a major factor in the retardation of cleavage when sea urchin sperm are UV-
irradiated, but protein damage is the dominant factor when sea urchin eggs are UV-
irradiated. The action spectra for UV -irradiated mammalian cells have been reviewed. (30)

1.6. REFERENCES

1. L. R. Koller, Ultraviolet Radiation, 2nd ed., Wiley, New York (1965).

2. J. A. Parrish, R. R. Anderson, F. Urbach, and D. Pitts, UV-A, Plenum Press, New York (1980).
3. J. G. Calvert and J. N. Pitts, Photochemistry, Wiley, New York (1966).
4. J. Jagger, Introduction to Research in Ultraviolet Photobiology, Prentice-Hall, Englewood Cliffs, N.J.
(1967).
5. O. Svelto, Principles of Lasers, 2nd ed., Plenum Press, New York (1982).
6. J. T. Verdegen, Laser Electronics, Prentice-Hall, Englewood Cliffs, N.J., (1981).
7. R. S. Longhurst, Geometrical and Physical Optics, 2nd ed., Wiley (1967).
8. R. W. Ditchburn, Light, 2nd ed., Interscience (1963).
9. F. A. Jenkins and H. E. White, Fundamentals of Optics, 4th ed., McGraw-Hill, New York (1976).
10. W. B. Allan, Fiber Optics, Plenum Press, London (1973).
11. L. Reynolds, C. C. Johnson, and A. Ishimaru, Diffuse reflectance from a finite blood medium: Applica-
tions to the modeling of fiber optic catheters, Appl. Opt. 15, 2059-2067 (1976).
12. c. A. Parker, Photoluminescence of Solutions, Elsevier, Amsterdam (1968).
13. C. S. Rupert, Dosimetric concepts in photobiology, Photochem. Photobiol. 20, 203-212 (1974).
14. D. Sliney and M. Wolbarsht, Safety with Lasers and Other Optical Sources, Plenum Press, New York
(1980).
Photophysics 45

15. C. G. Hatchard and C. A. Parker, A new sensitive chemical actinometer. II. Potassium ferrioxalate as a
standard chemical actinometer, Proc. Roy. Soc. (London) A235, 518-536 (1956).
16. L. O. Svaasand and R. Ellingsen, Optical penetration in human intercranial tumors, Photochem. Photobiol.
41, 73-76 (1985).
17. R. R. Anderson and J. A. Parrish, in: The Science of Photomedicine (J. D. Regan and J. A. Parrish, eds.),
pp. 147-194, Plenum Press, New York (1982).
18. G. Kortum, Reflectance Spectroscopy, Springer-Verlag, Berlin (1969).
19. W. L. Butler and K. H. Norris, The spectrophotometry of dense light scattering material, Arch. Biochem.
Biophys. 87, 31-40 (1960).
20. W. L. Butler, Absorption of light by turbid materials, J. Opt. Soc. Am. 52,291-299 (1962).
21. T. J. Dougherty, in Methods in Porphyrin Photosensitization (D. Kessel, ed.), pp. 313-328, Plenum Press,
New York (1985).
22. G. W. King, Spectroscopy and Molecular Structure, Holt, Rinehart and Winston, New York (1964).
23. N. J. Turro, Modern Molecular Photochemistry, Benjamin/Cummings, Menlo Park, Cal. (1978).
24. A. J. Pesce, c.-G. Rosen, and T. L. Pasby, Fluorescence Spectroscopy, Marcel Dekker, New York (1971).
25. R. V. Bensasson, E. J. Land, and T. G. Truscott, Flash Photolysis and Pulse Radiolysis, Pergamon Press,
Oxford (1983).
26. J. R. Lakowicz, Principles of Fluorescence Spectroscopy, Plenum Press, New York (1983).
27. C. R. Cantor and P. R. Schimmel, Biophysical Chemistry, Part II, W. H. Freeman, San Francisco (1980).
28. T. A. Moore, in: Photochemical and Photobiological Reviews, Vol. 7 (K. C. Smith, ed.), pp. 187-221;
Plenum Press, New York (1983).
29. A. C. Tam, in: Ultrasensitive Laser Spectroscopy (D. S. Kliger, ed.), pp. 1-108, Academic Press, New
York (1983).
30. T. P. Coohill, Action spectra for mammalian cells in vitro. In Topics in Photomedicine (K. C. Smith, ed.),
Plenum Press, New York (1984).
 2
Photochemistry

2.1. Fundamentals of Photochemistry. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47

2.1.1. Overall Reactions in Photochemistry ............................................ 49
2.1.2. Primary Reactions in Photochemistry .......................................... " 51
2.1.3. Photochemical Kinetics ....................................................... 54
2.2. Photochemistry of Proteins and Amino Acids ........................................... 57
2.2.1. Photochemistry of Amino Acids ................................................ 58
2.2.1.1. Initial Photochemistry of Aromatic Amino Acids........................... 58
2.2.1.2. Initial Photochemistry of Cystine, Histidine. and Peptide Bonds .............. 59
2.2.2. Photochemistry of Proteins .................................................... 60
2.2.2.1. Initial Photochemistry of Proteins ....................................... 61
2.2.2.2. Theories of Enzyme Inactivation ........................................ 62
2.2.2.3. Kinetics of Enzyme Inactivation ........................................ 64
2.3. Photochemistry of Nucleic Acids ..................................................... 65
2.3.1. Photochemistry of Nucleic Acid Constituents ..................................... 66
2.3.1.1. Deoxyribose......................................................... 66
2.3.1.2. Purines ............................................................. 66
2.3.1.3. Pyrimidines ......................................................... 66
2.3.1.3a. Hydration Products .......................................... 66
2.3.1.3b. Thymine Glycols...... ... . .. . .... . .. . .... ... .... . .... .... . .. 67
2.3.1.3c. Other Modifications of Pyrimidines ............................. 68
2.3.1.3d. Cyclobutane-Type Dimers of Thymine. Cytosine, and Uracil........ 68
2.3.1.3e. Spore Photoproduct .......................................... 70
2.3.1.3f. Pyrimidine-Pyrimidine (6-4) Adducts . . . . . . . . . . . . . . . . . . . . . . . . . . .. 71
2.3.1.3g. Heteroadducts of the Pyrimidines.. .... . .... ... .... . ... .... ... .. 71
2.3.2. Photochemistry of DNA ...................................................... 72
2.3.2.1. DNA Strand Breaks .................................................. 72
2.3.2.2. DNA-DNA Cross-Links. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 72
2.3.2.3. DNA-Protein Cross-Links ............................................. 72
2.3.3. Photochemistry of RNA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 74
2.3.4. Detection and Quantitation of Photochemical Lesions in DNA ....................... 75
2.4. References ....................................................................... 76

2.1. FUNDAMENTALS OF PHOTOCHEMISTRY

The central principle in photochemistry, first stated by Grotthus and Draper in 1818, is
that only absorbed light can produce a chemical change. The initial step in a pho-

Leonard I. Grossweiner • Physics Department, Illinois Institute of Technology, Chicago,

Illinois 60616. Kendric C. Smith • Department of Radiation Oncology, Stanford Univer-
sity School of Medicine, Stanford, California 94305.

47
48 Chapter 2

tochemical reaction at ordinary light intensities is the absorption of a single photon by a

molecule, and the production of an excited state in which one electron of the absorbing
molecule is raised to a higher energy level (Chapter 1). The subsequent events depend on
the molecular structure, the wavelength of the light, and the specific reaction conditions.
A certain fraction of the excited molecules may release their excess energy by a sequence
of "relaxation" processes, leading back to the original unexcited molecules. Alter-
natively, a photochemical process may take place at some stage of the relaxation chain,
either spontaneously or via interactions with other molecules. The quantum yield of a
given photochemical product is the probability that the absorption of one photon by the
system leads to one molecule of that product. The fIrst chemically reactive species formed
by light absorption are referred to as "primary products," typically metastable excited
molecular states and free radicals. In most cases light is not involved in the subsequent
reactions, and they are referred to as "dark" or "thermal" reactions. However, some
complex photobiological processes may utilize light in more than one stage, e.g., there
are two light-activated steps in photosynthesis by green plants (Chapter 12). A pho-
tobiological process takes place in stages, starting with the very fast processes at the
physical level, followed by dark chemical reactions, and terminating with relatively slow
events at the biological level (Fig. 2-1). The photophysical processes ofbiomolecules are
described in Chapter 1. Their photochemical reactions may be described by the well-
developed concepts of organic reaction mechanisms. (1) An alternative approach, to be
followed here, emphasizes the photochemistry of proteins and nucleic acids, which are the
two most important classes of functional biomolecules in cells. These "colorless" mac-

I fs I JLS Is I yr Myr

DARK REACTIONS OF PRIMARY PHOTOPRODUCTS

SINGLET STATES }
EXC ITED STATE REACTIONS
_ _ _ _ _ _ _ _ _ TRIPLET STATES

_ _ _ _ _ _ _ _ _ _ _ _ _ _ BIOLOGIC RESPONSES

_ _ _ _ _ _ _ _ _ _ _ TRIPLET EXCITATION TRANSFER

_ _ _ _ _ SINGLET EXCITATION TRANSFER

_ _ VIBRATIONAL RELAXATION

_ LIGHT ABSORPTION EVOLUTIONARY EFFECTS _ _ _ _ __

-15 -10 -5 o 5 10 15

log t(s)

Fig. 2-1. Time scales in photophysics, photochemistry, and photobiology. The lower scale is the logarithm of
the time (s) at which the events indicated by the horizontal lines take place. The upper scale locates the time
intervals in ordinary units. [Adapted from N. J. TUITO and A. A. Lamola, Chap. 3, in The Science of Pho-
tobiology, 1st ed. (K. C. Smith, ed.), Plenum Press, New York (1977).]
Photochemistry 49

romolecules absorb in the UV region of the spectrum. Although many different types of
proteins and nucleic acids are present in a given cell, characteristic photochemical reac-
tions have been identified that are relevant to biological damage induced by the exposure
of organisms to UV radiation. In addition, the effects induced by exposing cells to UV
radiation under controlled conditions provide a sensitive probe of biological responses to
molecular alterations, e.g., DNA repair. Other important classes ofbiomolecules absorb
both UV and visible radiations, including photosensitizing drugs and the chromophores of
natural processes. The photochemistry of "colored" molecules is considered in Chapter 3
and subsequent chapters on specific photoprocesses in biology.

2.1.1. Overall Reactions in Photochemistry

An overall photochemical reaction specifies only the initial reactants and stable
products. Some important overall reactions with examples in photobiology are as follows:
1. Linear addition to an unsaturated molecule:

o (2-1)
II NHz
~
CH3
HN
I I + HS-CHz~HCHCOOH I hv
)

o'l"' ........ N H
H

THYMINE CYSTEINE

o
" ~H2
~
S-CH2CHCHCOOH
HN CH 3 Ii
) H
o""'........ N H
H

5-S-CYSTEINYL-6-HYDROTHYMINE

The addition of cysteine to thymine serves as a model for the photochemical cross-linking
of DNA to protein (Section 2.3.2.3).
2. Cycloaddition of unsaturated molecules:

o o 0 (2-2)
II CH II H3C CH 3 I
2 HN~ 3 ___ h_V~) HN~H
o
~~N)
H O~~N~N)~
H H
THYMINE THYMINE DIMER

In cycloaddition reactions, two molecules or molecular groups on a macromolecule react

to form a ring product. Thymine dimers are an important class of molecular lesions
generated by the action of UV radiation on DNA (Section 2.3.1.3).
50 Chapter 2

3. Photofragmentation:

(2-3)

RIBOFLAVIN

H3

H C
3
C
r):
I ::CN)~# N
H
N
\I
NH

LUMIFLAVIN
o
II
+ CH 3CCHOHCHOHCH zOH

In photofragmentation reactions, the original molecule is converted to smaller products by

photolytic cleavage at one or a few chemical bonds. These reactions take place in essen-
tially all biomolecules, if exposed to UV radiation of sufficiently short wavelengths.
4. Photooxidation:

D5D
CH 3 CH z CH z (2-4)
' c / "- / "-
3C CH2 CH~C~:
hv
H3 C --4
DYE

HO ""

CHOLESTEROL

3B-HYDROXY-5a-HYDROPEROXy-~6-CHOLESTENE

The formation of this "5a" photooxidation product of cholesterol is a selective test for
the involvement of singlet molecular oxygen (Chapter 3).
5. Photohydration:
0
° (2-5)

HN:)
II

O~~N H2O
hv
~
~f"
HN

O""'~NH
--H
_OH
'-....
H
H

URACIL 6 ·HYDROXY -5' HYDROURACI L

Photochemistry 51

Photohydration products are another class of lesions induced by the UV irradiation of

DNA (Section 2.3.1.3).
6. Cis-trans isomerization:

(2-6)

This process leads to a 1800 rotation around a double bond. Photoisomerization of ll-cis-
retinal is the fIrst photochemical step in retinal vision (Chapter 9).
7. Photorearrangement:

(2-7)

hv
---'J>
SKIN

7-DEHYDROCHOLESTEROL

C~3
H -C-CH2- CH2-C72
H3 C H3C- C - H
I
CH3

HO
VITAMIN D3

In a photorearrangement reaction, the original and product molecules are made up of the
same atoms. The production of vitamin D3 in mammalian skin is initiated by the UV
irradiation of 7-dehydrocholesterol (Chapter 6).

2.1.2. Primary Reactions in Photochemistry

The overall photochemical reactions indicated in Section 2.1.1 lead to stable prod-
ucts, except when the products themselves undergo subsequent photochemical reactions,
e.g., thymine dimers are photodissociated by UV-C radiation (Section 2.3.1.3). The
primary processes usually lead to unstable intermediates that undergo additional reactions
under dark conditions. The excitation pathways of a typical organic biomolecule are
shown in a Jablonski diagram (Fig. 1-16). In addition to the physical relaxation processes
52 Chapter 2

indicated on the diagram, competing photochemical reactions may take place in the
excited states. In a unimolecular reaction, an excited state reacts without the intervention
of another molecule. The solvent may be involved in apparent unimolecular reactions in
condensed media, e.g., in the dissociation of an electronically excited acidic molecule,
the proton is transferred from the excited molecule to a molecule of the solvent. Most
elementary reactions in fluid media are bimolecular, and require proximity between the
two reacting species. The short lifetime of excited singlet states, typically 1-20 ns, limits
their bimolecular reactions to nearby reaction partners. The lifetime of triplet states may
exceed 1 ms, and they may diffuse significant distances prior to reacting. Excited state
reactions usually lead to short-lived free radical products, whose subsequent reactions are
"diffusion-limited," i.e., the slowest step is the time required for diffusional encounters
of the reaction partners. Diffusion-limited reactions are very fast in water and other low-
viscosity solvents, and may be quite slow in high-viscosity media such as organic glasses.
In contrast, many chemical reactions between stable molecules require "thermal activa-
tion" in order to provide the energy necessary to overcome potential energy barriers
between the reactants and the products. A small temperature rise can increase the rate of a
thermally activated reaction many fold, but it has a relatively small effect on a diffusion-
limited reaction, i.e., only to the extent that it affects the viscosity of the medium.
Two general methods have been used to study primary photochemical processes.
Fast reaction techniques make it possible to study short-lived species in real time, e.g., in
flash photolysis, detectable concentrations of the intermediates are generated by an in-
tense light flash, and their subsequent reactions are followed in real time by making fast
optical absorption measurements (Fig. 1-19). The picosecond time resolution available
with mode-locked lasers can probe virtually all photophysical and photochemical pro-
cesses. Alternatively, dry or frozen systems may be used to slow down the initial reactions
of photochemical intermediates, facilitating the use of conventional analysis methods,
e. g., optical spectroscopy and magnetic resonance. Some important types of primary
photophysical and photochemical processes are:
1. Singlet-singlet excitation energy transfer:

(2-8)

where * labels an electronically excited state, and the superscript indicates the spin
multiplicity (Section 1.4.1). Reactions of this type can occur by short-range (exchange)
and long-range (Forster) mechanisms. Exchange transfer is efficient only when the two
molecules are in contact, and when the excitation energy of the donor exceeds that of the
acceptor, i.e., the energetics are "downhilL" The rate of long-range transfer depends on
the spectral overlap between the fluorescence band of the donor (D) and the absorption
band of the acceptor (A), the distance between the two species, and their relative orientia-
tions. The effective range can be 5-10 nm for favorable conditions (Section 1.4.4).
2. Triplet-triplet energy transfer:

(2-9)
Photochemistry 53

In this process, an excited triplet state of the donor generates an excited triplet state of the
acceptor. Triplet-triplet transfer takes place by an exchange mechanism, requiring close
contact and downhill energetics, i.e., the triplet energy of the donor must be larger than
the triplet energy of the acceptor. The long lifetime of triplet states compared to excited
singlet states can make triplet-triplet transfer important in photochemical systems. Two
triplet states can react also, leaving one molecule in an excited singlet state and the other
in the ground state. This process is referred to as "triplet-triplet annihilation," and
generally requires two photons per transfer in order to generate the two triplet states.
However, the ground state of molecular oxygen is a triplet, and the excited triplet states of
many photosensitizers can transfer energy to molecular oxygen, thereby generating elec-
tronically excited molecular oxygen or "singlet oxygen" (Chapter 3).
3. Excited state electron transfer:

D* + A ~ D+ + A- (2-lOa)
A*+D ~ A-+D+ (2-lOb)

Many excited singlet and triplet states of photo sensitizers can act as electron donors
(reducing agents) or electron acceptors (oxidizing agents). The configuration diagram in
Fig. 1-15a shows that the S 1 and T, states have unpaired electrons in two energy levels. It
may be energetically favorable to either add or release an electron, leading to anionic or
cationic free radicals, respectively. The electron transfer act may be followed by a fast
proton transfer in polar solvents, in order to form the most stable acidity state of the
radical ion. The triplet states of some sensitizers can act as both energy and electron
transfer agents, e.g., the triplet state of the common dye, eosin Y, transfers energy to
molecular oxygen via Eq. (2-9), oxidizes aromatic compounds (e.g., tryptophan) via Eq.
(2-1 Oa), and reduces strong oxidants such as ferricyanide ion via Eq. (2-lOb). (2)
4. Photolytic production of free radicals:

(2-11)

Photocleavage may take place via different reaction pathways. One possible mechanism
involves a nonradiative transition from an unrelaxed (vibrationally excited) excited singlet
state to a "dissociative" state of the molecule, followed by bond rupture.
5. Photoionization:

(2-12)

Most molecules are photoionized in the gas phase by sufficiently energetic UV radiation,
typically in excess of 7 eV. Photoionization of inorganic anions (e.g., 1-, Br-, NO:3>,
small aromatic compounds (e.g., phenol, tyrosine, tryptophan), and some dyes occurs in
solution at photon energies lower than 4 eV. The ejected electron is temporarily stabilized
in the medium as a "solvated electron" (Section 2.2.1). Photoionization of aromatics and
dyes also takes place at low temperature in frozen aqueous and hydrocarbon media.
54 Chapter 2

6. Hydrogen atom abstraction by carbonyl compounds:

(2-13a)

RI-C-R2 + R3 H---7) R1-C- R2 + R; (2-13b)

1/ I
0* OH

This type of reaction involves organic ketones and an oxidizable constituent (R3 H).
Carbonyl derivatives have photochemical properties derived from (n,1T*) excited states
(Section 1.4 .2), in addition to the usual (1T, 1T*) excited states of unsaturated hydrocar-
bons. The lowest triplet state has (n,1T*) character in some carbonyl derivatives (e.g.,
acetone, acetophenone, benzophenone) and (1T ,1T*) character in others (e.g., 2-acetyl-
naphthalene, 4-phenylbenzophenone, fluorenone). Hydrogen abstraction is an important
reaction of (n,1T*) triplet states, e.g., irradiation of benzophenone in isopropanol from
320 to 360 nm leads to the disappearance of benzophenone with a quantum yield close to
2. In the first step, the (n,1T*) excited state ofbenzophenone abstracts an H atom from the
alcohol, thereby reducing the ketone to a ketyl radical and oxidizing the alcohol to a free
radical. Each alcohol radical then reduces another molecule of the ketone in a dark
reaction. This reaction exemplifies a process whereby the absorption of one photon leads
to the disappearance of more than one reactant molecule. In more complicated processes,
known as "chain reactions," an intermediate capable of further reacting with an initial
reactant is continuously restored by a sequence of intermediate steps. Photopolymeriza-
tion reactions are an important class of chain reactions.

2.1.3. Photochemical Kinetics

The order of a chemical reaction specifies how the overall rate depends on the
concentrations of the reactants. The rate of a first-order reaction increases linearly with a
concentration, the rate of a second-order reaction increases with the square of a concentra-
tion, etc. Reaction order is a macroscopic concept, and it is not always equivalent to the
molecularity of the elementary processes that combine to give the overall reaction. For
example, A ~ B is first order overall and first order with respect to species A. However,
the elementary reaction may not be unimolecular if solvent molecules are involved. The
reaction A + B ~ C is second order overall and first order with respect to A or B. It may
also be bimolecular if the reaction actually takes place by collisions between species A
and B. The rate constant of a reaction specifies the rate at unit concentrations of the
reactants, and given conditions of the solvent, temperature, and any other parameters that
affect the reactivity. Concentrations in dilute solution are usually expressed in molar
units, in which case first-order rate constants have units of s -I, and second-order rate
constant have units of liters mol - 1 S - I. The basic unit of light quantity in photochemical
dosimetry is fluence, defined as the number of photons of specified wavelengths in a
collimated beam incident in a perpendicular direction on a unit area of the reaction vessel.
The relationship between incident fluence and absorbed dose depends on the conditions of
the experiment. If a well-stirred photochemical system absorbs all incident light, then the
Photochemistry 55

absorbed dose equals the total number of incident quanta divided by the volume of the
system. The other extreme is an "optically dilute" system, whereby only a small fraction
of the incident photons are absorbed. In this case, the absorbed dose is proportional to the
product of the concentration of the absorbing molecules and the optical pathlength. (This
result follows from Beer's law; Section 1.5.1). The calculation of absorbed dose is more
complicated in intermediate cases where neither approximation applies.
The conventional analysis of photochemical kinetics will be illustrated with a typical
reaction, in which constant optical excitation of a reactant (A) generates an excited state
(A*), which may decompose to give a permanent product (A') in competition with the
relaxation of A * back to A:

A
hv

A*~
/ (2-14)

A' (product)

The kinetics analysis may be simplified by assuming that the processes by which the
short-lived intermediate A * is formed and relaxes back to A are much faster than the
reactions leading to permanent products, in this case species A'. This approximation is
usually valid when unstable intermediates are generated at ordinary light intensities, e.g.,
for metastable triplet states and reactive free radicals. Reaction kinetics are expressed
mathematically by equating the concentration changes of each species to the difference
between the rate at which that species is formed and the rate at which it disappears. For
species A * this relation is

d[A*]ldt = cf>o(dDldt)IN - (k l + k 2 )[A*] = 0 (2-15)

where (dD I dt)1N is the rate of light absorption by species A in units of quanta per liter per
second divided by the Avogadro number (N), cf>o is the quantum yield for the pho-
tochemical production of A *, kl (s - I) is the first-order rate constant for the decay of A *
to A, and k2 (s -I) is the first-order rate constant for the formation of the final product A'
from A *. The left side of Eq. (2-15) is approximately zero, because the concentration of
short-lived species A * is essentially constant. The "steady-state" concentration of A *
from Eq. (2-15) is

[A*] = cf>o(dDldt)/(k l + kz) (2-16)

where [A*] is much smaller than [A] and [A'] in the present approximation. The pho-
tolysis rate of the initial reactant A at any instant in time is given by

(2-17)
56 Chapter 2

The initial quantum yield for the disappearance of A, <I> -A' is defined as d[AJ/dD at t =
0, i.e.,

(2-18)

Equation (2-17) shows that the fraction of absorption events leading to the photolysis of A
equals the ratio of the rate constant for product formation to the sum of the rate constants
for product formation and relaxation of A * back to A. The relationship between absorbed
dose (D) and incident fluence (F) depends on the reaction conditions. If all of the incident
light is absorbed by species A, integration of Eq. (2-17) leads to a linear dependence of
[A] on the incident fluence:

(2-19)

where a is a constant equal to 1000/Nd, d is the pathlength of the irradiation vessel, [AJ o
is the initial concentration of A, and F is the incident fluence (quanta cm - 2 S - 1).
(Centimeter units are frequently employed in photochemical kinetics because extinction
coefficients are expressed in units of liters mol- 1 cm - 1.) The other limiting approxima-
tion applies to "optically dilute" solutions of A in an inert medium, such that the
absorbance of A is much smaller than unity (Section 1.5.1). In this case, integration of
Eq. (2-17) shows that the disappearance of A follows an exponential function of the
incident fluence:

[AJ/[AJ o = exp - b<l> -AF (2-20)

where b is a constant equal to (1000 10gelO)EA/N, and EA is the molar extinction coeffi-
cient of A at the irradiation wavelength. [The quantity b is the absorption cross-section of
species A, as defined by Eq. (1-9).]
The experimental implimentation of this kinetics analysis requires much effort and
attention to details. The incident fluence must be determined with appropriate radiometry,
and assays must be calibrated for the loss of reactant A and the formation of product A'.
The kinetics scheme of Eq. (2-14) can be tested by the fit of the yield data (i.e., A vs. For
A' vs. F) to either Eq. (2-19) for completely absorbing solutions or Eq. (2-20) for
"optically dilute" solutions. The fitting procedure also leads to the value of <I> _ A' which
should be constant for different values of [AJ o. The ratio of kl to k2 can be calculated from
Eq. (2-19), if <1>0 is known from other measurements, e.g., if A * is a triplet state, then <1>0
is the triplet yield of A. In this case, k) is the reciprocal of the triplet state lifetime, and the
value of k2 can be determined. In a thorough study, the measurements would be carried
out over a range of temperature, solvent properties, and possibly at several wavelengths.
Relatively few photochemical reactions have been completely analyzed. For most applica-
tions in photobiology, a minimally satisfactory level of understanding requires the identi-
fication of the more important photochemical intermediates and permanent products,
measurements of the key quantum yields and rate constants, and specification of reason-
able reaction mechanisms.
Photochemistry 57

2.2. PHOTOCHEMISTRY OF PROTEINS AND AMINO ACIDS(2)

Proteins are the most abundant molecules in cells, except for water, and the most
diverse of all biomolecules in size, form, and function. Functional proteins may be highly
sensitive to light because their biological properties are controlled by relatively small
regions of the entire macromolecule, e.g., the activity center of an enzyme. The absorp-
tion of light by a protein is localized at those chromophoric groups with significant
extinction coefficients at the incident wavelengths. However, the subsequent events may
not be confmed to the initial sites because excitation energy and electrons can migrate
within a large molecule. Protein photochemistry has been investigated with many tech-
niques, including assays of permanent residue destruction, flash photolysis, electron spin
resonance detection of photochemical intermediates, and circular dichroism studies on
conformation changes. A reasonable starting point for the study of protein photochemistry
is the photochemistry of the chromophoric amino acids that are responsible for most of the
light absorption. For wavelengths longer than 240 nm, the major absorbers in "colorless"
proteins are the aromatic amino acids, phenylalanine (Phe), tyrosine (Tyr) , and tryp-
tophan (Trp), with smaller contributions from histidine (His), cystine (eys), sulfhydryl,
and the peptide bonds (Fig. 2-2). (These abbreviations for chromophoric amino acids are
frequently used in the photochemical literature and may not be the same as those em-
ployed in standard texts on biochemistry.) The carboxyl and amino end groups absorb
below 240 nm, and all amino acids are strong absorbers of UV radiations shorter than
about 200 nm.

6000r---~-----.----.-----.-----.----.----~

.. 5000

. o
~

~ 4000

r.:
z
l.t.J
U 3000
iL
LL
l.t.J
o
u 2000
Z
o
i=
u
z
i=
x
l.t.J

Fig. 2-2. Molar extinction coefficients of aqueous tryptophan (Trp), tyrosine (Tyr), and phenylalanine (Phe).
[Adapted from Handbook of Biochemistry, 2nd Ed. (H. A. Sober, ed.), The Chemical Rubber Co., Cleveland
(1970).] These are the only common amino acids with significant absorptions above 240 nm. The histidine
absorption maximum is 211 nrn. Cystine has a strong absorption at 207 nm and a weak peak near 250 nm.
58 Chapter 2

2.2.1. Photochemistry of Amino Acids

Aqueous solutions of amino acids are decomposed by UV radiation, usually leading
to many products. Much of the early work on amino acid photochemistry was done with
254-nm radiation from low-pressure mercury arcs. For example, irradiation of His at 254
nm gave ammonia, histamine, imidazole, acetaldehyde, and other products.(3) Some of
the final products may result from the continued photolysis of earlier products, and
therefore their occurrence may not provide useful information about the photochemistry of
the starting amino acid. The early stages of amino acid photochemistry have been investi-
gated in frozen systems and with flash photolysis, especially for the aromatic amino acids.
There are numerous reports of protein and amino acid destruction by longer wavelength
radiations, where the extent of light absorption is very low, e.g., aqueous solutions in
glass vessels exposed to strong sunlight. These reactions may involve photochemical
processes initiated by the weak light absorption in the long-wavelength tail of the amino
acid spectra or, more likely, photosensitization by low concentrations of impurities that
absorb longer wavelength light.

2.2.1.1. Initial Photochemistry of Aromatic Amino Acids(4-6)

Prior to the 1960s, it was generally assumed that bond cleavage is the major UV
photolysis process in aromatic amino acids. However, early flash photolysis experiments
on aqueous solutions of aromatic amino acids led to the conclusion that photoionization
(electron ejection) is the most important primary act. (7) Flash photolysis transient spectra
of aqueous tyrosine led to a phenoxyl-type aromatic free radical (Tyr·), formed by disrup-
tion of the phenolic OH bond, and an intense, short-lived, broad absorption band peaking
in the red region. Aqueous tryptophan gave a different aromatic radical absorption (Trp·),
identified with photolysis of the indole ring, and a similar red absorption band. Flash
photolysis of aqueous phenylalanine led to the benzyl radical absorption (C 6 H5 CH 2 -) and
the red absorption band. The red band was observed only at short times after the irradia-
tion flash «1 /-ls) and was absent when oxygen, hydrogen peroxide, or nitrous oxide was
present. These properties characterize the "hydrated" electron, i.e., a solvated electron
in water, which was first identified by Hart and Boag in "pulse radiolysis" transient
spectra of water. (8) (Pulse radiolysis is similar to flash photolysis, except that the radiation
is a short, intense pulse of high-energy electrons or other ionizing radiation, instead of
light.) A hydrated electron is a quasi-free electron, trapped by the interactions of its
electric charge with water dipoles. It is an unstable chemical species, with characteristic
physical and chemical properties. Hydrated electrons are one of the strongest known
reducing agents in the absence of oxygen, and they react rapidly with oxygen to form the
superoxide radical anion (02)' the precursor of hydrogen peroxide. The absorption
spectra of these aromatic amino acid free radicals and the hydrated electron obtained by
flash photolysis are reproduced in Fig. 2-3. Further study has confirmed that photoioniza-
tion of aqueous aromatic amino acids and their derivatives also takes place at ordinary
light intensities. (4,5)
The stable reaction products reported for the UV photolysis of aqueous tryptophan in
air include tryptamine, N-formylkynurenine (NFK), kynurenine, aliphatic amino acids,
and ammonia. The formation of NFK is especially important in protein photolysis because
Photochemistry 59

16,000

14,000
Fig. 2-3. Absorption spectra of tran-
sient species identified by flash pho-

.g
tolysis of aqueous amino acids: half-
oxidized tryptophan radical (TIp") 12,000

[adapted from J. F. Baugher and L. I.

Grossweiner, Photolysis mechanism ..E
of aqueous tryptophan, 1. Phys. ... 10,000
Chern. 81, 1349-1354 (1977)], half-
~
oxidized tyrosine radical (Tyro) Z
W
[adapted from Ref. 23], benzyl radi- l) Cs Hs CH2
8,000
cal (C 6H5 CH 2o) [adapted from L. J. LL
IJ..
Mittal, J. P. Mittal, and E. Hayon, W
0 TyrO (x 2)
l)
Primary processes in the photochem-
istry of phenylalanine and derivatives
z 6,000 n.
0
in aqueous solution. Biphotonic pho- i= il
.Ni
l)
toionization and photodissociation z
i= \ ~ .
reactions, J. Am. Chern. Soc. 95,
6203-6210 (1973)], hydrated elec-
X
w
4,000
\! \
\/ i / \
/ \TrpO(X2)
tron (e aq) [adapted from J. K. Thom-
as, Methods of production of solv- 2,000 / \ .
ated electrons and their chemical and
\ / \
physical properties, Radiat. Res. \ \/ \
Revs. 1, 183-208 (1968)]. The esti- '-'( \
0
mated accuracy of the extinction co-
efficients is 10%.

this oxidation product of tryptophan can act as a photosensitizer of longer wavelength UV-
A radiation, e.g., NFK has been implicated in the aging of the ocular lens.(9) UV
photolysis of small tryptophan peptides in the absence of air, e.g., glycine-tryptophan-
glycine, led to unusual indole derivatives with fused, eight-member lactam rings and other
complex products. (l0.11) Photolysis of aqueous tyrosine gave many products, including
hydroxylated aromatics (especially dihydroxyphenylalanine or DOPA, the precursor of
melanin), bityrosine, aliphatic amino acids, and ammonia. The reaction mechanism has
not been worked out in detail. The permanent products reported for the continuous
irradiation of aqueous phenylalanine include tyrosine and other phenolics, including
DOPA, aliphatics, and ammonia.

2.2.1.2. Initial Photochemistry of Cystine, Histidine, and Peptide Bonds

Although cystine is a weakly absorbing amino acid, it is especially important in
protein photochemistry because cystine photolysis has a high quantum efficiency and may
lead to the splitting of interchain disulfide bridges. The reported permanent products of
cystine photolysis include sulfhdryl, disulfide, aliphatic derivatives, and hydrogen sul-
fide. (6) The probable primary events are the splitting of S-S bonds, leading to RSO-type
60 Chapter 2

radicals, and the splitting of C-S bonds, leading to -S-S'-type radicals. Hydrated elec-
trons react very rapidly with cystine to form a radical anion, in which an electron is
temporarily trapped at the S-S bond. This reaction has been observed also with cystinyl
bridges in proteins. The absorption by histidine above 240 nm is much weaker than the
other protein chromophores, and it is not an important primary absorber in proteins.
However, a protonated imidazole group is an electron trap, comparable to a disulfide
bridge, and can serve as a temporary site for electrons photoejected from aromatic resi-
dues. (12) The peptide bonds in proteins are weak UV absorbers, but their large numbers
can lead to a role in the protein photochemistry, especially in proteins without aromatic
amino acids, e.g., gelatin. Peptide bonds can act as electron traps, and may provide a path
for electron migration from optically excited aromatic residues to cystine and histidine
sites.

2.2.2. Photochemistry of Proteins

UV irradiation of an aqueous protein solution leads to extensive changes in almost all
of its properties. The "physical" effects include changes in the absorption spectrum,
optical activity, apparent molecular weight, electrophoretic pattern, solubility, heat sen-
sitivity, and other parameters. The "chemical" effects include changes in acid-base
titration curves, increase of digestibility by proteolytic enzymes, modification or destruc-
tion of residues, and changes in catalytic activity of enzymes. Attempts to relate the
overall damage to specific initial photochemical reactions are frustrated by the structural
complexity, e.g., the "small" hormonal protein insulin contains 16 different types of
amino acids, arranged on two polypeptide chains of 21 and 30 units, joined by two
disulfide bridges. The UV absorption of a typical "colorless" protein has a strong "end"
absorption peaking at 180-200 nm, a minimum at 240-250 nm, and the "aromatic" band
maximum near 280 nm. Protein absorption spectra are similar, but not identical, to the
composite spectra of mixtures of the constituent amino acids (Fig. 2-4). The differences
are caused by peptide bonding and interactions between the amino acid residues. As a first
approximation, it may be assumed that the fraction of light absorbed by a given amino
acid in a protein can be estimated from its concentration in the protein and the molar
extinction coefficient of the corresponding aqueous amino acid. However, the pho-
tochemistry of an amino acid residue in a protein may differ significantly from the
aqueous amino acid. The contributing factors include the following:

1. Excitation energy may be transferred from Phe ~ Tyr ~ Trp by the Forster
mechanism [Eq. (2-8)].
2. Electrons generated by the photoionization of aromatic residues [Eq. (2-12)] may
migrate and react at different sites.
3. Tyrosinyl residues can by oxidized indirectly by intramolecular electron transfer
from an intact tyrosine site to an oxidized tryptophan site. (12)
4. The photolytic splitting of disulfide bridges may lead to large changes in the
conformation of the protein, including denaturation.
5. Initial damage to an amino acid residue in a protein may be more stable than in the
free amino acid because the altered site cannot further react by diffusion.
Photochemistry 61

2.0

-:
~

w
u
z
<t
III
0:: 1.0
0
(/)
III
<t

Fig. 2-4. Absorption spectra of ex-chymotrypsin

and a mixture of its constituent amino acids.
(Adapted from Ref. 3.)

The identification of pathways leading from the initial photochemical reactions in a

protein to permanent damage is a challenging problem. Most work has been done with
lower molecular weight proteins, e.g., proteolytic enzymes, hormones, serum proteins,
and lens proteins, especially where there is information about the primary chain sequence
and the three-dimensional structure.

2.2.2.1. Initial Photochemistry of Proteins

Transient spectra obtained by flash photolysis of dilute, aqueous protein solutions
show aromatic free radical bands, similar to those identified by the flash photolysis of
tyrosine and tryptophan, and also the short-lived red absorption band of the hydrated
electron, e.g., the results in Fig. 2-5 reported for the laser flash photolysis of subtilisin
BPN' at 265 nm.(l3) The band maxima, identifying TyrO (410 nm), Trpo (510 nm), and
hydrated electrons (720 nm), have been observed in many other proteins.(2,14) Some
proteins with cystinyl residues had a transient band near 420 nm, identified with the
disulfide electron adduct, e.g., lysozyme, papain, and trypsin. These findings demon-
strate that the dominant initial photochemical reactions in proteins are the photoionization
of aromatic residues, and the temporary trapping of the ejected electrons in the water
medium and in some cases at disulfide bonds. The disulfide electron adduct band was not
suppressed by saturation with oxygen or nitrous oxide, which react rapidly with hydrated
electrons. This result indicates that the ejected electrons did not react with disulfide bonds
subsequent to their localization in the external aqueous medium as hydrated electrons.
Rather, the electrons migrate directly from an optically excited aromatic residue to a
disulfide bridge by a fast, intramolecular process. The absorption spectra of long-lived
tryptophan and tyrosine triplet states also have been observed in the flash photolysis
spectra of proteins. (14) Since triplet states may produce singlet oxygen by energy transfer,
62 Chapter 2

0.20
Fig. 2-5. Transient absorption spectrum of
aqueous, air-free subtilisin BPN', taken at
l00-ns delay after a 17-ns laser flash at 265
-=
Q)

0.15 run. The arrows show the absorption maxima

lLJ
u of half-oxidized tryptophanyl residues
Z
<I (Trp-), half-oxidized tyrosinyl residues
[IJ

~ 0.10
(Tyf"), and hydrated electrons (e. q) (see Fig.
<1l 2-3). This enzyme contains 3 tryptophanyl,
[IJ
<I 10 tyrosinyl, and no cystinyl residues. The
observed transient bands indicate that some
0.05 of the tryptophanyl and tyrosinyl residues
were photoionized, and the ejected electrons
were stabilized in the aqueous medium as
hydrated electrons. The laser irradiation also
led to loss of enzymic activity. None of the
tryptophanyl or tyrosinyl residues are located
near the active center. It was deduced that photolysis of a single tryptophanyl residue led to a small change in the
conformation that inhibited the ability of the enzyme to bind the synthetic substrate used assay the activity.
(Adapted from Ref. 13.)

which can then attack other protein sites, triplet states might be implicated in protein
photolysis reactions. However, quantum yields for the UV photo inactivation of enzymes
are usually independent of oxygen, which does not rule out the attack of singlet oxygen on
a "nonessential" site. Carbonic anhydrase is an exception whereby photoinactivation was
much faster under oxygen; evidence for the production of singlet oxygen by triplet energy
transfer from NFK to oxygen was obtained in this case. (15)

2.2.2.2. Theories of Enzyme Inactivation(2)

Many workers have attempted to relate the quantum yields for the UV inactivation of
enzymes to the amino acid composition. An early theory of Luse and McLaren is based on
the hypothesis that photolysis of any chromophoric amino acid residue in an enzyme
contributes additively to UV inactivation. (16) According to this approach, the inherent
photosensitivity of an amino acid is related to the product of its extinction coefficient (Ea)
and its photo destruction quantum yield (<Pa) in aqueous solution. The results in Table 2-1
show that cystine and tryptophan should be the most photosensitive residues at 254 nm. In
the Luse and McLaren theory, the contribution of each amino acid to the photoinactivation
quantum yield (<Pin) for the enzyme is determined with the additive relationship:

(2-21)

where na is the number of type a residues in the enzyme and Ee is the molar extinction
coefficient of the enzyme. The application of Eq. (2-21) to common enzymes gave results
in agreement with experimental data to within a factor of 2 in most cases, suggesting that
the concepts of inherent residue photosensitivity and additivity are valid. (17) However, the
Photochemistry 63

TABLE 2-1. Relative Photosensitivity of Amino Acids at 254 nm a

Compound eb «Jc (e x «J)d
Cystine (-S-S- bond) 270 0.13 35
Tryptophan 2870 0.004 12
Tyrosine 320 0.002 0.6
Phenylalanine 140 0.013 2
Histidine 0.24 <0.003 <0.007
Peptide bonds (acetylalanine) 0.2 0.05 0.01
aThe values in this table are based on data in Ref. 3.
bE indicates the molar extinction coefficient at 254 nm, which is the probability that light at 254
nm will be absorbed by the compound.
c<l> indicates the quantum yield, which is the probability that light absorbed at 254 nm will cause a
chemical change.
deE x <1» indicates a measure of the photochemical sensitivity of a particular amino acid.

earlier proposal of Setlow that <Pin correlates with the cystine content in enzymes gave
equally good results. (18) This correlation implies that conformation changes induced by
the rupture of disulfide bonds are the key factor leading to the loss of enzymic activity.
Augenstein and Riley attempted to rationalize the two divergent theories by proposing that
cystinyl residues can be destroyed by both direct photolysis and by energy transfer from
adjacent aromatic residues. (19) The same mechanism was extended by Dose to explain the
higher yields of cystine destruction in enzymes containing aromatic residues, (20)
The earlier work was based on the untenable assumptions that all residues of a given
type in an enzyme contribute equally to its activity and that the quantum yield for
destruction of an amino acid residue in a protein is the same as the aqueous amino acid, In
a recent extension of the McLaren and Luse approach, Grossweiner and coworkers
replaced na in Eq. (2-21) by the number of essential type a residues.(2I) In conventional
biochemistry, a residue is considered essential if it is directly involved in the catalytic
reaction, participates in substrate binding, or helps maintain the active enzyme conforma-
tion. In the photochemical process, the concept of "essential" may be extended to other
chromophoric residues, whose photolysis leads to loss of enzymic activity. In addition,
the <Pa in Eq. (2-21) were replaced by estimates of the initial photolysis quantum yields of
amino acid residues in proteins. It was found that only cystine and tryptophan need be
considered as photosensitive, and that only one set of <P a values is required at 254 and 280
nm. The results in Table 2-2 are based on the following relationship:

(2-22)

r
where is the fraction of essential tryptophanyl or cystinyl residues, f is the fractional
absorption by all tryptophanyl or cystinyl residues at the irradiation wavelength, and <Ptrp
and <Peys are the quantum yields for destruction of tryptophan or cystine in the protein,
respectively. An average value of <Ptrp = 0.05 was used at 254 and 280 nm, based on the
quantum yields for tryptophan destruction in proteins reported in the literature(2) and the
initial yields of Trp- formation in enzymes measured by laser flash photolysis. (21) The
values <P cys were taken as 0.20 at 254 nm and 0.13 at 280 nm, based on the quantum yields
for disulfide bond destruction in oxidized glutathione. (22) The results in Table 2-2 are in
64 Chapter 2

TABLE 2-2. Prediction of Photoinactivation Quantum Yieldsa

254 run 280 run
Enzyme f trpb fey/ <P in ( calC)d <Pin(exp)e <Pin(calc)d <Pin(exp)e
Lysozyme 3/6 2/4 0.028 0.024 0.023 0.023
Trypsin 114 2/6 0.015 0.020 0.009 0.010
Papain 115 0/3 0.006 0.006 0.006 0.003
Carboxypeptidase A 116 011 0.005 0.005
Subtilisin Carlsberg 011 0 0.000 0.007
Ribonuclease 0 2/4 0.029 0.030 0.004 0.007
aThe calculations in this table are based on Eq. (2-21).
bfUp indicates tbe fraction of essential tryptopbanyl residues in the enzyme based on biochemical infonnatioD and the structure.
efey, indicates the fraction of essential cystinyl residues in the enzyme based on biochemical infonnation and the structure.
d<Pin(calc) is the calculated photoinactivation quantum yield at the indicated wavelength as described in the text.
e<P in is the experimental photoinactivation quantum yield at the indicated wavelength based on literature cited in Ref. 21.

good agreement with the reported data, except for subtilisin Carlsberg, which has no cystine
and apparently no essential tryptophan. This model implies that energy and electron transfer
between residues are not major factors in photoinactivation, and supports the concept that
photosensitive residues may either be essential in the conventional biochemical sense or be
located in the structure where their photochemical disruption can affect essential resi-
dues.(2,14) Examples of photosensitive tryptophanyl residues identified in flash photolysis
studies include Trp 199 in trypsin, adjacent to the catalytic residue serine 198; Trp 177 in
papain, in contact with the active site residue His 159; Trp 73 in carboxypeptidase A,
adjacent to glutamic acid 72, which participates in the binding of essential zinc; Trp 62, Trp
63, and Trp 108 in lysozyme, which are part of the activite site. However, ribonuclease A
contains no tryptophanyl residues, and none of the tyrosinyl residues are essential. Pho-
to inactivation of ribonuclease A can be attributed entirely to the photolysis of two essential
cystinyl residues (Cys 26-84 and Cys 40-95), which is favored by the higher absorption in
cystine resulting from the absence of tryptophan. (23) Equation (2-22) is not applicable to the
subtilisins, which contain no cystine, and none of the aromatics are located near essential
residues. A laser flash photolysis study on subtilisin BPN' led to the suggestion that Trp 113
may be photosensitive because it is hydrogen-bonded to asparagine 113, which is located at
one end of a chain sequence associated with the aromatic substrate-binding sites. (13)

2.2.2.3. Kinetics of Enzyme Inactivation

The dependence of enzyme inactivation rates on dose exemplifies a relatively simple
type of "survival curve." Photoinactivation measurements are usually carried out by
exposing a dry or aqueous enzyme to monochromatic radiation, and assaying the remain-
ing activity at periodic intervals. The actinometry is more accurate for optically dilute
systems, where the enzyme absorbs a small fraction of the incident UV radiation. The
analysis may be simplified by assuming that inactivation does not affect the overall light
absorption by the enzyme, in which case the fraction of light absorbed by the active
Photochemistry 65

component is ClCo, where C is the concentration of active enzyme after incident fluence
F, and Co is the initial active enzyme concentration. A straightforward analysis leads to an
exponential survival curve:

(2-23)

where O"e is the absorption cross-section of the enzyme (Section 1.5.5). Equation (2-23)
predicts that a semilogarithmic plot of the surviving active enzyme fraction vs. incident
fluence will be a straight line, with the slope proportional to the product of the inactivation
quantum yield and the absorption cross-section. The quantity $inO"e is the "inactivation
cross-section, " which may be viewed as the effective cross-sectional area of the pho-
tosensitive targets on the enzyme molecule. An exponential dependence of surviving
fraction on dose characterizes a "one-hit model" because it implies that the absorption of
one photon by a "target" inactivates the enzyme molecule. Survival curves for cellular
systems may follow Eq. (2-23) if one photochemical reaction in a single cellular target
leads to the observed biological effect and this damage is not repaired during irradiation or
prior to the damage assay (Chapter 4).

2.3. PHOTOCHEMISTRY OF NUCLEIC ACIDS(24-38)

Early photochemical studies on the nucleic acids dealt largely with the destructive
cleavage of the pyrimidine ring, but in 1949 Sinheismer and Hastings showed that the
pyrimidines could undergo photochemistry without destroying the pyrimidine ring. They
reported the photochemical addition of a molecule of water across the 5,6 double bond of
the pyrimidines (pyrimidine hydrate). In the 1960s, Beukers and Berends observed that
UV irradiation caused the stable linkage of two adjacent thymine residues in a strand of
DNA. This discovery of thymine dimer formation stimulated a resurgence of interest in
UV photobiology that continues today. Unfortunately, as the importance of thymine
dimers in biological inactivation and mutagenesis became apparent, there developed a
tendency to give them credit for too much of UV photobiology. Many other types of
photoproducts are produced in the DNA of cells, and under certain experimental condi-
tions these photoproducts play the dominant role in UV photobiology (Chapter 4).
It is important to remember that DNA does not exhibit the same sensitivity to UV
radiation under all experimental conditions. The intrinsic sensitivity of DNA to pho-
tochemical alteration can be changed by a variety of biological (e.g., growth state of
cells), chemical (e.g., base analog substitution), and physical (e.g., denaturation, freez-
ing, drying) techniques. To give an example: one photoproduct that is produced in high
yield, and appears to be the major cause of death in UV -irradiated vegetative bacterial
cells, is not produced to a significant extent in bacterial spores. Thus, different types of
photoproducts appear to inactivate irradiated vegetative cells and spores, respectively.
Simple generalizations, therefore, cannot be made as to which photoproduct in DNA is
the most important to all irradiated cells under all experimental situations.
Furthermore, in addition to the intrinsic photochemical sensitivity of DNA, we must
also consider the ability of a cell to repair the damage produced in DNA (Chapter 4). If
66 Chapter 2

one type of lesion is repaired accurately by a cell it cannot be of biological importance to

that cell, while lesions that are not repaired, or are repaired inaccurately, may be of major
biological importance.

2.3.1. Photochemistry of Nucleic Acid Constituents

2.3 .1.1. Deoxyribose
Although carbohydrates make up about 41 % by weight of the nucleic acids, they
show essentially no UV absorption at wavelengths above 230 nm and therefore would not
be expected to undergo direct photochemical reactions when irradiated with light of
wavelengths greater than 230 run. However, indirect effects of UV radiation on deoxy-
ribose have been reported. For example, a cell is much more sensitive to killing by UV
radiation if 5-bromouracil (BrUra) replaces the thymine in its DNA. One effect of UV
radiation on BrUra in DNA is debromination, with the consequent production of a uracil
radical. In the absence of another hydrogen donor, a hydrogen atom will be abstracted
from the adjacent deoxyribose. This leads to the production of uracil, to the destruction of
the deoxyribose, and ultimately to a chain break in the DNA. (39) Chain breaks in DNA are
also produced in the presence of ketone sensitizers such as benzophenone. When certain
of these sensitizers are excited by light at 313 nm they can abstract hydrogen from water to
produce hydroxyl radicals, which then attack the DNA and produce chain breaks.(40)
Thus, under certain conditions, chemical alterations in the carbohydrates of the nucleic
acids can be brought about when the nucleic acids are exposed to UV and near-UV
radiation,(41) even though the carbohydrates do not absorb light at these wavelengths.

2.3 .1.2. Purines(42-44)

In pure solution, purines are approximately lO-fold more resistant to photochemical
alteration than the pyrimidines. Inside cells, however, the nucleic acids are not in pure
solution and, in fact, the purines (as well as the pyrimidines) participate very readily in
photochemical reactions with other molecules (heteroaddition reactions).
In general, however, the photochemistry of the purines has not been as well studied
as the photochemistry of the pyrimidines, but currently there is a resurgence of interest in
the photochemistry of the purines. A photodimer of adenine has been identified,(45) and
excisable purine photoproducts have been detected in a defined sequence ofUV-irradiated
human DNA. (46)
In addition, some of the photon energy absorbed by the purines can be transferred to
the pyrimidines or to the sugar-phosphate backbone of DNA, and thus result in chemistry.
For example, the transfer of energy from adenine to thymine has been implicated in the
formation of a thymine radical and in the phosphorescence of thymine in UV-irradiated
poly-dAT.

2.3 .1 .3 . Pyrimidines
2.3 .1.3a. Hydration Products(34)
When a solution of uracil (or one of its derivatives) is UV-irradiated, it loses its
characteristic UV absorption peak near 260 nm, but this can be largely regenerated by
Photochemistry 67

heat, alkali, or acid treatment. This reversible reaction was shown to be due to the
photochemical hydration of the 5,6 double bond of uracil to form 6-hydroxy-5-hydro-
uracil [Eq. (2-5)] (note that 5-hydroxy-6-hydrouracil is stable to heat). Photochemical
hydrates of cytosine and thymine are also formed, although with less efficiency than for
uracil. Pyrimidine hydrates appear to be formed through an excited precursor state in the
singlet manifold, since triplet quenchers and sensitizers do not change the photochemical
yield of hydrates.
The formation of a water addition photoproduct of cytosine in irradiated denatured
DNA has been inferred from the appearance of a heat-reversible absorption peak around
240 nm. Dihydrocytosine derivatives (i.e., lacking the 5,6 double bond) exhibit a charac-
teristic absorption peak at this wavelength. Irradiated native DNA, however, shows no
such heat-reversible absorption peak. These and other data suggest that hydrates of
cytosine are probably not formed in measurable yield in UV-irradiated, double-stranded
DNA, but they are formed in single-stranded DNA. However, other photoproducts (e.g.,
pyrimidine dimers) can distort the DNA helix and produce local denaturation (i.e.,
single-stranded regions), and thus may permit the formation of hydrates in otherwise
double-stranded DNA.
During replication and/or transcription of the DNA, there may be short regions of
single-strandedness, and in these regions the formation of pyrimidine hydrates may be of
importance. The possible role of pyrimidine hydrates in causing mutations has been
demonstrated in an in vitro model system. (47) When polycytidylic acid was UV -irradiated,
its coding properties in an RNA polymerase system were altered. The irradiated polymer
lost its ability to code for the incorporation of guanylic acid unless adenylic acid was also
added to the medium. The increase in adenylic acid incorporation in the polymer as a
function of UV radiation fluence was heat-reversible under conditions known to reverse
pyrimidine hydrates, and for this reason it was suggested that the code change might be
the result of the formation of cytosine hydrates (i.e., the cytosine hydrate appears to base-
pair with adenine rather than guanine). Therefore, the formation of pyrimidine hydrates in
single-stranded regions of the DNA may playa role in the production of mutations, but
this remains to be established in vivo.

2.3.1.3b. Thymine Glycols

Thymine glycols (Fig. 2-6) represent one of the common classes of DNA damage
produced by ionizing radiation (e.g., Refs. 48 and 49), and they are also produced in low
yield by UV radiation. They are thought to arise through the action of hydroxyl radi-
calS(41.50) and are much more stable than the pyrimidine hydrates. The glycosylase
activity of purified Escherichia coli endonuclease ill shows a specificity for excising
thymine glycols from UV-irradiated DNA (Chapter 4).

~
o ells
HN OR
~ OR
o N
H

Fig. 2-6. Thymine glycol.

68 Chapter 2

HN~~Ho
o~~
H

Fig. 2-7. 5,6-Dihydrothymine.

2.3 .1.3 c. Other Modifications oj Pyrimidines

The addition of two hydrogen atoms to the 5 and 6 positions of thymine in DNA to
form 5,6-dihydrothymine (Fig. 2-7) has been demonstrated in vitro. The biological impor-
tance of this reducing reaction in vivo is not known.
When thymine is UV-irradiated in solution, the oxidation products, 5-hydroxymethyl
uracil and 5-formyl uracil, are formed (Fig. 2-8). Aldehydes react readily with amino
groups. If 5-formyl uracil is formed in vivo, it has the interesting possibility of reacting
with amino groups in proteins to produce DNA-protein cross-links (see Section 2.3.2.3).

2.3.1.3d. Cyclobutane-Type Dimers oj Thymine, Cytosine, and Uracil(35,51)

When an aqueous solution of thymine is irradiated at 254 nm, it loses its charac-
teristic absorption properties at a rate about one-tenth that of uracil [the quantum yield (eI»
for thymine destruction is 0.4 x 10- 3 ]. If a solution of thymine is frozen and then UV-
irradiated, it shows a greatly increased photochemical reactivity (eI> = 0.2), and the major
product formed is a dimer. (N.B.: To be a dimer, a compound must have exactly twice the
molecular weight of the monomer, otherwise it should be called an adduct; see below.)
The probable effect of freezing is to bring the thymine molecules into an oriented jux-
taposition favorable for a bimolecular photochemical reaction.
To form the thymine dimer, two thymine molecules are linked to each other between
their respective 5 and 6 carbon atoms, thus forming a cyclobutane ring (four-carbon ring)
between the two thymines [Eq. (2-2)]. There are six possible isomers of the thymine
dimer, and these have been isolated from irradiated thymine oligomers (Fig. 2-9). The
type I (cis-syn) thymine dimer is the one formed between adjacent thymines in the same
strand of DNA. Certain of these isomers are stable to acid hydrolysis, while others are
not. (35) Since acid hydrolysis is the usual method for liberating photoproducts from
irradiated DNA, such labile photoproducts would be destroyed.
There is a wavelength dependency both for the formation and for the monomerization
of the cyclobutane-type thymine dimer. After a sufficient fluence of UV radiation, a
"photostationary" state is achieved, and a constant ratio of monomer to dimer is estab-
lished that is characteristic of the wavelength used. At the longer wavelengths of UV
radiation (near 280 nm) the formation of the dimer is favored, while at the shorter
wavelengths (near 240 nm) monomer formation is favored. This response is due to
differences in the absorption spectra of thymine and its dimer (Fig. 2-10), and in the

o o
HNJlyCH3 HN:XCHO Fig. 2-8. Photochemical formation of 5-hy-
O~N~H
H
crlN I H
H
droxymethyl uracil and 5-formyl uracil from
thymine.
 8
Photochemistry 69

NH
CO.H

~
H' C"'H H
C. C
co NH
co NH co NH
o
NH co NH co
(meso) (d,l)
1 n
cis-syn trans-syn

co

co

(d,i) (meso)

Fig. 2-9. Isomeric forms of the cyclobutane-type thymine m Dr

dimers. Optical iosmers are possible for types II and III. cis-anti trans-anti

quantum yields for the formation and splitting of the dimer. Thus, in UV-irradiated E.
coli, where 30-40% of the thymines could theoretically dimerize, only about 15% are
dimerized even after high fluences of UV radiation at 254 nm.
Five other cycIobutane-type dimers of the natural pyrimidines are also known. These
are the dimers of uracil and cytosine, and the adducts of uracil-thymine, cytosine-thy-
mine, and uracil-cytosine. The isolation of cytosine dimers is complicated not only by the
competition of the hydrate reaction but also by the fact that cytosine deaminates readily
when its 5,6 double bond is saturated. Cytosine dimers are therefore readily converted to
uracil dimers. It is apparent that if a cytosine dimer in the DNA of a cell were to deaminate

THYMINE
1.0 ,/

0.5
>-
u
z
<I:
o
co

IfJ
o.~o

~ 0.05
THYMINE

0.01

Q005

Fig. 2-10. The absorption spectra of thymine (25 ,""g/m!) and

of purified thymine dimer (34 ,""g/ml) [Modifed from R. Set- 220 260 300
low, Biochim. Biophys. Acta 49, 237 (1961).] WAVELENGTH (nm)
70 Chapter 2

to form a uracil dimer, and if this dimer were then split in situ by the photoreactivating
enzyme (Chapter 4), a mutation could result since the uracil residues would base-pair with
adenine rather than with guanine.
Dihydrocytosine (Le., cytosine without its 5,6 double bond) will react with amino
acids such that the amino group of dihydrocytosine is replaced by the amino group of the
amino acid, resulting in a covalent link between dihydrocytosine and the amino acid. (52)
Since the photohydrate of cytosine and the cyclobutane-type dimers of cytosine are
analogs of dihydrocytosine, the addition of protein amino groups to these cytosine pho-
toproducts might serve as another mechanism by which DNA (and RNA) and protein are
cross-linked by UV radiation (Section 2.3.2.3).
Up to now we have considered photochemical reactions that occur as the result of the
direct absorption of photons by the reacting species. Thymine dimers can be formed,
however, by wavelengths of light that are not absorbed by thymine, provided that the
thymine is in the presence of suitable molecules (i.e., photosensitizers) that do absorb
these wavelengths. This process occurs by triplet-triplet energy transfer [Eq. (2-9)] and
requires that the triplet state of the absorbing species (the photosensitizer) be higher in
energy than the triplet state of thymine. Upon interaction, the triplet energy of the
photosensitizer is transferred to the thymine, yielding thymine in its triplet state with the
possibility for the subsequent formation of thymine dimers. Examples of this situation are
the formation of thymine dimers by light of wavelengths >300 nm when DNA is irradi-
ated in the presence of 10 - 2 M acetophenone or when bacteria are irradiated while
suspended in 10% acetone. One advantage of the use of triplet state photosensitization to
drive a reaction is that it can be performed at wavelengths where the reverse reaction
(e.g., dimer splitting) does not occur. With this technique one should achieve essentially a
quantitative conversion of all adjacent thymine residues to dimers rather than only achiev-
ing a photo stationary state as obtained by the direct excitation of thymine residues (see
above). Certain metal ions (e.g., Ag+) photosensitize the formation of thymine dimers in
DNA, while others (e.g., Hg2+) quench this reaction.(37)

2.3.1.3e. Spore PhotoproductC 36 ,51)

After high fluences ofUV radiation (e.g., 2 x 104 J/m2 at 254 nm) about 30% of the
thymine in bacterial spores can be converted to what has been called the spore pho-
toproduct (Fig. 2-11). It is a thymine-thymine adduct, and appears to be formed through
the reaction of two different thymine radicals. The spore photoproduct does not exhibit the

Fig. 2-11. 5-Thyminyl-5,6-dihydrothymine, a major

photoproduct in UV-irradiated bacterial spores, is pro-
duced by the addition of two different radicals of thymine.
Photochemistry 71

short-wavelength reversal properties of the cyclobutane-type thymine dimers. Therefore,

in spores the yield of this product can approach the maximum determined by the number
of thymines that are nearest neighbors in the DNA.
When spores are irradiated at various times during germination (i.e., the transition
phase between the spore stage and the vegetative cell stage), the yield of spore pho-
toproduct decreases, the yield of cyclobutane-type pyrimidine dimers increases, and the
sensitivity to killing by UV radiation increases. The cyclobutane-type pyrimidine dimers
appear to be about 11 times more effective in killing vegetative cells than are the spore
photoproducts in killing spores. This could imply that the spore photoproduct is more
efficiently repaired than the cyclobutane-type dimers.

2.3.1.3f Pyrimidine-Pyrimidine (6-4) Adducts(25,36,51)

When a solution of DNA is UV -irradiated, or when a frozen solution of thymine is UV-
irradiated and thawed, the absorption spectrum of each shows a new peak at 320 nm. This
is due to the photochemical formation of (6-4) adducts (Fig. 2-12). The (6-4) adducts of
TC, CC, and IT are observed in UV-irradiated DNA, but the CT adduct is not. These
lesions cannot be split by reirradiation at short wavelengths, as can cyclobutane-type
pyrimidine dimers. Their chief diagnostic property is their lability to hot alkali. (53) The
(6-4) adducts are produced with various efficiencies at specific sites within DNA; how-
ever, on average they are produced about lOO-fold less efficiently than are pyrimidine
dimers. Recently, the (6-4) adducts have been shown to playa major role in UV radiation
mutagenesis at specific sites in DNA (Chapter 4).

2.3 .1.3g. H eteroadducts of the Pyrimidines

The pyrimidines (and purines) will also react photochemically with many other
compounds (for tabulations of these reactions, see Refs. 27 and 54). For example,
alcohols (R-OH) and the amino acid cysteine (R'-SH) photochemically add across the
5,6 double bond of the pyrimidines. Since some of the addition products of methanol and
ethanol can be reversed by acid, they appear analogous to the water (R-OH) addition
product described above, and therefore are expected to be attached to position 6. Cysteine
is known to add at the 5 position and forms a stable product [Eq. (2-1)], which may be an
important mechanism for the formation of DNA-protein cross-links (Section 2.3.2.3).
The pyrimidines can also form photoadducts with the purines (e.g., adenine-thymine)(44)
but their biological importance has yet to be determined.

Fig. 2-12. The (6-4) adduct of thymine [6,4'-{5'-methylpyrimidin-2'-one}-tbyminel.

72 Chapter 2

2.3.2. Photochemistry of DNA

2.3.2.1. DNA Strand Breaks

While X irradiation is very effective in producing single-strand and double-strand
breaks in DNA, (48) UV irradiation is very inefficient in this respect except in the presence
of sensitizers such as benzophenone (Section 2.3.1.1). However, DNA chain breaks are
produced enzymatically in UV-irradiated cells (and X-irradiated cells) as necessary steps
in the repair of damaged DNA (Chapter 4).

2.3.2.2. DNA-DNA Cross-Links(3l)

If DNA is UV-irradiated while dry or when it is very tightly packed, as in sperm
heads, DNA-DNA cross-links leading to gel formation have been observed. However,
this type of reaction (i.e., gel formation) appears to be of little importance to normal wet
cells.
A single interchain DNA cross-link causes the two strands of a molecule of DNA to
be connected so that they can no longer be separated when the DNA is denatured with heat
or formamide. Since no interchain cross-links were detected in normal phage T7 irradiated
to a survival of 1%, the biological importance of this lesion seems in doubt at low fluences
of UV radiation. However, this lesion may achieve a position of greater biological
importance at higher fluences of UV radiation in those cells that are relatively resistant to
radiation. The formation of interchain DNA cross-links has been studied in vitro. (31)
DNA can also be cross-lined indirectly using a photosensitizer (Chapter 3). The
photosensitizer (e.g., 8-methoxypsoralen) intercalates between the stacked bases of the
DNA and, when excited by the absorption of light energy, can react with pyrimidines on
each strand of the DNA to produce an interchain cross-link. This photochemical reaction
may be the basis for the photo chemotherapy of psoriasis (Chapter 6).

2.3.2.3. DNA -Protein Cross-Links(27,54,55)

There is a progressive decrease in the amount of DNA that can be extracted free of
protein from bacteria and mammalian cells following increasing fluences of UV radiation.
The DNA that becomes nonextractable due to UV irradiation can be quantitatively ac-
counted for in the precipitate containing the denatured proteins. Treatment of this precipi-
tate with trypsin, however, yields free DNA. These results suggested that UV irradiation
had caused the DNA to become cross-linked to protein. More direct proof came from
experiments showing that DNA and protein could be cross-linked in vitro by UV irradi-
ation.
The chemical mechanisms by which DNA and protein are cross-linked in vivo are
still under study; however, the initial isolation of a mixed photoproduct of thymine and
cysteine (5-S-cysteine-6-hydrothymine) [Eq. (2-1)] from the in vitro UV irradiation of a
solution of thymine and cysteine has served as a model for the cross-linking phenomenon.
In addition to cysteine, the following amino acids add photochemically to thymine:
arginine, lysine, tyrosine, tryptophan, and cystine. The following amino acids add pho-
tochemically to uracil: glycine, serine, cysteine, cystine, methionine, lysine, arginine,
Photochemistry 73

histidine, tryptophan, phenylalanine, and tyrosine. The other common amino acids were
unreactive under the conditions tested.
Fifteen amino acids have been found to react photochemically with DNA. Cysteine,
lysine, phenylalanine, tryptophan, and tyrosine were the most reactive; alanine, aspartic
acid, glutamic acid, serine, and threonine were unreactive. (56)
The experiments that best demonstrate the biological importance of DNA-protein
cross-linking are those using bacterial cells irradiated while frozen. E. coli Blr cells
showed marked differences in survival after UV irradiation as a function of the tem-
perature at which they were irradiated (Fig. 2-13). When the temperature was reduced
from +21 to -79°C, an increase in sensitivity to UV radiation was shown by both a
change in shoulder and a change in slope in the survival curves. At -196°C the cells were
not as sensitive as at -79°C but were more sensitive than at +21°C, as evidenced by the
absence of a shoulder on the survival curve. (N.B.: a shoulder on a survival curve, or a
decreased slope on the straight line portion of the curve, may be taken as indirect evidence
that cells have the capacity to repair DNA damage; see Section 4.1.2.)
The rate of formation of cycIobutane-type thymine dimers decreased progressively
as the temperature of the cells during irradiation was reduced from + 21 to - 79°C, and
then to -196°C. Therefore, there is no correlation between the production of thymine
dimers and the increased killing of E. coli by irradiation at - 79 and - 196°C. This
suggests that cyclobutane-type thymine dimers do not playas significant a role in the
events leading to the death of irradiated frozen cells as they appear to play at room
temperature. These results provide further evidence that the relative biological importance

u
c
~

'"c>
;::
::J
en 10-2

Fig. 2-13. Survival of E. coli Bfr thy as a

function of the fluence of UV radiation (254
nm) at different temperatures. [Modified
from K. C. Smith and M. O. O'Leary, Sci-
ence 155, 1024 (I967).J
74 Chapter 2

of a given photoproduct can change markedly, depending on growth or irradiation

conditions.
In contrast, however, the amount of DNA cross-linked to protein as a function of
temperature during irradiation followed the same relative pattern as the changes in sur-
vivaL Thus, the photochemical event that appears to correlate with viability when cells are
irradiated while frozen is the cross-linking of DNA with protein. Freezing may alter the
configuration or the proximity of the protein and the DNA within the cells, so that the
probability of forming DNA-protein cross-links by irradiation is greatly enhanced, thus
leading to the greater lethality observed under these conditions.
The action spectrum for the killing of Micrococcus radiodurans, one of the most
radiation-resistant organisms known, differs markedly from that for the more sensitive
organism E. coli (Fig. 1-23) in that it shows a high sensitivity to irradiation at 280 nm as
well as at 260 nm. Clasically, a response at 280 nm has indicated an involvement of
protein, while a response at 260 nm has implicated nucleic acid (Section 1.5.5). It has
been suggested that the resistance of this organism to UV radiation is due to its extraordi-
nary ability to repair pyrimidine dimers, but what ultimately kills the organism is damage
that involves both DNA and protein. The cross-linking of DNA and protein may constitute
one type of such damage.
It is reasonable to assume that a different type of photochemistry might arise when
protein and DNA are irradiated together as compared to when they are irradiated sepa-
rately. Since DNA and protein do not exist in cells as pure solutions of the separate
molecules but are in intimate contact with each other, it might be expected that the
photochemical interaction of DNA and protein would playa significant role in the inac-
tivation of UV -irradiated cells under certain conditions.
Proteins are not the only molecules that bind covalently to DNA due to the actions of
nonionizing (i.e., UV and visible) and ionizing (i.e., X and -y) radiation, or through direct
chemical attack (for tabulations of these reactions, see Refs. 27 and 54). In fact, an
international symposium was held to demonstrate that the seemingly diverse fields of
aging, carcinogenesis, and radiation biology are actually closely related, since the forma-
tion of DNA adducts appears to be an important initial event in each of these fields,
although the types of adducts produced differ in each field. (27)

2.3.3. Photochemistry of RNA

RNA differs from DNA by virtue of having ribose instead of deoxyribose, by having
uracil instead of thymine, and by having more single-stranded regions. These differences
do affect the photochemistry of RNA (e.g., more pyrimidine hydrates are formed), but in
general it is quite similar to that of DNA.
The photochemistry of RNA in vivo has not been studied as extensively as the
photochemistry of DNA. This is perhaps understandable in view of the central importance
of DNA to the survival of cells. Since RNA species are generally present within a cell in
multiple copies, the inactivation of a few percent of these molecules would not be
expected to have as adverse an effect on the survival of a cell as the photochemical
alteration of even a small portion of its DNA.
Nevertheless, under special circumstances the photochemical alteration of messenger
RNA and transfer RNA might be expected to have consequences for a celL For example,
Photochemistry 15

in bulk tRNA in E. coli, about 70% of the molecules contain 4-thiouridine (S4U) in place
of uridine in position 8 of the tRNA sequence. As the consequence of the substitution of
the oxygen atom at position 4 of uridine by a sulfur atom, the absorption maximum is
shifted from 260 to 330 nm. Therefore, S4U absorbs strongly in the near-UV region,
whereas most other nucleic acid bases show very little absorption in this region.
The absorption of near-UV radiation by tRNA species that contain S4U initiates a
very specific photochemical reaction that results in the intramolecular cross-linking of
S4U in position 8 to the cytidine present in position 13. Such modified tRNA species show
a much reduced ability to be charged with their specific amino acid; the result is a slowing
down of protein synthesis. (57,58)
In some situations this slowing down of protein synthesis by near-UV irradiation can
be beneficial to a cell. For example, prior near-UV irradiation protects E. coli B cells from
a subsequent exposure to 254-nm radiation ("photoprotection").(57,58) Photoprotection
has similarities with liquid-holding recovery (Chapter 4), where macromolecular syn-
thesis is inhibited after UV irradiation by holding cells in buffer for awhile before plating.
This liquid holding results in an increase in survival in UV-irradiated (254-nm) E. coli
cells that are deficient in postreplication repair but are proficient in excision repair. Thus,
if most DNA lesions can be repaired by excision repair prior to the resumption of DNA
replication, then there is less dependence by the cells on postreplication repair, which is
deficient in the strains that show photoprotection and liquid-holding recovery.
Some viruses contain RNA as their genetic determinant (e.g., tobacco mosaic virus).
For such viruses the photochemical alteration of their RNA leads to rapid inactivation.
The biological and chemical effects of UV radiation (254 nm) on RNA have been most
extensively studied with RNA viruses.( 5 9 ) '

2.3.4. Detection and Quantitation of Photochemical Lesions in DNA

A laboratory manual of research procedures for the study of DNA damage and repair
is available. (60)
Chromatography of various types (e.g., paper, column, thin-layer) has been used to
separate damaged purines and pyrimidines that have been released from UV -irradiated
DNA (usually specifically labeled with radioactivity) by acid or enzymic hydrolysis.
With the new techniques of molecular biology, one can start with defined sequences
of DNA (e.g., phage, plasmids, oligonucleotides), and then sequence the irradiated DNA,
and assay for the nature of the damage at defined sites.
Specific antibodies to DNA damage (both conventional antisera and monoclonal
antibodies) have been used effectively to detect and quantitate DNA damage.
Enzyme probes that cut the DNA strands at the sites of specific lesions are also very
useful. After the DNA has been cut by the enzyme probe, the number of enzymatically
induced DNA strand breaks are quantitated.
DNA strand breaks are generally quantitated by measuring the molecular weight of
DNA by centrifugation in sucrose gradients. Denaturing (i.e., alkaline) gradients are used
for measuring DNA single-strand breaks, and nondenaturing (i.e., neutral) gradients are
used for measuring DNA double-strand breaks. For mammalian cells, the techniques of
alkaline and neutral elution of DNA from filters is used to follow the formation and repair
of DNA single-strand and double-strand breaks, respectively.
76 Chapter 2

DNA-DNA and DNA-protein cross-links can also be measured by techniques sim-

ilar to those used to assay for DNA strand breaks.

2.4. REFERENCES

1. N. J. Turro, Modern Molecular Photochemistry, Benjamin/Cummings, Menlo Park, Cal. (1978).

2. L. I. Grossweiner, Photochemical inactivation of enzymes, Curro Topics Radiat. Res. Q. 11, 141-199
(1976).
3. A. D. McLaren and D. Shugar, Photochemistry of Proteins and Nucleic Acids. Macmillan, New York
(1964).
4. D. Creed, The photophysics and photochemistry of the near-uv absorbing amino acids. I. Tryptophan and
its simple derivatives, Photochem. Photobiol. 39, 537-562 (1984).
5. D. Creed, The photophysics and photochemistry of the near-uv absorbing amino acids. II. Tyrosine and its
simple derivatives, Photochem. Photobiol. 39, 563-575 (1984).
6. D. Creed. The photophysics and photochemistry of the near-uv absorbing amino acids. m. Cystine and its
simple derivatives, Photochem. Photobiol. 39, 577-583 (1984).
7. L. I. Grossweiner, G. W. Swenson, and E. F. Zwicker, Photochemical production of the hydrated electron,
Science 141, 805-806 (1963).
8. E. 1. Hart and 1. W. Boag, Absorption spectrum of the hydrated electron in water and aqueous solutions, J.
Am. Chem. Soc. 84, 4090-4095 (1962).
9. L. I. Grossweiner, Photochemistry of proteins: A review, Curro Eye Res. 3, 137-144 (1984).
10. 1. Dillon, The anerobic photolysis of tryptophan containing peptides, Photochem. Photobiol. 32, 37-39
(1980).
11. 1. Dillon, The anaerobic photolysis of tryptophan containing peptides. II, Photochem. P hotobiol. 33, 137-
142 (1981).
12. 1. Butler, E. 1. Lan, W. A. Prutz, and A. 1. Swallow, Charge transfer between tryptophan and tyrosine in
proteins, Biochim. Biophys. Acta 705, 150-162 (1982).
13. A. Blum and L. I. Grossweiner, Laser flash photolysis and photoinactivation of subtilisin BPN', Pho-
tochem. Photobiol. 36, 617-622 (1982).
14. L. L Grossweiner, A. G. Ka1uskar, and 1. F. Baugher, Flash photolysis of enzymes, Int. J. Radiat. Bioi.
29, 1-16 (1976).
15. P. Walrant and R. Santus, Ultraviolet and N-formyl-kynurenine-sensitized photoinactivation of bovine
carbonic anhydrase: An internal photodynamic effect, Photochem. Photobiol. 20, 455-460 (1974).
16. R. A. Luse and A. D. McLaren, Mechanism of enzyme inactivation by ultraviolet light and the photochemis-
try of amino acids (at 2537 A), Photochem. Photobiol. 2, 343-360 (1963).
17. A. D. Mclaren and O. Hidalgo-Salvatierra, Quantum yields for enzyme inactivation and the amino acid
composition of proteins, Photochem. Photobiol. 3, 349-352 (1964).
18. R. Setlow, A relation between cystine content and ultraviolet sensitivity of proteins, Biochim. Biophys.
Acta 16, 444-445 (1955).
19. L. Augenstein and P. Riley, The inactivation of enzymes by ultraviolet light. IV. The nature and involve-
ment of cystine disruption, Photochem. Photobiol. 3, 353-367 (1964).
20. K. Dose, Theoretical aspects of the U.V. inactivation of proteins containing disulfide bonds, Photochem.
Photobiol. 6, 437-443 (1967).
21. L. L Grossweiner, A. Blum, and A. M. Brendzel, Primary processes in the photochemistry of proteins, in:
Trends in Photobiology (C. Helene, M. Charlier, and Th. Montenay-Garestier, eds.), pp. 67-80, Plenum
Press, New York (1982).
22. T. K. Rathinasamy and L. G. Augenstein, Photochemical yields in ribonuclease and oxidized glutathione
irradiated at different wavelengths in the ultraviolet, Biophys. J. 8, 1275-1287 (1968).
23. 1. F. Baugher and L. I. Grossweiner, Photolysis mechanism of aqueous tyrosine and tyrosyl peptides,
Photochem. Photobiol. 28, 175-184 (1978).
24. S. Y. Wang (ed.), Photochemistry and Photobiology of Nucleic Acids, Vols. I and 2, Academic Press, New
York (1976).
Photochemistry 77

25. A. J. Varghese, Photochemistry of nucleic acids and their constituents, Photophysiology 7, 207-274
(1972).
26. R. O. Rahn, Ultraviolet irradiation of DNA, in: Concepts in Radiation Cell Biology (G. L. Whitson, ed.),
pp. 1-56, Academic Press, New York (1972).
27. K. C. Smith (ed.), Aging, Carcinogenesis and Radiation Biology: The Role of Nucleic Acid Addition
Reactions, Plenum Press, New York (1976).
28. A. D. McLaren and D. Shugar, Photochemistry of Proteins and Nucleic Acids, Pergamon Press, Oxford
(1964).
29. J. G. Burr, Advances in the photochemistry of nucleic acid derivatives, Adv. Photochem. 6, 193-299
(1968).
30. K. C. Smith, Photochemistry of the nucleic acids, Photophysiology 2, 329-388 (1964).
31. K. C. Smith, Physical and chemical changes induced in nucleic acids by ultraviolet light, Radiat. Res.
Supp. 6,54-79 (1966).
32. L. Kittler and G. LOber, Photochemistry of the nucleic acids, in: Photochemical and Photobiological
Reviews, Vol. 2 (K. C. Smith, ed.), pp. 39-131, Plenum Press, New York (1977).
33. A. Wacker, H. Dellweg, L. Trager, A. Kornhauser, E. Lodemann, G. Tiirck, R. Selzer, P. Chandra and M.
Ishirnoto, Organic photochemistry of nucleic acids, Photochem. Photobiol. 3, 369-394 (1964).
34. G. J. Fisher and H. E. Johns, Pyrimidine photohydrates, in: Photochemistry and Photobiology of Nucleic
Acids, Vol. 1 (S. Y. Wang, ed.), pp. 169-224, Academic Press, New York (1976).
35. G. J. Fisher and H. E. Johns, Pyrimidine photodimers, in: Photochemistry and Photobiology of Nucleic
Acids, Vol. 1 (S. Y. Wang, ed.), pp. 225-294, Academic Press, New York (1976).
36. M. H. Patrick and R. O. Rahn, Photochemistry of DNA and polynucleotides: Photoproducts, in: Pho-
tochemistry and Photobiology of Nucleic Acids, Vol. 2 (S. Y. Wang, ed.), pp. 35-95, Academic Press,
New York (1976).
37. R. O. Rahn and M. H. Patrick, Photochemistry of DNA; Secondary structure, photosensitization, base
substitution and exogenous molecules, in: Photochemistry and Photobiology ofNucleic Acids, Vol. 2 (S. Y.
Wang, ed.), pp. 97-145, Academic Press, New York (1976).
38. E. C. Friedberg, DNA Repair, W. H. Freeman, New York (1985).
39. F. Hutchinson, The lesions produced by ultraviolet light in DNA containing 5-bromouracil. Q. Rev.
Biophys. 6, 201-246 (1973).
40. R. O. Rahn, L. C. Landry, and W. L. Carrier, Formation of chain breaks and thymine dimers in DNA upon
photosensitization at 313 nm with acetophenone, acetone or benzophenone, Photochem. Photobiol. 19,75-
78 (1974).
41. R. Tyrrell, Damage and repair from non-ionizing radiations, in: Repairable Lesions in Microorganisms (A.
Hurst and A. Nasim, eds.), pp. 86-124, Academic Press, New York (1984).
42. D. Elad, Photoproducts of purines, in: Photochemistry and Photobiology of Nucleic Acids, Vol. 1 (S. Y.
Wang, ed.), pp. 357-380, Academic Press, New York (1976).
43. D. Elad, Photochemically-induced adducts of DNA, in: Aging, Carcinogenesis and Radiation Biology (K.
C. Smith, ed.), pp. 243-260, Plenum Press, New York (1976).
44. J. Cadet, L. Voituriez, A. Grand, F. E. Hruska, P. Vigny and L.-S. Kan, Recent aspects of the pho-
tochemistry of nucleic acids and related model compounds, Biochimie 67, 277-292 (1985).
45. F. P. Gasparro and J. R. Fresco, Ultraviolet-induced 8,8-adenine dehydrodimers in oligo- and poly-
nucleotides, Nucleic Acids Res. 14,4239-4251 (1986).
46. P. E. Gallagher and N. J. Duker, Detection of UV purine photoproducts in a defmed sequence of human
DNA, Molec. Cell. BioI. 6,707-709 (1986).
47. G. R. Banks, D. M. Brown, D. G. Streeter, and L. Grossman, Mutagenic analogues of cytosine: RNA
polymerase template and substrate studies, J. Mol. Bioi. 60,425-439 (1971).
48. F. Hutchinson, Chemical changes induced in DNA by ionizing radiation, Prog. Nucleic Acid Res. Mol.
Bioi. 32, 115-154 (1985).
49. L. H. Breimer and T. Lindahl, Thymine lesions produced by ionizing radiation in double-stranded DNA,
Biochemistry 24, 4018-4022 (1985).
50. M. G. Sirnic and S. V. Jovanovic, Free radical mechanisms of DNA base damage, in: Mechanisms of DNA
Damage and Repair (M. G. Simic, L. Grossman, and A. C. Upton, eds.), pp. 39-49, Plenum Press, New
York (1986).
78 Chapter 2

51. S. Y. Wang, Pyrimidine bimolecular photoproducts, in: Photochemistry and Photobiology of Nucleic
Acids, Vol. 1 (S. Y. Wang, ed.), pp. 295-356, Academic Press, New York (1976).
52. C. Janion and D. Shugar, Reaction of amines with dihydrocytosine analogues and formation of amino acid
and peptidyl derivatives of dihydropyrimidines, Acta Biochim. Polon. 14, 293-302 (1967).
53. W. A. Franklin, K. M. La, and W. A. Haseltine, Alkaline lability of fluorescent photoproducts produced in
ultraviolet light-irradiated DNA, J. Bioi. Chem. 257, 13535-13543 (1982).
54. K. C. Smith, the radiation-induced addition of proteins and other molecules to nucleic acids, in: Pho-
tochemistry and Photobiology of Nucleic Acids, Vol. 2 (S. Y. Wang, ed.), pp. 187-218, Academic Press,
New York (1976).
55. M. D. Shetlar, Cross-linking of proteins to nucleic acids by ultraviolet light, in: Photochemical and
Photobiological Reviews, Vol. 5 CK. C. Smith, ed.), pp. 105-197, Plenum Press, New York (1980).
56. M. D. Shetlar, J. Christensen, and K. Horn, Photochemical addition of amino acids and peptides to DNA,
Photochem. Photobiol. 39, 125-133 (1985).
57. A. Favre, E. Hajnsdorf, K. Thiam, and A. Caldeira de Araujo, Mutagenesis and growth delay induced in
Escherichia coli by near-ultraviolet radiations, Biochimie 67, 335-342 (1985).
58. J. Jagger, Solar-UV Actions on Living Celis, Praeger, New York (1985).
59. M. P. Gordon, C. Huang, and J. Hurter, Photochemistry and photobiology of ribonucleic acids, ribonucleo-
proteins, and RNA viruses, in: Photochemistry and Photobiology of Nucleic Acids, Vol. 2 (S. Y. Wang,
ed.), pp. 266-308, Academic Press, New York (1976).
60. E. C. Friedberg and P. C. Hanawalt (eds.), DNA Repair: A Laboratory Manual of Research Procedures,
Marcel Dekker, New York, Vol. 1, A and B (1981), Vol. 2 (1983), Vol. 3 (1988).
 3
Photosensitization

3.1. Introduction ..................................................................... 79

3.2. Photophysics and Photochemistry of Photosensitized Reactions. . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
3.2.1. Excited States of Sensitizer Molecules ................................. , .... . . .. 81
3.2.2. Photosensitization by Type I (Radical) Mechanisms ............................... 82
3.2.3. Photosensitization by the Type II (Singlet Oxygen) Mechanism ..................... 83
3.2.4. Anaerobic Photosensitized Reactions ........................................... 84
3.2.5. Diagnostic Techniques for Determining the Mechanisms of Photosensitized Reactions ... 85
3.2.6. Chemistry of Photosensitizing Compounds......... ..... ... . ..... ... ..... .... .... 86
3.3. Sensitized Photoeffects on Biomolecules in Vitro. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 88
3.3.1. Lipids .................................................................... 88
3.3.2. Purines, Pyrimidines, and Nucleic Acids. . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . 89
3.3.3. Amino Acids and Proteins. . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . 90
3.3.4. Other Biomolecules ......................................................... 91
3.4. Sensitized Photoeffects on Viruses ................................................... 92
3.5. Sensitized Photoeffects on Cells ..................................................... 93
3.5.1. Model Cell Systems ................................... , . . . . . . . . . . . . . . . . . . . . . 94
3.5.2. Microorganisms ............................................................ 94
3.5.3. Cells from Multicellular Organisms ............................................ 95
3.6. Sensitized Photoeffects on Multicellular Organisms ..................................... 97
3.6.1. Plants .................................................................... 97
3.6.2. Nonmammalian Animals ..................................................... 98
3.6.3. Mammals ................................................................. 98
3.6.4. Humans .................................................................. , 100
3.6.5. Sensitized Photocarcinogenesis ................................................ 100
3.7. Natural Photosensitivity ... , ........................................................ 101
3.8. Biological Applications of Photosensitized Reactions .................................... 101
3.8.1. The Health Sciences. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 102
3.8.2. The Agricultural Sciences .................................................... 103
3.9. Role of Phototoxic Phytochemicals in the Ecology and Coevolution of Plants and Herbivores. .. 104
3.10. Conclusions ..................................................................... 105
3.11. References ...................................................................... 105

3.1. INTRODUCTION

Several types of photochemical processes occur in living organisms that are important or
essential for their survival. In all of these reactions, the initial step is the absorption of
photons by specialized pigments, such as chlorophyll, in the photosynthetic processes of

John D. Spikes • Department of Biology, University of Utah, Salt Lake City, Utah 84112.

79
80 Chapter 3

plants, and carotenoid-protein complexes in animal vision. In addition to these normal

photobiological processes, cells and organisms can be injured and killed by light. This
occurs only if the radiation is absorbed (First Law of Photochemistry). UV-C radiation
(defined in Section 1.3) and, to a lesser extent, UV-B radiation are very injurious since
critical cellular components, including proteins and nucleic acids, absorb radiation very
strongly in these ranges (see Chapter 2). UV-A and visible radiation typically do not cause
direct damage to biological systems, since most types of biologically important molecules
absorb only weakly at these wavelengths. However, in the presence of certain kinds of
light-absorbing molecules termed photosensitizers, cells and organisms can be rapidly
injured or killed by UV-A or visible radiation. This results from the sensitized pho-
to alteration of critical biomolecules. Current research on photosensitized phenomena in
biology includes studies of the photophysics and photochemistry of photosensitizer mole-
cules, the mechanisms of the sensitized photodegradation of essential biomolecules, the
biochemistry and physiology of photosensitized viral inactivation and cell killing, the role
of endogenous sensitizers in the protection of plants against disease organisms and her-
bivores, the involvement of photosensitizers in plant-herbivore coevolution, and the ap-
plication of sensitized reactions in practical ways such as the photosensitized treatment of
diseases in humans. (1-16)
Although photosensitized effects on organisms were observed much earlier, it is
generally agreed that this area of photobiology started with research performed at the end
of the last century by Oscar Raab, a medical student in Munich. Raab found that very low
concentrations of certain colored compounds (usually termed dyes) such as acridines and
eosin, which had no effect in the dark, led to the rapid killing of the protozoan, para-
mecium, in the light; only those wavelengths of light absorbed by the sensitizer were
effective. (1) This discovery stimulated a large amount of research on the phenomenon that
was very early termed "photodynamic action" to distinguish it from simpler physical
phenomena such as the photosensitization of photographic plates by dyes. Within a few
years, it was shown that enzymes could be inactivated and that various types of cells as
well as small animals could be killed by illumination in the presence of a number of
different photosensitizersY) As described in the next section, a molecule must be able to
do more than just absorb light to be a photosensitizer; it must, in general, be capable of
being excited by light into a relatively long-lived, energy-rich form termed the triplet state
(Section 1.4). It is estimated that thousands of different natural and synthetic molecules
can function as photosensitizers. With most photosensitizers, biomolecules are degraded
in processes that involve the uptake of oxygen; these reactions do not take place under
anaerobic conditions. Many investigators confine the use of the term photodynamic action
to those reactions that require oxygen.(l) There are, however, several well-studied pho-
tosensitized reactions that do not require oxygen, as will be described later. (7 .17)
Research on photosensitized reactions in biology has increased sharply in recent
years, in part because of the rapidly· expanding use of a large variety of chemicals in
agriculture, industry, medicine, and the home that sensitize humans to injury by light.
Also, there is considerable interest in photochemotherapy, the use of photosensitizing
drugs plus light to treat disease, as in the utilization of psoralens plus UV -A radiation for
the treatment of psoriasis (PUVA), and in the use of porphyrins in the photodynamic
therapy of tumors (PDT) (see Chapter 6).
Photosensitization 81

3.2. PHOTOPHYSICS AND PHOTOCHEMISTRY OF PHOTOSENSITIZED

REACTIONS

In photosensitized reactions, photons are absorbed by one type of molecule (the

sensitizer); the resulting energy-rich state(s) of the sensitizer then undergoes reactions that
ultimately result in the chemical alteration of another molecule in the system (the sub-
strate). Photosensitized reactions are typically mediated by excited electronic states of the
sensitizer. In its primary reaction, the excited sensitizer molecule can react directly with
the substrate, or, alternatively, with some other molecule (frequently oxygen) in the
reaction mixture, giving products that in tum can react with the substrate. Thus, as with
photochemical reactions in general, photosensitized processes typically have an initial
"light step" followed by one or more "dark steps" (see Chapter 2). Many photosen-
sitized reactions are complex, and may involve several competing reaction pathways.

3.2.1. Excited States of Sensitizer Molecules(2-4,16)

Organic photosensitizer molecules in the dark (ground state = OSENS) are almost
always in the singlet state; in this state the molecule has no unpaired electron spins. On the
absorption of a photon, an electron in the sensitizer is promoted to a higher molecular
orbital without a change in its spin; thus, the first excited state, like the ground state, is a
singlet, ISENS. Very few photosensitized reactions are directly mediated by this state
because of its short lifetime (~1-100 ns). Excited singlet sensitizer states can decay
directly to the ground state, emitting heat or light. However, the singlet excited state of an
effective photosensitizer undergoes a fast spin inversion to give a metastable triplet state,
3SENS, which has two unpaired electrons (see Chapter 1 and Fig. 1-16). The triplet states
of sensitizer molecules typically have much longer lifetimes than singlet states (microse-
conds to seconds). Thus, triplet state sensitizers will undergo large numbers of collisions
with other molecules during their lifetimes and, as a result, can mediate photosensitized
reactions with high efficiencies. These various processes are shown schematically below.
The quantum yield of triplet formation for good photosensitizers, such as metal-free
porphyrins, can be as great as 0.9.

LIGHT
oSENS ISENS (3-1)
ISENS oSENS (3-2)
ISENS 3SENS (3-3)
3SENS photosensitized reactions (3-4)

In most photosensitized reactions, the triplet sensitizer ultimately returns to the ground
state and can absorb another photon, thus acting in a somewhat catalytic fashion. In a few
cases, the sensitizer is consumed stoichiometrically in the reaction. The triplet sensitizer
can undergo its primary reaction with molecules in its vicinity by an electron (or hydrogen
atom) transfer process (termed a type I reaction), or by an energy transfer reaction with
82 Chapter 3

oxygen (tenned a type II reaction). In a given system, both types of processes can take
place simultaneously in a competitive fashion. These processes will be described in the
next two sections. Triplet sensitizers can also decay directly to the ground state, or be
converted to the ground state by collisional interactions with certain types of molecules
(physical quenchers, Q).

3SENS -7 °SENS (3-5)

3SENS + Q -7 0SENS + Q (3-6)

3.2.2. Photosensitization by Type I (Radical) Mechanisms

In general, a molecule in its triplet state can both abstract electrons (or H atoms) from
and donate electrons (or H atoms) to other molecules more readily than when it is in the
ground state. The efficiencies of these reactions depend on the chemical structures of the
sensitizer and the substrate, as well as on the reaction conditions. The abstraction of an
electron from a reducing substrate, SUB (the most common type of substrate in biological
systems), by a sensitizer triplet gives semireduced sensitizer, SENS" or SENSH', and a
semioxidized fonn of the substrate, SUB·+ or SUB', as follows:

3SENS + SUB -7 SENS" + SUB.+ (3-7a)

3SENS + SUB -7 SENSH· + SUB· (3-7b)

These free radical products (which have odd numbers of electrons) are very reactive
chemically. In many cases, the resultant substrate radicals react with oxygen to give
oxidized products of various types; often these are peroxides, which can react further to
initiate free radical chain-type autooxidation processes. In many systems, a ground state
sensitizer can be regenerated by the reaction of the semireduced fonn with ground state
oxygen, 3°2 , to give a ground state sensitizer molecule and the superoxide radical anion,
Or (or its conjugate acid, HOi>,

SENS" + 30 2 -7 °SENS + Or (3-8a)

SENSH' + 30 2 -7 °SENS + HOi (3-8b)

which in tum can react with some kinds of biomolecules. (18,19) Some semireduced sen-
sitizers, such as flavins, undergo dismutation reactions producing one ground state and
one fully reduced sensitizer molecule. This latter species may react with oxygen to give
ground state sensitizer and hydrogen peroxide.

SENSH· + SENSH· -7 °SENS + SENSH2 (3-9)

SENSH2 + 302 -7 °SENS + H 20 2 (3-10)

Superoxide can also be produced with low efficiency by the transfer of an electron from
triplet sensitizer to oxygen(l8):

3SENS + 302 -7 SENS'+ + Or (3-11)

Photosensitization 83

The resulting semioxidized sensitizer, SENS;+- , can react with a reducing substrate to give
ground state sensitizer and semi oxidized substrate,

SENS'+ + SUB _ °SENS + SUB;+- (3-12)

which can react further as described above. Finally, in some cases, sensitizer and sub-
strate radicals can interact to produce covalent sensitizer-substrate and substrate-substrate
photoadducts:

SENS' + SUB;+- - SENS-SUB (3-13)

SUB· + SUB· - SUB-SUB (3-14)

Type I photosensitized processes are most efficient at high substrate and low oxygen
concentrations, since oxygen effectively competes with substrate for interaction with
triplet sensitizer. The formation of noncovalent sensitizer-substrate complexes before
illumination would be expected to increase the probability of type I reactions because of
the proximity of the sensitizer to the substrate molecules. Substrates most effectively
photodegraded in type I reactions include those that are readily oxidized (such as amines
and phenols) or reduced (such as quinones). An example of a type I reaction is the flavin-
sensitized photodegradation of the amino acid methionine. On illumination, triplet flavin
is generated, which abstracts one electron from the sulfur atom of the amino acid; the
resulting methionine radical undergoes deamination and decarboxylation to give meth-
ional (l3-methylmercaptopropionaldehyde). The flavin radical then undergoes reactions,
as in Eqs. (3-9) and (3-10), resulting in the formation of ground state flavin and hydrogen
peroxide.

3.2.3. Photosensitization by the Type II (Singlet Oxygen)

Mechanism(2-4,20-22)

In the type II photosensitized reaction, triplet sensitizer interacts by energy transfer

with ground state oxygen to give ground state sensitizer and an electronically excited
singlet state of oxygen,10 2 , as follows:

(3-15)

Since both the sensitizer and ground state oxygen are triplets (ground state oxygen is one
of only a very few stable triplet molecules), their interaction does not require a change in
spin direction and is very efficient. Oxygen can exist in two singlet excited states; the
longer lived form, lLl g , with an excess energy of 23 kcallmole (corresponding to a photon
energy of approximately 1 eV = 1270 nm radiation) is the principal species involved in
photodynamic reactions. It has a lifetime of approximately 4 f-LS in ordinary water and thus
rapidly decays to the ground state if it does not react chemically with another molecule. Its
lifetime is much longer in lipids and other nonpolar media. (20)
Ground state oxygen (a triplet) does not react very rapidly with most types of
biomolecules under ordinary conditions, since organic molecules are typically in the
84 Chapter 3

a H H H
R-C=C-C-R'
H H H
R-C-C=C-R'
H I
a
I
OH

Fig. 3-1. Some of the major pathways for

the interaction of singlet oxygen with bio-
molecules. (a) Shows the addition of singlet
oxygen to an olefin with allylic hydrogen
atoms (such as an unsaturated fatty acid) by
c an "ene" -type reaction to give the corre-
sponding allylic hydroperoxide. (b) Shows a
similar reaction in which singlet oxygen adds
to cholesterol to give the 5-a-hydroperoxide,
a specific singlet oxygen product. (c) Shows
the addition of singlet oxygen to the diene
system of a heterocyclic compound (an im-
d coo- idazole used as a model for histidine) to give
I
+H3N - C - H an endoperoxide. (d) Shows the addition of
I
CH 2 singlet oxygen to an organic sulfide, the ami-
+ Met-2 ~H2 no acid methionine (Met), to give an unsta-
I ble product which in tum reacts with another
+5-0-
molecule of methionine to give two mole-
I
CH3 cules of methionine sulfoxide.

singlet state and thus the interaction with oxygen is spin-forbidden. The reaction of singlet
oxygen with organic compounds, however, is not spin-forbidden. Furthermore, singlet
oxygen is more electrophilic than ground state oxygen and therefore can react rapidly,
albeit somewhat selectively, with electron-rich regions of many biomolecules to give
oxidized species. Some of the major reaction pathways are (1) the addition of singlet
oxygen to olefins with allylic hydrogen atoms (such as unsaturated fatty acids and choles-
terol) to give hydroperoxides, (2) the addition of singlet oxygen to diene systems in
heterocyclics (such as histidine) to give endoperoxides, and (3) the reaction of singlet
oxygen with organic sulfides (such as methionine) to form the corresponding sulfoxides
(see Fig. 3-1). Although the overall reaction of singlet oxygen with biomolecules some-
times appears to be simple, the process often involves several intermediate steps.

3.2.4. Anaerobic Photosensitized Reactions(7,23-27)

As mentioned earlier, a few types of light-absorbing molecules sensitize the pho-
toalteration of biomolecules in the absence of oxygen; in some cases the reaction is even
inhibited by oxygen. For example, some psoralens (furanocoumarins) intercalate between
the base pairs of DNA and RNA. On subsequent irradiation under anaerobic conditions
with wavelengths absorbed by the psoralen (usually UV-A), covalent photoadducts are
formed between the psoralen and pyrimidine residues in the nucleic acid. These can be
monoadducts (psoralen-base) or diadducts (base-psoralen-base) or both, depending on the
Photosensitization 85

structure of the psoralen and the illumination regime. These reactions have been studied in
detail because of the extensive use of psoralens as sensitizers in the photochemotherapy of
psoriasis, a disease of the skin in humans (see Chapter 6). Photochemical studies with
psoralens are complicated somewhat by the fact that the illumination of these sensitizers in
solution under aerobic conditions can result in the generation of singlet oxygen and the
resulting oxidation of proteins, unsaturated lipids, etc. Psoralens are not the only type of
anaerobic photosensitizer. For example, red blood cells are hemolyzed on illumination
under anaerobic conditions in the presence of chlorpromazine (a drug used in the treat-
ment of patients with certain mental disorders). The red blood cells are apparently lysed
chemically by stable, detergentlike photoproducts resulting from the illumination of
chlorpromazine.

3.2.5. Diagnostic Techniques for Determining the Mechanisms of

Photosensitized Reactions(5 ,21 ,28-31)
As described above, the sensitized photo alteration of biomolecules may involve one
or more of a large number of reactive species such as photoexcited states of molecules,
free radicals, and reactive forms of oxygen including singlet oxygen, superoxide, hydro-
gen peroxide, and possibly the hydroxyl radical. Even in the simplest case of homoge-
neous solutions, the unambiguous establishment of reaction mechanism(s) is not easy, and
in the complex environment of a living cell it becomes very difficult.
Most of the diagnostic approaches that have been used are designed to determine
whether singlet oxygen is produced in the reaction and whether this species is responsible
for the subsequent damage to the biological system. One technique is to examine the
effects of "inhibitors" (that efficiently react with or physically quench singlet oxygen) on
the rate of the photosensitized process under study. However, the use of inhibitors alone
rarely gives unambiguous results. For example, if a photosensitized reaction is inhibited
by low concentrations of azide (a very efficient quencher of singlet oxygen), this is often
interpreted as indicating that a singlet oxygen-mediated process is involved. However, for
such measurements to be useful, it must be established (as by flash techniques; see
Chapter 1 and Fig. 1-7) that the inhibitor does not quench the triplet state of the sensitizer
directly, since this would also decrease the reaction rate. Another kinetic technique is to
compare the rate of the sensitized reaction in ordinary water (H2 0) and in heavy water
(D 2 0). Since the decay lifetime of singlet oxygen is over 10 times greater in D2 0 than in
H2 0, a significant increase of the reaction rate in D 2 0 suggests the involvement of singlet
oxygen. Again, this technique must be used with care since it is valid only under reaction
conditions where singlet oxygen deactivation by the solvent is the main rate-limiting
factor. The photooxidation of "trap" compounds, such as furans and histidine, that are
efficiently oxidized by singlet oxygen is sometimes taken as evidence for the production
of this species in a reaction. However, many such compounds can also be oxidized to the
same products by free radical processes. Cholesterol is one of the few unambiguous traps,
since singlet oxygen reacts with it to form a single product, the 5-a-hydroperoxide [see
Chapter 2, Eq. (2-4»), whereas interaction with radicals gives a mixture of other products.
A rather new technique for the detection of singlet oxygen involves the measurement of
light emission at 1270 om, since one path of singlet oxygen decay is as follows:

(3-16)
86 Chapter 3

This luminescence emission can be measured with present instrumentation, which permits
the unambiguous determination of the relative yields of singlet oxygen by illuminated
photosensitizers; the technique also permits measuring the rates of reaction of substrates
and inhibitors with singlet oxygen. Finally, the technique of flash photolysis can be used
to determine the yields and absorption spectra of triplet states of photo sensitizers as well
as the reaction rates of the triplets with oxygen, inhibitors, quenchers, and substrates; this
technique has been of very great value in determining the mechanism(s) involved in many
photosensitized processes.

3.2.6. Chemistry of Photosensitizing Compounds

Hundreds of different organic, metalloorganic, and inorganic compounds have been
shown to be photosensitizers for biological systems, and there must be many, many more. (5-
7,15,32,33) A large number of sensitizers occur naturally in organisms, especially in plants.
These include certain alkaloids (berberine, harmaline, etc.), benzofurans, chlorophylls, flavins
(riboflavin and FMN, but not FAD), polyacetylenes, a number of polycyclic polyhydroxy-
quinones such as hypericin, protoporphyrin, psoralens (furanocoumarins), pyridoxal, quinones,
thiophenes, etc. (33) An enormous number of synthetic compounds are photosensitizers , includ-
ing acridines (acridine orange, proflavine, etc.), anthraquinones, azine dyes such as the
safranines, haloaromatics, ketones, phenothiazines (methylene blue, chlorpromazine, etc.),
phthalocyanines, many kinds of porphyrins, xanthenes (including eosin, rose bengal, etc.), and
so on. (5-7) Inorganic photosensitizers include cadmium sulfide, cupric ions, ferric ions,
ruthenium trisbipyridyl, titanium dioxide, and uranium compounds. (5-7) Many categories of
light-absorbing compounds, however, are not effective photosensitizers. These include antho-
cyanins, azo dyes, indophenols, iron-containing porphyrins, nitro dyes, nitroso dyes, oxazine
dyes, and triarylmethane dyes; presumably these compounds are not useful photosensitizers
because of their low triplet yields or short triplet lifetimes. In some applications, as in the textile
industry, only nonphotosensitizing dyes are used. (34) The structures of several photosensitizers
commonly used in photobiological studies are shown in Fig. 3-2. All of these compounds
sensitize via the type n mechanism. In addition, some of the compounds sensitize via type I
mechanisms, depending on the reaction conditions. For example, with alcohol as a solvent,
flavins sensitize the photooxidation of methionine to methionine sulfoxide by a singlet oxygen-
mediated process; in contrast, a free radical process predominates in aqueous solvents leading to
methional, as described above. The structures of several photosensitizing and nonphotosensitiz-
ing psoralens are given in Fig. 3-3.
In summary, a good photosensitizer must absorb photons efficiently (i.e., it must
have a high molar extinction coefficient), it must have a high quantum yield of triplet
formation, and the triplet must be long-lived. Efficient type I sensitizers must be able to
donate and/ or accept electrons easily, while good type II sensitizers must generate singlet
oxygen with high yields. The system in which a sensitizer is used can have a large effect
on its efficiency. For example, basic dyes are generally much better photosensitizers for
nucleic acids than are acid dyes because they associate more closely with the nucleic acid
by virtue of their charge. Some sensitizers that are very effective for substrate molecules
in solution are ineffective with cells because they do not penetrate into the cell. Appropri-
ate dark control experiments must always be done in photosensitization studies, since
certain sensitizers may react directly with substrates and cells in the dark.
Photosensitization 87

v M
OH 0 OH
M

OH

M OH

P ,
Phytol" COOM 0 P P
CHLOROPHYLL a HEMATOPORPHYRIN HYPERICIN

9 HzOH
(yHOH)3

H
3
coc~~HZ
v I "'( N::,......o

H3C:::"" NQ r NH
o
RIBOFLAVIN ACRIDINE ORANGE

Br Br
NoO NoO o

Br

CI

EOSIN Y ROSE BENGAL

(CH3) Z N ( X 1 S~N(CH3)Z
"'" C~
H
THIOPYRONIN METHYLENE BLUE

Fig. 3-2. Structures of some typical photosensitizers. E, M, P, and V in the structural formulas of chlorophyll a
and hematoporphyrin stand for ethyl, methyl, propionic acid, and vinyl groups, respectively. Hypericin is a
polycyclic polyhydroxyquinone. A number of effective sensitizers are based on fluorescein, including eosin Y
(tetrabromofluorescein) and rose bengal (tetrachlorotetraiodofluorescein). Although the three compounds shown
in the bottom row have rather similar structures, they differ in photosensitizing capability with thiopyronine
being most effective, methylene blue less so, and pyronine Y the least effective.

A few types of cells and organisms are inherently photosensitive, since they contain
naturally occurring sensitizers, as will be described later. In general, however, evolution-
ary pressures have ensured that most living material is well designed to minimize both
direct and sensitized photodamage. For example, chlorophyll is an efficient photosen-
sitizer, yet photosynthetic organisms are ordinarily highly resistant to chlorophyll-sen-
sitized photodamage. In part, this results from the structural organization of photo-
synthetic organelles that assures the very rapid transfer of chlorophyll singlet excitation
energy to regions where the electron transfer reactions of photosynthesis start. Thus, little
triplet chlorophyll is generated. In addition, the organelles contain high concentrations of
88 Chapter 3

I~\
rf°if'BY°'fo o,~'-../o,...,o

~
PSORALEN
I~/'-..~
"
ISOPSORALEN
r
(active) (slightly active)

I-
LCO
CH 3

I '" °'.-.0
.# .#
., L(X)
I I '"
OCH3

.,
"<.'>.

.# .#
0'.-.0 ..
LCO
OH

I I '" °'...:0
.-
.# .#
8-METHYLPSORALEN 8-METHOXYPSORALEN 8-HYDROXYPSORALEN
(very active) (very active) (Inactive I

Fig. 3-3. Structures of several photochemically active and inactive psoraiens. It will be noted that substitution
of a methyl or methoxy group in the 8 position of psoralen increases photosensitizing activity, whereas
substitution of a hydroxy group eliminates activity. A similar pattern is observed with substituents in the 5
position. Isopsoralen is also known as angelicin, while 8-methoxypsoralen is also termed methoxsalen or
xanthotoxin.

carotenoids that quench both triplet chlorophyll and singlet oxygen (as well as some types
of free radicals) efficiently. (35) Many other kinds of endogenous photosensitizing com-
pounds are present in some species of plants; however, the mechanisms by which plants
are protected against photodynamic damage by these compounds have not been deter-
mined. (33) Porphyrins in living organisms, such as the heme (ferric protoporphyrin) of
hemoglobin and most other hemoproteins, contain a coordinated paramagnetic ferric ion;
this metal ion shortens the lifetime of the porphyrin excited state to such an extent that it is
no longer an effective photosensitizer. (15)

3.3. SENSITIZED PHOTO EFFECTS ON BIOMOLECULES IN V1TRO(2,3.5)

A number of types of organic molecules characteristic of living material are suscepti-

ble to photosensitized alteration, including examples from the following categories: acids,
alcohols, amines, carbohydrates, esters, growth factors, ketones, nitrogen heterocyclics,
nucleic acids (and certain of their component bases), unsaturated lipids, phenols, proteins
(and certain of their component amino acids), pyrroles, steroids, many vitamins, etc. The
sensitivity of a particular molecule depends on its structure, on the properties of the
photosensitizer, and on the reaction conditions (in particular the pH, the reactant con-
centrations, and the nature of the solvent). In most cases the photosensitized reactions of
biomolecules tum out to be rather complex. Often the primary product will be unstable
and decay very rapidly to secondary products, which may be susceptible to further
sensitized photochemical changes. Reactions of a number of biologically important mole-
cules as studied in solution are described in the next sections.

3.3.1. Lipids(2,5)
Molecules in this category that are susceptible to photosensitized oxidation include
unsaturated fatty acids, unsaturated neutral fats, unsaturated phospholipids, and steroids.
Photosensitization 89

A large amount of research has been done with these compounds because of their impor-
tance in the photodynamic damage to cell membranes and their involvement in the
photosensitized degradation of some types of foods. (36,37) Unsaturated fatty acids can be
photooxidized by both type I and type II processes. Allylic hydroperoxides are initially
formed in both mechanisms, although the product isomer distributions and the reaction
kinetics are different in the two cases. The type I processes are complex, giving allylic
free radicals, which can react in various ways to give different hydroperoxides and, often,
other products such as alcohols, epoxides, and ketones. The main products of the sen-
sitized photooxidation of unsaturated triglycerides and phospholipids are hydroperoxides;
however, dark autooxidation of the products can result in the formation of more complex
mixtures of final products. In type I reactions, cholesterol is converted to a complex
mixture of products including the 7-0.- and 7-I3-hydroperoxides. As mentioned above, the
type II reaction is simpler, yielding the 5-0.-hydroperoxide, with only traces of other
products.

3.3.2. Purines, Pyrimidines, and Nucleic Acids

Although the photosensitized reactions of nucleic acids and their component bases
have been studied extensively, these processes are still not well understood. Purines are
photooxidized more readily than pyrimidines with most sensitizers; guanine and its deriv-
atives are the most sensitive under physiological conditions. Typically, the rates of
photooxidation increase with increasing pH. With guanine derivatives, the rate increases
rapidly above pH 7, attaining a maximum at pH 10.5. The pH curve has an inflection at
pH 9.2, corresponding to the pK for the dissociation of a proton from the NI atom of the
purine. Thus, the guanine anion is much more sensitive to photooxidation than is the
unionized base. The sensitized photooxidation of purines and pyrimidines results in a
complex array of reaction products. In the photooxidation of guanine and its derivatives,
both rings are ruptured and a number of secondary products are produced, including
guanidine, parabanic acid, urea, carbon dioxide, etc. It has been suggested that endo-
peroxides, hydroperoxides, and dioxetanes could be among the early photooxidation
products. Different photosensitizers give different patterns of oxidation products suggest-
ing that both type I and type II reaction pathways are involved. The mechanism of the
flavin-sensitized photooxidation of purines appears to be somewhat different from that
observed with other sensitizers. For example, adenine and its derivatives, which are
photooxidized only very slowly with eosin and methylene blue, are ph0100xidized more
rapidly with flavins. In the case of pyrimidines, the primary product of the photooxidation
of uracil, as observed at very low temperatures, appears to be a hydroperoxide. Certain
uracil derivatives, such as the 5-amino- and 5-hydroxyuracils, are much more rapidly
photooxidized than the parent compound. (2,5,38)
The photodynamic treatment of nucleic acids in solution can significantly alter the
chemical, physical, and biological properties of these compounds. The type of sensitizer
used may have an effect on the reaction mechanism. For example, the majority of the
most effective sensitizers for nucleic acids (acridines, methylene blue, thiopyronine, etc.)
are positively charged and bind to the nucleic acids. This close association may increase
the probability of type I reactions; the photogeneration of free radicals in the nucleic acid
moiety of DNA-sensitizer complexes has been reported. In contrast, certain porphyrins
sensitize the photooxidation of nucleic acids in singlet-oxygen-mediated reactions. Under
90 Chapter 3

physiological conditions, guanine residues in nucleic acids and synthetic polynucleotides

in solutions are more rapidly photooxidized than the other bases. The destruction of
guanine residues produces labile glycosylic bonds in the nucleic acid chain, which can
result in single- and double-strand breaks. Other physical-chemical changes in the nucleic
acids include a decrease in solution viscosity, a decrease in the melting temperature,
spectral shifts, conformational alterations, and an increase in susceptibility to enzymatic
degradation. Changes in the biological properties are also observed. For example, the
infectivity of tobacco mosaic virus RNA as well as the transforming capability of bacterial
DNA transforming principle is lost. Template, messenger, and translational activities of
nucleic acids are altered as well as their antigenic properties. (5,38,39)
On exposure to light, psoralens react with nucleic acid bases in solution, especially
the pyrimidines. Either the 3,4 or the 4' ,5' carbon-carbon double bond of the photoex-
cited psoralen adds across the 5,6 carbon-carbon double bond of the pyrimidine to give a
covalent cycloadduct. In studies using 365-nm radiation, psoralens can form 4' ,5' mono-
adducts with the pyrimidine bases in DNA; certain of these adducts can absorb another
photon and react further giving a 3,4 adduct with a pyrimidine in the complementary
strand of the DNA, thus cross-linking the two strands. (38,40,41) Recently, monoclonal
antibodies have been produced that specifically recognize DNA molecules modified by
illumination with UV -A radiation in the presence of 8-methoxypsoralen; one photoadduct
per 107 base pairs can be detected. (42) Some drugs, such as· chlorpromazine, form
covalent photoadducts with DNA. (27)

3.3.3. Amino Acids and Proteins(2,5,38)

Only five of the amino acids that occur in typical proteins are rapidly photooxidized
in sensitized reactions: cysteine, histidine, methionine, tryptophan, and tyrosine, all of
which have electron-rich side chains. The kinetics and mechanisms of these reactions
depend on the amino acid, the sensitizer, the solvent, the pH, and the other reaction
conditions. With most sensitizers, cysteine is photooxidized largely to the disulfide,
cystine. With crystal violet, however, the main product is cysteic acid. The rate of
photooxidation increases with increasing pH in a fashion indicating that the thiol group
must be unprotonated for the reaction to occur. Histidine is efficiently photooxidized,
typically by a type II process. The primary product is probably an endoperoxide; however,
this is very unstable and decomposes rapidly with the formation of a variety of final
products. There is a marked increase in the rate with increasing pH, showing that the
unprotonated imidazole ring of the molecule is the reactive site. Methionine is photoox-
idized by both type I and type II processes, depending on the sensitizer, the solvent, and
the reaction conditions; pH has only a small effect on the rate. The product in type I
reactions is methional, while the sulfoxide is produced by type II reactions, as described
above. A number of products result from the photooxidation of tryptophan; both free
radical and singlet oxygen pathways appear to be involved, and the array of reaction
products resulting can be complex. One well-established stable product is N-for-
mylkynurenine, which itself is an efficient photosensitizer. The detailed mechanisms of
tyrosine photooxidation are not known, but both type I and type II processes are probably
involved. The reaction products are not well established, although rupture of the tyrosine
ring occurs. The reaction rate increases at high pH indicating that the phenolate anion is
the most reactive form of the molecule. (2,5)
Photosensitization 91

Almost all proteins are rapidly photooxidized on illumination in the presence of

appropriate photosensitizers and oxygen. All classes of enzymes have been studied as well
as blood plasma proteins, hemoglobin, ovalbumin, bacterial toxins, protein hormones, the
structural protein collagen, etc. In most cases, the sites of damage are cysteinyl, histidyl,
methionyl, tryptophyl, and tyrosyl residues in the macromolecule. Photooxidizable resi-
dues exposed at the surface of the protein molecule are degraded more rapidly than
residues located in the interior of the molecule. Photosensitized reactions can be used to
probe the three-dimensional structures of protein molecules by using sensitizers or protec-
tive agents that bind at specific sites on the protein. The pattern of amino acid residue
photooxidation depends on the protein, the sensitizer, and the reaction conditions. For
example, histidyl residues are not photooxidized at low pH but are very sensitive at high
pH, as is the case for the free amino acid. Marked changes in the physical-chemical
properties of photodynamically treated proteins are often observed. These include spectral
changes, increased susceptibility to proteolytic enzymes and to denaturation by urea, and
changes in conformation, electrophoretic mobility, extent of cross-linking, solubility,
ability to bind cofactors, etc. (5)
In most cases, the normal biological function of a protein is changed or destroyed by
sensitized photooxidation. With the exception of horseradish peroxidase and superoxide
dismutase, all enzymes are rapidly inactivated; in a very few cases, enzymes show an
initial increase in activity on photooxidation, followed by inactivation. Peptide hormones
such as glucagon, insulin, etc., lose their biological function, and protein toxins from
bacteria, plants, and snake venoms are inactivated. Proteins lose their antigenicity and
ability to react with specific antibodies. Some photosensitizers form covalent photoad-
ducts with proteins, e.g., riboflavin photobinds to the enzyme lysozyme. In some cases,
covalent protein-protein cross-links are formed on illumination of proteins in the presence
of photodynamic sensitizers. For example, illumination of hemoglobin subunits in the
presence of methylene blue produces covalent dimers and higher aggregates. Sensitized
cross-linking of proteins to nucleic acids and various other molecules can occur. Some
proteins contain chromophores which act as photosensitizers. For example, bacterial
glutamate decarboxylases, which require pyridoxal-5' -phosphate for activity, are inacti-
vated on illumination in the presence of the cofactor. (5)
Psoralens interact photochemically with proteins via mechanisms that depend on the
reaction conditions. UV-A irradiation of solutions containing psoralen-protein mixtures
under anaerobic conditions results in the slow formation of covalent psoralen-protein
adducts. Under aerobic conditions, psoralens sensitize the photoinactivation of enzymes
by a singlet oxygen-mediated process. Finally, if psoralen solutions are illuminated under
aerobic conditions and then mixed with bovine serum albumin in the dark, covalent
psoralen-albumin adducts form; this suggests that reactive products are formed during
illumination that can subsequently react with proteins. (25)

3.3.4. Other Biomolecules

Alcohols have been studied as model compounds for the photosensitized oxidation of
carbohydrates. Anthraquinone and ketonic sensitizers, on illumination, abstract a hydro-
gen atom from the u carbon of the alcohol giving an alcohol radical; this in tum reacts
with oxygen to yield aldehydes or carboxylic acids for primary alcohols, and ketones for
secondary alcohols. Alcohols as well as simple sugars, polysaccharides such as cellulose,
92 Chapter 3

and complex carbohydrates (glycosaminoglycans) are photooxidized with rather low effi-
ciencies. Hexitols such as sorbitol are oxidized to the corresponding hexose, and then to
the hexonic acid. Much research has been carried out on cellulose because of its impor-
tance in the textile industry. Mechanistically, as for alcohols, cellulose is photooxidized
by a type I process, giving fibers with decreased mechanical strength. Thus, dyes used
with textiles should not be efficient photodynamic sensitizers, as mentioned earlier.
Complex polysaccharides such as hyaluronic acid, heparin, and sodium alginate are
photooxidized by free radical processes; these reactions result in chain scission giving
depolymerized products. A number of important molecules that occur in only small
amounts in living material are efficiently photooxidized, including ascorbic acid, biotin,
glutathione, folic acid, lipoic acids, tocopherols, and plant growth hormones such as
indoleacetic acid. (2,5,6)

3.4. SENSITIZED PHOTO EFFECTS ON VIRUSES(6)

Many viruses are efficiently inactivated by photosensitized treatment; this would be

expected since they are made up of nucleic acids and proteins, and, in some cases, lipids.
A large amount of research has been done in this area for a number of reasons, e.g., to
elucidate the mechanisms of viral inactivation, to use photosensitized reactions as probes
of the molecular structure of viruses, to examine the use of photodynamically inactivated
viruses in the preparation of vaccines and immunodiagnostic reagents, and to explore the
use of photosensitized reactions in the therapy of superficial viral infections in laboratory
animals and humans.
Sensitized photoeffects on bacterial viruses (bacteriophages) have been studied ex-
tensively. (6) Both nucleic acids and proteins in bacteriophage can be altered by pho-
todynamic treatment; in many cases nucleic acid strands are broken, DNA-protein cross-
links are formed, and mutant forms of the bacteriophage are sometimes produced. In some
cases, photodynamically damaged bacteriophage can be repaired in appropriate host
bacteria. Rates of photodynamic inactivation vary widely with different strains of bacte-
riophage because of differences in the permeability of the protein head membrane to
sensitizers. As might be expected with a substrate as complicated as bacteriophages,
photosensitized effects can be complex. For example, the inactivation of bacteriophage
with N-formylkynurenine as sensitizer results from singlet oxygen-mediated damage to
the purines of the DNA and to the tail proteins; inactivation also involves the direct
interaction of excited sensitizer with the bacteriophage. With basic sensitizers such as
acridine, acridine orange, proflavine, etc., the targets for bacteriophage P22 photoinac-
tivation are the injection proteins; these appear to be photosensitized by dye molecules
stacked in a region of the bacteriophage DNA located close to these proteins. (43)
Plant viruses, too, are inactivated by photodynamic treatment. For example, tobacco
mosaic virus is inactivated with a number of different sensitizers. The isolated infectious
RNA from tobacco mosaic virus is inactivated more rapidly and more completely than the
virus itself; this suggests that the protein coat of the virus can have a protective effect, as
in the case of some bacteriophages. The main site of photodynamic damage in tobacco
mosaic virus appears to be the guanine residues of the RNA. As with bacteriophage,
photodynamic treatment can produce mutations in plant viruses. (6)
Photosensitization 93

Animal viruses from a number of different families can be inactivated by photosen-

sitized treatment including adeno, herpes, orthomyxo (influenza), papova (SV-40), para-
myxo (measles, mumps, etc.), picorna (polio), pox (vaccinia), reo, toga (equine encepha-
litis), etc. (44) Some viruses, such as polio, are resistant to photosensitized inactivation
since photosensitizers do not penetrate the protein coat. However, if the virus is grown in
the presence of a sensitizer (which results in its incorporation into the viral structure) or if
it is treated with sensitizer at a high pH (which increases the permeability of the protein
coat) it becomes light-sensitive. As with bacteriophages and plant viruses, mutants of
animal viruses have been produced by photodynamic treatment. (6)
A dye-light procedure for the treatment of herpes simplex infections in humans has
been developed. In this, a solution of a photodynamic sensitizer, such as neutral red or
proflavine, is applied to the lesions; the area is then exposed to light. (45) However, there is
considerable concern over the risk/benefit ratio of this type of photochemotherapy since it
is reported that such treatment can transform mammalian cells (Chapter 6). (46,47) The
most common sensitizers used for photoinactivating animal viruses are positively charged
dyes (methylene blue, neutral red, and acridines such as proflavine and acridine orange)
which bind to the viral nucleic acids; psoralens, which efficiently sensitize the photoinac-
tivation of almost all kinds of viruses, intercalate into the viral nucleic acid. Thus, on
illumination in the presence of these sensitizers, the major effect is probably on the viral
nucleic acid. The use of negatively charged sensitizers, such as porphyrins, may result in
viral inactivation by effects on proteins, thus avoiding the production of virus mutants that
might induce tumors. (48) Also, in the case of those viruses containing lipids, the use of
lipid-soluble photosensitizers might avoid effects on nucleic acids and thus decrease the
possibility of producing oncogenic changes in the virus. (49)

3.5. SENSITIZED PHOTOEFFECTS ON CELLS

Phototoxicity is a general term for the damaging or killing of cells by photosensitized

reactions. Mechanistic studies on phototoxicity are difficult because of the extreme com-
plexity of cells. For example, the cell membrane acts as a differential barrier to the
penetration of photosensitizers into the cell because of its charge and hydrophobic core.
Also, cells are not merely tiny bags containing simple solutions but are composed of a
very large number of microregions. Because of the many kinds of molecules making up
cells, this provides microenvironments with a wide array of physical-chemical properties.
As a result, photosensitizer molecules penetrating into cells will not ordinarily be in
simple aqueous solution but, depending on their size, charge, and solubility properties,
will be bound noncovalently at various subcellular sites, be associated with the more
hydrophobic regions of membraneous organelles, etc. Thus, the microscopic pattern of
the cellular localization of sensitizers, and therefore the pattern of subcellular damage
resulting from illumination, can vary with the sensitizer, the cell type, the physiological
condition of the cell, etc. Furthermore, the mechanism, and thus the end product of a
photosensitized reaction, might be altered by the binding of the sensitizer to a cellular
component or by the dielectric constant, oxygen solubility properties, etc., of the micro-
environment. (50)
94 Chapter 3

3.5.1. Model Cell Systems(Sl)

One preliminary approach to the elucidation of reaction mechanisms in biological
systems is to use so-called model systems, systems that simulate in some ways, albeit
much oversimplified, the properties of the real system. For example, changing the di-
electric properties (and solvent properties) of aqueous solutions by the addition of a
miscible organic solvent such as methanol can have significant effects on the efficiencies
and mechanisms of sensitized photooxidations.(S2) In some cases, photosensitizers show
an increased efficiency when bound to nucleic acids or proteins. Bound dye can sensitize
photodamage to the molecule to which it is bound as well as to other molecules in the
reaction system. Sensitizer binding may also favor type I photoprocesses. (S,50) In aqueous
systems, surfactant (detergent) micelles have a charged, hydrophilic exterior and a hydro-
phobic interior, and thus crudely resemble biomembranes. Hydrophobic sensitizers lo-
calize in the inner part of the micelle, and hydrophilic sensitizers can associate with the
surface region; in both cases, on illumination, singlet oxygen can be generated with high
efficiency and oxidize substrates both inside and outside of the micelle. (SO) Similar types
of studies have been carried out with liposomes (microscopic, bilayered phospholipid
vesicles), which structurally are even more like cell membranes.(Sl,53) As additional
model studies are carried out, especially with the more complex organized assemblies, the
resulting information will further our understanding of the mechanistic aspects of the
photosensitized killing of cells and of photodynamic damage to tissues. (54) It should be
stressed that the actual mechanism of cell killing will depend on the sensitizer, the
reaction conditions, and the type of cell involved.

3.5.2. Microorganisms(6,SS-S7)
All types of procaryotic microorganisms (rickettsia, mycoplasmas, blue-green algae,
and bacteria) can be killed on illumination in the presence of appropriate photosensitizers;
most of the studies in this area have been carried out with bacteria. Many categories of
sensitizers are effective, including psoralens. Photosensitized damage can be mediated by
effects on nucleic acids, proteins, or unsaturated lipids. It appears that most sensitizers
have to bind to or penetrate into bacterial cells to be effective. However, Escherichia coli
cells are killed on illumination in the presence of rose bengal bound covalently to water-
insoluble polystyrene beads. This sensitizer remains outside of the cells, so killing proba-
bly results from the diffusion of singlet oxygen into the cells. With some bacteria, the cell
envelope of treated cells is more easily lysed. The swimming behavior of flagellated
bacteria, such as Salmonella typhimurium, is altered by photosensitized reactions; longer
treatment results in the complete loss of motility. A number of the metabolic activities of
bacteria can be inhibited by photodynamic treatment including glycolysis, protein syn-
thesis, respiration, etc.; in some cases, particular enzymes are inactivated. Carotenoids,
which occur in high concentrations in some kinds of bacteria, protect the organisms
against photosensitized damage. Some photosensitization studies have been carried out
with isolated bacterial organelles. For example, photodynamic treatment inactivates the
respiratory system located in cell membranes isolated from Sarcina [utea. Also, isolated
E. coli ribosomes are damaged by photosensitized treatment; both ribosomal RNA and
protein are altered, and with some sensitizers, covalent RNA-protein cross-links are
Photosensitization 95

formed. The photodynamic treatment of intact bacterial cells can produce DNA-protein
cross-links. Furthermore, mutations are produced in various strains of E. coli, S. ty-
phimurium, etc., using acridine dyes, methylene blue, phenothiazines, psoralens, and the
like as photosensitizers. Some strains of bacteria are able to repair certain kinds of
sensitized photodamage. E. coli strains carrying various combinations of genes that
control sensitivity to near-UV and far-UV radiation have been used to analyze the mecha-
nisms of photosensitized cell inactivation and mutagenicity of a number of sensitizers
derived from plants. (58)
Many kinds of photosensitization studies have been performed using eucaryotic
microorganisms. As indicated earlier, modem studies on biological photosensitization
began with the work of Oscar Raab, who used a ciliated protozoan, paramecium, as his
test organism. Since Raab's pioneering work other studies have been carried out with
paramecia; the effects of a number of different types of photosensitizers have been
examined, including carcinogenic hydrocarbons. Xanthene dyes sensitize paramecia to
photobehavioral responses, e.g., positive or negative phobophototactic behavior, depend-
ing on the sensitizer concentration, light intensity, etc. The photodynamic treatment of
paramecium also causes loss of motility, a delay in cell division, and induces a subsequent
increase in sensitivity to heat, Another ciliate, Tetrahymena pyriformis, has also been
used in a number of photodynamic studies. Trypanosomes, which are flagellated pro-
tozooans, and amebas have also been utilized. The green algae Chiorella and
Chlamydomonas, as well as the dinoflagellate Gymnodinium, are inactivated by pho-
todynamic treatment. A large amount of work has been done with yeasts using well-
defined genetic strains. Little is known of the genetic basis of sensitivity to photodynamic
killing in eucaryotic microorganisms, although both sensitive and resistant mutants of
yeast have been isolated. Different mechanisms can be involved in the photosensitized
damage to yeast cells depending on the sensitizer, reaction conditions, etc. Some por-
phyrins, for example, do not penetrate into yeast cells, and thus probably sensitize only
membrane damage. Other sensitizers, such as eosin and rose bengal, penetrate into the
cytoplasm and sensitize damage there, while some basic dyes, including acridine orange,
penetrate into the nucleus where they sensitize damage. Mutations of various types are
readily produced in yeast cells with a number of sensitizers. Many structurally different
psoralens have been studied as photosensitizers for yeast cells. Difunctional psoralen
derivatives efficiently sensitize both cell killing and the induction of nuclear mutations,
but are rather ineffective in the production of' 'petite" cytoplasmic mutations. In contrast,
monofunctional derivatives are active in the photoproduction of petite mutations, but are
poor sensitizers for cell killing and the induction of nuclear mutations. (6,55 ,57,59)

3.5.3. Cells from Multicellular Organisms(6,7)

Cells from multicellular organisms have been used extensively for photosensitization
studies. In the case of higher plants, cell division in isolated leaf mesophyll cells from the
plant Caiystegia (a morning glory) is inhibited by illumination in the presence of 8-
methoxypsoralen. (60) Cercosporin, a complex quinone-type photosensitizer produced by
certain types of plant pathogenic fungi, sensitizes the rapid photodynamic killing of
protoplasts isolated from tobacco cells grown in culture. Damage appears to be mediated
by the photooxidative destruction of membrane lipids that results in the bursting of the
96 Chapter 3

protoplasts. (61) Protoplasts isolated from bean leaf cells show membrane blebbing and
lysis on illumination in the presence of hematoporphyrin derivative as sensitizer. (62) Some
work has been done with cells from invertebrate animals. (6) For example, the treatment of
the sperm of marine invertebrates with photosensitizers causes a loss of motility and a
decrease in the fertilizing capability. Treatment of marine invertebrate eggs can stimulate
parthenogenetic cell division; longer treatment causes loss of fertilizability and blocks the
formation of the fertilization membrane. Extensive studies have been carried out on the
effects of photodynamic treatment on the axons of the nerve cells from the lobster, snail,
octopus, and squid. Large light doses in the presence of photosensitizers that penetrate
into the membrane block sodium channels, depolarize the membranes, and thus stop nerve
impulse propagation. Frog muscle is stimulated to contract by light in the presence of
appropriate photosensitizers. (59)
Mammalian erythrocytes have been a favorite experimental material ever since it was
shown in 1908 that rabbit red blood cells hemolyzed on illumination in the presence of
hematoporphyrin or chlorophyll preparations. (I) The mature mammalian red blood cell is
the simplest biological model for cell membrane studies, since the cells do not contain
nuclei or much in the way of the usual cytoplasmIc organelles. Red blood cells can be
processed to give resealed "ghosts" in which essentially only the plasma membrane
remains. Photodynamic treatment produces an increase in the passive permeability of the
cell membrane to small cations that results in the colloid osmotic lysis of the cells.
Membrane components damaged include unsaturated fatty acids, cholesterol, and en-
zymes; extensive cross-linking of membrane structural proteins also occurs. (55,57 ,59,63-65)
An enormous amount of research has been done in recent years on photosensitized effects
using normal and neoplastic cells from humans and other mammals in culture. (7 ,55,57 ,59,63)
The kinds of lesions expected in photosensitized cells can be predicted. For example, cell
membranes contain enzymes, structural proteins, and unsaturated lipids, all of which can be
photodegraded. Thus, sensitizers located in the membrane would cause alterations in
catalytic and transport processes and changes in membrane permeability and mechanical
properties. Sensitizers located in various regions of the cytoplasm could mediate the
photodegradation of enzymes, tRNA, and components of the various cell organelles.
Finally, with sensitizers that localize in the nuclear region, illumination would result in
guanine destruction in the DNA, nucleic acid strand breaks, and the formation of DNA-
protein adducts; these photochemical changes could in tum interfere with cell division and
produce mutations. All of these changes, which in many cases result in cell death, have
been observed, depending on the cell type, sensitizer, and reaction conditions. Cell survival
curves (see Chapter 2 and Fig. 2-14) typically have an initial curved region (shoulder)
followed by a linear section corresponding to the exponential killing of the cells. The width
of the shoulder, its shape, and the slope of the linear region depend on the sensitizer, the
reaction conditions, and the cell type and physiological status; some cell lines, as studied in
synchronous cultures, show a marked difference in photodynamic sensitivity during differ-
ent phases of the cell division cycle. (7)
With mammalian cells in culture, more is probably known about sensitized effects on
the plasma membrane than on other organelles and components. (7 ,64,66) Effects include
inhibition of membrane transport processes, leakage of potassium and other components
out of the cells, increases in the passive permeability of the membrane to many kinds of
molecules, changes in the mechanical properties of the membrane, alterations in the
binding and antigenic characteristics of the membrane, and changes in membrane mor-
Photosensitization 97

phology (bleb and hole formation). With many kinds of mammalian cells and with
moderately hydrophobic sensitizers such as protoporphyrin, deuteroporphyrin, etc., the
membrane is the principal site of damage. A number of kinds of damage are observed with
photosensitizers that penetrate into the cytoplasm of mammalian cells. In some cases, the
mitochondria show the flrst ultrastructural changes observed in photodynamically treated
cells. Mitochondrial, lysosomal, and cytosol enzymes are inactivated. Typically, both
protein and RNA synthesis are markedly decreased on the photodynamic treatment of
cells. A number of more detailed photodynamic studies have been performed using
organelles isolated from the cytoplasm of mammalian cells. For example, with isolated
mitochondria, a number of changes occur including swelling, structural disruption, un-
coupling and inhibition of oxidative phosphorylation, and inactivation of various en-
zymes. (63,67) Photodynamic treatment alters the membranes of isolated lysosomes permit-
ting lysosomal enzymes to leak out; with isolated micro somes , the mixed-function
oxidase system is inactivated and lipids are peroxidized. Finally, a number of charac-
teristic effects are observed with sensitizers that penetrate into the nucleus. The nuclear
membrane can be damaged, alkali-labile lesions are produced in the DNA, and chromo-
some breaks and sister chromatid exchanges occur. Guanine residues in the DNA strands
of intact cells are altered as demonstrated by immunofluorescence techniques. DNA
strand breaks in mammalian cells are usually repaired rapidly. (55) There are very few
reports of mutation production in mammalian cells using oxygen-requiring photosen-
sitizers. Psoralens appear to be more photomutagenic. (7) As this section clearly indicates,
photodynamically treated mammalian cells can be injured at one or more of numerous
sites, and by a variety of reaction mechanisms, depending on the cell type, sensitizer,
reaction conditions, etc.C7 ,55 ,57 ,59 ,63-67)

3.6. SENSITIZED PHOTOEFFECTS ON MULTICELLULAR ORGANISMS(6,8)

3.6.1. Plants

Under some conditions plants can be injured or killed by reactions photosensitized by

endogenous chlorophyll. For example, if photosynthesis in leaves is blocked by heat
treatment or carbon dioxide deprivation, illumination kills the photosynthetic tissues; no
photodamage is observed in the absence of oxygen. (6) Chloroplasts contain high con-
centrations of carotenoids, which protect against photodynamic damage under ordinary
conditions by quenching triplet chlorophyll and singlet oxygen. Superoxide dismutase,
which is also present in photosynthetic structures, appears to protect leaves against super-
oxide-mediated photodamage.(68) As mentioned in Section 3.2.6, many species of plants
contain endogenous photosensitizing compounds in addition to chlorophylls. It would be
of interest to understand the mechanisms by which such plants are protected against
photodamage, since some of these compounds are potent photosensitizers for plant
tissues. For example, berberine, a quinoline alkaloid synthesized by several plant species,
strongly inhibits the growth of onion root tips on illumination. (69) Illumination of pea
plant tissue in the presence of psoralens, which occur in many plant species, increases the
rate of synthesis of certain proteins in the tissue; this suggests a possible growth pho-
toregulatory function for such compounds. Leaves, stems, and roots of plants are injured
98 Chapter 3

on illumination after treatment with exogenous sensitizers. (6) For example, illumination
of pea leaf tissue after exposure to eosin inhibits photosynthesis, inactivates certain
chloroplast enzymes such as ribulose biphosphate carboxylase, bleaches chlorophyll, and
disrupts chloroplast structure. The inhibition of photosynthesis results from an inter-
ference with electron transport processes in Photosystem II (see Chapter 12).(70) Roots
sensitized with dyes such as erythrosin or rose bengal show a positive phototropism on
unilateral illumination due to growth inhibition on the illuminated side; illumination in the
presence of riboflavin decreases the rate of the hormonally controlled growth of pea stem
sections. Protoplasmic streaming in plant cells and cyclosis in barley root hairs is initially
stimulated and then inhibited by photodynamic treatment. Chromosome breaks in barley
and bean plant tissues are produced as a result of illumination in the presence of basic
photosensitizing dyes such as acridine orange and methylene blue. (6)

3.6.2. Nonmammalian Animals(6)

Relatively little work has been done on the responses of lower invertebrates to
photodynamic treatment, although hydra, several kinds of worms, and rotifers have been
used to a small extent as experimental material, primarily in photokilling studies. (1 ,6) The
nematode worm Aphelenchus avenae is killed efficiently on illumination in the presence
of dithiophene sensitizers and oxygen. (71) The aquatic larval forms of the blood fluke
Schistosoma mansoni are rapidly killed on illumination in the presence of low concentra-
tions of acridine orange and methylene blue. Light calises contraction of the adductor
muscle of larval mussels in the presence of some sensitizers. (6) Nauplii of the brine shrimp
Artemia salina lose motility on illumination in the presence of a number of polycyclic
aromatic hydrocarbons; the relative photosensitizing efficiencies of the compounds used
correlates well with their relative carcinogenic activities. (72) More recently, a number of
studies have been done with insects. The aquatic larvae of several different kinds of
mosquitoes are efficiently killed when illuminated in water containing low concentrations
of photosensitizers. A number of other types of insects including the fruit fly (Dros-
ophila), the housefly, the face fly, the imported fire ant, and the boll weevil can be
photokilled with many different types of sensitizers, including several xanthene dyes
(erythrosin B, rose bengal, etc.), acridine red, substituted butadienes, thiophenes, etc.(73)
The use of photosensitizing compounds in the control of insect pests and the role of
naturally occurring sensitizers in the resistance of plants to insects will be discussed in
Sections 3.8.2 and 3.9.
Only a few photosensitization studies have been reported with the nonmammalian
vertebrates. Tadpoles, newts, and small fish can be killed by photodynamic treatment,
and frogs have been used for some studies on the effects of photodynamic treatment of
muscle tissue and blood vessels. (1,6) Chickens, ducks, geese, and turkeys become light-
sensitive as a result of eating psoralen-containing seeds from several different plant
species. Ducks, which appear to be the most sensitive, suffer phototoxic damage to the
beak, foot webs, periorbital region, and retina; this can result in severe deformation of the
beak and feet, and blindness. (74,75)

3.6.3. Mammals(6,8)
Light responses in sensitized humans and some other mammals can involve two
different mechanisms, phototoxicity and photoallergy, depending on the photosensitizer
Photosensitization 99

and the organism involved. Phototoxic reactions are immediate and will probably occur in
all mammals if sufficiently high sensitizer and light doses are used. (76-79) The photo-
allergic response is delayed, and appears to result from the photosensitized fonnation of
protein "photoantigens," which in tum produce an immunological type of reaction (see
Chapter 6). (78) The phototoxic response in white mice, for example, involves a charac-
teristic array of symptoms on illumination including hyperactivity, skin itching, skin injury
(reddening, edema, necrosis, and ulceration), and retinal damage. In addition, intestinal
hemorrhage, decreased blood pressure, and circulatory collapse occur in severe phototox-
icity; these responses probably result from the photodynamic fonnation of toxic materials in
the skin that are then transported to all parts of the body in the bloodstream.<I) Other
mammals, including rats, guinea pigs, rabbits, and dogs, show similar responses.(77,79)
With most sensitizers, such as porphyrins, oxygen is necessary for phototoxic responses.
Psoralens can sensitize phototoxic skin reactions that are typically expressed as erythema
followed by significant skin darkening. (26) Some hereditary or drug-induced diseases in
mammals result in the accumulation of abnonnal amounts of iron-free porphyrins in the
organism. These conditions, tenned porphyrias, often make the animal highly photosen-
sitive. (80) For example, mice fed with large doses of the antibiotic griseofulvin accumulate
excessive amounts of protoporphyrin, a phototoxic porphyrin, in the liver and red blood
cells, with smaller amounts in the blood serum. On illumination, these mice show acute
phototoxic responses in the ears, tail, and depilated skin. (15) Some chemicals protect
mammals against photosensitized injury. For example, ~-carotene, which is an efficient
quencher of triplet photosensitizers, free radicals, and singlet oxygen, partially protects
mice against photosensitization with hematoporphyrin. (8\) Thiols and a-tocopherol also
have some protective action. (6)
Cows, sheep, and other grazing mammals sometimes show a type of photosensitivity
resulting from a disturbance in liver function. The photosensitizer involved is phyl-
loerythrin, a phototoxic derivative of chlorophyll produced in the animal gut by bacterial
action. This material is nonnally excreted in the bile; however, in a number of types of
liver dysfunction it accumulates in the body to serious photosensitizing levels. These
conditions can be produced by chemicals tenned hepatotoxins that occur in several kinds
of plants; in addition, some mycotoxins produced by fungi growing on plants cause liver
damage. In most cases, the chemical nature of the toxin has not been established. Some
viral diseases in animals damage the liver and thus induce a secondary photosensitization.
Finally, some breeds of sheep have a hereditary abnonnality in the phylloerythrin-excret-
ing mechanism of the liver. Ligation of the bile duct in sheep can result in photodynamic
death if the animal is fed green plants; if the animal is protected from light, or if it is fed
grain or some other chlorophyll-free food, photosensitivity does not develop. A hereditary
type of porphyria sometimes occurs in cattle, which results in the accumulation of pho-
totoxic porphyrins including protoporphyrin and uroporphyrin in the body. Cattle with
this condition (tenned "pink tooth," inherited as a simple recessive) show a reddish
brown pigmentation of the bones, teeth, and urine; hairless and lightly pigmented regions
of the skin become highly photosensitive. (75,82-84)
As mentioned above, many species of plants contain potent photosensitizers other
than chlorophylls. If grazing animals eat such plants they can become photosensitive; this
is a serious problem in certain areas of the world. Phototoxic effects are greatest in
unpigmented areas of the skin; historically, white sheep and white horses could not be
raised in some geographic regions. Plants of the genus Hypericum are especially trou-
100 Chapter 3

blesome since many species contain a potent photosensitizer, hypericin (or related com-
pounds); the structure of this sensitizer is shown in Fig. 3-2. Many Fagopyrum species
(buckwheat) also produce pigments (fagopyrins) that photosensitize grazing animals and
swine. Because of this, buckwheat leaves are not used as fodder; fortunately, the buck-
wheat grain contains little photosensitizer, so that eating buckwheat cakes should not be
hazardous to your health! Finally, members of several plant genera (including Ammi and
Cymopterus) contain psoralens, which are very effective photo sensitizers as described
above. In some areas these plants pose a significant threat to sheep, for example. (75,82,85)

3.6.4. Humans
Photosensitization in humans will be considered only very briefly since this area will
be covered in Chapter 6 on photomedicine. Light sensitivity in humans accompanied by
an excretion of abnormal porphyrins in the urine was observed in the last century.
Experimentally, photosensitization in humans was first studied in 1913 by Meyer-Betz, a
physician, who injected himself with 200 mg of hematoporphyrin, and then exposed parts
of his body to sunlight in controlled ways. Exposure resulted in an immediate burning
sensation in the exposed skin and in a few minutes erythema, edema, and some pain
started to develop; the severity of the reaction increased with increasing light dose.
Meyer-Betz's skin remained light-sensitive for two months.(8) As in the case of other
mammals, humans show both phototoxic and photoallergic responses.(78) Photosensitiza-
tion in humans is not uncommon and can result from ingestion of a sensitizer, from skin
contact with a sensitizer, from an injection of a photosensitizing drug, or from certain
diseases including hereditary and drug-induced porphyrias, pellagra, and lupus erythema-
tosus. Many individuals are photosensitized by commonly used drugs including certain
antibiotics, antihistamines, antiseptics, barbiturates, diuretics, sulfonamides, phe-
nothiazine tranquilizers such as chlorpromazine, etc. Also, some chemicals used in agri-
culture, industry, the home, and research are phototoxic. (76,78) For this reason, com-
pounds being considered for commercial use should be routinely screened for possible
photosensitizing activity. (79) Furthermore, adequate precautions should be taken in medi-
cal practice when using drugs that might photosensitize patients. Individuals with the
hereditary disease erythropoietic protoporphyria (inherited as a simple dominant) have
unusually high levels of protoporphyrin in their erythrocytes, feces, and serum. On
illumination with light that is absorbed by the porphyrin, many of these people have
itching and burning sensations in the exposed skin areas, followed by the development of
erythema and edema; over a period of time more severe skin damage can develop. Persons
with porphyria cutanea tarda (a type of drug-induced porphyria) accumulate abnormal
levels of uroporphyrin in the body; in many cases this results in high sensitivity to
light. (86,87) Some success has been obtained in using large doses of ~-carotene for the
treatment of patients with photosensitizing porphyrias, especially those with erythro-
poietic protoporphyria. (81)

3.6.5. Sensitized Photocarcinogenesis

It is well known that chronic exposure to UV-B radiation induces skin cancers in
laboratory animals (and presumably in humans). (88 .89) Cancers can also be induced by
visible or UV -A radiation in animals treated with certain photosensitizers. For example, it
Photosensitization 101

has been reported that laboratory animals treated with hematoporphyrin, proflavine, or
neutral red develop cancers on chronic exposure to light. None of these compounds act as
chemical carcinogens in the dark. (90) Animals develop cancers when exposed to UV-A
radiation following treatment with photosensitizing psoralens. (26) The mechanisms in-
volved in photosensitized carcinogenesis are not known. However, the phenomenon is
perhaps not unexpected since photosensitized treatment can produce chemical alterations
in DNA, chromosome breaks, and sister chromatid exchanges in mammalian cells. Such
treatment also transforms mammalian cells, and cells transformed by PUVA treatment can
produce tumors when implanted in animals. Many photosensitized reactions produce
hydrogen peroxide and singlet oxygen; it has been suggested that both of these species can
act as chemical carcinogens. (7.26)

3.7. NATURAL PHOTOSENSITIVITY

Some kinds of cells are killed or injured on illumination with cool white or daylight-
type fluorescent lamps (which emit visible light and some UV-A radiation). Cellular
damage in this case may be mediated by light absorbed by the flavin and! or heme
components of certain proteins. For example, light treatment of isolated rat hepatocytes
severely damages catalase (a heme-containing enzyme), microsomes, and mitochondria;
the cell membranes are not affected. Microsomes and mitochondria isolated from the cells
are also damaged by light. Photodamage is decreased by antioxidants such as ascorbic
acid and a-tocopherol. In many mammals, the retinal cells are very light-sensitive; recent
studies suggest that the damage is mediated by singlet oxygen generated by illuminated
retinal in the cells. Many kinds of mammalian cells exhibit light sensitivity when incu-
bated in tissue culture media that contain flavins. Cell killing, production of chromosome
breaks, formation of cross-links in DNA, and neoplastic transformation have been re-
ported. These effects may be caused by hydrogen peroxide produced in the flavin-
sensitized photodynamic degradation of tryptophan or tyrosine in the culture medium. The
most effective wavelengths for both the endogenous and medium-sensitized effects are in
the UV-A and blue regions; thus, mammalian cell cultures should probably be handled
and stored under lights emitting only at longer wavelengths. Natural photosensitivity,
presumably sensitized by endogenous chromophores, has been observed with many kinds
of microorganisms. Effects such as cell killing and mutagenesis commonly occur. Even a
few multicellular animals are injured or killed by visible light. For example, the small
aquatic annelid Tubifex is killed within a few minutes on illumination in the presence of
oxygen; this organism is normally found in habitats where both the light intensity and the
oxygen concentration are low. (6.7, IS)

3.S. BIOLOGICAL APPLICATIONS OF PHOTOSENSITIZED REACTIONS

In general, photosensitized reactions are harmful to living organisms. They are also
harmful in other ways. For example, some colored textiles are photodegraded in reactions
sensitized by the dyes used. Light degrades many kinds of beverages and foods via
sensitized reactions. Beer rapidly develops an unpleasant flavor on exposure to radiation
102 Chapter 3

in the UV -A and blue regions of the spectrum. This results from riboflavin-sensitized
photooxidation reactions that produce certain kinds of organosulfur compounds. For this
reason, the glass (typically brown) used in the manufacture of beer bottles is designed to
absorb radiation in the UV-A and blue regions. Milk also develops an off-flavor on
exposure to light, again by flavin-sensitized reactions. Many foods, in particular those
containing high concentrations of unsaturated lipids, are light-sensitive, including olive
oil, cooking oils, and snack foods such as potato and com chips.(36,37)
Although we tend to think of photosensitized reactions as being detrimental, a
number of attempts have been made to utilize such reactions in the solution of practical
problems. Some of these are described below.

3.S.1. The Health Sciences(91)

The most successful applications of photosensitized reactions in the health sciences
have been in the treatment of certain diseases (see Chapter 6). Light can be used in therapy
in two ways. In phototherapy, the light involved is absorbed by endogenous molecules in
the organism. The best known example of this is the use of blue light to treat premature
infants with hyperbilirubinemia. (92) In this condition, which is common in premature
infants, bilirubin (a normal breakdown product of hemoglobin) can accumulate in the
blood of an infant to levels that produce irreversible, brain damage. Illumination of
hyperbilirubinemic infants with high-intensity light of wavelengths absorbed by bilirubin
causes photochemical reactions that result in an increased excretion of bilirubin and thus
brain damage is prevented. Very large numbers of infants are treated with this technique
each year.
In photochemotherapy, an exogenous photosensitizing drug is first administered to
the patient; then after an appropriate time interval, the patient is illuminated with light of
wavelengths absorbed by the sensitizer. The most successful application of this technique
has been in the treatment of plaque-type psoriasis, an inflammatory disease of unknown
etiology in which increased epidermal proliferation results in the formation of raised,
scaly, red patches on the skin. Patients are typically given an oral dose of 8-methoxy-
psoralen, and then, after a short wait, the psoriatic areas are irradiated using UV-A lamps.
This type of treatment is usually termed PUVA therapy. In some cases, the sensitizer is
applied topically to the psoriatic areas of the skin before irradiation. Many thousands of
patients with psoriasis are treated each year with this form of therapy, and with a high
degree of success. Other skin diseases can also be treated with PUVA therapy. (10,93)
More recently, a photochemotherapeutic technique has been applied to the treatment
of tumors. It has been known for a number of years that certain porphyrins are selectively
retained in solid tumors in laboratory animals and humans. Since the porphyrin retained in
tumors fluoresces bright red on illumination with blue or UV-A radiation, this permits the
visual localization of tumors, especially small metastases, Illumination, usually with a
laser, of a tumor with retained photosensitizing porphyrins can destroy the tumor with
relatively little effect on the adjacent normal tissue. Several thousand patients have now
been treated by this technique, termed photodynamic therapy (PDT), with generally
encouraging results. It appears to be especially promising for endobronchial tumors and
superficial bladder tumors.(1l-15,94,95) Much research needs to be done in this area to
develop more specific site-selective photosensitizing drugs. One approach that looks very
Photosensitization 103

promising is to covalently attach a photosensitizer, such as a porphyrin, to monoclonal

antibodies directed toward particular types of tumor cells. (96)
Photosensitized reactions have been used in other ways in the health sciences. For
example, they have been used to inactivate viruses and snake venom toxins for use in the
preparation of vaccines and antivenins. In the public health area, it has been proposed that
photosensitized reactions might be an economical tool for water purification. In particu-
lar, it has been shown that the treatment of effluents from sewage disposal plants with
photosensitizers such as methylene blue followed by exposure to artificial light or sunlight
efficiently destroys both bacteria and viruses. Chemical pollutants, such as phenols in
industrial wastewater, can also be destroyed by photosensitized reactions. Only rapidly
degraded sensitizers should be used in these applications to avoid building up detrimental
levels of photosensitizing molecules in the environment. (91)

3.8.2. The Agricultural Sciences(73)

There is considerable interest at present in the use of photodynamic reactions to
control insect pests. Although many kinds of conventional insecticides have been success-
fully used, they often show undesirable side-effects, including toxicity to nontarget orga-
nisms such as fish and mammals, including humans; the induction of mutations and
cancers; and long-term persistence in the environment. A number of synthetic sensitizing
dyes, including acridine red, methylene blue, and xanthenes (especially halogenated
fluoresceins such as erythrosin, phloxine, and rose bengal), photosensitize the injury and
killing of the larvae and/or adults of some economically important insects. Susceptible
organisms include the mosquito, codling moth, house fly, cockroach, mealworm, import-
ed fire ant, boll weevil, face fly, cabbage butterfly, cabbage looper, and com earworm. In
preliminary field tests, it appears possible to control both mosquitoes and house flies
under certain conditions. Some sensitizers such as erythrosin B are inexpensive, appear to
be relatively nontoxic to mammals at the concentrations used, and are rapidly pho-
todegraded in the environment. However, more detailed toxicological studies need to be
carried out. Hopefully, synthetic photoinsecticides will have useful applications, es-
pecially in the control of insect pests around poultry-producing facilities, barns, feedlots,
manure storage areas, etc.(73,97,98)
Plants synthesize a large number of different phototoxic compounds.(33,99,lOO) Re-
cently, in addition to the work with synthetic sensitizers, considerable interest has devel-
oped in the use of some of these naturally occurring materials as photoinsecticides. For
example, studies have been made using a-terthienyl, a compound synthesized by many
plants of the aster family. When applied to experimental ponds at appropriate concentra-
tions, this compound sensitizes the photokilling of the larvae of the mosquito Aedes
intrudens with relatively little effect on nontarget organisms including the caddis fly ,
shrimp, snail, and trout. (101) The furanocoumarins, 8-methoxypsoralen, and sphondin,
which are synthesized by some plants, efficiently sensitize the delayed photokilling of
early instar larvae of the mosquito Aedes aegypti; later instar larvae are less sensitive. (102)
Although questions must be answered about the environmental safety of these types of
photosensitizers, the preliminary work is very encouraging. Also, there are many addi-
tional types of naturally occurring photosensitizers to be tested. (33)
A novel development in herbicides has occurred recently, based on knowledge of the
104 Chapter 3

chemical pathways involved in the biosynthesis of chlorophylls in plants. Application of a

tetrapyrrole precursor, such as aminolevulinic acid, to plants leads to the synthesis of
large amounts of the tetrapyrrole intermediates formed in the chlorophyll pathways. These
accumulate and then sensitize the photodynamic killing of the plant; this occurs within
only a few hours for susceptible species. Considerable selectivity of killing is observed,
depending on the plant species. For example, monocotyledenous plants such as barley,
corn, and wheat show low sensitivity, while dicotyledenous weeds including lambs-
quarter, purslane, and mustard are highly sensitive. These different responses appear to
depend on the tetrapyrrole biosynthetic pathways and the rates of turnover of tetrapyrroles
in the different plant species. (103)

3.9. ROLE OF PHOTOTOXIC PHYTOCHEMICALS IN THE ECOLOGY AND

COEVOLUTION OF PLANTS AND HERBIVORES

Photosynthetic plants produce and accumulate an enormous array of so-called sec-

ondary substances or metabolites (allelochemicals), most of which have no known meta-
bolic, nutritional, or physiological function in the plant itself. Large numbers of these
phytochemicals are toxic to herbivores, such as mammals and insects, as well as to
bacteria and fungi. Thus, they appear to play an important ecological role in the defense of
plants against other organisms, especially herbivores. These materials are highly diverse
chemically, including alkaloids, cyanogenic and triterpenoid glycosides, flavanoids, non-
protein amino acids, phenols, etc. (104) A large number of these secondary plant metabo-
lites are potent photosensitizers, typically with absorption peaks in the UV -A region of the
spectrum. Thus, in the presence of sunlight, phototoxic reactions may playa significant
ecological role in the defense of plants against predators. (99,105-109) Some of these plant
phototoxins, such as the thiophenes and furanocoumarins, are being studied as potentially
commercial photoinsecticides, as described in the preceding section.
It has been suggested that secondary plant substances are involved importantly in
herbivorous insect-plant coevolution. (110) In this type of evolutionary process it is envi-
sioned that, by mutation and recombination, photosynthetic plants would evolve capable
of producing compounds that, by chance, decrease their suitability as food; such plants
would suffer less predation than their ancestors. However, insects, by mutation, could
evolve mechanisms to resist the new phytochemicals. (Ill)
Recent experimental studies on the associations observed between furanocoumarin-
producing plants and certain insect species furnish good supporting evidence for the
process of coevolution. (111) Furanocoumarins occur in hundreds of species of higher
plants, almost all of which are in two plant families, the Umbelliferae (carrot-parsley
family) and the Rutaceae (citrus family). Furanocoumarins occur in two structural forms,
linear (psoralens) and angular (isopsoralens), as shown in Fig. 3-3. The angular forms
occur in lower concentrations in plants than the linear forms. Caterpillars of the black
swallowtail butterfly (Papilio polyxenes) feed exclusively on Umbelliferae that contain
high levels of linear furanocoumarins; in spite of this, they do not show phototoxic
responses. The caterpillars avoid plants that contain appreciable quantities of angular
furanocoumarins, which, in contrast to the linear compounds, are phototoxic to them. In
this species, tolerance to linear furanocoumarins results from the rapid and almost com-
Photosensitization 105

plete detoxification of the sensitizer by microsomal mixed-function oxidases located in

both the midgut and body cells. Angular psoralens are metabolized more slowly and thus
can apparently accumulate to photosensitizing levels.(112,113) In contrast to the resistance
of the black swallowtail caterpillars, many insects, such as the armyworm, which feed on
a variety of plants are lethally photosensitized by furanocoumarins; experiments show that
the armyworm detoxifies furanocoumarins only very slowly.(113) There is evidence that
the production of angular furanocoumarins by plants has evolved more recently than
that of the linear compounds. It has been suggested that this may be a coevolutionary
response to the selective pressure of herbivores, such as the black swallowtail, that can
detoxify the linear furanocoumarins. (114) In conclusion, it should be mentioned that some
insects avoid the phototoxic effects of plant photosensitizers by other methods. For
example, dark pigmentation in the cuticle of the insects could decrease light sensitivity.
Also behavioral responses are involved. Some insect species feed only at night and thus
avoid high light intensities. Others (leaf rollers) avoid light by wrapping themselves in
leaves, while some bore into the stems of plants where they are protected,o 1l.1l5)

3.10. CONCLUSIONS

Basic studies of photosensitized processes as described in this chapter are quite

justifiable in their own right in terms of increasing our understanding of this important and
rather commonly occurring photobiological phenomenon. In addition, from the examples
given, it can be seen that photosensitized reactions have become useful tools in some areas
including medicine and agriculture. Also, phototoxic reactions involving natural sen-
sitizers in plants are contributing to our understanding of some ecological patterns as well
as coevolutionary mechanisms.
We now understand the initial processes involved in photosensitized reactions rea-
sonably well, including the production and properties of the excited states of sensitizers,
the immediate reactions of triplet sensitizers by free radical and energy transfer (singlet
oxygen) pathways, and the chemical behavior of the various reactive oxygen species that
can be produced. Our understanding of the mechanistic organic chemistry involved in the
sensitized photodegradation of biomolecules is less complete. Finally, although much
excellent work has been done in recent years, our understanding of events at the cellular
and organismal levels is still very incomplete. Clearly, there is much research yet to be
done on photosensitized reactions in biological systems.

3.11. REFERENCES

1. H. F. Blum, Photodynamic Action and Diseases Caused by Light, Reinhold, New York (1941). (Reprinted
in 1964 with an updated appendix by Hafner Pub!., New York.)
2. C. S. Foote, Photooxidation of biological model compounds, in: Oxygen and Oxy-Radicals in Chemistry
and Biology (M. A. 1. Rodgers and E. L. Powers, eds.), pp. 425-439, Academic Press, New York
(1981).
3. C. S. Foote, Mechanisms of photooxygenation, in: Porphyrin Localization and Treatment of Tumors (D.
R. Doiron and C. 1. Gomer, eds.), pp. 3-18, Alan R. Liss, New York (1984).
4. G. Laustriat, Molecular mechanisms of photosensitization, Biochimie 68, 771-778 (1986).
106 Chapter 3

5. R. C. Straigh'; and J. D. Spikes, Photosensitized oxidation of biomolecules, in: Singlet 020 Vol. 4 (A. A.
Frimer, ed.). pp. 91-143, CRC Press, Boca Raton (1985).
6. J. D. Spikes, Photodynamic reactions in photomedicine, in: The Science of Photomedicine (J. D. Regan
and J. A. Parrish, eds.), pp. 113-144, Plenum Press, New York (1982). (This paper lists a large number
of earlier reviews on photosensitizatiou.)
7. J. D. Spikes, Photosensitization in mammalian cells, in: Photoimmunology (1. A. Parrish, M. L. Kripke,
and W. Morison, eds.), pp. 23-49, Plenum Press, New York (1983).
8. J. D. Spikes, Photosensitized reactions in mammals, in: Experimental and Clinical Photoimmunology,
Vol. I (R. A. Daynes and J. D. Spikes, eds.), pp. 70-80, CRC Press, Boca Raton (1983).
9. J. D. Regan and J. A. Parrish (eds.), The Science of Photomedicine, Plenum Press, New York (1982).
10. W. L. Morison, Phototherapy and Photochemotherapy of Skin Diseases, Praeger, Westport (1983).
II. D. Kessel and T. J. Dougherty (eds.), Porphyrin Photosensitzation, Plenum Press, New York (1983).
12. D. R. Doiron and C. J. Gomer (eds.), Porphyrin Localization and Treatment of Tumors, Alan R. Liss,
New York (1984).
13. A. Andreoni and R. Cubeddu (eds.), Porphyrins in Tumor Phototherapy, Plenum Press, New York
(1984).
14. G. Jori and C. Perria (eds.), Photodynamic Therapy of Tumors and Other Diseases, Libreria Progetto,
Padova (1985).
15. G. Jori and J. D. Spikes, Photobiochemistry of porphyrins, in: Topics in Photomedicine (K. C. Smith,
ed.), Plenum Press, New York (1984).
16. R. V. Bensasson, G. Jori, E. J. Land, and T. G. Truscott (eds.), Primary Photo-Processes in Biology and
Medicine, Plenum Press, New York (1985).
17. T. B. Fitzpatrick and M. A. Pathak, Historical aspects of methoxsa1en and other furocoumarins, J. Invest.
Dermatol. 32, 299-312 (1959).
18. M. A. J. Rodgers and E. L. Powers (eds.), Oxygen and Oxy-Radicals in Chemistry and Biology, Academ-
ic Press, New York (1981).
19. W. Bors, M. Saran, and D. Tait (eds.), Oxygen Radicals in Chemistry and Biology, Walter de Gruyter,
Berlin (1984).
20. M. A. J. Rodgers, Activated oxygen, in: Primary Photo-Processes in Biology and Medicine (R. V.
Bensasson, G. Jori, E. J. Land, and T. G. Truscott, eds.), pp. 181-195, Plenum Press, New York (1985).
21. N. I. Krinsky, Biology and photobiology of singlet oxygen, in: Oxygen Radicals in Chemistry and Biology
(W. Bors, M. Saran, and D. Tait, eds.), pp. 453-463, Walter de Gruyter, Berlin (1984).
22. A. A. Frimer (ed.), Singlet 020 4 vols., CRC Press, Boca Raton (1985).
23. P.-S. Song, Photochemistry and photobiology of psoralens, Photochem. Photobiol. 29, 1177-1197
(1979).
24. G. Rodighiero, F. Dall' Acqua, and M. A. Pathak, Photobiological properties of monofunctional
furocoumarin derivatives, in: Topics in Photomedicine (K. C. Smith, ed.), pp. 319-398, Plenum Press,
New York (1984).
25. F. Dall'Acqua, S. Caffieri, and G. Rodighiero, Photoreactions of furocoumarins (psoralens and an-
gelicins), in: Primary Photo-Processes in Biology and Medicine (R. V. Bensasson, G. Jori, E. J. Land,
and T. G. Truscott, eds.), pp. 259-272, Plenum Press, New York (1985).
26. F. Dall' Acqua and G. Rodighiero, Biological and medical aspects of furocoumarins (psoralens and
angelicins), in: Primary Photo-Processes in Biology and Medicine (R. V. Bensasson, G. Jori, E. J. Land,
and T. G. Truscott, eds.), pp. 277-293, Plenum Press, New York (1985).
27. I. E. Kochevar, Phototoxicity mechanisms: Chlorpromazine-photosensitized damage to DNA and cell
membranes, J. Invest. Dermatol. 76, 59-64 (1981).
28. C. S. Foote, F. C. Shook, and R. B. Abakerli, Characterization of singlet oxygen, in: Methods in
Enzymology, Vol. 105, Oxygen Radicals in Biological Systems (L. Packer, ed.), pp. 36-47, Academic
Press, New York (1984).
29. J. B. Verlhac, A. Gaudemer, and I. Kraljic, Water soluble porphyrins and metalloporphyrins as photosen-
sitizers in aerated aqueous solutions. I. Detection and determination of quantum yields of formation of
singlet oxygen, J. Chimie 8, 401-406 (1984).
30. R. V. Bensasson, E. J. Land, and T. G. Truscott, Flash Photolysis and Pulse Radiolysis: Contributions to
the Chemistry of Biology and Medicine, Pergamon Press, Oxford (1983).
Photosensitization 107

31. M. A. J. Rodgers, Instrumentation for the generation and detection of transient species, in: Primary Photo-
Processes in Biology and Medicine (R. V. Bensasson, G. Jori, E. J. Land, and T. G. Truscott, eds.), pp.
1-23, Plenum, New York (1985).
32. L. Santamaria and G. Prino, List of the photodynamic substances, in: Research Progress in Organic,
Biological and Medicinal Chemistry, Vol. 3, Part 1 (U. Gallo and L. Santamaria, eds.), pp. XI-XXXV,
North-Holland, Amsterdam (1972).
33. G. H. N. Towers, Photosensitizers from plants and their photodynamic action, Progr. Phytochem. 6, 183-
202 (1980).
34. H. E. A. Kramer, The lightfastness or the non-photochemistry of dyes, Chimia 40, 160-169 (1986).
35. N. I. Krinsky, Photobiology of carotenoid protection, in: The Science of Photomedicine O. D. Regan and
J. A. Parrish, eds.), pp. 397-407, Plenum Press, New York (1982).
36. J. D. Spikes, Photodegradation of foods and beverages, in: Photochemical and Photobiological Reviews,
Vol. 6 (K. C. Smith, ed.), pp. 39-85, Plenum Press, New York (1981).
37. I. Rosenthal, Photooxidation of foods, in: Singlet Db Vol. 4 (A. A. Frimer, ed.), pp. 145-163, CRC
Press, Boca Raton (1985).
38. J. Amagasa, Dye binding and photodynamic action, Photochem. Photobiol. 33, 947-955 (1981).
39. S. Kawanishi, S. Inoue, and S. Sano, Photodynamic guanine modification by hematoporphyrin is specific
for single-stranded DNA with singlet oxygen as a mediator, 1. Bioi. Chem. 261, 6090-6095 (1986).
40. G. D. Cimino, H. G. Gamper, S. T. Isaacs, and J. E. Hearst, Psoralens as photoactive probes of nucleic
acid structure and function: Organic chemistry, photochemistry, and biochemistry, Annu. Rev. Biochem.
54, 1151-1193 (1985).
41. J. Cadet, M. Berger, C. Decarroz, J. R. Wagner, 1. E. Van Lier, Y. M. Ginot, and P. Vigny, Photosen-
sitized reactions of nucleic acids, Biochimie 68, 813-834 (1986).
42. R. M. Santella, N. Dharmaraja, F. P. Gasparro, and R. L. Edelson, Monoclonal antibodies to DNA
modified by 8-methoxypsoralen and ultraviolet A light, Nucleic Acid Res. 13, 2533-2544 (1985).
43. J. L. Bryant, Jr., and J. King, DNA injection proteins are targets of acridine-sensitized photoinactivation
of bacteriophage P22, J. Mol. Bioi. 180, 837-863 (1984).
44. c. V. Hanson, Inactivation of viruses for use as vaccines and immunodiagnostic reagents, in: Medical
Virology, Vol. 2, Proc. Int. Symp. 1982, pp. 45-79, Elsevier, Amsterdam (1983).
45. J. L. Melnick and C. Wallis, Photodynamic inactivation of herpesvirus, in: The Science of Photomedicine
(1. D. Regan and J. A. Parrish, eds.), pp. 545-559, Plenum Press, New York (1982).
46. F. Rapp and J.-L. H. Li, Transforming activity of viruses after dye-light inactivation, in: The Science of
Photomedicine (1. D. Regan and J. A. Parrish, eds.), pp. 571-594, Plenum Press, New York (1982).
47. L. E. Bockstahler, T. P. Coohill, K. B. Hellman, C. D. Lytle, and J. E. Roberts, Critical review and risk
evaluation of photodynamic therapy for herpes simplex, in: Int. Encycl. Pharmacol. Ther. (Viral Chem-
other.), Vol. 1 (D. Shugar, ed.), pp. 479-509, Pergamon Press, Oxford (1984).
48. L. E. Schnipper, A. E. Lewin, M. Swartz, and C. S. Crumpacker, Mechanisms of photodynamic
inactivation of herpes simplex viruses: Comparison between methylene blue, light plus electricity, and
hematoporphyrin plus light, J. Clin. Invest. 65, 432-438 (1980).
49. W. Snipes, G. Keller, J. Woog, T. Vickroy, R. Deering, and A. Keith, Inactivation of lipid-containing
viruses by hydrophobic photosensitizers and near-ultraviolet radiation, Photochem. Photobiol. 29, 785-
790 (1979).
50. G. Jori and J. D. Spikes, Photosensitized oxidations in complex biological structures, in: Oxygen and Oxy-
Radicals in Chemistry and Biology (M. A. 1. Rodgers and E. L. Powers, eds.), pp. 441-459, Academic
Press, New York (1981).
51. J. D. Spikes, Photosensitizing properties of porphyrins in model cell systems, in: Porphyrins ill Tumor
Phototherapy (A. Andreoni and R. Cubeddu, eds.), pp. 51-60, Plenum Press, New York (1984).
52. E. Reddi, M. A. J. Rodgers, J. D. Spikes, and G. Jori, The effect of medium polarity on the hema-
toporphyrin-sensitized photooxidation of L-tryptophan, Photochem. Photobiol. 40, 415-421 (1984).
53. L. I. Grossweiner and G. Goyal, Photosensitization of liposomes by porphyrins, 1. Photochem. 25,253-
265 (1984).
54. J. D. Spikes and R. C. Straight, Photodynamic behavior of porphyrins in model cell, tissue and tumor
systems, in: Photodynamic Therapy of Tumors and Other Diseases (G. Jori and C. Perria, eds.), pp. 45-
53, Libreria Progetto, Padova (1985).
108 Chapter 3

55. T. Ito, Photodynamic agents as tools for cell biology, Photochem. Photobiol. Rev. 7, 141-186 (1983).
56. G. Bertoloni, A. Viel, A. Grossato, and G. Jori, The photosensitizing activity of haematoporphyrin on
mollicutes, J. Gen. Microbiol. 131,2217-2223 (1985).
57. J. Moan, Porphyrin photosensitization and phototherapy, Photochem. Photobiol. 43,681-690 (1986).
58. R. W. Tuveson, M. R. Berenbaum, and E. E. Heininger, Inactivation and mutagenesis by phototoxins
using Escherichia coli strains differing in sensitivity to near- and far-ultraviolet light, J. Chem. Ecol. 12,
933-948 (1986).
59. 1. P. Pooler and D. P. Valenzeno, Dye-sensitized photodynamic inactivation of cells, Med. Phys. 8,614-
628 (1981).
60. P. Ries and R. Beiderbeck, Photosensibilization of isolated mesophyll cells of Calystegia sepium L., Z.
Naturforsch. 34c, 1215-1217 (1979).
61. M. E. Daub, Peroxidation of tobacco membrane lipids by the photosensitizing toxin, cercosporin, Plant.
Physiol. 69, 1361-1364 (1982).
62. B. Kjeldstad, M. L. Olsten, T. Christensen, and A. Johnsson, Photodynamic damage to protoplasts with a
hematoporphyrin derivative, Plant Sci. 44, 139-143 (1986).
63. G. Moreno, Photosensitization of mammalian cells by psoralens and porphyrins, Biochimie 68, 869-873
(1986).
64. D. P. Valenzeno, Photomodification of biological membranes with emphasis on singlet oxygen mecha-
nisms, Photochem. Photobiol. 46, 147-160 (1987).
65. A. W. Girotti, J. D. Thomas, and J. E. Jordan, Prooxidant and antioxidant effects of ascorbate on
photosensitized peroxidation of lipids in erythrocyte membranes, Photochem. Photobiol. 41 267-276
(1985).
66. R. Santus and J. P. Reyfimann, Photosensitization of membrane components, Biochimie 68, 843-848
(1986).
67. T. M. A. R. Dubbelman, Porphyrin-photosensitized modification of subcellular structures, in: Pho-
todynamic Therapy of Tumors and Other Diseases (G. Jori and C. Perria, eds.), pp. 93-99, Libreria
Progetto, Padova (1985).
68. H. D. Rabinowitch and 1. Fridovich, Superoxide radicals, superoxide dismutases and oxygen toxicity in
plants, Photochem. Photobiol. 37, 679-690 (1983).
69. M. L. Molero, M. J. Hazen, and J. C. Stockert, Photodynamic effects of berberine sulfate on the growth
rate of Allium cepa roots, J. Plant Physiol. 120,91-94 (1985).
70. J. P. Knox and A. D. Dodge, The photodynamic action of eosin, a singlet oxygen generator. The
inhibition of photosynthetic electron transport, Planta 164, 30-34 (1985).
71. F. J. Gommers, J. Bakker, and H. Wynberg, Dithiophenes as singlet oxygen sensitizers, Photochem.
Photobiol. 35, 615-619 (1982).
72. D. D. Morgan and D. Warshawsky, The photodynamic immobilization of Artemia salina nauplii by
polycyclic aromatic hydrocarbons and its relationship to carcinogenic activity, Photochem. Photobio!. 25,
39-46 (1977).
73. J. R. Heitz and K. R. Downum (eds.), Light-Activated Pesticides, ACS Symp. Ser. 339, American
Chemical Society, Washington, D.C. (1987).
74. A. Shlosberg and M. N. Egyed, Photosensitization in ducklings induced by seeds of Cymopterus watsonii
and C. iongpipes, Avian Diseases 22, 576-583 (1978).
75. G. W. Ivie, Chemical and biochemical aspects of photosensitization in livestock and pOUltry, J. Natl.
Cancer Inst. 69, 259-262 (1982).
76. J. H. Epstein, Chemical phototoxicity in humans, J. Nati. Cancer Inst. 69,265-268 (1982).
77. K. L. Gabriel, P. D. Forbes, and R. E. Davies, Phototoxicity, J. Toxico!., Cutaneous Ocul. Toxicol. 4,
185-191 (1984).
78. L. C. Harber, 1. E. Kochevar, and A. R. Shalita, Mechanisms of photosensitization to drugs in humans,
in: The Science of Photomedicine (1. D. Regan and J. A. Parrish, eds.), pp. 323-347, Plenum Press, New
York (1982).
79. D. P. Valenzeno and J. P. Pooler, Phototoxicity, the neglected factor, J. Am. Med. Assoc. 242,453-454
(1979).
80. J. T. Hindmarsh, The porphyrias: Recent advances, Clin. Chern. (Winston-Salem, N.C.) 32,1255-1263
(1986).
Photosensitization 109

81. M. M. Mathews-Roth, Beta-carotene therapy for erythropoietic protoporphyria and other photosensitivity
diseases, Biochimie 68, 875-884 (1986).
82. S. J. Galitzer and F. W. Oehme, Photosensitization: A literature review, Vet. Sci. Commun. 2,217-230
(1978).
83. A. E. Johnson, Toxicologic aspects of photosensitization in livestock, J. Natl. Cancer [nst. 69, 253-258
(1982).
84. G. R. Ruth, S. Schwartz, and B. Stephenson, Bovine protoporphyria: The fIrst nonhuman model of this
hereditary photosensitizing disease, Science 198, 199-202 (1977).
85. A. C. Giese, Hypericism, Photochem. Photobiol. Revs. 5, 229-255 (1980).
86. I. A. Magnus (ed.), Dermatological Photobiology: Clinical and Experimental Aspects, Blackwell, Oxford
(1976).
87. H. Ippen, Photodermatoses, in: The Science of Photomedicine (1. D. Regan and J. A. Parrish, eds.), pp.
349-394, Plenum Press, New York (1982).
88. F. Urbach, Ultraviolet radiation and skin cancer, in: Topics in Photomedicine (K. C. Smith, ed.), pp. 39-
142, Plenum Press, New York (1984).
89. P. D. Forbes, R. E. Davies, F. Urbach, and C. Cole, Photocarcinogenesis, J. Toxicol. Cutaneous Ocul.
Toxico!. 4, 219-236 (1985).
90. L. Santamaria, A. Arnaboldi, L. Andreoni, A. Bianchi, and P. Daffara, Photocarcinogenesis by methox-
ypsoralen, neutral red, proflavine and long UV radiation, Cancer Detect. Prevo 8, 447-454 (1985).
91. J. D. Spikes, The historical development of ideas on applications of photosensitized reactions in the health
sciences, in: Primary Photo-Processes in Biology and Medicine (R. V. Bensasson, G. Jori, E. J. Land,
and T. G. Truscott, eds.), pp. 209-227, Plenum Press, New York (1985).
92. T. R. C. Sisson and T. P. Vogel, Phototherapy of hyperbilirubinemia, in: The Science of Photomedicine
(J. D. Regan and J. A. Parrish, eds.), pp. 477-509, Plenum Press, New York (1982).
93. J. A. Parrish, R. S. Stem, M. A. Pathak, and T. B. Fitzpatrick, Photochemotherapy of skin diseases, in:
The Science of Photomedicine (J. D. Regan and 1. A. Parrish, eds.), pp. 595-623, Plenum Press, New
York (1982).
94. D. Kessel (ed.), Methods in Porphyrin Photosensitization, Plenum Press, New York (1985).
95. T. J. Dougherty, Photosensitizers: Therapy and detection of malignant tumors, Photochem. Photobiol. 45,
879-889 (1987).
96. A. R. Oseroff, G. Ara, D. Ohuoha, J. Aprille, J. C. Bommer, M. L. Yarmush, J. Foley, and L. Cincotta,
Strategies for selective cancer photochemotherapy: Antibody-targeted and selective carcinoma cell pho-
tolysis, Photochem. Photobiol. 46, 83-96 (1987).
97. J. R. Heitz, Xanthene dyes as pesticides, in: Insecticide Mode of Action (1. R. Coates, ed.), pp. 429-457,
Academic Press, New York, (1982).
98. J. R. Robinson, Photodynamic insecticides: A review of studies on photosensitizing dyes as insect control
agents, their practical application, hazards and residues, Residue Rev. 88,69-100 (1983).
99. J. T. Amason, G. H. N. Towers, B. J. R. Philogene, andJ. D. H. Lambert, Theroleofnaturalphotosen-
sitizers in plant resistance to insects, ACS Symp. Ser. 208 (Plant Resist. Insects), 139-151 (1983).
100. K. R. Downum, Photoactivated biocides from higher plants, ACS Symp. Ser. 296 (Nat. Resist. Plants Pests),
197-205 (1986).
101. J. T. Amason, B. J. R. Philogene, P. Morand, J. C. Scaiano, N. Werstiuk, and J. Lam, Thiophenes and
acetylenes: Phototoxic agents to herbivorous and blood-feeding insects, ACS Symp. Ser. 339 (Light-
Activated Pesticides), 255-264 (1987).
102. J. Kagan, P. Szczepanski, V. Bindokas, W. D. Wulff, and J. S. McCallum, Delayed phototoxic effects of
8-methoxypsoralen, khellin and sphondin in Aedes aegypti, J. Chem. Ecol. 12, 899-914 (1986).
103. C. A. Rebeiz, A. Montazer-Zouhoor, J. M. Mayasich, B. C. Tripathy, S. M. Wu, and C. C. Rebeiz,
Photodynamic herbicides and chlorophyll biosynthesis modulators, ACS Symp. Ser. 339 (Light-Activated
Pesticides), 295-328 (1987).
104. G. A. Rosenthal and D. H. Janzen (eds.), Herbivores: Their Interaction with Secondary Plant Metabo-
lites, Academic Press, New York (1979).
105. D. Ashkenzay, Y. Kashman, A. Nyska, and J. Friedman, Furocoumarlns in shoots of Pituranthos
triradiatus (Umbelliferae) as protectants against grazing by Hyrax (Procaviidea: Procavia capensis syr-
iaca), J. Chem. Ecol. 11,231-239 (1985).
110 Chapter 3

106. G. H. N. Towers, Significance of phototoxic phytochemicals in insect herbivory, J. Chern. Ecol. 12,813-
821 (1986).
107. K. R. Downum and E. Rodriguez, Toxicological action and ecological importance of plant photosen-
sitizers, J. Chern. Eco!. 12, 823-834, (1986).
108. D. E. Champagne, J. T. Amason, B. J. R. Philogene, P. Morand, and J. Lam, Light-mediated al-
lelochemical effects of naturally occurring polyacetylenes from Asteraceae on herbivorous insects, J.
Chern. Ecol. 12, 835-858 (1986).
109. M. R. Berenbaum, Charge of the light brigade: phototoxicity as a defense against insects, ACS Symp. Ser.
339 (Light-Activated Pesticides), 206-216 (1987).
110. 1. B. Harborne (ed.), Biochemical Aspects of Plant and Animal Coevolution, Academic Press, New York
(1978).
111. M. Berenbaum, Coumarins and caterpillars: A case for coevolution, Evolution 37, 163-179 (1983).
112. G. W. Ivie, Biological actions and metabolic transformations of furanocoumarins, ACS Syrnp. Ser. 339
(Light-Activated Pesticides), 217-230 (1987).
113. D. L. Bull, G. W. Ivie, R. C. Beier, and N. W. Pryor, In vitro metabolism of a linear furanocoumarin (8-
methoxypsoralen, xanthotoxin) by mixed-function oxidases of larvae of black swallowtail butterfly and
fall armyworm, J. Chern. Eco!. 12, 885-892 (1986).
114. M. R. Berenbaum and P. Feeny, Toxicity of angular furanocoumarins to swallowtails: Escalation in the
coevolutionary arms race, Science 212, 927-929 (1981).
115. R. A. Larson, Insect defenses against phototoxic plant chemicals, J. Chern. Ecol. 12, 859-870 (1986).
 4
UV Radiation Effects
DNA Repair and Mutagenesis

4.1. Repair of Photochemical Damage to DNA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. III

4.1.1. Introduction ................................................................ III
4.1.2. Survival Curves ............................................................. 114
4.1.3. Recovery Phenomena. . . . . . . . . . . . . ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 115
4.1.3.1. Liquid-Holding Recovery .............................................. 115
4.1.3.2. Minimal Medium Recovery ............................................ 116
4.1.3.3. Medium-Dependent Resistance .......................................... 116
4.1.4. Inducible Responses. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 117
4.1.4.1. SOS System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 117
4.1.4.2. Adaptive Response ................................................... 118
4.1.4.3. Heat-Shock System ................................................... 119
4.1.4.4. Oxidation System .................................................... 119
4.1.5. Mechanisms of DNA Repair ................................................... 119
4.1.5.1. Photoreactivation .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 119
4.1.5.2. Excision Repair ...................................................... 120
4.1.5.3. Postreplication Repair ................................................. 123
4.1.5.4. Conclusions ......................................................... 126
4.2. UV Radiation Mutagenesis .......................................................... 126
4.2.1. Introduction ................................................................ 126
4.2.1.1. DNA Replication. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 127
4.2.1.2. DNA Repair. . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 127
4.2.1.3. Types of Mutations ................................................... 127
4.2.2. Mutation Assays. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 127
4.2.3. Kinetics of Induction ......................................................... 128
4.2.4. Genetic Control of Mutagenesis ................................................ 129
4.3. Spontaneous Mutagenesis ........................................................... 130
4.3.1. DNA Replication ............................................................ 130
4.3.2. DNA Repair ................................................................ 130
4.4. References ....................................................................... 131

4.1. REPAIR OF PHOTOCHEMICAL DAMAGE TO DNA

4.1.1. Introduction
The first indication that cells might have the capacity to recover from radiation damage
was the observation that minor modifications in the postirradiation treatment of the cells

Kendric C. Smith • Department of Radiation Oncology, Stanford University School of Medi-

cine, Stanford, California 94305.
111
112 Chapter 4

(e.g., growth media, temperature, etc.) had a marked effect upon the ultimate viability of
irradiated cells. Thus in 1937, Hollaender and Claus found that higher survival levels of
UV -irradiated fungal spores could be obtained if they were allowed to remain in water or
salt solution for a period of time before assaying for survival. Roberts and Aldous(l)
extended these observations by showing that the shapes of the UV radiation survival
curves for Escherichia coli could be changed drastically by holding the irradiated cells in
media devoid of an energy source for various times before plating on nutrient agar to assay
for survival (Fig. 4-1). The molecular basis of this phenomenon, known as liquid-holding
recovery, will be discussed along with other "recovery phenomena" in Section 4.1.3.
The isolation of a radiation-sensitive mutant of E. coli (B s _I ) by Ruth Hill in 1958
inaugurated the era of the genetic approach to the study of DNA repair. Genetic studies
have been responsible for most of the discoveries about the molecular basis of DNA
repair. For example, one can measure the initial yield of a particular type of radiation-
induced lesion in the DNA of a wild-type strain (i.e., radiation-resistant) and a mutant
strain (i. e., radiation-sensitive), and then follow the kinetics and extent of repair of that
lesion, and thereby determine if the mutation under investigation has an effect on the
repair of that type of DNA lesion. Then by constructing strains that carry two mutations
that affect DNA repair, one can determine through survival and DNA repair studies
whether the two mutations act in the same or in different biochemical pathways of repair.

10

~ 5
-0
>
.~
::l
if)

Fig. 4-1. UV radiation survival curves of E. coli B. After

irradiation the cells were suspended in a liquid medium
devoid of an energy source for the times indicated before
being plated on nutrient agar. It is evident that both the
o slopes and the shapes of the survival curves can be altered
by the postirradiation treatment of the cells. These data,
0.1 "----:'::---~--~----'
from 1949, provided evidence that cells could recover from
UV fluence (relative units) radiation damage. (Modified from Ref. I.)
UV Radiation Effects 113

Wild Type ~
,. I I I
• 20 40
10-1

'"<:
.;;
.~

'""
.!!l
Qj
(J

'0
<:
a
Fig. 4-2. UV radiation survival curves of DNA re- nc 10- 3
pair-deficient mutants of E. coli K-I2. The mutations It
are defined in Table 4-1. Note that the double mutant
is much more radiation-sensitive than either of the
uvrA6 recAI3
single mutants, suggesting that the uvrA gene and the
recA gene act in different pathways of DNA repair. It 10-4
is now known that the uvrA mutation blocks excision
repair, and the recA mutation blocks postreplication
repair (Section 4. I .5) and inducible processes (Sec-
tion 4.1.4). [Modifed from P. Howard-Flanders and o 5 10
R. P. Boyce, Radiat. Res. Suppl. 6, 156 (1966).] UV fluence IJ m· 2 )

For example, since the uvrA recA double mutant is much more UV radiation-sensitive
than are either of the singly mutant strains (Fig. 4-2), it suggests that the uvrA and recA
genes function in separate pathways of DNA repair. As will be seen later (Section 4.1.5),
a uvrA mutation blocks excision repair, and a recA mutation blocks postreplication repair
and inducible repair functions (Section 4.1.4).
The discovery of excision repair (Section 4.1.5.2) by Howard-Flanders and cowork-
ers and by Setlow and coworkers came from the observation that, after UV irradiation, the
wild-type strain broke down some of its DNA into acid-soluble material in a radiation
dose-dependent manner, but the radiation-sensitive strain did not. It did not seem logical
at first that DNA degradation could be beneficial to survival, but subsequently it was
shown that this DNA degradation was not random. UV radiation-induced lesions (i.e.,
thymine dimers) were selectively cut out of the DNA in the wild-type strain, but not in the
sensitive strain. It is this ability to make comparative measurements between the wild-type
strain and different radiation-sensitive strains that has allowed the rapid progress in our
understanding of DNA repair in E. coli. The paucity of radiation-sensitizing mutations in
mammalian cells has slowed progress in understanding DNA repair in mammalian
cells. (2-4) Many mutations that affect radiation sensitivity are known for yeast, (5) but the
understanding of their biochemical basis is far behind that for E. coli. This explains why
the bulk of the information for this chapter comes from work on E. coli.
A textbook on DNA repair,(2) the proceedings of an international meeting on this
subject, (6) and a laboratory manual of research procedures for studying DNA repair(7)
have recently been published.
114 Chapter 4

4.1.2. Survival Curves

A radiation survival curve is a plot of the logarithm of the surviving fraction of a
population of cells vs. the dose of radiation (a semilogarithmic plot). For ionizing radia-
tion (e.g., X rays) the dose is generally the absorbed dose (100 rad = 1 gray), but for
nonionizing radiation (e.g., UV radiation) it is generally the incident dose (fluence)
(joules/m2). This difference (i.e., absorbed vs. incident dose) simply reflects the types of
dosimetry in common use for these two types of radiation.
Before it was discovered that cells could recover from radiation-induced damage,
survival curves were interpreted in terms of target theory, i.e., in terms of physical targets
and inactivating hits within the targets. (8,9) If the survival curve is a straight line on a
semilogarithmic plot, it suggests that the cells were killed when one target within the cell
was hit once by the radiation. More complicated survival curves (e.g., shouldered curves
such as those labeled 2.5 and 4.5 in Fig. 4-1) were interpreted as being due to the presence
of mUltiple targets and/or the requirement for multiple hits within the targets.
In a form similar to the Lambert-Bouguer-Beer law (Section 1.5.1), one can ex-
press the simplest case of cellular inactivation (i.e., a straight line on a semilogarithmic
plot) by the relationship N / No = e - aD, where No is the number of live cells initially
present and N is the number of cells surviving after a dose D. The constant IT is called the
inactivation cross-section and is the product of the quantum yield (<1» for inactivation
(i.e., <1> = number of cells killed/number of photons absorbed) and the absorption cross-
section. The expression ITD is the average number of lethal hits per target. [The determin-
ation of IT over a range of wavelengths would constitute an action spectrum (Section
1.5.5) for inactivation.] The inactivation cross-section can be determined from the slope
of the survival curve and is the reciprocal of the dose to give 37% survival (N.B.: e- I =
0.37). At this dose (mean lethal dose), designated as D37 (dose) or F37 (fluence), there is
an average of one hit per sensitive target. A similar term for shouldered survival curves is
Do and is calculated from the linear portion of the curve. A description of the difference
between Do and D37 and the definition of other parameters that can be derived from
survival curves are shown in Fig. 4-3.
Most survival curves are not straight lines but come in a variety of shapes (e.g., Fig.
4-1). The mathematics of the curve fitting becomes much more complicated for such
curves, (8,9) and attempts to explain the mechanism(s) of cell lethality on the basis of such
curve fitting become even more difficult. A recent development (1972) in curve fitting is
the linear-quadratic model for inactivation. Many (but not all!) X-ray survival curves for
mammalian cells can be fit by the expression N / No = e - (rill + f3D2). This suggests that
part of the targets are inactivated by single-hit kinetics (i.e., a linear function of dose D)
and some by two-hit kinetics (i.e., dose squared, D2). Alpha and beta are constants.
Another interpretation of the diverse shapes of survival curves has tried to incorpo-
rate the concept of DNA repair. (e.g., 10) The simplest interpretation is that a shoulder on a
survival curve indicates the capacity to repair and a linear survival curve indicates that the
cells have no repair capacity. At high doses when a shouldered survival curve becomes
exponential (i.e., linear on a semilogarithmic plot), it is presumed that either the repair
systems themselves have been inactivated or they have been "swamped" by the large
amount of damage produced by such high doses. These interpretations are oversimplified,
however, since a linear survival curve does not necessarily mean that cells have no DNA
repair capacity.
UV Radiation Effects 115

Fig. 4-3. Examples of survival curves and their associ-

ated parameters. Straight-line survival curves (a and b)
can be characterized by a single parameter, the D37
(i.e., the dose to yield a survival of 37%). For a shoul-
dered survival curve (c), three parameters apply. The
extrapolation number (II) is one measure of the width of z
o
the shoulder, and is the point where an extrapolation of j:::
u
the linear portion of the survival curve crosses the ordi- «
0::
nate. Dq (quasi-threshold dose) is also a measure of the u..
width of the shoulder, and is the dose at which the c.:J
Z
extrapolation line, described above, intersects the 100% >
survival line. The Do is the same concept as the D 37 , but >
0::
it is determined on the linear portion of the survival ::::J
til
curve. While Do = D37 on an exponential survival
curve, these two parameters differ greatly on a shoul-
dered survival curve. Note that the Do for curve c is
equal to the D37 of curve b, i.e., the lines are parallel.
These parameters were largely developed for studies
using ionizing radiation and the term dose (i.e., ab-
sorbed dose). Most frequently for UV radiation studies, 10-3~~~--~~~~~~~~
fluence (incident dose) is used rather than dose, so the o 40 120
comparable parameters would be F o, F 37 , and Fw RADIATION DOSE

Such curve fitting has not greatly advanced our understanding of the molecular basis
of radiation biology. However, "mathematical analysis does compel one, at least, to
make explicit the assumptions and ideas used to describe the phenomena. "(11) Since such
mathematical treatments have played such a large part in the history of radiation biology,
a student of radiation biology (UV or ionizing radiation) should become knowledgeable in
this area, but should also be aware of its limitations.

4.1.3. Recovery Phenomena

4.1.3 .1. Liquid-Holding Recovery
Recovery phenomena have been an important part of radiation biology.o 2 ) The
recovery phenomenon now known as liquid-holding recovery (LHR) provided the first
evidence that cells could recover from radiation-induced DNA damage (Section 4.1.1).
The initial observation was that when E. coli cells were held in the absence of nutrients for
a period of time after UV irradiation, not only did the shape of their survival curve change
with time, but the cells also exhibited a much higher survival than if they were plated on
growth medium immediately after UV irradiation (Fig. 4-1).
Genetic studies have shown that there are two major requirements for observing
LHR. (13) First, the cell must be proficient in nucleotide excision repair (Section 4.1.5.2),
and second, it must be at least partially deficient in postreplication repair (Section
4.1.5.3). In fact, cells completely deficient in postreplication repair (i.e., recA cells)
show the greatest amount of LHR.
In a fully repair-proficient cell (i.e., wild-type), if a DNA lesion is not repaired by
the excision repair process it can usually be repaired by the postreplication repair process.
However, nucleotide excision repair can occur in buffer that is devoid of nutrients, but
116 Chapter 4

postreplication repair requires the presence of complete growth medium (i.e., mac-
romolecular synthesis is required). Therefore, by holding cells in buffer without essential
nutrients, both DNA replir;;ation and postreplication repair are prevented, and only excis-
ion repair can proceed. Such liquid holding is beneficial to the survival of cells deficient in
postreplication repair because the more lesions that are repaired by the excision repair
process before the cells are placed in growth medium where DNA replication can proceed,
the fewer lesions that will be left to cause the formation of DNA daughter strand gaps,
which are the substrates for a repair process in which the cells are deficient. LHR has also
been studied in yeast.
In mammalian cells, a similar phenomenon after X irradiation is called plateau phase
recovery, or the repair of potentially lethal damage. (14) In this process, cells that have
entered stationary phase (i.e., the cells are confluent and nutrients in the growth medium
are exhausted) show a higher survival after X irradiation if they are kept in stationary
phase for several hours rather than being immediately subcultured and plated in fresh
medium for the determination of survival.

4.1.3.2. Minimal Medium Recovery

UV -irradiated, excision repair-deficient cells of E. coli exhibit minimal medium
recovery (MMR), although a better name for this process would be rich medium lethality
(see below). The observation was that excision repair-deficient cells show a higher sur-
vival after UV irradiation if they are plated on minimal medium rather than on rich
medium, regardless of the type of medium in which they were grown before irradiation.
Since this process was described in excision repair-deficient cells, the presumption was
that some aspect of postreplication repair was inhibited by components of the rich growth
medium. This has proven to be the case. The major processes of postreplication repair
(i.e., the repair of DNA daughter strand gaps and the repair of DNA double-strand breaks)
are partially inhibited by rich growth medium. This inhibitory effect of rich medium can
be reproduced by just adding the common amino acids to minimal growth medium. A
mutation, mmrA, has been isolated that blocks this effect of rich medium on survival and
repair, but the mmrA gene product is not known,05) nor is the mechanism known by
which excess amino acids inhibit postreplication repair. (16)

4.1.3.3 . Medium-Dependent Resistance

E. coli cells grown to logarithmic phase in rich medium and plated on rich medium
are about three times (dose modification jactor, i.e., ratio of doses to produce equal
survival) more resistant to X radiation and about 1.5 times more resistant to UV radiation
than are cells grown in and plated on minimal medium, i.e., the cells show medium
dependent resistance (MDR). (17) Therefore, MDR reflects a selective effect on the repair
of X-ray-induced DNA damage. Cells that exhibit MDR also show a much greater
capacity for postirradiation protein synthesis and for the repair of DNA double-strand
breaks. MDR is an inducible process that is absent in recA, lexA, and recBC cells, and is
partially inhibited by other mutations. (18)
Such recovery phenomena (especially LHR) were particularly useful for gaining
insights about DNA repair in the period before mutations affecting DNA repair were
UV Radiation Effects 117

available. Now in combination with genetic studies, certain of these recovery phenomena
offer the possibility of studying the physiological control of DNA repair (e.g., MMR,
MDR).

4.1.4. Inducible Responses

Since the isolation ofthe first radiation-sensitive mutant of E. coli in 1958, numerous
mutations that affect radiation sensitivity have been isolated, and the biochemical defi-
ciency of many of these mutants has been determined (Table 4-1). Through the charac-
terization and comparison of these mutants have come most of the discoveries of the
multiple processes of DNA repair.
Another important discovery from such genetic studies was that some of the repair
systems are constitutive and some are inducible. Furthermore, certain of the constitutive
repair proteins (e.g., UvrA and UvrB proteins) can be induced to higher concentrations in
irradiated cells.

4.1.4.1. SOS System(2,19-21)

In 1974, Radman postulated that DNA damage, or the physiological consequence of

such damage (e.g., the temporary blockage of DNA synthesis), causes the induction of
repair processes in E. coli. He called this phenomenon "SOS" repair (N.B.: SOS is the
international distress signal). (reviewed in Ref. 2) Since UV radiation mutagenesis was only
observed under conditions of functional SOS repair, it was presumed that this mutagenesis

TABLE 4·1. Some of the More Important Genes of Escherichia coli

That Affect DNA Repair
Map
Gene position
symbol (miu)a Gene product and/ or function b
ada 48 Methyltransferase, adaptive response
alkA 43 3-Methyladenine-DNA glycosylase II (inducible)
lexA 92 "50S" gene repressor, inducible repair
/ig 52 DNA ligase, all "dark" repair
phr 16 DNA photolyase, photoreactivation
poLA 87 DNA polymerase I, ER, PRR
polB 2 DNA polymerase II, unknown
polC 4 DNA polymerase III, DNA replication
recA 58 50S repair, recombination, ER, PRR
reeB,C,D 61 Exonuclease V, recombination, ER, PRR
reeF 83 Recombination, ER, PRR
ssb 92 Single-strand DNA-binding protein
tag 72 3-Methyladenine DNA-glycosylase I (noninducible)
umuC,D 26 UV radiation mutagenesis, PRR
uvrA,B,C 92,18,42 Excision nuclease, ER
uvrD 85 Helicase II, ER, PRR
aFrom B. J. Bachmann, Linkage map of Escherichia coli K-12, ed. 7, Microbiol. Rev., 47, 180-230 (1983).
hER = nucleotide excision repair; PRR = postreplication repair.
118 Chapter 4

was due to error-prone DNA repair. The SOS phenomenon is under the control of the recA
and lexA genes. DNA-damaging agents other than UV radiation can also induce the SOS
system, e.g., ionizing radiation, and chemicals such as mitomycin C. Since the SOS
system is inducible, inhibitors of RNA and protein synthesis prevent its expression.
One candidate for the "signal" that causes the SOS system to be induced is the
sudden increase in the amount of single-stranded DNA that occurs in a cell due to the
distortion of the DNA helix by lesions in the DNA and/or due to the stalling of DNA
replication at the site of a lesion. In support of this idea, single-stranded DNA activates
the protease function of the RecA protein (product of the recA gene) in vitro. The LexA
protein (product of the lexA gene) is the repressor of the SOS genes. The activated RecA
protein catalyzes the cleavage of the LexA protein, and when the concentration of LexA
protein in the cell falls to a low level the SOS genes become depressed, including the recA
and lexA genes themselves. Other inducible SOS genes include the uvrA, uvrB, and uvrD
genes, which are involved in excision repair; the recN and ruv genes, which are involved
in recombination; a number of din (damage-inducible) genes, some of which are still
uncharacterized; and the umuC and umuD genes, which playa key role in UV radiation
mutagenesis but playa lesser role in X-ray mutagenesis (Section 4.2.4). (N.B.: Recent
evidence indicates that the activated RecA protein also catalyzes the cleavage of the
UmuD protein.) Thus, a large set of genes that perform diverse functions are induced in
the SOS process to optimize the cell's chances for survival.
It is still not totally resolved as to whether or not mammalian cells show an SOS
response. (e.g .• 22,23)

4.1.4.2. Adaptive Response(2,20,21)

Other inducible DNA repair processes are known that are independent of the SOS
system. The best characterized of these is the adaptive response, the induced resistance to
alkylating agents. In 1977, Samson and Cairns reported that cells of E. coli that had been
treated with low levels of the methylating agent N-methyl-N' -nitro-N-nitrosoguanidine
(MNNG) developed resistance to the lethal and mutagenic effects of a subsequent chal-
lenge with a higher dose of MNNG. Two types of processes are induced in the adaptive
response. One of these is related to base excision repair, and one is an entirely new
mechanism involving an enzyme called a methyltransferase (see below).
The product of the alkA gene is 3-methyladenine-DNA glycosylase II, and its con-
centration increases approximately 20-fold as a consequence of the adaptive response.
This enzyme has a broad range of activities for removing a number of methylated purines
and pyrimidines from DNA, as the fIrst step in base excision repair (see Section 4.1.5.2).
The product of the tag gene is 3-methyladenine-DNA glycosylase I. It is not inducible and
shows a high specificity for removing 3-methyladenine from DNA.
The product of the ada gene not only plays a role in regulating the adaptive response
by a mechanism not yet understood but also participates in the repair process as a
methyltransferase. Methyltransferase is a new kind of enzyme, a suicide enzyme. The
classical definition of an enzyme (an organic catalyst) is that it speeds up the reaction but
is not consumed in the reaction. The methyltransferases do not follow this rule. They
transfer a methyl group, most frequently from 06- methylguanine, to a specifIc cysteine
residue in their structure, and thereby inactivate themselves.
UV Radiation Effects 119

4.1.4.3. Heat-Shock System(20)

When cells are shifted to lethal temperatures for a sublethal period of time (i.e., they
are given a heat shock), they induce a number of heat-shock proteins (detectable by gel
electrophoresis) and become resistant to a subsequent exposure to high temperatures. The
definition of high temperature differs for different species, but for mammalian cells it is
around 43°C and for E. coli it is around 50°C. The gene that controls this response in E.
coli is the htpR gene. The molecular basis of this heat-shock response (i.e., is it due to
damage to DNA or to membranes, or both) is under active investigation in many types of
cells. This is especially true for human cells, since hyperthermia is now being used in the
treatment of certain types of human tumors. (24)
There is some overlap in these three inducible systems, e.g., MNNG can induce
some of the SOS proteins, and SOS-inducing agents (e.g., UV radiation, nalidixic acid,
etc.) can induce some of the heat-shock proteins.

4.1.4.4. Oxidation System

Evidence is accumulating for the existence of a fourth inducible system, which is
specific for oxidative damage to DNA, i.e., the prior treatment of cells with a low
concentration of H202 induces resistance to subsequent exposure to H20 2 but not to
MNNG. Thus, this response appears to differ in specificity from the adaptive response for
alkylating agents(25) but shows some overlap with the heat-shock response.(26) A similar
overlap exists between H202 and near-UV radiation. (27) This result is consistent with the
fact that many of the photochemical reactions involving near-UV radiation also require
oxygen (see Chapter 3).

4.1.5. Mechanisms of DNA Repair

4.1.5.1. Photoreactivation
Under certain conditions, the damaged part of a DNA molecule can be restored to its
functional state in situ (i.e., without breaking the sugar-phosphate backbone of the DNA).
This restoration can be either by chemical or enzymatic means. Chemical restoration can
result from the spontaneous "decay" of the damage to the original structure, e.g., the
dehydration of pyrimidine photohydrates (Section 2.3.1.3) or by the recombination of
radicals to yield a restored molecule (e.g., R' + H' ~ RH).
Photoreactivation, the enzymatic splitting of cyc1obutane-type pyrimidine dimers in
situ, is mediated by exposure to visible light, and was the first repair system to be
elucidated at the molecular level. (28) In 1949, Kelner observed that the survival of UV-
irradiated bacteria could be greatly enhanced if the cells were subsequently exposed to an
intense source of blue light. In 1958, Rupert and coworkers demonstrated the existence of
a photoreactivating enzyme (DNA photolyase) and established its basic properties. In the
dark, the enzyme binds tightly to a cyc1obutane-type pyrimidine dimer in UV -irradiated
DNA to form an enzyme-substrate complex. The absorption of light between 300 and 450
nm activates this complex, the substrate is converted from a cyc1obutane-type pyrimidine
dimer to two monomeric pyrimidines, and the enzyme is released. This reaction is shown
schematically in Fig. 4-4.
120 Chapter 4

Pyrimidine Dimer

1 J I I I I I I

Enzyme-DNA Complex

( I
, I I I I I I I I I I , ( I I I I

Absorption of Light and Repair

h"~
I
I
I
I
I
I
ffif!{t ~l-r"""1
I I I I I I I I I I I I 11
'I"T""'1rr

Release of Enzyme

~
Fig. 4-4. A general model for photoreactivation. The photoreac-
tivating enzyme (DNA photolyase) binds to a cyclobutane-type
pyrimidine dimer in UV-irradiated DNA to form an enzyme-
substrate complex. The absorption of light between 300 and 450
I I I I
nm activates the complex, the pyrimidine dimer is converted to
I I I I monomeric pyrimidines, and the enzyme is released.

As much as 90% of the lethal damage induced in bacteria by low fluences of UV

radiation at 254 nm can be photoreactivated, thus indicating the importance of pyrimidine
dimers as lethal lesions. The gene for the photoreactivating enzyme has been mapped in
E. COli(29,30) and shown to be a flavoprotein. (31)
Photoreactivating enzymes have been found in a wide range of species from the
simplest living cells, the mycoplasmas, to the skin and white blood cells of man. (32)

4.1.5.2. Excision Repair

Two types of excision repair are known, nucleotide excision repair and base excision
repair. The former system constituted the first discovery (in 1964) of a mechanism of dark
repair, i.e., "dark", to distinguish it from photoreactivation that is mediated by visible
light (Section 4.1.5.1). Base excision repair is a more recent discovery.
In nucleotide excision repair, (2) the damaged section of a DNA strand is removed
UV Radiation Effects 121

EXCISION
(v,rASe nuclease)

5:1!11!I!·II!I!!~~:

REPAIR REPLICATION
(DNA polymerase I)
Fig. 4-5. A model for the major pathway of nucleotide-excision repair in E.
coli. The UvrABC nuclease recognizes the lesion, shown here as a cyclobutyl-
pyrimidine dimer, and cuts on both sides of the lesion, producing a gapped 1IIIIIIiliiiiliiiilITTITT

structure that is about 12 nucleotides long. Repair replication (heavy line) fills
this gap, using the opposite strand of DNA as the template. Finally, the break
in the repaired DNA strand is sealed by DNA ligase. See the text for more REJOINING
(DNA ligase)
details. [Modified from A. Sancar and W. D. Rupp, Cell 33, 249-260
(1983).] lIillI"!!i1";;!! II!II!

and replaced with undamaged nucleotides to restore the normal function of the DNA. A
schematic representation of the probable steps involved in the major pathway of this
process is shown in Fig. 4-5.
The first step in nucleotide excision repair is the recognition of the damage in the
DNA. The UvrABC nuclease appears to recognize the distortion in the DNA helix due to
the presence of the damage rather than recognizing the damage itself. This conclusion is
based on the observation that this excision repair system functions on both radiation and
chemical damage whose molecular structures are very different. This complex enzyme,
coded for by the uvrA, uvrB, and uvrC genes (Table 4-1), produces breaks in the DNA
chain at the eighth phosphodiester bond on the 5' side and the fourth or fifth phos-
phodiester bond on the 3' side of a pyrimidine dimer. The uvrD gene product, helicase II,
then appears to cause the release of the UvrC protein (through some type of protein-
protein interaction rather than through its helicase activity), and DNA polymerase I
(product of the polA gene) then displaces the DNA fragment containing the lesion and fills
the gap using the DNA strand opposite the gap as the template. (33) When all of the
nucleotides have been replaced, DNA ligase (product of the fig gene) seals the two ends of
the chain as the final step in the repair process.
These results with the uvrD gene product were from in vitro studies. Studies in vivo,
however, give a somewhat different picture. Since a uvrD mutation sensitizes arecA
strain to UV radiation only very slightly, it suggests that the uvrD gene plays very little
role in recA-independent excision repair. However, a uvrD mutation causes a marked
sensitization of a uvrB strain, suggesting that the uvrD gene plays a significant role in
postreplication repair. This apparent contradiction between the in vivo and in vitro results
on the role of the uvrD protein in excision repair remains to be resolved.
Nucleotide excision repair has been divided into two pathways (Fig. 4-6). The major
pathway for the repair of excision gaps requires DNA polymerase I (polA), and repair can
proceed by the mechanism described in Fig. 4-5, even when the cells are in buffer. The
minor pathway (in terms of the number of lesions repaired) is dependent on a functional
recA gene and requires that the cells be in complete growth medium (see below).
Another feature that distinguishes these two pathways is the size of the repair
122 Chapter 4

UV·DAMAGED DNA

Fig. 4-6. The genetic and physiological

control of the different pathways of nu-
cleotide excision repair inE. coli. Most of
GROWTH GROWTH the excision gaps are repaired by the
MEDIUM MEDIUM growth medium-independent, palA-de-
INDEPENDENT DEPENDENT
pendent pathway by the mechanism
shown in Fig. 4-5. A small number of
lexA
gaps are repaired by the growth medium-
recA dependent, recA-dependent pathway.
pofA reeB,C The latter pathway appears to function
fig reeF only in the replicated portion of the chro-
polC mosome (see text and Fig. 4-7). The mu-
fig
tations are defmed in Table 4-1. Chloram-
'MPURE AGAR
{ CAP phenicol (CAP), dinitrophenol (DNP) ,
DNP and impure agar inhibit the growth medi-
um-dependent pathway. [Modified from
D. A. Youngs, E. Van der Schueren and
REPAIRED DNA K. C. Smith, 1. Bacterial. 117,717-725
(1974).]

"patches." Hanawalt and coworkers have shown that short patches (20-30 nucleotides
long) are produced in the polA-dependent process, while long patches (200-1500 nu-
cleo tides long) are produced in the reeA -dependent process. (34)
Based on recent knowledge about the multiple pathways of postreplication repair
(Section 4.1.5.3), a model for reeA-dependent nucleotide excision repair has been pro-
posed.(35) Consistent with this model, reeA-dependent excision repair only occurs in the
replicated portion of the chromosome where sister duplexes exist, and where intra-
chromosomal recombination can occur (Fig. 4-7); the repair of excision gaps that occur in
cells with unreplicated chromosomes is reeA-independent. The majority of this reeA-
dependent excision repair is reeF-dependent, and a small portion is reeB-dependent. The
reeF and reeB gene products are suggested to function in reeA-dependent excision repair
in a manner similar to their function in postreplication repair (see below), i.e., in the
portion of the chromosome replicated prior to UV irradiation, the RecF pathway repairs
excision gaps by a recombinational process that results in the formation of long repair
patches (Fig. 4-7C-E), and the RecB pathway repairs the DNA double-strand breaks that
arise at unrepaired gaps (Fig. 4-7F). This model for reeA-dependent excision repair will
become more clear after reading Section 4.1. 5.3 on postreplication repair.
A study of excision repair in mammalian cells has led to the observation of an
apparent correlation between carcinogenesis and the defective repair of DNA. (36) Fibro-
blasts from patients with the hereditary disease xeroderma pigmentosum exhibit much-
reduced levels of excision repair after UV irradiation. It has been suggested that this
deficiency in excision repair is related to the induction of fatal skin cancers in these
patients after exposure to sunlight. In general, excision repair appears to be more accurate
UV Radiation Effects 123

Fig. 4.7. The reeA-dependent repair

of excision gaps in UV -irradiated E.
coli. Lesions can be produced in both
the replicated and nonreplicated re-
gions of the chromosome (A, B), but
only the excision gaps produced in the
replicated region of the chromosome
(C) can be repaired by an intra-
chromosomal recombinational process
that is reeF-dependent (D). This pro-
cess leaves a gap in the homologous
sister duplex that can be filled by long-
patch repair replication (wavy line),
using the parental strand opposite the
gap as a template (E). If the daughter
l recA

strand opposite the excision gap is cut

(-/ /-; F), the resulting double- l recS
strand break is repaired primarily by a
REPAIR OF DOUBLE
recB-dependent process (see text for STRAND BREAKS
further discussion). (From Ref. 35.)

than postreplication repair. Therefore, if an individual is deficient in "error-free" repair,

then all DNA lesions must be repaired by the remaining "error-prone" system. Error-
prone DNA repair is the cause of UV radiation mutagenesis (Section 4.2.4), and muta-
tions are believed to be an early step in some mechanisms of carcinogenesis. The DNA
repair capacities of cells from patients with various genetic diseases that predispose them
to cancer have been reviewed, (2,3) as well as the immunological defects that are man-
ifested in these diseases. (4)
In base excision repair,o,37) altered bases (e.g., thymine glycol, 3-methyladenine,
uracil, hydroxymethyluracil, hypoxanthine, urea) are recognized by a class of enzymes
called DNA glycosylases, which are specific to varying degrees for particular types of
altered bases. These enzymes split off the altered base at the N-glycosylic bond that links
the base to the sugar in the DNA backbone. The site of the base loss is called an AP site
(for apurinic or apyrimidinic). These steps are shown schematically in Fig. 4-8. AP sites
are substrates for AP endonucleases, of which there are two types. One type cuts on the 5'
side and the other type cuts on the 3' side of an AP site. If the two AP endonucleases act in
concert, then in theory a gap can be formed in the DNA that is only one nucleotide long.
However, if after an initial cut by one of the AP endonucleases, an exonuclease degrades
the broken strand, the gaps could be much larger. These gaps are presumably repaired by
the same processes described above for the repair of the gaps formed during nucleotide
excision repair.

4.1.5.3. Postreplication Repair(2,38)

The first indication that excision repair was not the only mechanism by which cells
could repair radiation damage to their DNA in the dark was the observation that bacterial
124 Chapter 4

A B Fig. 4.8. Initial steps in base excision repair. An al-

3' T T T T T T T T tered base (dark circle in A) is recognized by a specific
o 0 0 0 0 0 0 0 enzyme called a DNA glycosylase. The base is removed

5' -
<2 ,
-
9 <2
- -
9
-
99
- -
from the DNA by hydrolyzing the N-glycosylic bond
that links the base to the sugar in the DNA backbone.
The site of the base loss (in B) is called an AP site (for apurinic or apyrimidinic). In subsequent steps (not
shown), an AP endonuclease cuts the DNA backbone at the AP site, and the resulting DNA gap is presumably
repaired by the same processes described for the repair of the gaps formed during nucleotide excision repair
(Section 4.1.5.2).

cells deficient both in excision repair (uvrA) and in genetic recombination (reeA) are much
more sensitive to killing by UV radiation than cells carrying either mutation alone (Fig.
4-2). These genetic studies suggested that the uvrA and reeA genes act in different
biochemical pathways of repair and that certain steps in genetic recombination must be
important in the repair of radiation damage. Postreplication repair was discovered in 1968
by Rupp and Howard-Flanders using excision repair-deficient cells.
The observations that led to the model for postreplication repair were that (1) the
DNA that is synthesized in E. coli immediately after UV irradiation has discontinuities
when assayed on alkaline sucrose gradients and (2) the mean length of the newly synthe-
sized DNA approximates the average distance between pyrimidine dimers in the parental
strand. With further incubation of the cells, these discontinuities (i.e., DNA daughter'
strand gaps) disappear, and the DNA approximates the molecular size of that from
unirradiated control cells. This process is shown schematically in Fig. 4-9.
This model is an oversimplification because it shows that the genetic transfers (Fig.
4-9c-d) result in lesion-free daughter strand DNA. However, in E. coli about half ofthese
genetic transfers result in the covalent joining of daughter strand DNA to parental DNA
that contains pyrimidine dimers. (38) Therefore, several cycles of DNA replication and
postreplication repair are needed before all of the dimers are "diluted" out of the daughter
DNA. The transfer of dimers from parental DNA to daughter DNA appears to be due to
the random resolution of the Holliday junction, an intermediate in recombination; if
resolved one way it yields dimer-free daughter strands, and if resolved in a second way it
transfers pyrimidine dimers to the daughter strands.(39) For mammalian cells, however,
very few pyrimidine dimers are detected in daughter strand DNA after postreplication
repair. (40) It is not clear whether this means that Holliday junctions are only resolved in
one way in mammalian cells (i.e., to yield dirner-free daughter strands) or that daughter
strand gaps are repaired in mammalian cells by a mechanism that differs from that found
in E. coli.
Studies on the genetic control of postreplication repair indicate that there are two
distinct processes; both require a functional reeA gene. (41 ,42) One process is for the repair
of DNA daughter strand gaps. About half of the DNA daughter strand gaps are repaired by
a process that is reeF-dependent, and about half are repaired by a process that is reeF- and
reeB-independent and is constitutive. (43) The genetic control of this latter pathway is not
established, although the umuC gene appears to play a small role, and may define an
error-prone (i.e., mutagenic; see Section 4.2.4) pathway of postreplication repair. The
poLA gene (especially the 5' ~ 3' exonuclease activity of DNA polymerase I) also plays a
role in this reeF-independent pathway for the repair of daughter strand gaps.
The second process of postreplication repair requires a functional reeB gene, and
repairs the DNA double-strand breaks that arise at unrepaired daughter strand gaps. The
UV Radiation Effects 125

(0)
.
<

Fig. 4-9. The postreplication repair of DNA daughter strand gaps in (b) ....---
• :10
-->-
UV-irradiated E. coli. (a) Dots indicate photochemical lesions produced
<
in the two strands of DNA. (b) DNA synthesis proceeds up to and then
skips past the lesions in the parental strands, leaving gaps in the daughter
strands. (c) Filling of the gaps in the daughter strands with material from (c)
the parental strands by a recombinational process that also requires a
functional recA gene. About half of the time, however, lesions are also
transferred from the parental strands to the daughter strands (not
shown).(39) (d) Gaps in the parental strands are repaired by repair rep- w w
lication. [Modified from K. C. Smith, Photophysiology 6, 209-278
"
(1971).]

reeB pathway for the repair of double-strand breaks is about as important to the survival of
UV-irradiated E. eoli as the reeF pathway for the repair of daughter strand gaps, as judged
by the fact that, in the uvrB background, a reeF or a reeB mutation sensitizes cells to
killing to about the same extent. These pathways of postreplication repair are shown
schematically in Fig. 4-10.
The recombination deficiency and radiation sensitivity of reeBC strains are sup-
pressed by an additional mutation in the sheB gene, which is the structural gene for
exonuclease I, a single-strand specific 3' ~ 5' DNA exonuclease. The presence of an
sheB mutation also restores the proficiency of reeBC cells to repair DNA double-strand
breaks, and this repair is dependent on the reeF and reel genes. Since the RecBCD
enzyme (Exo V) has a DNA helicase activity that requires blunt or nearly blunt ends of
DNA duplexes (i.e., it will not unwind DNA that has a long single-stranded tail), it
suggests that the double-strand breaks that are formed at daughter strand gaps are nor-
mally processed to blunt ends by Exo I (sheB) and Exo V (reeBCD) before being repaired
by the reeBCD-dependent recombination process. When daughter strand gaps containing
single-stranded tails are not degraded by Exo I and Exo V (i.e., in sheB reeBC cells), they
become substrates for a reeF-dependent recombination process. Therefore, depending on
the structure of a DNA double-strand break and the genetic background of the cell, a
double-strand break may be repaired by the RecBCD pathway (the primary pathway in a
wild-type cell) or by the RecF pathway, or both.

REPLICATION OF DAMAGED DNA

Fig. 4-10. Schematic diagram of the interaction of the multiple DAUGHTER DOUBLE
pathways of postreplication repair in E. coli. During DNA rep- STRAND STRAND
lication, noncoding DNA lesions are skipped, producing DNA GAPS BREAKS
daughter strand gaps (Fig. 4-9). These gaps are repaired by two
processes, one of which requires a functional recF gene; the one
marked with "?" is a pathway that functions in the absence of the recB
recF and recBC genes. If these gaps are not repaired, they can be
converted to DNA double-strand breaks, which are repaired pri-
marily by a process that requires a functional recB gene.
126 Chapter 4

The repair of DNA double-strand breaks in UV -irradiated E. coli is very complex

and poorly understood. Its complexity is best exemplified by the number of genes that
appear to control the repair of double-strand breaks. With the exception of the umuC gene,
practically all of the genes that have been implicated in postreplication repair are also
involved, to varying degrees, in the repair of double-strand breaks. While the reeF and
reel mutations appear to specifically affect the repair of double-strand breaks in UV-
irradiated uvrA recBC sbcB cells, mutations in reeA, recBC, recN, radB, ssb, uvrD, lexA,
and polA all produce a deficiency in the repair of double-strand breaks in sbeB + cells.
Presumably, some of these genes are preferentially involved in either the ReeF or the
RecBCD pathway for the repair of double-strand breaks, and some may be involved in
both pathways. A similar set of genes has also been observed to control the repair of X-
ray-induced DNA double-strand breaks; however, in this case there is a requirement for
the reeF and reel gene products in sbcB+ cells. This suggests that additional types of
double-strand breaks are formed by X irradiation.
The postreplicational formation and repair of DNA double-strand breaks after UV
irradiation has also been observed in cultured human cells. Therefore, this aspect of
postreplication repair appears to be similar for both E. coli and human cells.

4.1.5.4. Conclusions
One cannot help but be impressed with the multitude and sophistication of the repair
systems possessed by cells. A significant percentage of the energy of a cell is spent in
synthesizing enzymes to repair and maintain the integrity of the genetic information coded
in its DNA. There is evidence, however, to suggest that these repair systems have not
evolved just to repair damage produced in DNA by environmental agents, but rather they
appear to have a necessary function in the everyday life of a cell. This conclusion is based
on the observation that while certain mutants deficient in one DNA repair protein grow
normall y in the absence of radiation, certain mutants deficient in two repair proteins (e. g. ,
polA reeA) are not viable. Thus, normal growth processes appear to produce lesions in
DNA that must be repaired in order to maintain viability. Such normal metabolic damage
to DNA has also been shown to contribute to spontaneous mutagenesis (Section 4.3).
With the recent availability of a number of highly purified DNA repair proteins, e.g.,
RecA, RecBC, UvrA, UvrB, UvrC, UvrD, Phr, PoIA, and Lig, and of DNA-sequencing
gels and similar techniques, biochemical studies on DNA repair in vitro are under way in a
number of laboratories. As a result, major advances in our understanding of the bio-
chemistry of DNA repair and of recombination should be forthcoming.

4.2. UV RADIATION MUTAGENESIS

4.2.1. Introduction
The action spectrum for UV radiation mutagenesis in bacteria mimics the absorption
spectrum for nucleic acid (DNA). Therefore, to understand the molecular basis of UV
radiation mutagenesis one needs to understand the biochemistry of DNA in terms of both
replication and repair.
UV Radiation Effects 127

4.2 .1.1. DNA Replication

A UV-irradiated cell may suffer a mutation because a DNA lesion produces a coding
error during the replication of DNA. It has been mentioned (Section 2.3.1.3) that the
photohydrate of cytosine codes for adenine rather than for guanine in an in vitro replica-
tion system. Such miscoding lesions, which lead directly to mutations via DNA replica-
tion, must be rare events in vivo after UV irradiation, since mutations are not induced by
UV irradiation in the repair-deficient recA and lexA strains, where mutations can only be
produced by replication errors (see below). In support of this concept, mutations can be
induced in these strains by treatment with certain chemicals (e.g., ethylmethanesulfonate)
that produce altered DNA bases with different coding properties. The fact that a muta-
tional event is recA-independent has been regarded as diagnostic of the involvement of
miscoding lesions and of DNA replication in the mutagenic process.

4.2.1.2. DNA Repair

Noncoding lesions in DNA elicit a repair response. The second mechanism for
mutagenesis requires the induction of an error-prone DNA repair system that produces
changes in the base sequence of the DNA in the course of restoring viability. Evidence for
this mechanism of mutagenesis followed from the initial observation that certain DNA
repair-deficient strains of E. coli (i.e., recA and lexA) cannot be mutated by UV irradia-
tion, suggesting that error-prone DNA repair plays the major role in UV radiation muta-
genesis. (I 9) In general, postreplication repair is considered to be more error-prone than
excision repair.

4.2.1.3. Types of Mutations

One type of DN A alteration that leads to mutations is the substitution of one base pair
by another. Base substitution mutations can be of two kinds. In transitions a purine is
replaced by another purine (e.g., G ~ A), or a pyrimidine is replaced by another
pyrimidine (e.g., C ~ T). In transversions a purine is replaced by a pyrimidine or vice
versa [i.e., (G or A) ~ (C or T)]. If the base substitution causes a change of one amino
acid in the protein and the protein exhibits an altered function, this is called a missense
mutation. If the base substitution produces a chain-terminating codon (i.e., a triplet of
nuc1eotides that does not code for any amino acid), it is called a nonsense mutation, and
an incomplete protein is synthesized.
The second type of mutagenic change in DNA is the deletion or insertion of a base
pair (or a larger unit) to yield aframeshift mutation. The addition or removal of a base pair
would shift the reading frame of the genetic code such that a string of incorrect codons and
incorrect amino acids would result.

4.2.2. Mutation Assays

Bacteria have been popular subjects for the study of the molecular basis of muta-
genesis because they grow rapidly and can easily be obtained in large numbers. Most
bacterial studies have followed the acquisition of resistance to killing by bacteriophage or
128 Chapter 4

by antibiotics, and the reversion to prototrophy (i.e., an auxotrophic bacterium may

require an amino acid for growth, but a prototrophic revertant would not).
The reversion to prototrophy has been the most popular assay because low frequen-
cies of mutations can easily be detected and the same medium (partially enriched medium
containing -1 j.Lg/rnl of the required amino acid) can be used to score both mutations and
survival. When plating for survival at high dilution, auxotrophs form small (i.e., they run
out of the required amino acid) but visible colonies. When plated at high cell densities for
scoring for mutations, the auxotrophic population forms a limited' 'lawn" of cells against
which prototrophic mutants form large colonies. The irradiated cells need a small amount
of the required amino acid in order to "fIx" the mutation through error-prone DNA repair
before they become prototrophs.
With the new DNA-sequencing techniques, one no longer needs to depend on earlier
indirect techniques to classify the type of mutation (i.e., missense, nonsense, frameshift)
that is produced (e.g., based on the types of chemicals that are needed to revert the
mutation). Now the exact base change due to the mutation can be determined. Through
such DNA-sequencing procedures it was shown that the (6-4) pyrimidine adduct (Section
2.3.1.3) was the mutagenic lesion at certain "hot spots" in the lad gene, as opposed to
the previous conclusion that cyclobutane-type pyrimidine dimers were responsible for all
UV radiation mutagenesis. (44)
Mutation studies in mammalian cells are not as easy to perform, and there are
relatively few assays (i.e., loci) from which to choose. Recently, a new technique has
been introduced that overcomes many of these problems, and the mutations can be readily
assayed at the DNA sequence level. In one example of this procedure, the bacterial lad
gene, carried on a shuttle vector (i.e., an autonomously replicating plasmid), is introduced
into mammalian cells and allowed to replicate in the cell nucleus. The cells are then
treated with the mutagen, and one or two days later the vector DNA is recovered and
transfected back into E. coli for the analysis of mutations in the lad gene. (45) Other
shuttle vector techniques are also under development for the study of mutagenesis in
mammalian cells. (46)

4.2.3. Kinetics of Induction

Classically, UV radiation mutagenesis was categorized as being dependent on the
interaction of two radiation-induced events, i.e., it was a two-hit process, while X-ray
mutagenesis was categorized as being a one-hit process. That is, if one plots mutant
frequency against dose on a log-log plot, the slopes of the straight lines would be 2 and 1,
respectively. With the observation that UV radiation mutagenesis was inducible, there
was an attempt to explain the two-hit response as one hit in the gene to be mutated, and
one hit somewhere else in the cell to induce error-prone DNA repair. There is some
support for this hypothesis after low fluences of UV radiation. (19)
Like most generalities, the apparent distinction between X-ray mutagenesis and UV
radiation mutagenesis on the basis of the hit number is not strictly correct. Both radiations
show one-hit and two-hit mutagenesis. For UV radiation mutagenesis, the one-hit mode
has been correlated with the production of back mutants (i.e., the direct correction of the
initial genetic error), while the two-hit process has been correlated with the formation of
suppressor mutations [i.e., a second mutation in the cell, usually a mutation affecting the
UV Radiation Effects 129

anticodon of a transfer RNA (tRNA), suppresses the effect of the fIrst mutation].(47) Is
there some unique structural feature (i.e., base sequence) in tRNA genes that requires (or
predisposes them to) two-hit kinetics for UV radiation mutagenesis? For example, it
would require two hits to produce an overlapping DNA daughter strand gap. Very many
fundamental questions about UV radiation mutagenesis, such as this, remain to be
answered.

4.2.4. Genetic Control of Mutagenesis

The observation that the recA and lexA strains of E. coli could not be mutated by UV
irradiation led to the discovery that UV radiation mutagenesis is not a passive process but
is due to inducible, error-prone DNA repair.(19) Because the recA and lexA genes seemed
to be involved in the regulation of the complex SOS response, it was thought that there
might be other gene products that participate more directly in error-prone DNA repair. To
this end, the umuC and umuD mutations were isolated, which block all of targeted UV
radiation mutagenesis. (reviewed in 20) The umuC and umuD genes are part of the SOS
system but their mechanism of action is not known. A umuC mutation has a small effect
on the repair of DNA daughter strand gaps after UV irradiation, and it has been suggested
that the umuC gene may be involved in the repair of overlapping DNA daughter strand
gaps,(48) since such lesions cannot be repaired by the usual mechanism of postreplication
repair. However, if one of the gaps could be repaired by some unique process such as
translesion synthesis (see below), then the other gap could be repaired by classical
postreplication repair.
Actually, translesion synthesis was offered some years ago as a model to explain all
of UV radiation mutagenesis. (19) In this model, an inhibitor of the editing function of
DNA polymerase is induced, thus causing the polymerase to synthesize past noncoding
lesions, yielding modified DNA. Recently, the RecA protein has been found to bind to
UV -irradiated DNA and to inhibit the editing subunit of DNA polymerase III halo-
enzyme. (49) In addition, a model has been presented that suggests that while nucleotides
can be inserted opposite noncoding lesions in UV -irradiated cells, the UmuDC proteins
are required for the resumption of replication past the lesion. (50)
The mucA and mucB genes in the plasmid pKMIOI are analogs of the umuD and
umuC genes in E. coli. A nonmutable umuC strain can be made mutable by infection with
pKM 10 1, and a umuC+ strain can be made more mutable by infection with pKMlOl.(20)
The Ames tester strains, which are used for screening environmental mutagens, often
carry the plasmid pKMlOI to increase their sensitivities to mutagens.
While a umuC mutation knocks out all of targeted UV radiation mutagenesis, it only
knocks out part of X-ray mutagenesis. (51) Furthermore, it is the "oxygen effect" of X-ray
mutagenesis that is blocked by a umuC mutation [i.e., while wild-type cells show en-
hanced mutagenesis if they are X-irradiated in the presence of oxygen (vs. the absence of
oxygen), a umuC strain shows no such oxygen effect]. A small portion of oxygen-
independent X-ray mutagenesis is also umuC-dependent and leads to (AT or GC) ~ TA
transversion mutations. The umuC-independent, oxygen-independent mutagenesis results
in GC ~ AT and AT ~ GC transitions.
The fact that the umuDC genes only control a portion of X-ray mutagenesis raises the
possibility that other genes control the umuDC-independent portion of X-ray mutagenesis.
130 Chapter 4

Recently, three mutations (xmu, X-ray mutability) that block umuC-independent X-ray
mutagenesis have been isolated in this laboratory and await further study.

4.3. SPONTANEOUS MUTAGENESIS(2)

4.3.1. DNA Replication

Until very recently, it was generally believed that spontaneous mutagenesis was due
entirely to errors made during DNA replication, as a result of the inherent inaccuracy of
the replication system.
One theory for spontaneous mutagenesis is that when a hydrogen atom on a DNA
base shifts from one position to another, the base assumes a different configuration (i.e.,
undergoes a tautomeric shift) such that it can form an abnormal base pair. The occurrence
of such tautomeric shifts at the moment of DNA synthesis would result in mutations if not
subsequently repaired.
If a DNA polymerase does put in an incorrect nucleotide, it has an editing function
that can cut out the incorrect nucleotide, and then it can try again to insert the correct one.
However, DNA polymerase does occasionally make errors that it fails to correct.
The second line of defense against replication errors is mismatch repair. (2,20,53)
When a mismatched base pair is recognized shortly after replication, the incorrect base is
cut out and replaced with the correct one. In E. coli the DNA that is newly synthesized is
not yet methylated, whereas the parental strand is methylated at the adenine residues in
GATe sequences. This pattern of methylation is what tells the mismatch repair system
which strand of DNA to cut in order to remove the incorrect base. Mutations that inhibit
mismatch repair (e.g., mutH, mutL, mutS) greatly increase the level of spontaneous
mutagenesis.
In a mutant deficient in DNA adenine methylase (dam), which methylates the ade-
nines of newly synthesized DNA, the cell doesn't know which strand to cut for mismatch
repair, and has a 50: 50 chance of cutting the correct one. Therefore, the spontaneous
mutation level is increased in dam mutants. Furthermore, since neither strand is methy-
lated, the cell sometimes cuts both strands, producing a DNA double-strand break. The
production of such double-strand breaks in a dam strain can be prevented by an additional
mutL or mutS mutation that blocks mismatch repair. (54)

4.3.2. DNA Repair

About half of spontaneous mutagenesis in wild-type E. coli is due to error-prone
DNA repair. This conclusion is based in part on the observations that about half of
spontaneous base substitution mutagenesis is under the controi of the recA gene; and that
mutations in DNA repair genes that either reduce UV radiation mutagenesis (e.g., lexA,
umuC) or enhance UV radiation mutagenesis (e.g., uvrA, uvrB) have a similar effect on
spontaneous mutagenesis. This latter result has also been interpreted to mean that some of
the DNA lesions that lead to spontaneous mutagenesis are "UV-like" in their require-
ments for repair.
What is the source of this UV -like damage? Excited state intermediates are formed
enzymatically by certain oxidases. For example, the action of horseradish peroxidase on
UV Radiation Effects 131

isobutyraldehyde results in the production of acetone in its excited state. This enzymati-
cally generated, excited state acetone can transfer its energy to DNA and produce damage
in vitro. Recently, the metabolism of phenylalanine has been shown to produce UV -like
lesions in DNA (based on genetic studies) that result in mutations in E. coli. Phe-
nylalanine is only weakly mutagenic, which is a good thing since we can't live without it.
(N.B.: Oxygen is also mutagenic, and also necessary for life.) In contrast, the presence of
excess cystine, arginine, or threonine appears to be antimutagenic in E. coli.
DNA repair is not only important in spontaneous mutagenesis for E. coli, but also for
yeast, and for all other organisms where data are available on this topic. (52) Since excision
repair-deficient strains of E. coli show an enhanced level of spontaneous mutagenesis, it
allows one to speculate that humans that are deficient in excision repair (i.e., patients with
xeroderma pigmentosum) should show an abnormal amount or type of internal organ
cancer. There are some data to support this hypothesis.(e. g ., 55)
It is important to remember that the spontaneous mutation rate of a cell culture or of
an organism is the net result of a large number of both mutagenic and antimutagenic
processes that involve both DNA replication and DNA repair.

4.4. REFERENCES

1. R. B. Roberts and E. Aldous, Recovery from ultraviolet irradiation in Escherichia coli, J. Bacteriol. 57,
363-375 (1949).
2. E. C. Friedberg, DNA Repair. W. H. Freeman, New York (1985).
3. E. C. Friedberg, U. K. Ehmann, and J. I. Williams, Human diseases associated with defective DNA repair,
Adv. Radiat. BioI. 8, 85-174 (1979).
4. K. B. Hellman, G. B. Schuller, and W. R. Lewis, Immunological defects in human radiation sensitive
disorders, in: Topics in Photomedicine (K. C. Smith, ed.), pp. 143-182, Plenum Press, New York (1984).
5. R. H. Haynes and B. A. Kunz, DNA repair and mutagenesis in yeast, in: Molecular Biology of the Yeast
Saccharomyces: Life Cycle and Inheritance. Monograph I1A, pp. 371-414, Cold Spring Harbor Laborato-
ry, Cold Spring Harbor, New York (1981).
6. E. C. Friedberg and B. A. Bridges (eds.), Cellular Responses to DNA Damage, Alan R. Liss, New York
(1983).
7. E. C. Friedberg and P. C. Hanawalt (eds.), DNA Repair: A Laboratory Manual of Research Procedures,
Vol. 1, Marcel Dekker, New York, (1981), Vol. 2 (1983), Vol. 3 (1988).
8. D. E. Lea, Actions of Radiation on Living Cells, Cambridge University Press, London (1955) (Reprinted
1962).
9. W. Harm, Biological Effects of Ultraviolet Radiation, Cambridge University Press, Cambridge (1980).
10. R. H. Haynes, The influence of repair processes on radiobiological survival curves, in: Cell Survival After
Low Doses of Radiation: Theoretical and ClinicalImplications (T. Alper, ed.), pp. 197-208, Wiley, New
York (1975).
11. R. H. Haynes, F. Eckardt, and B. A. Kunz, The DNA damage-repair hypothesis in radiation biology:
Comparison with classical hit theory. Br. J. Cancer 49, suppl. VI, 81-90 (1984).
12. P. A. Swenson, Physiological responses of Escherichia coli to far-ultraviolet radiation, in: Photochemical
and Photobiological Reviews. Vol. 1 (K. C. Smith, ed.), pp. 269-387, Plenum Press, New York (1976).
13. M. Tang and K. C. Smith, The expression of liquid holding recovery in ultraviolet-irradiated Escherichia
coli requires a deficiency in growth medium-dependent DNA repair, Photochem. Photobiol. 32, 763-769
(1980).
14. M. Bertrand and D. F. Deen, Factors influencing the recovery from potentially lethal damage (PLD) in
mammalian cells in vitro and in vivo, Cancer Treat. Rev. 7, 1-15 (1980).
15. R. C. Sharma, N. J. Sargentini, and K. C. Smith, New mutation (mmrAl) in Escherichia coli K-12 that
affects minimal medium recovery and postreplication repair after UV irradiation, J. Bacteriol. 154, 743-
747 (1983).
132 Chapter 4

16. R. C. Sharma and K. C. Smith, A mechanism for rich-medium inhibition of the repair of daughter-strand
gaps in the deoxyribonucleic acid of UV-irradiated Escherichia coli K12 uvrA, Mutation Res. 146, 177-183
(1985).
17. N. J. Sargentini, W. P. Diver, and K. C. Smith, The effect of growth conditions on inducible, recA-
dependent resistance to X-rays in Escherichia coli. Radiat. Res. 93, 364-380 (1983).
18. N. J. Sargentini and K. C. Smith, Quantitation of the involvement of the recA, recB, recC, reeF, recJ,
recN, lexA, radA, radE, uvrD, and umuC genes in the repair of X-ray-induced DNA double-strand breaks in
Escherichia coli, Radiat. Res. 107, 58-72 (1986).
19. E. M. Witkin, Ultraviolet mutagenesis and inducible DNA repair in Escherichia coli, Bacteriol. Rev. 40,
869-907 (1976).
20. G. C. Walker, Mutagenesis and inducible responses to deoxyribonucleic acid damage in Escherichia coli,
Microbiol. Rev. 48, 60-93 (1984).
21. G. C. Walker, Inducible DNA repair systems, Annu. Rev. Biochem. 54,425-437 (1985).
22. M. Radman, Is there SOS induction in mammalian cells? Photochem. Photobiol. 32, 823-830 (1980).
23. T. G. Rossman and C. B. Klein, Mammalian SOS system: A case of misplaced analogies, Cancer Invest. 3,
175-187 (1985).
24. G. M. Hahn, Hyperthermia and Cancer, Plenum Press, New York (1982).
25. B. Demple and J. Halbrook, Inducible repair of oxidative DNA damage in Escherichia coli, Nature 304,
466-468 (1983).
26. P. C. Lee, B. R. Bochner, and B. N. Ames, AppppA, heat-shock stress, and cell oxidation, Proc. Natl.
Acad. Sci. USA 80, 7496-7500 (1983).
27. R. M. Tyrrell, A common pathway for protection of bacteria against damage by solar UVA (334 nm, 365
nm) and an oxidising agent (H 202 ), Mutation Res .. 145, 129-136 (1985).
28. W. Harm, C. S. Rupert, and H. Harm, The study of photoenzymatic repair of UV lesions in DNA by flash
photolysis, Photophysiology 6, 279-324 (1971).
29. D. A. Youngs and K. C. Smith, Genetic location of the phr gene of Escherichia coli K-12, Mutation Res.
51, 133-137 (1978).
30. A. Sancar and C. S. Rupert, Correction of the map location for the phr gene in Escherichia coli K-12,
Mutation Res. 51, 139-143 (1978).
31. A. Sancar and G. B. Sancar, Escherichia coli DNA photolyase is a flavoprotein, J. Mol. Bioi. 172, 223-
227 (1984).
32. E. C. Friedberg, K. H. Cook, J. Duncan, and K. Mortelmans, DNA repair enzymes in mammalian cells, in:
Photochemical and Photobiological Reviews, Vol. 2 (K. C. Smith, ed.), pp. 263-322, Plenum Press, New
York (1977).
33. I. Husain, B. van Houten, D. C. Thomas, M. Abdel-Monem, and A. Sancar, Effect of DNA polymerase I
and DNA helicase II on the turnover rate of UvrABC excision nuclease, Proc. Natl. Acad. Sci. USA 82,
6774-6778 (1985).
34. P. C. Hanawalt, P. K. Cooper, A. K. Ganesan, R. S. Lloyd, C. A. Smith, and M. E. Zolan, Repair
responses to DNA damage: Enzymatic pathways in E. coli and human cells, J. Cellular Biochem. 18, 271-
283 (1982).
35. K. C. Smith and R. C. Sharma, A model for the recA-dependent repair of excision gaps in UV-irradiated
Escherichia coli, Mutation Res. 183, 1-9 (1987).
36. J. E. Cleaver, Repair processes for photochemical damage in mammalian cells, Adv. Radiat. Bioi. 4, 1-75
(1974).
37. T. Lindahl, DNA repair enzymes, Annu. Rev. Biochem. 51, 61-87 (1982).
38. K. C. Smith, T. V. Wang, and R. C. Sharma, recA-Dependent DNA repair in UV-irradiated Escherichia
coli, J. Photochem. Photobiol., B:Biology 1, I-II (1987).
39. A. K. Ganesan, Persistance of pyrimidine dimers during post-replication repair in ultraviolet light-irradiated
Escherichia coli, J. Mol. Bioi. 87, 103-119 (1974).
40. A. J. Fornace, Recombination of parent and daughter strand DNA after UV-irradiation in mammalian cells,
Nature 304, 552-554 (1983).
41. R. H. Rothman, T. Kato, and A. J. Clark, The beginning of an investigation of the role of reeF in the
pathways of metabolism of ultraviolet-irradiated DNA in Escherichia coli, in: Molecular Mechanisms for
Repair of DNA (P. C. Hanawalt and R. B. Setlow, eds.), pp. 283-291, Plenum Press, New York (1975).
42. T. V. Wang and K. C. Smith, Mechanisms for reeF-dependent and recB-dependent pathways of postrepli-
cation repair in UV-irradiated Escherichia coli uvrB, J. Bacteriol. 156, 1093-1098 (1983).
UV Radiation Effects 133

43. R. C. Sharma and K. C. Smith, A minor pathway of postreplication repair in Escherichia coli is indepen-
dent of the recB, recC and recF genes, Mutation Res. 146, 169-176 (1985).
44. W. A. Franklin and W. A. Haseltine, The role of the (6-4) photoproduct in ultraviolet light-induced
transition mutants in E. coli, Mutation Res. 165, 1-7 (1986).
45. J. S. Lebkowski, S. Clancy, J. H. Miller, and M. P. Calos, The lael shuttle: Rapid analysis of the
mutagenic specificity of ultraviolet light in human cells, Proc. Natl. Acad. Sci. USA 82, 8606-8610 (1985).
46. M. P. Calos, Mutation of autonomously replicating plasmids, in: Gene Transfer (R. Kucherlapati, ed.), pp.
243-261, Plenum Press, New York (1986).
47. N. J. Sargentini, R. C. Bockrath, and K. C. Smith, Three mechanisms for ultraviolet radiation mutagenesis
in Escherichia coli K-12 uvrB5: Specificity for the production of back and suppressor mutants, Mutation
Res. 106,217-224 (1982).
48. T. V. Wang and K. C. Smith, Role ofthe umuC gene in postreplication repair in UV-irradiated Escherichia
coli K-12 uvrB, Mutation Res. 145, 107-112 (1985).
49. C. Lu, R. H. Scheuermann, and H. Echols, Capacity of RecA protein to bind preferentially to UV lesions
and inhibit the editing subunit (E) of DNA polymerase III: A possible mechanism for SOS-induced targeted
mutagenesis, Proc. Natl. Acad. Sci. USA 83, 619-623 (1986).
50. B. A. Bridges, R. Woodgate, M. Ruiz-Rubio, F. Sharif, S. G. Sedgwick, and U. Hubscher, Current
understanding ofUV-induced base pair substimtion mutation in E. coli with particular reference to the DNA
polymerase III complex, Mutation Res. 181, 219-226 (1987).
51. N. J. Sargentini and K. C. Smith, umuC-dependent and umuC-independent '1- and UV-radiation muta-
genesis in Escherichia coli, Mutation Res. 128, 1-9 (1984).
52. N. J. Sargentini and K. C. Smith, Spontaneous mutagenesis: The roles of DNA repair, replication, and
recombination, Mutation Res. 154, 1-27 (1985).
53. M. Radman and R. Wagner, Mismatch repair in Escherichia coli, Annu. Rev. Genet. 20, 523-528 (1986).
54. T. V. Wang and K. C. Smith, Inviability of dam recA and dam recB cells of Escherichia coli is correlated
with their inability to repair DNA double-strand breaks produced by mismatch repair, J. Bacterial. 165,
1023-1025 (1986).
55. K. H. Kraemer, M. M. Lee, and J. Scotto, DNA repair protects against cutaneous and internal neoplasia:
Evidence from xeroderma pigmentosum, Carcinogenesis 5,511-514 (1984).
 5
Environmental Photobiology

5.1. Introduction ...................................................................... 135

5.2. Solar Radiation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 136
5.2.1. The Solar Spectrum .......................................................... 136
5.2.2. Stratospheric Ozone and the Evolution of Terrestrial Organisms ...................... 138
5.2.3. The Solar Radiation Environment in Aquatic Ecosystems ........................... 140
5.2.4. The Solar Radiation Environment beneath Vegetation Canopies ...................... 141
5.3. How Organisms Protect Themselves from Solar UV Radiation ............................. 143
5.3.1. Terrestrial Plants ............................................................ 143
5.3.2. Animals and Aquatic Organisms. . . ... . . . . . . . . . . . .. . . . ... . . . . . . . . . . . . . . . . . . . . . .. 145
5.4. Biological Consequences of Stratospheric Ozone Depletion ................................ 147
5.4.1. The Response of Individual Plant Species and the Consequences for Ecosystem
Productivity ................................................................ 148
5.4.2. Effects on Terrestrial and Aquatic Animal Populations .............................. 151
5.5. Conclusions ...................................................................... 152
5.6. References ....................................................................... 152

5.1. INTRODUCTION

The abiotic and biotic components that make up the surroundings or environment of an
organism can exert considerable influence on the effects of light-mediated processes
within the organism. Environmental factors such as temperature, the availability of water
and nutrients, and interactions with other organisms modify the effects of light-mediated
processes within an organism and thereby ultimately affect its growth and survival in the
ecosystem. The examination of individual processes involving light allows us to under-
stand the mechanism of these processes within an organism. How radiation is absorbed,
what wavelengths are utilized, and what action or effect the radiation elicits in biological
systems are questions of primary concern for photobiologists. For example, studies on the
absorption of light by chlorophyll and the photochemical reactions of the photosynthetic
apparatus are important for understanding the fundamental mechanism of photosynthesis.
Important considerations in this mechanism would include the absorption spectrum for
chlorophyll and the number of photons required to provide sufficient energy for the light
reaction of photosynthesis. Although knowledge of how the photosynthetic mechanism

Ronald Robberecht • Department of Range Resources, University of Idaho, Moscow, Idaho

83843.

135
136 Chapter 5

works is essential, this alone does not allow us to predict how the whole organism would
respond in nature. This is because the actual capacity of a plant for carbon assimilation
and biomass production is determined by the complex interplay of plant genetics and
physiology, and the environment. The fundamental nature of processes involving light are
not changed, but rather their effect is modified by the way in which the organism interacts
with its environment. It is therefore important to consider light-mediated processes in the
context of the whole organism and its interaction with the environment. Environmental
photobiology thus provides a bridge between the understanding offundamental processes
involving light within an organism and the effects of these processes on the whole
organism in the ecosystem.
The spectral distribution and irradiance of incoming solar radiation can be signifi-
cantly altered in terrestrial ecosystems by the vegetation canopy. Further selective at-
tenuation will occur as radiation penetrates the leaf. Therefore, the position of a plant in a
forest canopy or leaves within an individual plant canopy will to a large extent determine
the wavelength quality and irradiance available for light-mediated processes. The situa-
tion is similar for aquatic ecosystems, where the selective attenuation of radiation by
water occurs. Water clarity and the location of aquatic plants and microorganisms are
critical variables in this ecosystem.
Environmental photobiology thus involves an understanding of how the general
environment alters the radiation regime for organisms as well as how the general environ-
ment influences the behavior of the organism in the ecosystem after light-mediated pro-
cesses have produced an effect. The field of environmental photobiology encompasses a
wide variety of topics, ranging from the more simple case of how one environmental
factor affects one organism, to the most complex level of the ecosystem where the
interaction among species and multiple environmental factors must be considered. An
environmental factor that has recently caused great scientific and public concern is the
potential intensification of UV radiation on earth due to a partial depletion of stratospheric
ozone. Because of the effectiveness of UV radiation to cause damage in biological
systems and because this environmental problem is of global concern, it will be used to
illustrate many of the aspects of environmental photobiology in this chapter.

5.2. SOLAR RADIATION

5.2.1. The Solar Spectrum

The solar energy incident on the earth's atmosphere is relatively constant at approx-
imately 1.39 kW/m2 and has a spectral distribution that approximates a 60000 K black-
body radiation curve (Fig. 5_1).(1,2) Because of reflection, molecular and particulate
scattering, and absorption of radiation in the atmosphere, solar irradiance is reduced as it
penetrates the atmosphere, so that on average only approximately one-half of the radiation
incident at the top of the atmosphere reaches the ground. The amount of solar radiation
incident at the surface of the earth is highly dependent on cloud cover and the clarity of the
atmosphere, as well as solar angle. In addition to reduced irradiance at the earth's surface,
the change in spectral distribution, or wavelength quality, as radiation penetrates the
atmosphere, vegetation canopies, or water is particularly significant. Since the action of
Environmental Photobiology 137

Fig. 5-1. Solar radiation at the top of the Infrared

atmosphere and at sea level (top). The so-
Visible 400-700
lar spectrum at the earth's surface is trun-
320-400
cated at approximately 295 mn by strat- 290-320
ospheric ozone. The absorption bands in 's 0.20 190-290
the infrared region are due to the absorp- N "

tion of solar radiation by water, oxygen, 's

0.15
carbon dioxide, and ozone molecules in Extraterrestrial solar radiation
the atmosphere. (Adapted from Ref. I.)
The absorption, transmittance, and u
..
reflectance characteristics for various chro- "'0o 0.10
Solar radiation at sea level
mophores and plant tissues (bottom). The
.~
leaf epidermis is a highly selective filter of

.
UV-B radiation as shown by the UV trans- o 0.05
U
mittance spectra for Argyroxiphium sand- a.
wicense (silversword) and Oenothera stric- rn
,I-"'~ ...\
ta (evening primrose). (Adapted from
1400 1800 2200 2600 3000
Refs. 23 and 26.) The highly pubescent
leaf surface of A. sandwicense and the Protein ~a~bs;;:o;jrpiiiliiOionn----_ _ _ __
\ Nucleic ac~
highly glaucous or wax-covered leaf sur- • ~V-
face of Dudleya brittonii (live-forever) .. . I \ ~e absorption Oenothero stricto

..
; } ........ (Epidermal Iran,millance)

I i.
show that these specialized leaf mor- o
phologies can result in high levels of UV
u

i ..............
l .L.j.:-:-:...:..... .
and visible reflectance. (Adapted from ~ J1: . / L Oudleya "bri'iionii
"0 0.50 /"" I: I
I
(Leaf refleclance)
Refs. 23 and 24.) The relatively low level
of UV-B transmittance for glabrous leaves ~ I ___ /~Argyrt:llfiPhium sondwlcense

such as those of O. stricta is primarily due 0.25 ,\. / (Leaf refleclance )

Flavonoid, / Epidermal transmillance
to the UV absorption characteristics of fla- absorption I ~

vonoid and related phenolic compounds o ...- ,/

(shown is the spectrum for 7-hydroxyiso- 200 300 400 500 600 700 800
flavone, adapted from Ref. 31). The ozone Wavelength (nm)
absorption coefficient, normalized to 290
nm, shows that extremely effective absorption occurs at wavelengths below 300 nm. (Adapted from Ref. 7.)
Although the absorption of UV radiation by nucleic acids and proteins is maximal in the UV -C waveband, these
chromophores do absorb UV-B. Partial depletion of stratospheric ozone is predicted to result in increased UV-B
irradiance as well as a shift in the solar spectrum at the ground surface toward shorter wavelengths. There would
thus be a greater degree of overlap between the UV -B waveband and the absorption spectra of nucleic acids and
proteins. (Adapted from Ref. 32.)

radiation on biological processes is highly wavelength-dependent, the spectral distribution

of radiation in the environment of the organism, and how the organism itself alters the
radiation through, e.g., pigmentation, will greatly influence the ultimate effect of radia-
tion on physiological processes in the organism.
The solar radiation curve shown in Fig. 5-1 indicates that the solar spectrum at the
ground surface can be partitioned according to the biological effectiveness of the major
wavebands. On earth, the solar spectrum is truncated in the UV waveband at approx-
imately 295 nm by the absorption of stratospheric ozone. In this region of the solar
spectrum, the UV-B waveband (290-320 nm) is of particular interest because of its
potential to cause damage to organisms. (3) The biological effectiveness of UV radiation
increases logarithmically with decreasing wavelength. This increase is primarily due to
the increased overlap between the UV-B waveband and the absorption spectra of nucleic
138 Chapter 5

acids and proteins as the wavelength decreases. (4) The visible radiation waveband at 400-
700 run, also referred to as photosynthetically active radiation (PAR), is primarily signifi-
cant for its role in photosynthesis and plant photomorphogenesis, although it also has
significant thermal and photodestructive effects. (5) The infrared (IR) region extends from
700 to approximately 4000 nm, and is primarily significant in regard to thermal effects on
organisms. References 1-7 are suggested for further reading on the quantitative aspects of
solar radiation.

5.2.2. Stratospheric Ozone and the Evolution of Terrestrial Organisms

Ozone forms a thin layer in the stratosphere with a maximum concentration between
20 and 26 km above sea level. It absorbs solar UV-B radiation with increasing effective-
ness at shorter wavelengths so that essentially no radiation below 295 nm penetrates
through the atmosphere to the ground. (2) Ozone (03) is constantly formed in the upper
atmosphere through the combination of molecular oxygen (02) and atomic oxygen (0).
The latter is formed from the photodissociation of 02 by short-wavelength UV -C radiation
«240 nm). (8) Other photochemical processes in the upper atmosphere, involving reac-
tions catalyzed by oxides of nitrogen and chlorine (e.g., NOx and Cl), result in the
breakdown of ozone. This natural formation and breakdown of ozone is in balance and
results in an effective shield against injurious UV radiation at the ground. Ozone pho-
tochemistry in the stratosphere is limited by oxygen availability and may be affected by
natural perturbations such as stratospheric temperature variations, solar proton events,
changes in solar irradiance, and chemical inputs from volcanic eruptions. (9)
Most of the ozone is produced above the equator where solar irradiance is maximal.
Ozone formed at these latitudes subsequently diffuses in the stratosphere toward the poles.
Thus, while ozone concentrations in the stratosphere above the equator may average 0.29
cm (the thickness of a vertical column of stratospheric ozone condensed to standard
temperature and pressure), ozone concentrations above the poles may exceed 0.40 cm at
the end of the winter season.(8) Considerable seasonal and daily fluctuations in ozone
concentration may occur, however. Concentrations of stratospheric ozone tend to be
highest in late winter and early spring, and lowest in late summer and early fall. Because
the smallest seasonal variation in solar irradiance occurs above the equator, stratospheric
ozone at this latitude also shows the smallest seasonal variation. At temperate latitudes the
seasonal variation in stratospheric ozone concentration appears to be small, and at first
glance not too biologically relevant. Ozone measured at Arosa, Switzerland, for example,
varied from a mean monthly high of 0.27 cm in April to a low of 0.205 cm in November
(Fig. 5-2).(2) The relatively small monthly fluctuations in ozone shown for this location,
however, can result in large variations in UV-B irradiance. This is because the absorption
of UV radiation by ozone increases exponentially with ozone thickness.
The significance of the stratospheric ozone layer and its seasonal variation becomes
more apparent when viewed in regard to the biological effectiveness of UV -B radiation.
Biological effectiveness refers to the capacity of UV -B radiation to cause damage to living
organisms and is based on the biological weighting factor used to quantify UV-B re-
sponse. The damage induced at each wavelength of the UV-B waveband to DNA mole-
cules, photosynthesis, or the whole organism has been used to develop biological weight-
ing factors. These weighting factors, also known as action spectra, are highly wavelength-
Environmental Photobiology 139

E 0.30 r - - - - - - - - - - - ,
$ Uean monthly ozone thickness
Aroaa. Switzerland
"c 0.28

~O.26 ~
Fig. 5-2. Mean monthly stratospheric ozone concentration at
standard temperature and pressure for Arosa, Switzerland. Be- Io
0.24
/ "'0 "',-
cause UV-B irradiance varies exponentially with ozone con-
centration, the apparently small monthly changes in ozone at this
.8 0.22 0",,- /
Ul~ .~
0.20 '--'--'--'--'--'---'---l.---l.-L---'--'
temperate latitude location result in large changes in UV -B irra- JFMAMJJASOD
diance. (Adapted from Ref. 2.) Month

dependent in the UV region and tend to increase exponentially with decreasing wave-
length. An equation that integrates the weighting factor and solar irradiance over the UV-
B waveband is:

Biologically effective UV-B =

I 320

280
leA) . E(A) . dA

where leA) is the spectral irradiance at wavelength A, and E(A) is the biological weighting
factor.(6) The weighting factor for DNA damage in response to UV-B irradiation, for
example, increases more than four orders of magnitude as the wavelength decreases from
320 to 280 nm. This indicates that the potential for UV radiation-induced damage increases
dramatically with small decreases in wavelength. When both the biological effectiveness of
UV radiation and the wavelength dependency of the absorption of UV radiation by ozone
are considered in respect to the natural latitudinal gradient of ozone thickness, along with
increased solar angles at higher latitudes, a pronounced gradient in biologically effective
UV -B radiation results. (6) When the weighting factorfor UV -B radiation-induced damage
to DNA is used in the equation above, the biologically effective solar UV-B irradiance at
low latitudes can exceed that at high latitudes by an order of magnitude.
Prior to the development of the stratospheric ozone layer, the earth's atmosphere was
transparent to short-wavelength UV radiation «290 nm), wavelengths of sufficient ener-
gy and overlap with the absorption spectra of nucleic acids and proteins to cause lethal
damage to organisms. The processes that led to the development of an ozone layer about
600 million years ago remain unclear. It has been hypothesized that oxygen in the earth's
atmosphere increased slowly, at first through the photodissociation of water and later
through photosynthesis. (10) Since oxygen is essential for the photochemical production of
ozone, a threshold concentration of atmospheric oxygen was necessary before ozone
could be formed in significant amounts. An ozone concentration of 1% of present atmo-
spheric levels may have been reached about 600 million years ago. This primeval atmo-
sphere, still rich in highly photoreactive wavelengths, precluded terrestrial organisms. (10)
Plant and animal life at that time was confined to aquatic environments where the penetra-
tion of UV radiation was restricted. Early aquatic organisms were confronted with the
dilemma of occupying aquatic habitats deep enough to confer protection from harmful
short-wavelength UV radiation, yet having sufficient visible radiation available for photo-
synthesis. It was not until approximately 400 million years ago when ozone concentra-
140 Chapter 5

tions are believed to have reached 10% of present levels that adaptation to the terrestrial
environment was possible. The earliest fossil records of land plants and animals correlates
well with this theory of ozone development. Although considerable debate exists on the
exact processes that led to the development of the ozone layer and over what time scale
this occurred, it is clear that UV-B irradiance in the early atmosphere was more intense
than present levels. Furthermore, this radiation environment would have presented con-
siderable selective pressure for adaptation in terrestrial organisms under this intense UV -B
regime.
Mechanisms that attenuated UV -B radiation before damage occurred in nucleic acids
would have been of protective value and of great adaptive significance for survival in the
intense UV-B radiation environment. The more effectively an organism was protected
from UV-B radiation, the less likely that the rate of damage would exceed the capacity of
enzymatic mechanisms to repair UV radiation-induced damage. Early land plants adapted
to the relatively high UV-B radiation environment through the development of a pigment
system that provided effective protection from UV-B-induced injury.(lO,Il) This system
consists of flavonoid and related phenolic compounds, which are colorless to light yellow
pigments with a strong absorption of UV radiation at wavelengths less than 320 nm (Fig.
5-1). These compounds are present in most tissues of higher land plants, and form an
effective UV-B radiation filter in the outertissue layers ofleaves, stems, and flower parts.
With a few exceptions, flavonoid compounds are not present in algae. (12)
Flavonoid and related phenolic compounds are synthesized in plants in response to
UV-B irradiation. Their presence in leaf epidermal tissue and around chloroplasts in the
mesophyll reduces the potential for radiation-induced damage to sensitive targets in the
plant. Although the present stratospheric ozone layer provides an effective filter for short-
wave UV radiation and presently limits this waveband to 295 nm (Fig. 5-1), flavonoid and
phenolic compounds in leaves and other plant tissues still appear to have adaptive signifi-
cance for plant survival at present levels of UV-B irradiance. These pigments as well as
stratospheric ozone are not only effective filters of the more actinic wavelengths below
295 nm, but are transparent to visible radiation.

5.2.3. The Solar Radiation Environment in Aquatic Ecosystems

Water clarity in the environment of aquatic organisms directly affects the degree of
penetration of solar radiation into water. (13) Figure 5-3 illustrates the degree of penetration
and the spectral distribution of solar radiation to the I-m water depth for distilled water
and for lakes varying in water clarity. In the IR waveband, at wavelengths above 700 nm,
water is relatively opaque to radiation. (14) This strong absorption of IR radiation is a
physical property of water and is not particularly affected by the level of impurities in the
water. Impurities in water such as dissolved organic compounds do, however, exert a
highly significant influence on the penetration of visible and UV radiation. The effect of
these impurities increases with decreasing wavelength, so that while distilled water trans-
mits approximately 98% of the incident solar radiation below 550 nm to a depth of 1 m,
lakes with very high concentrations of dissolved organic compounds absorb essentially all
radiation below 550 nm in the first meter of water. Thus, the radiation environment for
aquatic organisms can be one of low solar irradiance and of a spectral distribution greatly
different from that found at the surface. Aquatic organisms in oligotrophic lakes, rela-
Environmental Photobiology 141

Fig. 5-3. The penetration of solar radiation into water and veg-
etation canopies. Curves A-D illustrate the level of transmit-
tance of solar radiation into I-m-deep water of different degrees 1.0
A
of clarity (curve A is for distilled water, and curves B-D are for g
lakes with increasing levels of dissolved organic compounds
that tend to attenuate UV wavelengths more effectively than "0
I:
0
0.5
longer wavelengths). (Adapted from Ref. 14.) Curve E shows ~
'E
that vegetation canopies tend to selectively absorb the visible or III
I:
photosynthetically active wavelengths (400-700 nm) and are ~
highly transparent to near-infrared radiation (700-2000 nm). 0.0
The radiation environment beneath a canopy is therefore shifted 300 400 500 600 700 800 900
to the infrared region. (Adapted from Ref. 15.) Wavelength (nm)

tively clear and nutrient-poor waters, would be exposed to high levels of solar UV-B and
visible irradiance. Those in eutrophic estuaries or other coastal waters, which are aquatic
environments rich in dissolved organic matter and nutrients, would exist in an environ-
ment much reduced in UV and visible wavelengths.
The reduced penetration of photosynthetically useful wavelengths into eutrophic
estuaries, lakes, or coastal waters would be expected to reduce the primary productivity of
these aquatic ecosystems. (13) However, significant environmental and behavioral factors
can ameliorate the effects of a reduced solar radiation regime. The movement of water,
either as upwelling along coastal regions, stream flow, or the seasonal turnover of thermal
strata in lakes, is a major factor in the productivity of aquatic ecosystems. This movement
will not only influence the distribution of nutrients and dissolved organic matter in the
water, but also may influence the location and movement of planktonic communities.
Nonmotile phytoplankton may benefit from periodic water movements if that movement
results in transport to the water surface layers where exposure to solar radiation would be
higher. In contrast, motile and phototactic algae may overcome the effects of water
movement by their ability to migrate to the upper water layers to increase the daily
reception of solar radiation. Because photosynthetically useful wavelengths are absorbed
in the upper I-m layer of relatively eutrophic ecosystems (Fig. 5-3), aquatic macrophytic
algae that are attached to rocks or sediments in lakes or in coastal waters must rely on
efficient mechanisms for the absorption and utilization of radiation deficient in these
wavelengths. This may involve, for example, a greater reliance on accessory pigments for
photosynthesis (Chapter 12). Thus, the reduced radiation regime of eutrophic ecosystems
could be compensated for by organisms that are adapted for low-light conditions, or with
mechanisms involving periodic migration to regions of higher solar irradiation. Reference
13 is recommended for a more detailed discussion on light-mediated processes and aquatic
ecosystems.

5.2.4. The Solar Radiation Environment beneath Vegetation Canopies

The radiation environment beneath plant canopies is also quite different in intensity
and spectral distribution from the solar radiation regime at the top of the vegetation layer.
Absorption of solar radiation by successive layers of leaves within an individual plant
canopy or within a stand of vegetation, however, results in a radiation environment
distinctly different in spectral distribution from that in aquatic ecosystems (Fig. 5-3). The
142 Chapter 5

mechanisms involved in the alteration of solar radiation by vegetation is similar for

canopies of individual plants and for the complex canopy structure of whole plant
communities.
Wavelength selectivity of the leaf is the most significant factor determining the level
and quality of irradiance underneath a vegetation canopy. Since leaves absorb essentially
all UV-B radiation, the lower leaves in a plant canopy tend to be well protected from
exposure to this photochemically active waveband. Coincident with the absorption spectra
of photosynthetic chromophores such as chlorophyll a and b, and with the spectra of
accessory pigments such as carotenoids and xanthophylls, approximately 90% of the
visible radiation incident on the leaf surface is absorbed. With this high level of absorption
at each leaf layer in the canopy, the radiation available to leaves in the lower strata of a
plant canopy, or the understory of a plant community, is rapidly depleted of wavelengths
that are photosynthetically useful. Because leaves are highly transparent to near-IR radia-
tion, the selective absorption ofUV-B and visible radiation and the selective transmittance
of near-IR radiation results in a radiation regime for the understory of a plant community
that is depleted of PAR and rich in near-IR radiation (see Fig. 5-3 and Ref. 15).
The progressive spectral shift toward the IR with increasing depth into the plant
canopy can significantly affect plant growth and development. Understory plants have
adapted to the PAR-poor and lR-rich radiation environment through leaf structural and
functional modifications. Relative to leaves developed in full sunlight, attributes of these
shade-adapted leaves include the tendency toward larger leaves, reduced mesophyll thick-
ness and chlorophyll alb ratios, and increased density of chloroplast membrane systems.
In addition, the accessory pigments such as xanthophylls tend to be more important for
shade-adapted plants. Functionally, the shade-adapted leaves have greatly reduced rates
of respiration, photosynthesis, and transpiration. In obligate shade-adapted plants, these
characteristics result in efficient utilization of irradiance, but also tend to preclude the
capacity to acclimate to full insolation.
Radiation penetration into plant canopies is often described simply by a modified
version of the Lambert-Beer extinction law:

where I represents irradiance at some depth in the canopy, lois solar irradiance incident at
the top of the canopy, k is the extinction coefficient for a particular vegetation stand, and
LAI is the ratio of total leaf area to ground area (m 2 /m 2 ) above the depth at which the final
irradiance is estimated. While this relationship is instructive for viewing general aspects
of radiation attenuation in different types of vegetation stands and is useful for comparing
the attenuating capacity of different types of plant communities, accurate descriptions of
such attenuation are considerably more complex. It is of particular importance to note that
this equation does not describe the shift in spectral distribution to the IR waveband as
radiation penetration into the canopy occurs. Of particular importance is the fact that
characterization of radiation penetration into plant stands involves the integration of all
aspects of vegetation architecture and solar radiation. (16) Some of the more significant
considerations include leaf optical properties (reflectance, absorptance, and transmit-
tance), leaf orientation in relation to the sun and to other leaves, the arrangement of
foliage strata in the canopy, changes in solar angle throughout the day and year, and the
Environmental Photobiology 143

affect of changes in the direct and diffuse components of solar radiation. All of these
components would have seasonal variability, especially in plant communities dominated
by deciduous trees, and daily variability as solar angle and radiation penetration through
the atmosphere changes. The integration of all these aspects into mathematical models
forms a considerably greater task than utilizing the Lambert-Beer law, and necessitates
the use of computer modeling. The end result of differences in vegetation architecture is
that the potential for photosynthesis within the canopy will differ and thereby significantly
affect the productivity of the plant stand. References 16-21 are suggested for further
reading on the relationship between canopy architecture and solar radiation.
Competition for photosynthetically active radiation in plant stands is perhaps the
fullest expression of how light-mediated processes are linked to the functional aspects of
the organism in its environment. The competitive capacity of an organism relative to
neighboring plants can have profound effects on its growth, survival, and productivity.
While it is not possible to truly separate competition among plants for soil nutrients or
water from competition for light, plants that out-compete their neighbors for critical
resources in light-limited environments may be able to occupy a higher position in the
canopy where irradiance is higher. This would be achieved through relatively faster stem
and leaf development. Interception of radiation by plants in the higher strata of the canopy
would thus inhibit the growth of plants beneath them. The ability to increase plant height
faster than neighboring plants and the efficient display of foliage for the interception of
visible solar radiation are essential characteristics for competitiveness in plant communit-
ies. (21)

5.3. HOW ORGANISMS PROTECT THEMSELVES FROM SOLAR UV

RADIATION

Solar radiation can affect organisms only if the radiation is absorbed. And since the
effect on light-mediated processes is highly wavelength-dependent, environmental factors
or aspects of the organism itself that alter the spectral distribution of radiation before it is
absorbed may significantly influence how well the organism functions in its environment.
The potential damage to nucleic acids caused by UV-B exposure, and the subsequent
effects on growth, reproduction, and survival, can exert a considerable influence on the
natural selection for plants and animals with effective protective mechanisms against UV-
B radiation damage. Mechanisms that prevent damage, or at least reduce it to levels
within the organism's capacity to repair damage, would therefore have adaptive signifi-
cance for the organism.

5.3.1. Terrestrial Plants

Higher plants are confronted with the problem of balancing the need to maximize the
interception of visible radiation for photosynthesis and to minimize the interception of
UV-B radiation and its possible damaging effects. Foliage display for maximal intercep-
tion of visible radiation may be optimum for photosynthesis, but also exposes leaves to
possible high rates of damage by UV-B radiation. Avoidance of UV-B radiation through
changes in leaf orientation relative to the sun is a possible adaptation, but would simul-
144 Chapter 5

taneously reduce the interception of visible radiation by the leaf. Adaptations that involve
changes in leaf anatomy, such as leaf surface pubescence or thick cuticular waxes, tend to
decrease UV -B penetration into the leaf primarily by increasing the reflectivity of the leaf
surface. However, these anatomical adaptations are generally not wavelength-selective
and attenuate UV and visible radiation to the same extent. This often results in relatively
low rates of photosynthesis, growth, and annual productivity for such species.
Although leaf pubescence or thick cuticular wax layers can increase the reflectance
of UV -B radiation from the top of these layers, cuticular waxes and cell wall constituents
are themselves relatively transparent to UV-B and visible radiation. These plant tissues,
therefore, do not substantially contribute to the attenuation capacity of the epidermis. (22)
The geometry of cell walls, including the amount of air space within the leaf, may
increase multiple reflection of radiation within the leaf, and thereby reduce to some extent
the UV-B radiation incident on sensitive targets within the leaf.
The highly pubescent leaves of Argyroxiphium sandwicense DC. (silversword) or the
thick wax-covered leaves of Dudleya brittonii Johansen (live-forever) are notable exam-
ples of species occupying relatively high irradiance environments. (23,24) The leaf surfaces
of these plants are highly reflective in the UV-B and visible wavebands. Additional
absorption of radiation by flavonoid compounds in the epidermal layer results in essen-
tially no UV-B penetration to the mesophyll. However; these highly reflective leaf sur-
faces also greatly reduce the penetration of visible radiation for photosynthesis, which
may be a significant disadvantage. Because this type of protective system requires a
relatively permanent modification of leaf anatomy, it is not responsive to the daily or
seasonal fluctuations in UV -B irradiance. A lack of flexibility and wavelength selectivity
limits UV -B acclimation potential and the efficacy of such leaf surface modifications as a
UV-B protective mechanism for plants. Furthermore, species with highly pubescent or
wax-covered leaves tend to be relatively rare and, in general, the green glabrous or
smooth leaves are the dominant leaf type found in nature.
A UV-B protective mechanism that is highly wavelength-selective for this waveband
and the capacity to respond and acclimate to changes in the UV -B radiation environment
would provide a great adaptive advantage for terrestrial plants. The presence of flavonoid
and related phenolic compounds in the upper epidermis of leaves may be an essential part
of such a protective mechanism. The absorption characteristics of this class of compounds
produce an effective attenuation of UV -B and shorter wavelength radiation (Fig. 5-1) and
no absorption of visible radiation (with the exception of anthocyanins, which strongly
absorb radiation between 520 and 560 nm). Several studies have clearly shown the
relationship between UV-B irradiation and the increased synthesis of these compounds.
The link between UV-B irradiation and flavonoid induction has been indicated both at the
biochemical level, as increased levels of the enzyme phenylalanine ammonia-lyase that is
involved in the production of flavonoids, and at the whole-plant level, as increased UV-B
absorbance in the epidermis and mesophyll. (25,26) This direct link between UV -B irradia-
tion and increased flavonoid concentrations forms a protective mechanism that is poten-
tially responsive to seasonal and daily changes in the UV -B radiation environment.
Absorption characteristics of leaves indicate that the epidermis functions as a highly
selective filter of solar radiation. Reflectance of UV-B from green glabrous leaves is
relatively low at 3-5%, and increases in the middle part of the visible waveband to
approximately 15-20%. The outer layers of the leaf can attenuate over 90% of the UV-B
Environmental Photobiology 145

radiation incident on the leaf surface. With additional UV -B absorption in the mesophyll,
essentially no UV -B radiation penetrates through the entire leaf, and so leaves are opaque
to this waveband. This substantial attenuation of UV-B radiation in leaf tissue is due to
flavonoid and related phenolic compounds. With increasing wavelength into the UV-A
and visible regions, the epidermis becomes increasingly more transparent (Fig. 5-1). The
epidermis has maximal transmittance of more than 80% in the visible waveband; meso-
phyll absorption is maximal in this region. In the near-IR region the entire leaf is both
highly reflective and transparent, with the combined reflectance and transmittance of
near-IR radiation as high as 95%. This characteristic has adaptive significance in reducing
the thermal energy load for the leaf. The combined optical characteristics of the leaf
epidermis and mesophyll thus produce a highly wavelength-selective filter well suited to
protect the leaf from injurious UV-B radiation and IR radiation heat stress, and to max-
imize the penetration of photosynthetically active visible radiation.
The reproductive parts of plants may also be sensitive targets for UV-B radiation.
However, the effect of this radiation on pollen and anthers, and the protective role of the
surrounding flower petals, has been difficult to study. This is related both to the small size
of these reproductive structures and to the difficulty in predicting the length of time that
pollen is exposed to UV -B irradiation in nature. Flower petals and anther walls are largely
opaque to UV -B radiation. (27) Pollen is therefore completely protected from UV -B ex-
posure while the flower is closed and, after the flower opens, remains protected by the
UV-opaque anther wall. When exposed to direct UV-B irradiance, pollen is rather sen-
sitive to injury, as shown by significant reductions in the germination percentages of the
exposed pollen. The length of time that pollen is exposed to UV-B irradiation in nature,
that period between dehiscence of pollen from the anther and the penetration into the
stigma, is the critical factor. This potential exposure period is relatively short, however,
and current studies suggest that under natural field conditions pollen of temperate latitude
plant species is not adversely affected by current levels of UV-B irradiance.

5.3.2. Animals and Aquatic Organisms

Unlike plants, animals do not require sunlight to synthesize their food and are
mobile. Continuous exposure to solar radiation throughout an animal's lifetime is not
required, as in the case of plants. Rather, exposure is regulated by behavior ot' the animal.
When exposed to solar radiation, however, animals are subject to the damaging effects of
UV-B radiation. The adaptations that have evolved in animals to cope with damaging UV-
B radiation are in some ways similar to those in plants, i.e., they are protective in that they
minimize UV radiation-induced damage. Two major protective mechanisms for reducing
the potential for UV -B-induced damage involve optical properties of the outer skin layers
and surface coverings. Several chromophores in the epidermal layer of human skin will
absorb UV-B radiation. These include aromatic amino acids, urocanic acid, nucleic acids,
and melanin. The latter chromophore gives human skin a dark-pigmented appearance, and
people with reduced skin pigmentation are generally more susceptible to sunburn and UV-
B related skin cancers than are highly pigmented people. This is especially the case for
albinos.(9) Thus, as in the case for the protective role of UV-absorbing flavonoid com-
pounds in the plant leaf epidermis, the UV -absorbing chromophores in the human skin
epidermis provide a degree of protection from damaging UV-B radiation. This protective
146 Chapter 5

mechanism must be highly effective because only about 5% of the UV radiation incident
on human skin is reflected. Thus, the major portion of this incident radiation penetrates
the skin. Surface coverings, such as hair or fur, can substantially reduce UV radiation
incident on the skin. For human populations, clothing provides a suitable covering to
reduce the exposure of skin to sunlight. However, the recent rise in UV -related skin
cancers for light-skinned people in, for example, the United States of America, Europe,
and Australia highlight the need to modify recreational or work activities that involve
extensive exposure of skin to sunlight. In the past, exposure of the skin to sunlight was
restricted by the use of more clothing. The function of fur is primarily related to its
thermal properties, but it can also protect against physical damage to the skin and against
insects. In addition, the color or pigmentation of fur often functions as camouflage. While
these may be the primary functions of fur, a skin covering of fur can reduce the penetra-
tion of UV radiation to the skin, and thereby provide protection from damaging UV
radiation. The white fur of polar bears and baby harp seals is an interesting exception to
this. (28) The fur of these animals is actually composed of hollow, transparent hairs that are
entirely lacking in pigmentation. The rough inside surface of the hollow hairs reflects
visible radiation and thus appears white. Furthermore, only UV radiation is "funneled"
to the skin through the hollow-cored hairs. (28) The result is that the white fur of these
animals appears black when photographed with UV -sensitive film. Because of the low
solar angles and the relatively high concentrations of stratospheric ozone at arctic and
antarctic latitudes (Section 5.2.2), UV-B irradiance is relatively low in polar habitats.
Therefore, the present or predicted enhanced levels of UV-B radiation will probably not
be deleterious for arctic populations of polar bears and harp seals.
The mobility of animals is another effective mechanism by which these organisms
can reduce exposure to solar radiation. It is a mechanism of avoidance, and primarily
involves behavioral control over the duration of exposure and the time of day that an
individual is exposed. For humans, the desire for outdoor recreational activities has
outweighed caution against too much exposure. Avoidance of UV-B exposure by animals
is an indirect benefit of behaviors that minimize exposure to the midday sun and high
temperatures, such as nocturnal or early morning and evening feeding habits. The com-
bined affects of skin pigmentation, hair and fur, mobility, and behavior thus form an
effective defense against damaging UV radiation.
The primary form of protection from UV -B radiation for aquatic organisms involves
mobility, movements that change the position of the organism in the vertical profile of the
water column. As discussed above in Section 5.2.3, water circulation patterns or turbulent
mixing of water surface layers can transport planktonic communities to different locations
in the water. Also, motile and phototactic algae have the ability to migrate nearer to or
further from the water surface. These movements will directly determine the UV-B
exposure time for these organisms. Since with few exceptions flavonoid compounds are
not present in algae, these organisms lack the protective pigment system of higher vas-
cular plants. Regulation of UV-B exposure time through movement in the vertical profile
of their water habitat is thus significant for aquatic organisms. Because the water medium
itself acts as a filter of UV -B radiation (Fig. 5-3), aquatic organisms, particularly those in
estuaries or zones of upwelling where mixing of dissolved organic compounds and other
impurities may occur, are to some extent sheltered from the full intensity of UV-B
radiation.
Environmental Photobiology 147

5.4. BIOLOGICAL CONSEQUENCES OF STRATOSPHERIC OZONE

DEPLETION

A significant environmental problem of global magnitude is the prediction that

atmospheric pollutants will result in a partial depletion of the stratospheric ozone layer.
The release of pollutants such as chlorofluorocarbons (CFCs) and nitrous oxide (N 20)
into the atmosphere is predicted to reduce the equilibrium ozone column thickness at all
latitudes in the coming decades, i.e., the increased concentration of these pollutants will
increase the rate of ozone destruction relative to the rate at which ozone is naturally
produced.(8) Ozone is destroyed in chemical reactions involving oxides of nitrogen (e.g.,
N0 2), which are chemically active molecules formed from the photooxidation of nitrous
oxide (N20). The latter compound is produced naturally by bacteria in the soil and water.
Additional sources of N20 related to human activities include high-altitude aircraft ex-
haust, human and animal waste, and industrially produced nitrogen fertilizers. These
additional new sources of N20 have significantly increased the atmospheric concentration
of this molecule by 2.7% from 1964 to 1981, a trend that is expected to continue into the
future. (9) Chlorofluorocarbons (e.g., CF2Cl 2 and CFCI 3 ) are a commercially produced
class of compounds found in refrigeration systems and propellants. When these com-
pounds are decomposed by solar radiation in the upper atmosphere, atomic chlorine reacts
with ozone in chemical reactions similar to those involving nitrous oxides. (8) Other
potential sources of chlorine atoms include methyl chloroform (CH 3 CCI 3 ), an industrial
solvent, methyl chloride (CH 3 Cl), and carbon tetrachloride (CCI4 ).
Computer simulation models of atmospheric chemistry represent our primary source
for the prediction of ozone depletion rates and tend to be very complex. The degree of
ozone depletion predicted varies with input variables such as present and future pollutant
production rates, their rate of diffusion to the stratosphere, the length of time these
compounds are resident in the stratosphere, the photochemical reactions and their rates,
and the behavior of the stratosphere. Because of the complex nature of these interacting
variables as well as incomplete knowledge of atmospheric processes, computer simulation
models often vary in their predictions for the expected depletion rate of ozone. This has
resulted in predictions of stratospheric ozone depletion ranging from 5 to 20% for the
effect of CFCs alone. Refinements in the models have improved the estimates of ozone
depletion and, as reported by the National Research Council of the United States National
Academy of Sciences, the degree of ozone depletion from present levels is expected to be
5-9% due to CFCs alone and 10-16% due to a doubling of N20' (9) Such reductions in the
ozone layer will not only increase UV -B irradiance reaching the earth's surface but should
also shift the terrestrial solar spectrum slightly toward shorter, more photobiologic ally
effective wavelengths. Since this spectral shift would increase the overlap of UV-B
wavelengths with the absorption spectra of nucleic acids and proteins, the increase in UV-
B irradiance is expected to have biologically significant effects. When the combined
effects of a 15% decrease in ozone at temperate latitudes, the spectral shift toward shorter
wavelength, and increased UV-B irradiance weighted for DNA damage are considered,
the potential future radiation environment could be 44% greater in effective UV-B radia-
tion (see equation in Section 5.2.2).(6)
The discovery in 1985 of an ozone depletion zone over the continent of Antarctica,
an ozone "hole" that appears every antarctic spring, has renewed concern over the
148 Chapter 5

potential for ozone destruction and the intensification of UV -B irradiance. Observations

over Antarctica indicate that ozone concentrations during the spring season have declined
by up to 40% over the past decade. (29) While the cause of this substantial ozone depletion
zone above Antarctica is not yet fully understood, it is believed to involve a combination
of stratospheric cooling and circulation patterns and a higher abundance of chlorinated
molecules in this polar region. Public concern over this global environmental problem has
resulted in an international agreement, signed in 1987 by 23 nations and sponsored by the
United Nations Environment Program, for a 50% reduction in CFC production by 1999.

5.4.1. The Response of Individual Plant Species and the Consequences for
Ecosystem Productivity
The survival of organisms in the predicted enhanced UV-B radiation environment
may depend greatly on the effectiveness of their protective mechanisms and the efficiency
of mechanisms that repair UV-B induced damage. Effective protection from UV radia-
tion-induced damage is particularly important for plants, since these organisms depend on
solar radiation for photosynthesis and cannot avoid exposure to UV-B radiation. In an
enhanced UV-B radiation environment, the various mechanisms that repair UV-B-in-
duced damage to nucleic acids (Chapter 4) may not be able to keep pace with the rate at
which damage occurs. Therefore, organisms with the capacity to increase their attenuation
of UV-B radiation, such as plants with a relatively opaque leaf epidermis, may suffi-
ciently reduce the rate of damage below the point at which repair mechanisms are satura-
ted.
An increase in the attenuation capacity of the epidermis, through the synthesis of
increased amounts of flavonoid and related phenolic compounds, does occur under ar-
tificially enhanced UV-B environments. For example, flavonoids in the epidermal tissue
of Oenothera stricta Ledeb. (evening primrose) increased by up to 100% after exposure to
an enhanced UV -B environment. (26) This radiation regime also significantly reduced UV-
B transmittance through intact epidermal tissue by as much as 33%. Transmittance of
visible radiation through the epidermis was not affected. However, once the daily UV-B
exposure exceeded the highest levels normally found in this species' field habitat, photo-
synthesis was significantly decreased. This suggests that an upper threshold of exposure
exists above which the capacity of protective mechanisms to reduce injury is exceeded.
Oenothera stricta is native to South America at low-elevation, temperate latitude sites,
but has in recent years invaded high irradiance habitats in the Hawaiian Islands. The
ability of this species to colonize such habitats is evidence that it can acclimate to current
high UV-B irradiance levels. Whether sufficient acclimation capacity exists within this
species, or others like it, to tolerate the enhanced UV -B environment of the future is still
unclear.
The capacity for acclimation to higher UV -B irradiance levels varies among plant
species, and agronomic plants tend to be more sensitive to enhanced UV-B irradiation
than wildland plants. These conclusions, however, are based on studies in which filtered
fluorescent lamps were used to produce an enhanced UV-B environment. The lamp/filter
systems used do not perfectly simulate the natural spectral distribution of UV radiation in
sunlight, and plants studied in greenhouse or controlled environment chambers tend to be
more sensitive than field-grown plants to UV irradiation. This greater UV -B sensitivity
Environmental Photobiology 149

Fig. 5-4. The degree of UV transmittance through UV-B UV-A UV-8 UV-A
leaf epidermal tissue of Pisum sativum "Alderman" 80 Oena/hera Sfn~~~ _ --- Pisum solivum
(pea) and Oenothera stricta (evening primrose). The /
/ Haleakala
latter species has colonized high-irradiance habitats I
on the Haleakala Crater of the Hawaiian island of
Maui, and P. sativum is a common agronomic spe-
~
..
<J
60
/
/
I (extracted
epidermis-)
Logan, Utah
No UV·8 I'
, .
-----
c: (greenhouse) I •
/
cies. For plants examined under greenhouse condi- " 40 / / /

..
;:
tions at Logan, Utah, the reduction in epidermal E /
/
c:
transmittance for O. stricta was evident after 11 ~
days of a biologically effective UV-B exposure ap- f- 20 NoU

proximately 20% less than that experienced on Peru (field)

cloudless days in the Hawaiian habitat. Field-grown o~~~~~~===c==~~
plants exhibited the lowest level of epidermal UV-B 280 320 360 280 320 360 400
transmittance, and thus the greatest protection from Wavelength (nm)
UV-induced damage. After flavonoid and related
phenolic compounds were extracted from the epider-
mis, the capacity of this tissue to attenuate UV radiation was substantially reduced. A similar response to UV-B
irradiation was evident for P. sativum, a plant species relatively sensitive to enhanced UV -B radiation. Exposure
to UV-B radiation in controlled experiments reduced epidermal transmittance but, as with O. stricta, the greatest
attenuation of UV radiation occurred in field-grown plants. The high degree of plasticity in epidermal transmit-
tance exhibited by P. sativum may explain why this temperate latitude cultivar is successfully cultivated in high-
elevation fields of the Peruvian Andes, where biologically effective UV-B radiation is up to 30% higher than that
experienced by the greenhouse plants. (Adapted from Refs. 23 and 26.)

may in part be attributed to the lower level of UV -A and visible radiation available in
controlled environments for UV repair mechanisms requiring light (i.e., photoreactiva-
tion; see Chapter 4). Controlled experiments can, however, be used to detect differences
in UV -B radiation sensitivity and their capacity to acclimate to higher UV -B levels among
species. Based on reductions in photosynthesis and biomass after UV -B irradiation in such
experiments, Rumex patientia L. (dock), Pisum sativum L. (pea), Cucurbita pepo L.
(squash), and Glycine max (L.) Merr (soybean) are among the more UV-sensitive spe-
cies. (30) Some species under field conditions exhibit considerably greater capacity for
acclimation to UV-B radiation than when examined under experimental conditions. As
illustrated for O. stricta and P. sativum "Alderman" in Fig. 5-4, the degree of UV-B
transmittance is reduced in response to intensified UV -B irradiation under experimental
conditions. The reduction in epidermal UV-B transmittance, however, is substantially
greater in the plants grown in their high UV-B irradiance field habitats.
The consequences of an enhanced UV-B environment may be expressed in more
subtle ways than what has been demonstrated for individual plants cultivated under the
artificially enhanced UV-B regimes. Rarely in nature are organisms exposed to only one
environmental stress factor at a time. Usually several factors act on an organism to affect
its growth, survival, and reproductive capacity in the ecosystem. How the whole organism
is affected depends, however, to a great extent on factors secondary to the initial influence
of radiation on light-mediated processes. Growing season temperatures, solar radiation,
water availability, and nutrient supply are among the more significant environmental
factors that influence organisms. Particularly important is the interplay between these
factors and the developmental stage of the organism. The influence of interacting factors
such as UV-B irradiance, water and nutrient supply, and air temperature on a plant, for
example, would most likely be different during the seedling stage than at the mature plant
150 Chapter 5

or flowering stage. In addition to the effects of environmental factors, the availability of

resources in the environment that are critical to the organism as well as competition
among organisms for these resources should have a significant influence on the ultimate
response of organisms in an enhanced UV -B radiation environment. Obtaining answers to
how plants in nature will respond to an intensified UV -B environment requires multifactor
experiments. However, executing such multifactor experiments so that reliable predic-
tions can be made about plant behavior under a variety of situations has been quite
difficult.
Ecosystems, because of their large scale and complexity, render extrapolations from
experiments on individual plants to their behavior in whole ecosystems tenuous. Although
predictions about ecosystem level responses may be based on the relative sensitivity of
individual plant species, e.g., reduced photosynthesis and growth in some sensitive
species and acclimation in others, these predictions would have a large level of uncertain-
ty. Differences in UV-B sensitivity among species may indicate that overall primary
productivity of natural terrestrial ecosystems will not be significantly reduced under an
intensified UV -B environment. Rather, this future radiation environment may result in
more subtle changes in relative species composition and abundance. Growth and produc-
tivity of species with moderate to high UV-B sensitivity may be reduced so that they are
not as competitive with species more resistant to UV radiation-induced damage. Species
in this latter category could thus become more dominant and abundant in an ecosystem.
The actual situation is considerably more complex since various stages of plant growth
can be more sensitive to UV-B radiation than others, and will also respond differently to
environmental factors at each of these stages. Agricultural ecosystems are composed of
monocultures, and are thus generally less complex than natural ecosystems. Although
differences in soils, water and nutrient supply, species composition, and genetic variation
are minimized in agricultural systems, experimentation difficulties encountered with these
systems are similar to those with plants in natural ecosystems. Simulation of actual field
conditions in experiments has been difficult to achieve with the artificial UV -B radiation
systems that have been used. However, the yields of some economically important crops
may be reduced in an intensified UV -B environment, an assessment based largely on the
response of agronomic plants in greenhouse and controlled environment chambers. In
such experiments, G. max, P. sativum, Vigna unguiculata (L.) Walp. (cowpea), Phaseo-
Ius vulgaris L. (bean), and Brassica oleraceae L. "acephala" (collard) are among the
crops that have exhibited reductions in photosynthesis, leaf area, or productivity. (30)
Species that have been found to be relatively more resistant to UV-B radiation injury
include such important crops as Hordeum vulgare L. (barley), Oryza sativa L. (rice),
Triticum aestivum L. (wheat), and Zea mays L. (com). Of all the species examined in
numerous studies, approximately 20% have been found to be very sensitive to UV-B
irradiation, 50% exhibited moderate sensitivity, and 30% were relatively resistant to UV-
B radiation injury. (30) There are substantive inconsistencies among various studies in
regard to the sensitivity of species to UV -B radiation, however. The frequent contradicto-
ry findings among studies may stem from the lack of uniform experimental techniques,
variations in the UV radiation lamp/filter systems used to simulate an enhanced UV-B
radiation regime, and differences in UV-B radiation exposure, plant cultivation condi-
tions, and cultivars. Considerable caution should thus be used in the interpretation of the
results from studies on the UV-B radiation sensitivity of crop species. However, there is
Environmental Photobiology 151

wide agreement that species and varieties exhibit substantial differences in their response
to UV-B radiation. Important sources of variation include genetic differences among
species and among cultivars that determine the capacity for UV-B attenuation in the
epidermis and the efficacy of mechanisms for the repair of UV-B radiation injury. The
phenological stage at which the stress of enhanced UV -B radiation is initiated may also be
a significant factor in the apparent sensitivity of a species to UV -B radiation. Current
research sponsored by the United States Environmental Protection Agency on the re-
sponse of representative crops plants (e.g., wheat, corn, and soybeans) to enhanced UV-B
radiation should significantly increase our understanding of how agricultural ecosystems
will respond under the predicted intensified UV-B radiation environment of the future.

5.4.2. Effects on Terrestrial and Aquatic Animal Populations

The consequences of an intensified UV-B environment for animal populations in
terrestrial ecosystems is difficult to assess since this aspect has received little study.
Because of the protective pigments and fur, and particularly because of the behavioral
mechanisms that reduce exposure to damaging UV radiation, animals in natural terrestrial
ecosystems may be less affected by increased UV-B radiation than plants. The informa-
tion available for domesticated animals suggests that the food yield from domestic live-
stock will not be significantly reduced even at the most extreme predicted levels of ozone
depletion. (9) Also, the degree of UV radiation-induced damage found in experiments with
cattle, where only one breed has been found susceptible to UV-B-caused cancer eye,<9)
has not warranted significant concern about the deleterious consequences of increased
UV-B radiation for livestock.
There has been considerable concern, however, over the possible impact of UV-B
intensification on humans. Noncancerous and acute effects of UV-B radiation exposure
include sunburn and inflammation of the cornea, and in some individuals UV -B exposure
may alter aspects of the immune system. Long-term exposure to UV-B radiation, es-
pecially in light-skinned Caucasians, is either the direct cause of nonmelanoma skin
cancers (e.g., basal cell and squamous cell cancers) or a contributing factor in the case of
skin melanoma. (9) The incidence of both skin melanoma and other UV-B-induced skin
cancers is correlated with the increase in solar irradiance that occurs from northern to
equatorial latitudes. This suggests that UV -B radiation is a contributing causal factor.
Also, people whose occupations require day-long exposure to sunlight tend to exhibit
higher rates of UV -B related skin cancers on exposed body areas such as the head and
neck. Although the incidence of skin cancer is predicted to increase significantly in an
enhanced UV-B environment, it is unclear how changes in human behavior that reduce
exposure time will influence the incidence of such skin diseases.
In aquatic ecosystems the effect of higher levels of UV-B radiation will to a large
extent depend on the degree of UV -B attenuation by water and the exposure time for the
organism. Ecosystem level research \s difficult and predictions for the effects on eco-
system productivity currently rely on studies performed under controlled conditions. Such
studies show that even current levels of UV-B radiation are sufficient to depress primary
productivity near the water surface of marine ecosystems. (9) Because phytoplankton form
the base of food chains in aquatic ecosystems, any significant reduction in the productivity
of phytoplankton is expected to profoundly affect organisms at higher levels in the food
152 Chapter 5

chain, such as populations of anchovy, herring, shellfish, and crustaceans. The dynamic
nature of aquatic ecosystems, where UV-B attenuation by water (Fig. 5-3) is highly
variable and where variations in the vertical position of planktonic organisms can consid-
erably affect their level of UV-B exposure, results in large uncertainties for predictions of
ecosystem response to enhanced UV -B radiation.

5.5. CONCLUSIONS

The effects of light-mediated processes on the growth and development of an orga-

nism are highly influenced by environmental factors and how the organism responds to
environmental stress. In nature, organisms are simultaneously exposed to several interact-
ing environmental factors, e.g., solar radiation, water and nutrient availability, and tem-
perature, all of which can exhibit considerable daily and seasonal variation. The response
of an organism to these environmental factors also changes during its lifetime. Therefore,
experiments in environmental photobiology must consider both the dynamic nature of the
environment and the plasticity of an organism's response to these factors. Although
experiments that closely reflect field conditions are desirable, they tend to increase in
complexity as the number of environmental factors considered increases. This complexity
presents the environmental photobiologist with formidable problems in experimental de-
sign. Moreover, as the systems studied become progressively more complex from the
cellular level, to the whole organism, to the ecosystem, the experiments also tend to
increase in difficulty and expense. The degree of uncertainty in the results may also
increase as the number of uncontrolled variables increases at, say, the ecosystem level.
Because of the increased complexity of experiments involving mUltiple factors,
studies on the how the effects of light-mediated processes are influenced by the environ-
mental factors with which an organism interacts have tended to be single-factor experi-
ments. The effects oftemperature, nutrient or water stress, and radiation on photosynthesis,
for example, have generally been examined one factor at a time. This single-factor
approach has also been the primary experimental procedure used to examine the response of
organisms to enhanced UV-B radiation. While such experimental designs do indeed
increase our understanding of the mechanisms of processes involving light, and how the
effect of these processes are affected by a particular environmental factor, the results often
do not allow extrapolation to field conditions. Reliable extrapolations to field situations are
particularly crucial for anticipating the consequences of enhanced UV -B radiation for
natural and agricultural ecosystems. Therefore, the design of experiments that permit
predictions on how organisms would be affected under field conditions, where many
environmental factors interact and vary, poses a significant challenge to the field of
environmental photobiology.

5.6. REFERENCES

1. P. R. Gast, A. S. Jursa, J. Castelli, S. Basu, and J. Aarons, Solar electromagnetic radiation, in: Handbook
of Geophysics and Space Environments (S. L. Valley, ed.), pp. 16-1-16-38, U.S. Air Force Cambridge
Research Laboratories, McGraw-Hill, New York (1965).
Environmental Photobiology 153

2. L. R. Koller, Ultraviolet Radiation, Wiley, New York (1965).

3. 1. Jagger, Solar-UV Actions on Living Cells, Praeger, New York (1985).
4. R. B. Setlow, The wavelengths in sunlight effective in producing skin cancer: A theoretical analysis, Proc.
Natl. Acad. Sci. USA 71, 3363-3366 (1974).
5. J. Ross, Radiative transfer in plant communities, in: Vegetation and the Atmosphere, Vol. I O. L.
Monteith, ed.), pp. 13-55, Academic Press, London (1975).
6. M. M. Caldwell, R. Robberecht, and W. D. Billings, A steep latitudinal gradient of solar ultraviolet-B
radiation in the arctic-alpine life zone, Ecology 61, 600-611 (1980).
7. M. Iqbal, An Introduction to Solar Radiation, Academic Press, New York (1983).
8. G. Brasseur and A. De Rudder, Agents and effects of ozone trends in the atmosphere, in: Stratospheric
Ozone Reduction, Solar Ultraviolet Radiation and Plant Life (R. C. WOITest and M. M. Caldwell, eds.),
pp. 1-28, Springer-Verlag, Berlin (1986).
9. National Research Council, National Academy of Sciences, Causes and Effects of Stratospheric Ozone
Reduction: An Update, National Academy Press, Washington, D.C. (1982).
10. B. Lowry, D. Lee, and C. H6bant, The origin ofland plants: A new look at an old problem, Taxon 29,183-
197 (1980).
II. J. W. McClure, Physiology and functions of flavonoids, in: The Flavonoids (J. B. Harborne, T. J. Mabry,
and H. Mabry, eds,), pp. 970-1055, Academic Press, New York (1975).
12. T. Swain, Evolution of flavonoid compounds, in: The Flavonoids (1. B. Harborne, T. J. Mabry, and H.
Mabry, eds.), pp. 1096-1129, Academic Press, New York (1975).
13. R. G. Wetzel, Limnology, 2nd ed., Saunders, Philadelphia (1983).
14. H. R. James and E. A. Birge, A laboratory study of the absorption of light by lake waters, Trans. Wis.
Acad. Sci. Arts Lett. 31, 1-154 (1938).
15. D. M. Gates, H. J. Keegan, J. C. Schleter, and V. R. Weidner, Spectral properties of plants, Appl. Opt. 4,
11-20 (1965).
16, J. Ross, The Radiation Regime and Architecture of Plant Stands. Dr. W. Junk, The Hague (1981).
17. M. M. Caldwell, T. J. Dean, R. S. Nowak, R. S. Dzurec, and J. H. Richards, Bunchgrass architecture,
light interception, and water-use efficiency: Assessment by fiber optic point quadrat and gas exchange,
Oecologia 59, 178-184 (1983).
18. R. L. McCown and W. A. Williams, Competition for nutrients and light between the annual grassland
species Bromus mollis and Erodium botrys, Ecology 49,981-990 (1968).
19. H. M. Rawson, R. L. Dunstone, M. J. Long, and 1. E. Begg, Canopy development, light interception and
seed production in sunflower as influenced by temperature and radiation, Aust. 1. Plant Physiol. 11, 255-
265 (1984).
20. J. Warren Wilson, Analysis of growth, photosynthesis and light interception for single plants and stands,
Ann. Bot. 48, 507-512 (1981).
21. M. M. Caldwell, Plant architecture and resource competition, in: Potentials and Limitations of Ecosystem
Analysis (E. -D. Schulze and H. Zwolfer, eds.), pp. 164-179, Springer-Verlag, Berlin (1987).
22. A. Frey-Wyssling, The Plant Cell Wall, Gebriider Borntraeger, Berlin (1976).
23. R. Robberecht, M. M. Caldwell, and W. D. Billings, Leaf ultraviolet optical properties along a latitudinal
gradient in the arctic-alpine zone, Ecology 61, 612-619 (1980).
24. T. W. Mulroy, Spectral properties of heavily glaucous and non-glaucous leaves of a succulent rossette-
plant, Oecologia 38, 349-357 (1979).
25. E. Wellmann, UV dose-dependent induction of enzymes related to flavonoid biosynthesis in cell suspension
cultures of parsley, FEBS Lett. 51, 105-107 (1975).
26. R. Robberecht and M. M. Caldwell, Protective mechanisms and acclimation to solar ultraviolet-B radiation
in Oenothera stricta, Plant, Cell Environ. 6,477-485 (1983).
27. S. D. Flint and M. M. Caldwell, Influence of floral optical properties on the ultraviolet radiation environ-
ment of pollen, Am. 1. Bot. 70, 1416-1419 (1983).
28. R. E. Grojean, J. A. Sousa, and M. C. Henry, Utilization of solar radiation by polar animals: An optical
model for pelts, Appl. Opt. 19, 339-346 (1980).
29. R. S. Stolarski, The Antarctic ozone hole, Sci. Am. 258, 30-36 (1988).
30. A. H. Teramura, Effects of ultraviolet-B radiation on the growth and yield of crop plants, Physiol. Plant.
58, 415-427 (1983).
154 Chapter 5

31. T. J. Mabry, K. R. Markham, and M. B. Thomas, The Systematic Identification of Flavonoids, Springer-
Verlag, Berlin (1970).
3Z. M. M. Caldwell, Plant response to solar ultraviolet radiation, in: Plant Physiological Ecology. I. En-
cyclopedia of Plant Physiology, Vol. lZA (0. L. Lange, P. S. Nobel, and C. B. Osmond, eds.), pp. 169-
197, Springer-Vedag, Berlin (1981).
 6
Photomedicine

6.1. Introduction ..................................................................... 156

6.2. Optical Properties of the Skin ....................................................... 156
6.3. Acute Effects of Sunlight on the Skin ................................................ 158
6.3.1. Sunburn................................................................... 158
6.3.2. Microscopic Anatomy ....................................................... 160
6.3.2.1. Light Microscopy ................................................... 160
6.3.2.2. Electron Microscopy. . . . . . . . . . . . . . ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 160
6.3.3. Macromolecular Synthesis and Mitosis. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 161
6.3.4. DNA Repair ............................................................... 162
6.3.5. Melanogenesis ............................................................. 163
6.4. Chronic Effects of Sunlight on the Skin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 165
6.4.1. Solar (Actinic) Degeneration .......................................... , . . . . . .. 165
6.4.2. Carcinogenesis ............................,..............................., 166
6.4.2.1. Human Skin Cancer Formation .............. ,......................... 166
6.4.2.2. Experimental Carcinogenesis .......................................... 168
6.4.2.2a. Quantitative Investigations ................................... 168
6.4.2.2b. Qualitative Investigations .................................... 169
6.5. Adverse Cutaneous Reactions to Sunlight ........ ,.................................... 171
6.5.1. Photosensitivity ............................................................ 171
6.5.1.1. Phototoxicity ....................................................... 171
6.5.1.2. Photoallergy ............ , .................... , ...................... 174
6.5.2. Clinical Problems .. , ...................... , ............. , ................. ,. 175
6.5.2.1. Lack or Loss of Protection ........................ , ................ , .. 176
6.5.2.2. Photosensitized Reactions ......................... , ................. ,. 177
6.5.2.3. Protection Normal and No Known Photosensitizer. . . .. . . . . . . . . . . . . . . . . . . .. 178
6.6. Beneficial Effects of Sun and/or Artificial Light Energy ................................. 178
6.6.1. Vitamin D Formation ......................... , . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 178
6.6.2. Treatment of Skin Diseases ...... , ...................................... , . . . .. 179
6.6.2.1. Vitiligo ............... , ............. , .. , ................. , ... , ..... 179
6.6.2.2. Psoriasis .......... , ... , ......... , ........................ , ... , . . . .. 180
6.6.2.3. Herpes Simplex Infection ................. , ........................ ,.. 182
6.6.2.4. Photochemotherapy of Tumors ... , .. , . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 182
6.6.3. Treatment of Systemic Diseases (Neonatal Jaundice) .............................. 183
6.7. Photoprotection................................................................... 184
6.7.1. Sunscreens .............................................. , ................. 184
6.7.2. Physical Factors ............................................................ 185
6.7.3. Systemic Chemical Agents ................................ ,.................. 185
6.7.3.1. Antimalarials ...................... , ....... , . '" .................... 185
6.7.3.2. j3-Carotene ........... , ............ , ................................ 186
6.7.3.3. Systemic Psoralens .................................................. 186

John H. Epstein • Department of Dermatology, University of California School of Medicine,

San Francisco, California 94143-0536.

155
156 Chapter 6

6.7.4. Topical Psoralens ........................................................... 186

6.7.5. Quick-Tanning Preparations .................................................. 186
6.8. Photoimmunology ................................................................ 187
6.8.1. Immediate Hypersensitivity ................................................... 187
6.8.2. Delayed Hypersensitivity (DHS) Cell-Mediated Immune Response (CMI) ............ 187
6.8.3. Lymphocytes and Nonionizing Radiation .............................. , ....... ,. 188
6.9. Indirect Effects of Light ........................................................... 188
6.10. Ultraviolet Radiation Effects on the Eye .............................................. 189
6.11. Conclusion ...................................................................... 189
6.12. References ......... ,............................................................ 190

6.1. INTRODUCTION

The medical aspects of the science of photobiology are among the most important parts of
this discipline, since they are concerned primarily with direct effects of nonionizing
radiation on humans. Two organ systems are directly affected by sunlight: the eye and the
skin. The eye is discussed in Chapter 9. Therefore, the present discussion will be con-
cerned primarily with cutaneous reactions to the sun.(l-6)
The solar radiation that reaches the earth's surface ranges from around 290 nm (or
perhaps 288 nm) in the UV spectrum through the visible, infrared, and well beyond.
Almost all of the photobiological reactions that occur in the skin are induced by radiation
between 290 (or 288) and 320 nm (UV-B). These are the wavelengths that inhibit mitosis,
DNA, RNA, and protein synthesis, make vitamin D, induce skin cancer, stimulate pig-
ment formation, and produce the erythema response that we call "sunburn." These
wavelengths comprise what is generally termed the "sunburn spectrum." The longer
wavelength UV radiation between 320 and 400 nm (UV-A) does cause a few minor
photobiological effects such as immediate pigment darkening (IPD) and the immediate
erythema response. Major biological effects of UV -A rays include their accentuation of
the acute injury induced by, and enhancement of, the carcinogenic potential of the UV-B
spectrum. Furthermore, UV -A is responsible for the vast majority of the reactions pro-
duced by exogenous photosensitizers. An excellent review ofUV-A effects is presented in
Ref. 7. The rest of the sun's spectrum has little specific effect on the skin, though infrared
radiation can produce a thermal bum and can enhance photobiological responses initiated
by UV radiation.

6.2. OPTICAL PROPERTIES OF THE SKIN(I·6:pp.147-194)

For photobiological reactions to occur, three components are necessary: the biolog-
ical system, the radiation, and a radiation absorber in the biological system. The biolog-
ical system with which we are concerned is the skin (Figs. 6-1, 6-2). The radiation
emanates from the sun. The radiation-absorbing molecules in skin generally remain to be
determined; however, at least one important molecule would appear to be DNA.
If a photoactive molecule is present in or on the dead cells of the stratum corneum, it
will most likely act as a protective filter. If the molecule is in the Malpighian or ger-
minative layers of the epidermis, it may initiate a photoreaction after abosrbing radiation
Photomedicine 157

Fig. 6-1. Structure of skin. The skin consists of three

main layers. The epidermis is the outermost layer,
which is separated from the dermis by a basement mem-
brane (BM). The dermis consists primarily of connec-
tive tissue, mostly collagen, and some elastic tissue pro-
duced by fibrocytes. The dermis also contains blood and

00
lymph vessels, hair follicles, sweat glands (SW), Hair ide \
sebaceous glands (F), muscles, and nerves. Below this
is a fat layer not shown on the diagram.

of the proper wavelength. It is clear that the radiation must penetrate to the site of the
absorbing molecule for the reaction to occur. The penetration of radiation is governed, in
part, by the optical properties of the skin.
Skin reflects a large amount of incident visible and near-infrared radiation. White
skin reflects these radiations, as well as UV-A radiation, much more effectively than
black skin. Very little reflection of UV radiation shorter than 320 nm occurs with either
type of pigmentation.
Scattering and absorption play a greater role in the attenuation of sunburn radiation
penetration than does reflection. The flattened cells of the stratum corneum, melano-
somes, nucleic acids, proteins, lipids, histidine, urocanic acid, peptides, cholesterol, and
the like limit penetration of this UV radiation by both scattering and absorption.
Because of the importance of the shorter wavelength UV radiation in the production
of photobiological reactions, extensive transmittance studies have been conducted at these
wavelengths. Although these investigations have produced a number of conflicting re-

Fig. 6-2. Structure of epidermis. The epidermis consists of

four layers. The basal layer (B) is composed of a single or
double layer of germinative cells where cell division occurs
for the replacement of the epidermis. The Malpighian layer
(M) consists of multiple squamous or prickle cells (ker-
atinocytes) with intercellular bridges. These cells emanate
from the basal layer , and eventuate in the outer dead stratum
corneum layer. The Malpighian cells become the granular
layer (G) where specific proteins are formed, presumably
necessary for the formation of keratin. The cells then lose
their nuclei, contain large amounts of keratin, and form the
protective stratum corneum (C). They are then shed into the
environment. The melanocytes (P) are dendritic cells lo-
cated in the basal layer. Their dendritic (armlike) processes
extend up into the Malpighian layer. These cells form the
pigment melanin, which they pass into the Malpighian cells
through their dendrites. The melanocytes and associated
Malpighian cells compose the epidermal melanin unit.
158 Chapter 6

sults, it is generally agreed that the bulk of incident radiation in the sunburn or UV-B
spectrum (290-320 run) is absorbed in the stratum corneum. In Caucasian skin, at least
20% of this radiation reaches the Malpighian cells, and probably 10% penetrates to the
upper dermis. The stratum corneum of black skin absorbs a greater amount of this energy
due to the presence of melanin. A large proportion of the longer wavelength UV radiation
(UV-A) and of visible radiation penetrates into, and can be absorbed by, photosensitizers
in the dermis.
With the absorption of this energy, an excited state is induced in the chromophore.
The excited state may be deactivated by a number of methods including thermal decay or
emission of light energy, usually of a longer wavelength than that absorbed (Chapter 1).
Cutaneous photosensitivity reactions are brought about either by direct alteration in the
absorbing molecule or by the transference of this energy to other molecules or cellular
components such as membranes, mitochondria, nucleic acids, and the like.

6.3. ACUTE EFFECTS OF SUNLIGHT ON THE SKIN

6.3.1. Sunburn(!.3,6:pp.219-260)

The sunburn reaction in the skin is by far the most common adverse effect produced
by sunlight. The cutaneous changes induced by the erythemogenic radiation depend on the
amount of radiation, the degree of melanin pigmentation, and the thickness of the stratum
corneum. As would be expected, the morphological and microscopic responses are less
intense with less radiation and/or more protection. However, sunburn will occur even in
pigmented skin if exposed to enough radiation of the proper wavelength.
Erythema (reddening of the skin) is the most visually prominent aspect of the sun-
burn response. It appears in a biphasic pattern consisting of an immediate faint erythema
that occurs during exposure and disappears shortly thereafter. The delayed response
appears 2-4 h later, reaches a peak in 14-20 h, and persists for 24-48 h. The injury is
then followed by desquamation of the dead epidermal cells.
The action spectrum for sunlight-induced erythema is confined primarily to radiation
between 288 or 290 and 320 nm (Fig. 6-3). The standard erythemal curve described by
Coblentz and Stair in 1934 showed a maximum efficiency at 297 nm, and minimum
efficiency at 280 and 320 nm. Subsequent studies have provided various results with the
peak erythemogenic potential ranging between 290 and 294 nm. In addition, using param-
eters of efficiency and available energy, the most erythemogenic wavelengths in the sun's
spectrum lie between 305 and 308 nm.o:pp·184-187) A mild erythema can be produced by
radiation between 320 and 400 nm, but this requires a 1000 times greater exposure.
However, UV-A markedly enhances the erythema induced by UV_B.(4:pp.131-141,8)
Radiation at wavelengths shorter than the UV -B spectrum, produced by artificial
light sources, are also erythemogenic. In truth, UV -C radiation at 254 nm is more efficient
in producing erythema than is the UV-B spectrum. Though the various studies have not
produced exactly consistent curves, there is agreement that radiation at 254 nm is most
efficient with a rapid reduction in erythemal efficacy, especially at 270-280 nm. On this
point there is some discrepancy, some investigators noting a depression and others a
plateau at 270-280 nm, relative to 290 nm. The curve continues toward but not to zero at
Photomedicine 159

en
en 10
w
Z
w 9
>
i=
t)
8
w
u. 7
u.
W
...J 6
«
:2 5
w
:I:
I- 4
Fig. 6-3. The erythemal action spec- >-
0:
trum. This spectrum has not been com- w 3
w
pletely agreed upon, and this figure > 2
represents the different concepts noted i=
«
in Ref. 1. It is not clear whether there ...J
w
is a dip (dotted line) or a plateau (solid 0:
line) in the effectiveness of wave- 350
lengths around 280 nm. WAVELENGTHS IN NANOMETERS

320 nm. The erythema produced by radiation at wavelengths shorter than 290 nm differs
from the UV-B response in appearance and latent period, i.e., it is a pink rather than a
deep red color, and it reaches a peak at 8 h rather than 14-20 h. Therefore, it may well
represent a different phenomenon.
The term "sunburn" stands for a number of complicated cutaneous responses to
injury induced primarily by wavelengths between 288 or 290 and 320 nm. Obviously,
many changes occur simultaneously. In addition, it is even difficult to differentiate
primary and secondary effects. The nature of the chromophore that absorbs the light
energy in order to initiate the primary photochemical responses is not known. Proteins
containing aromatic amino acids have been considered likely candidates because of their
absorption spectra and the profound effects ofUV radiation on these molecules. However,
a number of other substances, including nucleoproteins (DNA and RNA), urocanic acid,
melanin, and even unsaturated fatty acids of phospholipids, may playa role in the initial
absorption of light energy. (4:pp.117-130,6:pp.225) Lipid peroxidation has been demonstrated
both in vitro and in vivo. Since lysosomal membranes are preferentially damaged by lipid
peroxides, their formation may play an important part in UV radiation-induced injury. It
seems quite likely that multiple chromophores are involved in this complex process.
The pathogenesis of the sunburn erythema has perhaps received the most attention.
However, it has not been established conclusively whether this is a primary or secondary
phenomenon. Though the latter has been considered more probable, direct UV damage to
the upper dermal blood vessels has been demonstrated and some studies suggest that there
is a broad-action spectrum from 250 nm into the long-wavelength UV region that causes a
direct effect on blood vessels. (l:pp.182, 183) In addition, there is a superimposed sharp peak
around 300 nm that induces the formation of a diffusable vasodilating substance.
This brings up the question as to the character of the chemical mediator of the
sunburn response. (4:pp.143-146,6:pp.221-224) The accumulated evidence indicates that one
of the most prevalent vasoactive substances in mammalian tissue, histamine, is not
responsible for this process. In addition, though kinins and possibly serotonin are released
160 Chapter 6

following exposure to sunburn energy, they are not likely to be related to the characteristic
delayed erythema response. More recent evidence suggests that the vasoactive mediator is
a prostaglandin. (9) However, further work is needed to establish the validity of this
concept since the information to date is primarily circumstantial.

6.3.2. Microscopic Anatomy

6.3.2.1. Light Microscopy
Histologically very little is seen in human skin for several hours after exposure to UV
radiation. In fair-skinned people the fIrst recognizable change consists of focci of dys-
keratotic cells (sunburn cells) in the epidermal Malpighian layer noted by 24 h. Within 48
h, damage is noted throughout the epidermis [e.g., abnormal staining (dyskeratotic) and
vacuolated cells and shrunken nuclei]. The dendrites of the melanocytes also become
quite prominent at this time. Regeneration is usually in progress, and the damaged cells
form a desquamating (sloughing) layer by 72 h. Hyperplasia (thickening) of the epidermis
is quite striking by this time (Fig. 6-4). At 96 h there is a great increase in dihydrox-
yphenylalanine (DOPA)-positive melanocytes, with marked arborization of their den-
drites. These melanocytic changes occur in pigmented skin as well, but the dyskeratotic
changes are not nearly as notable and cellular disorganization is much less prominent.
Histochemical and biochemical techniques have demonstrated a number of other
acute epidermal changes following irradiation. (l,4:pp.117-130,147-156,6:pp.219-260) These
include lysosomal membrane alterations and rupture, lipid peroxide formation, inhibition
of the activity of a number of enzymes, increases in both sulfhydryl and disulfide groups,
and the transformation of urocanic acid from a trans to a cis isomer. This last noted
change may well play an important protective role in dissipating sunburn energy. Also,
glycogen accumulates in the epidermal basal cells 12 h postirradiation, presumably to
supply energy for the subsequent burst of mitotic activity.
Dermal changes generally are not prominent. Vasodilatation occurs, and in rodents
an inflammatory infIltrate consisting of polymorphonuclear leucocytes can by seen as
early as 40 min postirradiation, which becomes maximal between 8 and 24 h. This cellular
reponse is minimal, if present at all, in human skin.
The action spectrum for the microscopic changes is found primarily in the UV-B
spectrum. However, as with the erythema response, the long-wavelength UV-A radiation
accentuates the microscopic damage initiated by the shorter wavelengths (UV-B). (8)

6.3.2.2. Electron Microscopy(4:pp.195-215,6:pp.219-260)

Electron microscopic studies have shown that structural changes occur well before
they are demonstratable by light microscopy. The earliest changes noted to date occur in
melanocytes associated with the immediate pigment-darkening phenomenon, with changes
in microtubules, melanosome patterns, and thick fIlaments. Within 2-3 h postirradiation,
marked changes are also noted in the cytoplasm, nucleus, and nucleolus of the epidermal
cells, including vacuole formation, fIlament aggregation, and apparent loss of
lysosomes.(lO,l1) Thus, it appears that a number of structural changes occur shortly after
UV exposure, which seem likely to result from injury to several sites simultaneously rather
than representing only a sequence of events.
Photomedicine 161

Fig. 6-4. UV-irradiated mouse skin 72 h postirradiation. (A) This shows the normal nonirradiated hairless
mouse epidermis that is usually three cell layers thick. There is one mitosis in the basal layer. E, epidermis; D,
dermis . (Hematoxylin and eosin: x 600.) (B) Demonstrates epidermal hypertrophy 72 h post-UV irradiation with
multiple mitoses in the basal layer. E, epidermis; D, dermis. (Hematoxylin and eosin: X600 .) (From Ref. 13.)

6.3.3. Macromolecular Synthesis and Mitosis

The sunburn spectrum has a profound influence on DNA, RNA, and protein syn-
thesis, and mitosis in mammalian skin in vivo. (12,13) Using autoradiographic, bio-
chemical, and metaphase-arresting techniques, it has been demonstrated that synthesis of
these macromolecules and mitosis are inhibited within the fIrst hour postirradiation with a
moderate erythema dose. This inhibition persists for at least 6 h. By 24 h, recovery occurs
followed by acceleration of these functions, reaching a peak at 48-72 h; the activity then
gradually subsides. However, increased mitotic activity in mouse skin has been demon-
strated up to 40 days after a single exposure. (14) The stage of increased mitotic activity is
associated with epidermal hyperplasia, which also gradually subsides over the next 2
months. The mechanism of this hyperplasia and increased mitoses is not established.
162 Chapter 6

Recent findings relating to the presence of epidermal mitotic inhibitors (chalone) and the
actions of cyclic AMP and GMP present interesting but as yet unproven possibili-
ties. (15.16) In addition, a mitotic stimulating substance has been identified postirradiation,
suggesting that this hyperplasia is due to a combination of the removal of inhibition and
the stimulation of growth.

6.3.4. DNA Repair(4:pp.67-77,299-315,6:pp.91-112,17,18)

The inhibition of epidermal cellular DNA synthesis is one of the earliest post-UV
irradiation events that occurs in mammalian skin in vivo, similar to responses noted in
cultured mammalian cells (Fig. 6-5). Among the number of DNA lesions that could
occur, pyrimidine dimer formation, primarily between thymine bases, has received an
extraordinary amount of attention. Currently, three repair mechanisms for DNA base
damage in bacteria have been described; photoreactivation, excision repair, and postrepli-
cation repair (Chapter 4). In mammalian cells, the excision repair system appears to be the
primary mechanism. (17, IS) Autoradiographic (Fig. 6-5), density gradient and chro-
matographic techniques have been used to demonstrate that thymine dimerization can be
induced by sunlight as well as by artificial UV radiation. The DNA can be repaired by

A B

Fig. 6-5. DNA repair 15 min postirradiation. (Al Autoradiographs utilizing tritiated thymidine (3H-TdR) as the
radioactive tracer. This figure shows a non-UV-irradiated human epidermis with the labeling (black dots) con-
centrated in two of the basal cells in a dense labeling pattern, which represents premitotic semiconservative DNA
synthesis. (Hematoxylin and eosin: xno.) (B) Autoradiograph of human skin labeled 15 min post-UV irradia-
tion using 3H_ TdR as the radioactive tracer. A sparse labeling pattern is seen throughout the epidermal cell
nuclei. This sparse labeling represents unscheduled DNA synthesis or repair replication that is absent or marked-
ly reduced in the skin of patients with xerodema pigmentosum. In normal human skin postirradiation, the upper
dermal fibrocytes and vascular endothelial cells show this unscheduled DNA synthesis, giving visual proof of
the penetration of the sunburn rays into the upper dermis. (Hematoxylin and eosin: x no.)
Photomedicine 163

excision of the dimers and their replacement by normal bases in the proper sequence by
the dark repair enzyme system. This damage and repair occurs in the epidermis in vivo and
in vitro, as well as in cell culture systems. The discovery that the human photosensitive
disease xeroderma pigmentosum (XP) is characterized by a defect in this repair system has
supplied a model for the evaluation of the importance of this repair system in human
skin. (4:pp.299-31S) It is of interest that bacteria with a defect in excision repair not only are
more sensitive to the killing effects of UV radiation, but also form increased numbers of
mutations postirradiation. A similar situation exists for cells from patients with XP. The
primary photosensitive problem in patients with XP is their inordinate sensitivity to the
carcinogenic effects of sunlight. Whether this represents cellular mutation or not remains
to be determined.
A postreplication repair process has been described in mouse L cells in culture, and is
apparently dependent on de novo DNA synthesis, and not recombination mecha-
nisms. (4:pp.91-106,19) Recent studies by Lehman et at. (20) indicate that this repair mecha-
nism is defective in cultured cells from four patients with xeroderma pigmentosum (XP
variant).

6.3.5. Melanogenesis( 1 ,4:pp.16S-21S.6:pp.219-260)

Melanins are black or brownish yellow pigments that appear to be complex polymers
of DOPA-quinone, 5,6-dihydroxyindole, and 5,6-dihydroxyindole-2-carboxylic acid at
various oxidation levels, held together by a variety of bond types. This pigment in
mammalian skin is produced from tyrosine and DOPA through enzymatic action in
specialized cells, i.e., melanocytes. This activity occurs on or in the cytoplasmic melano-
some system. Light energy can both induce the immediate pigment-darkening (IPD)
phenomenon and stimulate new pigment formation.
IPD is characterized morphologically by a darkening of the skin that occurs essen-
tially immediately upon exposure to radiation between 320 and 700 nm. It is presumably
due to the oxidation of reduced or partially reduced melanin already present in epidermal
melanocytes (Fig. 6-2). It reaches a maximum at 1-2 h and decreases between 3 and 24 h
postirradiation. Electron microscopic studies suggest that IPD is related to a distal dis-
tribution of the melanosomes toward the tips of the dendrites elicited by the active motor
force of microtubules and microfilaments, and an apparent transfer to keratinocytes (Mal-
pighian cells) without an obvious effect on the character of the melanosome. There is no
significant change in the melanogenic organelles necessary for the synthesis of new
melanosomes.
New melanin formation (delayed tanning) is characterized morphologically by a
brown pigmentation of the skin starting by 48-72 h postirradiation. It reaches a peak by
13-21 days and then gradually subsides over the next several months. The action spec-
trum for this response falls primarily in the sunburn range (290-320 nm). However, long-
wavelength UV and visible radiation, at least as long as 700 nm, can initiate new melanin
formation if enough energy is applied. UV radiation from 250 to 280 nm is also distinctly
less effective in inducing new melanin production. Light and electron microscopy corre-
lated with biochemical studies indicate that this new pigment formation occurs in or on
cytoplasmic organelles called melanosomes, which are formed in the epidermal melano-
cytes. Subsequently, the pigmented melanosomes are transferred to epidermal ker-
164 Chapter 6
Photomedicine 165

atinocytes, which together with the melanocytes make up the epidermal melanin unit (Fig.
6-2). The pigment is then carried to the stratum corneum where it provides the most
distinctive visually dark or hyperpigmented color effect. This new pigmentation provides
the most significant available natural barrier to sunlight-induced skin damage.
Hyperpigmentation of the skin in "delayed tanning" is associated with (1) prolifera-
tion of melanocytes and activation of dormant melanocytes, (2) melanocytic hypertrophy
and increased dendritic arborization, (3) increased melanosomal synthesis, (4) increased
rate of melanization of melanosomes, (5) increased transfer of melanosomes to ker-
atinocytes that is related to an increase in turnover of keratinocytes, (6) increased size of
melanosome complexes (more notable in Caucasoids and Mongoloids), and (7) tyrosinase
activation by a direct effect on tyrosinase-inhibiting sulfhydryl compounds in the epider-
mis. (4:pp.165-194)

6.4. CHRONIC EFFECTS OF SUNLIGHT ON THE SKIN(l,3,6:pp.261-292)

6.4.1. Solar (Actinic) Degeneration

Repeated exposures of Caucasoid skin over a number of years result in gross
changes, including wrinkling, atrophy, hyper- and hypopigmented spots, dilated blood
vessels, yellow raised areas, and actinic keratoses. A distinctive furrowed leathery change
may be noted on the back of the neck, especially of very fair-skinned people who have
received extensive sun exposure, such as sailors and farmers.
Histologically a shortening or flattening of the rete ridges (epidermal downward
projections, Fig. 6-1), a possible thinning of the epidermis, and the presence of many
abnormal cells in disorderly arrangement are seen in chronically sun-damaged human
epidermis. The melanocytes show great variations in size, distribution, and tyrosinase
content, although their absolute number generally remains unaltered. In addition, there
are greater differences in size and melanin-forming activity among the basal melanocytes
as compared to normal skin. Pigment transfer from melanocytes to keratinocytes also
appears to be impaired in areas of chronic solar damage. (21)
More dramatic changes occur after repeated exposure of hairless mouse skin under
controlled conditions. (22) These include a striking thickening of all of the layers of the
epidermis and the epidermal-dermal basement membrane (Fig. 6-6). In addition, there is
an increase in the number of basal cells in DNA synthesis and mitosis, and a decrease in
the transit time of cells through the epidermis.
In contrast to the relatively limited demonstrable epidermal changes, profound der-
mal alterations occur associated with solar radiation-induced degeneration in human skin.
There is a progressive degeneration in the upper portion of the dermis. Specific changes

Fig. 6-6. Effects of chronic UV radiation exposures to mouse skin. (A) Hairless mouse skin after repeated UV
radiation exposures for I month showing a benign hyperplasia with regular mild epidermal hypertrophy, and
moderate but regular thickening of the basement membrane (BM), and thickening of the dermis. E, epidermis;
D, dermis. (Hematoxylin and eosin: x600.) (B) Hairless mouse skin after 4 months of repeated UV irradiation
showing a stage of premalignant to malignant change. It is characterized by abnormalities of size, shape,
polarity, and chromatin patterns of the basal cells, and a marked thickening and clumping of the basement
membrane, which has become discontinuous at several points. (Hematoxylin and eosin: X600.) (From Ref. 26.)
166 Chapter 6

include the development of dilated blood vessels, the accumulation of acid mucopolysac-
charides and abnormal-appearing fibrocytes, the loss of mature collagen, but an increase
in the soluble component, and marked increase and degeneration of elastic tissue referred
to as actinic elastosis. Actinic elastosis, the most prominent and obvious connective tissue
alteration due to chronic solar damage, is a dynamic progressive process that has been
detected as early as the first decade of life. (4:pp.157-163) The earliest change appears to be
a simple increase in number of elastic fibers. Subsequent alterations include thickening,
curling, and increased branching of the fibers with eventual replacement of the dermis and
disorganization of the connective tissue into amorphous masses. Although some questions
may still exist as to the origin of the fibers that stain like elastic tissue, biochemical and
electron microscopic studies have confirmed that actinic elastosis is due to the accumula-
tion of elastic tissue.
The action spectrum for the induction of elastosis in experimental animals falls in the
sunburn range. It has been postulated that this connective tissue change is the result of
photochemically induced alterations in fibroblast function by UV radiation rather than
degradation of connective tissue elements. (1) In support of this concept, the direct injury
of human dermal connective tissue cellular DNA in vivo has been observed within a few
minutes after irradiation with wavelengths shorter than 320 nm.(23) Therefore, the radia-
tion that produces sunburn most likely also plays an important role in chronic solar
damage to dermal connective tissue. Whether the longer wavelengths also have a signifi-
cant influence must remain a matter of speculation at this time. However, recent studies
have shown that chronic UV -A and infrared radiation can produce connective tissue
changes in mouse skin. (24)

6.4.2. Carcinogenesis(3.4:pp.259-283.22,25-27)
6.4.2.1. Human Skin Cancer Formation(4:pp.266-28o,27:pp.488-489)
Examination of the sun's role in the production of human skin cancers does not lend
itself to direct experimentation. However, extensive astute observations have strongly
suggested the etiological significance of light energy in the induction of these tumors.
Skin cancers in Caucasians are generally most prevalent in geographic areas of the
greatest insolation (i.e., receiving the greatest amount of solar radiation) and among
people who receive the most exposure (i.e., people who work outdoors). They are rare in
Negroes and other deeply pigmented individuals who have the greatest protection against
UV-induced injury. Furthermore, the lightest complexioned individuals, such as those of
Scottish and Irish descent, appear to be most susceptible to skin cancer formation when
they live in geographic areas of high-UV fluence. When skin cancers do occur in the
darkly pigmented races, they are not distributed primarily in the sun-exposed areas, as
they are in light-skinned people. The tumors in these pigmented individuals are most
commonly stimulated by other forms of trauma such as chronic leg ulcers, irritation due to
not wearing shoes, the use of a kangeri (an earthenware pot that is filled with burning
charcoal and strapped to the abdomen for warmth), the wearing of a dhoti (loin cloth), and
so on. In contrast, the distribution of skin cancer in the Bantu albino and in patients with
xeroderma pigmentosum follows sun exposure patterns. The arguments supporting the
role of sunlight in human skin cancer formation may be summarized as follows: (1) skin
cancers do occur predominantly on the sun-exposed parts of the body, (2) they are more
Photomedicine 167

common in regions of the earth that receive the most sunlight, and (3) pigmented races are
much less susceptible to skin cancer formation than Caucasians.
Though these arguments do not constitute absolute proof and are themselves not
wholly established, there is a considerable body of circumstantial evidence supporting the
role of sunlight in at least three types of skin cancers: basal cell epitheliomas, squamous
cell carcinomas, and melanomas.
The formation of the most common skin cancers (and therefore the most common
human malignancies), the basal cell epithelioma and squamous cell carcinoma, is influ-
enced by four major factors. (4:pp.259-283)

1. Total lifetime of sunlight exposure. Exposure to sunlight for prolonged periods,

such as occurs in Texas and Australia, appears to be an important factor in the
development of skin cancer. However, a comparison of skin cancer formation in
Galway, Ireland, and Philadelphia indicates that hours of exposure does not
constitute the complete story. Cancers occurred in Philadelphia with much less
total exposure.
2. Intensity and duration of the UV component in sunlight. Recent studies indicate
that skin cancer formation can be correlated with the duration of exposure to the
erythemally effective radiation (290-320 nm) more accurately than the duration
oftotal sun exposure. Using this criteria, the discrepancy between cancer forma-
tion occurring in rural Galway, with much more total sunlight exposure, and
urban Philadelphia becomes easier to understand. The maximum exposure to the
sunburn spectrum possible in Galway would be 1.33 months/year and in Phila-
delphia it would be 4 months/year (if people were out of doors continuously
during daylight hours). On this basis, the "effective" exposure times of the
Philadelphia and Galway skin cancer groups become more similar.
3. Genetic predisposition. The condition xeroderma pigmentosum is the most strik-
ing example of genetic predisposition to skin cancer formation. (4:pp.299-315)
Those individuals who lack the ability to repair UV-damaged DNA generally
develop all varieties of sunlight-induced malignancies within the first two dec-
ades of life. However, genetic predisposition also appears to play an important
role in the formation of basal and squamous cell cancer in the general population.
A number of studies have shown a distinct association between these cancers and
light eye color, fair complexion, light hair color, poor ability to tan, ease of
sunburning, and a history of repeated sunburn reactions. In addition, individuals
of Celtic origin are more prone to this type of carcinogenesis than other light-
complexioned individuals.
4. Factors unrelated to sunlight. Exposure to sunlight, primarily to erythemally
effective wavelengths, appears to be the primary factor in the formation of these
skin malignancies. The morphological distribution of the most common cut-
aneous cancer, the basal cell epithelioma, indicates that other factors must also
playa role. Squamous cell carcinomas appear to be directly related to sun
exposure. In contrast, about one-third of basal cell epitheliomas occur on areas
of the skin receiving minimal exposure to sunlight. Thus, although sunlight
appears to be a dominant factor in the induction of this type of tumor, other as
yet undetermined influences must also participate.
168 Chapter 6

Melanomas represent the most dreaded type of skin cancer. Unlike basal cell and
squamous cell cancers, these lesions metastasize readily to other organs in the body and
have a comparatively high mortality rate. Fortunately, they are relatively uncommon.
Unlike the two common malignancies, the influence'of sunlight on melanoma formation is
not as well established.(26) Melanomas are not found primarily on sun-exposed areas, and
the protective effect of melanin pigment is not so obvious as it is in other forms of skin
cancer. However, a number of surveys suggest that sunlight does influence the develop-
ment of at least some of these malignancies. (28) Caucasians with melanomas statistically
tend to have light eyes, light hair, fair complexions, and spend more time outdoors when
compared to a control group of patients without melanomas. Surveys of the geographic
distribution of this cancer, even in genetically similar populations, have demonstrated a
much greater melanoma prevalence associated with high insolation (Le., exposure to
sunlight), as compared to low insolation. The occurrence of solar radiation-induced
melanomas in patients with xeroderma pigmentosum, and the production of melanomas
from benign pigmented lesions by chronic exposure to UV radiation in experimental
animals, further confirms the potential of sunlight to stimulate the production of these
tumors under proper circumstances. In addition, the distribution of melanomas developing
from circumscribed precancerous melanosi.s strongly suggests that the sun is at least
responsible in part for such lesions. However, the anatomical and geographic distribution
of melanomas indicates that factors other than sunlight must play an important role in the
etiology and increasing incidence of these tumors. (29)

6.4.2.2. Experimental Carcinogenesis

6.4.2.2a. Quantitiative Investigations
Traditionally, the skin of the ear of the albino mouse and, to a lesser extent, the
albino rat was used for the experimental production of tumors by UV radiation because
this area lacks the three main natural inhibiting factors: pigment, a thick stratum corneum,
and thick hair growth. The type of tumor produced in this tissue was primarily a sarcoma
that was quite adequate for the monumental quantitative studies accomplished by Blum
and coworkers.(25) Using the ears of an inbred, albino, haired mouse strain and a UV
radiation source primarily emitting radiation shorter than 320 nm, they demonstrated that
skin cancer development follows the law of reciprocity as far as dose-rate dependence is
concerned. Increasing the dose or shortening the intervals between exposures accelerated
tumor formation but did not alter the shape of the incidence curve.
A factor in the production of skin cancer by UV radiation that has apparently not
been explored is the influence of protraction of the daily dose on reciprocity. Various
experimenters have used exposures varying in length from a few seconds to several hours,
while exploring such variables as influence of chemicals, action spectrum, etc. There are
apparently no published data on whether a radiation source has an equally good chance of
producing skin cancer if a given fluence is delivered in a few seconds or is attenuated over
several hours. However, recent studies by Forbes et al. (19) demonstrated that the delivery
of the same dose in divided doses (Le., five exposures a week) was more effective in
producing tumors than when delivered once a week. The reason for this lack of the
reciprocity effect is not clear.
Photomedicine 169

Under controlled circumstances, skin cancers will develop if enough energy is deliv-
ered for a sufficient period of time.(25) However, tumors will not appear no matter what
energy is used if it is not applied long enough. Blum(25) surmised that UV radiation-
induced cancer formation is a continuous process that begins with the initial exposure.
The appearance of tumors within the lifetime of the animal depends on sufficient accelera-
tion of the growth process. In support of the concept that growth acceleration is an
important component of carcinogenesis, the production of squamous cell carcinomas in
hairless mouse skin with one exposure to UV radiation followed by croton oil promotion
has been reported. (26)
The mechanism of tumor growth acceleration remains to be established. The produc-
tion of mitosis-stimulating substances or, perhaps more importantly, the removal of
mitosis-inhibiting materials ("chalones") may well playa role in this process.

6.4.2.2h. Qualitative InvestigationS<4:pp.259-283,22,26)

The evaluation of progressive changes occurring in tissue during UV radiation-
induced epidermal cancer development was examined in the skin of the hairless mouse;
these tumors are identical to primary growths induced with sunlight radiation in human
skin. In addition, these animals proved useful in studying the effect of UV radiation on
melanoma formation. (26)
A number of investigators have considered dermal influence to be of prime impor-
tance in the development of epidermal skin cancers. Although traditionally actinic elas-
tosis was considered to play an important role in this respect, several studies have shown
that such changes are not essential to the formation of epidermal malignancies in humans
or experimental animals. (22,26) However, other changes, including dissolution of elastic
tissue and collagen, proliferation of young collagen, accumulation of acid mucopolysac-
charides, mast cells, and fibroblasts, formation of new elastic tissue, and alterations in
dermal vasculature, have been reported. (24,26) Some or all of these changes may be
associated with significant dermal influences on the development of epidermal tumors.
The progressive development from benign hyperplasia through stages of actinic
keratosislike lesions to frank invasive malignancy was evaluated in the hairless mouse
system, using histochemical and radioactive tracer eH-thymidine) techniques to examine
anatomic changes and cellular kinetics. (22,26) These studies demonstrated an accumula-
tion of acid mucopolysaccharides, a loss of insoluble collagen in the upper dermis, and the
proliferation of mast cells and fibrocytes. The most striking response occurred in the
epidermal-dermal basement membrane (BM) (Fig. 6-6). A progressive thickening of this
structure was noted with the development of epidermal hyperplasia. As irregular and
abnormal epidermal cell proliferation occurred, the BM became thicker, irregular, clum-
ped, and frayed in appearance, and with frank invasive malignancy, it disappeared al-
together. Similar breaks in the BM have been noted with the invasion of human skin
cancers and chemically induced experimental tumors. Electron microscopic studies have
suggested that the microprojection of celIs through the basal lamina may represent the
earliest stage of tumor invasion.
Epidermal cell kinetic studies demonstrated a progressive increase in the number of
germinative basal cells synthesizing DNA and dividing, with a shortening of the DNA
synthesis time and G2 period associated with the development of epidermal changes. (22)
170 Chapter 6

In addition, there was a progressive reduction in the cell transit time through the epidermis
despite its increasing thickness. With frank malignancy, the presence of a germinative cell
layer was lost, and mitoses, many abnormal in appearance, were present throughout the
tumor. Thus, the process of carcinogenesis was characterized by the acceleration of cell
formation, maturation, and turnover. Furthermore, the epidermal germinative basal cell
appeared to be the primary or initial site of abnormal proliferation.
Experimentally, the transformation of benign pigmented growths to malignant
melanomas in the skin of hairless mice has been induced with radiation shorter than 320
nm. (27) Repeated exposures over several months resulted in the production of large
invasive melanocytic tumors with histologic, autoradiographic, and electron microscopic
characteristics of malignancy. In addition, several of these tumors metastasized to the
regional lymph nodes. These results indicate that benign pigmented lesions can be trans-
formed to malignant growths in experimental animals by UV radiation. Whether a similar
process occurs in human skin remains to be established.
The enhancing effect of heat on the degree of cutaneous injury and the intensity of
erythema response to UV radiation has been well documented. (27) Furthermore, it has
been demonstrated that increased temperatures present at the time of exposure to UV
radiation accelerate tumor production. Clinical experience also suggests that heat does in
fact aggravate UV radiation-induced cancer formation in human skin.
The presence of carcinogenic and tumor-promoting chemicals in our environment
has made the experimental evaluation of chemical influences on UV radiation-induced
carcinogenesis of practical importance. (26) Recent studies have demonstrated that UV
radiation and chemical carcinogenic stimuli are additive in nature. In addition, repeated
applications of the noncarcinogen croton oil results in significant cancer formation follow-
ing a single UV exposure. These and other experimental findings suggest that environ-
mental chemicals may playa role in the development of skin tumors. (27)
The relationship of immunological factors to the control of cancer formation,
growth, invasion, and rejection presents another important aspect of carcinogenesis,
which is being examined experimentally. (30-32) Recent studies have demonstrated that
subcarcinogenic amounts of UV-B radiation will induce a specific population of sup-
pressor T lymphocytes, which prevent rejection of skin cancers that have been produced
by carcinogenic amounts of UV -B radiation in experimental animals. Whether such tumor
tolerance occurs in humans is not known at this time.
Perhaps the most intriguing aspect of UV radiation-induced carcinogenesis concerns
the relationship of the acute responses to the eventual cancer formation. (3,17,25-27) Light-
induced carcinogenesis occurs only when acute phototoxic erythema is produced as well.
The croton oil experiments also indicate that the process of UV radiation-induced cancer
formation begins with the initial exposure.(27) However, the early changes leading to
cancer formation remain undetermined. A number of studies have demonstrated a similar
but not identical effect of UV radiation and chemical carcinogens on certain vital func-
tions in mammalian epidermis, including DNA and RNA synthesis, cell turnover, and
mitotic rate. (13) The most intriguing findings to date relating acute photoinjury to cancer
formation have developed from the study of DNA repair systems and of the genetic
disease xeroderma pigmentosum (XP). As noted in Section 6.3.4, this is a rare genetic
disease in which photosensitivity is expressed as an inordinate susceptibility to the devel-
opment of sunlight-induced skin cancers (Fig. 6-7). (4:pp 299-315,20,33) The cells of these
Photomedicine 171

Fig. 6-7. Clinical picture of a patient with xeroder-

ma pigmentosum showing actinically damaged skin,
multiple actinic keratoses, and tumors. (Courtesy of
William Spencer, M.D.)

patients are defective in their ability to repair UV radiation-damaged DNA. (20,27 ,33)
Cultured XP cells do have a higher mutation rate than normal human cells following UV
irradiation. (34) It has been postulated that this defect may lead to a high mutation rate
following sunlight irradiation, and thus result in cancer formation. It should be noted that
even if this theory proves correct for the cancer susceptibility of XP patients, it may not
relate to sunlight-induced cancers in the general population.

6.5. ADVERSE CUTANEOUS REACTIONS TO SUNLIGHT (2-4,6,35)

Most of the cutaneous effects of sunlight are injurious. The following part of this
discussion is designed to define the pathogenesis of some of the adverse effects of this
energy. The explanations will of necessity be somewhat crude due to the limitations of our
knowledge at this time.

6.5.1. Photosensitivity(3)
Photosensitivity is the broad term used to describe adverse reactions to sunlight or
artificial light energy. Two types of photosensitivity reactions may occur. The reactions
may be phototoxic or photoallergic in nature.

6.5.1.1. Phototoxicity(35)
Light-induced damage in the skin that is not dependent on an allergic mechanism
may be considered to be phototoxic. Theoretically, these reactions could occur in every-
body if the skin were exposed to enough light energy of the proper wavelengths, and
172 Chapter 6

enough molecules that absorb these wavelengths were present. Of course, the radiation
must penetrate to the absorbing molecules for the reaction to occur. The sunburn reaction
is the classic example of a phototoxic response. Although all types of skin will sunburn,
much more energy is required to produce this response in deeply pigmented skin than in
very light skin. Clinically, phototoxic reactions usually are characterized by erythema and
at times by edema (swelling) that occurs within a few minutes to several hours after
exposure, followed by hyperpigmentation and desquamation (peeling) confined to the
exposed areas (Fig. 6-8).
Histologically, epidermal cell degeneration may be prominent when the photosen-
sitizer is in the epidermis. This is most notable after the application of an exogenous
photosensitizer to the skin. Edema with a mild to moderate inflammatory cell infiltration
into the dermis consisting primarily of polymorphonuclear leukocytes may be seen (Fig.
6-9).
Despite a remarkable amount of investigation, the mechanisms by which phototoxic
responses occur are not well understood. In the case of an exogenous photosensitizer,
either the molecule alone or a complex of the chemical and cellular organelles becomes
excited by the absorption of light. Triplet states and/or free radicals may thus be formed,
and the dissipation of this energy may result in a number of changes including peroxide

Fig. 6-8. Phototoxic reactions. Demonstrates a phototoxic reaction in two areas on the forearm, Band C, due to
the topical application of a psoralen compound followed by exposure to the action spectrum for this chemical
(UV radiation between 320 and 400 nm). It is characterized by erythema and edema at this point. The area
between B and C received the UV exposure without prior application of the psoralen compound. (Courtesy of H.
I. Maibach, M.D. and F. N. Manulli, M.D.)
Photomedicine 173

Fig. 6-9. Histology of the phototoxic reaction. Biopsy of the phototoxic reaction shown in Fig. 6-8 demonstrat-
ing a marked destruction of the epidermis with intracellular and extracellular edema (swelling), vacuole forma-
tion, and cell death. The dermis shows very little cellular infiltrate. E, epidermis; D, dermis. (Hematoxylin and
eosin: x50.) (From Ref. 3.)

formation, cell membrane or lysosomal membrane damage, and nuclear and/or mitochon-
drial injury.
It is likely that mechanisms vary with the photosensitizer. (36) In support of this idea,
there is evidence that the phototoxicity induced by the furocoumarins is associated with
the formation of an adduct with DNA. Chlorpromazine appears to form similar adducts,
primarily with RNA. Other chemical photosensitizers require the presence of oxygen.
This last type of phototoxicity has been termed "photodynamic action" (Chapter 3).
Certain dyes and chemicals such as methylene blue, acriflavine, rose bengal, and por-
phyrins produce phototoxic effects on living and nonliving substrates only in the presence
of oxygen. The photodynamically active substance becomes excited and forms a triplet
state or a free radical. Singlet oxygen formation may playa role. The excited chemical
also may form peroxides and then oxidize the substrate. Other possibilities include pass-
ing the energy from the excited chemical to the biological substrate, which then becomes
oxidized, or the activated chemical may be able to accept electrons resulting in oxidation
of the substrate. After excitation, the photosensitizing molecules return to the ground state
and are structurally unchanged. Thus, they may partake in the induction of photodynamic
reactions as long as they are associated with the substrate. Recent studies have also
indicated that complement, mast cells, and prostaglandins may also play an important role
in mediating some phototoxic responses.(36:p.47,37,38)
174 Chapter 6

Fig. 6-10. Solar urticaria. Clinical picture of urticarial wheal (hive) occurring shortly after exposure to UV
radiation and subsiding within 30 min. (From Ref. 3.)

6.5.1.2 . Photoallergy(35,39)
Photoallergy can be defined as an acquired, altered capacity of the skin to react to
light energy alone or in the presence of a photosensitizer that is presumably dependent on
the development of a circulating antibody or a cell-mediated immune response. These
reactions are generally uncommon, and the clinical patterns range from an immediate hive
response (Fig. 6-10) to delayed itching, scaling, or weaping lesions (Fig. 6-11). The
eruption frequently extends beyond the exposed areas, and eruptions in distant, previously
involved sites may occur. Under controlled conditions, the general characteristics of
allergic responses can be identified, including detection of an incubation period, spon-
taneous flare, and transfer of the process to normal SUbjects. Usually, less radiation
exposure is required to induce a photoallergic response than a phototoxic response, when
both are dependent on the same action spectrum.
Photoallergy also differs histologically from phototoxicity. The immediate urticarial
(hive) lesions show very little microscopically other than some edema and blood vessel
dilatation. The delayed reactions present a dense round-cell infiltrate around the blood
vessels in the dermis that is characteristic, though not diagnostic, of these responses (Fig.
6-12). Similar to phototoxicity, photoallergy can be produced with or without the pres-
ence of known exogenous photosensitizers.
Photo medicine 175

Fig. 6-11. Allergic photocontact dennatitis. Clinical picture of an eczematous photoallergic contact dennatitis
induced by utilization of a soap containing tribromosalicylanilide and exposure to the action spectrum (UV
radiation between 320 and 400 nm).

6.5.2. Clinical Problems

It is not appropriate in this chapter to dwell on the clinical responses that fall
primarily in the field of individuals with dermatological training. However, an outline of
these reactions hopefully may serve a useful purpose in putting medical effects of light
176 Chapter 6

.;•• ..f."~"-.-
.~
- ..'.
., ... ....
... ~. ~
,. '.

.'.
(
;
.......
\ "

.:
.'

Fig. 6-12. Histology of the allergic photocontact dermatitis. Biopsy of an allergic photocontact reaction show-
ing a dense perivascular round cell (collection of black dots in the dermis) infiltrate in the dermis characteristic of
a delayed hypersensitivity response. E, epidermis; D, dermis. (Hematoxylin and eosin: x80.) (From Ref. 3.)

energy in perspective. The adverse clinical effects may be divided into three large catego-
ries: (1) responses due to lack or loss of protection, (2) those due to the presence of a
photosensitizing chemical, and (3) those not due to a deficiency in protection or to the
presence of a known photosensitizer.

6.5.2.1. Lack or Loss of Protection(2,3)

Melanin pigment is the primary natural protective agent in the skin. Therefore,
almost all of the clinical problems in this category relate to a melanin deficiency. This
group includes patients with albinism who have a defect in the enzymatic tyrosinase
system and are therefore unable to produce sufficient melanin; vitiligo, in which there is a
loss of pigment cells and thus a loss of melanin; phenylketonuria (PKU), in which
pigment dilution occurs due to an inherited error in phenylalanine metabolism; Chediak-
Higashi syndrome, a genetic disorder with pigment dilution, apparently due to a structural
abnormality in the melanosomes; and, the most common of all photosensitive conditions,
i.e., fair complexion with light hair and eye color. This category also includes patients
with xeroderma pigmentosum, in whom the lack of melanin does not playa role (Fig.
6-7). As noted in Section 6.3.4, the defect in this disease relates to an inability to repair
UV radiation-damaged DNA.
The clinical reactions in this category consist of acute sunburning and, with chronic
repeated injury, skin degeneration, actinic keratoses (Fig. 6-13), and skin cancer forma-
Photomedicine 117

AK

AK

Fig. 6-13. Chronic sun (actinic) damage. Chronic actinic damage of the back of a hand with actinic keratoses
(AK).

tion. The responses are phototoxic in nature, and the action spectrum falls in the sunburn
range (290-320 nm).

6.5.2.2. Photosensitized Reactions(2,3,36)

A photosensitizer in this discussion will be considered to be a substance that, as the
result of absorption of the sun's energy, induces an adverse response in the skin. The
photosensitizer may be endogenous or exogenous in origin.
The porphyrins are the only well-established photosensitizers made by the human
body. These are pyrrole ring structures that are essential to cellular metabolism throughout
the body. Among other things, the heme part of hemoglobin is formed by these mole-
cules. However, in the group of diseases named the porphyrias, large amounts of por-
phyrins are made in the erythropoietic (bone marrow) and/or the hepatic (liver) tissues. In
these states, the porphyrins are not associated with iron and are potent photosensitizers.
As such, they induce cutaneous changes ranging from marked photodestruction of the skin
and underlying tissues to simple hyperpigmentation, depending on sun exposure and the
178 Chapter 6

amount of porphyrin molecules available. The cutaneous changes are due to photodyna-
mic phototoxic responses.
Exogenous photo sensitizers may get to the skin by topical application (contact) or
through the bloodstream (systemic).
(1) Topical chemicals may cause photosensitivity reactions advertently, as occurs
with tars and psoralen compounds in the treatment of psoriasis and vitiligo, or inadver-
tently, as occurs with the psoralen molecules in plants and perfumes, halogenated sali-
cylanilides in deodorant soaps, sulfonamide and phenothiazine medications, and so
on.(2,17) The reactions are usually phototoxic, though occasionally they are photo allergic
in type. The action spectrum usually falls in the range of 320-400 nm.
(2) Systemically administered chemicals also may induce photosensitivity reactions
advertently, as occurs with the use of psoralen compounds in the treatment of vitiligo and
psoriasis, or inadvertently, which is the usual case. These inadvertent photoreactions are
generally induced by commonly used medications including antibacterial sulfonamides,
thiazide diuretics, sulfonylurea antidiabetic drugs, phenothiazines, and the broad-spec-
trum antibiotic demethylchlortetracycline. (36) The vast majority are phototoxic in nature,
though occasionally delayed hypersensitivity responses may occur. The action spectrum
for both phototoxic and photoallergic reactions usually includes long-wavelength UV
radiation.

6.5.2.3. Protection Normal and No Known Photosensitizer

Photoreactions not due to deficient protection or the presence of known photosen-
sitizers comprise a catch basket of conditions, including allergic responses such as the
immediate antibody-mediated solar urticaria (hive response) and the apparently cell-
mediated polymorphous light eruption (PMLE), certain autoimmune diseases, vitamin
deficiencies, and a number of genetic problems. In general, these are uncommon diseases,
except for PMLE. The action spectrum and mechanisms vary with the conditions. Be-
cause of the complicated and primarily clinical nature of this category, it will not be
discussed further. (2,3)

6.6. BENEFICIAL EFFECTS OF SUN AND/OR ARTIFICIAL LIGHT ENERGY

Although adverse effects are much more common than beneficial responses, UV
radiation does have therapeutic uses.

6.6.1. Vitamin D Formation(l,2,4:pp.247-258,6:pp.195-218,40)

Perhaps the only completely established beneficial effect of UV radiation on normal

human skin is the formation of vitamin D. Vitamin D or, better stated, the D vitamins
affect the transport of calcium from the gut, and also affect bone metabolism. These
vitamins prevent rickets (softening of bones) in children, and osteomalacia (loss of bone
substance) in adults by ensuring proper bone calcification.
Vitamin D3 (calciferol) is made in the skin, most likely in the Malpighian cells (Fig.
Photomedicine 179

6-2), through conversion of 7-dehydrocholesterol by the action of UV radiation. Further

hydroxylation to 25-0H calciferol provides a more efficient compound. It has been
suggested that skin color, and therefore the amount of protection against UV radiation,
has been geographically regulated to prevent naturally occurring vitamin D intoxica-
tion. (41) Thus, the light skin of northern Europe was needed because of the limited amount
of natural exposure to UV radiation, and the dark skin of peoples who live in the tropics is
a necessary protection against producing too much vitamin D. However, as yet no natural
vitamin D intoxication has been reported in light-complexioned individuals who have
migrated southward. Thus, vitamin D formation by UV irradiation remains primarily a
beneficial effect.

6.6.2. Treatment of Skin Diseases(6:pp.511-532)

Niels Finsen, an outstanding physician and an early pioneer in photobiology, re-

ceived the Nobel prize in 1903 for his use of UV radiation in the treatment of cutaneous
tuberculosis. Though this form of treatment is no longer necessary due to medical ad-
vances, the therapeutic value of light energy has been described for a large number of
conditions including acne, eczema, pityriasis rosea, vitiligo, psoriasis, and herpes sim-
plex. Perhaps we have the most information as to such effects in the last three diseases.

6.6.2.1. Vitiligo
Vitiligo is a cutaneous disorder characterized by a loss of pigment cells and thus
pigment, the etiology of which is speculative at present (Fig. 6-14). Phototherapy repre-
sents almost the only available treatment, and success is limited at best. The procedure
consists of the topical or systemic administration of psoralen compounds followed by
exposure to radiation that includes the action spectrum of these chemicals, i.e., from 320
to 380 nm. (4:pp.335-368, 783-791) It should be noted that the action spectra for the stimula-
tion of pigmentation and the induction of erythema are quite similar, but they may not be
identical. Until recently it was felt that a significant phototoxic erythema effect was
necessary for the induction of repigmentation. However, certain studies indicate that the
pigment response may be separate from the phototoxic injury (as defined by the erythema
and visible damage). (42) The basic mechanism for the induction of hyperpigmentation
induced by psoralen compounds and long wavelength UV radiation has not been estab-
lished. However, the following observations have been noted following application of
4,5' ,8-trimethylpsoralen (TMP) and exposure to these wavelengths of radiation: (1) an
increased number of functioning epidermal melanocytes occurs due to proliferation and/ or
activation of inactive cells (mitotic activity has been detected between 48 and 72 h after
TMP plus UV-A irradiation), (2) melanocyte hypertrophy and increased extension of
dendrites around the keratinocytes, (3) increased number of melanosomes in melanocytes
and keratinocytes, (4) increased tyrosinase activity in melanocytes, (5) increased rate of
transfer of melanosomes to keratinocytes, and (6) dispersion of melanosomes in ker-
atinocytes in nonaggregated distribution. (This last effect has only been described in
Caucasoid skin.) This dispersed distribution of melanosomes may persist for more than 9
months, suggesting that some long-term gene depression may occur, perhaps related to
the photoaddition of psoralens to DNA.
180 Chapter 6

Fig. 6-14. Clinical picture of vitiligo showing depigmentation over the hands and feet areas caused by loss of
pigment cells.

6.6.2.2. Psoriasis
Psoriasis is a common dermatological disease of unknown etiology that affects 1-3%
of the world's population, and between 2 and 8 million people in the United States. The
clinical picture varies from localized scaling plaques to generalized exfoliation (Fig.
6-15). The most notable pathological characteristic of this disorder is a marked increase in
epidermal cell proliferation with rapid turnover of the germinative cells.
Photomedicine 181

Fig. 6-15. Clinical picture of psoriasis characterized by large, erythematous, scaling plaques.

Treatment of psoriasis has revolved around attempts at the inhibition of this increased
cellular proliferation, and phototherapy represents one of the most time-honored, effective
methods in this respect(4: pp .793-796,6: pp .51l-532) Astute clinical observations and subse-
quent controlled studies have established that radiation in the UV-B region (290-320 nm)
is therapeutically effective in this disease. In 1925, Goeckerman described increased
benefits from the topical application of crude coal tar to the sunburn exposure regimen.
Since the action spectrum of the coal tar photosensitization is in the UV -A range, the
increased effects had to be additive in nature, i.e., the tar and UV-B acted independently.
One of the disadvantages of the Goeckerman regimen was the messy nature of the tar,
which generally had to be used in a hospital setting. Within the past few years, regression
of lesions following topical application of psoralen compounds and UV -A radiation has
been reported. Most recently, an apparently extremely effective treatment has been de-
scribed using systemic 8-methoxypsoralen and UV-A radiation emitted from a newly
developed high-intensity system (PUVA).(43)
Though the mechanism of the effect of phototherapy on psoriasis has not been
completely established, a reduction in cellular proliferation most likely plays a part. UV
radiation shorter than 320 nm has a profound influence on epidermal cellular DNA
synthesis and mitosis, which are inhibited shortly after irradiation and remain so for
182 Chapter 6

several hours. (11) Similarly, psoralens plus long-wavelength UV radiation significantly

inhibit these functions, presumably due to photo addition between the chemical and thy-
mine bases of the DNA. It is also possible that phototherapy influences the dermal
vasculature either directly or indirectly, and thus alters the course of the psoriatic lesions.
Further study is obviously needed to clarify these concepts.

6.6.2.3. Herpes Simplex Injection(6:pp.545-559.571-594.44)

The herpes simplex virus is a common human pathogen that causes primary and
recurrent skin and mucosal infections, and occasionally may involve other organs. Re-
cently, a possible relationship to human cancer formation has been described. By far the
most common clinical problem is the recurrent herpes simplex infection that is charac-
terized by localized vesicular lesions involving the face, especially the lips, and the
genitalia. Two varieties of closely related viruses are responsible for the lesions: those that
occur above the waist are usually due to the type 1 virus, and those that appear below the
waist are generally due to the type 2 variety. The eruption may occur as frequently as at
monthly intervals or less, and thus presents a most disturbing and at times debilitating
problem.
A group of investigators have advocated the use of photoinactivating procedures
utilizing photodynamically active dyes, such as proflavine and neutral red, and visible
light to inactivate the virus in these lesions. This concept resulted from the fact that
photoinactivation of the viruses could be accomplished in tissue culture by this procedure.
At present the clinical efficacy and safety of this herpes simplex virus inactivation
has not been established. The clinical results are as yet not convincing due to the great
difficulty in obtaining enough patients to have adequate controls. In addition, in vitro
studies suggest that there may be a risk of inducing cancer by this procedure.(6:pp.571-
594.44) Thus, this treatment should be considered with some reservation until further
information is available.

6.6.2.4. Photochemotherapy oj Tumors(6:pp.625-638.45)

Photo chemotherapy of tumors using exogenous chemicals and photosensitization of
tumor tissue apparently was initially attempted in the early 1900s. (46) In the 1970s the
concept was examined further by a number of investigators. (6) Most of these reactions in
vivo are mediated through a transfer of energy from the triplet excited state of the
sensitizer to molecular oxygen producing a highly reactive singlet oxygen which is
cytotoxic. (6:pp.113-144) The chemical in wide use at this time is named hematoporphyrin
derivative (HpD). This chemical localizes to tumor cells apparently due primarily to one
of its components, di-hematoporphyrin ether (DHE).(47)
Though the major absorption peak for HpD is at the Soret band around 400 nm, the
630-nm band is the most important for therapy because tissue absorption at 630 nm is
minimal, which allows greater penetration into the tissues. (48) Filtered xenon and tungsten
lamps have been used to treat cutaneous and subcutaneous lesions. However, the argon
pumped dye laser at 630 nm and optical fiber delivery systems can be used in all body
cavities. (45)
This photochemotherapy has been and is being used to treat lung, esophageal,
Photomedicine 183

bladder, occular, head and neck, neurological, and gynecological tumors with partial to
complete response in 85% of the patients. (6:p.89) However, more cutaneous and sub-
cutaneous lesions have been treated to date. These include metastatic breast cancers,
melanomas, and basal cell epitheliomas. The results have been most encouraging.
It should be noted that the HpD is also being used for tumor detection because it does
localize in cancer and dysplastic tissue. Fluorescence of the HpD is usually stimulated by
light rays around 400 nm from a filtered mercury-arc lamp or a krypton ion laser. (45:p.87)
Neither HpD nor DHE are toxic in the dose utilized in the absence of the light
energy. However, they are retained in the skin and do induce a photosensitivity that may
last 3-4 weeks.

6.6.3. Treatment of Systemic Diseases (Neonatal Jaundice)

Historically, phototherapy has been prescribed for a variety of diseases. As noted
previously, there is evidence that UV radiation is beneficial for diseases of calcium
metabolism, such as rickets and osteomalacia, due to the photochemical formation of
vitamin D in the skin. Perhaps the one other clinical condition in which there is good
evidence for the benefits of phototherapy is neonatal jaundice. (4:pp.231-258,6:pp.477-
510,49)

Bilirubin (C33H3306N4)' the chief pigment in human bile, is formed from heme
when hemoglobin from red blood cells is broken down primarily by the Kupffer cells of
the liver. It appears to result from the opening of the porphyrin ring. In the neonatal
period, increased amounts of unconjugated bilirubin are present in the circulation. This is
due to the shorter survival of red blood cells, and to the functional immaturity of the
neonatal liver, such that it is limited in its capacity to convert bilirubin from a lipid-soluble
to a water-soluble compound (i.e., by conjugation) that can be excreted in the urine. In
addition, increased permeability of the blood-brain barrier makes the newborn infant
more susceptible to central nervous system damage by the deposition of bilirubin in brain
cells (kernicterus). The danger is most notable in the premature infant. At present, therapy
is designed to keep unconjugated bilirubin levels below 10-15 mg%. One of the mecha-
nisms for the reduction in circulating unconjugated bilirubin, which has received consid-
erable attention recently, is phototherapy.
In vitro studies indicate that light energy primarily between 450 and 500 nm will
cause photodegradation of bilirubin. The exposure of experimental animals (Gunn rats)
and infants to this energy has established that there is some alteration of bilirubin by the
light energy. In addition, there is a striking reduction in circulating unconjugated biliru-
bin. In vitro studies have demonstrated that wavelengths in the blue part of the visible
spectrum (primarily 450-500 nm) will photooxidize bilirubin. These in vitro photooxida-
tion products lack the toxicity of bilirubin. The in vivo derivatives are less lipophilic and
are more readily excreted without requiring conjugation. Therefore they are less likely to
accumulate in the central nervous system tissue where the damage occurs in kernicterus.
The mechanism of the beneficial effects of this phototherapy does not depend on pho-
todegradation of bilirubin as originally believed. In infants exposed to blue light the major
components of the excreted products are isomers of bilirubin. Thus, phototherapy en-
hances the excretion of bilirubin predominantly by photoisomerization. (49)
As with all therapies, the potential disadvantages must be considered. Fortunately,
184 Chapter 6

the photoproducts produced appear to be either nontoxic or so readily excreted that they
do not accumulate in the central nervous system. Photodynamic injury to cutaneous cells
must be considered as a possible detrimental effect, though no such response has been
observed as yet. In addition, a number of injurious effects on bacteria and purified DNA
have been recorded. Whether such fmdings relate to clinical responses remains to be
determined. Perhaps the most potentially disturbing aspect of this phototherapy relates to
possible retinal damage, which has been induced in newborn piglets exposed to blue
fluorescent lights. No such damage has been noted in human infants to date despite
extensive exposure. This is most likely due to the extensive use of eye protection.

6.7. PHOTOPROTECTION

The concept of cutaneous photoprotection encompasses protection against all of the

adverse effects of sun on the skin. These range from acute phototoxic, photoallergic,
idiosyncratic, and specific disease-oriented responses to chronic damage, and car-
cinogenesis. Obviously, the available sunscreens provide excellent methods for such
protection, and recent data indicate that these screens also will inhibit UV radiation-
induced cancer formation in experimental animals. Also, there are a number of non-
sunscreen photoprotective measures that can be used to reduce the incidence and severity
of acute adverse reactions, and thus perhaps reduce the chronic changes. These range
from physical avoidance of sun exposure to the use of certain systemic agents.

6.7.1. Sunscreens(50)
Sunscreens protect the viable cells of the skin against actinic damage by absorbing
and reflecting the radiation impinging on the skin. Most sunscreens are chemicals that
absorb UV radiation and are incorporated in a cream, lotion, or gel vehicle. They are
usually effective as an invisible thin film and are cosmetically acceptable to most people.
The most widely used chemical sunscreens contain para-aminobenzoic acid (PABA),
PABA esters (glyceryl PABA, amyldimethyl PABA, and octyldimethyl PABA), ben-
zophenones (oxybenzone and dixoybenzone), cinnamates (octylmethoxycinnamate and
cinoxate), salicylates (homomenthyl salicylate), and anthranilates. Physical sunblocks
(zinc oxide and titanium dioxide) are usually opaque formulations that reflect and scatter
radiation. They are cosmetically unacceptable to many individuals but may be essential
for those who are unusually sensitive to UV as well as visible radiation, and for protection
of limited areas such as the nose or lips, which receive a great deal of sun exposure.
Most chemical sunscreens (e.g., PABA and PABA esters) absorb and filter out the
sunburn-producing radiation (UV-B, 290-320 nm). Only benzophenones and, to a lim-
ited extent, anthranilates exhibit sufficient additional absorption in the UV -A (320-400
nm) range to label them as broad-spectrum sunscreens.
The topical sunscreens containing PABA or PABA esters that are effective in the
prevention of erythema reactions to sunburn radiation (290-320 nm) have not proved to
be highly effective in the prevention of phototoxic and photoallergic reactions induced by
near-UV and visible radiations. A combination of two sunscreens is usually recom-
mended: (1) one that absorbs UV-B radiation and (2) one that also absorbs UV-A radia-
tion. The topical psoralen-induced phototoxic reaction (8-methoxypsoralen and 4,5' ,8-
Photomedicine 185

trimethylpsoralen) caused by 320-380 nm radiation can be reduced by certain ben-

zophenone derivatives.
The topical formulations in oil base with emollients, containing two or more light-
absorbing ingredients such as 4% ethylhexyl-p-methoxycinnamate + 3% 2-hydroxy-4-
methoxybenzophenone + 5.5% 2-phenylbenzimidazole sulfonic acid appear to be very
effective formulations in the protection of human skin against the effects of sunburn UV
radiation. Such a formulation is also effective at high elevations during skiing or mountain
climbing, and against drug-induced phototoxic reactions.
Adverse effects from topical sunscreens include yellow discoloration of clothes by
PABA, especially when the clothing is exposed to the sun. A few patients develop
contact-type eczematous dermatitis from PABA, PABA esters, benzophenones, or other
sun-screening ingredients. Patients allergic to benzocaine, procaine, para-phenylenedia-
mine, and sulfanilamide may cross-react with PABA and exhibit allergic reactions. Im-
purities in the form of ortho- and meta-esters of PABA may cause contact and photocon-
tact dermatits. Certain sunscreens cause selective burning and smarting reactions on the
face while sunbathing.

6.7.2. Physical Factors(50)

Clothing provides an optical filter against radiation penetration. Hats are essential for
preventing damage to sparsely haired scalps, and broad brims protect ears and noses and
the like. Dark, especially black, clothes absorb the light and heat, and thus may be quite
uncomfortable. Tighter weaves tend to let less radiation through to the skin. Also one
should avoid using synthetic fabrics that allow the penetration of sunburning radiation.
The time of day also plays a role. The UV-B rays reach the earth's surface from the
sun predominantly in the middle of the day. They are most intense between 11:30 A.M.
and 12:30 P.M. Avoidance of exposure between 10:00 A.M. and 2:00 P.M. will reduce the
danger of UV-B radiation exposure by 50-60%.
Melanin pigment provides the most effective physiological protection produced by
our skins. It provides an optical filter and is a stable free radical that can quench other free
radicals, which may be generated during photoreactions. Dark-complexioned people re-
quire a great deal more energy to develop sunburns or other phototoxic responses than do
light-complexioned individuals. Also dark skins are much more resistant to photocarcino-
genesis.
The stratum corneum also provides a fair to good optical barrier for the passage of
UV radiation. Thickening of the stratum corneum does protect the lower vulnerable layers
of the skin.
Through the cis-trans isomerization process, urocanic acid provides some protection
against the damaging effects of UV-B radiation.

6.7.3. Systemic Chemical Agents

6.7.3.1. Antimalarials(51)
Chloroquine (Aralen), dihydroxychloroquine (Plaquenil), and quinacrine (Atabrine)
do not alter UV-B radiation-induced sunburn or other phototoxic responses. However,
they are effective in certain photosensitive diseases such as polymorphous light eruptions
(PMLE), lupus erythematosus (LE), and solar urticaria. The mechanisms responsible for
186 Chapter 6

these effects are not clear, though their antiinflammatory properties most likely play a
role. Regular ophthalmologic examinations are necessary, however, because of the possi-
ble retinal damage that might be induced by chloroquine or Plaquenil. However, recent
evaluations indicate that the dosage schedules in use are unlikely to produce such effects.
Complete blood count (CBC) evaluations are indicated since rare hematopoietic toxic
effects may occur. Other toxic effects are noted in the references. (50,51)

6.7.3.2. J3-Carotene(52-54)
J3-Carotene and phytone, a colorless precursor of J3-carotene, have been shown to
reduce the intensity of acute UV-B radiation-induced phototoxic reactions, and J3-carotene
has been shown to inhibit UV-B radiation-induced carcinogenesis experimentally. How-
ever, these effects are quite limited and may have little, if any, clinical significance. 13-
Carotene does appear to be at least partially effective in the treatment of certain of the
porphyrias, such as erythropoietic protoporphyria (EPP). Reports on its beneficial effects
on other photosensitive diseases such as solar urticaria and hydroa aestivale require more
thorough evaluation. Recent data on its effect on PMLE are encouraging. (55) J3-Carotene
appears to produce its beneficial effects through quenching singlet oxygen.

6.7.3.3. Systemic Psoralens(8,10,56,57)

Systemic 8-methoxypsoralen (8-MOP) and 4,5' ,8-trimethylpsoralen (trisoralen) with
UV-A radiation have been used to increase pigment production, and thus lead to greater
natural protection. In general, this is not a suggested approach except under unusual
circumstances, because of the potential damage to the skin and eyes. However, 8-MOP
plus UV -A radiation has proved to be an effective therapy for patients with PMLE, actinic
reticuloid, and solar urticaria. The mechanisms of these responses, at least in the former
two diseases, may relate in part to eliminating reactive cells in the dermis. Protection
and/or stabilization of mast cells may playa role in its effects on solar urticaria.

6.7.4. Topical Psoralens(ll,12,58,59)

5-Methoxypsoralen (5-MOP) has been incorporated into sunscreens in small amounts

to accelerate tanning. Experimental studies indicate that this ingredient is phototoxic and
photomutagenic. In addition, it is photocarcinogenic experimentally in the concentrations
present in these sunscreens. Thus, it does not appear to be an appropriate approach to
protection.

6.7.5. Quick-Tanning Preparations(50)

Topical agents containing 3-5% dihydroxyacetone and oral preparations containing
J3-carotene and canthoxanthine can be used to produce color in the skin. This approach
appears to be safe, though the long-term effects of canthoxanthine have not been com-
pletely evaluated. These agents do not produce any photoprotective effect from sunburn
damage; thus the color is not photoprotective.
Photomedicine 187

6.8. PHOTOIMMUNOLOGV(6:pp.293-322,39,60-65)

Photoimmunology could be considered the overall term for all of the light-influenced
immunological phenomena, including the cutaneous photoallergic reactions noted in Sec-
tion 6.5.1.2. However, for the purpose of this discussion, the term will represent the
effects of nonionizing radiation on immune phenomena excluding these photoallergic
reactions.

6.8.1. Immediate Hypersensitivity(6:pp.298-303,39:pp.636-637)

Immediate hypersensitivity reactions are mediated through antibody formation. Anti-

bodies are produced by B lymphocytes and plasma cells following stimulation by anti-
genic substances, and T-helper and suppressor lymphocytes modulate this antibody
formation. Large amounts of UV-C radiation in vitro have been shown to inhibit various
antibody activities, i.e., agglutination, complement fixation, etc. However, the amount of
energy needed makes the clinical relevance of these findings questionable. Though sur-
face markers and viability of Band T lymphocytes can be altered by psoralen and UV-A
photosensitization (PUV A) and UV radiation in vitro, the effect of UV radiation or PUVA
on antibody formation in vivo has not been clarified as yet.
Mast cells are an integral part of IgE-mediated immediate hypersensitivity responses.
These cells can be altered by UV radiation. In vitro studies have shown that UV-B and
UV -A radiation, in small enough doses to be potentially clinically relevant, can disrupt
these cells and release their granules. In contrast, erythemogenic exposures to UV-B
radiation over 4-32 days will markedly increase the number of mast cells in a mouse ear.
A mild increase was noted with chronic UV-B radiation and none with chronic erythro-
genic amounts of UV-A exposures. Perhaps from a clinically relevant point of view the
response of these cells to PUV A therapy in vivo is of the most significance. PUV A
treatment has been shown to stabilize mast cell membranes to prevent degranulation. (66)
This relationship requires further evaluation. PUV A therapy has been found to be of value
in urticaria pigmentosum, a disease characterized by the accumulation of mast cells in the
skin, and in solar urticaria, which at least in some cases appears to be an IgE-mediated
immediate hypersensitivity response.

6.8.2. Delayed Hypersensitivity (DHS) Cell-Mediated Immune Response

(CMI)(6:pp.303-305,39:p.637)

There is an increasing body of evidence that UV radiation and PUV A have a pro-
found influence on CMI responses and DHS. Clinically, chronic sun-exposed skin is less
reactive to allergens than nonexposed skin. UV -Band PUV A irradiation has been shown
to locally and systemically suppress contact DHS in animals that had been sensitized to
dinitrochlorobenzene (DNCB). Also, localized reduction in both the initiation or sen-
sitization and elicitation phase of DHS has been reported in PUV A-treated sites in experi-
mental animals. In addition, PUVA treatment of psoriasis has been reported to suppress
DHS to DNCB in humans. UV-B as well as PUVA have been shown to suppress contact
sensitization to nitrogen mustard. The mechanism and specificity of this suppression was
not determined.
188 Chapter 6

Recent studies in mice have demonstrated the inhibition of the afferent phase of
DHS, which is associated with the formation of antigen specific T -suppressor cells by
both UV -B and PUVA irradiation. (61) There is some evidence that this affect relates to
antigen presentation to splenic macrophages in the UV-B-treated mice.(62.63) This aspect
has not been examined in PUVA-exposed mice.
The role of Langerhans cells in these radiation effects on DHS must be considered.
Langerhans cells have been implicated as antigen-processing cells for antigens presented
through the skin. UV radiation (both UV-A and UV-B) in sufficient doses can effectively
abolish surface markers of Langerhans cells in human and mouse skin without destroying
these cells.(64) However, UV radiation-induced reduction in ATPase-positive cells does
not result in the consistent inhibition of DHS reactivity. Both UV -B and PUV A inhibit the
primary epidermal cell-lymphocyte proliferation reaction, which appears to be mediated
through epidermal Langerhans cells. It appears likely that the effects ofUV radiation and
photochemotherapy on DHS is dependent at least in part on the influences of the Lan-
gerhans cells.

6.8.3. Lymphocytes and Nonionizing Radiation(6:pp.299-303)

Lymphocytes play an essential role in both humoral and cell-mediated immune

responses. UV radiation and PUVA have been shown to profoundly influence these cells
in vitro and in vivo.
In vitro studies using trypan blue dye exclusion and responses to mitogens indicated
that UV-C is more toxic than UV-B, which is more toxic than UV-A radiation. PUVA has
also been shown to be toxic to these cells in vitro. T lymphocytes appear to be more
sensitive to UV-C radiation than B lymphocytes. However, the sensitivity ofT and B cells
to UV-B radiation is approximately equal.<67)
It is obvious that UV radiation and PUVA can alter lymphocyte structure and func-
tion both in vivo and in vitro. It seems likely that effects on these cells are related to
alterations in immune function that have been recorded after such irradiation. However,
our information is as yet too preliminary to make any judgments concerning mechanisms
of effects on human immune reactions and disease.

6.9. INDIRECT EFFECTS OF LlGHT(4:pp.231-258)

A number of miscellaneous indirect effects of light energy on animal functions have

been described. These include possible control of biological rhythms (Chapter 7), gonadal
function (Chapter 9), cortisol production, and the like. One of the best characterized
indirect effects, other than vision, concerns the inhibition of melatonin formation in the
pineal organs by light energy that impinges on the retina. What general influence this
response engenders remains to be determined, though recent studies suggest that the
pineal gland may playa most important physiological role in human hormonal and central
nervous system function. These indirect effects appear to be mediated through a retinal
receptor, as yet unidentified in mammals. Other effects include reduction in systolic and
diastolic blood pressure, increase in exercise tolerance, reduction in blood cholesterol,
and decrease in serum tyrosine.
Photomedicine 189

6.10. ULTRAVIOLET RADIATION EFFECTS ON THE EYE(7:pp.231-262)

The ocular structures, the cornea, aqueous, lens, and vitreous have no chromophores
to absorb all but the shortest of visible rays. This transparency allows the visible energy to
reach the retina to allow us to see. In contrast, these tissues, especially the lens, can
absorb varying amounts of UV radiation. UV radiation has been implicated in a number of
adverse effects. Acute photokeratitis due to reflection of UV radiation from snow results
in "snow blindness." This type of reaction may occur industrially in welders, glass
workers, etc. Other conjunctival and corneal changes apparently due to UV radiation
include pingueculae, pterygiums, exposure keratosis, nodular band-shaped keratopathy,
dysplasia, and intraepithelial carcinoma. (7:p.232) Since the human cornea and the aqueous
humor transmit almost all wavelengths longer than 300 nm, these untoward effects should
be due to wavelengths of 300 nm and less. (7:pp.232,253,254) Therefore at least in humans
the lens is exposed to UV radiation between 300 and 400 nm throughout the individual's
lifetime. This chronic UV radiation exposure over the years leads to the generation of
fluorescent chromophores, which are in part responsible for the yellow discoloration of
the lens nucleus as it ages. In about one-tenth of the aged population this leads to nuclear
cataracts. UV radiation-induced cataracts have recently been reported supporting this
concept. (68)
The vitreous and the retina are protected from shorter UV-B rays by the cornea and
from the UV rays longer than 300 nm by the lens. However, in very young eyes the lens
may not absorb the above noted UV radiation. In addition, when lens damage has
occurred, i.e., with cataract formation, the lens is removed. Under these circumstances
the vitreous and the retina are exposed to UV radiation between 300 and 400 nm. The
effect of this longer UV radiation on the vitreous requires clarification, though there is
some evidence suggesting that it may produce some damage to the protein and polysac-
charide structures of this gel. The shorter UV rays below 295 nm, which are filtered out
by the cornea, have been shown to cause shrinkage of the vitreous gel and collagen
network. (69) They also can damage the polysaccharide solution due to a breakdown of the
glucosidic linkages, increasing the reducing end groups and the formation of smaller
fragments. (70)
The retina is also exposed to these longer UV rays as well as more of the short blue
visible rays when the lens is removed. Recent data implicate the UV and short visible
radiation as being able to cause retinal damage. (7:p.235)

6.11. CONCLUSION

Light energy has profound effects on mammalian cell structure and function. These
influences playa key role in the relationship between humans and their environment. The
primary effects, both detrimental and beneficial, occur at the site of impact, i.e., the skin.
The skin is not only the most common site of photoresponse but also the most available
human tissue for examination and investigation. Thus, medical photobiologists have
unique opportunities. They can examine cancerous growths as they are being formed,
evaluate radiation and chemical interactions as they relate to disease states, study the
influence of light energy on immunological processes, and so on. Though the eye is
190 Chapter 6

primarily involved in vision responses, phototoxic damage also occurs to its structures.
Again the eye is more available for study than other, more internal organs. Though the
sophistication of methodology for the study of human responses is somewhat limited
compared to other disciplines in photobiology, the significance of such studies cannot be
overemphasized. The field of medical photobiology is wide open. At present we have just
scratched the surface of understanding.

6.12. REFERENCES

1. B. E. Johnson, F. Daniels, Jf., and I. A. Magnus, Response of human skin to ultraviolet light, in:
Photophysiology, Vol. 4 (A. C. Giese, ed.), pp. 139-202, Academic Press, New York (1968).
2. F. Urbach (ed.), The Biologic Effects of Ultraviolet Radiation Pergamon Press, New York (1969).
3. J. H. Epstein, Adverse cutaneous reactions to the sun, in: Yearbook of Dermatology (F. D. Malkinson and
R. W. Pearson, eds.), pp. 5-43, Yearbook, Chicago (1971).
4. T. B. Fitzpatrick (consult. ed.), M. A. Pathak, L. C. Harber, M. Seiji, and A. Kukita (eds.), Sunlight and
Man, University of Tokyo Press, Tokyo (1974).
5. J. A. Parrish, R. R. Anderson, F. Urbach, and D. Pitts (eds.), UV-A: Biological Effects of Ultraviolet
Radiation with Emphasis on Human Responses to Longwave Ultraviolet, Plenum Press, New York (1978).
6. J. D. Regan and J. A. Parrish (eds.), The Science of Photomedicine, Plenum Press, New York (1982).
7. F. Urbach and R. W. Gange (ed.), The Biological Effects of UVA Radiation, Praeger, New York (1986).
8. I. Willis, A. Kligman, and J. H. Epstein, Effects of long ultraviolet rays on human skin: Photoprotective or
photoaugmentative? J. Invest. Dermatol. 59, 416-420 (1972).
9. D. S. Snyder and W. H. Eaglstein, Intradermal antiprostaglandin agents and sunburn, J. Invest. Dermatol.
62, 47-50 (1974).
10. G. F. Wilgram, R. L. Kidd, W. S. Krawczyk, and P. L. Cole, Sunburn effect on keratinosomes, Arch.
Dermatol. 101, 505-519 (1970).
11. K. Wier, K. Fukuyama, and W. L. Epstein, Nuclear changes during light induced depression of ribonucleic
acid and protein synthesis in human epidermis, Lab. Invest. 25, 451-456 (1971).
12. K. Fukuyama, W. L. Epstein, and 1. H. Epstein, The effect of ultraviolet light on RNA and protein
synthesis in differentiated epidermal cells, Nature 216, 1031-1032 (1967).
13. J. H. Epstein, K. Fukuyama, and K. Fye, Effects of ultraviolet radiation on the mitotic cycle and DNA,
RNA and protein synthesis in mammalian epidermis in vivo. Photochem. Photobiol. 12, 57-65 (1970).
14. G. A. Soffen and H. F. Blum, Quantitative measurements in changes in mouse skin following a single dose
of ultraviolet light, J. Cell. Compo Physiol. 58, 81-96 (1961).
15. W. S. Bullough and E. B. Laurence, Tissue homeostasis in adult mammals, in: Advances in Biology of
Skin, Vol. 7 (W. Montagna and R. L. Dobson, eds.), pp. 1-36, Pergamon Press, Oxford (1966).
16. J. J. Voorhees, E. A. Duell, and W. H. Kelsey, Dibutyryl cyclic AMP inhibition of epidermal cell division,
Arch. Dermatol. 105, 384-386 (1972).
17. J. E. Cleaver, Repair of damaged DNA in human and other eukaryotic cells, in: Nucleic Acid-Protein
Interactions-Nucleic Acid Synthesis in Viral Infection CD. W. Ribbons, J. F. Woessner, and J. Schultz,
eds.), pp. 87-112, North-Holland, Amsterdam (1971).
18. J. E. Cleaver, Repair processes for photochemical damage in mammalian cells, Adv. Radiat. Bioi. 4, 1-75
(1974).
19. P. D. Forbes, F. Urbach, and R. E. Davies, Photocarcinogenesis in hairless mice: Dose-response and
influence of dose-delivery, Photochem. Photobiol. 34, 361-365 (1981).
20. A. R. Lehman, S. Kirk-Bell, C. F. Arlett, M. C. Paterson, P. H. M. Lohman, E. A. de Weerd-Kastelein,
and D. Bootsma, Xeroderma pigmentosum cells with normal levels of excision repair have a defect in DNA
synthesis after UV-irradiation, Proc. Natl. Acad. Sci. USA 2, 219-223 (1975).
21. R. L. Olson, J. Nordquist, and M. A. Everett, The role of epidermallysosomes in melanin pigmentation,
Br. J. Dermato!. 83, 189-199 (1970).
22. J. H. Epstein, K. Fukuyama, and R. L. Dobson, Ultraviolet light carcinogenesis, in: The Biologic Effects of
Ultraviolet Radiation (F. Urbach, ed.), pp. 551-568, Pergamon Press, New York (1969).
Photomedicine 191

23. W. L. Epstein, K. Fukuyama, and 1. H. Epstein, Early effects of ultraviolet light on DNA synthesis in
human skin in vivo, Arch. Dermatol. 100, 84-89 (1969).
24. L. H. Kligman, F. 1. Akin, and A. M. Kligman, The contributions ofUV-A and UV-B to connective tissue
damage in hairless mice, J. Invest. Dermatol. 84,272-276 (1985).
25. H. F. Blum, Carcinogenesis by Ultraviolet Light, Princeton University Press, Princeton (1959).
26. 1. H. Epstein, Ultraviolet carcinogenesis, in: Photophysiology, Vol. 5 (A. C. Giese, ed.), pp. 235-273,
Academic Press, New York (1970).
27. 1. H. Epstein, Photocarcinogenesis, skin cancer and aging, J. Am. Acad. Dermatol. 9,487-506 (1983).
28. A. W. Kopf, M. L. Kripke, and R. S. Stern, Sun and malignant melanoma, J. Am. Acad. Dermatol. 11,
674-684 (1984).
29. 1. H. Epstein, How important is the role of radiation from the sun in the rising incidence of melanomas?, in:
Controversies in Dermatology (E. Epstein, ed.), pp. 10-12, Saunders, Philadelphia (1984).
30. M. L. Kripke, Immunology of UV-induced skin cancer: Yearly review, Photochem. Photobiol. 32, 837-
839 (1980).
31. R. A. Daynes, E. J. Bernhard, M. F. Gurish, and D. H. Lynch, Experimental photoimmunology: Immu-
nologic ramifications of UV induced carcinogenesis, J. Invest. Dermatol. 77, 77-85 (1981).
32. R. D. Granstein, I. A. McAuliffe, D. 1. Waltenbaugh, and M. I. Greene, Immunologic inhibition of
ultraviolet radiation-induced tumor suppressor cell activity, Science 224, 615-617 (1984).
33. 1. E. Cleaver and D. Bootsma, Xeroderma pigmentosum: Biochemical and genetic characteristics, Ann.
Rev. Genetics 9, 19-38 (1975).
34. V. M. Maher, L. M. Oulette, R. D. Curren, and I. I. McCormick, Frequency of ultraviolet light-induced
mutations is higher in xeroderma pigmentosum variant cells than in normal human cells, Nature 261,593-
595 (1976).
35. I. H. Epstein, Photoallergy: A review, Arch. Dermatol. 106, 741-748 (1972).
36. I. H. Epstein and B. U. Wintroub, Photosensitivity due to drugs, Drugs 30, 42-57 (1985).
37. H. W. Lim, L. Young, M. Hagen, and I. Gigli, Delayed phase of hematoporphyrin-induced phototoxicity:
Modulation by complement, leukocytes and antihistamines, J. Invest. Dermatol. 84, 114-117 (1985).
38. H. W. Lim, M. Hagen, and I. Gigli, Phototoxicity in mast cell deficient, C5 deficient and leukopenic mice,
Clin. Res. 33, 23A (1985).
39. I. H. Epstein, Photoallergy and photoimmunology, in: Dermatologic Immunology and Allergy (1. Stone,
ed.), pp. 629-640, Mosby, St. Louis (1985).
40. H. F. DeLuca, Vitamin D: The vitamin and the hormone, Federation Proc. 33, 2211-2219 (1974).
41. W. F. Loomis, Skin-pigmentation regulation of vitamin D biosynthesis in man, Science 157, 501-506
(1967).
42. K. H. Kaidbey and A. M. Kligman, Photopigmentation with trioxsalan, Arch. Dermatol. 109, 674-677
(1974).
43. I. A. Parrish, T. B. Fitzpatrick, L. Tanenbaum, and M. A. Pathak, Photochemotherapy of psoriasis with
oral methoxsalen and longwave ultraviolet light, N. Engl. J. Med. 291, 1207-1211 (1974).
44. L. E. Bockstahler, C. D. Lytle, and K. B. Hellman, A review of photodynamic therapy for herpes simplex:
Benefits and potential risks, DHEW Publication No. (FDA) 75-8013. Available from the Bureau of Radi-
ological Health, Rockville, MD 20852 (1974).
45. D. R. Doiron and G. S. Keller, Porphyrin photodynamic therapy: Principles and clinical applications, in:
Therapeutic Photomedicine (H. Honigsmann and G. Stengl, eds.), pp. 85-93, Karger, Basel (1986).
46. A. lesionek, Tappenier zur Behandlung des Hautcarcinomes mit fluoreszierenden Stoffen, Munch. Med.
Wochenschr. 41, 2042-2044 (1903).
47. T. I. Dougherty, W. R. Potter, and K. R. Weishaupt, The structure of the active component of hema-
toporphyrin derivative, in: Porphyrin Localization and Treatment of Tumors (D. R. Doiron and C. 1.
Gomer, eds.), No. 170, pp. 301-314, Alan R. Liss, New York (1985).
48. D. R. Doiron, L. O. Svaasand, and A. E. Profio, Light dosimetry in tissue application to photoirradiation
therapy, in: Porphyrin Photosensitization (D. Kessel and T. I. Dougherty, eds.), p. 63, Plenum Press, New
York (1983).
49. Phototherapy in the Newborn, Final report of the Committee Division of Medical Sciences, National
Research Council, 2101 Constitution Ave., Washington, D.C. 20418 (1974).
50. M. A. Pathak, Sunscreens: Topical and systemic approaches for protection of human skin against harmful
effects of solar radiation, J. Am. Acad. Dermatol. 7, 285-312 (1982).
192 Chapter 6

51. L. Tanenbaum and D. L. Tuffanelli, Antimalarial agents: Chloroquine, hydroxychloroquine and quina-
crine, Arch. Dermatol. 16, 587-591 (1980).
52. M. M. Mathews-Roth, M. A. Pathak, T. B. Fitzpatrick, L. C. Harber, and E. H. Kass, Beta-carotene
therapy of erythropoietic porphyria and other photosensitive diseases, Arch. Dermatol. 113, 1229-1232
(1977).
53. J. H. Epstein, The effects of beta-carotene on ultraviolet induced cancer formation in the hairless mouse
skin, Photochem. Photobiol. 25, 211-213 (1977).
54. M. M. Mathews-Roth and N. I. Krinsky, Carotenoid dose level and protection against UV-B induced skin
tumors, Photochem. Photobiol. 42, 35-38 (1985).
55. C. T. Jensen, Oral carotinoid treatment in polymorphous light eruption: A cross-over comparison with
oxychloroquine and placebo, Photodermatol. 2, 166-169 (1985).
56. J. D. Imbraie, F. Daniels, L. Bergeron, C. E. Hopkins, and T. B. Fitzpatrick, Increased erythema threshold
6 weeks after a single exposure to sunlight plus oral methoxsalen, 1. Invest. Dermatol. 32,331-337 (1959).
57. R. S. Stem, N. Laird, J. Melski, J. A. Parrish, T. B. Fitzpatrick, and H. L. Bleich, Cutaneous squamous
cell carcinoma in patients treated with PUVA, N. Eng/. 1. Med. 310, 1156-1161 (1984).
58. M. J. Ashwood-Smith, G. A. Poulton, M. Baker, and M. Mildenberger, 5-Methoxypsoralen, an ingredient
in several sunscreen preparations had lethal metagenic and clastogenic properties, Nature 285, 407-409
(1980).
59. L. E. Cartwright and J. F. Walter, Psoralen containing sunscreen is tumorigenic in hairless mice, 1. Invest.
Dermato/. 80,306 (1983).
60. W. L. Morison, Photoimmunology, J. Invest. Dermatol. 77,71-76 (1981).
61. M. L. Kripke, W. L. Morrison, and J. A. Parrish, Systemic suppression of contact hypersensitivity in mice
by psoralen plus UVA radiation (PUVA), 1. Invest. Dermatol. 81, 87-92 (1983).
62. M.1. Green, M. S. Sy, M. L. Kripke, and B. Benacerraf, Impairment of antigen-presenting cell function by
ultraviolet radiation, Proc. Natl. Acad. Sci. USA 76, 6592-6595 (1979).
63. F. P. Noonan, M. L. Kripke, G. M. Pederson, and M. I. Green, Suppression of contact hypersensitivity in
mice by ultraviolet radiation is associated with defective antigen presentation, Immunology 43, 527-533
(1981).
64. W. Aberer, G. Shuler, G. Sting!, H. Honigsman, and K. Wolff, Ultraviolet light depletes surface markers
of Langerhans cells, 1. Invest. Dermatol. 76, 202-210 (1981).
65. G. A. Elmets and P. R. Bergstresser, Ultraviolet radiation effects on inunune processes, Photochem.
Photobiol. 36, 715-719 (1982).
66. K. Danno, K. Toda, and T. Horio, The effect of 8-methoxypsoralen plus long-wave ultraviolet (PUVA)
radiation on mast cells: PUV A suppresses degranulation of mouse skin mast cells induced by compound
48/80 on concanavalin A, J. Invest. Dermatol. 85, 1l0-1l4 (1985).
67. D. J. Cripps, S. Horowitz, and R. Hong, Spectrum of ultraviolet radiation on human B and T lymphocyte
viability, CUn. Exp. Dermatol. 3,43-50 (1978).
68. S. Lerman, Human UV radiation cataracts, Ophthalmic Res. 12, 303-314 (1980).
69. E. A. Balazs, T. C. Laurent, A. F. Howe, and L. Varga, Irradiation of mucopolysaccharides with
ultraviolet light and electrons, Radiation Res. 11, 149-164 (1959).
70. S. Lerman, K. Gardiner, J. Megew, and R. Borkman, The prevention of direct and photosensitized UV
radiation damage to the occular lens, Ophthalmic Res. 13, 284-292 (1981).
 7
Circadian Rhythms

7.1. Introduction ...................................................................... 193

7.2. General Features of Circadian Rhythms. . . . . . . . . . . . .. . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . .. 194
7.2.1. Entrainment to Artificial Light and Temperature Cycles .......... . . . . . . . . . . . . . . . . . .. 194
7.2.2. "Free-Running" (Nonentrained) Rhythms ........................................ 195
7.2.3. Temperature Compensation .................................................... 196
7.3. Entrainment to Environmental Cycles .................................................. 197
7.3.1. Formal Aspects of Entrainment by Light ......................................... 198
7.3.2. Nature of the Photoreceptor .................................................... 200
7.4. Cellular Mechanisms of Circadian Rhythms ....................................... ,..... 201
7.4.1. Biochemical Rhythms ........................................................ 202
7.4.2. Studies Using Inhibitors. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 203
7.5. Genetic Approaches to Circadian Rhythms ............................................. 204
7.5.1. Genetic Analysis of Clock Mutants ............................................. 204
7.5.2. Molecular Analysis of Clock Mutants ........................................... 206
7.6. Circadian Organization in Multicellular Organisms ....................................... 206
7.6.1. Localization of Circadian Clocks in Invertebrates .................................. 206
7.6.2. Localization of Circadian Clocks in Vertebrates ................................... 207
7.7. Circadian Rhythms in Humans ....................................................... 208
7.8. Conclusions ...................................................................... 211
7.9. References ....................................................................... 211

7.1. INTRODUCTION(1-5)

A basic concept in biology is that organisms are adapted to their environment. However,
the environment is a constantly changing and highly variable milieu that exhibits dramatic
fluctuations in such factors as light and temperature. An organism apparently well adapted
to the environment at one time may be poorly adapted only a short time later if it cannot
modify its physiology or behavior. Some of the environmental changes are quite unpre-
dictable, such as those associated with day-to-day changes in the weather and general
atmospheric conditions. However, a number of environmental changes are highly predict-
able because they result from specific planetary movements that exhibit precise peri-
odicities. For example, the daily day/night cycle is caused by the earth's rotation about its
axis, tidal and lunar cycles depend on the revolution of the moon around the earth, and
annual cycles reflect the revolution of the earth about the sun.

Jerry F. Feldman • Thimann Laboratories, University of California, Santa Cruz, Santa Cruz,
California 95064.

193
194 Chapter 7

Organisms that can adjust to changes in the environment are likely to exhibit a
greater degree of adaptiveness than those that cannot. For the unpredictable kinds of
changes, an organism needs systems that respond directly to environmental changes.
However, for the periodic and predictable fluctuations, specific mechanisms have evolved
that generate endogenous biological oscillations whose periodicities correspond directly
with certain environmental periodicities. These internal oscillations produce rhythmicities
in the organism that are generated by negative feedback systems wholly within the
organism, and thus do not depend directly on the fluctuations in the environment for their
rhythmicity. However, these systems use the periodic information from the environment
to couple the biological oscillation to the environmental cycle.
Thus, biological rhythms with daily, tidal, lunar, and annual periodicities have been
identified in nearly all eucaryotic organisms, including plants, animals, and micro-
organisms, that affect a remarkable diversity of functions, including sleep/activity pat-
terns, hormone levels, metabolic rate and body temperature in animals, photosynthesis,
leaf position and sap exudation in plants, and cell division, phototaxis, photosynthesis,
and bioluminescence in microorganisms. Among the rhythms that have been studied,
those analyzed in most detail are daily rhythms known as circadian rhythms. Circadian
rhythms are those daily rhythms that are the result of an endogenous timer, or circadian
clock, that oscillates with a periodicity of about one day and that can be synchronized, or
entrained, to the daily environmental cycle, using primarily information from the daily
light-dark cycle (or sometimes the daily temperature cycle). In this chapter we will
examine the general nature and properties of circadian rhythms and provide examples in
different organisms. We will examine what is known about the physiological and bio-
chemical mechanisms of the rhythmicity and the kinds of experimental approaches that
are being used to study this question. Finally, we will discuss some special and unique
questions that arise in the consideration of circadian rhythms in humans.

7.2. GENERAL FEATURES OF CIRCADIAN RHYTHMS(I-5)

As stated above, circadian rhythms have been the most intensively studied class of
biological rhythms and exhibit a number of unique properties that distinguish them from
other kinds of daily rhythms.

7.2.1. Entrainment to Artificial Light and Temperature Cycles

Circadian rhythms can be synchronized (entrained) to artificial light-dark cycles or
temperature cycles that are either exactly 24 h in length or close to 24 h. However, there
are rather severe limits to which the biological rhythm can be forced away from its
"natural" 24-h periodicity. In some organisms, such as invertebrates, entrainment to
cycles as short as 19 h or as long as 29 h can be achieved, but in mammals this range is
much narrower and is rarely wider than about 22-26 h.
In most cases the environmental factor that serves as the primary entraining agent is
the light-dark cycle (Fig. 7-1). It is because of the importance of the light-dark cycles
and the sensitivity of circadian clocks to both visible and ultraviolet light that pho-
tobiologists have become particularly interested in circadian clock mechanisms.
Considerable progress has been made in understanding both the conceptual strategy
Circadian Rhythms 195

o HOURS 24 o 24
I I I I

~24 HOURS~

1 HOUR UGHT

50

100
18 HOURS LIGHT

PER D~

150

DAYS

o HOURS 24 o 2.(

Fig. 7-1. The circadian rhythm of locomotry activity (measured on a running wheel) of the field mouse Per-
moyscus maniculatus under free-running and entrained conditions. A 1:23 light-dark cycle was imposed from
day 0 to day 59. The rhythm was entrained by day 6. From day 60 to day 92, the rhythm free-ran in constant
darkness with a period length slightly less than 24. From day 93 to day 132, an 18:6 light-dark cycle was
imposed, and the rhythm again entrained to the 24-h cycle. After day 132, the rhythm freeran in constant
darkness, again with a period length somewhat shorter than 24 h. (From Ref. 6.)

by which light-dark cycles entrain circadian clocks and the identification and nature of
the photoreceptor molecules involved in the photoreception process. Indeed, because so
little is known about the molecular mechanisms of the oscillator itself (Section 7.4), an
understanding of the molecular mechanisms of clock photoreception offers one of the
most important approaches to studying the biochemical mechanisms of circadian timing.

7.2.2. "Free-Running" (Nonentrained) Rhythms

When organisms are maintained in constant conditions in the laboratory (usually
constant temperature and constant darkness, although constant light is used with some
organisms) isolated from the external environmental cycle, rhythmicity persists with a
196 Chapter 7

period length that is close to, but not exactly equal to, the 24-h period of the earth's
rotation (Fig. 7-1). This small but significant deviation from 24 h of the "free-running"
(i.e., not entrained) period length results in the biological rhythm drifting out of syn-
chrony (out of phase) with the environmental cycle and is the strongest single piece of
evidence that the rhythm is caused by an endogenous oscillation rather than simply being a
response to periodic changes in the environment.
It is worth mentioning at this point the origin, or rationale, for two of the terms used
here. First, "circadian", coined by Halberg in 1959, comes from the Latin words circa,
meaning about, and diem, meaning day, and emphasizes the importance of the free-
running period length not being exactly 24 h. Second, the term "clock" is used to
describe the actual timing or oscillator mechanism because it serves a specific time-
keeping function for the organism, i.e., the entraining mechanism not only matches the
period length of the biological oscillator to the environmental cycle but also "phases" the
biological rhythmicities so that the times of maxima and minima occur at the most
"appropriate" time of day for each process. For example, organisms that show a circa-
dian rhythm of photosynthesis have their maximum about midday and their minimum
about the middle of the night. On the other hand, for rhythms of bioluminescence, the
phase of the rhythm is almost completely reversed, with the maximum occurring in the
middle of the night and the minimum in the middle of the day.

7.2.3. Temperature Compensation

The period length of a circadian rhythm in constant conditions (i.e., the free-running
period length) is the same over a wide range of physiological temperatures. This property
is known as temperature compensation and is expressed methematically by a parameter
called QlO' which is usually given as the rate of a process at temperature t divided by the
rate at temperature t - 10. For circadian rhythms the period length of the clock, rather than
its rate, is used in this calculation, and since period length is inversely related to rate, the
Q 10 is given as the period length at temperature t divided by the period length at tem-
perature t + 10. The value of the Q 10 for circadian rhythms is usually about 0.9-1.1,
indicating that the period length is essentially the same at different temperatures. This QlO
value contrasts sharply with the temperature dependence of most chemical, biochemical,
and physiological processes, whose QlO is often 2 or greater, i.e., their rate approximately
doubles for each 100 e increase in temperature.
The apparent temperature independence of the circadian clock theoretically could
occur if the rate-limiting step in clock timing were a physical process, such as diffusion,
whose rate is proportional to the absolute temperature. Within physiological temperature
ranges (approximately 288-308°K), the rate would therefore not vary significantly. On
the other hand, QIO values close to 1 could also be obtained by a temperature compensa-
tion mechanism, in which individual rate-limiting components of the system are tem-
perature-dependent reactions with relatively high QIO values but which interact in such a
way that the rate of the final output of the system does not vary much with temperature.
Several types of temperature compensation models have been proposed for circadian
clocks. In one type, (7) a primary reaction A --? B interacts with a secondary reaction e--?
D. D serves as an inhibitor of the A --? B reaction. When the temperature is raised, both
reactions speed up; the increased rate of the secondary reaction results in the accumulation
Circadian Rhythms 197

of a higher concentration of D. As a result, the increased inhibition by D of the primary

reaction compensates for the increased rate of A ---7 B produced by the temperature
increase.
A second type of temperature compensation modeiC 8 ) is based on the observation that
many organisms maintain a constant fluidity of their membranes at different temperatures.
This is accomplished by increasing the ratio of saturated to unsaturated fatty acids as the
temperature is increased. If the rate-limiting processes of the clock were dependent on the
fluidity of the membrane, the clock would operate at a similar rate at different tempera-
tures.
Several lines of evidence tend to favor temperature compensation rather than tem-
perature independence:
1. Q IO values of less than 1 have been reported in a number of circadian systems. (9)
This result is difficult to explain by a temperature independence model. However, if a
temperature compensation system was not absolutely precise (the compensating event did
not exactly match the primary event), it would be just as plausible to have a system
slightly "overcompensated" (QIO < 1) as one slightly undercompensated (QIO > 1).
2. Circadian rhythms can be reset, or phase-shifted, by temperature steps or tem-
perature pulses. (9) This shows that a circadian clock can detect differences in temperature
and can adapt to the new conditions rather than "ignore" them. In fact one mathematical
model for temperature compensation(lO) specifically predicts much of the data on phase
shifting by temperature steps. The primary notion is that there is a time lag associated with
the adaptation process. For example, in a temperature step up, the A ---7 B reaction (see
above) initially would run faster until the new, higher steady-state level of D is achieved.
This would result in a resetting of the clock "ahead," i.e., a phase advance. On the other
hand, a temperature step down would result in the primary reaction running slower until
the new, lower steady-state level of D was achieved and a phase delay would result. In
most cases where such experiments have been done, this is exactly the result that is seen;
temperature steps up cause phase advances, and temperature steps down cause phase
delays.
3. At the high and low ends of physiological temperatures for many organisms, the
temperature compensation is lost and the QIO rises significantly above 1. This indicates
that the system can operate efficiently only within certain limits. In the models above, for
example, there could be upper and lower limits for the concentration of substance D or
limits to the binding of D to the enzyme that controls the primary reaction A ---7 B, or
limits to the ratio of saturated to unsaturated fatty acids that control membrane fluidity.
4. Mutants have been isolated in which the temperature compensation mechanism is
either altered or entirely lost, although the clock appears to run normally or nearly
normally in most other respects.(ll) It is difficult to imagine how this might occur in a
temperature-independent clock system, but it is quite plausible for a temperature-compen-
sated system.

7.3. ENTRAINMENT TO ENVIRONMENTAL CYCLES(l2)

The synchronization, or entrainment, of the circadian clock to the environmental

cycle is a complex phenomenon whose mechanisms have not yet been elucidated. There
198 Chapter 7

have been two major approaches to the analysis of the mechanisms of entrainment of
circadian rhythms. One approach considers the clock as a "black box" and studies the
general conceptual strategies by which the light information is utilized by the circadian
oscillator. In such studies, pioneered primarily by Colin Pittendrigh and his co-workers,
the specific cellular or molecular details of the circadian system are not at issue; rather,
what have been called the "formal properties" of the system have been examined. For
example, what kinds of mathematical equations or models can be constructed to explain
the behavior of the system in response to different types of light cycles and light interrup-
tion, or perturbation, experiments? Does the light cycle entrain the biological oscillator by
changing the rate of the biological oscillation or by periodically resetting it once or twice
each cycle?
The second type of approach is complementary to the first in that it is specifically
attempting to identify the cellular and molecular components involved in the photorecep-
tion mechanism and the transducing of information from the photoreceptor to the clock
oscillator. A variety of studies of this type have been carried out, some of which will be
described here; others are discussed in Chapter 8, "Extraretinal Photoreception."

7.3.1. Formal Aspects of Entrainment by Light(13,14)

One of the most important conclusions of these studies has been the recognition that
in a normal light-dark cycle, dawn (lights on) and dusk (lights off) provide the key signals
for entrainment. These transitions result in a rapid (less than 1 h) resetting of the clock
mechanism, the amount and direction being determined by the time of day at which the
transition occurs (dawn or dusk) and the free-running period of the circadian oscillator.
Usually, but not always, dawn produces an advance (+) phase shift while dusk produces a
delay (-) phase shift. Thus, if an organism has a clock whose free-running period length
is 23 h, in order to entrain to a 24-h cycle, the net amount of resetting of its clock must be
-1 h each day. For an organism whose clock has a free-running period of 25 h, the signals
must produce a net advance of + 1 h each day.
Evidence for resetting as the primary mechanism of entrainment has come from
several types of experiments. One of the first involved what are called "skeleton"
photoperiods. Skeleton photoperiods mimic "normal" light-dark cycles in the following
way: If a normal cycle was 10 h light, 14 h dark (LD 10: 14), a skeleton LD 10: 14 would
be one in which a brief pulse of light was given at the beginning and end of the pre-
sumptive 1O-h "day" phase (to simulate the dawn and dusk signals), but between these
two brief pulses, darkness would prevail. In Drosophila an extensive series of experi-
ments were carried out in which the phase of emergence time of the adult fly (eclosion),
which is controlled by the circadian clock, was determined for different normal and
skeleton photoperiods. Remarkably, the times of eclosion under the two light regimes,
which differ dramatically in the amount of light seen by the organism, were quite similar
for a large number of different photoperiodic cycles, ranging from LD 1 : 23 to LD 13 : 11.
For photoperiods with longer light times, the organism always interpreted the shorter
interval as the day phase, but with this adjustment, agreement between the two types of
cycles extended to much longer photoperiods up to LD 23 : 1.
These results indicate that the light transition provides a signal, either a strong
increase in intensity (dawn) or a strong decrease in intensity (dusk), that resets the
Circadian Rhythms 199

~ +10
o
oS

~~
Fig. 7-2. Phase response curves from three organisms showing the a I
I
phase-shifting effects of single light pulses. (abscissa) Time of day at
which the light pulse was administered; (ordinate) phase shift pro-
duced by the light pulse (+ = advance; - = delay). (Redrawn from
Ref. 15.) Time of Light Pulse (hours)

biological clock twice a day in a manner that synchronizes it to the external environment
cycle. Since the skeleton photoperiod produced entrainment in a manner similar to the
normal (full) photoperiod, this hypothesis also suggests that the continued presence of
light during the day phase plays only a very minimal role in the entrainment of the clock.
A curious but still unexplained aspect of this model is that the direction of the light
transition (on or off) is not important in conveying phase information to the clock.
Further evidence for the "resetting" role of light in entrainment has been obtained
by determining the effects of single pulses of light when such pulses are given to an
organism otherwise kept in constant darkness. In a typical experiment, a set of organisms
(or cultures of organisms) would be subjected to a single light pulse, each culture receiv-
ing the pulse at a different phase of the cycle. The effect of the light pulse at different
phases of the cycle would be determined by comparing the phase of each pulsed culture
with that of a control culture that had not received any light signals during the experiment.
In most cases light pulses given during the organism's late night or early day phase
(organism time is called "subjective" time) produce an advance phase shift while light
pulses given during the late subjective day or early subjective night produce a delay phase
shift. These results can be plotted as a "phase response curve" in which the amount of the
phase shift is plotted against the time in the cycle when the pulse was administered (Fig.
7-2). The data from the single pulse experiment (i.e., the phase response curve) success-
fully predict the behavior of the culture (i.e., the phase of the rhythm) in the skeleton
photoperiods and therefore the phase of the rhythm in normal light dark cycles.
On the other hand, there are a number of experiments that indicate that the continu-
ous action of light does have an effect, if only minor, on the behavior of the circadian
clock. The best documented cases involve comparison of the period length of circadian
rhythms in organisms maintained in constant darkness or at different intensities of con-
stant light. Such experiments show that the intensity of light affects the period length, i.e.,
the rate at which the clock runs, with differences of several hours observed in the most
extreme cases. The direction of change of period length depends on whether the animal is
nocturnal or diurnal, and these relationships have been summarized in what are called
200 Chapter 7

Aschoff's rules, after their discoverer, Professor Jurgen Aschoff. Aschoff's rules are
discussed in more detail in Chapter 8.

7.3.2. Nature of the Photoreceptor

In Chapter 8 there is a discussion of photoreceptor pigments in several animal and
plant systems, and the evidence for the role of rhodopsin, or rhodopsinlike pigments, for
circadian clocks in a number of animals is presented. In these cases, identification of the
clock photoreceptor has rested only on action spectrum data, in which the efficacy of
different wavelengths of light in resetting the clock has been measured. Because of the
inherent difficulties of measuring phase shifting with great accuracy in many organisms,
such experiments at best provide a general indication of the type of molecule involved and
frequently are unable to distinguish among several classes of molecules.
However, in the filamentous fungus Neurospora crassa, clock photoreceptor studies
have been carried out in much more detail with the use of mutant strains deficient in
various types of pigment molecules. Neurospora has a circadian rhythm of asexual spore
formation (conidiation), which can be easily assayed in cultures growing across an agar
medium (Fig. 7-3). Because the culture leaves a permanent record of the position, and
hence the time, of formation of the conidia, no automated equipment is needed to monitor
the culture continuously, and as a result a large number of cultures can be handled rapidly
and inexpensively. (17)
Action spectra for the Neurospora clock have indicated that the primary photorecep-
tor belongs to a class of photoreceptors widespread in higher plants and microorganisms
known as the "blue light" photoreceptor. In general, action spectra alone for blue light
photoreceptors have not been able to identify a unique class of molecules as the pho-
toreceptor, and in particular to distinguish whether the molecule is a carotenoid or a
flavin. Experiments using Neurospora mutants point strongly to the involvement of a
flavin in clock photoreception.
1. An action spectrum for the suppression of the conidiation rhythm by constant
light indicated effective wavelengths of 350-520 nm, with a maximum effect at 465 nm,
and other peaks at 485,415, and 375 nm (Fig. 7-4). An albino strain carrying mutations at
two different steps in carotenoid biosynthesis and containing no detectable carotenoids
showed no alteration in this light response. (19)

RACE TUBE CULTURE (side view)

Fig. 7·3. The circadian rhythm of conidiation in the fila·
mentous fungus Neurospora crassa. The culture grew
from left to right in the diagrams. The "decision" to form

>.-;"
24 hours of growth 21.5-hour period conidia is made just behind the growth front at times of
inoculation point ~ ~ day determined by the circadian clock. Thus. the position
'-.. "';b<C C; J.·.4•.,.... of the conidial bands reflects the position of the growth
1 1 front at the "decision time" and can therefore be used
conidial band present growth front directly to determine the time of day at which the conidia
RACE TUBE CULTURE (top view) were formed. (From Ref. 16.)
Circadian Rhythms 201

NEUROSPORA CRASSA

PECTINOPHORA GOSSYPIELLA

Fig. 7-4. Action spectra for setting the circa-

dian rhythm of conidiation in the filamentous
fungus Neurospora crassa and the egg-hatching
rhythm of the moth Pectinophora gossypiella.
(From Ref. 15, data for Pectinophora from Ref. 400 480 560 640
18, for Neurospora from Ref. 19.) WAVELENGTH (nm)

2. A plasma membrane fraction of Neurospora has been isolated that exhibits a

flavin-mediated blue light-induced absorbance change indicative of the photoreduction of
a b-type cytochrome. Similar blue light-induced absorbance changes have recently been
found in other membrane fractions of Neurospora, including the endoplasmic reticulum
and the mitochondria. This in vitro system has been correlated with in vivo clock pho-
toreception in several ways. First, in the respiratory mutant poky, the threshold intensity
for inhibition of banding in continuous light was about 50-fold higher than in the wild-
type strain. Correlated with this reduction in light sensitivity was a reduction in the
amount of nonmitochondrial cytochrome to 16% of that in the wild-type strain. The
absence of a direct quantitative correlation between light sensitivity and cytochrome levels
in these experiments could have been due to the use of different growth conditions under
which the two measurements were made, since poky undergoes significant biochemical
changes during an adaptation process. (20)
Second, addition of the photodynamically active dye methylene blue to the in vitro
plasma membrane preparation resulted in the sensitization of cytochrome reduction to red
light. In a similar way, addition of methylene blue to the medium of a culture growing in
race tubes resulted in the sensitization of its clock to red light for both the suppression by
constant light and phase shifting by light pulses. (20)
Evidence for the involvement of the flavin component of the flavin/b-cytochrome
complex in the clock photoreceptor has also come from experiments with mutants. Two
riboflavin auxotrophs, rib-l and rib-2, were grown under conditions where riboflavin was
limiting for growth. In both mutants a reduction in the levels of FAD and FMN in the
mycelia correlated with a reduction in the sensitivity of the clock to light for both the
phase-shifting and damping responses. Furthermore, addition of exogenous riboflavin
analogs to the medium restored light sensitivity to these strains. (21)

7.4. CELLULAR MECHANISMS OF CIRCADIAN RHYTHMS(22)

In general, two types of approaches have been used to try to identify the biochemical
components of the clock oscillator system. In one approach circadian rhythms in bio-
chemical parameters have been identified and their role in the clock analyzed. In the
202 Chapter 7

second approach, specific drugs or inhibitors have been tested to determine which ones
can perturb circadian timing systems and thus implicate the target of the inhibitor in the
clock mechanism. These approaches have some shortcomings in experimental design and
interpretation that have limited their contributions to our understanding of clock mecha-
nisms, but some important and interesting results have nevertheless emerged from them
and are summarized below.

7.4.1. Biochemical Rhythms(2.17)

Biochemical rhythms have been identified in a wide variety of parameters in many

different organisms. For example, in microorganisms there are rhythms in DNA, RNA,
and protein synthesis, in both the substrate and enzyme associated with bioluminescence,
in photosynthesis, in pathways of intermediary metabolism, including glycolysis, the
hexose monophosphate shunt, and the Krebs cycle, in the levels of ATP and cofactors
such as NAD and NADP, and in fatty acids. A similar diversity of rhythms in higher
organisms also exists. In higher plants, for example, there are rhythms in dark CO 2
fixation and protein synthesis, while in mammals there are rhythms in the levels of
coenzyme A, glycogen, and succinic dehydrogenase, in tyrosine aminotransferase activity
in the liver, and in tyrosine levels in the blood. In fact, this is only a small sample of the
various parameters that show circadian oscillations, and there are now documented liter-
ally hundreds of enzymes, subtrates, cofactors, hormones, and other substances that show
circadian oscillations in various organisms.
However, the existence of a rhythm in a particular biochemical parameter does not in
itself implicate that parameter as part of the clock mechanism. An important experiment
involving the phtotsynthesis rhythm in the marine dinoflagellate Gonyaulax illustrates this
point. (23) In this experiment, photosynthesis was blocked by the addition of dichloro-
phenylidimethylurea (DCMU), but when the inhibitor was washed out, photosynthesis
resumed in a rhythmic manner, and the phase of the rhythm was identical with that of a
control culture that had not be exposed to DCMU. This result was obtained independent of
the time of addition of DCMU, time of its removal, or duration of the pulse, and indicates
that although the process exhibiting the overt biochemical rhythm was inhibited, the
timing mechanism underlying the rhythmicity persisted throughout the experiment.
Therefore, the biochemical events of photosynthesis itself are not part of the timing
mechanism. As a result of this experiment and others like it, a distinction has been made
between the "hands of the clock" (e.g., photosynthesis) and the clock mechanism.
Similar results have been obtained with nearly all other biochemical rhythms, and as a
result it is now generally accepted that the existence of a biochemical rhythm alone tells
little about the mechanism of circadian timing.
On the other hand, several interesting studies have been carried out using bio-
chemical rhythms as starting points for analyzing clock mechanisms. The rationale of
such studies has been to identify the immediate cause of the biochemical fluctuation, and
then to determine the cause of that rhythm and so forth, ultimately "working back" from
the external hand of the clock to the actual mechanism. Although this has not yet been
fully accomplished in any organism, the initial steps have been taken in two systems,
photosynthesis and bioluminescence. In both Gonyaulax and Euglena circadian rhythms
in photosynthesis have been traced to changes in the light reactions. (24.25) More specific
Circadian Rhythms 203

identification of the specific components of the light reactions involved in photosynthetic

rhythms awaits further study.
In the case of bioluminescence, rhythmicity has been found in both the level of
substrate (luciferin) and the activity of the enzyme involved (luciferase). Recent studies
using an antibody against purified luciferase have shown that the rhythm in luciferase
activity is the result in changes in the amount of enzyme present in the cell at different
times of the day, with low levels during the day phase when luminescence is at its
minimum and high levels during the night when it is at its maximum. Thus, the luciferase
rhythm is the result of rhythms in the synthesis and! or degradation of the enzyme. (26)

7.4.2. Studies Using Inhibitors(22)

An alternative biochemical approach has been to apply a variety of drugs or inhib-
itors to organisms expressing a circadian rhythm in constant conditions to deter-
mine whether the drug can perturb the circadian timing system. In such studies one asks
whether the continued presence of the drug can change the free-running period length
or whether a pulse of the drug can reset (phase-shift) the clock. In such experiments one
hopes to be able to identify the known cellular target of the drugs as part of, or closely
coupled to, the clock mechanism.
A wide variety of inhibitors has now been shown to perturb the clock in different
organisms, although results vary widely from one system to another. The strongest
generalization is that inhibitors of cytoplasmic (nonorganelle) protein synthesis alter
circadian timing. Drugs such as cycloheximide and anisomycin induce a change in either
the period length (continuous application) or the phase (pulse) of the clock in essentially
all organisms tested, including Euglena, Neurospora, Gonyaulax, the marine mollusc
Aplysia, and Acetabularia. Inhibitors of organelle protein synthesis, such as chloram-
phenicol and streptomycin, do not produce these effects. This result may seem to suggest
that periodic synthesis of specific proteins is part of the basic oscillatory cycle. However,
recent data with clock mutants of Neurospora (Section 7.5) favor the notion that alteration
of the clock by inhibition of protein synthesis occurs because of the existence of proteins
with high turnover rates that must be replenished at least once per cycle before they are
needed. None of these experiments, however, suggest the identity or even the general
nature of the kinds of proteins involved.
A second class of inhibitors that induces phase shifts in a number of organisms are
those that affect respiratory metabolism. Drugs such as azide, cyanide, dinitrophenol, and
DCCD perturb the clock of certain organisms (e.g., Neurospora) but not others (e.g.,
Gonyaulax). Again, in Neurospora one might have interpreted this result as indicating a
direct involvement of some mitochondrial component or the process of oxidative phos-
phorylation in the generation of circadian oscillations. However, the level and duration of
a cyanide pulse needed to produce a significant phase shift is about lO-fold higher than
that needed to completely inhibit oxygen consumption or to significantly deplete the ATP
pools. This indicates that components of oxidative phosphorylation themselves are proba-
bly not components of the clock but rather that some critical step in the clock is dependent
on this process. Unfortunately, since Neurospora is an obligate aerobe, nearly every
metabolic process in the cell shows such a dependency, and few clues as to the nature of
the clock components are obtained from this result.
204 Chapter 7

Finally, drugs affecting membrane structure and/or function, including various

ionophores that alter ion transport, have been shown to perturb circadian timing. Anes-
thetics that disrupt membrane structure induce phase shifts in Gonyaulax, as does the K +
ionophore valinomycin, while in Neurospora the Ca2 + ionophore A23187 also produces
phase shifts. The Neurospora result is especially interesting since the direction of the
phase shift (advance or delay) at certain times in the cycle can be correlated with whether
Ca2 + is entering the cell or leaving it. This is the only case of this type presently known
and is the kind of result expected if the oscillations in the concentration of Ca2 + were part
of the clock mechanism. Additional data suggesting a role for Ca2 + come from phase
shifts produced by a!1tagonists of the calcium activator protein calmodulin, such as tri-
fluoropyrizine, in both Euglena and Neurospora. (27 ,28)
Finally, there have been a number of studies in which inhibitors of cyclic AMP
metabolism alter circadian timing. Once again, however, there is no evidence indicating
that this molecule is directly involved in circadian timing.
In many of the studies reported here, investigators have proposed biochemical mod-
els for the clock based on the specific kinds of drug inhibition under study. For example,
effects of protein synthesis inhibitors have led to models involving periodic synthesis and
degradation of (unknown) clock proteins,<29.30) results with respiratory inhibitors have
suggested models involving periodic coupling and uncoupling of oxidative phosphoryla-
tionpl) and phase shifting by ionophores has been the basis of models involving periodic
accumulation and depletion of ions as a result of rhythmic changes in membrane structure
or function. (8,32) Further studies are necessary to distinguish among these hypotheses.

7.5. GENETIC APPROACHES TO CIRCADIAN RHYTHMS(33)

Because of some of the inherent difficulties associated with the biochemical ap-
proaches described above, studies have been initiated in several organisms to use genetic
analysis for studying the clock mechanisms. Genetic analysis and the use of mutants have
been important tools for dissecting complex biological phenomena, including biochemical
pathways, gene regulation, bacteriophage assembly, cell cycles, and behavioral re-
sponses. Genetic analysis also opens the door to the powerful techniques of recombinant
DNA research.
Genetic approaches to clocks have involved the isolation of clock mutants in several
organisms, including Drosophila, Neurospora, and Chlamydomonas. Most of these mu-
tants have alterations in the free-running period length of their circadian clock. In other
studies, mutants with specific biochemical lesions have been isolated and their clock
assayed to determine the effect, if any, of the lesion on clock function. Examples of this
type of approach have been presented above in Section 7.3.2 in our discussion of the
nature of the clock photoreceptor in Neurospora.

7.5.1. Genetic Analysis of Clock Mutants

In Drosophila melanogaster, approximately 15 mutants have been isolated that alter
the free-running period length of its circadian clock. (34,35) In these studies both the
rhythm of eclosion (adult emergence) and the rhythm of adult locomotory behavior have
Circadian Rhythms 205

been studied, and in all cases both rhythms are altered in similar, although not identical,
ways. The most interesting of these includes a set of five mutants that map to a single
genetic locus on the X chromosome called per. Mutants at the per locus exhibit one of
three phenotypes: short period, called per (19-h period), long period, called perl (29-h
period), or arrhythmic, called pe,o (no rhythm).
The cells or tissues responsible for the behavior controlled by the per gene have been
identified by a technique known as fate mapping. This procedure involves the construc-
tion of a fly that is a genetic mosaic, in which some of its cells are mutant while others are
wild type. One can identify the cells carrying the mutant per gene with the use of visible
markers, such as those affecting body color, or histological markers such as those staining
for a particular enzyme, that are linked to the per gene. Those cells in the fly that are
mutant for the visible marker will also be mutant for the clock gene, while those that are
wild type for the visible marker will also be wild type for the clock gene. If a given fly
exhibits a mutant clock phenotype, one can say that among those cells carrying the mutant
allele are the ones that control the clock function. By examining a sufficient number of
flies, it has been possible to show that clock periodicity controlled by the per gene is
determined by brain tissue in the head of the fly. In fact, when one side of the head is
mutant and the other side wild type, the fly behaves as if it had two clocks running
simultaneously. Similar results indicating a duplication of clock organization on two sides
of the brain have been obtained from ablation experiments in other invertebrates, such as
the cockroach and cricket (Section 7.6). (36)
Additional evidence that the brain exerts control over clock periodicity in Drosophila
also has come from experiments using mutants. Transplantation of a brain from a short
period pers mutant into the abdomen of an arrhythmic pe,o host resulted in the expression
of a short-period rhythm in the host. (37) This experiment also provided evidence that clock
information is communicated from the brain to the rest of the animal through the secretion
of a diffusible hormone.
A surprising result has been the finding that mutations at the per locus also altered the
periodicity of a high-frequency rhythm (period length about 1 min) in the interval between
pulses of the Drosophila courtship song. Furthermore, the per mutations altered this high
frequency rhythm in the same manner (i.e, shorter, longer, or arrhythmic) as they did the
circadian rhythm. Apparently, the per gene product plays a significant role in· both
oscillations. However, fate mapping of the effect of the per mutations on the courtship
song rhythm has shown that the focus of the mutation is in cells of thoracic ganglia rather
than the brain. Thus, different neural cell types are apparently differentiated in ways to
produce different types of oscillations, but at least one gene product is common to
both. (38)
In Neurospora, approximately 15 clock mutants have been isolated that alter the free-
running period of the Neurospora clock. As in Drosophila, half of the Neurospora clock
mutants map to a single genetic locus calledfrq. Most mutants at thefrq locus have either
short period or long period, but one mutant has lost its temperature compensation (i.e., it
has a temperature-dependent circadian clock). Again, several different rhythms in Neu-
rospora (DNA synthesis, RNA synthesis, CO 2 fixation, and conidiation) are all affected
in the same manner by each of the mutations. (33)
An important genetic result from the Neurospora studies has come from the analysis
of heterokaryons, strains that contain a mixture of different nuclei, some of which carry a
206 Chapter 7

mutantfrq allele and some of which carry the wild-typefrq+ allele. Heterokaryons with
different ratios of the two types of nuclei showed that there was a "gene dosage" effect of
the frq gene, i.e., the change in period length was proportional to the fraction of mutant
nuclei in the heterokaryon. This result suggests that the rate at which the Neurospora
clock runs is proportional to the amount of frq gene product present in the cells and argues
that the frq gene product plays an important role in the organization of the Neurospora
clock. (39)

7.5.2. Molecular Analysis of Clock Mutants(40-44)

An important feature of genetic analysis is that it points the direction for the use of
recombinant DNA techniques to clone clock genes. The cloning of such genes offers an
important starting point for the eventual identification of the biochemical nature and
function of the gene products of such clock genes. This approach should inevitably lead to
the understanding of the role of specific biochemical events in the function of the circa-
dian clock.
The most significant progress along these lines has been made with the per locus of
Drosophila. This gene has been cloned (40,41) and its effect as a clock gene has been
conflrmed by transforming cloned wild-type DNA into an arrhythmic perO recipient. (42).
Such transformed flies show restored rhythmicity, often with the normal 24-h period
length. DNA sequencing of the per gene, as well as isolation of the per protein with
antibodies, suggests that it is a proteoglycan. (43,44)

7.6. CIRCADIAN ORGANIZATION IN MULTICELLULAR ORGANISMS

7.6.1. Localization of Circadian Clocks in Invertebrates(45-52)

Important questions in multicellular organisms are whether each individual cell has a
circadian clock, and whether there exists a master clock or clocks that control and
integrate circadian rhythmicity for the entire organism. The flrst of (hese questions cannot
be answered at the present time, although efforts to examine circadian rhythmicity in cell
culture are proceeding along several lines. (45) For example, in the marine molluscs
Aplysia and Bulla, circadian rhythms in the rate of spontaneous flring of the optic nerve
have been studied in considerable detail. (46-48) When the eye is dissected along with the
nerve and maintained in organ culture, the rhythm of spontaneous firing persists and can
be entrained to 24-h light-dark cycles in vitro (Fig. 7-5). This shows that the eye and its
attached optic nerve have all of the organization and machinery necessary to generate a
circadian oscillator. By removing successively larger parts of the distal portion of the eye
from the culture preparation, it has been shown that as few as six cells are necessary to
maintain circadian output, and it seems likely that circadian clock organization exists
within the individual cells of the molluscan eye.
Similar results have been obtained with cultured cells from avian pineal glands, in
which rhythms in cyclic nucleotide metabolism have been observed and entrained by
external light-dark cycles. (50)
On the other hand, it also seems likely that one or more master oscillators exist that
control the circadian behavior of the organism and serve to couple individual rhythms in
Circadian Rhythms 207

Fig. 7-5. Diagram of the Aplysia eye and the

arrangement for recording and stimulating indi-
vidual cells or the optic nerve (ON). A tubing
electrode on the optic nerve can be used either
for recording or for stimulation. Intracellular electrodes can be used to record (V) or stimulate (I) a single cell.
The parts of the eye are the lens (L) and a pigmented retina (R). (Redrawn from Ref. 49.)

the organism to each other as well as to couple the rhythmicity of the organism to the
external environment. As mentioned in Section 7.5.1, the technique of fate mapping in
Drosophila using the clock mutants at the per gene has shown that neural tissue on both
sides of the brain of the fly are the site at which the per gene product exerts its control over
circadian periodicity, and that transplanting the brain from a rhythmic fly into an ar-
rhythmic one restores rhythmicity in the arrhythmic mutant. These experiments indicate
that the brain of the fly possesses a master oscillator that communicates its rhythmic
information to the rest of the animal by the secretion of a neurohormone or other diffusible
product.
Surgical ablation experiments have been carried out in great detail with the brain of
the cockroach and have shown clearly that the optic lobes of the brain of this animal are
essential for the expression of the circadian rhythm of locomotor activity, and are there-
fore the site of a master oscillator. Once again, the lobes on both sides ofthe brain seem to
possess oscillator characteristics, but in contrast to the behavior of the Drosophila brain,
the oscillators on the two sides seem much more tightly coupled than in the fly. Transplan-
tation of the optic lobes of a rhythmic animal to the brain of an arrhythmic animal that had
its optic lobes ablated restores rhythmicity. (51.52)

7.6.2. localization of Circadian Clocks in Vertebrates(53-57)

Clock localization experiments in vertebrates have depended mostly on a variety of
surgical procedures, since few clock mutants exist among any of these organisms. Early
studies involving removal of a wide variety of endocrine and exocrine glands, including
the adrenal, pituitary, pancreas, and thyroid, failed to eliminate rhythmicity in a number
of different mammals. In recent years, however, these studies have focused on two
structures, the pineal gland and the suprachiasmatic nucleus region of the hypothalamus,
since their removal has abolished circadian rhythmicity in a number of different organ-
Isms.
In house sparrows, the circadian rhythms of both perching activity and body tem-
perature are abolished by the removal of the pineal gland. The pineal exhibits a circadian
rhythm of melatonin secretion, and the continuous exposure to moderate to high levels of
exogenous melatonin causes a significant disruption of the perch hopping rhythm. A
number of additional experiments support the general notion that the pineal possesses an
endogenous oscillator that controls many of the circadian rhythms of the bird by periodic
melatonin secretion: (1) Cutting the sympathetic innervation to the pineal from the superi-
or cervical ganglion fails to abolish rhythmicity. This indicates that pineal rhythmicity is
not dependent on periodic neural input from other parts of the central nervous system. (2)
208 Chapter 7

If a bird is first made arrhythmic by pinealectomy, rhythmicity can be restored by

transplantation of a pineal from a rhythmic bird to the anterior chamber of the eye in the
pinealectomized bird. Furthermore, the phase of the newly restored rhythm in the host is
similar, if not identical, to the phase of the donor pineal. This result not only supports the
notion of an autonomous oscillator in the pineal but also that it exerts its control of
rhythmicity by secretion of a diffusible substance, since rhythmicity returns to the host
animal long before any neural connections could have been regenerated (see Chapter 8).
Further analysis of the pineal oscillator has been pursued with in vitro techniques, by
which the pineal has been maintained in organ culture for several weeks. Under these
conditions the pineal continues to exhibit a circadian rhythm of melatonin output under
constant light and temperature conditions, and can be entrained to external light-dark
cycles. The rhythm in melatonin secretion has been traced to a rhythm in melantonin
synthesis, which itself is controlled by a rhythm in the activity of the enzyme N-acetyl-
transferase, the rate-limiting step in melatonin biosynthesis. The rhythm in N-acetyltrans-
ferase activity has also been detected in dispersed pineal cell cultures. It seems likely that
this culture system will provide an important tool for future studies on the mechanisms of
pineal circadian rhythmicity. (49)
In mammals the role of the pineal appears to be somewhat different than that in
birds. (53) Although the gland exhibits a circadian rhythm in N-acetyltransferase activity,
the rhythm does not persist when innervation to the pineal is destroyed. On the other hand,
ablation of the suprachiasmatic nucleus (SCN) of the hypothalamus eliminates free-
running circadian rhythmicity in the rat, including the rhythm of melatonin secretion from
the pineal (Fig. 7_6).(54.55) Furthermore, the SCN itself shows a circadian rhythm of
electrical activity which persists when the region is physically isolated from the rest of the
brain by surgical techniques. (57)
An important result further implicating the SCN as a controlling site for circadian
rhythmicity was the finding of the retinohypothalamic nerve tract (RHT) , a direct innerva-
tion from the retina of the eye to the hypothalamus. It has been suggested that the
transmission of light information from the environment, received by the eye, to the clock
in the SCN occurs along this pathway. Perhaps most impressive of all has been the finding
that transplantation of a donor SCN to animals that were made arrhythmic by ablation of
their own SCN can restore rhythmicity to the host animals. (54)

7.7. Circadian Rhythms in Humans(4.58-62)

Anyone who has taken a commercial airline flight across several time zones has
personally experienced the effect of circadian rhythmicity on human physiology and

Fig. 7-6. Diagram of a coronal section through the brain of a golden

hamster. SCN = suprachiasmatic nucleus; SON = supraoptic nu-
cleus; OC = optic chiasm; ill = third ventricle; AC = anterior com-
missure; F = fornix; LV = lateral ventricle; CC = corpus callosum.
(Redrawn from Ref. 56.)
Circadian Rhythms 209

behavior. The general malaise or discomfort that is encountered is the result of a disrup-
tion in one's widespread patterns of circadian rhythmicity. In humans, as in other orga-
nisms, the circadian clock controls a vast array of biochemical, cellular, and physiological
parameters that oscillate in precise phase relationships to both the external environment
and to each other. Circadian rhythmicity exists in such functions as sleep/wake cycles,
body temperature, and metabolic rate. Urine excretion of water, Na +, K +, Ca2 + ,
Mg2+ , and Cl- show circadian fluctuations, while circadian rhythms are found in the
levels of a many hormones, including cortisol, growth hormone, aldosterone, prolactin,
testosterone, thyrotropin, luteinizing hormone (LH), and follicle-stimulating hormone
(FSH).
Each of these parameters oscillates with a well-defined peak and trough, and the
relationship of the oscillation among the different rhythms is critical to the normal func-
tioning of the individual. For example, among the urinary excretion rhythms, the water,
K + , Ca2 + , and Mg2 + rhythms have maximum around midday, while the Na + and CI-
rhythms have a peak at dusk. When individuals travel to a new time zone, their clocks
must reset to the new environmental time, and this resetting can take 2 days to 2 weeks
depending on the amount and direction (advance or delay) of the phase shift. For reasons
that are completely unknown, advance phase shifts take longer to complete than delays.
During the period of adjustment, the rhythms do not reset in a totally coordinated and
synchronized manner. Indeed, the kinetics of resetting are quite different f<?r different
rhythms. so that certain parameters (e.g., sleep/wake cycles) may reset fairly quickly,
while others, such as K + excretion, may reset very slowly. As a consequence, during this
period the different rhythms do not maintain their normal phase relationship, and a
phenomenon called internal desynchronization results. It is believed that this internal
desynchronization is what is responsible for the feelings of ill health and poor perfor-
mance exhibited by nearly everyone who must reset their clock. This phenomenon has
become known as "jet lag. " Although there have been a number of suggestions regarding
how to avoid or minimize jet lag, at the moment the only way individuals can have
significant impact on this condition when arriving at their destination is to "preadapt" by
gradually resetting their clock several days prior to departure. (58)
Internal desynchronization has implications for other aspects of human physiology.
Shift workers are a particular target of the ill effects of resetting human circadian clocks,
since each change from one 8-h shift to a different one, if accompanied by a change in
sleep and eating habits, is the equivalent of a flight across eight time zones and leads to a
general physiological disruption for as much as 10 days to 2 weeks. It is clear that if one
changes shifts frequently, one spends a considerable portion of one's life under conditions
of internal desynchronization. There are also suggestions, although no real proof as yet,
that animals whose cycles are continually or frequently shifted exhibit a greater tendency
to develop a wide range of systemic diseases such as cancer and heart disease, and show
faster rates of aging. Internal desynchronization is also a problem for individuals who
have unusually long periods of work, followed by long periods of rest. These include
flight crews, especially those on long international flights, who may be required to
perform their functions for 20 or more hours at a time. The fact that they are then rested
for a similar long period does not compensate for the disruption in their circadian phys-
iology, and the consequent implications for their health and performance abilities.
At the behavioral level there is considerable evidence that both mental and physical
210 Chapter 7

80

20

Fig. 7-7. Circadian rhythm in the sensitivity of mice to endotoxin from

12:30 12:30 the bacterium E. coli. The same dose of endotoxin was administered at
Time (hours) each time of day. (Redrawn from Ref. 60.)

perfonnance shows significant circadian fluctuations, and that an individual's ability to

carry out a variety of simple and complex tasks varies dramatically during the day. A
number of serious disasters that have been traced to human error, including the nuclear
crisis at Three Mile Island, have occurred in the early morning hours when individuals are
at their lowest levels of perfonnance ability. (58,59)
Finally, there is much documentation of circadian rhythms in the sensitivity and
responsiveness to a variety of drugs or toxic agents, such as anesthetics, sedatives,
antibiotics, and pain killers (Fig. 7-7). In some cases, as with anesthetics, both the desired
positive effects of the drug and unwanted side-effects show rhythms, and often these two
rhythms are out of phase with each other. If these rhythms are not taken into account in
tenns of the timing of the administration of the drug, one might be given an anesthetic at a
time when a high dose is needed to sedate the individual but at the same time that
maximum side-effects are seen. On the other hand, if the time of administration is altered
to take into account these circadian rhythmicities, a much lower dose of drug can be
applied simultaneously with the phase of least side-effects. (61)
Rhythms in drug sensitivity have been important in a new area of medicine that has
developed in recent years called chronopharmacology. Most notable have been studies
involving cancer chemotherapy, in which anticancer drugs such as cytosine arabinoside
are administered at different levels at different times of the day. Since most of these drugs
operate by inhibiting DNA synthesis in growing malignant tissue, attempts are made to
take advantage of phase differences in the mitotic cycle between malignant and nonnal
tissues. Thus, administration of an anticancer drug at relatively higher levels during the S
phase of the malignant tissue and lower levels during the S phase of nonnal tissues allows
a maximum effectiveness of inhibition of the cancer cells with a minimum in~ibitory
effect on the nonnal cells. Significant increases in survival rate using this type of approach
have been reported for a number of different types of cancer. (62) In addition to rhythms in
DNA synthesis, the physiological or biochemical basis for other rhythms in drug sen-
sitivity has been traced to rhythms in liver detoxifying enzymes, in urinary excretion of
the drugs, in uptake and utilization of the drug, and in blood volume that affects plasma
concentration. (62)
In general the significance and impact of circadian rhythmicity in human physiology
Circadian Rhythms 211

and pathology has not received widespread attention in the medical profession. Ob-
viously, such factors significantly complicate both diagnosis and treatment of a large
number of disorders as well as influencing normal patterns of behavior. However, the
existing data are clearly sufficient to warrant more serious concern in all of these areas,
and it can only be hoped that additional studies with more convincing and unequivocal
data will gain the broader attention and concern they deserve.

7.S. CONCLUSIONS

Nearly all eucaryotic organisms exhibit daily rhythms in a wide variety of bio-
chemical, physiological, and behavioral characteristics. These rhythms are controlled by
an endogenous timing mechanism called a circadian clock that serves to couple the
biological rhythms with the environmental day/night cycle. The circadian clock uses the
light-dark cycle as the primary source of information concerning environmental time, and
the clock photoreceptor molecules and structures have been identified in several organ-
isms.
In order to keep proper time, the circadian clock has evolved a mechanism of
temperature compensation whereby the rate of the clock remains the same over a wide
range of physiological temperatures. The mechanism of temperature compensation and
the mechanism of the circadian oscillator itself are not known. Several approaches to this
question are being used in animals, plants, and microorganisms. Biochemical rhythms
have been identified and their basis analyzed; drugs that perturb circadian timing have
been used to focus attention on specific molecular events in the cell; and clock mutants
have been isolated that alter circadian timing, and at least one clock gene has been cloned
and the structure of the gene and the protein that it encodes have been analyzed.
In both invertebrates and vertebrates, specific parts of the central nervous system
have been identified as the site of a master control oscillator that regulates circadian
rhythmicity in the animal. The optic lobes of invertebrate brains, the eye of certain
molluscs, the pineal gland of birds, and the suprachiasmatic nucleus region of the hypo-
thalamus in birds and mammals have all been implicated in this regard.
Human circadian rhythms have been identified in many physiological and bio-
chemical functions, and disruption of normal human circadian patterns due to jet travel or
change in work shift produce ill effects of various sorts. Circadian rhythm of sensitivity to
drugs is widespread and is being used to improve the efficacy of cancer chemotherapy and
treatment of other disorders.

7.9. REFERENCES

1.E. Bunning, The Physiological Clock, 3rd ed., Springer-Verlag, New York (1973).
2.A. Chovnick (ed.), Biological clocks, Cold Spring Harbor Symp. Quant. Bioi. 25 (1960).
3.B. M. Sweeney, Rhythmic Phenomena in Plants, Academic Press, London (1969).
4.M. C. Moore-Ede, F. M. Sulzman, and C. A. Fuller, The Clocks That Time Us, Harvard University Press
(1982).
5. D. S. Saunders, An Introduction to Biological Rhythms, BIackie, London (1977).
6. C. S. Pittendrigh, Biological Clocks, in: Science in the Sixties. Air Force Office of Scientific Research,
Washington, D.C., pp. 96-111 (1966).
212 Chapter 7

7. J. W. Hastings and B. M. Sweeney, On the mechanism of temperature independence in a biological clock,

Proc. Natl. Acad. Sci. USA 43: 804-811 (1957).
8. D. Njus, F. M. Sulzman, and J. W. Hastings, Membrane model for the circadian clock, Nature 248, 116-
120 (1974).
9. B. M. Sweeney and J. W. Hastings, Effects of temperature upon diurnal rhythms, in: Cold Spring Harbor
Symp. QlUJnt. Bioi. 25, 87-104 (1960).
10. T. Pavlidis, W. F. Zimmerman, and J. Osborn, A mathematical model fot the temperature effects on
circadian rhythms, J. Theor. Bioi. 18: 210-221 (1968).
II. G. F. Gardner and J. F. Feldman, Temperature compensation of circadian period length in clock mutants of
Neurospora crassa, Plant Physiol. 68, 1244-1248 (1981).
12. V. G. Bruce, Environmental entrainment of circadian rhythms, in: Cold Spring Harbor Symp. Quant. Bioi.
25, 29-48 (1960).
13. C. S. Pittendrigh, The circadian oscillation in Drosophila pseudoobscura pupae: A model for the pho-
toperiodic clock, Z. PJlanzenphysiol. 54, 275-307 (1966).
14. C. S. Pittendrigh, Circadian organization and the photoperiod phenomena, in: Biological Clocks in Re-
productive Cycles (B.K. Follett, ed.), Wright, Bristol, England (1981).
15. B. M. Sweeney, Chronobiology, in: Science of Photobiology (K.c. Smith, ed.), pp. 209-226 Plenum
Press, New York (1977).
16. J. F. Feldman, Genetics of circadian clocks, BioScience 33: 426-431 (1983).
17. J. F. Feldman and J. C. Dunlap, Neurospora crassa: A unique system for studying circadian rhythms,
Photochem. Photobiol. Revs. 7, 319-368 (1983).
18. V. G. Bruce and D. H. Minis, Circadian clock action spectrum in a photoperiodic moth, Science 163, 583-
585 (1969).
19. M. L. Sargent and W. R. Briggs, The effects of light on a circadian rhythm of conidiation in Neurospora,
Plant Physiol. 42, 1504-1510 (1967).
20. R. Brain, D. O. Woodward, and W. R. Briggs, Correlative studies of light sensitivity and cytochrome
content in Neurospora crassa, in: Carnegie Inst. Washington Yearbook 1976, pp. 295-299 (1977).
21. J. Paietta and M. L. Sargent, Photoreception in Neurospora crassa: Correlation of reduced light sensitivity
with flavin deficiency, Proc. Natl. Acad. Sci. USA 78, 5573-5577 (1981).
22. J. W. Hastings and H.-G. Schweiger (eds.), The Molecular Basis of Circadian Rhythms, Dahlem Kon-
ferenzen, Berlin (1976).
23. J. W. Hastings, Biochemical aspects of rhythms: Phase shifting by chemicals, in: Cold Spring Harbor
Symp. Quant. Bioi. 25, 131-144 (1960).
24. T. Lonergan and M. L. Sargent, Regulation of the photosynthesis rhythm in Euglena gracilis. I. Carbonic
anhydrase and glyceraldehyde-e-phosphate dehydrogenase do not regulate the photosynthesis rhythm, Plant
Physiol. 61, 150-153 (1979).
25. B. Prezelin and B. M. Sweeney, Characterization of photosynthetic rhythms in marine dinoflagellates. II.
Photosynthesis-irradiance curves in in vivo chlorophyll a fluorescence, Plant Physiol. 60, 388-392 (1977).
26. C. H. Johnson, J. F. Roeber, and J. W. Hastings, Circadian changes in enzyme concentration account for
rhythm of enzyme activity in Gonyaulax, Science 223, 1428-1430 (1984).
27. K. Goto, D. L. Laval-Martin, and L. N. Edmunds, Biochemical modeling of an autonomously oscillatory
circadian clock in Euglena, Science 228, 1284-1288 (1985).
28. H. Nakashima, Phase shifting of the circadian conidiation rhythm in Neurospora crassa by calmodulin
antagonists, J. Bioi. Rhythms 1: 163-169.
29. C. F. Ehret and E. Trucco, Molecular models for the circadian clock. I. The chronon concept, J. Theor.
Bioi. 15, 240-262 (1967).
30. H.-G. Schweiger and M. F. Schweiger, Circadian rhythms in unicellular organisms: An endeavor to explain
the molecular mechanism, Int. Rev. Cytol. 51,315-342 (1977).
31. D. P. Delmer and S. Brody, Circadian rhythms in Neurospora crassa: Oscillation in the level of an adenine
nucleotide, J. Bacteriol. 121, 548-553 (1975).
32. B. M. Sweeney, A physiological model for circadian rhythms derived from the Acetabularia rhythm
paradoxes, Int. J. Chronobiol. 2, 25-33 (1974).
33. J. F. Feldman, Genetic approaches to circadian clocks, Ann. Rev. Plant Physiol. 33, 583-608 (1982).
34. R. J. Konopka, Genetic dissection of the Drosophila circadian system, Fed. Proc. 38,2602-2605 (1979).
35. R. J. Konopka, Genetics and development of circadian rhythms in invertebrates, in: Handbook of Behav-
ioral Neurobiology, Vol. 4 (J. Aschoff, ed.), pp. 173-181, Plenum Press, New York (1981).
Circadian Rhythms 213

36. R. J. Konopka, and S. Benzer, Clock mutants of Drosophila melanogaster, Proc. Natl. Acad. Sci. USA 68,
2112-2116 (1971).
37. A. M. Handler and R. J. Konopka, Transplantation of a circadian pacemaker in Drosophila, Nature 279,
236-238 (1979).
38. C. P. Kyriacou and J. C. Hall, Circadian rhythm mutations in Drosophila melanogaster affect short term
fluctuations in the male's courtship song, Proc. Natl. Acad. Sci 77, 6729-6733 (1980).
39. G. F. Gardner and J. F. Feldman, Thefrq locus in Neurospora crassa: A key element in circadian clock
organization, Genetics 96, 877-886 (1980).
40. T. A. Bargiello and M. W. Young, Molecular genetics of a biological clock in Drosophila, Proc. Natl.
Acad Sci 81, 2142-2146 (1984).
41. P. Reddy, W. A. Zehring, D. A. Wheeler, V. Pirotta, C. Hadfield, J. C. Hall, and M. Rosbash, Molecular
analysis of the period locus in Drosophila melanogaster and identification of a transcript involved in
biological rhythms, Cell 38, 701-710 (1984).
42. W. A. Zehring, D. A. Wheeler, P. Reddy, R. J. Konopka, C. P. Kyriacou, M. Rosbash, andJ. C. Hall, P-
element transformation with period locus DNA restores rhythmicity to mutant, arrhythmic Drosophila
melanogaster, Cell 39, 369-376 (1984).
43. M. Rosbash and J. C. Hall, Biological clocks in Drosophila: Finding the molecules that make them tick,
Cell 43, 3-4 (1985).
44. F. Rob Jackson, T. A. Bargiello, S. -H. Yun, and M. W. Young, Ptoduct of per locus of Drosophila shares
homology with proteoglycans, Nature 320, 185-188 (1986).
45. J. S. Takahashi and M. Menaker, Circadian rhythmicity: Regulation in the time domain, in: Biological
Regulation and Development (R. F. Goldberger and K. R. Yamamoto, eds.), pp. 285-303, Plenum Ptess
New York (1984). Ann. Rev. Neurosci 1, 19-34 (1978).
46. J. W. Jacklet, Neurobiology of circadian rhythm generators, Trends Neurosci. 8, 69-73 (1985).
47. G. D. Block and S. F. Wallace, Localization of a circadian pacemaker in the eye of a mollusc, Bulla,
Science 217, 155-157 (1982).
48. G. D. Block and D. G. McMahon, Cellular analysis of the Bulla ocular circadian pacemaker system. III.
Localization of the circadian pacemaker, J. Compo Physiol. 155A, 387-395 (1984).
49. J. W. Jacklet, L. Schuster, and C. Rolerson, Electrical activity and structure ofretunal cells of the Aplysia
eye: I. Secondary neurones, J. Exp. BioI. 99, 369-380 (1982).
50. L. M. Robertson and J. S. Takahashi, The circadian clock in cell culture, J. Neurosci. 8, 12-21.
51. T. L. Page, Transplantation of the cockroach circadian pacemaker, Science 216, 73-75 (1982).
52. T. L. Page, Clocks and circadian rhythms in insects, in: Comprehensive Insect Biochemistry, Physiology,
and Pharmacology. VI. Sensory Physiology (G. A. Kerkut and L. I. Gilbert, eds.), pp. 577-652 Pergamon
Ptess, Oxford (1984).
53. M. Menaker and S. Binkley, Neural and endocrine control of circadian rhythms in the vertebrates, in:
Handbook of Behavioral Neurobiology, Vol. 4 (J. Aschoff, ed.), pp. 243-255, Plenum, New York (1981).
54. R. Y. Moore, Organization and function of a central nervous system circadian oscillator: the suprachiasma-
tic hypothalamic nucleus, Fed. Proc. 42, 2783-2789 (1983).
55. M. C. Moore-Ede, The circadian timing system in mammals: Two pacemakers may preside over many
secondary oscillators, Fed. Proc. 42, 2802-2808 (1983).
56. F. W. Turek, Neurobiology of circadian rhythms in mammals, BioScience 33, 439-444 (1983).
57. T. Deguchi, A circadian oscillator in cultured cells of chicken pineal gland, Nature 282,94-96 (1979).
58. J. Aschoff, Circadian rhythms in man, in: Biological Timekeeping (J. Brady, ed.), pp. 143-157, Cam-
bridge University Ptess, Cambridge (1982).
59. J. Aschoff, Features of circadian rhythms relevant for the design of shift schedules, Ergonomics 39, 739-
754 (1979).
60. F. Halberg, E. A. Johnson, B. W. Brown, and J. J. Bittner, Susceptibility rhythm to E. coli endotoxin and
bioassay, Proc. Soc. Exp. Bioi. Med. 103, 142-144 (1960).
61. M. C. Moore-Ede and F. M. Su1zman, Internal temporal order, in: Handbook of Behavioral Neurobiology,
Vol. 4 (J. Aschoff, ed.), pp. 215-241, Plenum Ptess, New York, (1981).
62. F. Halberg and E. Halberg, Chronopharmacology and further steps toward chronotherapy, in: Pharmacoki-
netic Basis for Drug Treatment (L. Z. Benet, N. Massoud, and J. G. Gambertoglio, eds.), pp. 221-248,
Raven Ptess, New York (1984).
 8
Extraretinal Photoreception

8.1. Introduction ...................................................................... 215

8.2. Extraretinal Photoreception in Invertebrates .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 217
8.3. Extraretinal Photoreception in Vertebrates .............................................. 218
8.4. Extraretinal Photoreception in the House Sparrow. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 221
8.4.1. Effects of Light on the Circadian Clock: Entrainment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 221
8.4.2. Effects of Light on the Circadian Clock: Aschoffs Rule ............................ 223
8.4.3. Effects of Light on the Circadian Clock: Stopping the Clock with Bright
Constant Light .............................................................. 224
8.4.4. Effects of Light on the Reproductive System: Photoperiodism ........................ 224
8.5. Possible Sites of Extraretinal Photoreception ............................................ 227
8.6. Do Mammals Have Extraretinal Photoreception? ......................................... 227
8.7. Conclusions ...................................................................... 228
8.8. References ....................................................................... 229

8.1. INTRODUCTION

Light plays many roles in the lives of organisms other than that which we normally
associate with the visual perception of patterns in the environment. Therefore, it is useful
to distinguish between those situations in which (1) light is important to organisms
primarily because of its energy content and (2) light functions as a signal, stimulus, or
"trigger" and, from a biological viewpoint, energetic considerations are secondary.
The clearest example of the former is surely photosynthesis (Chapter 12), in which it
is specifically the energy of light that the photosynthetic organism is "after"; indeed, the
photosynthetic machinery is exquisitely evolved to capture and utilize this energy with
maximum efficiency. On a priori grounds, one might suppose that the chlorophyll mole-
cules utilized by plants to capture the energy of sunlight are also employed to perceive
sunrise and sunset. As we shall see, this assumption, although plausible, is incorrect.
The most familiar signaling function of light is pattern vision. In this activity the
amount of light energy involved is of relatively minor importance. Most eyes have not
evolved to maximize the amount of light absorbed; in fact, in most visual situations,
structures in the eye operate to control and often reduce the amount of energy reaching the
pigment molecules. Spatial and temporal patterning, small variations in intensity, and

Michael Menaker • Department of Biology, University of Virginia, Charlottesville, Virginia

22901.

215
216 Chapter 8

differences in wavelength are of much greater importance to the visual process than are
large-scale changes in the energy contained in the stimulating light. It is not quite so
obvious that these same considerations apply to a variety of other processes in which light
plays a signaling, stimulating, or triggering role. These processes include pho-
tomorphogenesis (Chapter 10), photomovement (Chapter 11), circadian rhythms (Chapter
7), and reproductive cycles (below), and can reasonably be said to involve photorecep-
tion, as distinct from the simple absorption of light quanta by photopigments.
It is characteristic of photoreception that the biological activities initiated by it
require the processing of energy in amounts completely out of proportion to the energy in
the light stimulus itself. Thus, the number of joules required to escape from the tiger that
you perceive to be stalking you is infinitely greater than that involved in conveying his
image to your retina, and bears no necessary relationship to whether the event is illumi-
nated by the moon or by the noonday sun. It is less evident, but equally true, that a plant
bending toward the light or flowering in response to long days is processing energy in
amounts that are unrelated to the energy contained in the triggering light stimuli. On the
other hand, the amount of carbon fixed in photosynthesis and the number of cells killed by
ultraviolet radiation depend much more directly on the number of quanta in the light that
produces these effects. All these processes are mediated by the effects of light on biolog-
ical molecules; however, photoreception is mediated'by photopigments usually contained
within photoreceptors; relatively complex organelles or organs composing a larger system
that functions to amplify the biological response to light.
Having drawn the above distinction, it becomes clear that a single organism might
well be expected to possess many photopigments, some of which function within pho-
toreceptors, while others are involved primarily in energy transfer. Hence, there is no
reason to assume that green plants must perceive the sunrise with their cholorphyll
molecules.
The possession of multiple photopigments, and also of multiple photoreceptors that
mediate different responses to light, is the rule among organisms. This should not be
surprising in view of the above discussion, but our own subjective photoreceptive experi-
ence depends so heavily on our eyes that we are usually surprised to learn that most highly
evolved organisms with complex image-forming eyes have, in addition, other much less
obvious photoreceptors that are used to monitor crucially important aspects of the photic
environment. The basic observation leading to this conclusion is that even though the eyes
of an animal have been surgically removed or its optic tracts severed, the animal still
responds in a variety of ways to light in the visible portion of the spectrum. In order to
counteract our own pattern vision bias, let us consider the kinds of information about the
environment that are available to an organism that has only simple photoreceptors incapa-
ble of forming images. The discrimination of day from night enables the entrainment of
the circadian system along with the consequent adaptive regulation of the timing of many
physiological and behavioral activities. The seasons of the year are most unambiguously
signaled by the regular annual change in day length, which controls annual reproductive
cycles, some aspects of migration, diapause, and other important physiological responses
in a wide variety of organisms. (This phenomenon is called photoperiodism.) The inten-
sity of the light to which an organism is exposed will provide cues concerning the cloud
cover, whether the organism is in full light or in shadow, the time of day, and even, if
moonlight can be perceived, the phase of the lunar cycle. Simple photoreceptors are also
Extraretinal Photoreception 217

capable of detecting the spectral composition of light that impinges on them and, crudely,
the direction from which it comes. These latter capacities improve rapidly with slight
increases in complexity of the receptor structure ..
Although plants and various simple animals extract many kinds of information from
the photic environment, utilizing very simple photoreceptive structures, the discovery that
vertebrates and complex invertebrates with image-forming eyes also have extraretinal
photoreceptors was surprising, even to most photobiologists. After all, granted that it does
not require an eye to make such responses, it would seem that once eyes had evolved as a
result of selection pressure for the perception of images, they should also serve perfectly
well for these other, simpler tasks.

8.2. EXTRARETINAL PHOTORECEPTION IN INVERTEBRATES

It has been experimentally demonstrated that many invertebrate species make use of
extraretinal photoreception. A few examples will suffice to illustrate the diversity of
activities controlled in this way.
Because of its large size and the relative simplicity of performing surgery on it, the
giant silk moth has long been a favorite experimental subject of insect endocrinologists.
Considerable progress has been made in understanding the hormonal basis of develop-
mental events such as the induction and termination of diapause (a state of greatly reduced
metabolic activity that allows the insect to overwinter), which are regulated by environ-
mental light cycles. Several of the hormones involved in the control of diapause and
related events are known to be produced and/or stored in the brain and other central
nervous system structures. The photoreceptors that mediate these responses are also
located in the brain.
In an ingenious experiment, Williams and Adkisson(l) fIrst established that the head
end of the diapausing silk moth pupa contained the photoreceptors used by the animals to
perceive the long days that trigger the onset of adult development (the termination of
diapause). They then showed that simply by transplanting the brain (which does not
contain the developing eyes) from its normal site in the head to a new location in the tip of
the abdomen, photosensitivity was transferred from the anterior to the posterior half of the
animal. In a similar experiment, Truman and Riddiford(2,3) demonstrated that the brain of
the silk moth contains photoreceptors that synchronize the circadian rhythm of eclosion
(emergence of the adult moth from the pupal case) with the light-dark cycle, and also
with the clock that times the event.
There is extensive literature on extraretinal photoreception in the control of insect
photoperiodism and circadian rhythms. Indeed there are so many examples of these
phenomena in diverse groups of insects that the few well-documented cases in which
photoreception involves the exclusive use of the compound eyes and ocelli (e.g., cock-
roaches and crickets) are of particular interest. (3)
Some time ago, investigators studying the ventral nerve chord of crayfIsh with
electrophysiological techniques observed that changes in the electrical activity of fIbers in
the chord could be produced by illuminating the 6th abdominal ganglion, but not by
illuminating any of the other ganglia. Further work showed that the 6th abdominal
ganglion contains a single pair of photoreceptive neurons. The action spectrum for these
218 Chapter 8

photoreceptors indicates that the photopigment involved is a rhodopsin, which is very

similar to that which mediates photoreception by the crayfish eye. (4) Cave crayfish, which
no longer have functional eyes, retain these so-called caudal photoreceptors. Their func-
tion remains unknown. Apparently, they are not involved in the entrainment of circadian
rhythms, although there is reason to believe that they may exert some influence on
locomotor behavior. On the other hand, crayfish have other extraretinal photoreceptors,
not yet precisely localized (but probably in the supraesophageal ganglion), which are
involved in entrainment of the circadian rhythm of locomotion. Page and Larimer<5) also
studied a circadian rhythm in the electroretinogram (ERG) amplitude of the crayfish eye.
They have found that even this rhythm, which is after all a rhythm of sensitivity to light of
the retinal elements themselves, can be synchronized via extraretinal photoreceptors
probably located in the brain.
Perhaps the most dramatic example of extraretinal photoreception among the inverte-
brates has been described by Block and McMahon(6) in the cloudy bubble snail, Bulla
gouldiana. The eyes of this mollusc, like those of its better known relative Aplysia, can be
cultured in vitro for a week or more, and under these conditions display a circadian
rhythm in the frequency of spontaneously produced compound action potentials. This
rhythm can be entrained by light cycles and phase-shifted by single light pulses while the
isolated eye is maintained in culture. Peculiarities in the anatomy of this preparation make
it possible to remove unambiguously the entire complement of organized (retinal) pho-
toreceptors, leaving behind a small number of electrically coupled cells at the base of the
eye with no obvious specialized photoreceptive structures. Preparations surgically re-
duced in this manner still express circadian rhythms and, furthermore, still respond to
light cycles and pulses. Thus, even within the eye itself, circadian photoreception is
accomplished by a distinct set of extraretinal photoreceptors. It is interesting to compare
this situation with that in the mammalian eye described in Section 8.6.

8.3. EXTRARETINAL PHOTORECEPTION IN VERTEBRATES

Extraretinal photoreception by some birds and fish has been documented for over 50
years; however, it is only recently that the extent, complexity, and importance of the
phenomenon have become apparent. At least some species in all five vertebrate classes are
known to regulate important aspects of their physiology and behavior with reference to
extraretinally perceived light. Least is known about this phenomenon in fish (because
there has been very little work) and in mammals (because it appears that extraretinal
photoreception may be limited to newborn animals,C7) but see Section 8.6).
The pineals and, when present, associated structures such as the parietal eye and
frontal organ of all nonmammalian vertebrates are probably photoreceptive. Solid elec-
trophysiological evidence that this is so exists for some fish, amphibians, and reptiles.
The pineal glands of some birds and lizards have been maintained in vitro for many days
by continuously superfusing them with oxygenated culture medium.{8,9) If the culture
medium is collected at frequent intervals after it has passed over the pineal, it can be
assayed by radioimmunoassay for melatonin, a hormone synthesized by the gland.
Melatonin synthesis by isolated, cultured pineal glands follows a circadian rhythm, and
this rhythm can be entrained in the culture by a light-dark cycle. Furthermore, the
entrained rhythm bears the same phase relationship to the light cycle (melatonin is high
Extraretinal Photoreception 219

:::1
10 3 ..:~........: ......-.................
.:-" ......
102
105

104

103

102
Fig. 8-1. Circadian rhythms of melatonin pro- 101
duced by four individual isolated pineal glands of
Anolis held in constant darkness at four different 4
temperatures. Each point wa,s determined by radi 0- 103 ]
10 ...... ...........:....\.
immunoassay of a superfusate sample. Super- ••'~" ....... ",M" "':,:' ......' ........... "~ .................- ............
fusion was continuous and all medium was col- 102
lected in 90-min aliquots. On the ordinate is
plotted the log of the melatonin concentration in lol~~i--~~I--~--rl--~~I--~~I~-r-
24 72 120 168 2 I6
pg/ml. Note that periods of the rhythms are short-
er at the higher temperatures. (From Ref. 9.) TIME IN HOURS

during the dark portion of the cycle) as that in the intact animal (Figs. 8-1 and 8-2). These
pineal glands, like the brains of silk moths and the eyes ofthe cloudy bubble snail, contain
both circadian oscillators and photoreceptors. Pineal glands of nonmammalian vertebrates
contain a variety of photoreceptive structures ranging from scattered simple outer-seg-
ment-like organelles to highly ordered arrays of photoreceptors that look much like a
simple retina and, like the retina of the lateral eye, are served by a lens and are connected
with the brain by a substantial nerve tract. (I 0- J 2) Whether these organized photoreceptors
are in fact responsible for circadian photoreception in the pineal is not known. We are
certainly not entitled to assume that they are, as the example of the Bulla eye makes clear.

Fig. 8-2. Melatonin rhythm entrained in vitro

by a light cycle (diagrammed on the abscissa).
Temperature was a constant 2TC ± O.5°C.
Samples were obtained and plotted as de-
scribed in the legend to Fig. 8-1. Note that the
period of the rhythm is indistinguishable from
24 h and that its phase relationship to the light
cycle is as expected, i.e., melatonin is high 24 72 120
during the dark. (From Ref. 9.) n~INHOURS
220 Chapter 8

There have been several studies of action spectra of extraretinally mediated responses
to light. Deguchi(l3) found that brief light exposure inhibits melatonin synthesis in
cultured chicken pineal glands. He obtained an action spectrum for this effect that strongly
suggests that the photopigment involved is rhodopsinlike. Foser and Follett(l4) illumi-
nated the brains, but not the eyes, of Japanese quail, measured the effect of light on the
gonads, and derived a similar action spectrum. On the assumption that the photoreceptor
is hypothalamic, they calculated that its sensitivity is within an order of magnitude of that
of the dark-adapted human eye and that its photopigment is rhodopsinlike. The tentative
general conclusion that extraretinal photoreception is based on rhodopsinlike photopig-
ments is supported by the effects of vitamin A deprivation on the ultrastructure of lizard
parietal eye photoreceptors and by the finding of opsin immunoreactivity in the pineals of
several reptiles and birds. The preliminary nature of this conclusion should be empha-
sized; no one has yet extracted any photopigment from an extraretinal photoreceptor nor
successfully subjected one to microspectrophotometric analysis.
We know as much about extraretinal photoreception in the house sparrow (Passer
domesticus) as in any other vertebrate species. In order to convey some idea of the
complexity of this sensory capacity, a detailed description of the phenomenon in that
species follows.

TO OTHER CAGES -

o \
0-

o
-"
°i
....

0' 1\
0
I
."
-0
....
0

0
TI ME R FOR
PAN EL LIGHT

II ilfr r
I :: . . . ". ":' . -I-
RECORDER

'~ p
/'
:lftl
J
<;J

I ~.
."
~
,
1 I
; ..c
,
F~

~"J ! pj"
,!I ;
I}!
miI ~
~ h
:==~
--:a..
- .
(~[ I
~ :,
I' Li in
-i -
t

t!I·. 1III
'-- I--

/'
!
"'"
L..J
I
PERCH SWITCHES i i-
Fig. 8-3. Arrangement for the continuous recording of perching behavior in sparrows. The bird's weight
depresses the switch connected to the perch and electrically activates a single pen (of a 20-pen recorder), which
writes on a strip of paper moving at 18 in /day. Each day's record is separated from those of other birds, and
each bird's daily records are displayed to yield cumulative data of the kind shown in subsequent figures. While
such data are only semiquantitative with respect to amount of activity, they give a very accurate picture of the
temporal distribution of activity over long periods of time. (From Ref. 15, Copyright 1972 by Scientific
American, Inc. All rights reserved.)
Extraretinal Photoreception 221

DAYS

-
~ ; : : : ; ; ; ; .:;: : : : : : : : : : : : : : : : : : : :: :.:. : : : : : : : : : : ; : 151 -- 14
.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:. 37
10 HOURS

-. 15

...
-
37

Fig. 8-4. The circadian locomotor rhythm of the house sparrow in its free-running and entrained states. During
the first 14 days of the record, the bird was maintained in continuous darkness, and its rhythm free-ran with a
period of about 25 h. Beginning on day 15, it was presented with a daily light cycle (diagrammed at the top of the
figure; LD 12: 12, intensity during the light portion of the cycle = 0.1 Ix) to which its rhythm synchronized
(entrained). Although this bird was blinded by the removal of both eyes prior to the start of the experiment, its
record is indistinguishable from that of a normal sparrow. (From Ref. 16.)

8.4. EXTRARETINAL PHOTORECEPTION IN THE HOUSE SPARROW

8.4.1. Effects of Light on the Circadian Clock: Entrainment

Circadian rhythms of locomotor activity can be most easily assayed in perching
birds, such as Passer, by continuously recording the switch closures produced when the
bird hops on a perch. In this way, one can automatically record the temporal distribution
of activity for weeks or months (Fig. 8-3). In the presence of a light cycle, the birds are
normally active primarily when the light is on, beginning their activity at very nearly the
same time each day (if the light-dark cycle has a period of exactly 24 h). They are said to
be entrained (or synchronized) to the light cycle (see Chapter 7). When placed in constant
darkness, the bird's locomotor activity remains conspicuously rhythmic, but since it no
longer has access to an environmental cue, it "free-runs" with a period close to, but
different from, 24 h. In practice, most sparrows have free-running periods longer than 24
h and begin their activity progressively later each day. The difference between entrained
and free-running locomotor rhythmicity is dramatically clear in the raw perching data
(Fig. 8-4). Since the bird can only entrain to light cycles it can perceive, entrainment
provides an unambiguous and easily recorded assay of photoreception.
Sparrows continue to entrain to light cycles even after their eyes and pineal organ
have been surgically removed, and many carefully controlled experiments have estab-
lished that it is visible light rather than some other periodic feature of the regime, such as
222 Chapter 8

o HOURS

C.B.
Injection

Light
Cycle B

}- Light
Out

Blinded

Fig. 8-5 . Demonstration of the contribution of both

the eyes and brain photoreceptors to entrainment of
the circadian locomotor rhythm in the house sparrow.
This figure is the activity record of a sparrow exposed
to the light regimes (LD 12 : 12; 0.03 Ix) diagrammed
at the top. Entrainment to light cycle A continues after
carbon black (C.B.) is injected under the skin of the
head. Reestablishment of the entrained steady state
following the 6-h phase delay (light cycle B) requires
Head 36 days . After blinding, the activity rhythm free-runs
Scraped through the light cycle twice indicating a contribution
of retinal photoreception. When the carbon black is
removed, the blind bird reentrains to the light cycle
using extraretinal photoreceptors. (From Ref. 17.)
Extraretinal Photoreception 223

heat, noise, or electrical fields, to which they are responding. The threshold intensity to
which they will entrain is surprisingly low: about 50% of blinded sparrows will entrain to
a cycle consisting of 12 h of darkness alternating with 12 h of light at an intensity of 0.11x
(approximately the intensity of bright moonlight). This level of illumination can thus be
thought of as a crude threshold of extraretinally mediated entrainment.(I6) Sparrows with
intact eyes will entrain to cycles in which the light portion is considerably less intense,
demonstrating that both the eyes and the extraretinal receptors have inputs to the entrain-
ment mechanism, and that these inputs are, to a first approximation, additive (Fig. 8-5).
Experiments in which the intensity of light reaching the brain has been either increased by
plucking the head feathers or decreased by injecting opaque substances (e.g., india ink)
under the skin of the head have shown that the extraretinal photoreceptors are located in
the brain. (17)

8.4.2. Effects of Light on the Circadian Clock: Aschoff's Rule

Several different effects of light on circadian rhythms are discussed in Chapter 7. In
addition to producing entrainment when it is presented in cycles (light-dark, LD), light
affects free-running periods of circadian rhythms when it is presented continuously (light-
light, LL). As the intensity of constant light is increased, the free-running periods of
diurnal organisms exposed to it decrease whereas those of nocturnal organisms increase.
This empirical generalization and several others related to it are known collectively as
Aschoff's rule (after Dr. Jurgen Aschoff who has been responsible for much of the work
in this area). The rule is well supported by data from a variety of very different organisms,
although there are exceptions to it.
Sparrows obey Aschoff's rule and continue to do so after blinding. As in extra-
retinally mediated entrainment, the receptors responsible have been shown to reside in the
brain and to act in concert with the image-forming eyes in the intact bird. The free-running
period of a bird with both its eyes and its brain photoreceptors normally exposed shows
greater changes with varying intensities of LL than does that of a bird in which the eyes
have been removed or the brain shielded from light. (18) Figure 8-6 illustrates a brain

HOURS
o 24 48
1
0.1 Lu.

Fig. S-6. The effects of changing the in- - H.od

Plu,k,d
tensity of light reaching the brain on the
free-running period of the locomotor
rhythm of a sparrow with intact eyes held
in constant light (0.1 Ix). The period is
initially slightly longer than 24 h. Plucking - C. B. lnJ
feathers from the top of the head shortens
the period, whereas subsequent injection
of carbon black (C.B.) under the skin of
the head reverses this effect. (From Ref.
IS.)
224 Chapter 8

photoreceptor-mediated effect of light on free-running period. In sparrows, then, extra-

retinal brain photoreceptors are sufficient for both entrainment and Aschoff's rule, but in
nature they probably work together with the eyes (presumably the retinal photoreceptors,
although there has been no direct demonstration of this) to mediate the response to light.
There is, however, another general response of circadian rhythms to LL for which, at least
in sparrows, the brain photoreceptors are not sufficient.

8.4.3. Effects of Light on the Circadian Clock: Stopping the Clock with
Bright Constant Light .
In many organisms, the clock is stopped by constant light when its intensity is raised
above some threshold, which varies considerably from one species to another. Binkley(l9)
showed this to be true for sparrows. One might expect that this effect represents simply an
extension of the effects that LL of lower intensities has on the free-running period, but this
does not appear to be so in sparrows, at least in terms of the photoreceptive mechanism.
Although the intensity of constant light required to produce aperiodic locomotor activity
in sparrows (the indication that the clock is stopped) lies between 10 and 100 lx, blind
sparrows do not become aperiodic in constant light even when its effective intensity is
raised to 20,000 Ix. From these experiments, it appears that brain photoreceptors are not
sufficient to mediate this effect of light on the sparrow's clock. This conclusion is
especially strong in view of the fact that we know from entrainment experiments that the
brain photoreceptors are sensitive to light at intensities of about O.llx, so that in exposing
them to 20,000 Ix we are above threshold by about 2 x 105 . Therefore, the eyes are
apparently necessary for this particular effect of light on the circadian clock.
To further complicate the issue, it can be shown that, in intact sparrows, the brain
photoreceptors are involved in the response to bright, constant light, although they are not
by themselves sufficient to mediate it. If intact sparrows are brought to aperiodicity by
constant light, the intensity of which is just above the required threshold (-10 Ix),
rhythmicity can be made to recur without changing the light intensity by simply shielding
their brain photoreceptors. (20)
The biological meaning of the different patterns of interaction between the eyes and
the brain photoreceptors in mediating the effects of light on the circadian system is still
obscure, but the experimental results, which have been briefly described above, empha-
size the complexity at the photoreceptor level of one system in one species to the photic
environment. Although it is beyond the scope of this chapter to explore the comparative
aspects of this phenomenon, it is worth mentioning that a similar analysis of the effects of
light on the circadian clock has been performed in several species of lizard. While the
results of experiments with lizHds are generally similar to those with birds, several of the
important details, especially concerning interaction of eyes and brain photoreceptors, are
quite different not only when birds and lizards are compared, but even when one lizard
species is compared with another. (21)

8.4.4. Effects of Light on the Reproductive System: Photoperiodism

Many animals and plants and most temperate zone birds use the information inherent
in the cyclic annual change in day length to synchronize their reproductive cycles with the
appropriate season (Fig. 8-7). Such responses involve the ability both to perceive and
"measure" day length, and are collectively termed photoperiodic. Perhaps the simplest
question that can be framed concerning photoperiodic responses is, with what photorecep-
tors does the organism perceive the photoperiod? When Professor Jacques Benoit posed
this question in 1935 with specific reference to the perception of day length, as indicated
by the testicular responses of ducks, the answer that he obtained came as a great surprise
to the scientific community; so great in fact that 20 or 30 years elapsed before it was fully
accepted. By hindsight, and in the context of this chapter, it should not surprise you that
this answer was, at least partially, that ducks use extraretinal photoreceptors in their brains
to monitor the environmental light cycles that regulate their reproductive physiology.
Since Benoit's pioneering work, photoperiodic photoreception by brain photorecep-
tors has been directly demonstrated in several species of birds not closely related to each
other, and it probably occurs in most, if not all, photoperiodic members of the class Aves.
Photoperiodic photoreception has been studied extensively in the house sparrow with
particular attention to the possible role of the eyes. It is easy to demonstrate that sparrows
employ brain photoreceptors to perceive day length. If male birds are brought into the
laboratory during the winter, when their testes are small and nonfunctional, and are
exposed for several weeks to artificial light cycles that simulate the long days of spring
(e.g., 14 h or more of light per 24-h day), their testes grow dramatically. Not only do the
226 Chapter 8

gonads increase in weight by as much as SaO-fold, but the testis tubules become packed
with spermatids where before there were only spermatogonia. In contrast, the testes of
control birds held on short days for the duration of the experiment remain unchanged. If
this experiment is repeated using blind birds in both the long-day and short-day groups,
the results are identical. In fact, it has not been possible to find any differences between
the photoperiodic responses of blind and intact birds on the basis of gonad weight, rates of
gonadal growth, or testis histology. (22.23)
These findings led to the speculation that, in this particular aspect of photoreception,
the eyes might have no role at all. Some weight was lent to this suggestion by the
discovery that the threshold intensity required to elicit photoperiodic testis growth in
sparrows is of the order of 10 lx, or about 100 times the threshold intensity for extra-
retinally mediated entrainment, and thousands of times the threshold intensity for vision.
A critical experiment was designed in which sparrows with their eyes intact were exposed
to long days at a light intensity only slightly above this threshold. Half the birds had
feathers plucked from the tops of their heads, thus increasing the amount of light that
penetrated to their brains, while the other half were injected with opaque material beneath
the skin of their heads, thus decreasing light to their brains. The testes of the plucked birds
grew dramatically, whereas those of the injected birds did not grow at all, in spite of the
fact that their eyes were exposed to long days at light intensities above threshold(24,25)
(Fig. 8-8). In sparrows at least, photoperiodic photoreception is apparently performed
exclusively by brain photoreceptors, perhaps by brain photoreceptors that are different
from those that mediate the effects oflight on the circadian clock. Note that in the course
of investigating four different responses of the house sparrow to the photic environment-
three responses of the clock and one of the reproductive system-three different patterns
of interaction between the eyes and the brain have been uncovered. Two responses were
observed for which brain photoreceptors are sufficient but to which the eyes contribute

0-
E
l-
I

'":s:
W

(f)

i=
(f)
W
I-

Fig. 8-8. Effects of manipulating the intensity of light reaching

the brain on the testis weight of sparrows with intact eyes. A,
control group sacrificed at the start of the experiment; B, india
ink-injected group; C, group with head feathers plucked. See
Section 8.4.4 for additional details. (Figure modified from Ref.
24.)
Extraretinal Photoreception 227

when present (entrainment and Aschoff's rule); one for which the eyes are necessary
(arrhythmicity in bright constant light); and one in which only the brain is involved
(photoperiodic reproductive response). Furthermore, there is reason to believe that there
are as yet undiscovered responses to light in this as well as other species.

8.5. POSSIBLE SITES OF EXTRARETINAL PHOTORECEPTION

None of the extraretinal responses of sparrows to light that have so far been discussed
depend on the presence of the pineal gland; they all proceed unaltered in its absence. On
the other hand, there is direct evidence that avian pineal glands as well as those of other
nonmammalian vertebrates are photoreceptive. Direct photosensitivity of the iris muscle
has been reported in some vertebrates including mammals. One report in the literature
describes a behavioral response of young, but not of adult, pigeons that is thought to be
mediated by skin photoreceptors.(26) Skin photoreception is fairly common among am-
phibians and reptiles, and other birds may well display it when juvenile. Our list of
photoreceptors in the vertebrates is thus already quite diverse even before we include the
eyes and an undetermined number of brain photoreceptors.
In looking for photoreceptors that control circadian rhythms and reproduction, it is
possible that we have only scratched the surface of the phenomenon of extraretinal
photoreception. These responses to light are among the easiest to measure. But what of
those that are a bit more difficult to assess? Perhaps important aspects of the effects of
light on growth and development depend on extraretinal photoreception. Possibly light
perceived in this way affects enzymatic activity in the brain and elsewhere in the body as
well. The answers to these and a host of other questions await the results of future
research.

8.6. DO MAMMALS HAVE EXTRARETINAL PHOTORECEPTION?

Although light penetrates the fur, skin, and skulls of mammals and reaches the brain
in significant amounts, all attempts to demonstrate extraretinal photoreception in adult
mammals have produced negative results. (27) Perhaps because they passed through a long
nocturnal bottleneck in their evolutionary history, mammals appear to have consolidated
all their photoreceptors in their eyes, where they are most likely to detect the dim light at
dawn and dusk. If that supposition were true, then one would expect that the photorecep-
tors mediating the effects of light on the circadian and reproductive systems of mammals
might retain some differences from those supporting vision even though now located in
the same organ. This speculation receives support from the fact that retinal fibers carrying
visual information arise from different kinds of retinal ganglion cells and terminate in
different brain areas from those carrying circadian and reproductive information.
It was recently shown that some of the properties of the photoreceptors that mediate
the effects of light on the circadian system of the golden hamster are unusual and perhaps
quite different from those involved in vision. (28) Although their action spectrum is that of
a rhodopsinlike photopigment, they are much less sensitive to light than the rhodopsin-
containing rods of the hamster retina, and, unlike rods, they do not appear to light-adapt.
228 Chapter 8

1.6
Fig. 8-9. Comparison of the effects of

.5
1.2- J IS-min, monochromatic (SIS nm) light
pulses of increasing radiance, in phase-
I-
If I shifting of the circadian locomotor
I
u..
:E
I/)
w
.8 f h I rhythms of mice with normal retinas
(filled circles) and those with genet-
I/) I
«
f .4 rf ically degenerate retinas (open circles).

I 1I
The mutant animals have lost virtually
all of their rods, although they retain
I r some cones. The two data sets are re-
markably similar, indicating that the
10
8
10
9
10
10
10 " 10
i2
10
13

rods contribute very little to circadian

RADIANCE (photons cm-esec- f .r- I) photoreception. The differences at ra-
diances of 3 x 10 10 and 6 x 10 10 are
significant, and may indicate either a minor contribution of rods to the normal process or a loss of sensitivity in
the mutant animals resulting from other retinal damage. (Hudson, Pullela, and Menaker, unpublished data.)

These findings suggested that the visual rod photoreceptors, which account for 98% of
photoreceptors in the retinas of nocturnal rodents, are not mediating the effects of light on
the circadian system. This suggestion was confirmed by experiments that made use of a
strain of mice carrying a genetic defect that results in a progressive loss of retinal rods.
Mice that have lost their rods as a result of this defect are just as sensitive to the effects of
light on the phase of their locomotor rhythms as mice with normal retinas (Fig. 8-9),
although they are visually blind. Either the sparse, scattered cones or some retinal cell
type not previously recognized as photoreceptive must be the circadian photoreceptor in
these animals.
Thus, it appears that mammals, in common with all other vertebrates, may have
photoreceptors that are specialized for nonvisual photoreception. Such photoreceptors,
which are extraretinal in nonmammalian vertebrates, are apparently located in the retinas
of mammals. As we learn more about them we may find that such receptors have common
physiological properties wherever they may be localized, and share little beyond pho-
topigments with visual photoreceptors.

8.7. CONCLUSIONS

It should be emphasized that as far as extraretinal photoreception is concerned, we do

not as yet have any complete cases. Either we have identified and studied the physiology
of such receptors but do not completely understand their function (e.g., the pineals of
nonmammalian vertebrates), or we have identified extraretinally mediated functions, such
as entrainment, without precise localization of physiological information about the recep-
tors themselves. Not until we know both the function and the identity and physiological
characteristics of several such receptor systems will we be able to proceed much further
with our analysis.
It is clear from our discussion that many organisms possess and use multiple, discrete
photoreceptors to monitor the photic environment. The puzzling question remains: What
Extraretinal Photoreception 229

are the selective advantages of partitioning out photoreceptivity in this way? Although we
cannot yet answer this question, we can note that the wide distribution of specialized
photoreceptors and the functional complexity of some of the better known retinal-extra-
retinal systems, such as that of the sparrow, argues strongly for major adaptive signifi-
cance. Indeed it may be most useful to think of extraretinal photoreception as a separate
sensory modality independently evolved and distinct from vision.

S.S. REFERENCES

1. C. M. Williams and P. L. Adkisson, Physiology of insect diapause. XIV. An endocrine mechanism for the
photoperiodic control of pupal diapause in the oak silkworm Antherea pernyi, BioI. Bull. 127, 511-525
(1964).
2. J. W. Truman and L. M. Riddiford, Neuroendocrine control of ecdysis in silkmoths, Science 67, 1624-
1626 (1970).
3. J. W. Truman, Extraretinal photoreception in insects, Photobiology 23, 215-225 (1976).
4. J. L. Larimer, D. L. Trevino, and E. A. Ashby, A comparison of the spectral sensitivities of caudal
photoreceptors of epigeal and cavernicolous crayfish, Compo Biochem. Physiol. 19, 409-415 (1966).
5. T. L. Page and J. L. Larimer, Extraretinal photoreception in entrainment of crustacean circadian rhythms,
Photochem. Photobiol. 23, 245-251 (1976).
6. G. D. Block and D. G. McMahon, Cellular analysis of the Bulla ocular circadian pacemaker system III.
Localization of the circadian pacemaker, J. Compo Physiol. 155A, 387-395 (1984).
7. M. Zweig, S. H. Snyder, and J. Axelrod, Evidence for a nonretinal pathway of light to the pineal gland of
newborn rats, Proc. Natl. Acad. Sci. USA 56, 515-520 (1966).
8. J. S. Takahashi, H. Hamm, and M. Menaker, Circadian rhythms of melatonin release from individual
superfused chicken pineal glands in vitro. Proc. Natl. Acad. Sci. USA 77, 2319-2322 (1980).
9. M. Menaker and S. Wisner, Temperature-compensated circadian clock in the pineal of Anolis, Proc. Natl.
Acad. Sci. USA 80, 6119-6121 (1983).
10. R. M. Eakin, The Third Eye, University of California Press, Berkeley (1973).
11. H. G. Hartwig and H. Oksche, Neurobiological aspects of extraretinal photoreceptive systems: Structure
and function, Experientia 38, 991-996 (1982).
12. H. Meissl and S. R. George, Electrophysiological studies on neuronal transmission in the frog's photosen-
sory pineal organ, Vision Res. 24, 1727-1734 (1984).
13. T. Deguchi, Rhodopsin-like photosensitivity of isolated chicken pineal gland, Nature 290, 702-704 (1981).
14. R. G. Foster and B. K. Follett, The involvement of a rhodopsin-like photopigment in the photoperiodic
response of the Japanese quail, J. Compo Physiol. A 157, 519-528 (1985).
15. M. Menaker, Nonvisual light reception, Sci. Am. 226, 22-29 (1972).
16. M. Menaker, Extraretinal light perception in the sparrow. I. Entrainment of the biological clock, Proc.
Natl. Acad. Sci. USA 59, 414-421 (1968).
17. J. P. McMillan, H. C. Keatts, and M. Menaker, On the role of eyes and brain photoreceptors in the
sparrow: Entrainment to light cycles, J. Compo Physiol. 102, 251-256 (1975).
18. J. P. McMillan, J. A. Elliott, and M. Menaker, On the role of eyes and brain photoreceptors in the sparrow:
Aschoff's rule, J. Compo Physioi. 102, 257-262 (1975).
19. S. Binkley, Constant light: Effects on the circadian locomotion rhythm of the house sparrow, Physiol. Zool.
3, 170-181 (1977).
20. J. P. McMillan, J. A. Elliott, and M. Menaker, On the role of eyes and brain photoreceptors in the sparrow:
Arrhythmicity in constant light, J. Compo Physiol. 102, 263-268 (1975).
21. H. Underwood and M. Menaker, Extraretinal photoreception in lizards, Photochem. Photobiol. 23, 227-
243 (1976).
22. M. Menaker and H. C. Keatts, Extraretinallight perception in the sparrow. II. Photoperiodic stimulation of
testis growth, Proc. Natl. Acad. Sci. USA 60, 146-151 (1968).
23. H. Underwood and M. Menaker, Photoperiodically significant photoreception in sparrows: Is the retina
involved? Science 167, 298-301 (1970).
230 Chapter 8

24. M. Menaker, R. Roberts, 1. A. Elliott, and H. Underwood, Extraretinallight perception in the sparrow. III.
The eyes do not participate in photoperiodic photoreception, Proc. Natl. Acad. Sci. USA 67, 320-325
(1970).
25. 1. P. McMillan, H. Underwood, 1. A. Elliott, M. H. Stetson, and M. Menaker, Extraretinallight perception
in the sparrow. IV. Further evidence that the eyes do not participate in photoperiodic photoreception, J.
Compo Physiol. 97, 205-213 (1975).
26. M. B. Heaton and M. S. Harth, Non-visual light responsiveness in the pigeon: Developmental and com-
parative considerations, J. Exp. Zoo/. 188, 251-264 (1974).
27. R. 1. Nelson and T. Zucker, Absence of extraocular photoreception in diurnal and nocturnal rodents
exposed to direct sunlight, Compo Biochem. Physio. 69A, 145-148 (1981).
28. 1. S. Takahashi, P. 1. DeCoursey, L. Bauman, and M. Menaker, Spectral sensitivity of a novel photorecep-
tive system mediating entrainment of mammalian circadian rhythms, Nature 308, 186-188 (1984).
 9
Vision

9.1. Introduction ...................................................................... 232

9.1.1. Overview of the Vision Process ............................................... 232
9.1.2. The Importance of Vision .................................................... 232
9.1.3. Phylogenetic Organization of Visual Systems .................................... 233
9.1.4. Overview of the Structural Organization and Functional Mechanisms in the Vertebrate
Eye ...................................................................... 233
9.1.5. Objectives of This Chapter ................................................... 236
9.2. Structure of the Vertebrate Retina. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 236
9.2.1. Retina Structure as Observed with the Light Microscope ........................... 236
9.2.2. Photoreceptor Ultrastructure as Observed with the Electron Microscope . . . . . . . . . . . . . .. 238
9.2.3. Molecular Structure as Observed by Chemical Studies and Low-Resolution Diffraction
Measurements ........ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ... 241
9.3. Pigment Spectra and the Chromophore ................................................. 246
9.3.1. Chromophore Structure and Spectra ............................................ 246
9.3.2. Visual Pigment Spectra ...................................................... 248
9.3.3. The Question of Wavelength Regulation of Visual Pigment Spectra .................. 251
9.3.4. Chromophore Analogs ....................................................... 253
9.3.5 . Photochemistry............................................................. 255
9.4. Measurable Responses to Light Stimuli ................................................ 255
9.4.1. Spectra and Properties of "Bleaching" Intermediates. . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 255
9.4.2. The Visual Cycle ........................................................... 257
9.4.3. Psychophysics and Action Spectra ............................................. 258
9.4.4. The Electroretinogram (ERG) ................................................. 258
9.4.5. Early Receptor Potential (ERP) ..................................... : . . . . . . . . .. 259
9.4.6. Intracellular Recording of Photoreceptor Responses ............................... 260
9.4.7. Extracellular Microelectrode Recording ......................................... 261
9.4.8. Rod Outer Segment Ion Currents Measured by Osmotic Responses .................. 261
9.4.9. Direct Measurements of the Photocurrent of Single Rods ........................... 262
9.4.10. Patch-Clamp Experiments Are Powerful Tools ................................... 262
9.5. The Mechanism of Visual Excitation .................................................. 262
9.5.1. The Internal Transmitter Hypothesis .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 262
9.5.2. Requirements for a Transmitter ................................................ 263
9.5.3. Evidence for the Role of Calcium .............................................. 263
9.5.4. Evidence against the Calcium Hypothesis for Visual Excitation ...................... 264
9.5.5. Cyclic GMP as Internal Transmitter ............................................ 264
9.6. Light Damage to the Retina ......................................................... 266
9.7. References ....................................................................... 267

Edward A. Dratz Chemistry Department, Montana State University, Bozeman, Montana

59717.

231
232 Chapter 9

9.1. INTRODUCTION

9.1.1. Overview of the Vision Process

The process of vision starts with absorption of light by a chromophore, ll-cis-retinal,
located in a pocket deep in the protein rhodopsin. The ll-cis-retinal has an energetically
unfavorable conformation that acts like a cocked spring, which is released by light. The
released spring forces the rhodopsin molecule to change conformation. The rhodopsin
molecules are embedded in membranes in the photoreceptor cell. Light-excited rhodopsin
stimulates a large number of guanosine triphosphate (GTP)-binding proteins on the mem-
brane surface to bind GTP. The GTP-filled GTP-binding proteins then trigger enzymes
that degrade cyclic guanosine monophosphate (cGMP). Each excited rhodopsin molecule
ultimately leads to the consumption of a large number of cGMP molecules, which in tum
control the conductance of a large number of ion channels in the cell surface membrane.
The changing membrane conductance alters the electrical voltage across the cell mem-
brane, and this voltage changes the rate of neurotransmitter release by the synaptic end of
the photoreceptor cells. Complex control mechanisms, which are just beginning to be
understood, shape and shut off the electrical response of the cell after the initial excitation.
Signals from the photoreceptor cells are processed by other cells in the retina, and neurons
pass signals on to the brain where visual perception occurs.

9.1.2. The Importance of Vision

A fascination with how vision works is quite natural. Light provides animals with
very detailed information about their surroundings through the visual sense. Humans are
especially visually oriented. There is great practical importance in understanding the inner
workings of the eye, so that dysfunctions of this key sensory organ can be avoided or
treated more effectively. Visual photoreceptors are also of more broad-reaching impor-
tance because they appear to function by mechanisms that are very similar to the receptors
for many of the hormones and neurotransmitters. The visual receptors are packaged in
specialized organelles, and are easier to isolate and study than most other receptors.
Vertebrate eyes contain two types of photoreceptor cells: the rods and the cones. The
rods, responsible for dim-light vision, transmit signals to the brain that we perceive as
colorless objects in shades of gray. The cones function in brighter light and enable us to
perceive the colors of the spectrum. An eye is said to be "dark-adapted" when it has
become well adjusted to the dark. If only 5-14 photons are absorbed in a cluster of 500
dark-adapted rods, we perceive the light. At this low photon flux, none ofthe rods absorbs
more than one photon. The dark-adapted rod is an efficient single-photon detector! How
such a sensitive detection is accomplished is a puzzle that appears to be approaching
solution (Section 9.5). Coincidence between signals from several excited rods is used by
the brain to discriminate against false signals due to the occasional thermal excitation of a
rod in the dark. When rods and cones are exposed to a steady background light, they
become light-adapted and are much less sensitive, but after returning to the dark, the
maximum sensitivity soon returns. The mechanisms of this adaptation are still poorly
understood, however.
The wavelength range of visible light is defined by the wavelengths that can be
Vision 233

detected by the human eye. The limits of the visible part of the spectrum are taken to be
from about 380 to 700 nm. The wavelength dependence of the sensitivity of the eye
follows the absorption spectra of the visual pigments (Section 9.3.2). The maximum
sensitivity of a dark-adapted vertebrate eye (called scotopic or dim-light vision) is near
500 nm, at the maximum absorption of the retinal rod cells. Under bright-light conditions
(called photopic vision) the maximum sensitivity occurs at about 550 nm, which is near
the absorption maximum of the most prevalent cone photoreceptors. The wavelength
sensitivity of human photopic vision is shown in Fig. 1-1. The sensitivities of the visual
systems of all animals fall fairly close to this curve except for insects, which can see much
deeper into the near-ultraviolet region with high sensitivity.
The light is absorbed by photoreceptor cells that are located in the retina, a thin layer
of tissue at the back of the vertebrate eye. The retina is considered to be an outpost of the
brain, since there is a considerable amount of signal processing in the retina after light
detection. Several aspects of signal processing in the nervous system have been studied
profitably by the use of the accessible and easily stimulated retina.

9.1.3. Phylogenetic Organization of Visual Systems

All vertebrates have similar visual photoreceptor cells. Photoreceptors of animals as
diverse as frogs, rats, and cows appear to be comparable, even identical in many aspects.
Furthermore, much of the information derived from studies of animal photoreceptors
appears to be directly applicable to the understanding of human photoreceptors.
Invertebrates, on the other hand, have evolved a variety of morphologically distinct
visual systems that are very different from those of the vertebrates. (I ,2) In the most
primitive photoreceptors, light acts directly on pigmented cells without passing through a
focusing apparatus. Rudimentary refractive mechanisms appear in many invertebrates.
Cephalopods have a complete cornea-lens system similar to that of the vertebrates (Fig.
9-1). Invertebrate eyes develop from epithelial (skin) tissue, and are arranged in single
cells or collections of cells in flat, cup-shaped, and compound or faceted structures in
different species. The study of invertebrate vision is an active field. Some invertebrates
have extremely large visual cells that are easier to study in detail with microelectrode and
microinjection methods than the physically smaller visual cells of the vertebrates. Exten-
sive knowledge of the genetics of some invertebrates, such as the fruit fly, provides a
potentially powerful research tool. Mutations have been produced that alter the develop-
ment and functional mechanisms of the invertebrate visual system. The availability of
such mutants suggests numerous new approaches in research on vision. Invertebrate
photoreceptors are not within the scope of this chapter, but leading references to their
comparative structural organization and some of the studies of their functional mecha-
nisms are provided. (1-5)

9.1.4. Overview of the Structural Organization and Functional

Mechanisms in the Vertebrate Eye
The optical function of the vertebrate eye can be compared to a modem camera.
Figure 9-1 is a diagram of a cross-section of the human eye. Light enters through a
compound lens (composed of the cornea and the focusable lens), passes through a variable
234 Chapter 9

(a)

(b)

Fig. 9-1. (a) Cross-section of the human eye. The parts of the eye are discussed in the text. The space inside the
eye is filled with a clear, jellylike substance called the vitreous humor. The point at which the optic nerve leaves
the retina is devoid of photoreceptors and is responsible for the eye's "blind spot." (b) A simplified optical
diagram of the eye looking at a tree. There is a point in the center of any lens system where light rays are not
deflected from their incident paths. In the human eye this point, labeled Pin the diagram, is about 17 mm in front
of the retina. The size of the image on the retina can be determined using simple trigonometry: in this example
the image size = 17 mm x 30 m/ 100 m = 5.1 mm, and the visual angle Ol = arctan (30/100) "" 16°. (From Ref.
89.)

aperture (the iris), and is focused on the thin, light-sensitive layer (the retina). Higher
(simian) primates and many birds have a tiny specialized portion of the retina, the fovea,
which has the most closely spaced photoreceptors in the retina and gives the highest
resolution image. The lower part of Fig. 9-1 is a simplified optical representation of an
eye looking at a tree, showing an inverted image on the retina, which is analogous to the
image of an object on film in a camera.
The eye is, however, functionally much more complex than a camera. It rapidly
delivers an image in "real time" to the brain and can control its sensitivity to yield useful
signals over a 1010 range of light intensities. The "mechanical" iris aperture of the eye
can account for, at most, 102 of the intensity range of sensitivity adaptation of the eye.
When completely dark-adapted, the eye is much more sensitive than camera film. Man
has not constructed an apparatus to equal the eye in overall performance, but the closest
Vision 235

type of device would be an ultrasensitive black-and-white TV camera combined with a

"hi-fi" color TV camera that produces useful signals over an enormously wide range of
light intensities.
The electrical response of the photoreceptor is processed by a series of intercon-
nected cells in the retina and is coded into .a train of electrical pulses that are passed on to
the brain. Figure 9-2 shows a simplified drawing of the various interconnected cells that
receive, process, and transmit signals from the retina. A curious feature of the vertebrate
retina is that it is oriented "backward" with respect to the incident light. Light is normally
incident from the bottom of the page in this drawing, and passes through several layers of
other cellular material before reaching the receptors. The light receptor cells are at the top
of the drawing and are located on the "backside" of the retina. The backward orientation
of the retina seems to have little consequence for the efficient light detection by the

PIGMENT
EPITHELIUM

RECEPTOR CELLS

OUTER SYNAPTIC
LAYER
HORIZONTAL
CELLS

BIPOLAR CELLS

AMACRINE CELLS

INNER SYNAPTIC
LAYER

'~~~~;;;:~;:;;;;;;;;;:;:5i;;;;~;;:;~ OPTIC-NERVE
FIBERS
OUTPUT (NEURAL IMAGE)

Fig. 9-2. Simplified drawing of the vertebrate retina based on a micrograph of a retina prepared by the Golgi
method, which selectively stains a few cell bodies and their processes. The various cell types and the major
layers of the retina are labeled. Note that light is normally incident from the bottom of the picture, as discussed in
the text, and is absorbed by the rod and cone receptor cells at'the top. Receptor signals are communicated across
the retina by the horizontal cells, while the bipolar cells carry information to the ganglion cell layer. The
amacrine cells also communicate across the retina. The ultimate output to the brain is transmitted through the
axons of the ganglion cells. [From F. A. Werblin, Sci. Am. 228,70-79, (1973).)
236 Chapter 9

photoreceptor cells. The layers of the retina closest to the front of the eye, which are
traversed first by light, have little or no visible light absorption. The pigment epithelium
cells are situated above the photoreceptors. The pigment epithelium cells supply nutrients
to the photoreceptors. The pigment epithelium cells absorb much of the light that is not
captured by the photoreceptor cells, and therefore cut down on the degradation of the
image resolution that would be caused by light scattering between photoreceptors.
The layer below the receptors contain the bipolar and horizontal cells, each of which
is connected to several photoreceptors and interconnected to one another. This layer
together with the lowest layer comprises the signal-processing "brain" function of the
retina. The processed signals are communicated to the lower region of the retina contain-
ing the amacrine cells. The ganglion cells send pulsed signals to the brain through the
optic nerve fibers. After the initial detection of light, an extraordinarily varied and
complex chain of neurobiological events is activated.

9.1.5. Objectives of This Chapter

In this chapter we will first consider the structure of the receptor cells; second, the
current knowledge of the visual pigment rhodopsin; third, the various measurable light
responses; and finally, a current view of the mechanism of action of the visual system.
Useful contributions to the understanding of visual photoreceptors have come from a
variety of fields that range from quantum mechanics, photochemistry, and physical chem-
istry, through biochemistry and electrophysiology. Much has been learned about pho-
toreceptors in recent years, and this information has also revealed the vast territory that yet
must be explored.

9.2. STRUCTURE OF THE VERTEBRATE RETINA

9.2.1. Retina Structure as Observed with the Light Microscope

The sizes and shapes of cells and their placement in a tissue may be studied with the
light microscope. To prepare cells for microscopic observation, the structural integrity of
the tissue is first stabilized by cross-linking it with a variety of chemicals (e.g., formalde-
hyde and/or glutyraldehyde). The tissue is then embedded in a wax or plastic supporting
medium, and thin sections are cut with a sharp knife on a microtome. A number of
different dyes or stains may be used to enhance the visibility of various components of the
tissues.
Figure 9-3 shows a light micrograph of a thin section of monkey retina tissue. The
retina has a layered appearance, and different types of cells or parts of cells occur in each
layer. The pigment epithelium, which nourishes the photoreceptors, is the thin layer of
cells (one cell thick) that extends from the edge of the retina, from 100 to about 105% on
the retina depth scale in Fig. 9-3. The receptors near the top of this micrograph extend
from about 45 to 100% of the width of the retina. Each cell has a single nucleus, and the
relatively large number of nuclei (dark spots between 60 and 80% thickness) show the
photoreceptors to be the most numerous cell type in the retina.
There is a lower density of cells in the layer below the receptors in the region of 20-
45% of the width of the retina, as can be seen from the number of cell nuclei in the inner
Vision 237

Pi gment Epithelium ..,E=,.....c= h.:...l~;;.;.:,O\i..Ic..:..:U ..:.(llr..(..,:..,.~t..!.'

Outer Segments
Inner Segments
80 ~
o
Receptor Cells 70 :
o
60 ~
Outer Synoptic Loyer c= -50 ;
Inner Nucleor Loyer -.-402-..,
-30 _
o
-20 .!;
Ganglion Cell Loyer
;;
- 10 a:
- 0
100~m

Fig. 9-3. Light micrograph of a thin section of a rhesus monkey peripheral retina stained with hematoxylin and
eosin, which makes the cell nuclei appear as dark blobs. The pigment epithelium at the back of the eye is at the
top of this picture, and light is normally incident from the bottom as in Fig. 9-2. The identification of the cell
layers is discussed in the text. [Adapted from K. T. Brown, K. Watanabe, and M. Murikami, Cold Spring
Harbor Symp. Quant. Bioi. 30, 457-482 (1965).]

nuclear layer (located at 30-40% of the retina depth). The cells in the layer below the
photoreceptors include the bipolar and horizontal cells, each of which receives input from
several receptors. The amacrine and the ganglion cells are found in the lowest layer (at 0-
20% of the width of the retina), and they are relatively sparse as can be seen by the small
number of nuclei around the blood vessels at around 10%. The fractional thickness of each
of these layers varies from the central to the peripheral retina. The signals detected by the
receptors are processed in the lower layers of the retina, and the outputs of the photorecep-
tors are coded into the firing rate of a smaller number of output channels, the neurons. In
experimental animals it has been found that the signal processing is such that some output
neurons respond to specific small moving spots, while some respond to light/dark edges
of various angular orientations.
Photoreceptor cells have several morphological domains. Outer segments contain the
visual pigments and are embedded in fingerlike structures that protrude from the pigment
epithelium cell layer at the top of Fig. 9-3 (back of the eye). The outer segments are found
in the space between 90 and 100% of the width of the retina. The so-called inner segments
occupy the space between 80 and 90% of the width of the retina; the outer nuclear layer,
between 60 and 80%; a fiber layer, between 50 and 60%; and a synaptic region near 45%.
Under the light microscope, at least two morphological types of photoreceptors are seen in
most portions of the retina in nearly every vertebrate species, although they cannot be
distinguished in Fig. 9-3. Cells called rods usually have larger outer segments and thinner
inner segments, and cells called cones usually have shorter outer segments and much
thicker inner segments.
The distribution of rods and cones, and their morphological appearance, is not
constant across the retina. Figure 9-4 shows the rod/cone distribution across the human
retina. The rods predominate in the outer portions of the retina, where only a small
number of cones are present. The cones in the central fovea are very closely packed and
238 Chapter 9

180,000

160,000
..f···..

/\
..
...
Ii)

Z
140,000
...................
0
u 120,000 ........ \
'"
0
./ :. ...
I-
oa- .....

/.....
Ii)
100,000
0
0 li)
....
...'"
0
80,000
RODS ./ \
\
o
~ ...............
CD
0
Z 60,000 ....
40,000

20,000
CONES

80 0 100 0
TEMPORAL NASAL

Fig. 9-4. The distribution of rods and cones across the human retina. The central point at 0° is the fovea, which
is densely packed with cones. There are few cones in the peripheral retina, which is rich in rods. The blind spot
is devoid of photoreceptors, since this is where the optic nerve leaves the retina, and the blood vessels pass
through. [After G. Osterberg, Acta OphthaE. Suppl. 6, 1-102 (1935).]

have long, thin outer and inner segments, so that they look similar to the rods in the
periphery. The closely packed cells provide high spatial resolution, which produces high
visual acuity. Many species show a specialized central region of the retina that is less
developed than the fovea. Every retina has a blind spot, as shown in Fig. 9-4. Here there
are no receptors because this is where the optic nerve leaves the retina and blood vessels
pass through. The brain somehow "fills in" the image in the blind spot, so that we are
normally unaware of it.
There are no blood vessels in the receptor layer. The receptors must get their
nutrients from the dense bed of blood vessels that run behind the pigment epithelium
located just above the point marked 100% retinal thickness in Fig. 9-3. Most mammals
have retinal blood vessels that run through the deep layers of the retina far from the
receptors. Two blood vessels can be seen in the section of monkey retina in Fig. 9-3. They
appear as large openings in the ganglion cell layer near the level marked 10% of the
thickness of the retina. Many species have no blood vessels in the retina, and all retinal
nutrients are obtained from behind the pigment epithelium.

9.2.2. Photoreceptor Ultrastructure as Observed with the Electron

Microscope
The electron microscope provides much more highly magnified views of cellular
structure than the light microscope. Figure 9-5 shows a thin section of a portion of the
Vision 239

photoreceptor layer of a retina. The light-sensitive outer segments are made up of dense
stacks of membranes (the three large objects in the figure covered with dark lines). The
adjoining portion of each photoreceptor cell above the outer segment in Fig. 9-5 is called
the inner segment, and is extremely rich in cellular "power houses," the mitochondria.
Numerous mitochondria are seen near the top of Fig. 9-5, and one mitochondrion is
labeled M.
Figure 9-6 shows an even higher magnification view of a thin section of a rod outer
segment that clearly reveals the stack of membranes surrounded by a plasma membrane
that is marked with an arrow in the figure. The internal membranes occur in flattened
structures called disks. That each flattened disk is hollow can be seen by the presence of
open loops at the ends, where the disk membrane turns the comer, and by a thin light
space down the center of some of the disks. X-ray diffraction measurements on pho-
toreceptors in living retinas clearly show an aqueous space within the disk, (6) but this
space has been partially or totally collapsed during the dehydration of the specimen in
preparation for observation in the electron microscope. The space between the inner
surfaces of the disk membrane is about 2.5 nm under physiological salt conditions and has
been shown to respond to osmotic variations. (6) The diagram in Fig. 9-7 shows the
membrane spacing. Each disk is separated from its neighbor by about 15 nm, and is about
15 nm thick, making the disk-to-disk repeat distance approximately 30 om.

Fig. 9-5. Electron micrograph of an

oblique section through a glutyralde-
hyde-osmium-fixed guinea pig retina
showing portions of three outer segments
and their attachment to the inner seg-
ments. The inner segments contain a pair
of centrioles (C), one of which gives rise
to a modified cilium composed of a con-
necting piece (CP), which gives rise to
the outer segment during embryological
development. The outer segments are
made of dense stacks of disk membranes
(D), and the inner segments are densely
packed with mitochondria (M). [From A.
W. Clark and D. Branton, Z. ZellJorsch.
Mikrosk. Anat. 91, 586-603 (1968).]
240 Chapter 9

Fig. 9-6. (a) Electron micrograph of a

cross-section of a portion of a rod outer
segment showing the stacked arrange-
ments of the disk membranes. The out-
er enveloping plasma membrane is in-
dicated by the arrows (X24,OOO). (b)
Magnified view of the edges of the disk
membrane and the plasma membrane.
Note the loop in the cross-section
where the disk membranes fold back
on themselves to form closed bags
(X24,OOO). [After M. 1. Hogan, J. A.
Alvarado, and J. E. Weddell, Histo-
logy of the Human Eye, Saunders,
Philadelphia (1971). J

The disk membranes are formed by infolding the plasma membrane. In the rods,
most of the disks are pinched off and sealed to form closed membrane structures that are
not contiguous with the plasma membrane. The disk membranes appear to be anchored in
position at the edges by thin filaments that presumably are made of cytoskeletal proteins
(perhaps similar to actin, tubulin, etc.). The cone disks do not pinch off, but remain
contiguous with the plasma membrane. The general outlines of the synthesis and turnover
of the photoreceptor membranes has been worked out using radioactive precursors and
autoradiography. (7) The rod disks are replaced every 10 days. New disk components are
made in the inner segment, and new disks are added to the rods at their base just above the
junction with the inner segment. There is a circadian rhythm process that controls the
steady-state length of the photoreceptors. Small portions of the tips of the rod outer
segments are shed from the cell each morning, and are then phagocytized (internalized)
and digested by the pigment epithelium. There is evidence that disk components that are
still in an undamaged condition are recycled and reused for new disk synthesis. The
pigment epithelium cells also phagocytize the tips of the cone outer segments, but this
appears to occur in the evening. (8)

T
29.5 nm Fig. 9-7. Membrane separations in retinal rod outer seg-
~ ments. The disks show a 29.S-nm repeat, are separated by
about 15 nm, and have about a 2.S-nm space inside the

t
disk. The crosshatched oblongs represent the approximate
disposition of rhodopsin in the membrane. The plasma
Disk Plasma membrane is diagranuned on the right, and its rhodopsin
Membrane Membrane content is not shown.
Vision 241

The rod outer segments are larger than the cone outer segments, and in retinas of
most species the rods are much more numerous; consequently, there is much more rod
pigment than cone pigment in most retinas. Furthermore, the rod outer segments are more
easily isolated and purified than those of the cones; consequently nearly all the work on
the properties of the photoreceptor outside the intact retina has been done on rod material.
Studies of the composition, structure, and mechanisms ofthe cone color receptors remains
largely for the future.
The outer segment portion of the photoreceptor cell is highly membranous, with its
membranes arranged in regularly spaced stacks. The edges of two disk membranes are
shown in Fig. 9-7, which diagrams the spacings between membranes and the relation of
the disk membrane to the plasma membrane. This organization of regularly stacked
membranes is superficially similar to the lamellae of the chloroplasts of green plants, and
both types of membranes contain light-absorbing pigments. Each membrane is very thin
(4-5 nm), and in both the outer segment and the chloroplast lamellae, the probability of
light absorption has been increased by the evolution of stacks of membranes. The func-
tions of the two types of membranes are, of course, very different. The chloroplast
converts the energy of absorbed photons to chemical energy (Chapter 12), while the outer
segment acts as a detector and, as will be seen in Section 9.5, uses absorbed photons as
signals to control the flow of metabolic energy.
Recently, a bacterial membrane protein system was characterized(9) that has several
properties in common with the visual pigment in the outer segment membrane(lO) and
whose function is very similar to that ofthe chloroplast. The membranes are not stacked but
occur as sheets on the surface membrane of the bacterium Halobacterium halobium. This
organism and other related species grow in very high salt, and are called halophiles. The
light-absorbing molecule, called bacteriorhadopsin, appears to have structural similarities
to the visual pigment rhodopsin. (1 0 ,11) However, the primary function of bacteriorhodopsin
is to pump protons across the membrane in the light in the process of photosynthesis rather
than to act as a light sensor. Related membrane proteins in H. halobium pump chloride ions
across the membrane in the light, and others function as sensory light receptors by which the
motile organisms avoid blue light and are attracted to red light. (9) Further discussion of
bacteriorhodopsin and related proteins is beyond the scope of this chapter, and additional
information can be found in the references cited. (9-11)

9.2.3. Molecular Structure as Observed by Chemical Studies and

Low-Resolution Diffraction Measurements
The electron microscope falls short of providing structural information at the mo-
lecular level. Fortunately, detailed information is obtainable by the use of other tools. The
intact retina is isolated by gently peeling off the eye cup. A crude suspension of rod outer
segments may be prepared by gently shaking the retina in physiological salt, sucrose, or
other solution. In this sort of preparation, the outer segments often break off the retina
with a portion of the inner segments attached. These preparations contain considerable
mitochondrial contamination, since the inner segments are rich in mitochondria. The
purest preparations of outer segment material have been obtained by a more vigorous
homogenization procedure, which breaks the rod outer segments (ROS) into large frag-
242 Chapter 9

ments. The fragments are purified on sucrose density gradients, since the ROS disk
membranes appear to have a unique mass density. (12) Purified disks can be separated by
floatation on Ficoll (a sucrose polymer) at very low osmotic pressure. (13)
The ROS are bright pink and "bleach" to a pale yellow if exposed to light from most
of the visible spectrum. The pink color is maintained under infrared or dim red light. Most
of the preparations and manipulations are carried out under dim red light for structural
studies. However, if very detailed studies are being carried out on physiological mecha-
nisms, manipulations must be carried out in complete darkness using infrared image
converters.
The pigment portion of the ROS cannot be extracted into aqueous buffers but remains
tightly bound to the membranes. The pigment can, however, be solubilized in a number of
mild detergents without changing its characteristic absorption spectrum. The pigment
color is found to be due to a protein called rhodopsin. If a very powerful detergent, such as
sodium dodecyl sulfate (SDS), is used to extract the pigment, the native protein structure
largely unfolds, and the color of the rhodopsin solution bleaches in the dark. In SDS-
treated suspensions, the components of the ROS can be separated by molecular weight
using polyacrylamide gel electrophoresis. Many ROS proteins are free in the cytoplasm or
are weakly bound to the membrane surface and can be washed off in low-salt solutions
free of divalent ions. Figure 9-8 shows a densitometer scan of a separation of the proteins
of purified ROS membranes, which have been washed free of most of the components that
are not tightly membrane-bound. At least 90% of the protein of the ROS appears in a
single dense band on the gel, which has been shown to be opsin. Opsin is the protein
portion of the visual pigment that has been stripped of its visible light absorbing chro-
mophore by the SDS extraction. The identities of the smaller peaks have not been fully
established. There are several enzyme activities in the ROS, as will be discussed in
Section 9.5.5. Some of the small peaks correspond to ROS enzymes. Cattle rhodopsin
moves on the gel with an apparent molecular weight of 35,000; however, the amino acid

2.7

2.4
Fig. 9-8. Polyacrylamide gel electrophoresis pattern of purified
2.1
bovine rod outer segment membranes dissolved in sodium dodecyl
1.8 sulfate (SDS). The gels were stained with coomassie blue after elec-
trophoresis to visualize the protein components. A photograph of the
E 1.5 stained gel is shown at the top, and a densitometer scan of the gel (at
c:
o the stain absorption maximum) is plotted below. The major peak is
:g 12 opsin, the protein portion of of rhodopsin. The very sharp peaks at 2,
<l 4, and 8 cm are needle marks made for length calibration. The tiny
09 peak at 9 cm is the Pyronine Y tracking dye (P.Y.), which indicates
the distance traveled by this rapidly moving marker. The other small
06 peaks are protein components. SDS polyacrylamide gels separate by
molecular weight, the smallest components having the highest mobil-
ity. Bovine opsin moves with an apparent molecular weight of
35,000, and nearly all of the other minor components have larger
apparent molecular weights. [After D. S. Papermaster and W. J.
em Dreyer, Biochemistry 13, 2438-2444 (1974).]
Vision 243

sequence shows the molecular weight to be about 41,000 daltons including 2000 daltons
of covalently bound carbohydrate. Numerous protein components that may be washed off
the membrane are present on the left side of the gel (Fig. 9-8). The functions of some of
the minor components (a GTP-binding protein, rhodopsin kinase, and a phos-
phodiesterase) have been identified and will be discussed in 9.5.
The disk membrane contains about 43% protein, 53% lipid, and 4% carbohydrate by
dry weight. The lipid portion of the membrane can be extracted with organic solvents and
separated by thin-layer chromatography. This treatment also releases the chromophore, a
derivative of vitamin A (Section 9.3.1). Most of the lipids are phosphatidy1choline,
phosphatidylethanolamine, phosphatidylserine, and phosphatidylinositol, which occur in
the ratio 40: 45 : 15: 1. These lipids each have different, hydrophilic, phosphodiester
"head" groups esterified to a glycerol molecule that is also esterified to two long-chain
hydrophobic fatty acids. There is about 5% by weight of cholesterol present, and a trace of
sphingomyelin. Small amounts of glycolipids may be present, but they have not been
characterized.
The fatty acids in the phospholipids of the membrane are strikingly polyunsaturated;
in cattle, rat, and frog ROS, about 45% of the fatty acids are docosahexaenoic acid (22
carbons and 6 double bonds, designated a 22: 6 fatty acid). (14) Most of the other fatty
acids found are the saturated palmitic (16: 0) and stearic (18: 0) acids. A typical phos-
pholipid has a saturated fatty acid esterified to one of the hydroxyl groups of glycerol and
a 22: 6 fatty acid esterified to the other. (15)
When polyunsaturated fatty acids are removed from a rat's diet, the ROS tenaciously
retain their polyunsaturated fatty acids, (17) while most other cells and organs in the
animals lose their polyunsaturated fatty acids under these conditions. The exceptionally
high level of 22 : 6 fatty acids and the tenacity with which they are retained by the ROS
and by brain synaptic endings suggests that these fatty acids may be important for the
functioning of both of these membrane systems. (17)
The regular array of stacked membranes in the ROS diffracts X-rays and neutrons.
Analyses of X-ray or neutron diffraction patterns of the ROS in the intact retina have been
carried out by several groups to a resolution of about 2.5 nm. These studies show the
membrane spacings present in live unfixed tissue (Fig. 9-7) and indicate that much of the
lipid in the disk membrane appears to be arranged in a lipid bilayer. A bilayer consists of
two sheets of lipids arranged with their polar head groups in contact with water, the
nonpolar hydrocarbons of the fatty acids forming an oily interior. Both X-ray and neutron
diffraction analyses on rod membranes agree that about 50% of the mass of rhodopsin is
embedded in the hydrocarbon region of the membrane and that a substantial mass pro-
trudes from the membrane'p S ,19) Models of the diffraction pattern are most consistent
with the rhodopsin molecule fully spanning the disk membrane, as shown schematically in
Fig. 9-7.
The localization of rhodopsin in the disk membrane has been studied by several other
methods. Membranes have a region of structural weakness where the hydrocarbon chains
meet in the center of the two halves of membrane bilayers. Freeze-fracture electron
microscopy reveals the textures of the cleavage planes that pass through this region of
weakness. Proteins that protrude into the hydrocarbon region are observed as bumps on
the fracture surfaces. Rhodopsin remains attached to the hydrocarbon layer on the outer
half of the disk membrane when samples are frozen and fractured. These experiments
244 Chapter 9

show that rhodopsin is asymmetrically associated with the two halves of the disk
membrane.
Several laboratories have shown that treatment of the disk membrane with a number
of proteases, which presumably do not penetrate the membrane, cleaves the peptide chain
of rhodopsin. (18,20) Rhodopsin has two short oligosaccharide chains, consisting of man-
nose and N-acetylglucosarnine, which protrude into the aqueous region. (20) Chemical
labeling experiments imply that the lipids are very asymmetrically disposed across the
membrane, with most of the phosphatidylethanolarnine and much of the phos-
phatidylserine distributed toward the outside surface of the membrane.(21) The bulk of the
phosphatidylcholine is presumably on the inside surface.
The amino acid sequence of rhodopsin reveals seven nonpolar stretches that are
adequate to span the hydrocarbon portion of the membrane. This and other evidence has
led to the model for the structure of rhodopsin (Fig. 9-9), which consists of seven slightly
bent transmembrane helices with a retinal molecule oriented perpendicular to the helix
axis in a "basket" of bent helices. (10)
The proposed structure has now been found to be consistent with sequences of seven
different visual pigments from fruit flies, sheep, horses, pig, and human rod and three
cone pigments. (11) In addition, similar structures have been proposed for the l3-adrenergic
and muscarinic membrane receptors. (11) The bovine sequence, fitted into the model
shown in Fig. 9-10, suggests a strongly asymmetric charge distribution across the mem-

DI~K
MEMBRANE

(i 'II' "', i)
ROD
OUTER
SEGMENT

INNER
SEGMENT
IN1RADlSKAL
SURfACE

Fig. 9-9. Diagram of a cross-section of the rod outer segment. The drawing is not to scale as there are 600-2000
disks/rod and 2 x 1()4 to 8 x 105 rhodopsins/disk depending on species. A magnified portion of the rod outer
segment is shown on the left, which is a schematic model of the structure of rhodopsin and its association with
the lipid bilayer of the rod outer segment disk membrane. Rhodopsin is represented as an elongated bundle of
somewhat irregular helices (bent where prolines occur in the helices) embedded in the lipid bilayer. The ll-cis-
retinal-binding site, oriented nearly parallel to the membrane plane, is schematically shown. The seven known
phosphorylation sites are marked with the letter "P." [After E. A. Dratz and P. A. Hargrave, Trends Biochem.
Sci. 8, 128-131 (1983).)
 <
iii'
0'
:I

Cytoplasmic
Surface

---- ... ---- -.. ------ ~ ~... .,."

Bilayer
Bilayer
Headgroup
Hydrocarbon
Center - Center
Thickness
Thickness

intradiskal
Surface

Fig. 9-10. A model for the organization of the polypeptide chain of bovine rhodopsin in the disk membrane bilayer. The carboxyl terminus is exposed to the
rod cell cytoplasm. Amino acids with potentially positively charged side chains are shown as outlined shaded circles, and amino acids with potentially
negatively charged side chains are shown as shaded squares. Carbohydrates are indicated by the strings of smaller circles near the amino terminal end. The
crosshatched areas at either side indicate the lower polarity part of the bilayer. The central crosshatched part represents the width of the hydrocarbon region of ~
the bilayer. [After E. A. Dratz and P. A. Hargrave, Trends Biochem. Sci. 8, 128-131 (1983).] UI
246 Chapter 9

brane with a positive charge on the cytoplasmic surface from the protein(lO) and a negative
charge from the lipid. (21)
The two-dimensional arrangement of rhodopsin molecules in the plane of the disk
membrane appears to be a dynamic one. Spectroscopic measurements on the rhodopsin
chromophore have provided information on the rotation and translation of rhodopsin in the
membrane (Section 9.3.2).

9.3. PIGMENT SPECTRA AND THE CHROMOPHORE

9.3.1. Chromophore Structure and Spectra

Wald showed many years ago that the chromophore in rhodopsin is the ll-cis isomer
of the polyene vitamin A aldehyde. This molecule is commonly called ll-cis-retinal, and
its structure (I) together with the structures of some related molecules are shown in Fig.
9-11. When rhodopsin is exposed to light before the chromophore is isolated, retinal is
found to be isomerized to the all-trans form (Structure III). In Wald's laboratory, Bownds
found that retinal could be fixed to the E-amino group of a lysine in rhodopsin by sodium
borohydride reduction after light exposure, and therefore is bound by a Schiff base linkage

II
~IO ~12 ~ ~
13
~14 -9
15
I 1I #
H ~016 0 H
H H H

~O ~ ~ ~ OH

I2:

~ ~

# #

'lZ: # :2l H........ h

N H N H
I I(±)
R R
HH H

~ ~ ~ NHR ~ ~ ~ ~O

:lZII :mIl

Fig. 9-11. Structures of different vitamin A derivatives important in vision. I, ll-cis-retinai. II, ll-cis-12-s-cis-
retinal. III, all-trans-retinal. IV, all-trans-retinol, the alcohol form of this compound. V, II-cis-12-s-cis-retinal
Schiff base of II-cis-N-retinylidene-alkylamine. VI, protonated ll-cis-12-s-cis-retinal Schiff base or protonated
ll-cis-12-s-cis-retinylidene-alkylamine. VII, reduced all-trans-retinal Schiff base or all-trans-N-retinylalkyl
amine. VIII, all-trans-3-dehydroretinal or vitamin Az aldehyde.
Vision 247

to the protein before reduction. A Schiff base linkage, shown in Structure V of Fig. 9-11,
is equivalent to Structure II combined with an R-NH2 group. The reduced Schiff base
linkage of all-trans-retinal is shown in Structure VII of Fig. 9-11. Borohydride is water
soluble and cannot reach the chromophore linkage (buried deep in the protein) in the dark,
but when rhodopsin is exposed to light, the structure opens up and the borohydride can
reach the linkage and reacts rapidly.
Several different geometrical isomers of retinal can be prepared, including 7-cis,
9-cis, ll-cis, 13-cis, 9,1l-di-cis, 9,13-di-cis, 1l,13-di-cis, and 9,1l,13-tri-cis. The natu-
rally occurring II-cis isomer is energetically the least favorable mono-cis isomer because
of considerable steric hindrance between the 13-methyl and the lO-hydrogen (Fig. 9-11,
Structure I). If retinal in solution is thermally isomerized to an equilibrium mixture by
iodine catalysis or is photoisomerized, much less ll-cis isomer is formed than any other
mono-cis isomer, and ll-cis is exceeded in amount by some of the di-cis forms.
Theoretical calculations led to the proposal that the steric strain in II-cis retinal is
relieved by twisting about the 12-13 single bond. A twisted nonplanar conformation close
to the 12-s-cis conformation (Fig. 9-11, Structure II) (the s-cis refers to the cis conforma-
tion about a single bond) is predicted to be of the lowest energy, and a nonplanar
conformation close to the 12-s-trans (Fig. 9-11, Structure I) has a slightly higher energy.
Both forms were also predicted to be highly twisted about the 6-7 bond and to be closest
to a 6-s-cis conformation. The crystal structure of ll-cis-retinal was deduced soon after,
and these theoretical predictions were borne out quantitatively in the crystal state. A
nonplanar 12-s-cis conformation was found with a 40° twist out of plane, and the 6-7
bond was twisted 135° from the s-trans orientation, close to a 6-s-cis conformation.
Nonplanar conformations about these bonds also persist in solution, as shown by nuclear
magnetic resonance (NMR) measurements. Bacteriorhodopsin contains all-trans-retinal
as its chromophore. Recent evidence from solid state NMR indicates that the all-trans-
retinal in bacteriorhodopsin is in the unexpected 6-s-trans conformation, (22) and this
should be examined in rhodopsin.
One other closely related polyene chromophore is found in some visual pigments.
This chromophore is called 3-dehydroretinalaldehyde or vitamin A2 aldehyde (Fig. 9-11,
Structure VIII) and contains an additional double bond in the six-membered ring. A useful
shorthand for vitamin A2 aldehyde is A2 and, correspondingly, AI for vitamin A al-
dehyde. Visual pigments from hundreds of species have now been analyzed. They all
contain either the AI or A2 chromophore. The retinas of some species contain both Ac
and A2-based visual pigments, and the relative portion depends on age, season, and
habitat. (23)
Schiff base derivatives of AI aldehyde chromophores may be easily prepared with
aliphatic amines in organic solvents, and are found to have absorption maxima in the near-
UV region. The solid and dashed lines in Fig. 9-12 show spectra for Schiff bases of AI in
the ll-cis and all-trans conformations, respectively. Analogous compounds of the A2
chromophore absorb maximally at somewhat longer wavelengths than those that contain
the AI chromophore. The II-cis chromophores exhibit two peaks on the short-wavelength
side of the strongest band; these are called cis peaks and are found in every cis polyene.
The all-trans chromophore shows a much stronger main band, and weaker short-wave-
length bands than the cis forms. It retinal Schiff bases are protonated by the addition of
acid, the spectral maxima are shifted to longer wavelengths. The dotted line in Fig. 9-12
248 Chapter 9

..
o
~
0.8

0.6
Fig. 9-12. Absorption spectra of retinal Schiff bases made
............. with l-amino-2-propanol. Dashed curve, all-trans-N-reti-
'"g0.4 nylidene-l-amino-2-propanol. Solid curve, ll-cis-N-retinyli-
J:I
<I: dene-l-amino-2-propanol. Dotted curve, protonated Schiff
0.2 base made with equimolar HCI0 4 added to 11-cis-N-reti-
.... nylidene-l-amino-2-propanol. All spectra are in 1,2-di-
300 400 500 600 chloroethane. [Redrawn from data of J. O. Erickson and P. E.
Wavelength (nm) Blatz, Vision Res. 8, 1367-1375, (1968).]

shows a spectrum of the protonated Schiff base of ll-cis Al whose structure is shown in
Fig. 9-11, Structure VI. The chromophore in most visual pigments absorbs considerably
farther to the red than do the protonated Schiff bases in solution; the additional red or blue
shift due to the protein environment will be discussed in Section 9.3.3.

9.3.2. Visual Pigment Spectra

The absorption spectra of visual pigments (VP) may be measured in four ways. First,
VP may be extracted from isolated retinas with a mild detergent solution. Second, they
may be measured in situ in single-photoreceptor outer segments by direct spectrophotome-
try using microbeams. Third, they can be estimated in intact animals (including man) with
a technique where the amount of light that is reflected from the back of the eye is
measured before and after an exposure to bright light, which bleaches a large fraction of
the pigment. (24) Finally, absorption spectra can be measured via the electrical responses
of individual cells as a function of different colors of light. (25) All VPs show similarly
shaped absorption spectra with large extinction coefficients close to 40,000 liters mol- I
cm - 1, and all of the native VPs appear to be bleached with high quantum efficiency (near
0.67).
A striking feature of VP spectra is the great range of wavelengths of maximum
absorption that are observed in photoreceptors of different types in various species. The
absorption maxima range from about 435 nm in the blue to about 625 nm in the red.
Mixtures of different VPs are present in intact retinas of most species and in the detergent
extracts of their retinas. The composition of the mixtures and the absorption maxima of
the components can be estimated by a technique called "partial bleaching," if the max-
ima of the different pigments are not too close to one another. This method measures the
difference in absorbance between visual pigment samples that have been exposed to
different sets of light conditions. "Bleaching" is a term used for the loss of visible
absorption that occurs after exposure to light. In the typical approach, the visual pigments
in a mixture that absorb farthest to the red are first bleached by successive exposures to
bright, far-red light. Difference spectra are recorded after each bleach until no further
change is observed. Then the bleaching light is decreased in wavelength, and the process
is repeated. The peaks in the difference spectra are close to the VP absorption maxima, if
the wavelengths of the bleaching lights are varied sufficiently finely to provide reasonable
assurance that predominantly one form of VP is being bleached at each stage.
Absorption spectra of purified bovine rhodopsin and of the isolated Il-cis-retinal are
Vision 249

shown in Fig. 9-13. The large peak at 498 run is called the a peak, the small peak near 340
nm is called the 13 peak, and the strong peak at 278 nm is called the 'Y peak. The 'Y peak is
largely due to the aromatic amino acids, tyrosine and tryptophan, in the protein, although
there is probably some contribution by the chromophore at these wavelengths. The wave-
length sensitivity of rod vision matches the absorption spectrum of rhodopsin, with the
exception that humans do not see UV light (although some insects do) because it is
strongly cut off through absorption by the cornea, lens, and vitreous humor (see Fig. 9-1).
Humans see rather well in UV light if the lens is removed from the eye and the focus is
corrected with UV light-transmitting glasses.
In recent years microspectrophotometry on single isolated photoreceptors has been
particularly useful for visual pigment spectral studies. Single cones have distinct spectra
and cones with red, green, and blue absorption maxima are present in many species. (27)
These observations support the trichromatic theory of color vision: three types of cones
exist, each with a visual pigment of a different color. Some species appear to have only
one or two different colors of cone visual pigments. Some species, such as birds and
amphibians, also have highly colored oil droplets in the inner segment that the light must
pass through to reach the cone outer segments. Different cones in the same species may
have different colored oil droplets, which apparently act as color filters. Therefore, two
types of spectral sensitivity controls are found in cones: wavelength regulation of the
visual pigment absorption spectra and, in some cases, an oil drop color filter.
In simian primates the central region of the retina is called the macula. This region
includes the cone-rich fovea and a surrounding rod-rich area. The macula contains a
mixture of yellow carotenoic pigments called lutein and zeaxanthin. 26 The absorption
spectra of the macular pigment shows a series of relatively sharp peaks in the blue and
green, and the spectral sensitivity of the human eye shows corresponding dips that are due

70,000

60,000

€ 40,000 rhodopsin _____

30,000 II-cis
, ,
retinal--") .....

20,000 ,, \
\

" '-- ../

/ \
\
\
10,000

300 350 400 450 500 550 600 650

A (nm)

Fig. 9-13. The absorption spectrum of purified bovine rhodopsin solubilized in the detergent (solid line) and
ll-cis-retinal (dashed line) in Ammoxy LO detergent. [From T. G. Ebrey and B. Honig, Q. Rev. Biophys. 8,
129-184 (1975).]
250 Chapter 9

___ V
ENDVIEW

RED

~RED

COLORLESS

RED~
Fig. 9-14. Schematic diagram illustrating the dichroism of
outer segments, when viewed from the side with polarized
light; and the lack of dichroism, when viewed from the end
SIDE/ (see text for more details). The rhodopsin chromophores are
VIEW
highly restricted and must lie in the plane of the disk mem-
brane, but are randomly oriented within that plane.

to the screening of the photoreceptor by absorption by this pigment. The function of the
macular pigment is not yet understood, but may protect against energetic blue light or
quench toxic oxygen species. (26)
Microspectrophotometry, using polarized light, has shown that the rhodopsin chro-
mophore is highly oriented in the disk membraneP7) Figure 9-14 shows the geometry of
viewing the rod either "end on," as light is incident on the intact eye, or "side on," as
can be done experimentally in isolated photoreceptors. If the photoreceptor is viewed side
on with polarized light, rhodopsin absorbs light much more strongly if the electric vector
of the light is polarized in the plane of the membrane than if it is polarized perpendicular
to the membrane. Viewed visually side on in the light microscope, the rods appear nearly
colorless when the light is polarized parallel to the long axis of the rod. However, the rods
are a distinct red when the light is polarized parallel to the plane of the disk membrane
(Fig. 9-14). Viewed end on, as they normally function, rods are equally red with any
direction of polarization. The red color results because the visual pigment preferentially
absorbs the green and blue (see Fig. 9-13), and allows the red to pass through. The ratio of
the absorbances for the parallel and perpendicular orientation of the polarized light (the
dichroic ratio) is about 5. This large dichroism requires that the chromophore be strongly
oriented in the plane of the membrane.
Light incident on an intact eye always hits the photoreceptor end on and travels
parallel to the long axis of the photoreceptor. Rhodopsin is optimally oriented to absorb
this light, for the electric vector of the end-on incident light lies in the plane of the disk
membrane and has no component perpendicular to it. The rod appears uniformly red,
independent of the polarization (i.e., there is no dichroism), when viewed end on, as
shown in Fig. 9-14. Consequently, vertebrate eyes are not sensitive to the plane of
polarization of the light, as are some insect eyes.
If a brief, bright, polarized flash is incident on the rod end on, a large transient
dichroism is induced that quickly dies away. (28) The initial dichroism is due to photoselec-
tion of rhodopsins that happen to be oriented optimally for absorption of the polarized
Vision 251

flash and are efficiently bleached. However, the rhodopsin molecules are rapidly rotating
in the plane of the membrane, and therefore the dichroism quickly disappears. The half-
time for the rotational disorientation is about 20 Jl.S at 5°C. An increase in temperature of
lOoC reduces the half-time for disorientation by about a factor of 2.5 (the QIO)' The lateral
motion of rhodopsin in the plane of the membrane has also been measured spec-
trophotometrically. (29,30) The lateral diffusion constant is 4 x 10 - 9 cm 2/s at 25°C with a
QIO of 2. It is clear that the lipid matrix of the disk membrane presents a relatively fluid
medium for rhodopsin movement. The diffusion constants of rhodopsin have been used to
estimate an effective viscosity of about 2 poise in the plane of the membrane, which is 200
times that of water and is of the order of the bulk viscosity of olive oil. (28-30) Glu-
tyraldehyde fixation stops the rotational and the translational motion of rhodopsin. The
electrical response of the retina is abolished by glutyraldehyde fixation, although rhodop-
sin can still be bleached, and appears to undergo its normal conformational changes after
fixation (see the discussion of the early receptor potential in Section 9.4.5). It is thought
that movement of rhodopsin molecules in the plane of the membrane is required for
photoexcitation. (34)

9.3.3. The Question of Wavelength Regulation of Visual Pigment Spectra

The wide range of absorption maxima found in visual pigments correspond to a
chromophore excitation energy difference, which is "tuned" over nearly 0.75 eV in the
excitation energy of the chromophore. Part of this range is spanned by the use of two
slightly different chromophores, Aj and A2. However, the Aj pigments themselves show
a range of nearly 0.5 eV (430-575 nm), which is a very large range for a single chro-
mophore. Different protein environments around the chromophore-binding site in differ-
ent visual pigments must be able to cause a change of the absorption maximum of the
chromophore from blue to red. This influence of the protein on the chromophore is called
wavelength regulation.
When a VP protein is coupled with the A2 chromophore, the VP always absorbs at
longer wavelengths than if the same protein is coupled to the Aj chromophore. If these
A2-A j shifts are compared for a number of VPs, the difference is largest for the pigments
that absorb furthest to the red, and gets smaller for pigments that absorb to the blue. In
fact, a plot of the A2-A j shift vs. the absorption maxima of either the Aj or the A2 VP is
smooth and nearly linear, and the A2-A j difference converges to near zero at the short-
wavelength end. This phenomenon is called "convergence."
A large red shift relative to the isolated chromophore (Fig. 9-12) can be obtained by
protonation of the Schiff base linkage to rhodopsin (see Fig. 9-11, Structure VI). Reso-
nance Raman spectroscopy experiments have shown that the Schiff base linkage in rho-
dopsin is protonated. The nature of the counterion to the protonated Schiff base can have a
large effect on the spectral maximum. Carboxyl groups are probably the only available
anions in the protein environment, and variations in the properties of the anion are
probably not sufficient to explain the range of absorption maxima, although the separation
distance of the counterion from the protonated Schiff base may have an important effect.
Additional negative ions (Qr dipoles) in the oily region near the retinal have been proposed
to complete the wavelength tuning. (31)
The amino acid sequences of the human red, green, and blue cone pigments and the
252 Chapter 9

Cytoplasmic Surface

10

···l
8110Ve,
Hvdrocorbon
Thlckn."

. . .J.

Intradiskal Surface

Fig. 9-15. Model for the transmembrane arrangement of amino acids in vertebrate pigments in which each
amino acid is represented by a circle. 34 The structure is composed of seven transmembrane helices (I-VII) with
connecting aqueous loops and N- and C-terminal "tails." The degree of shading of the different residues
represents the extent of conservation of the amino acids in seven sequences, including two fly pigments, bovine
rhodopsin, human rhodopsin, and three human cone pigments. Black residues are identical in all seven se-
quences, open residues labeled with a letter are identical in at least four sequences, and shaded residues are very
similar in chemical properties in at least four sequences. The one-letter code is: A = alanine, C = cystine, D =
aspartic acid, E = glutamic acid, F = phenylalanine, G = glycine, H = histidine, 1 = isoleucine, K = lysine, L
= leucine, M = methionine, N = asparagine, P = proline, Q = glutamine, R = arginine, S = serine, T =
threonine, V = valine, W = tryptophan, Y = tyrosine. Rat Lys296 denotes retinal. [After P. A. Liebman, K.
R. Parker, and E. A. Dratz, Ann. Rev. Physiol. 49, 765-791 (1987).]
Vision 253

rod pigment have recently been obtained. (32,33) The absorption maxima of these pigments
range from 440 to 540 nm. The sequences are highly homologous and are perfectly
consistent with the transmembrane model originally proposed for bovine rhodopsin,(10)
which has seven highly hydrophobic helices spanning the membrane with hydrophilic
loops connecting the helices on both the inside and outside aqueous surfaces (Fig. 9-10).
This model also accommodates several other visual pigment sequences that have now
been determined. (1 1,31-34) Figure 9-15 shows the consistent features of the amino acid
sequences in comparing seven visual pigment sequences. Most of the charged groups in
the sequences are assigned to loops between helices located in the aqueous phase. Very
few charged groups are present in the regions assigned to the intramembrane helices, but a
few charged groups are thought to be present.(1l,34)
The Glu-Arg or a Asp-Arg at 134-135 are found in the region assigned to helix III in
every pigment. The opposite paired charges are stabilized in the oily membrane region.
The negative end of the dipole from the paired charges may be oriented toward the retinal,
but since the dipolar pair of charges is present in visual pigments absorbing over the full
range of colors, it cannot have a differential effect on the colors of the different pigments.
The strongest support for the role of hydrophobic charges in contributing to the color
of the visual pigments comes from analyzing the human rod, blue-cone, green-cone, and
red-cone sequences. (32,34) Figure 9-16 shows a diagram of the charge location superim-
posed on the transmembrane model of the sequence. The blue-cone visual pigment has
only the dipolar pair of charges on helix III and lysine 296. The human rod pigment adds
two negative charges on helices II and III, and possibly a positive charge on helix V (a
histidine that mayor may not be protonated under physiological conditions). A doubly
negative charge site on helix II is added to both the green and red pigments. Both
pigments lose a negative site on helix III and lose the potentially positive histidine charge
site on helix V.
The general picture that charges in the hydrophobic regions around retinal control
visual pigment color is attractive, but some other mechanism must also contribute, e.g., to
differentiate between the green and red pigments. If aromatic amino acid side chains are
placed in the protein-binding pocket near retinal, they can also contribute to significant
shifts in the absorption maximum. (35,36) Therefore, aromatic side chains might be used
for "fine-tuning" the colors of the visual pigments.

9.3.4. Chromophore Analogs

Comparisons of the properties of the VPs formed from isomers of the naturally
occurring Al and A2 retinals provide some testable criteria for theories of VP spectra. The
9-cis and 9,13-di-cis isomers of these chromophores form pigments with the protein, in
addition to the native II-cis isomer. For some time chemists have been making synthetic
vitamin A aldehyde analogs that differ in chain length and steric and conformational
properties to test the restrictions in the retinal-binding site. It has been found that the
rhodopsin-binding site is quite restrictive as to what sorts of alterations of the chro-
mophore it will accept. Longer or shorter polyene molecules do not form VPs, and the six-
membered ring must be present. Anywhere from zero to three double bonds in the ring is
accepted. None of the methyl groups is required for the formation of VPs; however, some
of the analogs that are missing methyl groups are not bound as tightly as the native
254 Chapter 9

Fig. 9-16. The charged amino acids in the

human blue, green, and red cone pigments
(after Ref. 32) and rod pigment (after Ref.
33) are shown by the EEl and e symbols in
the sequence. The sequence is arrayed ac-
cording to the transmembrane folding model
proposed for bovine rhodopsin (Fig. 9-10),
and fit by all known visual pigment se-
quences (e.g., Fig. 9-15). Most of the
charged groups are in contact with the aque-
ous media outside the membrane. The few
charges in the hydrophobic region of the
membranes are proposed to contribute to
"tuning" the color of the visual pigment.
The small numbers 1-5 denote the in-
tron! extron boundaries in the genes.
Vision 255

chromophores. The pigment formed from the 9-desmethyl (lacking the 9-methyl in Fig.
9-11, Structure I) has a significantly different absorption maximum and circular dichroism
intensity from that of the native chromophore. Most of the structural variations of the
chromophores produce less stable pigments with opsin and also show a reduced photosen-
sitivity compared with the native pigment. It is beyond the scope of this chapter to discuss
these results in detail, but much of the research has been extensively reviewed in other
works. (37-40) Retinal analogs may be very useful in future research on the interactions
between the protein and the chromophore.

9.3.5. Photochemistry
The quantum efficiency of stimulation of rhodopsin is very high, near 0.67. It was
pointed out in Section 9.3.1 that the bleaching of rhodopsin causes the isomerization of
the chromophore from the II-cis isomer (Fig. 9-11, Structure I) to the all-trans conforma-
tion (Fig. 9-11, Structure III). Conformational changes initiated by photon absorption
trigger a series of changes that are discussed in the next section. After a photon is
absorbed, experiments show that all the other reactions in vertebrate photoreceptors occur
in the dark. The photochemistry in the system then centers around the isomerization
mechanism, and whether the isomerization occurs by a singlet or a triplet pathway. The
fIrst detectable spectral change in rhodopsin appears in less than 6 ps (6 X 10- 12 s). This
observation does not explain the mechanism, but it excludes the involvement of large
simultaneous changes in the protein-binding site. While II-cis-retinal isomerizes pre-
dominantly by a triplet mechanism in solution, the protonated Schiff base isomerizes
much more rapidly through a singlet mechanism. (40) These observations indicate that
retinal may be a poor model system of rhodopsin photochemistry, and that the protonated
Schiff base has important, different properties.

9.4. MEASURABLE RESPONSES TO LIGHT STIMULI

9.4.1. Spectra and Properties of "Bleaching" Intermediates

When rhodopsin is exposed to light, it undergoes a series of color changes, from its
original reddish-pink to yellow in a series of steps. As has been mentioned, this entire
process is called "bleaching." Figure 9-17 shows the series of intermediate steps in the
bleaching process that have been identified, their absorption maxima, and the approxi-
mate half-times of interconversion in the retina or in isolated outer segments. The first
change from rhodopsin to prelumirhodopsin is light-induced, all other changes being dark
reactions. All the intermediates may photobackreact to reform rhodopsin; however, it has
been claimed that a form of metarhodopsin I does not photobackreact. (41) A single bright
flash, no matter how bright the flash, does not bleach all the rhodopsin because of the
photobackreactions. The amount of bleaching in a saturating flash depends on the relative
absorbance of rhodopsin and the intermediates integrated over the spectral distribution of
the flash, the photobackreaction quantum efficiency, and the lifetimes of the intermediates
relative to the flash length. The lifetimes of the intermediates, and presumably the pho-
tobackreaction quantum efficiency, depend on the environment of rhodopsin. For exam-
256 Chapter 9

Rhodopsin (498 nm)

T 20.C :; 6 X 10- 125 ~ hV

1' •
20 C
~ 3 x 1:::~umirhOdrSin :5:~I::~e
Lumirhodopsin (497 nm)

1'20 • C '" 10-6 S ~ T 2: -40·e

if
Metarhodopsin I (478 nm)

3
1"20·C '" 10- 5 +H+ T 2: -15·e
1"37.C '" 0.25,10-3 5

Metarhodopsin n (380 nm»)

7'20.C '" 102 S ~ T 2: o·e
Metarhodopsin ill (465nm) 7'20.c",10 2 s

'1'20• C '" 10 s
2 +
+
All-trans Retinol (387nm) + Opsin

1'20•C '" .
II-cis Retinal ~
+ Opsin Isomerase
s
103. -
{
~l oxidoreductase

AII-trons Retinol (330nm)

Fig. 9-17. Stages in the bleaching and regeneration of rhodopsin. The initial photoevent is shown as a wavy
line, and the subsequent thermal, dark reactions are shown as solid lines. The absorption maximum of each
species is shown in parentheses after the name of the species, and is the value observed for bovine rhodopsin.
The approximate half-times for interconversion of each species at 20 and 37°e are denoted by 200 e and 37°e,
respectively. The approximate temperatures above which the interconversions occur are indicated by "T."
More than one step may be involved in the interconversions of metarhodopsin I to metarhodopsin n, meta-
rhodopsin III to all-trans-retinal + opsin, and the interconversion of all-trans-retinal to ll-cis-retinal.

pIe, in many detergents the meta I ~ meta II rate is accelerated about WOO-fold. (42) In
these detergents, nearly complete bleaching can be obtained in a single I-ms flash. In
contrast, in the lipid environment of the intact membrane it is difficult to obtain more than
a 50% bleach in a single bright flash.
The kinetics of the spectral changes are rather strongly temperature-dependent. At
37°e, meta II appears with a half-time of 0.25 ms in the rat retina, whereas below ooe
meta II appears very slowly, and meta I may be thermally trapped below -15°C. Bovine
rhodopsin, which has been studied in most detail, exhibits the same general pattern of
temperature-dependent kinetics, with some quantitative differences. For example, lumi-
rhodopsin can be trapped below about -40oe, and prelumirhodopsin below -140°C.
Prelumirhodopsin can also be formed at liquid nitrogen temperature (-196°C), and can be
photobackreacted to rhodopsin at this temperature; however, some 9-cis pigment (called
isorhodopsin, "-max = 486 nm in cattle rhodopsin) is also formed in the backreaction under
these conditions. The prelumirhodopsin formed at -196°e completes the bleaching se-
quence when it is warmed to room temperature in the dark, resulting in isomerization of
the retinal to the all-trans form. The chicken cone pigment, called iodopsin, behaves
differently; prelumi-iodopsin formed at -I96°e backreacts to iodopsin upon warming,
and does not therefore get distorted enough in prelumi-iodopsin to attain isomerization.
Vision 257

A proton is taken up by rhodopsin during the meta I ~ meta II transition. This is

surprising because the absorption spectrum of meta II rhodopsin is similar to that of an
unprotonated Schiff base and that of meta I rhodopsin is similar to that of a protonated
Schiff base. In fact, Raman and Fourier transform infrared vibrational spectroscopy shows
that the chromophore loses a proton upon meta II formation. The protein must take up this
proton, as well as one additional proton during the meta I ~ meta II transition.
A dependence of the ratio of meta I1meta II on pH can be observed at 5°C, where
bovine meta II has a lifetime of several hours. Meta II predominates at acid pH, and the
ratio shows a simple Henderson-Hasselbach equilibrium relationship with a pK near 6.5.
Meta III and later products form at higher temperatures. The color of the bleaching
product, formed after meta III, is also pH-dependent in a manner that appears to reflect
variable protonation of a retinal Schiff base linkage.
The meta I1meta II ratio is pressure-dependent at SoC, when rhodopsin is located in
the lipid environment of the disk membrane. (43) Application of pressure increases the
proportion of meta I at the expense of meta II. Complete reversal of the meta I ~ meta II
equilibrium to meta I is obtained at about 3 x 106 kg/m2. These observations show that
the formation of meta II results in a volume expansion of the membrane. This volume
expansion depends on the lipid-protein interaction in the membrane because the pressure
effect is absent if rhodopsin is extracted into detergent micelles. In detergent extracts,
rhodopsin also exposes some additional sulfhydryl groups to reaction with sulfhydryl
reagents upon formation of meta II. In the disk membrane, however, rhodopsin does not
appear to expose additional sulfhydryl groups upon the formation of meta II. (44) These
observations are some of several indicating that detergent extracts are not adequate models
for studying the behavior of the formation of meta II, and that the membrane environment
is very important. The steps up to and including the meta I to meta II interconversion are
particularly significant because spectral changes subsequent to meta 11 formation are much
too slow to account for the brief time delay of visual excitation after light absorption.
It was mentioned in Section 9.2.3 that the retinal photoreceptor membranes contain a
large concentration of highly unsaturated fatty acids with 22 carbons and 6 double bonds
(22:6), which are also rich in synaptic endings.(l7) It has recently been found that 22:6 is
required in membrane phospholipids to reconstitute the pressure dependence of the meta I
~ meta II reaction. (45) The double bonds are thought to coil up like a spring, and are able
to accommodate conformational changes of the protein at meta II that expand the area of
the protein in the membrane plane.

9.4.2. The Visual Cycle

The bleaching of rhodopsin leads to the slow liberation of all-trans-retinal. In the
eye, bleached rhodopsin is regenerated by the enzymatic conversion of all-trans-retinal
into ll-cis-retinal. The ll-cis-retinal spontaneously combines with bleached rhodopsin
(opsin) to regenerate rhodopsin. There are many intermediate steps in the formation of
ll-cis-retinal that are diagrammed in Fig. 9-15. The entire process from the bleaching of
rhodopsin to its regeneration is called the visual cycle. In isolated retinas, purified ROS,
or detergent extracts, there is little or no spontaneous regeneration. If ll-cis-retinal is
added to the bleached rhodopsin in a membrane environment or in one of a very few
detergents (e.g., digitonin, Tween-80, octylglucoside, nonylglucoside, or dodecyl mal-
258 Chapter 9

toside), regeneration of the visual pigment color may be accomplished. Retinal and other
aldehydes make membranes leaky to ions if the membranes contain the lipid phos-
phatidylethanolamine, but the corresponding alcohols do not. (46) The ROS membranes
are rich in phosphatidylethanolamine. Retinal does not build up in the retina upon bleach-
ing because an oxidoreductase activity (Fig. 9-17) reduces it rapidly to retinol (Fig. 9-11,
Structure IV) with the consumption of reduced nicotinamide adenine dinucleotide phos-
phate (NADPH). Some of the retinal-reducing activity survives in isolated ROS, but the
NADPH is largely depleted.
The events in the visual cycle, after retinal formation, have not been fully established
in detail. If a large amount of pigment is bleached, the retinal is reduced to retinol; the
retinol is esterified to a fatty acid and transported to the pigment epithelium. The visual
cycle is completed by the isomerization to II-cis, return to the ROS, ester hydrolysis, and
oxidation of retinol to retinal; but the localization and sequences of the events are not
entirely clear. There is some evidence for a "short visual cycle" where retinal is iso-
merized without leaving the ROS, if bleaching of the visual pigments occurs at fluences
closer to physiological light levels. (47)

9.4.3. Psychophysics and Action Spectra

One powerful way to study the sensory system is to ask an intact animal to report in
some way what it detects, or if it detects a carefully controlled stimulus. This method is
called psychophysics. Another powerful method, widely used in photobiology, is to
measure the quantum efficiency of a response as a function of wavelength of excitation.
The resulting curve is called the action spectrum of the response (Section 1.5.5).
Using psychophysical methods, it has been shown in numerous experiments that the
action spectrum of vision matches precisely the absorption spectrum of the visual pig-
ments present. These measurements are straightforward in pure rod or pure cone retinas,
but are more complex in mixed retinas that have both rods and cones. The sensitivity of a
mixed retina matches the absorption of rod visual pigment under dim light, and matches
the absorption spectra of cone visual pigment under brighter light stimuli. After stimula-
tion with bright light, the photoreceptors show a marked decrease in sensitivity; this
phenomenon is called light adaptation. It is a familiar experience to be momentarily
blinded by going from a dim to a brightly lighted environment, then to adapt rapidly to
seeing well in the bright environment. Similarly, in going from bright to dim light the
photoreceptors regain sensitivity. This phenomenon is called dark adaptation and is also
familiar from everyday experience.
Dark adaptation shows complex kinetics. The cones recover sensitivity rapidly;
however, while the cones are active, they seem to repress the recovery of the rods. (24) The
rods recover sensitivity much more slowly than the cones, with a half-time of about 5-8
min. Complete dark adaptation of the rods takes about 30 min. Understanding the mecha-
nisms of light and dark adaptation will require more knowledge of the internal mecha-
nisms of the photoreceptors (Section 9.5).

9.4.4. The Electroretinogram (ERG)

The electrical responses of the photoreceptor cells to light stimuli may be measured
in a number of ways. A simple and surprisingly informative method is to measure the
Vision 259

potential generated across the whole eye in intact animals, excised eyes, or across isolated
retinas, in response to light stimuli. Plots of these potentials are called the electroretino-
gram (ERG) and may be measured with gross electrodes (cotton wicks soaked in saline) or
with extracellular microelectrodes. These responses have been separated into components
that come from the photoreceptor cells (the a wave) or P III, from deeper layers of the
retina (b wave), and from the pigment epithelium (c wave). Therefore, the a wave of the
ERG allows the functioning of the photoreceptors to be studied in intact retina.
The a wave can be isolated in retina preparations by the addition of the amino acid,
sodium aspartate, which serves to deactivate the horizontal cells below the receptors.(48)
The a wave is highly sensitive to anoxia and to the removal of glucose, and therefore
requires oxidative energy sources. Furthermore, the a wave is suppressed by the removal
of Na + or by the addition of Ca2+. The appearance of the a wave shows a several-
milliseconds delay after a stimulus flash.

9.4.5. Early Receptor Potential (ERP)

Very bright stimulus flashes produce a potential that appears with no detectable delay
after the stimulus. (49) This response has been called the early receptor potential (ERP). In
contrast with the a wave of the ERG, the ERP is highly resistant to anoxia and is not
dependent on the identity of the salts present. The ERP depends linearly on the amount of
rhodopsin bleached, whereas the other potentials generated by the retina are highly
nonlinear in the amount of rhodopsin bleached. (50) Utilizing the photobackreactions of the
bleaching intermediates, it has been shown that the ERP appears to monitor the intercon-
version of rhodopsin intermediates. (51)
The ERP (which Hagins calls the fast photovoltage, or FPV) appears to correspond to
charge displacements across the plasma membranes of the rod. (52) Presumably rhodopsin
changes its conformation during interconversion between bleaching intermediates, and
these conformational changes lead to a change in the distribution of charges across the
membrane. The rhodopsin in the pinched-off disk membranes cannot contribute any net
voltage to the ERP, so the ERP is due only to changes of the rhodopsin in the enveloping
plasma membrane (including, of course, the disks that have not yet pinched off at the base
of the outer segment). The spectral changes after bleaching result primarily from the
rhodopsin that is in the disk membrane, since 97-99% of the ROS membrane area is due
to the disk. The close similarity of the kinetics of the spectral changes and the ERP
changes, in both the forward and backward directions, indicate that rhodopsin in the disk
and plasma membranes behave very similarly.
The ERP tells us little about the mechanism of visual excitation but does indicate that
rhodopsin undergoes a conformational change associated with charge movement after
bleaching and that a conformational change is particularly evident upon formation of meta
II rhodopsin. Evidence for conformational changes upon meta II formation have been
detected in a number of experiments,(34) including the chemical reactivity experiments
described in Section 9.4.1. All the intermediates up to, but not including, meta II form in
dehydrated gelatin films prepared from digitonin micelles. If the light-exposed films are
hydrated in the dark, meta II is formed and is therefore the first intermediate that requires
an aqueous environment. Meta II formation is also blocked in lipid-stripped rhodopsin. (42)
All these considerations suggest that the formation of meta II requires a substantial
260 Chapter 9

conformational rearrangement and the formation of meta II is a leading candidate for the
step that triggers visual excitation. (34)

9.4.6. Intracellular Recording of Photoreceptor Responses

Intracellular recording involves piercing a cell with a glass micropipet electrode of
very small diameter and directly measuring the electrical potential across the cell mem-
brane. Electrical contact is provided by filling the micropipet with a concentrated salt
solution. Intracellular recording has been very useful in understanding the electrical
properties and ion flows in a variety of cells, the most prominent examples being nerve
axons and muscle cells. The relatively small size of photoreceptor cells has made intra-
cellular recording difficult. However, evolution of the technique has made it possible to
obtain intracellular recordings from several different species of amphibian retinas, which
happen to have particularly large photoreceptors. Numerous studies have also been made
of the intracellular responses of other cells deeper in the retina that are stimulated by the
photoreceptors. (53)
The photoreceptors do not give spike responses like axons but give smoothly graded
responses to graded light stimuli. (48,54) This behavior is not unusual for short neurons.
The photoreceptors have a negative potential inside, as do all cells; however, they are
unusual in that they hyperpolarize (become more negative inside) rather than depolarize in
response to stimuli. The graded hyperpolarizing property upon stimulation is also shown
by the horizontal and bipolar cells deeper in the retina. The ganglion cells behave like
"standard" neurons; they depolarize in response to stimuli and have an output that
consists of a train of spikes.
Intracellulear recordings have been made from both cones and rods, as shown by
spectral sensitivities, characteristic saturation behavior, and intracellular dye marking
after the recordings. The rods show much slower electrical kinetics than cones, particu-
larly in the recovery phase after a brief stimulus. Rods have been shown to be electrically
connected to neighboring rods in many species and to receive additive signals from them.
This additive behavior contributes to the high sensitivity of rod vision, although it must
cause reduced acuity (spatial resolution) in rod vision. With broad-field stimulation of
neighboring rods, individual rods give detectable signals at much less than I quantum
absorbed per rod. Rods give responses that increase with intensity up to about 300 quanta
absorbed per rod. Above this intensity, the rod response saturates in amplitude and shows
increasingly slow recovery after brighter supersaturating stimuli. The voltage response
amplitude of photoreceptors follows the relation V/Vmax = In/(ln + Kn), where V is the
voltage response, I is the stimulus intensity, n is a constant that ranges from 0.8 to 2.0
depending on the photoreceptor; K is the stimulus intensity that produces half-saturation,
and has a value of about 30 quanta for rods.
Dark-adapted cones give detectable responses from about 100 photons absorbed per
cone, and increase with intensity up to about 104-105 photons per cone.(54) The half-
saturating intensity is about 1200-1300 photons per cone. These numbers are only ap-
proximate, but it is clear that the cones are much less sensitive than the rods. Upon
exposure to conditioning background lights, cones are desensitized and give detectable
responses to flashes above the background, up to about 108 quanta absorbed per cone. The
Vision 261

two types of receptors, rods and cones, together cover a useful dynamic range of inten-
sities of about 1010.

9.4.7. Extracellular Microelectrode Recording

The spatial localization of currents may be measured in slices of retina. (55) This
method measures voltage gradients between two relatively low-resistance extracellular
microelectrodes. The extracellular medium has a small but significant resistance, and
currents between two regions of the tissue produce a small voltage between two electrodes
placed in these regions. The spatial distribution of voltage gradients along the photorecep-
tors has been used to locate sources and sinks of current, and to measure the magnitude of
these currents.
In the invertebrate retina there is an increase in Na + current into the photoreceptor
outer segment in the light,(56) but the opposite result is found in vertebrate photorecep-
tors. (5) For example, in the rat retina there is a large Na + current (= 107 Na + ions rod - 1
S - 1) from the inner segment to the outer segment in the dark, and this "dark current" is
decreased by light. One photon absorbed per rod decreases the current by about 1%. The
decrease is linear up to about half the maximum response, approximately 30 photons
absorbed per rod. The full response curve to flash illumination follows the relation V/Vmax
= I/(J + 30),(59) which is similar to that found by intracellular recording on other rods.
A high concentration of Ca2 + (about 10 mM) reversibly mimicked the effect of light
in repressing the photoreceptor Na + current. (57) It was proposed that Ca2 + was the
intracellular signal that controls Na + permeability but that extracellularly applied Ca2 +
did not get into the cell very effectively. The Na + dark current runs down in the presence
of ouabain (a Na+-K+ ATPase inhibitor) and the rate of rundown is slowed by light
exposure or by extracellular Ca2 +. The "calcium-coupling" hypothesis proposes that
Ca2 + is the intracellular signal for visual excitation. More will be said about this in
Section 9.5.3.

9.4.8. Rod Outer Segment Ion Currents Measured by Osmotic Responses

Fresh isolated frog or rat ROS shrink rapidly after hyperosmotic shock. (58) The rods
remain shrunken in most salt solutions. However, in NaCl solution in the dark, the rods
rapidly recover to their native length. If shrunken rods in NaCI are illuminated or exposed
to high Ca2 + concentrations (10 mM), they do not recover their length. These and other
experiments show that light and extracellular Ca 2 + can control the Na + flux into outer
segments. The recovery after a hyperosmotic NaCI shock is easily lost if the preparation is
not kept highly concentrated in ROS or if the ROS age more than about 15 min after the
retina is dissected.
The rods act as if they have a "spring" in the 15-nm space between the disks; this
space collapses in high salt but returns to 15 nm in NaCI solutions in the dark, as shown by
freeze-fracture measurements of the disk membrane spacing. The ROS plasma membrane
appears to be permeable to NaCl in the dark, which allows recovery from the osmotically
induced shrinkage. From the rate of recovery, it has been calculated that 109 Na+ ions/s
pass through the plasma membrane of each frog rod in the dark and that the absorption of
1 quanta/rod decreases the Na+ permeability by 1_3%.(58)
262 Chapter 9

9.4.9. Direct Measurements of the Photocurrent of Single Rods

Studies of the photocurrent of retinal rods has been facilitated by a technique where-
by the outer segment is sucked into a snug-fitting pipet tip. (59,60) Most of the plasma
membrane current of the rod passes through the suction pipet. This large pipet has a much
lower resistance than the tiny intracellular electrodes, and therefore the suction pipet has a
far superior signal-to-noise ratio. The rod response to single photons is very reproduci-
ble, (60) and this indicates that a sufficiently large number of internal transmitter molecules
(greater than many hundreds) must be produced per photon to keep the statistical fluctua-
tion in the response so low. A single absorbed photon decreases the light sensitive
membrane current by 3_5%,(55,60) and fewer than 100 photons saturate the response of
the dark-adapted cell. This same method has been adapted to cones and has been used to
obtain excellent data on the spectral sensitivity of cones. (61)

9.4.10. Patch-Clamp Experiments Are Powerful Tools

Recently, "patch-clamp" experiments have been utilized to great advantage in the
study of retinal rods. Large pipets are tightly sealed with a very high electrical resistance
(so called "gigaohm seals") to the rod plasma membrane.
The plasma membrane patch may be pulled off to study its electrical properties. (62)
Alternatively, the plasma membrane patch may be blown out of the pipet while maintain-
ing the seal to allow diffusion of substances from the large patch pipet in the so-called
"whole-cell" configuration. (63,64) Results ofthis method will be discussed more fully in
the next section' on the mechanism of visual excitation.

9.5. THE MECHANISM OF VISUAL EXCITATION

9.5.1. The Internal Transmitter Hypothesis

The topological organization of the rod cell requires that an intracellular "transmit-
ter" carry a signal from the site of light absorption in the disk membrane to the site of
Na + conductance change on the outer plasma membrane. Only a few newly synthesized
disks at the base of the rod are connected with the plasma membrane, and most of the
disks have pinched off and are not connected. The plasma membrane accounts for only
about 1-3% of the total outer segment membrane area depending on the species. The
remaining 97-99% of the membrane area is made up of the disks. Therefore, the disks
must contain most of the rhodopsin, since rhodopsin could not be much more closely
packed in the plasma membrane than it is in the disk. Since the rod is so sensitive, the
disks not connected to the plasma membrane must be used to cause the cell to respond.
This argument leads to the requirement for a signal or transmitter between the disks and
the plasma membrane. It is highly unlikely that this signal is a brute force electrical
coupling, which would require the disks to take up or release sufficient amounts of an ion
to make a large contribution to the membrane potential. A more detailed discussion of the
implausibility of electrical coupling can be consulted. (5) The high sensitivity of the rod is
consistent with the intracellular transmitter being a chemical with specific receptors on the
plasma membrane. A specific chemical transmitter could effect the plasma membrane
Vision 263

Na + permeability by binding to specific sites, even if it were released or taken up in very

small amounts.

9.5.2. Requirements for a Transmitter

A transmitter would have to be produced or depleted sufficiently rapidly in the light
to change the concentration at the plasma membrane. The time involved is about the time
for a peak response to a dim flash (100-200 ms).
After a single photon is absorbed, a transmitter would have to be produced or
depleted in sufficient quantity to detect the change in concentration above the background
leakage of the rod and above the thermal noise in the conductivity of the plasma mem-
brane.
A transmitter would have to bind to the plasma membrane and the binding must
affect the Na + conductance. Experimental manipulation of the transmitter concentration
in the dark should mimic the effect of light.
In the dark, the transmitter concentration must return to its initial level rapidly
enough to explain the restoration rate of plasma membrane voltage. Depending on the
species, the restoration rate has a half-time of 0.15-2.0 s with dim stimuli and gets much
longer with bright stimuli.
Excited rhodopsin in the disk membrane must control the level of the transmitter in
the light. Since the photoreceptor cell has such a high sensitivity, it is likely that the dark-
adapted level of the intracellular transmitter is very low. In this way a relatively small
change in the absolute amount of transmitter could cause a relatively large fractional
change in the transmitter. From studies of photovoltages in turtle cones, Baylor and
Fuortes(65) proposed that light produces a "substance which decreases the permeability of
membrane channels acting as a shunt of the membrane in darkness." Yoshikami and
Hagins(66) proposed that this substance was free calcium ions in the cytoplasm.

9.5.3. Evidence for the Role of Calcium

Calcium clearly plays a role in photoexcitation (see Ref. 67 for a general review of
the roles of calcium in cellular regulation and Ref. 68 for a review of the role of calcium in
photoreceptors). For a time it seemed likely that free cytoplasmic calcium was the internal
transmitter because of t..'1e following observations:

1. Elevated external calcium mimics light effects on the dark current and calcium
ionophores potentiate this effect. (66)
2. Electrophoretic injection of calcium hyperpolarizes the rod membrane just as does
light. (69)
3. In very low external calcium solutions the rod depolarizes greatly to about 10
mY, the potential at which the light-sensitive current reverses as would be ex-
pected if the light-sensitive conductance became larger and dominant. Light then
drives the membrane potential to the same hyperpolarized level as measured at
normal calcium levels, indicating that only the light-sensitive conductance has
been altered by calcium (see Ref. 68).
4. Injection of calcium chelators increases the outer segment membrane current, (64)
264 Chapter 9

which is expected if the lowering of internal free calcium opens the light-sensitive
channels.
5. Light causes an increase in extracellular calcium,C70) which is expected if light
causes a transient rise in internal calcium. Thus, the calcium hypothesis seemed
rather well established, but several results since 1984 have cast very serious doubt
on the calcium hypothesis (Section 9.5.4).

9.5.4. Evidence against the Calcium Hypothesis for Visual Excitation

The light-sensitive channels were found to be quite permeable to calcium, and large
Ca2 + influxes do not immediately mimic the light-induced dark current suppression. (71)
There is an enormous flood of Ca2 + that enters through the light-sensitive channels in the
dark current that would dwarf the light-induced Ca2 + release required by the Ca2 +
hypothesis. (68)
Infusion of large amounts of calcium into the rod through whole-cell patch pipets
(Section 9.4.10) does not decrease the light sensitivity. (64) Infusion of calcium chelators
increased rather than decreased light sensitivity, (64) a finding in total conflict with the
calcium hypothesis.
Free calcium has been found to decrease during the light response,<72) a result
consistent with considerable other evidence. (68) Taken together, the evidence against
Ca2 + as the internal transmitter is very strong. (68) However, Ca2 + has profound and
rapid effects on rod transduction and this must be accounted for in any complete theory of
transduction.

9.5.5. Cyclic GMP as Internal Transmitter

The rod outer segment contains a set of enzymes that regulate cGMP. Light-excited
rhodopsin activates a highly amplified series of enzymatic reactions, the "cGMP cas-
cade" that leads to the destruction of cGMP and the hyperpolarization of the cell. (34)
Relationships between the components of the cGMP cascade hypothesis for visual
excitation are given in Fig. 9-18, where part a shows the activation reactions and part b
shows the deactivation phase.
Light-excited rhodopsin (R*) interacts transiently with a GTP-binding protein(73) on
the cytoplasmic face of the membrane in vertebrate photoreceptors. The GTP-binding
protein in the rod photoreceptor is denoted Gy in Fig. 9-18. The GTP-binding protein
associated with R* exchanges GDP for GTP(74,75) which excites the GTP-binding protein
(Gy-GTP). Large numbers of GTP-binding proteins are excited per R* ,(75) and many of
the excited GTP-binding proteins turn on cGMP phosphodiesterase. (76) The excited
cGMP phosphodiesterase in turn hydrolyzes cyclic GMP.
The presence of cGMP opens a cation channel in the outer plasma membrane(62,77)
and in the disk membrane.(34,75,79) This channel is not controlled by Ca2 + nor is phos-
phorylation required to control it. The plasma membrane channel opened by cGMP is
thought to be closed in the light due to transient depletion of cGMP by the light-stimulated
phosphodiesterase, resulting in hyperpolarization of the photoreceptor cell membrane.
Figure 9-18b shows steps in the shutoff of R*. R* is heavily phosphorylated (7-9
phosphates/R*).(80) The phosphorylation of R* is essential to rapid reduction in R*
Vision 265

_3_Na~.
r;, +~ 'i:I,~~=-! ..
--- \..J -- -cGMP
r'H J-
cGMP
Plasma Membrane

3Na+ Ca++ Na+

~
.,,*;' I~:t: '::"2Pi

AlP GD~MP Pi

Gt: ~P

b
II-CiS~
all-trans 1Li hI

9PIY R~R·
V9ATP
TR.t;~Pgk 9ADP
Fig. 9-18. (a) The cGMP cascade model for the activation of the electrical response of the vertebrate rod. Light
converts rhodopsin to R * , which forms a transient complex with the GTP-binding protein (G v ) and catalyzes the
exchange of GTP for GDP. The Gy-GTP complexes the inhibited phosphodiesterase (D-I ) to activate the
hydrolysis of cGMP. The reduced level of cGMP closes the Na+ -Ca2 I,XXI channels on the plasma
membrane, which hyperpolarizes the cell and stops the entry of Ca2/ ,xXl. A Na+-Ca2 + exchanger in the
plasma membrane reduces the Ca2 + concentration in the cytoplasm, which accelerates the production of cGMP,
presumably by activating the guanylate cyclase-calmodulin complex (CM-GC). Other activities required to
maintain the cGMP and GTP levels are guanylate kinase (GK), nucleotide diphosphokinase (NDPK), and
pyrophosphatase (P-Pase). Gv-GTP is spontaneously inactivated by an endogenous GTPase activity to form Gy-
GDP plus Pi. (b) The inactivation of the excited receptor (R*). R* is rapidly phosphorylated by ATP and
rhodopsin kinase (RK) with up to nine phosphates/R* (R*-Pg). R*-Pg is fully inactivated by binding the S
antigen (48°K or arrestin). The detailed pathway is poorly known after this point, but probably involves first the
spontaneous inactivation of R*-P9 to RO-P9 (R O denotes the inactive conformation of R), loss of S antigen, loss
of phosphate by phosphatase (Pase), and replacement of all-trans-retinal by II-cis-retinal. [Taken from P. A.
Liebman, K. R. Parker, and E. ,,,"- Dratz, Ann. Rev. Physiol. 49, 765-791 (1987).]
266 Chapter 9

activity and to ultimate blockage by binding a 48-kDA protein (arrestin). (81) A hypo-
thetical complete restoration cycle is shown but the later steps are poorly understood.
Evidence for a large increase in metabolic flux of cGMP in whole retina in the light
and for an increase in free cGMP in the light has been presented. (82) This recent work
implies that the simple form of the model presented in Fig. 9-18a that requires a decrease
in cGMP in the light might not be adequate. (82) Changes in the flux of cGMP rather than
the change in level of cGMP may be important in photoreceptor physiology, although the
detailed mechanism is obscure. (82)
The cGMP cascade hypothesis for visual excitation (Fig. 9-18) is now very widely
accepted. The detailed mechanism of operation of the cGMP cascade and its control is a
rich field of investigation. (34,68) There has been considerable speculation as to whether
the powerful effects of calcium and calcium chelators on the photoreceptor (Section 9.5.3)
might be explained by interaction with the cGMP cascade. The leading hypothesis for this
mechanism is that raising calcium may inhibit the guanylate cyclase (CM/GC in Fig.
9-18a), which would drop the concentration of cGMP and thus mimic light. Lowered
Ca2+ would then stimulate guanylate cyclase, raising cGMP and opening plasma mem-
brane channels. (68) A potential problem with this mechanism is that the isolated cyclase is
only weakly inhibited by calcium. (83)
The kinetics of many of the enzymatic reactions in the retinal rod are strongly
concentration-dependent, as if weakly bound factors that can affect rates are released
when the cells are disrupted and diluted. (84) In particular, the GTP-binding protein, which
shuts itself off by hydrolysis of GTP, hydrolyzes GTP very slowly (1-3/min) when
studied in dilute form, but speeds up about lOO-fold near physiological concentration. (84)
All the excitation reactions (Fig. 9-18a) are concentration-independent, thus all the in-
teracting components appear to be membrane-bound. In contrast, all the shutoff reactions
(Fig. 9-18b) appear to be concentration-dependent, and evidently require weakly bound
factors that may be diluted out. (84)
It is likely that Ca2+ acts as a negative feedback signal derived from the dark
current. Ca2+ enters the light-regulated channel and is extruded by the Na/Ca exchanger
(Fig. 9-18a). Cytoplasmic Ca2 + drops in the light because the influx is stopped and the
exchanger keeps pumping it out. Calcium chelates alter the time constant of the feedback
signal and can give an electrical overshoot beyond the normal steady-state voltage. (64)
The calcium feedback may cut off the excitation and account for the faster electrical
properties of cones, and also provide a basis for explaining light and dark adaptation.
There may be other mechanisms for Ca2 + regulation in photoreceptors. The mem-
brane lipid, phosphatidylinositol diphosphate (PIP2), is reported to decrease in the light in
both vertebrate(85) and invertebrate(86) photoreceptors. The PIP2 hydrolysis product, in-
ositol trisphosphate (IP 3 ), increases in the light in invertebrate photoreceptors.(86) There
are reports of protein kinase C activation in the light(87) and inositol trisphosphate modula-
tion of calcium metabolism in retinal rods. (88)

9.6. LIGHT DAMAGE TO THE RETINA

As basic knowledge of the visual system biochemistry is collected, we may hope to

gain insight into medical problems of the retina. For example, many types of retinal
Vision 267

photoreceptor degeneration are known. Some types are quite common in older people, but
some types, such as retinitis pigmentosa, all too often afflict young and old alike. One
striking property of normal photoreceptors is that they can be severely damaged by
exposure to excessive light. (89,90) In many different animal species the light levels for
severe damage are only slightly above the levels commonly encountered by humans.
Damaging effeets are marked in some animals exposed to the light levels equivalent to a
bright beach for 8 h or to normal ambient illumination for 2-3 days. (91-93) In experimen-
tal animals the light-induced damage is largely reversible, but studies have not yet been
done to see if there is a cumulative effect of repeated light damage.
The highly unsaturated membrane fatty acids in isolated, purified rod photoreceptor
membranes are extremely sensitive to damage by atmospheric oxygen in the dark. (94) The
regenerability of opsin is irreversibly lost by incubation in the presence of oxygen in the
dark. The photoreceptor membranes are clearly very sensitive to damage by oxygen.
Furthermore, light might be expected to facilitate oxygen damage to the photoreceptors by
a singlet oxygen-mediated photodynamic mechanism. (26)
High levels of vitamin E in the isolated membrane are helpful in retarding this
oxygen damage. However, vitamin E is consumed in air and other potent mechanisms
must be present in the intact system to oppose oxygen damage. (95) The photoreceptors are
the most metabolically active tissues in the body, containing unusually large numbers of
mitochondria, and must absorb large quantities of oxygen from the surrounding medi-
um.(26) Oxygen metabolism by mitochondria can leak reactive free-radical intermediates,
which pose a threat to the tissue. (26)
Even potent protective mechanisms against oxygen and light damage must have
limits. This may be the reason that light levels slightly above those commonly encoun-
tered are damaging to animal photoreceptors. Potentially damaging light levels are en-
countered by some people, and might be expected to produce cumulative damage that is
evident in advanced age. Nutritional factors (e.g., vitamin E and other antioxidants)
would also be expected to have a role in maintaining the long-term health of the pho-
toreceptors. (26) More drastic forms of photoreceptor degeneration may be due to genetic
or disease-related deficiencies in the normally potent protective mechanism against lipid
oxidation. The above hypotheses have potential medical importance that can be tested
with the present knowledge of photoreceptors. (26)

9.7. REFERENCES

1. R. A. Eakin, Structure of invertebrate photoreceptors, Handbook Sensory Physiol. 7(1), 625-684 (1972).
2. T. H. Goldsmith, The natural history of invertebrate visual pigments, Handbook Sensory Physiol. 7(1),
685-719 (1972).
3. M. F. Moody, Photoreceptor organelles in animals, Bioi. Rev. 39, 43-86 (1964).
4. M. G. F. Fuortes (ed.), Physiology of photoreceptor organs, Handbook Sensory Physiol. 7(2) (1972).
5. W. A. Hagins, The visual process: Excitatory mechanisms in the primary receptor cells, Ann. Rev. Biophys.
Bioeng. 1, 131-158 (1972).
6. A. E. Blaurock and M. H. Wilkins, Structure of frog photoreceptor membranes. Nature 223, 906-909
(1969).
7. R. W. Young, Biogenesis and renewal of visual cell outer segment membranes, Exp. Eye Res. 18,215-223
(1974).
268 Chapter 9

8. M. 1. Hogan, I. S. Wood, and R. H. Steinberg, Phagocytosis by pigment epithelium of human retinal

cones, Nature 252, 305-307 (1974).
9. W. Stoeckenius and R. A. Bogomolni, Bacteriorhodopsin and related pigments of halobacteria, Ann. Rev.
Biochem. 52, 587-616 (1982).
10. E. A. Dratz and P. A. Hargrave, The structure of rhodopsin and the rod outer segment disk membrane,
Trends Biochem. Sci. 8, 128-131 (1983).
11. M. Applebury and P. A. Hargrave, Molecular biology of the visual pigments, Vision J1es. 26, 1881-1895
(1986).
12. D. S. Papermaster and W. 1. Dreyer, Rhodopsin content in the outer segment membrance of bovine and
frog retinal rods, Biochemistry 13, 2438-2444 (1974).
13. P. P. Nemes and E. A. Dratz, The preparation of isolated osmotically intact bovine rod outer segment disk
membranes, Methods. Enzymol. 81(H), 116-123 (1982).
14. W. L. Stone, C. C. Farnsworth, and E. A. Dratz, A reinvestigation of the fatty acid content of bovine, rat
and frog rod outer segments, Exp. Eye Res. 28, 387-397 (1979).
15. G. P. Miljanich, L. A. Sklar, D. L. White, and E. A. Dratz, Disaturated and dipolyunsaturated phos-
pholipids of the bovine retinal rod outer segment disk membranes, Biochim. Biophys. Acta. 552,294-306
(1979).
16. R. M. Benolken, R. E. Anderson, and T. G. Wheeler, Membrane fatty acids associated with the electrical
response in visual excitation, Science 182, 1253-1254 (1973).
17. E. A. Dratz and A. J. Deese, The role of docosahexaenoic acid (22:6w3) in biological membranes:
Examples from photoreceptors and model membrane bilayers, in: Health Effects of Polyunsaturated Fatty
Acids in Seafoods (A. Simopolous and R. Kifer, eds.), pp. 319-351, Academic Press, New York (1986).
18. E. A. Dratz, G. P. Miljanich, P. P. Nemes, J. E. Gaw, and S. Schwartz, The structure ofrhodopsin and its
disposition in the rod outer segment disk membrane, Photochem. Photobiol. 29,661-671 (1979).
19. M. Chabre, X-ray diffraction studies of retinal rods. I. Structure of the disk membrane, effect of illumina-
tion, Biochim. Biophys. Acta. 382, 322-335 (1975).
20. P. M. Hargrave, Rhodopsin chemistry, structure and topography, Prog. Ret. Res. 1, 1-51 (1982).
21. G. P. Miljanich, M. F. Brown, S. Mabrey-Gaud, E. A. Dratz, and J. M. Sturtevant, Thermotropic behavior
of retinal rod membranes and dispersions of extracted phospholipids, J. Membrane Bio!. 85, 79-86 (1985).
22. G. S. Harbison, S. O. Smith, J. A. Pardoen, C. Winkel, J. Lugtenburg, 1. Herzfeld, R. Mathies, and R. G.
Griffin, Dark-adapted bacteriorhodopsin contains 13-cis, 15-syn and all-trans, IS-anti retinal Schiff bases,
Proc. Natl. Acad. Sci. USA 81, 1706-1709 (1984).
23. C. D. B. Bridges, The rhodopsin-porphyropsin visual system, Handbook Sensory Physio!. 7(1),417-480
(1972).
24. W. A. H. Rushton, The Ferrier lecture: Visual adaptation, Proc. Roy. Soc. B 162, 20-46 (1965).
25. A. L. Zimmerman and D. A. Baylor, Single channel currents from the cyclic GMP sensitive conductance of
retinal rod outer segments, Nature 321,70-72 (1986).
26. G. J. Handelman, E. A. Dratz, C. C. Reay, and F. J. G. M. van Kuijk, Carotenoids in the human macala
and whole retina, Invest. Ophtha!. & Vis. Sci. 29, 850-855 (1988).
27. P. A. Liebman, Microspectrophotometry of photoreceptors, Handbook Sensory Physiol. 7, 481-528
(1972).
28. R. A. Cone, Rotational diffusion of rhodopsin in the visual receptor membrane, Nature [New Biol.J 236,
39-43 (1972).
29. M. Poo and R. A. Cone, Lateral diffusion of rhodopsin in the photoreceptor membrane, Nature 247, 438-
441 (1974).
30. P. A. Liebman and G. Entine, Lateral diffusion of visual pigment in photoreceptor disk membranes,
Science 185, 457-459 (1974).
31. B. Honig, U. Dinu, K. Nakanishi, V. Balogh-Nair, M. Gawinowicz, M. Arnabo1di, and M. Motto, The
external point charge model for wavelength regulation in visual pigments, J. Am. Chem. Soc. 101,7084-
7086 (1979).
32. J. Nathans, D. Thomas, and D. S. Hogness, Molecular genetics of human color vision: The genes of
inherited variation in human color vision, Science 232, 193-202 (1986).
33. J. Nathans and D. S. Rogness, Isolation and nucleotide sequence of the gene encoding human rhodopsin,
Proc. Nat!. Acad. Sci. USA 8, 4851-4855 (1984).
Vision 269

34. P. A. Liebman, K. R. Parker, and E. A. Dratz, The molecular mechanism of visual excitation and its
relation to the structure and composition of the rod outer segment, Ann. Rev. Physiol. 49,765-791 (1987).
35. D. S. Kliger, S. 1. Milder, and E. A. Dratz, Solvent effects on the spectra of retinal Schiff bases. I. Models
for the bathochromic shift of the chromophore spectrum in visual pigments, Photochem. Photobiol. 25,
277-286 (1977).
36. H. Kakitani, T. Kakitani, H. Rodman, and B. Honig, On the mechanism of wavelength regulation in visual
pigments, Photochem. Photobiol. 41, 471-479 (1985).
37. B. Honig and T. G. Ebrey, The structure and spectra of the chromophore of the visual pigments, Ann. Rev.
Biophys. Bioeng. 3, 151-177 (1974).
38. A. Kropf, B. P. Whittenberger, S. P. Goff, and A. S. Waggoner, The spectral properties of some visual
pigment analogs, Exp. Eye Res. 17, 591-606 (1973).
39. R. S. H. Liu, A. E. Asato, M. Denny, and D. Mead, The nature of restrictions in the binding site of
rhodopsin: A model study, 1. Am. Chern. Soc. 106, 8298-8300 (1984).
40. E. L. Menger and D. S. Kliger, Photoisomerization kinetics of II-cis retinal, its Schiff base, and its
protonated Schiff base, 1. Am. Chern. Soc. 98, 3975-3979 (1976).
41. T. P. Williams, An isochromic change in the bleaching of rhodopsin, Vision Res. 10, 525-533 (1970).
42. M. L. Applebury, D. M. Zuckerman, A. A. Lamola, and T. M. lovin, Rhodopsin purification and
recombination with phospholipids assayed by the metarhodopsin I ~ metarhodopsin II transition, Bio-
chemistry 13, 3448-3458 (1974).
43. A. A. Lamola, T. Yamane, and A. Zipp, Effects of detergents and high pressures upon the metarhodopsin I
~ metarhodopsin II equilibrium, Biochemistry 13,738-745 (1974).
44. W. 1. DeGrip, G. L. M. Van der Laar, F. 1. M. Daemen, and S. L. Bonting, Biochemical aspects of the
visual process. XXill. Sulfhydryl groups and rhodopsin photolysis, Biochim. Biophys. Acta 325,315-322
(1973).
45. E. A. Dratz, N. Ryba, A. Watts, and A. Deese, The role of highly unsaturated fatty acids in membranes,
submitted for publication.
46. S. L. Bonting and A. D. Bangham, On the biochemical mechanisms of the visual process, Exp. Eye Res. 6,
400-413 (1967).
47. W. Zimmerman, M. Yost, and F. 1. M. Daemen, Dynamics and function of vitamin A compounds in the rat
retina after a small bleach of rhodopsin, Nature 250, 66-67 (1974).
48. T. Tomita, Electrical activity of vertebrate photoreceptors, O. Rev. Biophys. 3, 179-222 (1970).
49. K. T. Brown and M. Murakami, Biphasic form of the early receptor potential of the monkey retina, Nature
204, 739-740 (1964).
50. R. A. Cone and W. L. Pak, The early receptor potential, Handbook Sensory Physiol. 1,345-365 (1971).
51. R. A. Cone, Early receptor potential: Photoreversible charge disp'lacement in rhodopsin, Science 155,
1128-1131 (1967).
52. W. A. Hagins and H. Ruppel, Fast photoelectric effects and the properties of vertebrate photoreceptors as
electric cables, Fed. Proc. 30,64-78 (1971).
53. F. S. Werblin and 1. E. Dowling, Organization of retina of the mudpuppy, Necturus maculosus. II.
Intracellular recording, 1. Neurophysiol. 32, 339-355 (1969).
54. D. Baylor, A. Hodgkin, and T. Lamb, The electrical response of turtle cones to flashes and steps of light, 1.
Physiol. (Lond.) 242, 685-727 (1974).
55. W. A. Hagins, R. D. Penn, and S. Yoshikami, Dark current and photocurrent in retinal rods, Biophys. 1.
10, 380-412 (1970).
56. W. A. Hagins, H. V. Zona, and R. G. Adams, Local membrane current in the outer segments of squid
photoreceptors, Nature 194, 844-847 (1962).
57. S. Yoshikami and W. A. Hagins, Control of the dark current in vertebrate rods and cones, in: Biochemistry
and Physiology of Visual Pigments (H. Langer, ed.), pp. 245-255, Springer-Verlag, Berlin (1973).
58. 1. E. Korenbrot and R. A. Cone, Dark ionic flux and the effects of light in isolated rod outer segments, 1.
Gen. Physiol. 60, 20-45 (1972).
59. D. A. Baylor, T. D. Lamb, and K. W. Yau, The membrane current of single rod outer segments, 1.
Physiol. 288, 589-611 (1979).
60. D. A. Baylor, T. D. Lamb, and K.-W. Yau, Responses of retinal rods to single photons, 1. Physiol. 288,
613-634 (1979).
270 Chapter 9

61. B. J. Nunn, J. L. Schnapf, and D. A. Baylor, Spectral sensitivity of single cones in the retina of Macaca
fascicularis, Nature 309, 264-266 (1985).
62. E. E. Fesenko, S. S. Kolesnikov and A. L. Lyubarsky, Induction by cyclic GMP of cationic conductance in
plasma membrane of retinal rod outer segment, Nature 321,66-70 (1986).
63. W. H. Cobbs and E. N. Pugh, Jr., Cyclic GMP can increase rod outer-segment light-sensitive current 10-
fold without delay of excitation, Nature 313, 585-587 (1985).
64. T. D. Lamb, H. Matthews and V. Torre, Changes in the photocurrent of salamander retinal rods during
incorporation of calcium buffers, J. Physiol. 372, 315-349 (1986).
65. D. A. Baylor and M. G. F. Fourtes, Electrical response of single cones in the retina of the turtle, J. Physiol.
207, 77-92 (1970).
66. S. Yoshikami and W. H. Hagins, Ionic bases of dark current and photocurrents ofretinal rods, Biophys. J.
10, 60a (1970).
67. H. Rasmussen, Calcium and cAMP as Synarchic Messengers, Wiley, New York (1981).
68. E. N. Pugh, Jr., The nature and identity of the internal excitation transmitter of vertebrate phototransduc-
tion, Ann. Rev. Physiol. 49, 715-741 (1987).
69. J. E. Brown, J. A. Cole and L. H. Pinto, Effects of injections of Ca2+ and EGTA into the outer segment of
the retinal rods of Bufo Marinus, J. Physiol. 269, 707-722 (1977).
70. G. H. Gold and J. I. Korenbrot, Light-induced calcium release by intact retinal rods, Proc. Natl. Acad. Sci.
U.S.A. 77, 5557-5561 (1980).
71. K.-W. Yau and K. Nakatani, Electrogenic Na-Ca exchange in retinal rod outer segment, Nature 311,661-
663 (1984).
72. P. A. McNaughton, B. J. Nunn and L. Cerretto, Measurement of intracellular free Ca2 + concentration in
salamander rod outer segments, Nature 322, 261-263 (1986).
73. H. Kuhn, Light- and GTP-regulated interaction of GTPase 'and other proteins with bovine photoreceptor
membranes, Nature 283, 587-589 (1980).
74. B. B.-K. Fung, and L. Stryer, Photolyzed rhodopsin catalyzes the exchange of GTP for bound GDP in
retinal rod outer segments, Proc. Natl. Acad. Sci. U.S.A. 77, 2500-2504 (1980).
75. P. A. Liebman, and E. N. Pugh, Jr., Gain, speed and sensitivity of GTP binding vs. PDE activation in
visual excitation, Vision Res. 22, 1475-1480 (1984).
76. P. A. Liebman, and E. N. Pugh, Jr., The control of phosphodiesterase in rod disk membranes: Kinetics,
possible mechanisms and significance for vision, Vision Res. 19, 375-380 (1979).
77. J. H. Stern, U. B. Kaupp, and P. R. Macleish, Control of the light-regulated current in rod photoreceptors
by cyclic GMP, calcium and l-cis-diltiazem, Proc. Natl. Acad. Sci. U.S.A. 83, 1163-1167 (1986).
78. A Cavaggioni and R. T. Sorbi, Cyclic-GMP releases calcium from disk membranes of vertebrate pho-
toreceptors, Proc. Natl. Acad. Sci. USA 78, 3964-3968 (1981).
79. K.-W. Koch and U. B. Kaupp, Cyclic GMP directly regulates a cation conductance in membranes of bovine
rods by a cooperative mechanism, J. Bioi. Chem. 260, 6788-6800 (1985).
80. U. Wilden and H. Kuhn, Light-dependent phosphorylation of rhodopsin: Number of phosphorylation sites,
Biochemistry 21, 3014-3022 (1982).
81. U. Wilden, S. W. Hall, and H. Kuhn, Phosphodiesterase activation by photoexcited rhodopsin is quenched
when rhodopsin is phosphorylated and binds the intrinsic 48-kDa protein of rod outer segments, Proc. Natl.
Acad. Sci. USA 83, 1174-1178 (1986).
82. N. D. Goldberg, A. Ames, III, J. E. Gander, and T. F. Walseth, Magnitude of increase in retinal cGMP
metabolic flux determined by 180 incorporation into nucleotide o.-phosphoryls corresponds with intensity of
photic stimulation, J. Biol. Chem. 258, 9213-9219 (1983).
83. D. Fleischman and M. Denisevich, Guanylate cyclase of isolated bovine retinal rod axonemes, Biochemis-
try 18, 5060-5066 (1979).
84. E. A. Dratz, J. W. Lewis, L. E. Schaechter, K. R. Parker, and D. S. Kliger, Retinal rod GTPase rate
increases with concentration: A key to the control of visual excitation?, Biochem. Biophys. Res. Commun.
146, 379-386 (1987).
85. A. Ghalayini and R. E. Anderson, PhosphatidylinositoI4,5-bisphosphate: Light-mediated breakdown in the
vertebrate retina, Biochem. Biophys. Res. Commun. 124, 503-506 (1984).
86. J. E. Brown, L. J. Rubin, A. J. Ghalayini, A. P. Tarvar, and R. F. Irvine, Myo-inositol polyphosphate
may be a messenger for visual excitation in Limulus photoreceptors, Nature 311, 160-163 (1984).
87. c. L. Kapoor and G. J. Chader, Endogenous phosphorylation of retinal photoreceptor outer segment
Vision 271

proteins by calcium-phospholipid-dependent protein kinase, Biochem. Biophys. Res. Commun. 122, 1397-
1403 (1984).
88. K. R. Parker, 1. A. Briggs and E. A. Dratz, Inositol triphosphate stimulates Ca2 + release from toad retinal
rod outer segment preparations, Biophysics 49, 31a (1986).
89. W. K. Noell, U. S. Walker, B. S. Kang and S. Berman, Retinal damage by light in rats, Invest.
Ophthalmol. 5, 450-472 (1966).
90. T. Kuwabara and R. A. Gom, Retinal damage by visible light, Arch. Ophthalmol. 79, 69-78 (1968).
91. K. V. Anderson, F. P. Coyle and W. K. O'Steen, Retinal degeneration produced by low intensity colored
light, Exp. Neurol. 35, 233-238 (1972).
92. T. Lawwill, Effects of prolonged exposure of rabbit retina to low-intensity light, Invest. Ophthalmol. 12,
45-51 (1973).
93. 1. Marshall, 1. Mellerio, and D. A. Palmer, Damage to pigeon retina by moderate illumination from
fluorescent lamps, Exp. Eye Res. 14, 164-169 (1972).
94. C. C. Farnsworth and E. A. Dratz, Oxidative damage of the rod outer segment (ROS) disk membrane and
the role of vitamin E, Biochim. Biophys. Acta 443, 556-570 (1976).
95. G. 1. Handelman and E. A. Dratz, The role of antioxidants in the retina and retina pigment epithelium and
the nature of prooxidant-induced damage, Adv. Free Rad. Bioi. Med. 2, 1-89 (1986).
 10
Photomorphogenesis

10.1. Introduction ..................................................................... 273

10.1.1. Light as a Source of Information ............................................. 273
10.1.2. Diversity of Photomorphogenesis ............................................. 274
10.2. Phytochrome-Mediated Morphogenesis ............................................... 278
10.2.1. Discovery ................................................................ 278
10.2.2. Events under Phytochrome Control ............................................ 280
10.2.2.1. Detection of the Onset of Light ...................................... 280
10.2.2.2. High-Irradiance Responses .......................................... 281
10.2.2.3. Sensing of Light Quality. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 282
10.2.3. Properties of Phytochrome ................................................... 285
10.2.3.1. Physicochemical Properties. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 285
10.2.3.2. Photochemical Reactions ........................................... 286
10.2.3.3. Nonphotochemical Reactions ........................................ 287
10.2.3.4. Differences between Pr and Pfr . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 288
10.2.3.5. Phytochrome from Green Plants. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 289
10.2.4. Primary Mode of Action .................................................... 290
10.3. Blue-Light-Mediated Morphogenesis ................................................. 291
10.3.1. Events under Blue-Light Control ............................................. 292
10.3.2. Photoreceptor(s) ........................................................... 294
10.4. UV Radiation-Mediated Morphogenesis ............................................... 296
10.4.1. Events under UV Radiation Control ........................................... 296
10.4.2. Photoreceptor(s) ........................................................... 297
10.5. Photochromic Adaptation ........................................................... 298
10.6. The Future ...................................................................... 299
10.7. References ...................................................................... 300

10.1. INTRODUCTION

10.1.1. Light as a Source of Information

An organism can utilize the energy oflight in either of two ways (Fig. 10-1). Light can be
used as a source of free energy to facilitate a process, such as photosynthesis (see Chapter
12), that otherwise would be thermodynamically unfeasible. Alternatively, light can be
used as a source of information. It is this second aspect of light with which we will be
concerned.

Lee H. Pratt • Department of Botany, University of Georgia, Athens, Georgia

30602. Marie-Michele Cordonnier • Biotechnology Research, CIBA-GEIGY Corpora-
tion, Research Triangle Park, North Carolina 27709-2257.

273
274 Chapter 10

Fig. 10-1. Light as a source of both energy and

photosynthesis photo morphogenesis information.

It would seem reasonable that an organism can be best adapted to its environment
only if it can sense that environment and respond appropriately to this sensory input. For
example, we can sense potentially harmful ultraviolet radiation and respond to it by
accumulating more melanin in our epidermis, thus protecting ourselves against a subse-
quent exposure to this radiation (Chapter 6). As a motile organism, however, we can also
respond by moving to another, potentially more suitable environment. In contrast, a
nonmotile organism such as a plant or fungus does not have this second alternative
available to it. Instead, it must adapt to its existing environment rather than move to a
better one. The ability to sense incident light energy and to respond to it in appropriate
ways is therefore especially important to such organisms. This ability is most important to
a green plant, since via photosynthesis it also utilizes light as its sole source of free
energy.
Photomorphogenesis encompasses phenomena whereby an organism senses and re-
sponds appropriately to incident light energy. The precise meaning of this term derives
from its three Greek roots. Photomorphogenesis is therefore the development (genesis) of
form (morph) of an organism as influenced by light (photo). Consequently, photomorpho-
genesis deals with events that occur at both cellular and multicellular levels. The underly-
ing phenomena that lead to changes in morphology, however, occur at subcellular and
molecular levels. Hence, this chapter will consider responses to light that occur at all
levels, including the molecular. Furthermore, because of the unique importance of pho-
tomorphogenesis to nonmotile organisms, coverage here will be restricted to photomor-
phogenesis among fungi and plants, including algae.

10.1.2. Diversity of Photomorphogenesis

There are at least four independent aspects of light that can serve as a source of
information. They are its (1) quantity, (2) quality, (3) spatial asymmetry (i.e., direction
from which the light comes), and (4) periodicity (Fig. 10-1). In a few instances, orga-
nisms also respond to the plane of polarization of incident light. A notable example of
such a response is polarotropism of the chloronema of Dryopteris felix-mas. The direction
of the growth of the chloronema, which is a filamentous growth stage of the haploid
generation of this fern, can under certain growth conditions be strictly determined by the
plane of polarization of incident light. When this is the case, the chloronema grows in
such a direction that its long axis is normal to the plane of vibration of the electrical vector
of incident light. Because such responses to polarization of light are limited to rather
special cases, and because their significance in the natural environment is generally
uncertain, they will not be included here.
Photomorphogenesis 275

A photomorphogenic response to the quantity, or fluence rate, of incident light can

be, in the simplest case, a response to its presence or absence. Most notably, a plant
exhibits vastly different morphology depending on whether it is grown in the presence or
absence of light (Section 10.2.2.1). The soybean seedling serves as a representative
example of this difference among dicotyledonous angiosperms, which are those flowering
plants that possess two cotyledons, or seed leaves. The cotyledons are typically used for
storage of food reserves that sustain the seedling until it develops the ability to perform
photosynthesis and become self-sufficient. When grown in total darkness (Fig. lO-2c),
such a seedling has relatively little pigmentation; a long, spindly hypocotyl, which is that
part of the shoot that is below the cotyledons; and small leaves still enclosed within the
cotyledons, together with the apical meristem that will eventually give rise to the mature
shoot. When grown in the light, however, the same seedling will have a shorter hypo-
cotyl, expanded cotyledons and leaves, and considerably more pigmentation (Fig. 1O-2d).
The situation in the monocotyledonous angiosperms, which are flowering plants with a
single cotyledon, is somewhat different. A typical monocotyledonous grass seedling when
grown in darkness (Fig. 1O-2b) also contains relatively little pigmentation. Moreover, its
coleoptile, which is a hollow sheath that covers the growing tip and the primary leaves of
the seedling, remains intact for a considerable period of time and its mesocotyl, which is
that part of its stem below the node from which the coleoptile and leaves of the seedling
arise, is quite long. In the light, however, the mesocotyl is vastly reduced in size and the
bulk of the plant body consists of highly pigmented, expanded leaves (Fig. 1O-2a). More
subtly, an organism can also respond to differences in the fluence rate of incident radia-
tion, as in the case of mustard seedlings that accumulate anthocyanins in amounts that are
proportional to the fluence rate of light that is incident upon them (Section 10.2.2.2).
The capacity to respond to the quality, or wavelength distribution, of incident light
energy is of special importance to photosynthetically active organisms (Sections 10.2.2.3
and 10.5). By doing so, these organisms can improve the efficiency with which they trap
incident radiant energy and make it available for photosynthesis. Some cyanobacteria, for
example, have the ability to produce photosynthetically active pigments that absorb light
maximally in different spectral regions, a process known as photochromic adaptation
(Section 10.5). Presumably, it would be to their advantage to synthesize pigments that
absorb most efficiently the wavelengths of light that are available to them, which is
precisely what they do.
Organisms can also sense the direction from which incident light comes. An example
of a response to this asymmetry in incident light is a leaf that changes its orientation
throughout the day so that its surface is always at a right angle to the direction from which
incident light comes. (1) This "tracking" of the sun, which is observed with plants such as
cotton, soybean, and alfalfa, presumably maximizes the absorbance of photosynthetically
active light by the pigments in the leaf. In addition to movements of this type, an organ
can also move as a consequence of differential rates of cell elongation on the sides nearest
to and furthest away from a source of light. These irreversible growth responses, which
are generally toward (in the case of shoots) or away from (in the case of roots) the source
of light, are known as phototropisms. Because photomovement is dealt with as a separate
topic elsewhere in this volume (Chapter 11), coverage here is restricted.
The final aspect of light to be considered is its periodicity. The rotation of the earth
leads to an ubiquitous diurnal variation in the light environment. Moreover, the length of a
276 Chapter 10

~ coleoptile

\
cotyledons

/
y
mesocotyl

a b

Fig. 10-2. Representative examples of seedlings grown for 10 days at 25°C in the presence or absence of light.
Light-grown (a) and dark-grown (b) maize (Zea mays L.) seedlings. The position of the node from which the
coleoptile and primary leaves arise is indicated with an unlabeled arrow. Dark-grown (c) and light-grown (d)
soybean (Glycine max L.) seedlings.
Photomorphogenesis 277

daily photoperiod varies as a function of both latitude and time of year. Since these
variations are highly reproducible, an organism can use this information as a dependable
environmental cue. For example, plants often respond to defined changes in photoperiod
by flowering, a photomorphogenic response that involves a change from vegetative to
reproductive morphology. (2) Because this aspect of light is dealt with elsewhere (Chapters
7 and 8), discussion of photoperiodism here is minimal.
The ultimate goal in studying photomorphogenesis is to explain how the absorbance
of light leads to the expression of morphogenic responses. To achieve this goal, one must
(1) identify the photoreceptor, (2) determine the initial biochemical event that follows
photoreceptor excitation, and (3) elucidate the subsequent chain of events that leads to the
response. Since the energy content of the light to which an organism responds is insuffi-
cient by itself to bring about a response, the explanation must include a mechanism for
signal amplification. Furthermore, since a single exposure to light typically leads to a vast
array of sometimes seemingly unrelated responses, the explanation must also include one
or more mechanisms whereby the transduction chain branches. Since photoreception itself
and the events that occur most immediately after the absorption of light are of most
general interest, the focus here will be on the identification and characterization of
photomorphogenically active pigments, and on suggestions of how they might function at
the molecular level.
The expression of photomorphogenesis is, of course, modulated by a host of param-
eters other than light. For example, the outcome of a potentially effective exposure to light
may depend on the time at which it is given during a diurnal cycle. Similarly, whether a
given light treatment will result in a given response depends on the ability of the organism
to respond. As emphasized by Mohr, (3) during the development of a young seedling this
potential to respond varies in a predictable fashion. Thus, even though photomorphogene-
sis will be treated here as an isolated topic, the reader should bear in mind that all
responses to light are modified by other factors.
In contrast to the difficulty of identifying photoreceptors and the initial steps leading
to photomorphogenic responses, it is relatively easy to describe the responses themselves.
Even a decade ago, it was possible to reference several hundred photomorphogenic
responses. (4) It is therefore important to subdivide them into appropriate categories rather
than try to deal with them individually. While it would be convenient to group them by
photoreceptor, phytochrome is the only one that has been identified. It is therefore
necessary to categorize other photomorphogenic phenomena in some alternative fashion.
Thus, responses to blue light will be dealt with together, as will be those to ultraviolet
light. It should become apparent, however, that this classification is artificial. Not only
are there extensive interactions among all photomorphogenic systems, but there are proba-
bly mUltiple photoreceptors for each of these spectral regions. Because this approach to
subdividing the subject omits the interesting phenomenon of chromatic adaptation, it will
be discussed separately in Section 10.5. And, while green light can apparently also be
morphogenically active (e.g .• 5) SO little is known about its effects and how it might be
sensed that they will be ignored here.
The diversity of photomorphogenesis requires that it be covered here in cursory
fashion. In particular, the wide range of pigments that might serve as receptors for
photomorphogenesis will not be evaluated independently. Although these pigments,
which include flavins and flavoproteins, carotenoids and carotenoproteins, biliproteins,
278 Chapter 10

cytochromes, and chlorophylls, represent a diverse range of chromophores, it will become

evident from the following discussion that their status as photomorphogenically active
pigments remains, for the most part, uncertain. Phytochrome receives the greatest atten-
tion here only because it is the photomorphogenic system about which most is known.
More extensive coverage of both phytochrome and other photomorphogenic systems is
available elsewhere. (6)

10.2. PHYTOCHROME-MEDIATED MORPHOGENESIS

10.2.1. Discovery
The experiments that led to the discovery of phytochrome were performed at the
U.S. Department of Agriculture laboratories in Beltsville, Maryland.C7- 9 ) H. A. Borth-
wick, S. B. Hendricks, and their colleagues were interested in the question of how
photoperiodically sensitive plants detect day length. They worked with both long-day
plants, which flower in response to nights that are shorter than a critical length, and short-
day plants, which flower in response to nights that exceed a critical length. In the case of a
long-day plant such as barley, interruption of a long-night period by a brief exposure to
light induces flowering that otherwise would not occur. Conversely, in the case of a short-
day plant such as cocklebur, interruption of a long-night period prevents flowering that
otherwise would occur. Action spectra (see Chapter 1 for definition and discussion) for
these seemingly opposed effects of light indicated that red light was most effective (Fig.
10-3). Shortly thereafter, the same investigators obtained similar action spectra for the
inhibition of barley stem elongation, initiation of pea leaf expansion, and promotion of
lettuce seed germination, indicating the existence of a common photomorphogenic system
that controls a wide variety of phenomena. (9)
It was the study of lettuce seed germination, however, that resulted in the critical
observation that led directly to the discovery of phytochrome. Following up on observa-
tions reported in 1935 by Flint and McAlisterC lO) and in 1936 by Meischke,ol) Borthwick
and Hendricks found that not only would far-red light suppress the germination of light-
sensitive lettuce seed, but it would also reverse the promotive effect of red light (Fig.
10-3). Moreover, these antagonistic effects of red and far-red light were repeatedly
reversible through as many as 100 cycles. The final outcome was a function solely ofthe

:
..
t:
Fig. 10-3. Effectiveness of light as a function of wavelength on (a)
'"
~ preventing flower induction in the short -day plant cocklebur (Xan-
o
! \d
£ thium saccharatum Wallr.), (b) promoting germination of light-sen-

~
'" sitive lettuce seed (Lactuca sativa L.), (c) reversing red-light-induced
'"
>
.+= germination of lettuce seed, and Cd) reversing red-light-induced sup-
c
a; pression of flowering in cocklebur (since only three datum points are
0::
available, they are connected by a dotted line). Relative effectiveness
600 700 800 values beyond 690 nm have been multiplied by 10. (Adapted from
Wavelength, nm Ref. 9.)
Photomorphogenesis 279

promote germination of
/ light-sensitive seed
""- • Induce flowering of
/ "" lang-day plants
p. red >- p ___ a Inhibit potassium uptake
r far-red fr ~
"
re d-0bs or b'"Q
f eel b b"
or-r -0 lor mQ /
./
_
Induce anthocyanin synthesis
inactive active ~ Induce gene transcription

Modulate bioelectric potentials

Fig. 10-4. Schematic representation of phytochrome function, showing its photoreversibility between an inac-
tive, red-absorbing form (Pr) and an active, far-red-absorbing form (Ph)' A few examples are given of the
diverse range of photoreversible responses that have been investigated. '

color of light given last. Subsequently, it was learned that other photomorphogenic
responses to red light were similarly photoreversible, reinforcing the notion that a single
photoreceptor was responsible for all of them and that this photoreceptor must exist in two
photointerconvertible forms (Fig. 10-4). One form (Pr ) should absorb red light most
strongly, while the other (Pfr) should absorb far-red light best.
Given the existence of this photoreversible pigment, two possibilities existed. Either
the red-absorbing, Pr form was active, preventing the display of photomorphogenic re-
sponses, or the far-red-absorbing, Pfr form was active, promoting their display. As
Hendricks(9) explains elegantly for dark-grown plant tissues, it must be Pfr that is active
for the following reason. The photoconversion of less than 0.1 % of Pr to Pfr can lead to a
significant biological response. Since this photoconversion represents an insignificant
change in the relative Pr level (100% ~ 99.9%), it is difficult to imagine how the change
could be sensed by the plant. This same change, however, represents a relatively large
increase in Pfr level that is from nothing to a detectable value. Thus, the plant must
respond to the appearance of Pfr' not the disappearance of Pr (Fig. 10-4).
A summary of the characteristics of classical phytochrome-mediated responses, of
which hundreds have been described,<4,6) includes the following points: (1) they are
potentiated by red light; (2) the effects of red light are reversed by far-red light; (3) the red
and far-red light effects are repeatedly reversible; (4) both red and far-red light are
effective at low fluences; and (5) the requirements for both red and far-red light exhibit
reciprocal relationships between fluence rate and time of exposure, such that the outcome
reflects the product of the two parameters.
Consideration of action spectra for the red and far-red light effects (Fig. 10-3) led
Warren Butler, who was then associated with Borthwick and Hendricks, to search for a
blue pigment that would undergo photoreversible changes in its absorbance properties. (7)
Utilizing a newly designed spectrophotometer created for this purpose, Butler and his
colleagues were able to detect such a pigment, which they shortly thereafter named
phytochrome.(12) They were then able to isolate phytochrome and to begin its purification
and characterization. (13)
Phytochrome has since been found in representatives of all major groups within the
plant kingdom, including angiosperms, gymnosperms, ferns, bryophytes, and algae.
While occasional reports have indicated that it may be present in at least some fungi, these
reports remain unconfirmed.
280 Chapter 10

10.2.2. Events under Phytochrome Control

Phytochrome is often associated only with photoreversible responses that exhibit the
characteristics summarized above. It has become evident, however, that plants also
"use" phytochrome in other ways. It is therefore preferable to discuss phytochrome-
mediated morphogenesis with respect to the type of information to which a plant is
responding (Fig. 10-1) rather than with respect to specific morphogenic changes. From
this perspective, phytochrome mediates responses to two aspects of light quantity: its
onset and its fluence rate. Those that occur as a consequence ofthe onset of even dim light
are the classical photoreversible responses that led to its discovery. Those that are propor-
tional to fluence rate are referred to as high-irradiance responses. In addition to these two
functions, phytochrome also provides a mechanism whereby a plant can sense light
quality. While these different sensory functions will be treated independently, it is impor-
tant to note that in some cases they can all contribute to the same photomorphogenic
outcome.

10.2.2.1. Detection of the Onset of Light

Phytochrome can mediate responses to exceedingly low levels of light, as in the
examples of enhancement of coleoptile elongation and suppression of mesocotyl elonga-
tion (Fig. 10-2) in grass seedlings. (14) Responses such as these, which are called very low
irradiance or very low fluence responses, can be to the quantity of light provided by a few
firefly flashes. Many responses to the onset of light, however, require an irradiance that is
104 times greater. Nevertheless, since even these responses are to low fluences, they are
known as low-irradiance or low-fluence responses. All of these responses to the onset of
light have obvious significance at two stages in the life cycle of a plant: (1) at the time of
seed germination and (2) when deetiolation occurs as a seedling emerges from the soil into
the light.
Many seeds remain dormant in the absence of light, as would be the case, for
example, when buried deeply in the soil. Following hydration, only a brief exposure to
light is required to initiate germination, which can then proceed in total darkness. (15) This
simple observation emphasizes an important point. It is the induction of a process, rather
than its expression, that requires light, at least when considering those responses that
occur as a consequence of the onset of light. It is this photosensitivity of many seeds that
gives rise to the gardener's frequent observation that each time the soil is turned over, a
new crop of weeds seems to develop. Teleologically, one can explain this phenomenon as
a survival mechanism. It presumably would prevent the germination of a seed unless it
were near enough to the soil surface to allow the resultant seedling to become photo-
synthetically competent before its food reserves were exhausted, as might happen if the
seed germinated too far below the surface.
A more general photomorphogenic response to the onset of light is the process of
deetiolation. A seedling normally begins life as a heterotrophic (i.e., etiolated) organism,
which means that it cannot manufacture its own organic compounds and must therefore
subsist on stored food reserves in the seed. Upon emerging from the soil into the light,
however, it undergoes a dramatic change to a photoautotrophic mode of existence (Fig.
10-2), which is to say that it develops the capacity to utilize light energy via photo-
Photomorphogenesis 281

synthesis to manufacture from inorganic sources all of the organic compounds that it
requires. All of the changes that accompany this morphogenic transformation can be
grouped under the heading of deetiolation. Among the more prominent of these changes
are the synthesis of a host of new gene products that are required for the development of
photosynthetically competent chloroplasts, (16) as well as those that are required for the
visually obvious morphogenic changes that occur (Fig. 10-2). Many of the morphogenic
features of an etiolated seedling can be explained in terms of their unique value to a
seedling as it emerges from the soil. For example, the hook near the shoot apex of an
etiolated dicotyledonous seedling (Fig. 1O-2c) presumably provides protection for the
sensitive apex, which is still enclosed within the cotyledons, as the shoot pushes up
through the soil. Deetiolation thus involves hook opening, which produces the normal
upright shoot apex that is associated with a green plant (Fig. 1O-2d). Interestingly, in
some instances these photomorphogenic events reflect acceleration of developmental
events that ultimately would occur even in the absence of light. As a result it seems
important that an etiolated seedling be able to modify its response to a given light
exposure as a function of its developmental stage, such that its response at any given time
is most appropriate to its presumed needs. (17) There is considerable evidence that these
changes in the competence of a seedling to respond do take place. (3 ,17)
While the vast majority of photo morphogenic responses to the onset of light fall into
the category of deetiolation, others have been equally well studied. In particular, there is
the photoreversible effect of light given in the middle of a night period on the pho-
toperiodic behavior of a plant. (2,9) This effect of light is in simplest terms interpreted as
the Pfr-dependent division of a night into two parts. If an otherwise photoperiodically
inductive night is divided by a pulse of light into two shorter dark periods, neither of
which is long enough to be perceived by the plant as a "long night," then the inductive
effect of that dark period is lost. In the case of a long-day plant, the outcome is the
induction of flowering; in the case of a short-day plant, the inhibition of flowering.

10.2.2.2. High-Irradiance Responses

Some photomorphogenic responses, known as the high-irradiance responses
(HIR),(l8) are to irradiations given over long periods of time. The magnitude of a HIR is
typically a function of both the fluence rate and duration of the incident light. A reciprocal
relationship between fluence rate and time of exposure is normally not observed. While
phytochrome-mediated responses to the onset of light generally result from considerably
less than one one-hundredth of the energy in all visible wavelengths provided by sunlight
during a I-min period, the HIR require at least 100 times more energy per unit time, given
over a much longer period. Even though the photoreversibility that is typically associated
with phytochrome-mediated morphogenesis (Fig. 10-4) is not observed, there is compel-
ling evidence that these responses are mediated at least in part by phytochrome. (18 ,19)
One of the most productive model systems for the study of the HIR has been light-
induced synthesis of anthocyanins (Fig. 10_5),(20) which are red to blue pigments found
throughout a plant, most obviously in many brightly colored flowers and fruits. The
dependence on fluence rate is well illustrated by this system. As is generally the case for
the HIR, light in both the red/far-red and blue spectral regions is effective. It is well
established that phytochrome is solely responsible for mediating the response to red/far-
282 Chapter 10

~
:g 0.2 -1000

«'"
E
Q)
E
80.1 Fig. 10-5. Time course for anthocyanin accumulation in mustard
(Sinapis alba L.) seedlings as a function of fluence rate during con-

I
c
tinuous irradiation with broad-band far-red light. Onset of irradiation
was at 36 h after sowing. The relative irradiance is indicated next to
c
« 0 :3 each curve; a relative irradiance of 1 is 3.5 mW/m2 • (Adapted from
Time after onset of irradiation, h Ref. 21.)

red light, (19) but the photoreceptor for the blue region has not yet been identified, although
in at least some cases it too may be phytochrome.(20)
The most detailed action spectrum for an HIR is that for inhibition of lettuce hypo-
cotyl elongation (Fig. 10-6). Since those wavelengths that.are most effective are absorbed
well by both Pr and P fr (Section 10.2.3.2), and since the magnitude of a typical HIR is
both fluence rate- and time-dependent, hypotheses to explain how phytochrome mediates
the HIR have generally included a requirement for cycling of the pigment between its two
forms. Thus, the magnitude of a response would be proportional to the extent of cycling
that had occurred. While a variety of hypotheses along this general theme have been
proposed,(l9,2o,23) none has yet been found acceptable in all respects. Moreover, action
spectra for the HIR sometimes indicate a lack of involvement of red/far-red light, and, in
at least some cases, the action spectrum for a given response changes markedly during the
development of a seedling, (24)

10.2.2.3. Sensing of Light Quality

Probably the most pervasive function of phytochrome, even though it has been least
studied, is the ability of a green plant to sense the wavelength distribution of incident
light,(25,26) an observation that derives from early work on seed germination by
Meischke. (11) This function is of most obvious benefit to a green plant that must compete
with other green plants for available photosynthetically active radiation. When sunlight is
filtered through a green-plant canopy, the pigments within the leaves forming the canopy
remove most of the photosynthetically active radiation (Fig. 10-7). In contrast, relatively

Fig. 10-6. Effect oflight via the high-irradiance response (HIR) on

400 600 BOO the inhibition of lettuce (Lactuca sativa L.) hypocotyl elongation.
Wavelength, nm (Adapted from Ref. 22.)
Photo morphogenesis 283

Fig. 10-7. Wavelength distribution of normal daylight (--) and

of shadelight beneath a canopy of broadleaf deciduous vegetation 400 600 800
( ...... ). (Adapted from Ref. 26.) Wavelength, nm

little far-red light is removed, which markedly reduces the ratio of red to far-red light. The
effect of such a change on phytochrome should be to decrease significantly the rate of
phototransformation of Pr to Pfr' while decreasing only slightly the back transformation of
Pfr to Pr (Fig. 10-4). Consequently, at photoequilibrium the steady-state concentration of
P fr' expressed as the ratio of P f/Ptot (Ptot = P r + P fr)' should be reduced. The amount by
which the ratio is reduced should in tum be a function of the density of the canopy. Direct
measurements of Pf/Ptot ratios indicate that this expectation is fulfilled (Table 10-1). In
fact, under a dense sugar beet canopy, the steady-state level of Pfr approaches that produced
by monochromatic far-red light.
The biological significance of these phytochrome-sensed changes in light quality has
been documented largely by the work of H. Smith and his colleagues. (25,26) For example,
they have shown that the appearance of Chenopodium album plants changes dramatically

TABLE 10-1. Proportion of Phytochrome Present as Ptr

at Photoequilibriuma
Lighting R/FRb Pr!P'otC
Midday daylight 0.99-1.17 0.64-0.67
Incandescent 0.71 0.60
Fluorescent 13.5 0.83
Red 20.1 0.88
Far-red 0.002 0.04
Under a wheat canopy 0.21 0.36
Under a sugar beet canopy 0.033-0.045 0.07-0.11
aData from reference 29.
bR/FR is the ratio of incident fluence rates at 660 and 730 nm, respectively.
cPr/Ptot is the ratio of Pfr to total phytochrome at photoequilibrium under the indicated lighting.
The data have been recalculated to reflect a value of 0.88 under monochromatic red light.(33)
284 Chapter 10

Fig. 10-8. Chenopodium album L. seedlings grown in white light as a function of the red to far-red fluence rate
ratio, which decreases from left to right. The proportion of phytochrome that is calculated to be present as Pfr
(lJ!c) decreases from 0.65 to 0.36, as indicated. (Adapted from Ref. 26.)

as a function of the ratio of red to far-red light that is incident upon them (Fig. 10-8). All
of the changes are those that would be predicted to help the plants compete for available
photosynthetically active radiation. Moreover, recent work(27,28) has shown that plants
respond in similar fashion to a change in light quality that occurs as the result of nearby,
neighbor plants. With phytochrome, a plant can thus determine not only whether it is in
the shade of another plant, but whether it is growing at high or low plant density. The
most obvious change in both situations is an increase in the rate of internode elongation,
which presumably would help the plant outgrow its competitors in order to reach open
sunlight. While only some species respond in this way to a changing red-to-far-red ratio, it
nevertheless seems likely that this function of phytochrome is pervasive, perhaps being
manifested in other ways in other species. For example, in order to germinate, many light-
sensitive seeds require a level of Pfr that is higher than that produced under'a dense green
canopy. These seed would germinate only when the canopy is eliminated, as happens
naturally under a deciduous plant canopy in the autumn or after a fire. Thus, seedlings
would develop only when sufficient photosynthetically active radiation would be available
to sustain their growth. (15,30)
Photomorphogenesis 285

10.2.3. Properties of Phytochrome

10.2.3.1. Physicochemical Properties
Characterization of phytochrome as a molecule has until recently been restricted to
the chromoprotein as it is isolated from etiolated plants.(31,32) There are two reasons for
this situation. First, green plants have 10- to 100-fold less phytochrome than etiolated
plants of the same genotype. Second, the abundance of chlorophyll in a green plant
prevents the spectral assay of phytochrome in them. Unfortunately, the characterization of
phytochrome even from etiolated plants has been difficult because of its lability to pro-
teases in crude extracts. Thus, the first report concerning purification of phytochrome to
homogeneity indicated that its monomer size was 60 kDa, whereas subsequent reports
indicated a size of about 118 kDa. It is now evident, however, that full-size phytochrome,
at least from oats, is 124 kDa in size and that earlier reports described partially degraded
molecules. (31 ,32)
Full-size phytochrome is now understood to be a dimer of two very similar, if not
identical, monomers, each bearing a linear tetrapyrrole chromophore joined to the protein
moiety via a thioether linkage (Fig. 10-9). It is the variable interaction between the Pr and

Protein
I
S

A
Photoconversions I I
I
I
I
Red~ I Far-red
I

Protein

Fig. 10-9. Proposed structures for the phytochrome chro-

mophore in the Pr and Pfr forms, indicating its thioether linkage
to the protein moiety. R
286 Chapter 10

"g
o
-e
..,5l
<t
Fig. 10-10. Absorbance spectra for highly purified oat
phytochrome after saturating far-red (--) and red
(- - ) irradiation. A' calculated spectrum for a
solution of P fr equimolar to the PT solution obtained by
irradiation with far-red light is also shown (----). The
Pfr/Ptot ratio that is established at different wavelengths
( ...... ) is calculated from the absorbance spectra that are
Wavelength, nm shown. (Adapted from Ref. 33.)

Pfr forms of the protein moiety with the two forms of the chromophore that give rise to
their different absorption spectra (Fig. 10-10). The absorption bands in the red/far-red and
the blue regions are due to the chromophore, whereas that at 280 nm arises primarily from
the protein moiety. Apart from the presence of a chromophore and about one phosphate
residue per monomer, there is no indication that the protein moiety undergoes posttransla-
tional modification. The amino acid composition of phytochrome is typical of a water-
soluble protein, indicating that it does not function as an integral membrane protein. This
assumption is reinforced by recent sequence data, which indicate that the protein moiety
does not possess a hydrophobic domain large enough to permit its insertion into the lipid
core of a membraneP4)

10.2.3.2. Photochemical Reactions

Both photoconversion pathways, Pr - ? Pfr and Pfr - ? Pr, occur via several intermedi-
ates (Fig. 10-11). Following the initial excitation of either Pr or Pfr' the first intermediate
that is produced in each case decays to the opposite form of phytochrome via different sets
of thermally unstable intermediates (I,.'s and If;s). The slowest reactions are those leading
to Pr, which occur on a millisecond-to-second time scale. Since the thermal decay of
intermediates represents the rate-limiting step in phototransformation, appreciable levels
of phytochrome can be present in intermediate form at steady state under bright illumina-

inhibition Fig. 10-11. The phytochrome system. Synthesis ofphy-

~
Y
Synt~siS Ir's"-..\ /
biological
activity
tochrome, which occurs as Pro is inhibited by Pfr' The
photoconversion pathways between Pr and Pfr involve
several intermediates (Ir's and Ifr's). Pfr, which yields
• Pr ,..rPfr biological activity by an as yet unknown mechanism, dis-
\ ~~ I•• ("hll / "-.. destruction appears from the cell via both thermal reversion to Pr and
enbanced proteolytic degradation of its protein moiety, a
reversion process that is referred to as destruction.
Photomorphogenesis 287

tion. (32,35) In principle, therefore, the possibility exists that under certain circumstances,
such as in the case of the HIR, a phototransformation intermediate may have biological
activity.
A second feature of the photochemical reactions of phytochrome that has attracted
considerable interest is the proportion of phytochrome present as Pfr (Pf/Ptot) at photo-
equilibrium as a function of wavelength (Fig. 10-10). Except in the far-red region, where
Pr absorbs negligibly, the two forms of phytochrome have overlapping spectra. Thus,
while it is possible to obtain essentially pure Pr' it is impossible to obtain pure Pfr . Instead,
under conditions of constant illumination, phytochrome will be cycling continuously
between its two forms. The proportion that is present as Pfr is then strongly wavelength-
dependent (Fig. 10-10; see Sections 10.2.2.2 and 10.2.2.3). It is these considerations that
provide a sensible explanation for the ability of phytochrome to sense light quality, as
discussed in Section 10.2.2.3.

10.2.3.3. Nonphotochemical Reactions

Phytochrome is involved in four important, nonphotochemical processes. These
include (1) de novo synthesis of Pr, (2) destruction, preferentially of Pfr' (3) nonphoto-
chemical reversion of Pfr back to Pr, and (4) the as yet unknown reaction(s) that lead to
biological activity (Fig. 10-11). Because of its importance, this fourth aspect will be dealt
with separately (Section 10.2.4).
Phytochrome is synthesized in darkness as Pr and accumulates until a plateau is
reached. This plateau presumably corresponds to a balance between synthesis and turn-
over of the chromoprotein. Interestingly, one of the most rapid Pfr-mediated responses is
an inhibition of Pr synthesis. (36) A 5-s irradiation with red light induces within 15 min a
measurable decline in the level of phytochrome mRNA. Within 5 h of a pulse of red light,
about 80% of the phytochrome mRNA is gone. Restriction map polymorphisms, together
with partial sequence data, indicates that there are at least four different genes expressing
phytochrome mRNA in etiolated oat seedlings. (34) It is not known, however, whether
these genes are expressed differentially, either with respect to tissue or to age. If either
were the case, the physiological implications could be considerable.
The photoconversion of Pr to Pfr in situ leads to a rapid decline in the level of
spectrophotometric ally detectable phytochrome; this decline is commonly referred to as
destruction (Fig. 10-11).(37) It was subsequently learned that destruction also results in a
loss of immunochemically detectable phytochrome. The process is apparently en-
zymatically mediated, exhibiting an appropriate temperature dependence and requiring an
input of free energy. It presumably results from a targeted, proteolytic degradation of
phytochrome, although the precise nature of this process has not been elucidated. It has
recently been suggested that phytochrome is targeted for destruction by ubiquitin,(38) a
ubiquitous protein that in other instances is known to serve this function. However, this
potential role of ubiquitin in phytochrome destruction remains to be firmly established.
Furthermore, while it has often been said that destruction is specific for Pfr' this is not the
case. In some instances, when light conditions are such that Pr is first converted to Pfr and
then back to Pr, this cycling of the pigment leads to a limited, but comparably enhanced,
rate of destruction of Pro (37) It is as though, in some cases, phytochrome molecules that
had been targeted for destruction as Pfr remain marked for destruction even after being
288 Chapter 10

converted back to Pr, such that Pr will also undergo limited destruction. Together with the
Pfr-induced decrease in phytochrome mRNA level, it is this Pfr-initiated destruction of
phytochrome that results in such low levels of this pigment in light-grown plants.
Thermal reversion of Pfr to Pr, which generally occurs over a span of hours and
which was once widely implicated as a timing function to explain the role of phytochrome
in photoperiodism, (9) is a puzzling event. It has been described for phytochrome in vivo in
etiolated dicotyledonous seedlings, but then involves only a limited pool of the pigment. It
has never been observed in vivo in etiolated grass seedlings, nor, apart from one excep-
tion, has phytochrome isolated from etiolated grass seedlings been found to exhibit
reversion in vitro, unless the protein has been partially degraded. (32) The exception
derives from the observation that a monoclonal antibody, when bound to a domain on
phytochrome that is apparently about 50 amino acids from its amino terminus, (39) induces
thermal reversion of Pfr to Pr' This domain had previously been shown to be important for
the maintenance of phytochrome in a spectrally native Pfr form. (32) This observation leads
to the speculation that plant cells might regulate thermal reversion, and thereby control the
level of the active Pfr form of phytochrome during the night, by controlling the availability
of a ligand that interacts with phytochrome at this location. Apart from the observation
that only a limited fraction of the phytochrome in dicotyledonous cells undergoes thermal
reversion, however, there are no independent data to support this speculation. At present,
it is simply impossible to assess the significance, if any, of this process. Nevertheless, it is
evident that thermal reversion, as well as phytochrome destruction, provides a mechanism
whereby a cell can rid itself of the active form of phytochrome during the night.

10.2.3.4. Differences between Pr and P fr

Since Pfr is active and Pr is not (Fig. 10-4), considerable attention has been given to
the question of how the two forms of phytochrome differ from one another. Such dif-
ferences might provide a clue as to how Pfr functions.(31,32,37) Unfortunately, in spite of
the effort expended on answering this question, little of substance has been learned. The
bulk of the evidence indicates that there is no gross conformational difference between Pr
and Pfr' although recent data do indicate a slightly different behavior by size exclusion
chromatography, with Pfr exhibiting a slightly greater hydrated radius than Pr .(32) While
many discrete differences between the two forms have been reported, they have for the
most part not been related to the structure of the chromoprotein. The recent production in
several laboratories of domain-specific monoclonal antibodies to phytochrome should
facilitate identification of differences between P rand Pfro Already a limited number of
monoclonal antibodies have been identified that recognize domains that change upon
phototransformation.(32,40) For example, three antibodies that bind several fold more
tightly to Pr than to Pfr have been found. (39) These antibodies bind to a domain that is
apparently about 50 amino acids from the amino terminus of phytochrome, a region that
previously had been shown to be important for proper interaction between the protein and
chromophore moieties of this pigment. (32,40) Interestingly, it was one of these antibodies
that was also found to facilitate reversion of Pfr to Pr, as mentioned above. Much more
will have to be learned, however, before even this information will assist us in under-
standing how Pfr functions (e.g., see Section 10.6).
Two other significant differences between Pr and P rr have been described. While Pr
Photomorphogenesis 289

behaves as a soluble protein in crude plant extracts, P fr associates preferentially with

particulate, subcellular debris. (41-43) It has been suggested that this enhanced association
of Pfr with sedimentable subcellular material, which is termed enhanced pelletability,
results from an interaction of phytochrome with membranes. (44) There is, however, little
hard evidence to support this conclusion. (42,43) Independently, immunocytochemical ob-
servations have led to the conclusion that while Pr is uniformly distributed throughout the
cytosol, consistent with its presence as a soluble protein, P fr is associated with discrete
subcellular regions. (37) This redistribution of phytochrome after photoconversion to P fr'
which has been called sequestering, has also led to the suggestion that it results from the
binding of Pfr to a membrane receptor. Again, there is no hard evidence to support this
suggestion. Instead, recent electron microscope observations of immunogold-labeled phy-
tochrome indicated that both sequestering and enhanced pelletability of P fr involve an
association of phytochrome with structures that appear not to be bounded by a mem-
brane. (45,46) Even though there seems to be no detectable association between Pfr and a
membrane, both sequestering and enhanced pelletability occur in oats within 5-10 s after
the appearance of P fr within the cell, which is sufficiently rapid to leave open the pos-
sibility that they result from an initial step in the mode of action of phytochrome.

10.2.3.5. Phytochrome from Green Plants

The development of sensitive immunochemical assays for phytochrome which elimi-
nate interference by chlorophyll has recently made it practical to begin the physicochemi-
cal characterization of phytochrome from green-plant tissue. Although indirect evidence
had led to the suggestion that phytochrome in green plants would be different from that in
etiolated plants, the recent discovery that phytochrome from green oats was both spec-
trally and immunochemically distinct from phytochrome from etiolated oats was neverthe-
less somewhat surprising.(47,48) With respect to spectral properties, Pfr from green oats is
similar to that from etiolated oats, whereas the absorbance maximum for Pr is shifted
about 10 nm to shorter wavelengths. With respect to immunochemical properties, the
majority of monoclonal antibodies to phytochrome from etiolated oats fail to recognize
phytochrome from green oats(48,49) and vice versa. (50) Moreover, phytochrome from
green oats is not only different from that from etiolated oats, but preliminary indications
are that it may itself consist of multiple populations. (40,48,49) Even though phytochrome
from green oats is about the same size as that from etiolated oats; they yield different
peptide maps upon proteolysis, reinforcing the conclusion that they are different pro-
teins.(47,49) Whether these two proteins are products of different genes or are differen-
tially modified products of the same gene is presently a matter of intensive investigation.
The significance of two kinds of phytochrome in the same plant is unknown. While it
is evident that one of the two, namely, that which has previously been well characterized,
is only abundant in etiolated tissue, tools with which to determine how the abundance of
the other might be regulated are only now becoming available. (50) Initial evidence does
indicate that the levels of the two phytochromes are differentially regulated. (51) It may
even be that these two phytochromes have different biological activities. It would be
interesting to determine, for example, whether the chromoprotein that is abundant in
etiolated oats is responsible for detecting the onset of light (Section 10.2.2.1) whereas the
chromoprotein detected in green oats is responsible for sensing light quality (Section
290 Chapter 10

10.2.2.3). In addition, it will be important to detennine whether the observations made

with oats can be generalized. Initial work with peas indicates that they can be. (48,52)

10.2.4. Primary Mode of Action

It has long been evident that Pfr is the active fonn of phytochrome.(9) While this
conclusion has occasionally been challenged, the challenges have never withstood scru-
tiny. It has also been proposed that there are multiple primary reactions for Pfro (53) While
this hypothesis has merit, there is at present no compelling reason to accept it. In some
respects, it would seem more prudent to begin with the simpler assumption, at least with
respect to phytochrome itself, that it has a single primary function. The mUltiplicity of
events controlled by phytochrome would then arise by branching in the transduction chain
between Pfr and the responses, after the initial function of Pfr has been expressed (Fig.
10-4). Alternatively, however, it is conceivable that there are mUltiple pools of phy-
tochrome (e.g., Section 10.2.3.5) and that, while each pool exerts only a single primary
function, that function is different for each pool. Irrespective of these considerations, the
key question to be asked is, what is it that Pfr does that P r does not or cannot do? The
answer to this question must incorporate answers to other, more specific questions as well:
(1) How is the signal that is detected by phytochrome amplified? (2) How does the
primary event involving Pfr result in such a vast array of seemingly unrelated responses?
(3) How is it that the fonnation of P fr within a cell gives rise to responses that are a
function of both the tissue and the morphological age of the cell in which they occur?
Three major hypotheses concerning the primary action of Pfr have been proposed.
While each of these three hypotheses has its unique merits, the following discussion is
intended to make clear that none has been proven correct.
The first hypothesis derives directly from the observation that phytochrome is a
chromoprotein. Since proteins are often enzymes, it seemed sensible at the outset to
propose that P fr exhibits an enzyme activity that Pr lacks.(9) No such activity, however,
has yet been described for Pfro While interest in this hypothesis has not been very great,
there is presently no good reason to discard it. Indeed, the recent evidence that highly
purified phytochrome is associated with a protein kinase activity(54) is rekindling interest
in this hypothesis. Nevertheless, as the authors of this work note, it remains to be
detennined whether the kinase activity belongs to phytochrome or an otherwise un-
detected contaminant.
The origin of the second hypothesis coincided with the rapid increase in knowledge
concerning the differential expression of genetic infonnation. With respect to phy-
tochrome, for a given cell at a given developmental stage there should be four components
of the genome(53): (1) genes that are induced by Pfr' (2) genes that are repressed by Pfr, (3)
genes that are never expressed, irrespective of the fonn of phytochrome, and (4) genes
that are expressed independently of phytochrome control. It has become abundantly clear
that phytochrome does influence gene expression as hypothesized, (55) but whether it does
so by interacting directly with the genome or indirectly by means of a second messenger
has not been established. The existence of promoter regions responsible for phytochrome-
mediated gene expression has been established, (55,56) but direct control of such promoters
by phytochrome has not been demonstrated. While it is conceivable that phytochrome
does regulate gene expression directly, such a primary function for phytochrome cannot
Photomorphogenesis 291

account for the most rapid phytochrome-mediated events that have been described (see
below), nor has there been any verifiable demonstration that phytochrome exists in the
nucleus.
The third hypothesis, for which experimental support is the strongest, arose at-a time
when interest in, and knowledge of, the organizational and functional complexity of
membranes was increasing rapidly. Logically, one should focus on those events that occur
most rapidly after the appearance of Pfr if one wants to learn about its primary function.
By doing so, one is almost invariably led to the study of phytochrome-mediated responses
that indicate an early effect of Pfr on membrane properties.(44,57) For example, trans-
membrane ion flux changes, notably of K + , begin within at most a few minutes after the
formation of Pfro Effects of phytochrome on bioelectric potentials are even more rapid. In
one of the most elegant of such studies, Newman(58) demonstrated that Pfr initiates an
electric potential change in the etiolated oat coleoptile beginning only 4.5 s after its
appearance within the cell. One is thus led to the hypothesis that phytochrome functions in
a primary sense by altering one or more membrane properties or functions.(57) Unless
phytochrome has different primary activities, it is evident that it cannot be directly
regulating the expression of genetic information.
Numerous variants of the membrane hypothesis have been suggested. It has been
proposed that phytochrome might function as an integral membrane protein, although its
biochemical properties (Section 10.2.3.1) are inconsistent with this possibility. Alter-
natively, it has been suggested that phytochrome might interact with a membrane by
binding to a receptor on its surface and that this interaction is specific for the Pfr form.
Superficially, both red-light-enhanced phytochrome pelletability and immunocytochemi-
cally detected sequestering of Pfr appear to be consistent with this possibility.(44) Never-
theless, as already discussed (Section 10.2.3.4), neither of these phenomena is likely to
involve phytochrome binding to a membrane receptor. A third variant, which at present
would seem to be the most reasonable, is that the effect of Pfr on membranes may be
mediated via a second messenger such as Ca2 + . As yet, however, there are no compelling
data to indicate that such a second messenger is, in fact, obligatorily involved in phy-
tochrome function.
Thus, while the hypothesis that phytochrome exerts its primary effect at the mem-
brane level is presently the most acceptable of the three that have been considered, even
this hypothesis does not enjoy overwhelming experimental support. It is at least equally
possible that Pfr functions as an enzyme, as originally proposed, or in some as yet
unimagined fashion.

10.3. BLUE-LIGHT-MEDIATED MORPHOGENESIS

Since blue light is absorbed by phytochrome, it should not be surprising that phy-
tochrome can serve as a photoreceptor for blue-light-mediated morphogenesis. Neverthe-
less, there are blue-light-absorbing photoreceptors quite independent of phytochrome.
Morphogenic phenomena that are mediated specifically by blue light differ in a few
important aspects from those mediated by phytochrome: (1) They are induced most
effectively by blue and, in many instances, near-UV radiation, but not by light in other
spectral regions, notably the red. (2) Whereas mediation by phytochrome of blue-light
292 Chapter 10

effects requires relatively high fluence rates of blue light, mediation by a specific blue-
light photoreceptor often, but not always, involves much lower incident energies. (3)
Unlike many phytochrome-mediated events, responses to low fluence rate blue light are
not photoreversible.
No chemically defined blue-light photoreceptor has yet been identified. At least three
reasons can be given for this failure: (1) The discovery of phytochrome was facilitated
greatly by the fact that it is a reversibly photochromic pigment, thus permitting the
development of a sensitive spectrophotometric assay for its detection in vitro.(7,13) Con-
versely, there is no firm evidence that any blue-light photoreceptor is comparably pho-
tochromic, which means that development of an assay for its detection in vitro will be
difficult at best. (2) A typical cell contains an abundance of blue/near-UV-absorbing
pigments. A search for a physiologically significant blue-light photoreceptor is thus
equivalent to looking for a needle in a haystack. (3) There are likely to be many blue-light
photoreceptors, (59) perhaps even within the same organism and involved in the same
response. (60) It is for this last reason that the term cryptochrome(61) is not used here.
Inherently, this term implies that there exists a single, chemically definable pigment that
is responsible for blue-light-mediated morphogenesis. Since this almost certainly is not
the case, more generic terminology will be used to minimize the possibility that a reader
will come to the conclusion that a single pigment is responsible for the wide range of blue-
light-mediated responses that have been described.

10.3.1. Events under Blue-Light Control(62)

The number of morphogenic responses to blue light that have been characterized is
considerably more limited than the number known to be mediated by phytochrome.
Conversely, the taxonomic distribution of blue-light photoreceptors is more extensive. As
is true for phytochrome, blue-light photoreceptors are found in angiosperms, gym-
nosperms, and other nonflowering plants, including green algae. Unlike phytochrome,
however, they are also widespread among the fungi. In part because of the absence of a
single well-defined photoreceptor, it is difficult to provide a uniform description of a
typical blue-light photoresponse. In general, such a photoresponse is insensitive to wave-
lengths of light longer than the blue, and it typically exhibits sensitivity to near-UV
radiation as well. Given this rather loose definition, the following are selected examples
of blue-light-mediated morphogenesis.
One of the best studied blue-light responses is the phototropic curvature of plant
shoots toward the light (Chapter 11). Positive phototropism of the coleoptile, which is an
organ that is best developed in young grass seedlings grown in total darkness (Fig. 1O-2b) ,
has long served as a model system. (63)
A second growth response of shoots to blue light, which appears to be distinct from
that leading to phototropism, is the blue-light-mediated inhibition of stem elongation, best
described for the etiolated cucumber hypocotyl. (64) In this instance, blue light of relatively
high fluence rate initiates a decrease in growth rate that begins after a lag of as little as 20-
30 s. This reduced growth rate persists as long as the blue light is on (Fig. 10-12). The
fluence rate required is roughly 1000-fold greater than that required to induce a pho-
totropic curvature of the grass coleoptile. Since the response begins after such a brief lag,
Photomorphogenesis 293

Fig. 10-12. Growth response of the cucumber (Cucumis sativus L.)

stem to irradiation with blue light (5 W/m2). A rapid decrease in
growth rate is seen beginning within less than I min after the onset of
light ( t ). The growth rate returns to the dark level within several
O~O----L-~6~O~~--~1~20~
minutes after the blue light is turned off ( i). (Adapted from Ref.
64.) Time, min

it is evident that it is relatively closely coupled to photoreceptor excitation, although the

photoreceptor and the nature of the transduction chain itself are completely unknown.
Simple shading experiments indicate that the photoreceptor is in the region of the hypo-
cotyl that exhibits the response, indicating that there may be no need for signal transmis-
sion in this instance as there is, say, in at least some cases of phototropism. (63)
The study of blue-light-induced pigment synthesis has also provided considerable
information about blue-light-mediated morphogenesis.(62,65) Accumulation of both carot-
enoids and anthocyanins are at least in part dependent on blue light. Light-induced
anthocyanin synthesis is an especially good example of the interactions that can occur
among different photoreceptor systems. The extent of blue-light-mediated anthocyanin
synthesis is itself regulated in a photoreversible fashion by phytochrome. If red light is
given after the termination of a prolonged exposure to a relatively high fluence rate of blue
light, such that the level of Pfr is high, then the quantity of anthocyanin that is accumulat-
ed is high. In contrast, if the blue-light exposure is terminated by far-red light such that the
level of Pfr is low, then little anthocyanin is accumulated. If only blue light is provided,
the response is then a function of the amount of Pfr that happens to be present when the
exposure to blue light is terminated. Since red light given in the absence of a preceding
exposure to blue light is by itself relatively ineffective, it is evident that a blue-light
photoreceptor is in some way interacting synergistically with phytochrome to produce the
final response.
Not all interactions are synergistic. For example, red light that is absorbed by
phytochrome reduces by a factor of 10 the sensitivity of the maize coleoptile to pho-
totropically active blue light, even though the red light itself is phototropically inactive. (66)
Other well-studied blue-light-mediated responses include (1) induction of a transition
from one-dimensional to two-dimensional growth of the fern protonema, which is an
early, filamentous growth stage of the haploid generation of this group of plants, (2)
formation in certain fungi of fruiting bodies known as perithecia, (3) regulation in higher
plants of stomatal apertures, which are variably sized pores in the epidermis of leaves
through which gas exchange occurs, and (4) changes in membrane potentials of cells near
the apex of young bean shootS.(59,61,62) It is therefore evident that blue light induces a
diverse range of morphogenic responses.
294 Chapter 10

10.3.2. Photoreceptor(s)
As already noted, it is evident that not all blue-light-mediated morphogenesis in-
volves the same photoreceptor. Nevertheless, it seems reasonable that most blue-light
photoreceptors might belong to the same pigment category.
Action spectra for typical blue-light-mediated responses, such as phototropism of the
oat coleoptile and carotenogenesis in the fungus Fusarium (Fig. 10-13), indicate that the
photoreceptor in each case has three absorbance bands in the blue region of the spectrum,
in addition to a band in the near-UV region. What pigments are candidates for such a
photoreceptor? Two possibilities have been given serious consideration: carotenoids,
which have a typical three-peaked absorbance spectrum in the blue region, and flavins,
which exhibit absorbance in both the blue and near-UV regions. Carotenoids typically do
not have absorbance in both the blue and the near-UV regions, nor do flavins usually
exhibit the fine structure in the blue that is seen in many action spectra. Examples of each
group that possess all requisite absorbance properties to serve as a blue-light photorecep-
tor do exist, however. Given these two distinctly different possibilities, it is perhaps only
natural that there has been considerable controversy over the question of whether blue-
light photoreceptors are flavins or carotenoids.
A consensus of opinion now suggests, however, that blue-light photoreceptors are
most likely flavins in association with protein, that is to say flavoproteins. (59,61,62) For
example, flavins are well known as photosensitizers (Chapter 3). In addition, chemicals
that interact with photoexcited flavins, such as phenylacetic acid, inhibit specifically one
of the classic blue-light-mediated events, namely, phototropism.(69) Moreover, "carot-
enoid-deficient" mutants have recently been found to exhibit a phototropic response to
blue light, albeit with slightly decreased photosensitivity. Thus, while a blue-light pho-
toreceptor has not yet been identified, it appears that the search should focus on flavopro-
teins as likely candidates. (59)
One approach to the question of how to identify a blue-light photoreceptor is poten-
tially fruitful. In 1970, Berns and Vaughn(70) proposed that one consequence of light
absorption by a blue-light photoreceptor might be a change in absorbance of either the
blue-light photoreceptor itself or another, closely associated pigment. This light-induced
absorbance change (LIAC) might then serve as the basis of an assay for the isolation,
purification, and characterization of a blue-light photoreceptor. In order for such a LIAC

'"'"
'"c:
'">
n
~
'"
~
+=
o
a;
0::
Fig. 10-13. Action spectra for first positive phototropic
curvature of the oat (A vena sativa L.) coleoptile (--)
and for carotenogenesis in the fungus Fusarium ( ...... ).
Wavelength, nm (Adapted from Refs. 67 and 68.)
Photomorphogenesis 295

Fig. 10-14. Difference spectrum (--)

showing a blue-light-induced absorbance
Q)
change (LIAC) in a membrane-enriched <.>

fraction from cauliflower (Brassica

"'................ c
o
..c
o
~.

oleracea L.). The LIAC was obtained by \ '"

..c
subtracting an absorbance spectrum mea- <i
sured in the absence of any irradiation ,
from one measured after irradiation for 2
min with blue light at a fluence rate of 15
:
en
..
,' ,
____ Jl'~ " ... --,
IT
" ' , I
W/m 2 . Both spectra were measured at CD I \
55 I ,
17K. Part of the LIAC is expanded to \
1
.~,'
show greater detail ( ...... ). Calibrations ti
:t / \
I \

in absorbance units for the original (--) Q) / \

CD \ 0.002 0.01
and the expanded ( ...... ) difference spec- > \
trum are given at the lower right. The '"5 '\.
action spectrum that gives rise to the
£ ,~

LIAC is shown for comparison (---- ). 400 500 600

(Adapted from Ref. 72.) Wavelength, nm

to serve as an assay for a potential blue-light photoreceptor, it should have an action

spectrum like that for blue-light-mediated photomorphogenic responses (Fig. 10-13).
Since the LIAC need not result from a change in absorbance of the blue-light photorecep-
tor itself, however, it could be displayed at any wavelength.
LIACs with these appropriate characteristics have been identified from both fungal
and higher plant sources. (71) The action spectra are similar to those for typical blue-light-
mediated morphogenic responses, while the LIACs themselves are indicative of the
reduction of a b-type cytochrome (Fig. 10-14). Isolation and partial purification of sub-
cellular fractions that exhibit a blue-light-induced absorbance change indicate that all
necessary components are membrane-associated and that the membrane with which they
are associated is most likely the plasma membrane. Purified, putative plasma membrane
fractions exhibiting the LIAC have been shown to possess flavin by both fluorescence
excitation and emission assay. It thus seems apparent that the LIAC is a flavin-mediated
photoreduction of a b-type cytochrome.(59,71,72)
The question that remains is whether this LIAC derives from the activity of a
biologically relevant blue-light photoreceptor. Unfortunately, this question is exceedingly
difficult to answer unequivocally. Given that commercially obtained horse heart cyto-
chrome c and either flavin mononucleotide or riboflavin can exhibit a comparable LIAC
under wholly artificial conditions, (73) it is possible to suggest that the naturally isolated
LIACs might be biologically irrelevant. Alternatively, it is evident that even in purified
membrane fractions, it is a specific b-type cytochrome that is photoreduced. The bulk of
the cytochromes that are present are not photoreduced. Moreover, a Neurospora mutant
that is deficient in cytochrome b is more than 100-fold less sensitive than the wild-type
strain to blue light, which in this instance induces conidiation. Similarly, methylene blue
sensitizes several biological systems to red light, such that action spectra for typical blue-
light-mediated events are extended artificially to include red light that is absorbed by the
methylene blue. Of interest is the correlated observation that this photosensitizer also
facilitates the production in vitro of a LIAC in response to red light and that the LIAC is
296 Chapter 10

specific for the same b-type cytochrome that is photoreduced by the endogenous system.
It is evident, therefore, that there is considerable, albeit indirect, evidence to support the
hypothesis that the LIACs that have been described might serve as an assay for a blue-
light photoreceptor, eventually allowing its complete purification and chemical identifica-
tion. (71) Nevertheless, it is important to bear in mind that the evidence is far from
conclusive. Whereas the indefinite photoreversibility of phytochrome, which is also ex-
pressed as a LIAC, provides a ready assay for its identification, the LIAC in question here
cannot be associated so unambiguously with a biologically relevant blue-light photorecep-
tor.

10.4. UV RADIATION-MEDIATED MORPHOGENESIS

That part of the UV spectrum that is significant with respect to photomorphogenesis

is from about 300 to 390 nm. Wavelengths between 320 and 390 nm constitute the UV-A
region and, as we have already seen, can be active via a blue-light photoreceptor and
phytochrome, as well as via a separate UV radiation photoreceptor. The region from 280
to 320 nm is known as the UV-B region. This region is of special interest because of its
potentially harmful effects on both plants and animals (Chapters 5 and 6). Normally, UV-
B radiation is for the most part selectively removed from solar radiation by ozone present
in the stratosphere. Recent indications that human activities are resulting in a depletion in
the amount of ozone in this layer, which would lead to increased UV-B radiation reaching
the surface of our planet, have generated increased interest in the effects of UV-B
radiation on biological systems.
As in the case of blue-light-mediated morphogenesis, there appear to be mUltiple
UV-absorbing photoreceptors that control a variety of morphogenic events.C7 4 ) Since
specific photomorphogenic responses to UV -A radiation are unknown in higher plants, (74)
the emphasis here will be on the effects of UV -B radiation. Unfortunately, since there is
no good candidate for a specific UV-B radiation photoreceptor, it is difficult to provide a
focus for even this limited discussion of UV radiation-mediated morphogenesis. As in the
case of blue-light-mediated morphogenesis, it will only be possible to indicate with a few
examples the wide range of responses to UV radiation that have been described and to
consider the possible nature of what is perhaps the best described UV radiation pho-
toreceptor system, which has been termed mycochrome since it is present uniquely in the
fungi. The extensively characterized photochemical effects of UV radiation, especially on
nucleic acids, and the photorepair of this damage, both of which are perhaps best excluded
from a discussion of photomorphogenesis in any case, are covered in Chapter 4.

10.4.1. Events under UV Radiation Control

Although UV radiation-mediated morphogenesis is best known among the fungi,
UV-B radiation does have significant morphogenic effects on higher plants. It not only
interferes with the photosynthetic activity of isolated chloroplasts, (75) but it also decreases
the photosynthetic activity of intact plants. (76) Furthermore, UV -B radiation leads to a
general decrease in plant growth, including decreases in leaf area and overall dry weight,
as well as enhanced flavonoid synthesis. (74) Since the newly synthesized flavonoids are
Photomorphogenesis 297

located in outer cell layers, and since they absorb solar UV-B radiation strongly, it has
been speculated that their function is to protect underlying tissues from the harmful effects
of UV-B radiation, such as impairment of the photosynthetic machinery.(74)
In contrast to UV radiation-induced changes in higher plants, the mycochrome
system in fungi has been better defined, although it is still far from being understood. (77)
At a particular stage of the development of spore-producing structures known as con-
idiophores, UV radiation and blue light have opposing effects on the development of
spores, referred to in this instance as conidia. If UV radiation and blue light are given in
alternating fashion, conidia develop if the last exposure is to UV radiation, whereas they
do not develop if blue light is given last. This photoreversibility is reminiscent of the
red/far-red photoreversibility seen with phytochrome. Nevertheless, the UV /blue effects
do not result from the photointerconversion of a single pigment between two forms, (77) as
is the case with phytochrome.

10.4.2. Photoreceptor(s)
There are serious practical difficulties associated with obtaining action spectra in the
UV region. Both the relatively low fluence rates that are available from monochromatic
sources and the massive screening that results from the absorption of UV radiation by
plant tissues interfere significantly. For these and other reasons,(74) very few detailed
action spectra have been obtained. Consequently, it has not even been possible in most
instances to predict with any precision the absorbance properties of potential photorecep-
tors.
The mycochrome system has been the one that is most amenable to analysis. (77) An
action spectrum for the induction of conidia formation exhibits a single, reasonably well-
defined maximum at about 300 nm, with a shoulder at about 320 nm (Fig. 10-15). The
reversion of this UV -B radiation effect by blue light exhibits maximum activity near 450
nm, with a shoulder or subsidiary peak on either side and with a second peak of activity at
about 380 nm. If both the UV-Bradiation and blue-light effects were mediated by a single
photoreversible pigment analogous to phytochrome, it might have been relatively easy to
use this photoreversibility as an assay to permit its isolation and purification. As already
noted, however, it seems evident that separate pigments are responsible for the opposing
effects of light. The chemical nature of these two pigments is completely unknown. The
obvious similarity between the action spectra for the effect of blue light in the my-

1.0

"'"'c:
Q)

Q)
.~
"0
Q)

~ 0.5
Q)
>
~
Q;
a::
Fig. 10-15. Action spectra for the induction of conidia-
tion ( - - ) and its reversal by blue light (-----) in
Alternaria tomato. (Adapted from Ref. 78.)
298 Chapter 10

cochrome system and in other blue-light-mediated morphogenic responses (Fig. 10-13)

indicates that a flavoprotein would be a good candidate for this photoreceptor. It has been
proposed that the two pigments involved in the mycochrome system interact in a coupled
oxidation-reduction process, (77) but the evidence so far available in support of this
hypothesis does not appear to be very strong.

10.5. PHOTOCHROMIC ADAPTATION

Cyanobacteria, which are phototrophic procaryotes, utilize phycobiliproteins as their

major light-harvesting pigments in photosynthesis (Chapter 12). Both the phycocyanins
and the phycoerythrins absorb light maximally at wavelengths that chlorophyll absorbs
poorly. The ability of the cyanobacteria to produce selectively pigments that are comple-
mentary to the color of light incident upon them at any given time would be expected to
help them utilize available photosynthetically active radiation with maximum efficiency.
Indeed, as early as 1883, Engelmann(79) observed just such an adaptation among the
cyanobacteria, which was detected as changes in cell color corresponding to changes in
the color of ambient illumination. These color changes, now known as complementary
chromatic adaptation, were later shown to result from modifications in the relative
amounts of the red-colored phycoerythrins and the blue-colored phycocyanins that were
synthesized. (80)
It is customary to divide the cyanobacteria into three groups, based on their ability to
exhibit chromatic adaptation. (81) Group I includes those organisms that do not adapt
chromatically. Organisms in group II exhibit unidirectional adaptation in that they can
control the synthesis only of phycoerythrin. Within group ill, however, the synthesis of
both phycoerythrin and phycocyanin is regulated by the wavelength distribution of inci-
dent light. These organisms are said to exhibit bidirectional chromatic adaptation. Photo-
control of phycoerythrin synthesis in this case is strictly a function of green light. Control
of phycocyanin synthesis, however, is more complicated. A total of four phycocyanin
subunits are found, each of which appears to be a unique gene product. Two subunits,
which are 17.8 and 20.2 kDa, are expressed constitutively and when assembled yield
phycocyanin that is designated as PC-I. The other two subunits, which are 17.4 and 21. 7
kDa, are found only in cells grown in red light or in darkness. Phycocyanin composed of
these two subunits is known as PC-2. It appears, therefore, that phycocyanin synthesis
results from the differential expression of four genes, only two of which are light-
regulated.
Action spectra for chromatic adaptation of representatives from both group II and
group ill indicate maxima between 540 and 550 nm for potentiation of phycoerythrin
synthesis by green light, and between 640 and 660 nm for potentiation of PC-2 synthesis
by red light. In every case, only a few minutes of actinic light is required. When
successive exposures to green and red light are given, it is always the last exposure that
determines the response. This system is thus analogous to the photoreversible responses
that are mediated by phytochrome.
Regulation of chromatic adaptation occurs at the level of transcription. (81) While it is
evident that the photoreceptor is not involved in energy transduction, it has not yet been
identified. A chromoprotein that exhibits absorbance changes consistent with a role as a
photoreversible detector of chromatic adaptation was first described by Scheibe. (82) Sub-
Photomorphogenesis 299

sequently, Bjorn and Bjorn(83) described four photoreversible pigments that they called
phycochromes a, b, c, and d. Recent work, however, has led to the conclusion that these
photochromic pigments, which are derived from native phycobiliproteins, are not in-
volved in chromatic adaptation. (81 ,84)
In a recent synthesis of available data, Tandeau de Marsac(81) proposed a model for
regulation of chromatic adaptation that assumes the existence of a photoreceptor in two
interconvertible forms, one absorbing primarily in the green (PG) and the other in the red
(PR). According to her model, both forms of the photoreceptor are active, with each
regulating a different event, in marked contrast to the phytochrome system, in which only
one form of the chromoprotein is active. An alternative hypothesis (Fig. 10-16) suggests
that only the PR form is active and that it functions both as a repressor of the genes coding
for PC-2 and as an inducer of phycoerythrin synthesis. This alternative model accounts for
the existence of PC-2 both in complete darkness and in red light, since expression of PC-2
would occur whenever PR is absent. Conversely, PR would act as an activator or de-
repressor of phycoerythrin synthesis.
To learn more about the control of complementary chromatic adaptation at a mo-
lecular level, it is evident that methods for the functional incorporation of foreign genetic
material into the cyanobacteria are needed. As these methods have just become available,
it is likely that they will soon be used for this purpose. A few pigmentation mutants,
which may reveal important information concerning chromatic adaptation, have already
been isolated. Information obtained with eight "blue mutants," each of which simul-
taneously either lost or had impaired ability both to photoinduce phycoerythrin and to
photorepress phycocyanin,(84) is consistent with the model presented here (Fig. 10-16).
Ultimately, however, isolation and characterization of the photocontrolled genes, and in
particular their promotor/repressor regions, will be required to determine the mode of
action of the photoreceptor system.

10.6. THE FUTURE

Apart from those photomorphogenic responses that can be ascribed to phytochrome,

the most compelling need is for the identification of photoreceptors. This task will be
difficult both for blue-light and UV-B radiation photoreceptors for reasons that have
already been mentioned. Since a direct approach to identifying photoreceptors appears
unlikely to be successful, it is evident that a different strategy is needed. Given the rapid
advances made in recent years in molecular genetics, a genetic approach would appear to
be most promising. Since strategies are being developed for the identification of defective
genes in the absence of prior knowledge of their products, it may become possible to
identify "photoreceptor genes" in the absence of any information about the chemical

Fig. 10-16. Hypothesis for the control of complementary chromatic green ~ Phycoerythrin
PG~PR
adaptation by green and red light via a hypothetical photochromic inactive red active~
pigment that exists in both an inactive, green-absorbing form (PG) PC-2
and an active, red-absorbing form (PR). PR would function by simultaneously promoting synthesis of phy-
coerythrin and repressing synthesis of PC- 2, which is phycocyanin that is composed of the two subunits that are
photoregulated.
300 Chapter 10

identity of the photoreceptor. Once relevant genes have been identified, methods for
subsequent determination of their products are already available. Thus, it should eventual-
ly be possible to identify gene products that are in some way related to a photoreceptor
without knowing anything in advance about its chemical structure. The most serious
limitation to this approach at present is that it will be necessary to work with an organism
for which there are available highly efficient methods for the introduction of foreign
genetic material under circumstances that will permit development of a mature organism
that has stably integrated the introduced DNA into its own genome. Keeping in mind the
present state of technology in this area, it is unlikely that this approach will be undertaken
until that technology has improved.
With respect to phytochrome, two questions seem most pressing. First, phytochrome
present in green oat shoots has been shown to be different from that which predominates
in etiolated oat shoots.(47,48) Similarly, pea shoots contain at least two distinct popula-
tions of phytochrome, one that is relatively more abundant when plants are grown in the
dark, the other more abundant when they are grown in the light. (52) Thus, it becomes
important to determine whether the chromoproteins that are most abundant in dark- and
light-grown tissues, respectively, are differentially modified products of the same gene, or
products of different genes. Moreover, it is evident that all of the information gained
about phytochrome so far, virtually all of which has been obtained with the chromoprotein
that is abundant only in plants grown for extended periods in darkness, might be of little
value when considering phytochrome-mediated morphogenesis in photosynthetically
competent plants grown under natural environmental conditions. Clarification of the
question of the nature of phytochrome in green plants is needed.
Second, we still know far too little about phytochrome itself, even from etiolated
plant tissues. What, precisely, are the differences between Pr and Pfr? On which part(s) of
the molecule are these differences to be found? What domains on phytoch~ome are likely
to be important with respect to its biological function? Since these domains are likely to
have been highly conserved throughout evolution, it is important to identify those regions
that are common to phytochromes isolated from a wide variety of taxonomically distinct
organisms. Monoclonal antibodies, which recognize well-defined, relatively small re-
gions on an antigen, are ideally suited to both purposes. Antibodies that discriminate
between Pr and Pfr have already been described,<39,40) as has been an antibody that
recognizes an antigenic determinant on a polypeptide the size of phytochrome from both
angiosperms and algae.(85) This work, however, represents only a beginning. Finally,
even though the primary structures of phytochrome from etiolated oats and zucchini have
recently been predicted from their cDNA sequences, (34,86) equally significant advances
will also have to be made concerning knowledge of its secondary, tertiary, and quaternary
structures if we are eventually to understand how it might function in a primary sense.
Consequently, continued biochemical and biophysical investigations of phytochrome it-
self are required.

10.7. REFERENCES

I. A. Schwartz and D. Koller, Phototropic response to vectorial light in leaves of Lavatera cretica L., Plant
Physiol. 61, 924-928 (1978).
2. D. Vince-Prue, Photoperiodism in Plants, McGraw-Hili, London (1975).
Photomorphogenesis 301

3. H. Mohr, Pattern specification and realization in photomorphogenesis, Encycl. Plant Physiol., New Ser.
16A, 336-357 (1983).
4. D. L. Correll, J. L. Edwards, and W. Shropshire, lr., Phytochrome: A Bibliography with Author, Biolog-
ical Materials, Taxonomic and Subject Indexes of Publications Prior to 1975, Smithsonian Institution
Press, Washington, D.C. (1977).
5. Z. Lechowski and 1. Bial~zyk, Interaction between green and far-red light on the low fluence rate chlo-
roplast orientation in Mougeotia, Plant Physiol. 85, 581-584 (1987).
6. W. Shropshire, lr., and H. Mohr (eds.), Encycl. Plant Physiol., New Ser. 16A and 16B, Springer-Verlag,
Berlin (1983).
7. W. L. Butler, Remembrances of phytochrome twenty years ago, in: Photoreceptors and Plant Development
(1. A. DeGreef, ed.), pp. 3-7, Antwerpen University Press, Belgium (1980).
8. H. A. Borthwick, The biological significance of phytochrome, in: Phytochrome (K. Mitrakos and W.
Shropshire, Jr., eds.), pp. 3-23, Academic Press, London (1972).
9. S. B. Hendricks, Photochemical aspects of plant photoperiodicity, Photophysiology I, 305-331 (1964).
10. L. H. Flint and E. D. McAlister, Wavelengths of radiation in the visible spectrum inhibiting the germination
of light-sensitive lettuce seed, Smithson. Misc. Collect. 94, 1-11 (1935).
11. D. Meischke, Uber den Einfluss der Strahlung auf Licht- und Dunkelkeirner. Jahrb. Wiss. Bot. 83, 359-
405 (1936).
12. H. A. Borthwick and S. B. Hendricks, Photoperiodism in plants, Science 132, 1223-1228 (1960).
13. W. L. Butler, K. H. Norris, H. W. Siegelman, and S. B. Hendricks, Detection, assay, and preliminary
purification of the pigment controlling photoresponsive development of plants, Proc. Natl. Acad. Sci. USA
45, 1703-1708 (1959).
14. D. F. Mandoli and W. R. Briggs, Phytochrome control of two low-irradiance responses in etiolated oat
seedlings, Plant Physiol. 67, 733-739 (1981).
15. B. Frankland and R. Taylorson, Light control of seed germination, Encycl. Plant Physiol., New Ser. 16A,
428-456 (1983).
16. P. Schopfer and K. Apel, Intracellular photomorphogenesis, Encycl. Plant Physiol., New Ser. 16A, 258-
288 (1983).
17. V. Gaba and M. Black, The control of cell growth by light, Encycl. Plant Physiol., New Ser. 16A,358-400
(1983).
18. A. L. Mancinelli and I. Rabino, The high irradiance responses of plant photomorphogenesis, Bot. Rev. 44,
129-180 (1978).
19. K. M. Hartmann, A general hypothesis to interpret high energy phenomena of photomorphogenesis on the
basis of phytochrome, Photochem. Photobiol. 5, 349-366 (1966).
20. A. L. Mancinelli, Light-dependent anthocyanin synthesis: A model system for the study of plant pho-
tomorphogenesis, Bot. Rev. 51, 107-157 (1985).
21. H. Lange, W. Shropshire, lr., and H. Mohr, An analysis of phytochrome-mediated anthocyanin synthesis,
Plant Physiol. 47, 649-655 (1971).
22. K. M. Hartmann, Ein Wirkungsspektrum der Photomorphogenese unter Hochenergie-bedingungen und
seine Interpretation auf der Basis des Phytochroms (Hypokotyl-wachstumshemmung bei Lactuca sativa L.),
Z. Natuiforsc. Ser. B 22, 1172-1175 (1967).
23. L. Fukshansky and E. Schafer, Models in photomorphogenesis, Encycl. Plant Physiol., New Ser. 16A,69-
95 (1983).
24. C. 1. Beggs, M. G. Holmes, M. labben, and E. Schafer, Action spectra for the inhibition of hypocotyl
growth by continuous irradiation in light and dark-grown Sinapis alba L. seedlings, Plant Physiol. 66,615- .
618 (1980).
25. H. Smith, Light quality, photoperception, and plant strategy, Annu. Rev. Plant Physiol. 33, 481-518
(1982).
26. H. Smith and D. C. Morgan, The function of phytochrome in nature, Encycl. Plant Physiol., New Ser. 16B,
491-517 (1983).
27. M. 1. Kasperbauer, Far-red light reflection from green leaves and effects on phytochrome-mediated assimi-
late partitioning under field conditions, Plant Physiol. 85, 350-354 (1987).
28. C. L. Ballare, R. A. Sanchez, A. L. Scopel, 1. 1. Casal, and C. M. Ghersa, Early detection of neighbour
plants by phytochrome perception of spectral changes in reflected sunlight, Plant Cell Environ. 10, 551-
557 (1987).
302 Chapter 10

29. M. G. Holmes and H. Smith, The function of phytochrome in plants growing in the natural environment,
Nature 254, 512-514 (1975).
30. 1. D. Bewley and M. Black, Physiology and Biochemistry of Seeds, Vol. 2, Viability, Dormancy and
Environmental Control, Springer-Verlag, Berlin (1982).
31. L. H. Pratt, Phytochrome: The protein moiety, Annu. Rev. Plant Physiol. 33, 557-582 (1982).
32. J. C. Lagarias, Progress in the molecular analysis of phytochrome, Photochem. Photobiol. 42, 811-820
(1985).
33. J. M. Kelly and J. C. Lagarias, Photochemistry of 124-kilodalton Avena phytochrome under constant
illumination in vitro, Biochemistry 24, 6003-6010 (1985).
34. H. P. Hershey, R. F. Barker, K. B. Idler, J. L. Lissemore, and P. H. Quail, Analysis of cloned cDNA and
genomic sequences for phytochrome: Complete amino acid sequences for two gene products expressed in
etiolated Avena, Nucl. Acids Res. 13, 8543-8558 (1985).
35. R. E. Kendrick and C. J. P. Spruit, Phototransformation of phytochrome, Photochem. Photobiol. 26,201-
214 (1977).
36. H. P. Hershey, R. F. Barker, J. T. Colbert, J. L. Lissemore, and P. H. Quail, Phytochrome: Molecular
properties, feedback regulation of mRNA levels and genomic clone isolation, in: Molecular Form and
Function of the Plant Genome (L. van Vloten-Doting, G. S. P. Groot, and T. C. Hall, eds.), pp. 101-lll,
Plenum Press, New York (1985).
37. L. H. Pratt, Phytochrome: Function and properties, Photochem. Photobiol. Rev. 4,59-124 (1979).
38. J. Shanklin, M. Jabben, and R. D. Vierstra, Red light-induced formation of ubiquitin-phytochrome conju-
gates: Identification of possible intermediates of phytochrome degradation, Proc. Natl. Acad. Sci. USA 84,
359-363 (1987).
39. M.-M. Cordonnier, H. Greppin, and L. H. Pratt, Monoclonal antibodies with differing affinities to the red-
absorbing and far-red-absorbing forms of phytochrome, Biochemistry 24, 3246-3253 (1985).
40. M. -M. Cordonnier, Recent advances in the study of the protein moiety of phytochrome, in: From
Photobiophysics to Photobiology (A. Favre, R. Tyrrell and J. Cadet, eds.), pp. 229-244, Elsevier, Amster-
dam (1987).
41. P. H. Quail, D. Marme, and E. Schafer, Particle-bound phytochrome from maize and pumpkin, Nature
[New Biol.J 245, 189-191 (1973).
42. L. H. Pratt, Molecular properties of phytochrome, Photochem. Photobiol. 27,81-105 (1978).
43. P. H. Quail, Intracellular localization of phytochrome, in: Trends in Photobiology (C. Helene, M. Charlier,
and T. L. Garestier, eds.), pp. 485-500, Plenum Press, New York (1982).
44. D. Marme, Phytochrome: Membranes as possible sites of primary action, Annu. Rev. Plant Physiol. 28,
173-198 (1977).
45. D. McCurdy and L. H. Pratt, Irnmunogold electron microscopy of phytochrome in Avena: Identification of
intracellular sites responsible for phytochrome sequestering and enhanced pelletability, J. Cell Bioi. 103,
2541-2550 (1986).
46. V. Speth, V. Otto, and E. Schafer, Intracellular localisation of phytochrome in oat coleoptiles by electron
microscopy, Planta 168, 299-304 (1986).
47. J. G. Tokuhisa, S. M. Daniels, and P. H. Quail, Phytochrome in green tissue: Spectral and immu-
nochemical evidence for two distinct molecular species of phytochrome in light-grown A vena sativa L.,
Planta 164, 321-332 (1985).
48. Y. Shimazaki and L. H. Pratt, Immunochemical detection with rabbit polyclonal and mouse monoclonal
antibodies of different pools of phytochrome from etiolated and green Avena shoots, Planta 164, 333-344
(1985).
49. M.-M. Cordonnier, H. Greppin, and L. H. Pratt, Phytochrome from green Avena shoots characterized with
a monoclonal antibody to phytochrome from etiolated Pisum shoots, Biochemistry 25,7657-7666 (1986).
50. L. H. Pratt and M.-M. Cordonnier, Phytochrome from green Avena, in: Phytochrome and Photoregulation
in Plants (M. Furuya, ed.), pp. 83-94, Academic Press, Tokyo (1987).
51. J. G. Tokuhisa and P. H. Quail, The levels of two distinct species of phytochrome are regulated differently
during germination in Avena sativa L., Planta 172, 371-377 (1987).
52. H. Abe, K. T. Yamamoto, A. Nagatani, and M. Furuya, Characterization of green tissue-specific phy-
tochrome isolated immunochemically from pea seedlings, Plant Cell Physiol. 26, 1387-1399 (1985).
53. H. Mohr, Lectures on Photomorphogenesis, Springer-Verlag, New York (1972).
54. Y.-S. Wong, H.-C. Cheng, D. A. Walsh, and J. C. Lagarias, Phosphorylation of Avena phytochrome in
vitro as a probe of light-induced conformational changes, J. BioI. Chem. 261, 12089-12097.
Photomorphogenesis 303

55. E. M. Tobin and 1. Silverthorne, Light regulation of gene expression in higher plants, Annu. Rev. Plant
Physiol. 36, 569-593 (1985).
56. G. Morelli, F. Nagy, R. T. Fraley, S. G. Rogers, and N.-H. Chua, A short conserved sequence is involved
in the light-inducibility of a gene encoding ribulose 1,5-bisphosphatase carboxylase small subunit of pea,
Nature 315, 200-204 (1985).
57. S. B. Hendricks and W. 1. Van der Woude, How phytochrome acts: Perspectives on the continuing quest,
Encycl. Plant Physiol., New Ser. 16A, 3-23 (1983).
58. I. A. Newman, Rapid electric responses of oats to phytochrome show membrane processes unrelated to
pelletability, Plant Physiol. 68, 1494-1499 (1981).
59. H. Senger and W. R. Briggs, The blue light receptor(s): Primary reactions and subsequent metabolic
changes, Photochem. Photobiol. Rev. 6, 1-38 (1981).
60. G. LOser and E. Schafer, Are there several photoreceptors involved in phototropism of Phycomyces
blakesleeanus? Kinetic studies of dichromatic irradiation, Photochem. Photobiol. 43, 195-204 (1986).
61. 1. Gressel, Blue light photoreception, Photochem. Photobiol. 30,749-754 (1979).
62. H. Senger (ed.), Blue Light Effects in Biological Systems, Springer-Verlag, Berlin (1984).
63. W. R. Briggs, The phototropic responses of higher plants, Annu. Rev. Plant Physiol. 14, 311-352 (1963).
64. D. 1. Cosgrove, Rapid suppression of growth by blue light: Occurrence, time course, and general charac-
teristics, Plant Physiol. 67, 584-590 (1981).
65. A. L. Mancinelli, The photoregulation of anthocyanin synthesis. Encycl. Plant Physiol., New Ser. 16B,
640-661 (1983).
66. H. P. Chon and W. R. Briggs, Effect of red light on the phototropic sensitivity of com coleoptiles, Plant
Physiol. 41, 1715-1724 (1966).
67. K. V. Thimann and G. M. Curry, Phototropism and phototaxis, in: Comparative Biochemistry, Vol. 1 (M.
Florkin and H. S. Mason, eds.), pp. 243-306, Academic Press, New York (1960).
68. W. Rau, Untersuchungen iiber die Lichtabhlingige Carotinoidsynthese. I. Das Wirkungsspektrum von
Fusarium aquaeductuum, Planta 72, 14-28 (1967).
69. R. D. Vierstra and K. Poff, Mechanism of specific inhibition of phototropism in com seedlings by
phenylacetic acid, Plant Physiol. 67, 1011-1015 (1981).
70. D. S. Berns and 1. R. Vaughn, Studies on the photopigment system in Phycomyces, Biochem. Biophys.
Res. Commun. 39, 1094-1103 (1970).
71. W. R. Briggs, A blue light photoreceptor system in higher plants and fungi, in: Photoreceptors and Plant
Development O. A. DeGreef, ed.), pp. 17-28, Antwerpen University Press, Belgium (1980).
72. S. Widell, R. J. Caubergs, and C. Larsson, Spectral characterization of light-reducible cytochrome in a
plasma membrane-enriched fraction and in other membranes from cauliflower inflorescences, Photochem.
Photobiol. 38, 95-98 (1983).
73. W. Schmidt and W. L. Butler, Flavin-mediated photoreactions in artificial systems: A possible model for
the blue-light photoreceptor pigment in living systems, Photochem. Photobiol. 24,71-75 (1976).
74. E. Wellmann, UV radiation in photomorphogenesis, Encycl. Plant Physiol., New Ser. 16B, 745-756
(1983).
75. W. Iwanzik, M. Tevini, G. Dohnt, M. Voss, W. Weiss, P. Graber, and G. Renger, Action of UV-B
radiation on photosynthetic primary reactions in spinach chloroplasts, Physiol. Plant. 58, 401-407
(1983).
76. W. B. Sisson and M. M. Caldwell, Photosynthesis, dark respiration, and growth of Rumex patientia L.
exposed to ultraviolet irradiance (288 to 315 nanometers) simulating a reduced atmospheric ozone column,
Plant Physiol. 58, 563-568 (1976).
77. T. Kumagai, Mycochrome system in the induction of fungal conidiation, in: Blue Light Effects in Biological
Systems (H. Senger, ed.), pp. 29-38, Springer Verlag, Berlin (1984).
78. T. Kumagai, Action spectra for the blue and near ultraviolet reversible photoreaction in the induction of
fungal conidiation, Physiol. Plant. 57,468-471 (1983).
79. T. W. Engelmann, Farbe und Assimilation, Botanische Zeitung 41, 1-13 (1883).
80. K. Boresch, Die komplementlire chromatische Adaptation, Arch. Protistenk. 44, 1-70 (1922).
81. N. Tandeau de Marsac, Phycobilisomes and complementary chromatic adaptation in cyanobacteria, Bull.
Inst. Pasteur 81, 201-254 (1983).
82. J. Scheibe, Photoreversible pigment: Occurrence in a blue-green alga, Science 176, 1037-1039 (1972).
83. G. S. Bjorn and L. O. Bjorn, Photochromic pigments from blue-green algae: Phycochromes a, b, and c,
Physiol. Plant. 36, 297-304 (1976).
304 Chapter 10

84. J. G. Cobley and R. D. Miranda, Mutations affecting chromatic adaptation in tbe cyanobacterium Fre-
myella diplosiphon, J. Bact. 153, 1486-1492 (1983).
85. M.-M. Cordonnier, H. Greppin, and L. H. Pratt, Identification of a highly conserved domain on phy-
tochrome from angiospenns to algae, Plant Physiol. 80,982-987 (1986).
86. R. A. Sharrock, J. L. Lissemore, and P. H. Quail, Nucleotide and amino acid sequence of a Cucurbita
phytochrome cDNA clone: Identification of conserved features by comparison with Avena phytochrome,
Gene 47,287-295 (1986).
 11
Photomovement

II. 1. Introduction ..................................................................... 306

11.2. Tenninology and Methods ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 307
11.2.1. Terminology .............................................................. 307
11.2.1.1. Photokinesis. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 308
11.2.1.2. Photophobic Responses. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 309
11.2.1.3. Phototaxis ....................................................... 309
11.2.1.4. Photoaccurnulation and Photodispersal ................................ 310
11.2.1.5. Photodinesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 310
II. 2.1. 6. Phototropism ..................................................... 31 0
11.2.1.7. Terminology Summary ............................................. 310
11.2.2. Methods ................................................................. 311
11.2.2.1. Individual Cell Methods ............................................ 311
11.2.2.2. Cell Population Methods ............................................ 315
11.2.2.3. Methods for Measuring Phototropism ................................. 316
11.2.2.4. Methods Summary ................................................ 316
II. 3. Photosensory Transduction Mechanisms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 317
11.3.1. PhotomovementofWholeCells .............................................. 318
II. 3.1.1. Cyanobacteria .................................................... 319
11.3.1.la. Photoresponses .......................................... 319
11.3.l.lb. Receptor Pigments for Photomovement ...................... 319
11.3.1.lc. The Primary Photoreaction ................................ 320
11.3.1.ld. Motor Responses ........................................ 322
11.3.1.2. Halobacterium halobium . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 323
II. 3.1. 2a. Photoresponses ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 323
11.3.1.2b. Receptor Pigments for Photomovement ...................... 323
11.3.1.2c. The Primary Photoreaction ................................ 324
11.3.1.2d. Motor Responses ........................................ 325
11.3.1.3. Chlamydomonas .................................................. 325
II. 3.1. 3a. Photoresponses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 325
11.3.1.3b. Receptor Pigments for Photomovement ...................... 325
11.3.1.3c. The Primary Photoreaction ................................ 328
11.3.1.3d. Motor Responses ........................................ 329
11.3.1.4. Stentor coeruleus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 329
11.3.1.4a. Photoresponses .......................................... 329
11.3.1.4b. Receptor Pigments for Photomovement ...................... 329
11.3.1.4c. The Primary Photoreaction ................................ 330
11.3.1.4d. Motor Responses ........................................ 332
11.3.2. Photomovement of Organelles (Photodinesis) ................................... 332
II. 3.2.1. Photoreceptor Pigments for Photodinesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 333

Pill-Soon Song • Department of Chemistry and Section of Molecular Plant Biology, School of
Biological Sciences, University of Nebraska, Lincoln, Nebraska 68588-0304. Kenneth L.
Poff • MSUIDOE Plant Research Laboratory, Michigan State University, East Lansing, Michi-
gan 48824-1312.

305
306 Chapter 11

11.3.2.2. Polarotropism of Chloroplasts ....................................... 333

11.3.2.3. Transduction Mechanisms of Polarotropism ............................ 335
11.3.3. Phototropism.............................................................. 337
1 i.3.3.1. Photoreception .................................................... 337
11.3.3.2. Signal Generation and Transduction .................................. 339
11.3.3.3. Measurement of Light Direction ..................................... 339
11.4. Ecological Significance ............................................................ 341
11.5. References ...................................................................... 342

11.1. INTRODUCTION

Organisms are subjected to a variety of environmental stimuli to which they respond and
adapt. A schematic relationship between the stimulus and the response is illustrated in
Fig. 11-1. Each stimulus (e.g., light, chemical) in some way interacts (e.g., absorption)
with a discrete receptor in a step referred to as perception. The perceived stimulus signal is
then processed through a chain of reactions, leading to an outcome such as a change in
motility. This entire process from the perception of a stimulus to the response is referred
to as sensory transduction.
As might be expected from the variety of stimuli listed in Fig. 11-1, there is variation
in the complexity of the transduction chains, and even more variation in the level at which
we understand the particular response mechanism. For example, bacterial chemotaxis
(swimming toward or away from a chemical stimulus) is perhaps the best characterized
and understood of the sensory transduction systems. (1) Thermotaxis (moving toward or
away from heat) of certain bacteria and amoebae reflects the surprising ability of an
organism to respond to extremely small temperature changes using some presently un-

Stimulus: Chemical, including 02

photons
heat
mechanical force
electrical stimulant
gravitational force
magnetiC field, etc.

Transduction
"black box"

Fig. 11-1. Transduction processes of different stimuli through a "black box" that mayor may not be com-
monly shared for different stimulus-response systems. Stimulus perception (circle) is carried out by a respective
receptor (e.g., photoreceptor for photons as the stimulus).
Photomovement 307

known biothermometer. (2) Some stimuli may evoke a single, highly specific response. In
contrast, in some cases, other stimuli such as a gravitational force, or mechanical and
electrical stimulation may evoke several different sensory responses. (3,4) No matter what
the stimulus, it appears that at least one organism can be found with a behavioral response
to that stimulus. For example, some organisms are able to align themselves with respect to
the magnetic field of the earth in the so-called magnetotaxis phenomenon. (5)
In this chapter, we will deal exclusively with that aspect of photosensory transduc-
tion known as photomovement. Photomovement refers to any light-induced motility or
behavioral response involving the spatial displacement of all or part of an organism. Most,
if not all, procaryotic and eucaryotic organisms exhibit some form of motile/behavioral
reaction to light direction, intensity, and wavelength (for a review, see Ref. 6). However,
the reactions to light stimuli known as photomovement are not confined to freely motile
organisms. For example, many sessile plants respond to light stimuli through the move-
ment or spatial displacement of a portion of the organism, the movement of cells, or the
movement of cellular organelles. Phototropic curvature by a plant toward or away from
the light stimulus (cf. Section 11.3.3) is included as a well-known example of the
displacement of a portion of an organism (for reviews, see Refs. 7-16). In eucaryotic
organisms, cells and cellular organelles move as a result of conformational changes of
certain critical proteins of the contractile elements and membranes involved. For example,
cytoplasmic streaming in plant cells is well known. Many of these cell and organelle
movements are sensitive to light. One of the best characterized responses of cell organell-
es to light is an intracellular chloroplast rotation, which is discussed in Section 11.3.2 for
the green alga Mougeotia.
In addition to the diversity in photomovement responses between organisms, there
may also be diverse responses in the same organism. Many organisms ranging from
procaryotes and unicellular eucaryotes to higher animals, such as birds and fish, exhibit
several responses under different wavelengths of light or different light intensities.
The reader should keep in mind throughout this chapter that the various examples of
photomovement discussed have an adaptive advantage for the organism. For example, an
exposure to light can be either beneficial or deleterious to an organism. In such a case, one
can understand the advantage to the organism of sensing the light and coupling this
perception with a movement response such that the organism is optimally positioned in
light or darkness. Thus, an organism may use light through photosynthesis for its energy
content, but to optimize the process of photosynthesis, it may use a nonphotosynthetic
photomovement mechanism and position itself close to the optimum light intensity for
photosynthesis.

11.2. TERMINOLOGY AND METHODS

11.2.1. Terminology
Although there are many problems with definitions, it is necessary to briefly cover
the commonly used terms to make our discussion of photomovement easier. One specific
problem of which the student should be wary is the tendency to use a term to describe a
phenomenon or mechanism instead of seeking an understanding of that phenomenon or
308 Chapter 11

mechanism. However, the student of photomovement frequently encounters a major

problem with terminology, since the most widely accepted terminology is based on
mechanism. Thus, if the precise mechanism of a photomovement response under observa-
tion is not clearly known, it is difficult to define such a photomovement response with a
generally acceptable term. Moreover, there is frequently no general agreement among the
workers in this field for an acceptable set of definitions to describe a particular pho-
tomovement phenomenon.
In 1976, a group of photobiologists proposed a set of terms describing the various
movement responses in photosensitive organisms. (17) In this chapter, we use these funda-
mental photomovement terms, which are based on the concept of a photoresponse.
A photoresponse may be defined as any light-induced change in the organism's
motor activity, resulting in an alteration of the movement or orientation of the organism.
The three major classes of photoresponses of motile organisms are termed photokinesis,
photophobic responses, and phototaxis. Each of these types of photoresponse may lead to
exactly the same photomovement, since the term refers not to the final response but rather
to the mechanism for achieving that response. A photomovement that is restricted to a
cellular organelle is termed a photodinesis, while the orientation to light of a nonmotile
organism is termed phototropism. These terms are defined below, and the photoresponses
are represented diagrammatically in Fig. 11-2.

11.2.1.1. Photokinesis
Imagine a population of organisms swimming faster in dim light than in bright light,
and in an environment that is half shaded. Each individual would swim quickly through
the shaded portion but would decrease its speed upon entering the unshaded portion. The
organism would swim through the unshaded portion and back into the shaded portion, and
so on, but, as a result of the change in swimming speed, would spend a greater percentage
of time in light than in shade. Thus, if this population were observed after a time, more
organisms would be in bright light than in dim light. This is an example of photokinesis.

'0
....ro~......o

C
'0- 1'*0

hI'
owe
b

d
",..0
...
vo'-"'O.......o.....-o"""01.l\oo()f

Fig. 11-2. Types of photoresponses: (a) Positive

0
W photokinesis. The time interval between the suc-
cessive positions of a cell is constant; thus, the cell
t ~~:- is moving faster in the light (inside the circle) than in
~ darkness. (b) Photoklinokinesis. The organism
~'VV'O~~

changes direction more frequently in light (inside

--
the circle) than in darkness. (c) Positive phototaxis.
e
--
The organism turns toward the light source. (d)
hy

~
Step-up photophobic response. The cell stops, and
changes direction upon entering a light field (inside
the circle). (e) Phototropism. A plant seedling grows
toward light.
Photomovement 309

Photokinesis is a response in which the steady-state rate of activity ofthe organism is

affected, without adaptation, by the steady-state light stimulus. If the activity parameter
measured is swimming velocity, it may be called "photoorthokinesis," but this is com-
monly referred to simply as photokinesis. If the activity parameter measured represents
the frequency of directional change, then it may be referred to as "photoklinokinesis"
(e.g., a bacterium may tumble both in light and in darkness, but tumble more frequently in
light).
A positive photokinesis refers to an activity rate, e.g., swimming rate, that is higher
in the presence of the stimulus light than in its absence (dark control). A negative
photokinesis refers to an activity rate, e.g., swimming rate, which is lower in the presence
of the stimulus light than in its absence. Thus, in the example given above in which the
organism swam slower in bright light than in dim light, the photoresponse would be called
a negative photokinesis.

11.2.1.2. Photophobic Responses

Imagine another population of organisms in an environment that is half shaded.
Imagine also that each organism swims smoothly when going from dim to bright light.
However, upon going in the opposite direction, from bright light to dim light, the swim-
ming pattern is altered; the organism stops, tumbles briefly, and then continues swim-
ming. The tumble results in a change in swimming direction, which may be into the dimly
lighted area but may also be into the brightly lighted area. This will result in a higher
probability for the organism entering the brightly lighted area than for the organism
leaving that area. Thus, if this population were observed after a time, more organisms
would be in bright light than in dim light. This is an example of a photophobic response.
Notice that the end result is the same for this example as for that given above for
photokinesis, but that the mechanism is different.
Photophobic response refers to a transient alteration in the activity, e.g., directional
change or swimming velocity, of the organism elicited by a change in the stimulus light
intensity. The photophobic response most commonly observed is, as in the example, a
stop response, typically followed by a change in movement direction. As a consequence,
an organism either moves into the area of higher light intensity or away from the area,
depending on the given organism, the light stimulus, and other variables.
A photophobic response may be due either to a step-up stimulus (a sudden increase in
light intensity) or a step-down stimulus (a sudden decrease in light intensity) stimulus.
Thus, in the example given above, the organism exhibited a photophobic response upon
encountering a decrease in light intensity so the photoresponse could be referred to as a
step-down photophobic response.

11.2.1.3. Phototaxis
Imagine yet another population of organisms in an environment that is half shaded.
Imagine that each organism can in some way determine the light direction, and orient
itself so that it swims toward the light, and thus stay within the brightly lighted area. If this
population were observed after a time, more organisms would be in bright light than in
dim light. This is an example of a phototactic response. The end result, an accumulation
310 Chapter 11

of organisms in light, is the same for this example as for those given above for pho-
tokinesis and for the photophobic response; only the mechanism is different.
Phototaxis is the oriented movement of an organism toward or away from the source
of a light stimulus, referred to as positive or negative phototaxis, respectively. In the
example above, the organism exhibited a positive phototaxis, moving toward the light
source. The direction of the taxis often depends on the stimulus intensity, with many
organisms showing positive phototaxis at low light intensity and negative phototaxis at
high light intensity.(l8) At an intermediate light intensity, the organism may in fact move
perpendicular to the source of the light stimulus; such a response is called transverse taxis
or diaphototaxis. (17)

11.2.1.4. Photoaccumulation and Photodispersal

In the above examples, the populations of organisms, through different photore-
sponses, accumulated in a lighted area. The terms photoaccumulation and photodispersal
are used to describe the end result of photokinesis, a photophobic response, or phototaxis
on populations of organisms. Responses that result in organisms accumulating or dispers-
ing from a region of higher fluence rate are referred to as photo accumulation and pho-
todispersal, respectively. Photodispersal is a behavioral consequence of positive pho-
tokinesis, a step-up photophobic response, or negative phototaxis. Similarly,
photoaccumulation is a behavioral consequence of negative photokinesis, a step-down
photophobic response, or positive phototaxis. Note that the terms photoaccumulation and
photodispersal describe the effect of a photoresponse on a population of organisms, and
not the photoresponse itself.

11.2.1.5. Photodinesis
The term photodinesis refers to the phenomenon in which light triggers or modulates
the movement of intracellular organelles. A photodinesis can be an acceleration of
existing streaming (as in cytoplasmic streaming) or an induction of movement of resting
organelles in response to light stimuli. An example of photodinesis will be discussed in
Section 11.3.2.

11.2.1 .6. Phototropism

Charles Darwin and his son Francis observed that canary grass seedlings grew toward
a lighted window. (l9) This growth-mediated directional response of a sessile organism or
organ to an asymmetric external light stimulus is phototropism. The organ may bend
toward or away from the direction of light stimulus in what are termed positive and
negative phototropism, respectively. (The bending of an organ relative to the electric
vector of a polarized light stimulus is polarotropism, and is usually excluded from the
definition of phototropism in the strict sense. (20))

11.2.1.7. Terminology Summary

Photoaccumulation or photodispersal by a population of organisms may result from
either one of three different photoresponses: photokinesis, phototaxis, or a photophobic
Photomovement 311

response. If the photomovement is restricted to a cellular organelle, it is referred to as a

photodinesis. If, however, a sessile organism or organ orients with respect to light through
a growth response, it is called phototropism. The reader is referred to the treatise by Haupt
and Feinleib(21) for a discussion of terminological problems in the field of photomove-
ment. However, at the same time, the reader is cautioned that the objective is not to
concentrate on a set of terms but rather to understand the mechanisms of photomovement.
Terms used incorrectly may obscure that goal.

11.2.2. Methods
A suitable method of observation and an appropriate choice of movement param-
eter(s) is critical for a study of photomovement. Not surprisingly, a considerable number
of methods for the measurement of photomovement have been described in the literature.
These methods can be divided into two categories: those based on the response of a single
cell or organ to the light stimulus, and those based on the response and/or behavior of a
population of cells in response to the light stimulus. Some of the most commonly used
methods for measuring photomovement are described below.

11.2.2.1. Individual Cell Methods

An individual cell and its photoresponse can be observed using a microscope and
recorded manually, or recorded automatically using a videomicroscopy system, such as
the one shown in Fig. 11-3. In this device, actinic light (i.e., light of a proper wavelength
and fluence rate to produce a response) is provided from the side of a culture chamber, and
the directional movement of the cells is then photographed or video-recorded using a
nonactinic monitoring beam from under the transparent culture chamber. A typical result
is presented in Fig. 11-4, showing six individual cells of the biflagellate Chlamydomonas
moving randomly in the control exposed only to nonactinic red light (a), and moving away
from the source of stimulus with actinic blue-green light, i.e., exhibiting negative pho-
totaxis (b). Large numbers of cells can be traced for their tactic response using such a
videornicroscopy recording system. In addition, automated motion analysis systems have
been developed to analyze the precise movements of the video-recorded cells.
To quantitate the phototaxis of a population of cells or organisms, one can measure
the direction of each organism with respect to the direction of the stimulus light. For
example, one organism may be directed 45 0 from the stimulus light. Another organism
may be directed directly toward (0°) the stimulus light. A number of such data may be
combined using the orientation mathematics described by Batschelet(24,25) to arrive at the
mean direction of the popUlation and, more importantly, the directedness of the popula-
tion. The mean direction of a population is between 0 and 3600 , and directedness ranges
from 0.0 for a completely random distribution to 1.0 for a distribution in which every
individual is oriented in exactly the same direction.
The application of such an analysis to the recordings shown in Fig. 11-4 yields a
value for directedness of 0 and 0.94 for the dark control and light-stimulated cells,
respectively. Thus, these represent a random distribution and a highly directed distribu-
tion, respectively. (22,24,25)
Figure 11-5 shows another example of videornicroscopy applied to observe and
record a step-up photophobic response by cells of the ciliate Stentor coeruleus using a
312 Chapter 11

Video
monitor

Video

camera

Trinocular microscope

c 3 m mI
• 45mm. /
L········.·;·;.;.·········.·;·;.::7.J __ J ha~
0
D····.·.·.·;.:········.·.· ..;.;V---'I._..::...:::.;'::-.__..I>
Fig. 11-3. Block diagram of a vid-
eomicroscopic recording system for
observing the photomovement of in-
dividual cells. The monitoring light

t L is usually from a dim infrared source

that does not act as an actinic stim-
Heat filters ~ ulant but is sufficient for vid-
eorecording by an IR-sensitive cam-
Red filter I era. C: sample chamber; L: actinic

Monitoring light i I light source (e.g., laser or Xe arc

plus monochromator or interference
filter). (Taken with minor modifica-
tion from Ref. 22.)

b
RED UGHT + BLUE-GREEN STIMULUS

..
Fig. 11-4. Videomicroscopic tracings of swimming paths (for 2 s) exhibited by cells of Chlamydomonas. Each
circle represents the cell position, at intervals of 1/3 s. (a) Control with nonactinic red background light. (b) In
actinic blue-green light incident from the left. (Adapted from Ref. 23.)
Photo movement 313

Nlit;C)·<:::~~~f.l~t;
..;~;:~~. :~'.
:.::;~.i· .

light trap

Fig. 11-5. Step-up photophobic response of Stentor coeruleus. The position of two cells has been traced every
0.24 s from a videomicroscopic recording. The stippled area represents darkness, while the clear circle repre-
sents an area lighted at 700 W/m 2 . (Drawing by Il-Hyun Kim.)

setup similar to that shown in Fig. 11-3. Response parameters such as the fraction of
responding cells, stop response at the dark-light border, rotation and swimming speeds,
and reversal of the ciliary beating direction, etc., can be determined from video recordings
of a sample of Stentor coeruleus.
Stentor coeruleus also shows negative phototaxis. (26) However, to establish the
phototactic response, it is necessary to distinguish between a true phototactic orientation
and a photophobic response of the organism. A Stentor cell might exhibit movement away
from a light source, with direction maintained through a series of step-up photophobic
314 Chapter 11

Focusing lens

-.
L i g ht
--to

Fig. 11-6. Focused light beam used to distinguish between a phototactic response and a photophobic response.
Cells of Stentor coeruleus move away from the light source even through and beyond the focal point of the lens.
The stippled area represents darkness. (Modified from Ref. 27.)

responses. Thus, movement in the "wrong" direction would elicit a step-up photophobic
response permitting a course correction. On the other hand, the cell might have the
capacity for phototaxis, having a mechanism for measuring the direction of the light, and
maintaining its course relative to that light direction. A focused light beam, such as that
shown in Fig. 11-6, can be used to distinguish between phototaxis and a step-up pho-
tophobic response as the mechanism for movement of cells away from the light. Light is
focused by a convex lens such that the focal point is within the culture chamber. The light
direction is from the left in this apparatus. However, in the absence of an absorbing
medium, the fluence rate is highest at the focal point of the lens. It then follows that a cell
possessing only a step-up photophobic response, and lacking a negative phototactic re-
sponse, would be incapable of moving through the region of highest fluence rate at the
lens focal point. Thus, since cells move within the light beam through the focal point, they
must in fact measure light direction itself, and therefore exhibit phototaxis in addition to
the more easily demonstrated step-up photophobic response. (28)
A second technique to demonstrate a true phototactic response is illustrated in Fig.
11-7. In this case, an organism, such as Chlamydomonas. moving toward one light source

:111 off
,
,
I

r~
hVZ on

Fig. 11-7. Test for a phototactic response in Chlamydomonas. The alga

abruptly changes its swimming direction from the original light direction (lro \)

~ to the new light direction (Jro 2 ) turned on normal to the first source at the same
time that the first source is turned off.
Photomovement 315

can be subjected to a second light source applied from a direction normal to the direction
of the first at the same time that the first source is turned off. If the intensity of the two
sources is the same, then the fluence rate experienced by the cell is constant, so no
photophobic response should be elicited. If the cell changes its direction smoothly toward
the second light source, then the response is a true positive phototactic response. Similar
tests can be applied to other organisms that exhibit phototaxis.

11.2.2.2. Cell Population Methods

There is a variety of population methods for measuring the photomovement re-
sponses of motile organisms. One classical method, devised by Engelmann, (29) uses a
light-trap. A small spot of light is projected onto an otherwise dark field of organisms.
After a period of time, the cells may accumulate within or disperse from the lighted area,
and the results can be easily seen with the unaided eye. Keep in mind that this demon-
strates only the presence or absence of a photoresponse, and gives no information con-
cerning the mechanism for that response.
One of the more common methods for measuring photomovement automatically
measures the photoaccumulation of cells or organisms through the use of a device called a
phototaxigraph. (30) In the phototaxigraph, an infrared monitoring beam passes through a
chamber containing a population of cells and is photoelectrically detected by means of a
photomultiplier tube. Accumulation or dispersal of the cells is induced by an actinic light

TIME {min)

Fig. 11-8. A device for measuring phototaxis of Stentor utilizing an infrared monitoring beam. (1) IR light
source. The light emerges through a circular hole in the side of the lamp housing, is filtered by Flo and focused
by condensing optics (2) onto an optical fiber bundle (3). The IR beam is directed through a cuvette (4)
containing Stentor cells, passes through a diffuser plate (D), and is detected with a CdS photocell connected to
an appropriate amplifier. The cells can also be exposed to an actinic light beam from light source (5) connected
to a transformer (6). The wavelength of the actinic light is defined by filters F2 and FJ . Inset shows a typical x-y
recorder (7) tracing from the photocell amplifier. (Modified after Ref. 31.)
316 Chapter 11

and measured as a change in the amount of radiant energy from the monitoring beam
striking the photomultiplier tube.
A relatively simple device has been used to measure the photoaccumulation and
photodispersal of Stentor coeruleus (Fig. 11-8). For the measurement of cell density at the
distal portion of the chamber away from the stimulus source, a monitoring beam from an
optical fiber is detected with a photocell. The inset in this figure is a typical recording,
showing that the cells accumulate maximally within about 2 min at a given fluence rate of
stimulus light. As expected, the microscopic counting of the cells in the accumulated
population matched the photocell tracing shown in Fig. 11-8. Such a device can also be
used to measure the average swimming speed of the cells.
It should be cautioned that the population methods employing photoaccumulation
alter the stim~lus strength and directionality due to the absorbance caused by the ac-
cumulating cells. Therefore, the net result will be a compromise that is time- and con-
centration-dependent. Moreover, these methods necessarily result in high cell concentra-
tions. Since the cells are certainly metabolically active and modifying their environment,
and since the response of the cell is expected to be sensitive to its environment, the
photomovement responses of a population of cells should not be expected to always
reflect the photomovement responses of single cells. (31,32) In addition, it must be remem-
bered that little information concerning mechanism may be obtained using a population
method. The major value of the population methods is their simplicity. Their use is
important but should not replace studies of the behavior of single cells.

11.2.2.3 . Methods for Measuring Phototropism

Phototropic curvature of an organ is measured by determining the angle of curvature
toward or away from the light stimulus direction. Figure 11-9 shows a time lapse display
(from a photographic recording) of the phototropic bending of the fungus Phycomyces
toward a blue-light source. From such data, the angle of curvature can be measured with a
protractor, while elongation-type growth can be measured quite simply with a ruler.
Transducers have been described with which one can automate the measurement of
curvature and growth.(33,34)

11.2.2.4. Methods Summary

Several of the more commonly used methods for measuring photomovement have
been described. In general, those methods based on the direct observation of individuals

Fig. 11-9. Phototropic bending of the sporangiophore of the fungus Phy-

comyces blakesleeanus toward the source of actinic blue light. Position shifts
of the sporangiophore represent 5-min intervals. (Adapted from Ref. 33 as
modified by Ref. 34.)
Photomovement 317

yield the most information relevant to mechanism. Those methods based on the response
by a large population are the most easily automated. Choice of the appropriate method
may be critical for the study of any particular photomovement. A review of some of the
methodology for measuring photomovement should be consulted for more detail. (35)

11.3. PHOTOSENSORY TRANSDUCTION MECHANISMS

A very general scheme of photosensory transduction was shown in Fig. 11-1. Let us
now consider a somewhat more complex version (Fig. 11-10) of that scheme, recognizing
at least some ofthe complexity that we know must exist. In the perception step, the light
stimulus is absorbed by a photoreceptor pigment that then participates in some bio-
chemical reaction, setting into operation a sequence of events that includes at least some
amplification and processing and ends in a response. In most cases of sensory physiology,
it is not solely the stimulus that results in a response. Rather, the organism is in some
manner extracting some piece of information from the stimulus, and responding based on
that information. For example, in phototaxis or phototropism, the organism is not re-
sponding solely to light but to light direction. Similarly, the organism may respond to
fluence rate or a change in fluence rate. Thus, processing in many cases must include a
mechanism for the extraction of information from the signal.
The physical and chemical basis of the absorption of light is described in Chapter 1.
For the purposes of this discussion, we need only remind the reader that absorption of a
quantum by a photoreceptor pigment causes an electron in that pigment to be raised to the
so-called excited state. Since the energy requirements for the electron transitions are
specific, only specific wavelengths of light are absorbed by a particular pigment. Because
absorption must occur to initiate the photosensory transduction process, it follows that the
response is inherently wavelength-dependent. The spectral features of light absorption by
various photobiological receptors have been discussed elsewhere. (36,37)
The excited photoreceptor pigment must interact with some biochemical to convert
the perceived stimulus into a signal that can act as the trigger to elicit the subsequent
transduction chain reactions. The generation of such a signal usually begins with an
ultrafast primary reaction, ~, of the excited photoreceptor pigment. The specific nature
of the primary reaction depends on the photoreceptor pigment system involved. To ensure
a high photosensitivity of the organism, the primary reaction must be fast enough to
effectively compete with the other relaxation processes of the excited state of the pigment
molecules. If this were not the case, a large percentage of the absorbed quanta would be
"wasted" in one of these other relaxation processes, and not productively elicit the
response. Extending this line of reasoning, one can also argue that primary reactions are
more likely to originate from the singlet excited state (SI in Fig. 11-11), than the triplet

Fig. 11-10. Perception, signal generation, and sig-

nal processing in photosensory transduction. The
black box in Fig. 11-1 is divided into signal genera-
tion and signal processing and amplification.
318 Chapter 11

Fig. 11-11. Modified Jablonski energy level diagram

for a photoreceptor pigment. kp: rate constant for fluo-
rescence, 108 - 109 s - }; kic: rate constant for internal
k
~ conversion, 107-1010 S - 1; kisc,k'ise: rate constants
----"'Photosensory
\ .. c for intersystem crossing, 106 - 108 s - }; kp: rate con-
'-.--- T, transduction
stant for phosphorescence, 10 - 2_ 102 S - 1; krx: rate
hv k, k,c
constant for biochemical reaction; So: ground state;
~sc kp
S 1: lowest singlet excited state; T}: lowest triplet ex-
cited state.

state (T 1). This is because the efficiency of the triplet state reaction is limited by the
relatively slow rate of intersystem crossing from the S 1 to the T 1 state. On the other hand,
the triplet state reaction will be more important in bimolecular processes in which the
reaction requires the longer lifetime of the triplet state.
Once the signal is generated, it must be processed and sufficiently amplified to
trigger the cellular/organelle responses, usually mechanical in nature, such as reversal of
the ciliary beat direction in protozoa and reversal in flagellar rotation in flagellates and
bacteria. The processing and amplification steps, where known, also depend on the
specific system involved.
In the following sections, we will discuss a few selected examples of photomove-
ment systems rather than describing a great number of diverse organisms that display
photomovement. We hope to show the student a bit of the diversity of photosensory
transduction mechanisms and illustrate the use of various experimental approaches in
elucidating these mechanisms.

11.3.1. Photomovement of Whole Cells

The initial act of photosensory transduction, leading to a photomovement response in
an organism, is the absorption of stimulus light quanta by the photoreceptor(s) (Fig.
11-10). When one begins work with a previously undescribed photoresponse, the pho-
toreceptor pigment and the transduction chain are unknown and, in addition, cannot be
directly observed; only the response itself can be directly observed or measured. For-
tunately, this may permit one to approach the identification of the photoreceptor pigment
through the measurement of an action spectrum. Such a spectrum is a plot of the response,
usually expressed in relative quantum efficiency or the reciprocal of fluence rate needed
for a given response, e.g., 50% response, as a function of wavelength. An action spec-
trum is of value if it reflects the spectral absorption characteristics of the photoreceptor
pigment in vivo. Because of various errors in the action spectral measurements and optical
biases in vivo, such as light scattering, reflection, refraction, absorption by screening
pigments, and polarization of incident light, the action spectrum usually does not quan-
titatively match the absorption spectrum of the photoreceptor pigment in vitro. Neverthe-
less, action spectroscopy permits one to eliminate many of the pigments in the organism,
leaving a few or, optimally, one candidate for the physiological photoreceptor pigment.
For this reason, action spectroscopy is the primary method used by the photobiologist for
the identification of the photoreceptor pigment for any photobiological process, including
photomovement responses (refer to Chapter 1 for a more detailed discussion of action
Photomovement 319

spectroscopy). Action spectra for some photomovement responses are described below,
and the possible photoreceptor chromophores are discussed.
The mechanisms of photosensory transduction in whole cells will be illustrated with
two procaryotes, photosynthetic Cyanobacteria (blue-green algae) and the nonphoto-
synthetic flagellate Halobacterium halobium, and two eucaryotes, the photosynthetic
flagellate Chlamydomonas and thenonphotosynthetic ciliate Stentor coeruleus. For each
of the four organisms, we will describe the photoresponse, the photoreceptors, signal
generation and amplification, and the mechanisms of movement.

11.3 .1 .1. Cyanobacteria

11.3 .1.1 a. Photoresponses
Cyanobacteria exhibit the three major classes of photomovement responses. They
show positive and negative phototaxis and photophobic responses, and also show pho-
tokinesis in response to light of photosynthetically active wavelengths. Reviews of these
photomovement responses in Cyanobacteria such as Phormidium and Anabaena should
be consulted for a more extensive discussion. (18,38)

11.3 .1.1 b. Receptor Pigments for Photomovement

Figure 11-12 shows action spectra for three photoresponses in the filamentous
cyanobacterium Phormidium uncinatum. (39-41) The action spectrum for positive pho-
tokinesis exhibits maxima in the red and blue regions of the spectrum. This spectrum is
very similar to the absorption spectrum of chlorophyll a, which shows absorption bands in
the red and blue. These data are consistent with the suggestion that this pigment is the
photoreceptor chromophore for the photokinesis response. However, the absorption of
light by chlorophyll a drops rapidly at wavelengths longer than 700 nm, in contrast with
the photokinesis action spectrum, which shows a relatively high efficiency at these wave-
lengths. There are three possible explanations for this discrepancy: (1) The efficiency of
wavelengths of light below 700 nm may be suppressed because of the screening effect of
other bulk pigments, such as the photosynthetic accessory pigments. This would dispro-
portionately exaggerate the efficiency of the far-red wavelengths where screening by other

Fig. 11-12. Action spectra for (A) photokinesis, (B) pho-

totaxis, and (C) photophobic response in Phormidium un- 400 500 600 700 600
cinatum. (Modified after Refs. 39-41.) wavelength, nm
320 Chapter 11

pigments is minimal. (2) The chlorophyll involved in photokinesis may represent a special
type of chlorophyll a rather than the bulk chlorophyll a in the organism. It has been
suggested that the chlorophyll involved here is that of photosystem I. (18) (3) The pho-
tokinesis action spectrum could be some pigment other than chlorophyll a. There is
inadequate information at present to decide which of these possibilities is correct. Not
surprisingly, other Cyanobacteria also show action spectra for photokinesis suggestive of
chlorophyll as the photoreceptor chromophore.
Although the shapes of the action spectra for photokinesis and the photophobic
responses are dramatically different (Fig. 11-12), chlorophyll a also has been suggested as
a photoreceptor pigment for the photophobic responses in Phormidium uncinatum. How-
ever, for the photophobic responses, the suggestions have been that both photosystem I
and photo system II chlorophyll a function as the primary photoreceptor pigment, with
other photosynthetic accessory pigments (particularly chlorophyll b and the phycobilins
with phycocyanobilin and phycoerythrobilin as typical chromophores) contributing to the
action spectrum in the 500- to 650-nm region (Fig. 11-12). It is not known why these
accessory pigments contribute to the action spectrum of the photophobic responses but not
to the action spectrum of photokinesis, if indeed photo system I chlorophyll a is responsi-
ble for the photokinetic response in Phormidium uncinatum.
In contrast to the action spectra for photokinesis and for the photophobic responses,
the phototaxis action spectrum shows maxima in the green, blue, and near-UV regions
(Fig. 11-12). Flavins absorb over much of this spectral region. However, the phototaxis
action spectrum extends well beyond 550 nm, where flavins do not absorb. Thus, it is
unlikely that a flavin is the primary photoreceptor pigment for phototaxis in Phormidium.
It is possible that phototaxis in Phormidium uncinatum uses a pigment complex consisting
of a primary photoreceptor pigment and antenna pigments. If the concentration of the
primary photoreceptor were negligibly low compared to that of antenna pigments, its
absorbance would not contribute significantly to the action spectrum (Fig. 11-12) relative
to the contribution of the antenna pigments. Thus, a pigment absorbing at wavelengths
longer than 600 nm, such as chlorophyll a, could be the primary photoreceptor pigment
without any significant indication of this in the action spectrum. The action spectrum
would then represent the contribution of the light-harvesting antenna pigments, which for
the green, blue, and near-UV region could be flavins, carotenoids, and/or phycobilins. It
is interesting that the cyanobacterium Anabaena shows different action spectra for
positive and negative phototaxis, with the former probably mediated by C-phycocyanin
and the latter mediated by chlorophyll a as the primary photoreceptor pigment. (42,43)

11.3 .1.1 c. The Primary Photoreaction

In the photokinetic responses of Cyanobacteria, if one accepts the action spectrum
(Fig. 11-12) as reflecting the absorption characteristics of chlorophyll a in the photo-
synthetic systems, the primary photoreaction is most likely to be photoionization and
charge separation in the photosynthetic reaction center. This primary reaction leads to the
photophosphorylation of ADP (Chapter 12). The ATP formed then serves as an energy
source and accelerates the motile activity of Cyanobacteria in the photokinetic
response. (40,41)
Photomovement 321

The nature of the primary reactions in the phototactic and photophobic responses of
Cyanobacteria are more ambiguous than that for photokinesis. In addition, the action
spectra for phototaxis and the photophobic responses are substantially different. There-
fore, the nature of their primary photoreactions cannot be discussed with any certainty.
Perplexingly, in the Anabaena, the positive and negative phototaxes exhibit different
action spectra, suggesting that the photoreceptors for these opposing phototactic responses
are not identical (see Section II. 3 .1.1 b).

~~------------------------~
~Photolynthetic system

Proton flux; PMF

~
Membrane potential change

trichome in light

Light

- - Light/Dark border

Dark

•
PMF collapse; signal

Plasma membrane depolarization

Voltage-sensitive Ca 2+ channels

Ca 2+ influx; amplification

Part of the trichome in dark

Step-down response

Fig. 11-13. Proposed transduction steps involved for the step-down photophobic response in Phormidium. As
part of the trichome moves into darkness (represented by the lower rectangle), the photosynthetically generated
membrane potential and PMF collapse, electrically polarizing the two ends (one in light and one in darkness) of
the trichome. A membrane depolarization in the darkened part leads to an opening of a Ca2 + channel and
amplification of the electric signal.
322 Chapter 11

Based on the similarity between the photophobic action spectrum (Fig. 11-12) and
that for COz fixation, a model has been proposed in which both photosystems I and II
reaction centers (Chapter 12) participate in the primary photoreaction in the photophobic
response in Phormidium. The basic unit in Phormidium is a trichome, a linear array of
cells that is motile. As a Phormidium trichome crosses from a lighted area into a dark
area, the trichome stops and the reversal movement is elicited, i.e., a step-down pho-
tophobic response occurs. In the lighted area, the trichome's photosynthetic electron
transport system is activated. Thus, a pH gradient and a membrane potential (proton
motive force, or PMF) can be expected to have been generated across the thylakoid
membrane. Accompanying the generation of the PMF, a change in plasma membrane
potential occurs establishing a steady-state potential that is different from the resting
potential in darkness. As part of the trichome moves into the dark side, the photosynthetic
electron transport ceases, and the PMF collapses. It is suggested that the PMF collapse is
taken as a signal by the trichome, which induces depolarization of the plasma membrane.
At this point, the trichome's front end and rear (shaded) end are electrically polarized.
The membrane depolarization causes a voltage-sensitive Caz + channel to be opened
and Caz + ions to flow in from the extracellular medium. The light signal is thus ampli-
fied; the change in lighting resulting in a massive Caz + ion influx. This influx triggers the
contraction of micro filaments necessary for a stop response (Section 11.3.1.3). This
model is schematically illustrated in Fig. 11-13.
Indirect evidence supports the transduction scheme shown in Fig. 11_13.(44-46) Four
main lines of evidence can be cited: (1) The photophobic response in Phormidium can be
stimulated by an abrupt external pH change in the absence of a light stimulus. This
supports the involvement of a change in PMF, possibly as a primary signaI.<47) (2) Light-
induced membrane potential changes can be measured in Phormidium. The action spec-
trum for the induction of these changes is similar to the action spectrum for the pho-
tophobic response shown in Fig. 11-12. (48,49) These data are consistent with membrane
depolarization being involved as in the amplification step shown in Fig. 11-13. (3) A
change in the direction of movement can be induced by an external DC electrical field.
This is further evidence that electrical signals are involved in the transduction sequence.
(4) Light-induced calcium uptake has been shown in Phormidium uncinatum using radio-
active Caz + . (50) This result is consistent with the involvement of Caz + ion influx as in
the amplification of the sensory signal. Thus, considerable evidence supports the model in
Fig. 11-13 for the photophobic response in Phormidium.
Whether or not a common transduction mechanism is utilized by both photophobic
and phototactic responses in Cyanobacteria should be considered an open question.

11 .3 .1.1 d. Motor Responses

Cyanobacterium moves using a gliding mechanism that is not well understood.
Exactly how the reversal of the gliding direction is dictated by a front-rear electrical
polarization (Section 11.3.1.2) (Fig. 11-13) is not known. One can conjecture that the
contractile micro filaments undergo a significant conformational transformation upon
binding Caz + ions. Such a hypothesis is worthwhile if it generates thoughts and experi-
ments that take us closer to a true understanding of the mechanism.
Photo movement 323

11.3.1.2. Halobacterium halobium

11.3.1.2a. Photoresponses
Halobacterium halobium shows a unique wavelength dependence of its photomove-
ment which has been called "color sensing. "(51) This flagellate bacterium exhibits re-
sponses to both step-up and step-down stimuli. On a cellular level, the response is a
change in the direction of flagellar rotation from clockwise to counterclockwise or vice
versa.(52) If the stimulus is in the blue-UV region of the spectrum, a step-up stimulus
results in the photodispersal ("repellent response") of the population. However, if the
step-up stimulus is in the yellow-green region of the spectrum, the result is a photoac-
cumulation ("attractant response ") ofthe popUlation. In contrast, a step-down stimulus in
the blue-UV region results in photoaccumulation, while a step-down stimulus in the
yellow-green region results in photodispersal. Thus, the bacteria accumulate in yellow-
green light and disperse from blue-UV light (Fig. 11-14). Reviews of these photomove-
ment responses in Halobacterium should be consulted for a more extensive discus-
sion. (51-53)

11.3.1.2b. Receptor Pigments for Photomovement

As a result of the presence of rhodopsinlike compounds on the Halobacterium
halobium membrane, these membranes are purple. Two of these pigments probably
mediate the two opposing photoresponses in this organism. Thus, the photoreceptor
pigment chromophore for each of these photoresponses would be a retinal, linked to an
opsin protein via a Schiff's base.
A proposed scheme that could account for the sensory responses of Halobacterium
halobium is shown in Fig. 11-15. In addition to bacteriorhodopsin and halorhodopsin,
which are involved in proton and chloride ion transport, respectively, a new rhodopsinlike
molecule has been detected in isolated purple membrane preparations, based on data from
flash photolysis experiments. (54) This new species is called slow-cycling rhodopsin
(SR587 ) because of its relatively slow photochemical cycle. It has been proposed(51) that
SR587 is the photoreceptor pigment for the attractant response of the bacterium, and that
the blue-UV photoreceptor pigment for the repellent response of the cell is S373' a
photocycling intermed~ate of SR587 . Supporting evidence for SR587 and S373 as the pho-

"c:'"
o
c.
'"
"
a: 10-"

Fig. 11-14. Action spectra for the attractant (broken line) and re-
pellent responses (solid line) in Halobacterium halobium. The ordi-
nate gives the sensitivity for the photophobic reaction (i.e., re-
ciprocal of photons/mm2/s required for a given response).
(Redrawn from Ref. 53.) WAVELENGTH, nm
324 Chapter 11

l
~hllSt
oms b
510 two-photon IIEPEllE

J
cycle SIGNAl TRAN liT
h)l
SRS97 ------s690~S373 l ATT!lA~
SOUCTfO

~ 800m.
one-photon ~SOUCT10Ii

J
cycle

Fig. 11-15. Photochemical cycle for slow-cycling rhodopsin (sR587 ) and its relationship to the mechanism of
color sensing based on the dual role of sR as the attractant and repellent photosensory receptors. (Modified after
Ref. 51.)

toreceptor pigments has been provided by the observation that sensitivity to near-UV light
is reduced when the photoresponse is observed under conditions that cause less excitation
of SRS87 - Hence, in this experiment less S373 intermediate is formed to act as the pho-
toreceptor for the repellent response. (S1) Though this hypothesis is attractive, others do
not agree that the S373 and SRS87 pigments are the primary photoreceptor pigments_ (52)

J1.3.1.2c_ The Primary Photoreaction

Assuming that SRS87 (Fig. 11-15) is the photoreceptor pigment for the attractant
response, one can conjecture that the primary reaction of SRS87 , leading to its intermediate
S373' is likely to be a configurational isomerization of the retinylic chromophore in the
excited state of the SRS87 ' An alternative photoisomerization path can then be invoked for
the repellent response mediated by S373' The signal generated from the primary reaction
could then be a conformational change in the sR protein embedded in the membrane _This
could then modulate an ionic flux as a processible form of the signal. It should be
emphasized that the molecular details of the primary reactions are not known. However,
these possibilities are consistent with what is known about rhodopsin in higher organisms.
The flagella motors in procaryotes appear to be driven by a PMF. (SS) By analogy
with cyanobacterial photosensory transduction, we speculate that the signal generated by
the photoreceptors, e.g., SRS87 ' is processed, resulting in a change in the PMF, and is
possibly amplified through a changed Ca2 + ion flux. This process ultimately triggers a
change in the direction of rotation of the flagella, from clockwise to counterclockwise
rotation or vice versa.
The photoreceptor pigment in its excited or metastable state may interact with a
methyl-accepting protein (MaP), which acts as the site of signal integration. This is based
on the observation that specific membrane proteins in Halobacterium halobium become
methylated upon exposure to light, which would function as an attractant. In contrast,
repellent light causes demethylation of these proteins. (S2) It is suggested that the MaP then
lowers the level of a hypothetical regulator, which serves as the flagellar rotor switch. It is
known that Ca2+ ions are essential for the reversal of swimming direction, presumably
through their involvement in the methylation/demethylation cycle. Whether or not Ca2+
ions directly affect the flagella switch regulator and its conformation, and whether or not
any role is played by PMF is subject to conjecture.
Photomovement 325

1l.3.1.2d. Motor Responses

Bacteria employ one or more flagella as the motor apparatus. Each flagellum is
driven by a rotary motor called the M ring (Fig. 11-16). As mentioned above, PMF's are
the driving force for the rotary motor. Protons flow through the interstice between the M
ring and the S ring (a stator), which powers the rotary motion of the motor apparatus. In
Halobacterium halobium, rhythmic spontaneous reversals are controlled by an os-
cillator, (56) and the photoresponse is either a suppression or an induction of a reversal in
the flagellar rotary motor. It is not known whether Ca2 + alone or Ca2 + plus protons are
directly involved in the possible conformation changes of the proteins of the rotary motor
in Halobacterium halobium.

11.3 .1.3. Chlamydomonas

1l.3.1.3a. Photoresponses
Chlamydomonas reinhardtii, a green flagellate, exhibits step-up photophobic re-
sponses and a negative phototactic response. Thus, these cells disperse from a lighted
area. Several comprehensive reviews(22,57) of the photoresponses of Chlamydomonas are
available for the interested reader.

11.3 .1.3b. Receptor Pigments for Photomovement

Chlamydomonas shows a maximum positive phototactic response to light of 500-
510 nm and 440 nm, as shown in Fig. 11-17. For comparison, an action spectrum for
photoaccumulation of Euglena gracilis(30) is also shown. The action spectrum for Eu-
glena closely matches the absorption spectrum of flavin, but the differences between the
absorption spectrum of flavin and the action spectrum for Chlamydomonas are consider-
able. Although the identity of the Chlamydomonas photoreceptor pigment is uncertain,
two suggestions have been made.

Fig. 11-16. The flagellar rotor of a bacterial

flagellut'l
flagellum. Top: Flagellum on a gram-negative
bacterium such as E. coli. Bottom: the basal end
of a bacterial flagellum representing the flagellar
rotor apparatus consisting of disks (rings) that
function in the flagellar rotation. The flagellum
is composed of the protein flagellin assembled in
a helical filament wound around the central core.
~~~rOd
'<Q S-ring

The bacterium is propelled by rotation of the M-ring

flagella about their long axes. The flagellar rotor

is driven by energy (PMF) supplied from the
cytosol. (Redrawn from Ref. 55.)
326 Chapter 11

~6
~
,-,
/ \
!l/IY'\\
/ \
=4 J'-",/
~ I \ I t

I \,
~ \
~ 2 \,

OL-__~__~~__~~\~L- Fig. 11-17. Action spectra for phototaxis of Chlamydomonas re-

350 450 inhardtii (solid line) and photoaccumulation of Euglena gracilis(30)
WAVELENGTH, nm (broken line). (Redrawn from Ref. 44.)

1. The Chlamydomonas photoreceptor pigment may be a naturally occurring flavin

other than the most common flavins such as riboflavin, FMN, and FAD. These more
common flavins show a broad, single absorption band around 450 nm. Others, e.g.,
deazaflavins and hydroxyflavins,(58) show absorption maxima on either side of the 450-
nm peale Furthermore, optical biases such as reflection and scattering, and screening by
other light-absorbing substances, may result in an action spectrum that is different from
the absorption spectrum of the photoreceptor pigment. Thus, a flavin cannot be ruled out
as the primary photoreceptor.
2. A quarter-wavelength interference reflector may radically alter the light absorbed
by the photoreceptor pigment such that the action spectrum for the response differs
dramatically from the absorption spectrum of the pigment. A quarter-wavelength inter-
ference reflector is any structure composed of layers of alternating refractive indices and
of approximately the thickness of a quarter of the incident light wavelength. Light re-
flected from such a structure is constructive, resulting in a reinforced light intensity for a
particular wavelength, and destructive, resulting in a decreased light intensity for other
wavelengths. Such quarter-wavelength interference phenomena in living organisms have
been described by Land. (59) To explain this suggestion, we first must describe the
Chlamydomonas stigma.
The stigma or "eye spot" is an optically dense body at the base of the flagellum on
one side of the Chlamydomonas cell. Historically, the stigma in Chlamydomonas and
other flagellates has been thought to have a role in light sensing. In the classic model, the
stigma was thought to function as a shading device. More recently, it has been suggested
that the stigma may function as a quarter-wavelength interference reflector. Although both
models ascribe a role to the stigma, that role is dramatically different for the two models.
In the shading model, the stigma functions solely as an optically dense body, peri-
odically shading the photoreceptor pigment as the cell rotates about its long axis. The
intensity transiently perceived by the photoreceptor pigment during the periodic shading
by the stigma causes a shift in swimming direction. Thus, the cell effectively measures
light direction through two measurements with the same photoreceptor pigment at two
times in its rotation. (22)
In the quarter-wavelength model, the stigma serves as a directional antenna, which is
rotated somewhat like a radar antenna as the cell rotates while swimming. (60) This model
is based on the observation that the granular layers of the stigma can provide alternating
Photomovement 327

layers of different refractive index media, with the thickness or the spacing between the
layers being about 120 om (Fig. 11-18). Thus, this could be a quarter-wavelength inter-
ference reflector for a wavelength of 480 om, the maximum in the Chlamydomonas action
spectrum (Fig. 11-17). It has been suggested that the photoreceptor pigment is located in
the plasma membrane in front of the granular rows of the stigma. (61) If this is so, then the
two stacked layers of stigma granules shown in Fig. 11-18 can in fact act as a reflector,
and the photoreceptor apparatus in the membrane in front Can be excited by the incident
and the reflected light beams. Since the latter is strongly dependent on the direction of the
incident light beam, the mechanism would be highly directional.
Both stigma-based models would result in action spectra different from the absorp-
tion spectrum of the photoreceptor pigment. On the one hand, if the shading model is
correct, then the action spectrum must be a function not only of the absorption by the
photoreceptor pigment, but also of the shading pigment. On the other hand, if the quarter-
wavelength interference reflector model is correct, the action spectrum must be a function
of absorption by the photoreceptor pigment, reflectance of the stigma granular layers, and
absorption by those layers. There are insufficient data at present to unequivocally say
whether or not the quarter-wavelength interference reflector model is valid for Chlamy-
domonas. If it can be demonstrated that this model is valid, it will be the first example that
such a mechanism serves in an aneural organism as it does in the eye of animals. (62)
Returning to the original question, what is likely to be the primary photoreceptor
pigment of Chlamydomonas? The phototaxis action spectrum (Fig. 11-17) resembles the
absorption spectrum of a carotenoid more closely than that of a flavin. However, a flavin
chromophore can be accommodated given substantial screening, etc., or a "nontypical"
flavin. (58)
The photomovement response (a photodispersal) can be restored in a carotenoidless
mutant of Chlamydomonas by exogenously added retinal isomers. (60) Retinals are the
photoreceptor chromophores of animal vision and their photochemistry is well estab-
lished, so it would not be surprising if the photoreceptor pigment in Chlamydomonas were
a retinal. Unfortunately, these experiments do not completely rule out photodynamic and
thermal effects of the exogenous retinals in the alga. For example, Paramecium without
apparent pigmentation can respond to light when artificial dyes such as methylene blue
and eosin are added to the cells.(63,64) Similarly, the nonphotosensitive, colorless species
Stentor polymorphus becomes phototactic when pigmented with symbiotic Chlorella. (57)

Fig. 11-18. (A) Schematic diagram showing the array of stigma

globules of the eye spot or stigma in Chlamydomonas. C: chloroplast,
CW: cell wall. P: plasmalemma. (B) Multilayer stack as a reflector.
b~ - - -
The stack consists of alternative quarter-wavelength layers of high (n\)
and low (n2) refractive index media. The refractive index of air is
assumed to be no = I.
328 Chapter 11

These examples illustrate the point that an exogenous pigment can induce or modify the
photoresponses of the organisms. In summary, the best evidence at present indicates that a
retinal is the photoreceptor pigment for phototaxis of Chlamydomonas. Caution should
always be exercised in the interpretation of action spectra, since the difficulties encoun-
tered in the identification of the Chlamydomonas photoreceptor pigment are similar to
problems involved in other systems.

1l.3.1.3c. The Primary Photoreaction

A step-up light stimulus causes a transient reversal in the flagellar strokes in Chlamy-
domonas. Although the primary reaction and the subsequent signal processing are not well
understood, what is qualitatively known at present is summarized in Fig. 11-19. Two
important steps in the photosensory transduction chain are depolarization of the plasma
membrane following a primary reaction of the photoreceptor, and Ca2 + ion influx, which
triggers a reversal of the flagellar rotor. A similar scheme has been proposed for the
phototactic response in Chlamydomonas. (44)

hy ,
Perception by Photoreceptor

l
Primary Reaction
(Conformation Change 1)

I ",,0' ,""'""'0
!
Depolarization of Membrane

Signal Amplification

Ca 2+ Influx

I
Flagellar Rotor Reversal

;:....
..... l
'. Response
Fig. 11-19. Schematic diagram of pho-

~
tosensory transduction in Chlamydomonas.
hI! ---+
Note the key role of the Ca2 + ion influx in
reversing the flagellar rotor in response to
the light stimulus.
Photomovement 329

11.3.1.3d. Motor Responses

The "motor" of the flagellated Chlamydomonas is considerably more complex than
that of the flagellated bacterium. In general, the flagella and cilia of eucaryotic cells
possess a characteristic protein assembly composed of nine doublet rnicrotubules sur-
rounding a pair of singlet rnicrotubules ("9 + 2 axoneme"). It is reasonable to suggest
that a light-induced influx of Ca2 + ions (Fig. 11-19) triggers a conformational change in
the axoneme, resulting in the reversal in flagellar stroke in Chlamydomonas. For further
information concerning the flagellar motor of eucaryotic cells, consult the excellent
review volume edited by Brokaw and Verdugo.(65)

11.3 .1.4. Stentor coeruleus

11.3.1.4a. Photoresponses
Cells of the ciliate Stentor coeruleus exhibit a photophobic response and also a
phototactic response (Section 11.2.2.1). Comprehensive reviews on the photomovement
responses of this organism(22,57) should be consulted for further information.

11.3 .1.4b. Receptor Pigments for Photomovement

The action spectrum for the step-up photophobic response in Stentor coeruleus (Fig.
11-20) is similar to the action spectrum for the negative phototactic response, suggesting
that the same photoreceptor chromophore is involved in both photoresponses. (26,66,67)
Moreover, these action spectra closely match the absorption spectrum of whole Stentor
coeruleus cells, the absorption of which is mainly due to the pigment stentorin. This
compound, which has a hypericinlike chromophore, is located in pigment granules,
presumably in the cell membrane. (68) Although the photoreceptor chromophores of the
photophobic and phototactic responses in Stentor coeruleus appear to be identical, it is not
established whether or not the same set of photoreceptors is utilized in both responses. If
the phototactic response is a consequence of a series of photophobic responses in Stentor
coeruleus, as discussed below, it then follows that the ultimate basis for the two responses
is the same.

r-------------...,.1.0
100 O.B .,
c:
0.6 ;;
~

o
10 0.4 ..
E .c
Fig. 11-20 Action spectrum (solid circles; left ordinate) ~ 5
for the step-up photophobic response in Stentor coe- <5'"
0.2

ruleus. An absorption spectrum of the cell (right ordi-

nate) at 77°K is also shown for comparison. (Redrawn 400 450 500 550 600 650
from Ref. 66.) Wavelen gth [nm]
330 Chapter 11

1l.3.1.4c. The Primary Photoreaction

Based on the available data, it has been proposed that the photophobic response in
Stentor coeruleus is triggered by a light-induced proton release from the photoreceptor
pigment in the pigment granules, which are located in the ectoplasm. (57) The intracellular
and/or intraciliary pH drop causes a depolarizing effect that triggers the opening of Ca2 +
channels, as outlined in Fig. 11-21.
It is suggested that the photoreceptor pigment ejects protons efficiently during its
excited state lifetime. (57 ,69) If this proton release serves as an initial signal to trigger the
photosensory transduction chain as proposed here, then the photoresponse of the cells
should be pH-dependent. Results indeed show that the photoresponse is pH-depen-
dent.(32) A sharp drop in response is seen at pH values lower than 6.0, with the midpoint
of the drop being at 5.4. Less than 5% of the maximum photoresponse is observed at pH
values below 5.0. It is not certain whether the midpoint of the pH profile represents an
apparent pK I of the photoreceptor chromophore or its protein residue( s), but the pH effect
itself is striking. Given the strong dependence of the photoresponse on pH, one would
expect proton ionophores to show strong inhibitory effects on the photoresponse of
Stentor coeruleus.
A slow cytoplasmic pH drop under steady-state photic stimulation has been ob-
served. (57) The fact that the photoresponse occurs only at pHon! > 6 (extracellular pH)
suggests that proton release from the excited photoreceptor organelles (pigment granules)
must be sufficient to generate a transient pHin < 6 to be an effective transducing signal.
Thus far, no experiments have been carried out to monitor transient cytoplasmic pH
changes upon flash photic stimulation of Stentor coeruleus. Such experiments could test
the hypothesis that a change in pH acts as the initial transducing signal following the
photoexcitation of the photoreceptor.

!III
stimulus
signal
signal a Fig. 11-21. Tentative scheme for a step-up pho-
perception tophobic response in Stentor coeruleus. (a) The
+ initial step, signal perception, involves absorp-
H+ release to signal b tion of light by the photoreceptor chromophore
cytoplasm generation
located in the pigment granules. (b) The excited
(l'.pH) :
state chromophores release protons into the
+ cytoplasm, thus generating a signal in the form
depolarization- of a pH gradient. (c) The pH produced in step b
sensitive
Ca+:channel :
activates the depolarization-sensitive Ca2 +
Signal C channel in the plasma/ciliary membranes, result-
+ amplification
a ing in Ca2 + influx that further depolarizes the
Ca++ influx: membranes, thus amplifying the signal. (d) Cili-
transmisSion
t ary beating is reversed as a result of the de-
ciliary reversal: mechano- d polarization, causing the organism to tum and

[;;;:p.
t transduction swim in the opposite direction, thus giving a
phobic response. It is assumed that a Ca2 +
h)l ~
pump operates to cause the efflux of Ca2 + ions
following the transient ciliary reversal, thus
H phobic
Co - pump response causing the resumption of clockwise ciliary beat-
efflux ing. (Redrawn from Ref. 57.)
Photomovement 331

TABLE 11-1. Summary of the Effects of Inhibitors on the Photophobic Response

of Stentor coeruleus(26,31,57,67)

Effect
Drug Specificity Inhibition on motility
Valinomycin K+ None None
2,4-Diaminopyridinea K+ None None
Ouabaina Na+ None None
TPMP+ (triphenylmethylphos- PMFb Strong Weak
phonium)
FCCP (carbonyIcyanide-p-tri- H+ Strong Weak
fluoromethoxyphenylhydra-
zone)
CCCP (carbonyIcyanide-m- H+ Strong Weak
chlorophenylhydrazone
Nigericin H+ Strong Weak
Gramicidin H+/K+ Strong Weak
Calimycin Ca 2 + Strong None
Ruthenium red Ca2 + Strong Weak
Methoxyverapamil Ca2 + Mild Weak
Verapamil Ca2 + Mild Weak
Lanthanum Ca2 + Mild Weak
POIY-L-lysine Ca2 + Mild Weak
aUnpublished data (I. H. Kim and R. K. Prusti).
bAn electric potential component of proton motive force (PMF).

The role of Ca2 + in the transduction sequence has been examined using Ca 2 +
inhibitors and ionophores. Ca2 + blockers including ruthenium red, methoxyverapamil,
and the Ca2 + ionophore calimycin specifically inhibit both the step-up photophobic and
phototactic responses in Stentor. (57) Results from the inhibition studies suggest that during
the response, depolarization-sensitive Ca2 + channels are opened, permitting a sudden
influx of Ca2 + ions into the celL Table 11-1 summarizes the inhibitory effects of different
drugs on the photophobic response in Stentor coeruleus.
Several photoresponse-enhancing agents also exist (Table 11-2). Calimycin may
stimulate the photoresponses of Stentor coeruleus under certain conditions. Plausible

TABLE 11-2. Agents That Enhance the Photophobic Response

of Stentor coeruleus(31,66,70,71)

Effect
Agent Specificity Enhancement on motility
Calimycin Ca 2 + ionophore Mild a None
D 2 0 (50% vi v) Primary reaction (?) Strong Weak
Caffeine Ca2 + permeability Strong None/weak
Ol-Phosphatidate Ca2+ permeability Strong None
aOnly at low concentrations (:51 nM). At higher concentrations of calimycin, a ciliary reversal occurs even in the dark without a
light stimulus.
332 Chapter 11

mechanisms for the observed effects of D2 0, caffeine, and <x-phosphatidate have been
described,<66,70,7l) It is particularly noteworthy that D2 0 substantially sensitizes the cell
to a step-up phobic stimulus, while it inhibits negative phototaxis. It appears that D2 0
influences both the primary photoprocess of stentorin (Le., H+ dissociation) and the rate
of cell body rotation and action potentials. (70) Though less likely, the enhanced pho-
tophobic response in heavy water might arise from a higher concentration of singlet
oxygen in D2 0 than in H2 0 (Chapter 3), as a photophobic response can be induced by
photodynamic action in some protozoa. (63,64)
Both caffeine and <x-phosphatidate enhance the photophobic response of Stentor
coeruleus, apparently by increasing the permeability of Ca2 + across the cell membrane.
This interpretation is also consistent with a calimycin-induced ciliary reversal in the dark.
The effects of caffeine as a Ca2 + permeability promoter and <x-phosphatidate as a Ca2 +
ionophore are well known in neurobiology. (66,71)
Is the scheme shown in Fig. 11-21 applicable to the negative phototactic response in
Stentor coeruleus? If the phototactic response results from a series of photophobic re-
sponses, the same transduction mechanism would account for phototaxis and the pho-
tophobic response. However, while heavy water enhances the step-up photophobic re-
sponse in Stentor coeruleus (Table 11-2), it inhibits the negative phototactic responses. In
addition, if the same transduction mechanism accounts for phototaxis and the photophobic
response, one would expect any inhibitor of the step-up photophobic response to inhibit
phototaxis also. In addition, ions of lanthanum, which is competitive with calcium,
strongly inhibit the step-up photophobic response (Table 11-1), but the phototactic re-
sponse is not affected. In fact, the inhibition of the phototactic response by heavy water is
partially restored by lanthanum ions. (70) These results also are consistent with the exis-
tence of different transduction mechanisms for the photophobic and phototactic responses
in Stentor coeruleus. Thus, there is no evidence at present that the photophobic response
is the basis of the phototactic response, so the scheme of Fig. 11-21 for the photophobic
response must not be applicable to the phototactic response.

1l.3.1.4d. Motor Responses

The "motor" of the ciliated Stentor, like that of the flagellated Chlamydomonas, is
considerably more complex than that of the flagellated bacterium. As stated above (Sec-
tion 11.3.1. 3d), the eucaryotic cilium is composed of nine doublet microtubules surround-
ing a pair of singlet microtubules (9 + 2 axoneme). It is suggested that a light induced
influx of Ca2 + ions triggers a conformational change in the axoneme, resulting in the
reversal in ciliary stroke.

11.3.2. Photomovement of Organelles (Photodinesis)

The cellular organelles of eucaryotes are structurally dynamic, and often exhibit
varied forms of movement within the cell. Even nonmuscle cells and their intracellular
organelles are capable of movement, largely controlled by filamentous proteins, e.g.,
through actin-myosin interactions or assembly to disassembly cycles of actin molecules.
Light strongly affects many of these intracellular movements. Photodinesis, the light-
induced or modulated movement of cytoplasm, and other intracellular organelles such as
Photomovement 333

I
Fig. 11-22. Action spectra for the induc- INDUCTION -1'-- CANCEL
tion of chloroplast movement in Mou-

-e:~~I ~if\
geotia at low fluence rate (red light) and -~ I
Oi 60
the cancellation of the movement induction ~o20 _____
(far-red light). Ordinate, q, reb is relative
quantum efficiency. (Redrawn from Ref.
o ' ~ ~
; ......-----.... •
~ _ m !
_
--.,

79.) WAVELENGTH (nm)

the nucleus and chloroplasts will be briefly described using chloroplast movement in the
photosynthetic alga Mougeotia as a specific example. (See Ref. 72 for a review of
chloroplast and nuclear movements.)

11.3.2.1. Photoreceptor Pigments for Photodinesis

Generally, there are two regions of spectral sensitivity in the photodinesis of several
algae that have been investigated. Chloroplast migration or local accumulation in Vau-
cheria sessilis, (73) Mougeotia, (74) Mesotaenium, (75) Selaginella martensii, (76) Lemna tri-
sulca, (77) and Vallisneria spiralis(78) show an action spectrum maximum in the blue
region at 440-480 nm. A flavin is most likely the chromophore.
Chloroplast movement in many algae also exhibits a maximum in the red region at
600-800 nm. The red sensitivity is particularly important at low fluence rates. Some algae
such as Mougeotia exhibit a maximum sensitivity of chloroplast movement in the blue
region at higher fluence rates, as shown in Fig. 11-22. However, even these algae show
red sensitivity at low fluence rates.(79) The fact that chloroplast movement in response to
polarized light, "polarotropism," in Mougeotia is red/far-red-reversible suggests that the
photoreceptor pigment is phytochrome. (80)

11.3.2.2. Polarotropism of Chloroplasts

Each cell of a trichome of the filamentous alga Mougeotia contains one flat chlo-
roplast. The rectangular plate-shaped chloroplast turns its flat face toward light of rela-
tively low intensity within minutes of irradiation, so that the plate-shaped chloroplast
faces approximately perpendicular to the direction of actinic red light of 660 nm (' 'face

Fig. 11-23. Chloroplast rotation in Mougeotia induced by

low fluence rate red light (Iru) polarized along the axis
shown by the double-headed arrow (E). The chloroplasts in
the cells oriented perpendicular to the electric vector E tum
to the face orientation. (Modified from Ref. 80.) n
334 Chapter 11

orientation"; Fig. 11-23). However, at high fluence rates, the flat chloroplast rotates such
that its edge faces the actinic beam (the "profile orientation" shown in Fig. 11-23). The
low fluence rate response (rotation to the face orientation) shows a maximum spectral
sensitivity in the red region (Fig. 11-22), whereas at high fluence rates the profile orienta-
tion is most effectively induced by blue light. Thus, the orientations of chloroplasts are
opposite at high and low fluence rates, and different photoreceptor pigments appear to
regulate the mechanism at high and low fluence rates. In darkness, Mougeotia chlo-
roplasts remain in the orientation established by the previous exposure to light. Intensity-
dependent chloroplast movement/rotation in Mougeotia and other algal cells has been
reviewed elsewhere. (81)
Photoreversibility of the low fluence response has been established (Fig. 11-22), and
the action spectra for induction and reversal approximate the absorption spectra of phy-
tochrome in solution (Chapter 10). Phytochrome from higher plants has been isolated and
substantially characterized (Chapter 10). Since this chromoprotein serves as the red-light
receptor for numerous photomorphogenic and developmental responses in plants, it is
reasonable to assume that the basic structural and functional characteristic of phy-
tochromes are probably common to both higher plants and Mougeotia.
There is a distinct possibility that phytochrome in Mougeotia is membrane- or
organelle-bound. This suggestion is based on the fact that chloroplast rotation in Mougeo-
tia is polarotropically controlled (Fig. 11-23). Thus, red light polarized perpendicular to
the long axis of the cell is preferentially absorbed at the front and back sides of the cells,
converting the oriented and dichroic phytochrome (Pr) to its Pfr form (Figs. 11-22 and
11-23). As Pfr builds up at the front and back sides ofthe cell, the chloroplast edges are
turned away from the higher Pfr gradient (Fig. 11-24). On the other hand, red light
polarized parallel to the long axis of the cell is not effective for chloroplast rotation.
By means of microbe am irradiation, it has been shown that phytochrome is not freely
diffusable, translationally or rotationally. The face orientation of chloroplasts achieved
with red light shown in Fig. 11-24 can only be reversed with far-red light of parallel
polarization (Eli), followed by a dark period. Perpendicularly polarized light (E 1-)' is not
effective in reversing the low fluence rate response of Mougeotia. (82)

Fig. 11-24. Polarotropic chloroplast rotation. In

response to red light (Iru) polarized along the short
axis of the M ougeotia cell (D ~), the chloroplast
rotates with its edges turning toward the low P fr
concentration. The difference in P fr concentration
is established as the result of preferential excita-

"
tion of the Pr transition dipoles (double arrows
parallel to the cell surface) in the proximal and
distal sides of the cylindrical cell. Note that the Pr

+,~::,+(r_'e
transition dipoles in the upper and lower flanks of
the cylindrical cell are not favorably oriented for
the absorption of the short-axis polarized light,
resulting in low Pfr concentrations at the flanks.
Red light polarized along the long axis of the cell
/
is not effective in inducing the chloroplast rotation
1
no rotation
because the Pr transition dipoles are orthogonal to
the long axis. (Modified from Ref. 80.)
Photomovement 335

Based on the results with polarized light and those obtained using microbe am irradia-
tions, the dichroism of phytochrome most likely results from the transition moments of Pr
and Pfr that are parallel and perpendicular to the cell surface, respectively. Thus, it is
likely that phytochrome is bound to the plasma membrane. (80)

11.3.2.3. Transduction Mechanisms of Polarotropism

The polarotropic chloroplast movement in Mougeotia cells involves at least two
features: (I) inversion of the dichroism of phytochrome and (2) the primary event follow-
ing the phototransformation of the dichroically oriented phytochrome molecules.
There are several mechanisms by which this inversion of dichroism could occur.
Assuming that the Qy-transition dipole (660-nm absorption band) of Pr lies parallel to the
cell surface of Mougeotia, (80) the following mechanisms can be suggested for the Qy-
transition dipole of Pfr orienting perpendicular to the cell surface: (1) The direction of the
Qy-transition dipole changes by about 90° upon phototransformation from Pr to Pfr , with
the phytochrome molecule fixed to the membrane either as a peripheral or an integral
membrane component. (2) The whole protein rotates by about 90° upon phototransforma-
tion, with the chromophore Qy-transition dipoles of Pr and Pfr remaining fIxed with
respect to the chromophore binding site/crevice. (3) The chromophore moves and re-
orients itself (by about 90°) relative to the chromophore binding site/crevice, with the
protein moiety more or less fIxed to the membrane. (4) An extensive conformational
change of the protein moiety occurs, resulting in the Pfr chromophore oriented in space
perpendicular to the original orientation of the Pr chromophore.
Let us now consider the above mechanisms in relation to the action dichroism of
chloroplast movement in Mougeotia. The Qy-dipole orientation (mechanism 1) can
change upon phototransformation of phytochrome if the chromophore undergoes a gross
conformational/geometric isomerization. Analysis of the absorption spectra of Pr and Pfr
suggests that such a gross conformational isomerization is unlikely, (83) although a lo-
calized geometric isomerization at either end of the tetrapyrrole chromophore ring system
can OCCUr.(84) Calculations of the Qy-transition dipole orientation in Pr and the Qy-
transition dipole orientation in several assumed structures of Pfr do not change signifI-
cantly with respect to the axis of polarization in Pr.
Recently, it has been reported that the direction of the Qy-transition dipole changes
by a substantial angle upon phytochrome phototransformation from Pr to Pfro (85) This
suggests that there is a signifIcant movement of the chromophore during the phototrans-
formation of phytochrome. These results are qualitatively consistent with the observation
that the chromophore topography also changes in the phototransformation from Pr to
Pfro (86)
The phototransformation of phytochrome from Pr to Pfr does not seem to increase the
membrane fluidity of liposomes. (87) If membrane fluidity changes significantly in
Mougeotia plasmalemma, it is unlikely that phytochrome dichroism will be preserved or
that the dipole orientation will be changed significantly upon phototransformation. One
can postulate that Pr as a peripheral membrane protein rotates upon its phototransforma-
tion to the Pfr form, which then binds to the lipid bilayer core via the newly generated
hydrophobic surface. Since both electrostatic and hydrophobic forces contribute to the
binding of phytochrome to membrane,(88-90) an anisotropic rotation of the protein mole-
336 Chapter 11

cule can be envisioned. Mechanism 3 is an alternative to the above mechanism, in that

only the chromophore reorients; the topography of membrane binding remains fixed for Pr
and Pfr. In view of the chromophore movement,(85,86) it is likely that both the chro-
mophore and the protein moiety rotate upon phototransformation. Mechanism 4, involv-
ing an extensive conformational change of the protein moiety upon the phototransforma-
tion from Pr to Pfr' is not likely but cannot be ruled out entirely. It should be noted that the
chromophore reorientation depicted in mechanism 3 implicitly entails at least local con-
formational changes around the chromophore and/or binding crevice.
Following the phototransformation of Pr to Pfr by red light, chloroplast movement is
elicited in M ougeotia cells. The primary molecular event leading to chloroplast movement
remains largely obscure, although the phytochrome phototransformation accompanies a
rapid Ca2 + influx and accumulation within the Mougeotia cell. It has been suggested that
phytochrome elicits chloroplast movement in Mougeotia by modulating the level of bound
Ca2 + ions. (91)
Based in part on the observation that Ca2 + uptake is enhanced by red-light irradia-
tion of Mougeotia, (92) a tentative scheme for the transduction chain involved in chlo-
roplast movement can be formulated. Haupt and Wagner<80) proposed that the pho-
totransformation from Pr to Pfr triggers changes in ion fluxes, with an enhanced Ca2 +
influx and a release of Ca2 + from an internal sequestering system. Calcium ions then
regulate actomyosin filament activity in eliciting chloroplast movement (Fig. 11-25).
However, the regulatory mechanism of the actin-myosin interactions and their coupling
to the membrane dynamics are not yet understood.

hv
t
""T h
<- " - 'h

l
Change in Ion Fluxes
ca 2+ Influx
2+
Ca Release from Internal Stores

Actomyosin Filaments

Actomyosin filaments connect

Chloroplast edge to Plasmalemma

Contractile activity regulated by

Ca 2+ distribution
Fig. 11·25. A schematic outline of the possible transduction
steps in the chloroplast movement of Mougeotia. (Adapted
Chloroplast Movement from Ref. 82.)
Photomovement 337

11.3.3. Phototropism
Over 100 years ago, Charles Darwin accurately described many aspects of plant
phototropism, the process whereby plants and many fungi grow toward, or orient with
respect to, a source of light. (19) This process permits a given plant to optimize its capacity
to harvest light through photosynthesis, and may enable the fungus to position its spore-
bearing structure to facilitate spore dispersal. The description by Darwin provided the
beginning of the field of plant physiology, and also showed an appreciation for the great
importance of phototropism. Although phototropic responses may be observed in flower-
ing plants, ferns, mosses, algae, and fungi, the phenomenon has been studied mainly in
the seedlings of a few grasses (corn and oats) and in the fungus Phycomyces blakesleeanus
(Fig. 11-9).
The mechanism through which the plant directs its growth with respect to light
direction can be divided into three aspects: photoreception, the transduction sequence
between photoreception and the regulation of growth on the two sides of the plant, and,
finally, the procedure through which the plant measures light direction itself. Although
considerable information is available concerning all three areas, the last is best under-
stood. These topics will be discussed in sequence.

11.3 .3 .1. Photoreception

If one exposes the shoots of young seedlings, such as corn, to light from one side,
after a brief delay period the seedling shoots will begin to grow toward the source of the
light. The final curvature attained is a function of the intensity, the duration of the
exposure, and the wavelength of the light, and the opposing tendency of the plant to grow
upward in response to the force of gravity. Given that gravity is constant in an earth-based
experiment (assuming no special centrifugal forces), we can study phototropism by vary-
ing the intensity, duration, and wavelength of light, while measuring the curvature made
by the plant in response to the stimulus. In this way, we can determine an action spectrum
for phototropism that should be roughly equivalent to the absorption spectrum that we
would measure for the photoreceptor pigment regulating the process.
The maxima from several such action spectra for phototropism are presented in Table
11-3 together with absorption maxima for [3-carotene as an example of a carotenoid and
riboflavin as an example of a flavin, representing the two classes of compounds that agree
best with the data from action spectra. Two points are evident from these data. First, the
action spectra for phototropism for many different organisms are surprisingly similar. In
fact, even the action spectrum for the fungus Phycomyces is essentially the same as the
action spectra for phototropism in flowering plants. All of the action spectra show a peak
at about 370 om, and a series of peaks between 400 and 490 om. Based on these
similarities, it is generally assumed that a single photoreceptor pigment regulates pho-
totropism in all of these organisms. Second, it is evident that neither carotenoid nor flavin
absorption spectra exactly match these action spectra.
The peak separation in the blue is mimicked either by a carotenoid or by a flavin in a
hydrophobic environment, such as oil, but the 370-nm peak is best explained by a flavin.
Neither carotenoids nor flavins show a 280-nm maximum, but the latter absorb in the 260-
 tl
00

TABLE 11-3. Comparison of the Action Maxima for Phototropism and Absorption Maxima
for Riboflavin and All-trans Il-Carotene
Material Action/absorption maxima (nm)a
Phycomyces (phototropism) 280b 370" (385") 430c 445 c (455 d ) 470 c (485")
Avena (phototropism) 280- 38St 42St 44St 47St
J3-Carotene in ethanol, room temp. (absorption)" 427 449 475
J3-Carotene in ethanol, nOK, all-transk 3881 411 434 468 501
J3-Carotene in ethanol, nOK, IS,IS'-cisk 3S1 414 432 465 493
Riboflavin in aqueous buffer at 23°C (absorption) 266h 373" 43S i 445" 48S;
Riboflavin in ethanol at 77°K (absorptionY 270 3S0 418 447 472
aNumbers in parentheses refer to shoulders. bRef. 93. cRef. 94. dRef. 95. <Ref. 96. /Ref. 97. ,Ref. 98. hRef.99. fRef. 100. jRef. 101. kRef. 102.
The visible absorption band of f3-carotene is resolved into sharp vibrational peaks at 77°K. fA weak absorption band.

o:::r"
AI

1..
......
Photomovement 339

nm region, and the 280-nm action maximum could result from the attachment of the
pigment moiety to a protein or be due to screening by proteins and nucleic acids. Based on
data such as these, it is thought that a flavin attached to a protein and constrained in the
hydrophobic matrix of a membrane is the most probable candidate for the photoreceptor
pigment. This conclusion, i.e., that the photoreceptor pigment is a flavoprotein, is sup-
ported by data from other experiments. For example, those compounds known to specifi-
cally inhibit phototropism all bind to or in some way interfere with the action of fla-
vins. (103) These and other data strongly support the flavin hypothesis, but none of the data
can be considered conclusive proof. Ultimately, this could be provided by a mutant with a
single gene change, lacking or with a reduced amount of one specific flavoprotein, and a
comparable decrease in the phototropic response.

11.3.3.2. Signal Generation and Transduction

Although very little work has been done on the transduction sequence in Phy-
comyces, enough information exists to suggest a possible sequence for converting pho-
toreception into the directed growth response. If light were directly stimulating the
activity of a rate-limiting step in growth, a higher light intensity at the distal wall of the
organism would increase growth on the side away from the light, resulting in growth
toward the light. The cell wall is composed primarily of chitin, the synthesis of which is
stimulated by light. (104) If light were directly stimulating the activity of a rate-limiting
enzyme in chitin synthesis, and if chitin synthesis were the rate-limiting step in growth, a
higher light intensity at the distal wall of the organism would result in increased chitin
synthesis on the side away from the light, and growth toward the light. Unfortunately,
although light is focused onto the distal side (Section 11.3.3.3), there is some internal
screening, so that the total light energy absorbed is higher on the proximal side than on the
distal side. Hence, any asymmetry in chitin synthesis must depend on a more complex
spatial and temporal pattern of light intensity in Phycomyces.
For many years, there has been convincing evidence that growth in plants is regu-
lated by hormones such as indoleacetic acid (IAA). This growth regulator is produced by
cells at the tip of the shoot and is transported down the shoot, regulating growth of the
cells below. The Cholodny-Went theory for phototropism suggests that IAA also is
transported across the shoot from the illuminated side to the shaded side. (See Ref. 14 for
a more extensive discussion of this theory.) As a result of this redistribution, cell elonga-
tion is inhibited on the illuminated side and stimulated on the shaded side, and thus the
shoot grows toward the lighted side or twoard the unilateral light. The best evidence
supporting this theory are data showing that growth is indeed stimulated on the shaded
side and inhibited on the lighted side of the shoot. (105)

11.3.3.3. Measurement of Light Direction

One of the most interesting aspects of phototropism is the mechanism used to
measure light direction. This is also the aspect that is generally considered to be best
understood. To measure light direction, some gradient must exist in light absorption, i.e.,
there must be some method that results in a greater number of quanta absorbed on one side
of a unilaterally irradiated organism than on the other side. Two obvious models accom-
340 Chapter 11

(Al
Fig. 11-26. (A) Diagram demonstrating the path of light in the cross-section
of a cylindrical organism such as Phycomyces with low internal absorption.
The straight lines with arrows represent three rays of light incident from the
screen
left. Refraction occurs for each ray not normal to the surface of the organism
as that ray passes from air with a low index of refraction to cytoplasm with
its relatively high index of refraction. With the low light losses within the
cytoplasm, the intensity of light is higher at the distal surface (12) than at
the proximal surface (1 1), (B) Diagram demonstrating the path oflight in the
cross-section of a cylindrical organism such as a corn seedling with high
internal absorption. As in Part A, the straight lines with arrows represent
three rays of light incident from the left. Screening is indicated by the
stippled area. As a result of the screening by absorption or scattering within
the organism, the intensity of light is lower at the distal surface (12) than at
(B) the proximal surface (11)'

plish this, one through refraction and one through screening. Either could be employed by
the organism to determine light direction.
In the refraction or "lens" mechanism, light is refracted as it passes from air with a
relatively low index of refraction (1.0) to cytoplasm, which is mostly water with its
relatively high index of refraction (1.34), and is thereby focused onto the distal side of the
cylindrical organism (Fig. 11-26A). Given low light losses across the organism (e.g., low
absorption and low scattering), the intensity of light will be higher on the distal side than
on the proximal side of the cylindrical organism (Fig. 11-26A). If the organism could
measure the light intensity in those two places, this "information" could be translated
into light direction. For instance, wall deposition could be related to the number of quanta
absorbed, and would be greater on the distal side resulting in growth toward the light.
The refraction mechanism is at least the dominant mechanism used by Phycomyces to
measure the direction of visible light. It is evident that the difference in index of refraction
between the air and the organism is the crucial element in the refraction model. It should
not, then, be surprising that manipulation of the index of refraction has been used to
obtain the most convincing evidence for the refraction mechanism in Phycomyces. The
index of refraction has been changed by immersing the cylindrical sporangiophore in oil
with an index of refraction roughly equal to that of the cytoplasm. Without the difference
in index of refraction, focusing does not occur, and the phototropic response is lost. (106)
The effect of the focusing within the organism has also been changed by placing a
cylindrical glass rod between the organism and the unilateral light. The light, if focused
by the additional lens such that the normal focusing by the organism does not result in an
increased light intensity on its distal side but rather on the proximal side, causes the
organism to grow away from the light.
Photomovement 341

In the screening mechanism, light is attenuated as it passes through the organism

such that the intensity of light is highest on the side of the organism facing the light (Fig.
11-26B). This attenuation may be from absorption by pigments within the organism
absorbing over the same wavelength range as the photoreceptor pigment, or though
scattering of the light within the organism by subcellular particles such as mitochondria or
surfaces such as cell walls. It should be clear from a comparison between Fig. 11-26A and
B that the effect of attenuation is opposite for the two mechanisms. For the refraction
mechanism, light attenuation across the organism would diminish the gradient established
through focusing. For the screening mechanism, light attenuation across the organism
would supplement the gradient. This is the basis for the most convincing evidence that the
screening mechanism is used by plants to measure light direction. Using inhibitors,
genetic mutants, and exogenous dyes, the absorption-based screening has been varied,
and found to be directly related to the capacity of the plant to respond phototropically. (107)
Had the plant used the refraction mechanism, the relationship between absorption and
phototropism would have been inverse.
In summary, phototropism in both plants and fungi is thought to use a flavoprotein as
a photoreceptor pigment. The end result of the transduction sequence in Phycomyces may
be an increase in chitin synthesis, while in flowering plants, light may affect the lateral
transport of IAA, which in turn regulates cell elongation. Although both the flowering
plant shoot and the fungus Phycomyces are positively phototropic, the optical mechanism
for measuring light direction is different for the two cases. Phycomyces uses a refraction
mechanism to establish the gradient in light intensity and measure light direction, while
the gradient in light intensity is established in flowering plants by a screening mechanism.

11.4. ECOLOGICAL SIGNIFICANCE

How light-mediated behavior benefits a species is often obvious by the nature of the
response. There are some situations, however, where the ecology of a species may be
influenced in very subtle ways by the organism's behavior. For example, phototaxis for
the flagellates such as Chlamydomonas is an ecologically important response because it
places the photosynthetic cell in an optimal light environment. On the other hand, one can
argue that it is to the advantage of the Stentor cell to exhibit a step-up photophobic
response, thereby avoiding intense light that could be photodynamically detrimental,
although it is not clear what the advantage is to the cell to contain a powerful photosen-
sitizing dye. (108)
Phototropism permits the plant to orient its shoots and leaves with respect to light.
The leaves contain the photosynthetic machinery on which the plant depends, so it seems
clear that plants with the ability to array that machinery and thereby optimize light capture
in photosynthesis would have a survival advantage.
The vertical migrations of several marine dinoflagellates correspond with the
daylight cycle; the organisms rise near the surface during the daylight hours and descend
at night. (109) If these cells were not phototactically active and remained at some fixed
depth as determined by the physical topography, would they be functionally capable of
reproducing themselves? Would they be able to carry on their photosynthetic processes
efficiently enough to produce the energy "surplus" necessary for fission? This can be
342 Chapter 11

tested and requires further investigation before we can understand the role of phototaxis in
the dynamics of population growth. One can ask other questions of this system as well.
For example, what environmental factors can alter the "normal" photoresponse? How
does temperature affect the photoresponse at the behavioral, physiological, or bio-
chemical level? These questions have not been studied systematically. The interactions
between the two branches of photobiology, photoecology and photomovement, will play
an important role in answering these questions. It is in the examination of ecological
problems that the universal importance of the light behavior of organisms will be appre-
ciated.

11.5. REFERENCES

1. R. M. Macnab and S. -1. Aizawa, Bacterial motility and the bacterial flageller motor, Annu. Rev. Biophys.
Bioeng. 13,51-84 (1984).
2. K. L. Poff, in: Sensory Perception and Transduction in Aneural Organisms (G. Colombetti, F. Lenci, and
P. -So Song, eds.), pp. 299-307, Plenum Press, New York (1985).
3. W. Haupt, in: Sensory Perception and Transduction in Aneural Organisms (G. Colombetti, F. Lenci, and
P. -So Song, eds.), pp. 1-18, Plenum Press, New York (1985).
4. H. Machemer, in: Sensory Perception and Transduction in Aneural Organisms (G. Colombetti, F. Lenci,
and P. -So Song, eds.), pp. 179-209, Plenum Press, New York (1985).
5. R. P. Blakemore and R. B. Frankel, Magnetic navigation in bacteria, Sci. Am. 245, 58-65 (1981).
6. W. Nultsch, Phototaxis and photokinesis, in: Primitive Sensory and Communication Systems (M. J.
Carlile, ed.), pp. 29-90, Academic Press, New York (1975).
7. K. V. Thimann and G. M. Curry, Phototropism and phototaxis, in: Light and Life. (W. McElroy and B.
Glass, eds.), pp. 647-670, The Johns Hopkins Press, Baltimore (1961).
8. W. R. Briggs, The phototropic responses of higher plants, Annu. Rev. Plant Physiol. 14,311-352 (1963).
9. D. S. Dennison, Phototropism, in: Physiology of Movements. Encyclopedia of Plant Physiology, New
Ser., Vol. 7 (W. Haupt and M. E. Feinleib, eds.), pp. 506-566, Springer-Verlag, Berlin (1979).
10. R. D. Firn and J. Digby, The establishment of tropic curvatures Lll plants, Ann. Rev. Plant Physiol. 31,
131-148 (1980).
11. H. Senger and W. R. Briggs, The blue light receptor(s): Primary reactions and subsequent metabolic
changes, Photochem. Photobiol. Rev. 6, 1-38 (1981).
12. K. L. Poff and C. B. Hong, Photomovement and photosensory transduction (yearly review), Photochem.
Photobiol. 36,749-752 (1982).
13. W. Haupt, Light-mediated movement of chloroplasts, Annu. Rev. Plant Physiol. 33, 205-233 (1982).
14. P. Galland and E. D. Lipson, Photophysiology of Phycomyces blakesleeanus (yearly review), Photochem.
Photobiol. 40, 795-800 (1984).
15. U. Pohl and V. E. A. Russo, Phototropism, in: Membranes and Sensory Transduction (G. Colombetti and
F. Lenci, eds.), pp. 231-329, Plenum Press, New York (1984).
16. H. Senger (ed.), Blue Light Effects in Biological Systems. Springer-Verlag, Berlin (1984).
17. B. Diehn, M. E. Feinleib, W. Haupt, E. Hildebrand, F. Lenci, and W. Nultsch, Terminology ofbehav-
ioral responses of motile microorganisms, Photochem. Photobiol. 26, 559-560 (1977).
18. D. -Po Hader, in: Physiology of Movements. Encyclopedia of Plant Physiology, New Ser., Vol. 7 (W.
Haupt and M. E. Feinleib, eds.), pp. 268-309, Springer-Verlag, Berlin (1979).
19. C. Darwin, The Power of Movements in Plants, John Murray, London (1880).
20. D. S. Dennison, in: Advanced Plant Physiology (M. B. Wilkins, ed.), pp. 149-162, Pitman, London
(1984).
21. W. Haupt and M. E. Feinleib (eds.), in: Physiology of Movements. Encyclopedia of Plant Physiology,
New Ser., Vol. 7, Springer-Verlag, Berlin (1979).
22. M. E. Feinleib, in: Sensory Perception and Transduction in Aneural Organisms (G. Colombetti, F. Lenci,
and P. -So Song, eds.), pp. 119-146, Plenum Press, New York (1985).
Photomovement 343

23. N. M. L. Morel-Laurens and M. E. Feinleib, Photomovement in an "eyeless" mutant of Chlamydomon-

as, Photochem. Photobiol. 37, 189-194 (1983).
24. E. Batschelet, Statistical Methods for Analysis of Problems in Animal Orientation and Navigation. AlBS,
Washington, D.C. (1965).
25. E. Batschelet, Statistical methods in the analysis of problems in animal orientation and certain biological
rhythm, in: NASA Symposium on Animal Orientation and Navigation (S. R. Galles, K. Schmidt-Koenig,
G. J. Jacob, and R. F. Belleville, eds.), pp. 61-91, NASA, Washington, D. C. (1972).
26. P. -So Song, D. -Po Hader, and K. L. Poff, Step-up photophobic responses in the ciliate, Stentor
coeruleus, Arch. Microbiol. 126, 181-186 (1980).
27. P. -So Song, D. -Po Hader, and K. L. Poff, Phototactic orientation by the ciliate, Stentor coeruleus.
Photochem. 32,781-786 (1980).
28. P. Halldal, Factors affecting light response in phototactic algae, Physiol. Plant. 12, 742-752 (1959).
29. Th. W. Engelmann, Bacterium photometricum. Ein Betrage zur vergleichenden Physiologie des Licht-
und Farbensinnes, PflugersArch. Ges. Physiol. 30, 95-124 (1883).
30. B. Diehn, Phototactic responses of Euglena to single and repetitive pulses of actinic light: Orientation time
and mechanism, Exp. Cell Res. 56, 375-381 (1969).
31. T. -H. Chang. E. B. Walker, and P. -So Song, Effects of ionophores and ruthenium red on the phototaxis
of Stentor coeruleus as measured by simple devices, Photochem. Photobiol. 33,933-936 (1981).
32. E. B. Walker, M. Yoon, and P. -So Song, The pH dependence of photosensory responses in Stentor
Coeruleus and model system, Biochim. Biophys. Acta 634, 298-308 (1981).
33. K. Bergman, P. B. Burke, E. Cerda-Olmedo, C. N. David, M. Delbruck, K. W. Foster, E. W. Goodell,
M. Heisenberg, G. Meissner, M. Zalokar, D. S. Dennison, and W. Shropshire, Jr., Phycomyces, Bac-
teriol. Rev. 33, 99-157 (1969).
34.W. Hand, in: The Science of Photobiology (K. C. Smith, ed.), 1st ed., pp. 313-328, Plenum Press, New
York (1977).
35. W. G. Hand and D. Davenport, in: Photobiology of Microorganisms (P. Halldal, ed.), pp. 253-282,
Wiley-Interscience, New York (1970).
36. P. -So Song, The electronic spectroscopy of photoreceptors (other than rhodopsin), in: Photochem.
Photobiol. Rev., Vol. 7 (K. C. Smith, ed.), pp. 77-140, Plenum Press, New York (1983).
37. P. -So Song, in: Sensory Perception and Transduction in Aneural Organisms (G. Colombetti, F. Lenci,
and P. -So Song, eds.), pp. 47-59, Plenum Press, New York (1985).
38. W. Nultsch and D. -Po Hader, Photomovement of motile microorganisms, Photochem. Photobiol. 29,
423-438 (1979).
39. W. Nultsch, Der Einfluss des Lichtes auf die Bewegung der Cyanophyceen. 1. Mitt. Phototopotaxis von
Phormidium autumnale, Planta 56, 632-647 (1961).
40. W. Nultsch, Der Einfluss des Lichtes auf die Bewegung der Cyanophyceen. II. Photokinesis bei Phor-
midium autumnale, Planta 57, 613-632 (1962).
41. W. Nultsch, Der Einfluss des Lichtes auf die Bewegung der Cyanophyceen. III. Photophobotaxis von
Phormidium uncinatum, Planta 58, 647-663 (1962).
42. W. Nultsch, H. Schuchart, and M. Hohl, Investigation on the phototactic orientation of Anabaena
variabilis, Arch. Microbiol. 122, 85-91 (1979).
43. W. Nultsch, Effects of sodium azide on phototaxis of the blue-green alga Anabaena variabilis and
consequences to the two-photoreceptor systems-hypothesis, Arch. Microbiol. 134, 33-37 (1983).
44. W. Nultsch, in: Topics in Photobiology (H. O. Kim and P. -So Song, eds.), pp. 95-107, Jeju National
University, Jeju (1983).
45. F. Lenci, D. -Po Hader, and G. Colombetti, in: Membranes and Sensory Transduction (G. Colombetti and
F. Lenci, eds.), pp. 199-229, Plenum Press, New York (1984).
46. W. Nultsch, in: Sensory Perception and Transduction in Aneural Organisms (G. Colombetti, F. Lenci,
and P. -So Song, eds.), pp. 147-164, Plenum Press, New York (1985).
47. D. -Po Hader, Electrical and proton gradients in the sensory transduction of photophobic responses in blue-
green alga, Phormidium uncinatum, Arch. Microbiol. 130, 83-86 (1981).
48. D. -Po Hader, Extracellular and intracellular determination of light-induced potential changes during
photophobic reactions in blue-green algae, Arch. Microbiol. 119,75-79 (1978).
49. D. -Po Hader, Effects of inhibitors and uncouplers on light-induced potential changes triggering pho-
tophobic responses, Arch. Microbiol. 120, 57-60 (1979).
344 Chapter 11

50. D. -Po Hader and K. L. Poff, Dependence of the photophobic response of the blue-green alga, Phormidium
uncinatum, on cations, Arch. Microbiol. 132, 345-348 (1982).
51. J. Spudich, in: Sensory Perception and Transduction in Aneural Organisms (G. Colombetti, F. Lenci, and
P. -So Song, eds.), pp. 113-118, Plenum Press, New York (1985).
52. E. Hildebrand and A. Schimz, in: Sensory Perception and Transduction in Aneural Organisms (G.
Colombetti, F. Lenci, and P. -So Song, eds.), pp. 93-111, Plenum Press, New York (1985).
53. E. Hildebrand and N. Dencher, Two photosystems controlling behavioral responses of Halobacterium
halobium, Nature 257,46-48 (1975).
54. R. A. Bogomolni and J. L. Spudich, Identification of third rhodopsin-like pigment in phototactic Halobac-
taium halobium, Proc. Natl. Acad. Sci. USA 79, 6250-6254 (1982).
55. H. C. Berg, Chemotaxis in bacteria, Annu. Rev. Biophys. Bioeng. 4, 119-136 (1975).
56. A. Schimz and E. Hildebrand, Response regulation and sensory control in Halobacterium halobium based
on an oscillator, Nature 317, 641-642 (1985).
57. P. -So Song, Protozoan and related photoreceptors: Molecular aspects, Annu. Rev. Biophys. Bioeng. 12,
35-68 (1983).
58. V. Massey, in: Photoreception and Sensory Transduction in Aneural Organisms (F. Lenci andG. Colom-
betti, eds.), pp. 253-270, Plenum Press, New York (1980).
59. M. F. Land, The physics and biology of animal reflectors, Progr. Biophys. 24,77-105 (1972).
60. S. O. Mast, Light and the Behavior of Organisms, Wiley, New York (1911).
61. M. Melkonian and H. Robenek, Eye spot membranes of Chlamydomonas reinhardii: A freeze-fracture
study, J. Ultrastr. Res. 72, 90-102 (1980).
62. W. H. Miller, Ocular optical filtering, in: lnvenebrate Photoreceptors Handbook of Sensory Physiology,
Vol. VIl/6A (H. Autrum, ed.), pp. 69-143, Springer-Verlag, Berlin (1979).
63. A. Lui, Photoresponse of Paramecium provoked by photosensitizing dyes, Bioi. Glasn. Jugosl. 8, 11-32
(1956).
64. L. K. Lozina-Lozinsky, Adaptive behavior of Paramecium caudatum to action of photodynamic dyes,
Acta Protozool. 18, 609-628 (1979).
65. C. J. Brokaw and P. Verdugo (eds.), Progress in Clinical and Biological Research, Vol. 80, Mechanism
and Control of Ciliary movement, A. L. Liss, New York (1982).
66. I. -H. Kim, R. K. Prusti, P. -So Song, D. -Po Hader, and M. Hader, Phototaxis and photophobic responses
in Stentor coeruleus: Action spectrum and role of Ca2 + fluxes, Biochim. Biophys. Acta 799, 298-304
(1984).
67. P. -So Song, D. -Po Hader, and K. L. Poff, Phototactic orientation by the ciliate, Stentor coeruleus,
Photochem. Photobiol. 32, 781-786 (1980).
68. P. -So Song, K. J. Tapley, Jr., and J. D. Berlin, in: The Biology of Photoreception (D. Cosens and D.
Vince-Prue, eds.), pp. 503-520, Cambridge University Press, Cambridge (1983).
69. P. -So Song, E. B. Walker, R. A. Auerbach, and G. W. Robinson, Proton release from Stentor pho-
toreceptors in the excited states, Biophys. J. 35, 551-555 (1981).
70. K. Iwatsuki and P. -So Song, Deuterium oxide (020) enhances the photosensitivity of Stentor coeruleus,
Biophys. J. 48, 1045-1048 (1985).
71. R. K. Prusti, P. -So Song, D. -Po Hader, and M. Hader, Caffeine-enhanced photomovement in the ciliate,
Stentor coeruleus, Photochem. Photobiol. 40, 369-375 (1984).
72. S. J. Britz, Chloroplast and nuclear migration, in: Physiology of Movements. Encyclopedia of Plant
Physiology, New Ser., Vol. 7 CW. Haupt and M. E. Feinleib, eds.) , pp. 170-205, Springer-Verlag, Berlin
(1979).
73. G. Fischer-Arnold, Untersuchungen iiber die Chloroplastenbewegung bei Vaucheria sessilis, Protoplasma
56, 495-520 (1963).
74. E. Schonbohm, Untersuchungen iiber die Starklichtbewegung des Mougeotia-Chloroplasten, Z. Bot. 51,
233-276 (1963).
75. W. Haupt and R. Gartner. Die Chloroplasten-Orientierung von Mesotaenium in starkem Lict, Naturwiss.
53, 411 (1966).
76. F. Mayer, Lichtorientierte Chloroplasten-Verlagerung bei Selaginella martensii, Z. Bot. 52, 346-381
(1964).
77. J. Zurzycki, The action spectrum for the light dependent movements of the chloroplasts in Lemna trisulca
L., Acta Soc. Bot. Polon. 31, 489-538 (1962).
 Photomovement 345

78. K. Seitz, Zur Frage der Jodid-Wirkung auf die Starklichtbewegungen der Chloroplaten von Vallisneria
spiralis ssp. torta, Z. Pflanzenphysiol. 62, 63-69 (1970).
79. W. Haupt, Die Chloroplastendrehung bei Mougeotia. I. Mitteilung: iiber den quantitativen und qualitati-
ven Lichtbedarf der Schwachlicht bewegung, Planta 53, 484-501 (1959).
80. W. Haupt and G. Wagner, in: Membranes and Sensory Transduction (G. Colombetti and F. Lenci, eds.),
pp. 331-375, Plenum Press, New York (1984).
81. E. Schonbohm, in: Blue Light Syndrome (H. Senger, ed.), pp. 69-96, Springer-Verlag, Berlin (1980).
82. W. Haupt, in: Handbook ofPhycological Methods: Development and Cytological Methods (E. Gantt, ed.),
pp. 196-204, Cambridge University Press, Cambridge (1980).
83. P. -So Song, Q. Chae, and J. G. Gardner, Spectroscopic properties and chromophore conformations of the
photomorphogenic receptor: Phytochrome, Biochim. Biophys. Acta 476, 479-495 (1979).
84. W. Rudiger, F. Thummler, E. Cmiel, and S. Schneider, Chromophore structure of the physiologically
active form (Pfr) of phytochrome, Proc. Natl. Acad. Sci. USA 80, 6244-6248 (1983).
85. N. G. A. Ekelund, C. Sundqvist, P. H. Quail, and R. D. Vierstra, Chromophore rotation in 124-kilodalton
Avena phytochrome as measured by light-induced changes in linear dichroism, Photochem. Photobiol. 41,
221-223 (1985).
86. T. R. Hahn, P. -So Song, P. J. Quail, and R. D. Vierstra, Tetranitromethane oxidation of phytochrome
chromophore as a function of spectral form and molecular weight, Plant Physiol. 74, 755-758 (1984).
87. I. -So Kim, E. S. Kim, and P. -So Song, Interactions of phytochrome with detergents and liposomes,
Biochim. Biophys. Acta 747, 55-64 (1983).
88. M. Furuya, J. H. Freer, A. Ellis, and K. T. Yamomoto, Electrostatic binding of proteins and phytochrome
to differently charged liposomes, Plant Cell Physiol. 22, 135-144 (1981).
89. I. ·S. Kim and P. -So Song, Binding of phytochrome to liposomes and oat protoplasts, Biochemistry 20,
5482-5489 (1981).
90. G. Hermann, A. Hattenbach, and E. Muller, Further characterization of the interaction between phy-
tochrome and liposomes, Biochem. Physiol. Pflanzen 181, 61-67 (1986).
91. G. Wagner and K. Klein, Differential effect of calcium on chloroplast movement in Mougeotia, Pho·
tochem. Photobiol. 27, 137-140 (1978).
92. E. M. Dreyer and M. H. Weisenseel, Phytochrome-mediated uptake of calcium in Mougeotia cells, Planta
146, 31-39 (1979).
93. G. M. Curry and H. E. Gruen, Action spectra for the positive and negative phototropism of Phycomyces
sporangiophore, Proc. Nat!. Acad. Sci. USA 45, 797-804 (1959).
94. G. V. Thimann and G. M. Curry, Phototropism and phototaxis. in: Comparative Biochemistry: A Compre-
hensive Treatise, Vol. 1 (M. Florkin and H. S. Mason, eds.), pp. 243-309, Academic Press, New York
(1960).
95. M. Delbruck and W. Shropshire, Jr., Action and transmission spectra of Phycomyces, Plant Physiol. 35,
194-204 (1960).
96. G. M. Curry, K. V. Thimann, and P. M. Ray, The base curvature of Avena seedlings to the ultraviolet.
Physiol. Plant. 9, 429-440 (1956).
97. W. Shropshire, Jr., and R. B. Withrow, Action spectrum of phototropic tip-curvature of Avena. Physiol.
Plant. 33, 360-365 (1958).
98. B. H. Davies, Carotenoids, in: Chemistry and Biochemistry of Plant Pigments, Vol. 2 (T. W. Goodwin,
ed.), pp. 38-165, Academic Press, New York (1976).
99. P. -So Song and T. A. Moore, On the photoreceptor pigment for phototropism and phototaxis: Is a
carotenoid the most likely candidate?, Photochem. Photobiol. 19, 435-441 (1974).
100. H. Beinert, Flavin coenzymes, in: The Enzymes, Vol. 2 (P. D. Boyer, H. Lardy, and K. Myrback, eds.),
pp. 339-416, Academic Press, New York (1960).
101. R. Matthews and V. Massey, Forms of old yellow enzyme, in: Flavins and Flavoproteins (H. Kamin,
ed.), pp. 329-348, University Park Press, Baltimore, and Butterworths, London (1971).
102. M. Sun, T. A. Moore, and P. -So Song, Molecular luminescence studies offlavins. I. The excited states of
flavins, J. Am. Chem. Soc. 94, 1730-1740 (1972).
103. R. D. Vierstra and K. L. Poff, Mechanism of specific inhibition of phototropism by phenylacetic acid in
corn seedling, Plant Physiol. 67, 1011-1015 (1981).
104. Y. N. Jan, Properties and cellular localization of chitin synthetase in Phycomyces blakesleeanus. J. Bioi.
Chern. 249, 1973-1979 (1974).
346 Chapter 11

105. M. fino and W. R. Briggs, Growth distribution during fIrst positive phototropic curvature of maize
coleoptiles, Plant Cell Environ. 7,97-104 (1984).
106. W. Shropshire, Jr., The lens effect and phototropism of Phycomyces. J. Gen. Physiol. 45, 949-958
(1962).
107. R. D. Vierstra and K. L. Poff, Role of carotenoids in the phototropic response of corn seedlings. Plant
Physiol. 68,798-801 (1981).
108. K. C. Yang, R. K. Prusti, P. S. Song, M. Watanabe, and M. Furuya, Photodynamic action in Stentor
coeruleus sensitized by endogenous pigment stentorin, Photochem. Photobiol. 43, 305-310 (1986).
109. R. B. Forward, Jr., Light and diurnal vertical migration: Photobehavior and photophysiology of plankton,
in: Photochem. Photobiol. Rev. Vol. 1 (K. C. Smith, ed.), pp. 157-209, Plenum Press, New York (1976).
 12
Photosynthesis

12.1. Introduction .................................................................... 348

12.1.1. How Did Photosynthesis Evolve? ....... ,", ................................. 348
12.1.2. An Overview of Photosynthesis ............................................. 349
12.2. The Photosynthetically Active Pigments .............................................. 351
12.2.1. The State of Chlorophyll in Vivo ............................................ 351
12.2.2. Distribution of Pigments in Photosynthetic Organisms ........................... 353
12.2.2.1. Purple and Green Photosynthetic Bacteria. . . . . . . . . . . . . . . . . . . . . . . . . . . .. 353
12.2.2.2. Cyanobacteria (Blue-Green Algae) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 354
12.2.2.3. Red Algae ...................................................... 354
12.2.2.4. The Brown Algae, Diatoms, and Dinoflagellates ....................... 354
12.2.2.5. Green Algae and Higher Plants ..................................... 355
12.3. Light Harvesting and Photochemical Conversion ....................................... 355
12.3.1. Absorption of Light and Fluorescence of Pigments .............................. 355
12.3.2. Energy Transfer .......................................................... 356
12.3.3. The Reaction Center ...................................................... 357
12.3.4. Delayed Fluorescence ..................................................... 357
12.3.5. The Photosynthetic Unit ................................................... 358
12.4. Anoxygenic Photosynthesis of Green and Purple Bacteria ............................... 359
12.4.1. Purple Bacteria. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 359
12.4.1.1. Purple Sulfur Bacteria (Chromatiaceae) .............................. 360
12.4.1.2. Purple Nonsulfur Bacteria (Rhodospirillaceae) ......................... 360
12.4.2. Green Bacteria ........................................................... 362
12.4.2.1. Green Sulfur Bacteria (Chlorobiaceae) ............................... 364
12.4.2.2. Green Gliding Bacteria (Chloroflexaceae) ..... . . . . . . . . . . . . . . . . . . . . . . .. 364
12.5. Oxygenic Photosynthesis of Cyanobacteria, Algae, and Higher Plants ..................... 365
12.5.1. Photosystem I .......... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 365
12.5.1.1. P700 ........ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 366
12.5.1.2. Primary Acceptor of Photosystem I .................................. 367
12.5.1.3. Ferredoxin and Ferredoxin-NADP+ Reductase ........................ 367
12.5.1.4. The Cytochrome blc Complex ...................................... 367
12.5.2. Photosystem II ........................................................... 368
12.5.2.1. P680 ...... , .......................... '" ....................... 369
12.5.2.2. Pheophytin ...................................................... 369
12.5.2.3. QA and QB ..................................................... 370
12.5.2.4. Plastoquinone.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 370
12.5.2.5. Donors to P680 .................................................. 371
12.5.2.6. Cytochrome b559 • . • . • . . • • • . . . . . . . . . . • • . . . . . • . . . . . . . . . . . • . • . . • . . . . 371
12.5.2.7. Oxygen Evolution ................................................ 371
12.6. Photosynthetic Phosphorylation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 373
12.6.1. Noncyclic and Cyclic Photophosphorylation ................................... 373
12.6.2. The Chemiosmotic Hypothesis .............................................. 373

David C. Fork • Department of Plant Biology, Carnegie Institution of Washington, Stanford,

California 94305.

347
348 Chapter 12

12.7. Structural Organization of Photosynthesis ............................................ 375

12.7.1. The Chloroplast .......................................................... 375
12.7.2. Lamellae, Grana, and the Stroma ........ " ., ........................ " ...... 375
12.7.3. Granal and Agranal Chloroplasts ............................................ 376
12.8. Carbon Dioxide Fixation .......................................................... 377
12.8.1. The Calvin-Benson-Bassham or C3 Carbon Cycle .............................. 377
12.8.2. The C4 Cycle ............................................................ 377
12.8.3. Crassulacean Acid Metabolism (CAM) ....................................... 379
12.8.4. Photorespiration .......................................................... 379
12.9. Photosynthesis as a Source of Energy ............................................... 380
12.10. Looking Ahead .................................................................. 381
12.11. References ..................................................................... 382

12.1. INTRODUCTION

Photosynthesis, the conversion of light energy into stabilized chemical energy, involves
the absorption of light by a pigment, energy transfer, energy trapping or stabilization by
reaction centers, and the initiation of chemical reactions from donor to acceptor mole-
cules. The process continues with a sequence of oxidation-reduction reactions that com-
pose the electron transport that leads to the formation of reduced nicotinamide adenine
dinucleotide phosphate (NADPH) in plants or reduced NAD (NADH) in bacteria and
adenosine triphosphate (ATP). Reactions leading to the fixation of carbon are powered by
the energy available in the molecules of NADPH (or NADH) and ATP.
The study of photosynthesis is receiving ever-expanding attention. The Sixth Interna-
tional Congress held in Brussels in 1983 was attended by more than 1100 scientists. The
proceedings of this congress alone resulted in the publication of the proceedings in four
volumes containing over 3400 pages and authored by more than 1200 contributors. (1) The
Third Congress held three years earlier in Greece produced six volumes(2) and earlier
congresses also yielded a wealth of new results.(3-6)
Photosynthesis is a multifaceted subject that can be studied from many points of
view, including those of the physicist, physical chemist, biochemist, biophysicist, biolo-
gist, ecologist, and agronomist. This chapter is an introduction to a complex process that
is both challenging and important to study and understand. A number of reviews and
books have appeared which, in addition to the above-mentioned works, can be consulted
for more detailed information than can be presented here. (7-25)

12.1.1. How Did Photosynthesis Evolve7(26,27)

Conditions on the primitive earth were drastically different from those that exist
today. The Russian biochemist Oparin(28) suggested that no 02 existed in the atmosphere,
but rather hydrogen and methane were the dominant components with lesser amounts of
ammonia, nitrogen, hydrogen sulfide, and water being present. When a mixture of these
substances is exposed to UV radiation, radioactivity, heat, or electric discharges, a wide
variety of substances can be produced, such as HCN, sugars, amino acids, peptides,
nucleosides, and nucleotides, as well as polyphosphates and porphyrins. (29,30)
Since no living organisms would have been around to metabolize these substances on
primeval earth, it is conceivable that substantial amounts could have accumulated over
Photosynthesis 349

long periods of time. When amino acids are heated they fonn proteinlike molecules that
can aggregate into small spheres about the size of bacteria. Nobody knows just how living
fonns may have arisen from such aggregations, but they may have appeared as early as 4
billion years ago. (3l ,32)
These primitive organisms could obtain all the energy they required simply by
utilizing substances that were found abundantly in the environment. With an increase in
the number of such "living organisms," the supply of these compounds would eventually
run out. Compounds from which it is easy to obtain energy would be used (oxidized) firSt.
Survival depended on the primitive bacterialike cells being able to oxidize the remaining
compounds that were more difficult to oxidize, such as H2S and H20. In order to do this,
these "cells" would be greatly assisted if a method could be developed by which the
energy in light could be used. Thus, a primitive photosynthetic system probably evolved
that could utilize light energy to oxidize an electron donor and thereby synthesize an
energy-rich substance for later metabolic use. Since water is one of the most difficult
compounds to oxidize, the ability of an organism to accomplish the photosynthetic oxida-
tion of water granted a great evolutionary advantage over other life fonns, since it would
no longer have to depend on the availability of prefonned energy-rich compounds.
The Oparin idea of the origin of life in the primitive ocean has been challenged by
Woese,(3l,32) who proposed that life arose in the atmosphere of early earth because the
extremely hot surface of the earth had almost no liquid water. Violent stonns could cause
dust to be carried into an atmosphere rich in CO 2 and thought to contain appreciable
quantities of water vapor and some methane, hydrogen sulfide, ammonia, as well as N2.
This atmosphere could act like a chemical reflux column where droplets containing many
absorbed and dissolved substances would provide the necessary surface area for a "mem-
brane" type of chemistry to occur. Macromolecules would partition at interfaces where
some types could absorb and convert solar energy into chemical energy. The synthesis of
these molecules would tend to be autocatalytic. Woese(31,32) proposed that organisms
capable of metabolizing methane (methanogens, one of the most primative known orga-
nisms) brought about the fonnation of water from the large amounts of CO 2 and H2 in the
atmosphere. .
Photosynthesis using H20 as the donor of electrons to reduce carbon to carbohydrate
and releasing 02 as a waste product appears to have evolved very early in the earth's
history. The fossil record shows that the 02-evolving cyanobacteria (blue-green algae)
appeared about 3 billion years ago. The ability of the blue-green algae to use H 20 as their
electron donor gave them great flexibility in the habitats that they could occupy and
provided them with a great evolutionary advantage over photosynthetic bacteria that were
confmed to areas where they could obtain simple organic compounds or H2S. Oxygen-
evolving photosynthesis became the dominant type of energy-converting system of plants,
giving rise to the accumulation of 02 in the atmosphere and, eventually, to the protective
ozone (03) layer in the upper atmosphere via the photochemical alteration of 02' thereby
forever changing the course of evolution on earth.

12.1.2. An Overview of Photosynthesis

The time scale for photosynthesis is large. The absorption of a photon produces the
exicted state in photosynthetic pigments in about a femtosecond (1 fs = 10 - 15 s) and
350 Chapter 12

charge separation occurs in about 1 picosecond (1 ps = 10 - 12 s). The enzymatic reactions

of CO2 fixation and cellular synthesis take place within seconds or minutes. The very first
step in photosynthesis may be considered to be the light-harvesting act. The absorption of
a photon raises chlorophyll (Chl) to an excited state, whereupon it may lose its energy by
radiationless transition to the ground state, by emitting fluorescence (having a longer
wavelength than the photon absorbed), by transferring its energy to another pigment, or
by inducing photochemistry. It is the latter process, in a specialized ChI protein termed a
reaction center, that brings about the conversion of light to chemical energy. Upon
receiving excitation energy, the reaction center ChI expels an electron and becomes
oxidized. The electrons expelled by reaction centers are used to reduce compounds termed
primary acceptors. The reaction centers containing a specialized Chl or·bacteriochloro-
phyll (BChl) (or dimers of these pigments) are composed of specific proteins that span the
width of the photosynthetic membrane. Since reaction centers are relatively few in
number, they are associated with light-harvesting complexes (LHC) that either are located
within the photosynthetic membranes, form layers, or are assembled in more complex
structures such as the phycobilisomes of the red and blue-green algae. (33-35) A photon
absorbed by any of the pigments in the antenna is transferred to other pigments having
longer wavelength absorption maxima, and eventually reaches the reaction center where a
charge separation occurs across the membranes. An electron is transferred from the
primary donor to a transient primary acceptor, and from there usually to a stable quinone-
type acceptor. The reduced quinones reduce other electron carriers and eventually transfer
an electron back to the oxidized primary donor of the reaction center, in the case of
photosynthetic bacteria (Section 12.4). Plants and cyanobacteria have two photosystems
(I and II). In photosystem II (PSII), reduced quinones transfer electrons to a different
reaction center, photosystem I (PSI) (Section 12.5). In PSI an absorption of another
photon induces a charge separation across the membrane, the reduction of a different
acceptor (an iron-sulfur protein), and, finally, the reduction of NADP+ to NADPH. The
energy-rich end products, NADPH and ATP, are used in the dark reactions of CO 2
fixation to yield carbohydrates (CH 2 0). Thus, photosynthesis in green plants and
cyanobacteria removes electrons (and hydrogen) from water and adds them to CO 2 ,
leaving O2 as a waste product. The electron flow, either cyclic or noncyclic, induces the
transport of H + across the photosynthetic membrane. The movement of the H + out of the
membrane is coupled to a system that results in the formation of ATP.
Bacterial photosynthesis uses only one light conversion step that is connected to an
electron transport sequence involving several components whereby an electron is trans-
ferred back to the oxidized primary donor of the reaction center, as noted above. This
cyclic flow also induces H + transport across the photosynthetic membrane, which is
coupled to ATP synthesis. In addition to the cyanobacteria, certain species of bacteria,
such as the green and purple bacteria, can reduce CO2 to CH 2 0. Green and purple
photosynthetic bacteria are unable to oxidize water, but instead must use compounds of
low redox potential such as H2 S, H2 , and Na2 S2 0 3 as their electron (and proton) supply,
and thus do not evolve 02.
An equation was provided by Van NieJC36) as a general formulation of photosynthesis
in plants and bacteria, by which oxidizing and reducing power produced by the pho-
tochemical steps leads to the subsequent oxidation of a hydrogen donor and the reduction
of CO 2 :
Photosynthesis 351

light

Chi or BChl

For green plants and cyanobacteria H2A represents water, and for photosynthetic
bacteria H2A either represents an inorganic sulfur compound such as H2S, organic acids,
or other organic hydrogen donors. Measurements(37,38) with oxygen isotopes are in agree-
ment with Van Niel's concept that the 02 released in photosynthesis originates from water
and not CO 2,(32,33)

12.2. THE PHOTOSYNTHETICALLY ACTIVE PIGMENTS(39-41)

The light reaching the earth's surface has a wavelength distribution extending from
about 290 to 1100 nm. Biologically injurious UV light below about 290 nm is screened
out of sunlight by the ozone layer, while wavelengths above about 1100 nm are absorbed
by water in the atmosphere. As outlined below, there are pigments in the different types of
photosynthetic organisms that absorb over this entire wavelength range.

12.2.1. The State of Chlorophyll in Vivo(42)

Chemically, Chi consists of four pyrrole rings joined to fonn a flat ring that is
complexed with magnesium in its center (Fig. 12-1A). Removal of the Mg and substitu-
tion of H produces a substance tenned pheophytin (Ph). The tetrapyrrole ring of ChI has
attached to it a long and almost completely saturated hydrocarbon phytol tail. The flat
head part of the ChI molecule is more soluble in water (hydrophilic), whereas the phytol
tail is more fat-soluble (hydrophobic).
ChI a has the CH 3 side chain in ring II in the position of X (Fig. 12-1A), while ChI b
has the CHO group in this position. In bacteriochlorophyll a (BChl a) the CH 3 group
occurs in position X, and it substitutes for CH 2 CH3 in ring II. A CH3CO group substitutes
for H2C=CH in ring I, and rings II and IV are reduced. As in ChI a and b, BChl a is
esterified with phytol, but in some species geranylgeraniol is found instead. Some bacteria
contain BChl b, which has its absorption in the near-infrared region in vivo near 1020 nm
one of the longest wavelength absorption bands known for a chlorophyll.
ChI extracted by organic solvents yields absorption spectra that are characteristic of
the species of ChI present and the particular solvent used. (43) If one measures the absorp-
tion spectra of intact material, the absorption spectrum of ChI is complex, being com-
posed of several bands, each absorbing at slightly different wavelengths. These so-called
fonns of ChI apparently result from the association of Chi with specific protein molecules.
The ability of plants to elaborate Chi-protein complexes absorbing at many different
wavelengths is a very important feature of the photochemical apparatus, for a photon will
be absorbed and transferred by a Chi having a short wavelength maximum to a Chi having
a longer wavelength maximum (Section 12.3.2). In addition, the so-called accessory
pigments such as ChI b and ChI c, the red and blue phycobilin pigments of red and blue-
green algae, as well as certain carotenoids can absorb short wavelengths of light and
352 Chapter 12

A B

CH3

,.;
,.; CH 3

,.;
,.; CH 3

,.;
,.;
CH 3
,.;
,.;
CH3
,.;

(3- Carotene

CH3~HNH 0

CH3CH ~

CH 3 --
NH

Chlorophyll

Phycoerythrobilin

Fig. 12-1. Molecular structures of chlorophyll (A), J3-carotene (B), and phycoerythrobilin (C).

transfer them to ChI a, which absorbs at longer wavelengths. The different accessory
pigments as well as the different associations between ChI and protein thus function as an
effective antenna to collect quanta from a large assemblage of pigments absorbing at
shorter wavelengths, and to transfer them to a reaction center ChI protein that absorbs at
the longest wavelength and initiates the photochemical reaction.
Photosynthesis 353

12.2.2. Distribution of Pigments in Photosynthetic Organisms(39,44,45)

12.2.2.1. Purple and Green Photosynthetic Bacteria
BChl serves as the photosynthetically active pigment in bacteria. Five distinct mo-
lecular species of BChl are known: BChl a, b, c, d, and e. These pigments derive their
characteristic absorption spectra from the various side groups that are attached to several
of the carbon atoms in the structure of ChI a. Most purple bacteria contain BChl a, which
has its red absorption bands in vivo between 800 and 900 nm and its blue band between
350 and 400 nm (Fig. 12-2A). Some species of purple bacteria contain BChl b, which has
its main absorption band in vivo at 800-1020 nm. Photosynthetically active carotenoids,
such as spheroidene, are commonly found in photosynthetic bacteria. The contribution to
absorption of carotenoids can be clearly seen as the triple peaks in the blue region between
440 and 540 nm (Fig. 12-2A).
BChis c, d, or e, whose characteristics are more closely related to higher plant ChI

Rhodopseudomonas A
B Chi

B
Anacystis
(Blue-green)

cUG'e
Phycoerythrin

/ 1 c~ u~
0 MaripellaC
IRed)

Chi 0 Chi c F... co~onlhm Chi a

or.~~
D

Fig. 12-2. Absorption spectra made in vivo for the major kinds of
Chi a
photosynthetic organisms. Spectrum A was measured using chro-
matophores of the nonsulfur purple bacterium Rhodopseudomonas
E
spheroides. Spectrum B was obtained using cells of the cyanobac-
Ulva
terium (blue-green alga) Synechococcus leopoliensis (Anacystis ':Green)
nidulans). Spectrum C was measured using the thallus of the deep-
water red alga Maripelta rotata which grows in submarine canyons
off California at depths up to 30 m. Spectrum D was obtained from
the thallus of the intertidal brown alga Endarachne binghamiae and
spectrum E from the green alga U/va lobata. Spectrum F was
obtained from cells of the yellow-green alga Pleurochloris sp. The
approximate absorption maxima and, in some cases, the range of
absorption of photosynthetic pigments are shown. Car., carotenoid, 350 450 550 650 750 850 950
PC, phycocyanin. (From Fork, unpublished.) Wavelength, nm
354 Chapter 12

rather than BChl, are present in the green bacteria. Thus, the main red absorption band of
the green bacterium Chlorobium is found between 700 and 800, and the blue maximum
occurs between 400 and 500 nm.

12.2.2.2. Cyanobacteria (Blue-Green Algae)

These primitive algae probably evolved from a progenitor of the photosynthetic
bacteria and, like the bacteria, do not have their photosynthetic pigments localized in
chloroplasts as do other kinds of algae and higher plants. Blue-green algae have flattened
sacklike structures termed lamellae or thylakoids extending throughout the cytoplasm of
the cell.
The blue-green algae contain ChI a as their main photosynthetic pigment. The
maximum absorption bands of ChI a occur in vivo near 680 nm in the red and around 440
nm in the blue part of the spectrum. Figure 12-1B shows the absorption curve for cells of
the blue-green alga Anacystis. In addition to l3-carotene (Fig. 12-1C), which has been
found in all photosynthetic plants, the blue-green algae contain myxoxanthophyll, a
carotenoid not found in other organisms. The blue-green algae also contain phycocyanin
(PC), absorbing maximally near 626 nm, and allophycocyanin (APC), absorbing at 650
nm, which serve as "accessory pigments. " PC as well as the pigment phycoerythrin (PE)
belong to a class of water-soluble proteins called the phycobilins (algal bilins). PC, APC,
and PE constitute the major accessory pigments of blue-green and red algae, and are
present in other less well-known algal groups as well. The chromophores of phycobilins
are related to ChI, but here the four pyrrole rings are arranged linearly (Fig. 12-1C). This
arrangement is the fundamental structure for the bilin pigments, so called because they
were first isolated from animal bile. These pigments serve by absorbing green to orange
light, wavelengths not effectively absorbed by ChI.

12.2.2.3. Red Algae

Red algae live largely in marine habitats, and frequently occur in deep waters where
red and blue wavelengths are largely filtered out by the upper water layers. Red algae
containing the phycobilin pigment PE can absorb the remaining green light in deep water
and use it effectively for photosynthesis. Red algae contain, in addition, varying amounts
of the pigments PC and APC. Red algae living in deep water have more PE than PC. The
combination of PE with ChI a permits red algae to absorb almost all the wavelengths in the
visible part ofthe spectrum (Fig. 12-2C). The absorption bands of Chi a in vivo at 440 and
678 nm can be seen, as can the peaks at 500,540, and 566 nm produced by PE absorption.
In addition to l3-carotene, the red algae contain a.-carotene that has been found in no other
plants examined so far, except for certain kinds of green algae (Siphonales). Lutein is the
principal xanthophyll of red algae.

12.2.2.4. The Brown Algae. Diatoms. and Dinojlagellates(46)

Diatoms and dinoflagellates living in the upper ocean layers account for about a third
of the world's photosynthesis. In addition to ChI a, these algae contain ChI c as well as the
characteristic xanthophylls fucoxanthin (brown algae and diatoms) and peridinin (dino-
Photosynthesis 355

flagellates) that serve as antenna pigments for photosynthesis (for a review of carotenoids
functioning in photosynthesis, see Ref. 47). The absorption maxima of ChI c occur in vivo
near 460 and 640 nm, while fucoxanthin and peridinin absorption maxima are near 490
nm (Fig. 12-2D). When in its native state, probably attached to a protein, the absorption
maxima of fucoxanthin and peridinin are shifted by about 40 nm to longer wavelengths,
compared to the absorption maxima of these pigments in organic solvents. Together with
ChI c, these xanthophylls make available a large fraction of the green light that otherwise
would be lost for photosynthetic purposes.

12.2.2.5. Green Algae and Higher Plants

These plants constitute the most commonly seen photosynthetic organisms in every-
day life. In these plants, ChI a is found in conjunction with ChI b, the latter having blue and
red maxima in vivo near 470 and 650 nm, respectively (Fig. 12-2E). Unlike the blue-green
algae, diatoms, and dinoflagellates, the higher plants and green algae do not have special
pigments to absorb light between about 500 and 600 nm, and so reflect these wavelengths
and appear green to the eye. In addition to the universall3-carotene, the higher plants and
green algae contain lutein as their major xanthophyll, and lesser amounts of violaxanthin
and neoxanthin. Unlike xanthophylls such as fucoxanthin and peridinin, the xanthophylls
of higher plants and green algae are not particularly active in absorbing light used for
photosynthesis.
There are certain algae that contain only ChI a. An example is given in Fig. 12-2F for
Pleurochloris (a yellow-green alga in the class Eustigmatophyceae). This spectrum clear-
ly shows the contribution to absorption of only ChI a and carotenoids.

12.3. LIGHT HARVESTING AND PHOTOCHEMICAL CONVERSION(l4,48)

12.3.1. Absorption of Light and Fluorescence of Pigments

In order for a light reaction to take place, it is necessary for a pigment to absorb light.
When a substance absorbs a photon of light, it acquires additional energy, and is described
as being in the electronic excited state. According to the laws of quantum mechanics, the
transition from the ground state to the excited state can take place only via certain discrete
energy states. To be absorbed, the energy content of the photon must be exactly the same
as the difference in energy between two electronic states. For this reason, only certain
wavelengths of light can be absorbed by a particular substance (Chapter 1).
Usually the energy levels of the fIrst excited singlet state and the ground states are not
overlapped by sublevels, as are the fIrst and higher excited states. Therefore, a transition
from the ground to the first excited singlet state requires the absorption of a photon, and
likewise, the transition from the fIrst excited singlet state to the ground state results in the
emission of a photon of light. This light emission is termed fluorescence. Fluorescence
emission comes only from the lowest excited singlet state and not from any higher excited
state because internal conversions (transitions producing heat) are rapid and take place
within a few ps, while fluorescence emission occurs in about a ns (1 ns = 10- 9 s). Also,
small rapid internal conversions (energy losses) occur between sublevels of the lowest
356 Chapter 12

excited singlet state. For these reasons, fluorescence is seen at longer wavelengths than
absorption. The photosynthesis sensitized by the absorption of high-energy blue-light
quanta is no more of an advantage to the plant than is the absorption of lower energy red-
light quanta because the higher energy state produced, e.g., by absorption of blue light, is
degraded rapidly to the lower energy state before the slower photochemical reactions have
a chance to take place.

12.3.2. Energy Transfer(49-52)

Before a quantum that is absorbed by a ChI molecule can be trapped by a reaction
center, it can be transferred from one molecule to another, again and again, much like one
billiard ball hitting another until it eventually falls into a pocket. A photon that is absorbed
by any of the 400 or so ChI molecules that compose the "antenna" molecules for the
photochemical trapping center is rapidly transferred from one ChI molecule to another, and
is eventually trapped by a reaction center; the whole process takes place within 10- 9 s.
Energy transfer can be demonstrated experimentally in a green alga such as Chlorella
by illuminating it with a wavelength of light (i.e., 650 om) that is absorbed in large
measure by Chl b, and by measuring the fluorescence of ChI a. The overlap between the
absorption of red bands of Chl b and ChI a is somewhat like the diagrammatic representa-
tion given in Fig. 12-3. The absorption maximum of ChI b occurs in vivo at a shorter
wavelength (650 nm) than the absorption maximum of ChI a (678 nm). In the above
experiment, only the characteristic fluorescence of ChI a is seen; no fluorescence can be
measured that corresponds to that of ChI b, even though this substance is strongly
fluorescent in solution. In fact, studies done in vivo with Chlorella demonstrate that the
transfer of energy from ChI b to ChI a is nearly 100% efficient. Similarly, the phycobilins
can transfer their energy to ChI with efficiencies ranging from 80 to 90%. By contrast, the
efficiency of energy transfer from carotenoids to ChI range from 20 to 50%, depending on
the particular carotenoid involved.
F6rster(53,54) provided one of the theoretical foundations for understanding the me-
chanics of energy transfer between molecules. He suggested that substantial overlap is
needed between the fluorescence of the donor molecule and the absorption of the acceptor
molecule (Fig. 12-3), and that the transfer rate is proportional to the reciprocal of the sixth
power of the distance between the two molecules, so that the rate of transfer has an
extremely great dependence on the separation between molecules. The rate of transfer also

B A
"
U
Absorption
./ r,
t::
"[i
I"
A
B /i Fig. 12-3. Diagrammatic representation of pigments
"(; Absorption / \Fluorescence having absorption and fluorescence bands close to-
"
;;:::

,, gether as is the case, e.g., with Chi b and Chi a in

\
(; \

.,
t::
.o
green algae and higher plants. Since the fluorescence
Co
(;
.0
" of B overlaps the absorption of A, there is a good
probability that a photon absorbed by B will be trans-
« ferred to A. This can be demonstrated by exciting B
and observing only the fluorescence of A (Section
Wavelength, nm _ 12.3.2).
Photosynthesis 357

depends on the mutual orientation of the molecules (55,56, for reviews see 57,58). For a
probability of 50% for energy transfer to occur from one molecule to another, the calcu-
lated distance between molecules should be of the order of 5 nm. In chloroplasts the
distances are much smaller than this, so that the probability of transfer is very high.

12.3.3. The Reaction Center

The photochemical reactions that convert the energy of an absorbed photon into a
stabilized chemical potential take place in specialized Chi proteins in the photosynthetic
apparatus called reaction centers. All reaction centers are embedded within the membrane
matrix, and receive photons directly by absorption or indirectly from other pigments that
act as an antenna to absorb and transfer their absorbed energy to the reaction centers. An
electron in ChI, elevated from the ground level to a higher state after absorption of a
photon, can be received by a suitable acceptor. The transfer of an electron to an acceptor
causes the latter to become reduced and Chi to become oxidized. ChI can be reduced when
it, in turn, accepts an electron from a suitable donor in an energetically downhill process.
If the acceptor molecule is at a lower electrochemical potential than the excited state of
Chi, this transfer will also be with the thermodynamic gradient (downhill), and proceed
spontaneously.
There are essentially four kinds of reaction centers: those found in the purple and
green bacteria (Section 12.4) and those of PSI and PSII in higher plants, algae, and
cyanobacteria (Sections 12.5.1.1, 12.6.2.1). A common feature of all the reaction centers
is the occurrence of a ChI (or BChl) that acts as the primary electron donor and is
associated with a specific protein. The reaction centers are termed P870 and P840 for
purple and green bacteria, respectively, or P680 for PSII and P700 for PSI in plants,
algae, and cyanobacteria, on the basis of the wavelength of their maximum absorption in
the near-infrared region (e.g., P680 has its maximum absorption at 680 nm). In most
purple and green bacteria a dimer of two molecules of BChI a acts as the reaction center
for P870 and P840, respectively. By analogy to the bacterial reaction center, the P680 in
PSII of plants, algae, and cyanobacteria may be a dimer of ChI a. However, P700 of PSI
is apparently a Chi a monomer, at least in its oxidized form.
Evidence suggests that either ChI (BChl) or Ph (or BPh) acts as the first transient
electron acceptor for the initial charge-separation event that occurs in the reaction center
within about 1 ps. The electron is then transferred within a few hundred picoseconds to a
secondary acceptor that acts to stabilize the separated charge and prevent it from recom-
bining again with the primary donor.
Perhaps the best understood reaction center is that found in the purple bacterium
Rhodopseudomonas viridis. Crystallized reaction centers that are photochemically active
have been prepared from this bacterium (Section 12.4).

12.3.4. Delayed Fluorescence(59-62)

If a plant is transferred to the dark after a period of illumination, it can continue to emit
a faint red glow. This glow is only detectable with sensitive light-sensing equipment, and
was discovered unexpectedly by William Arnold and Bernard Strehler in 1951 in an
experiment intended to demonstrate the light-induced formation of ATP during photo-
358 Chapter 12

synthesis in algae using an extraction of fIrefly tails that glows when ATP is present. They
found that these algae emitted a glow even when no firefly extract was added. It appears that
all photosynthetic organisms emit this "afterglow" or luminescence that has been termed
"delayed fluorescence" (or delayed light emission) to distinguish it from "prompt fluores-
cence," which stops within about 10 - 9 S after the exciting light is turned off.
The two types of emission have the same spectral distribution, indicating that they
both originate from ChI (or BChl). Delayed fluorescence appears to be produced by the
back-reaction of primary (or secondary) electron acceptors with the oxidized Chl that is
formed by light at the reaction centers. This back-reduction of ChI apparently returns it to
its singlet excited state, from which light can be emitted. Moreover, delayed fluorescence
seems to be a phenomenon that is associated predominantly with the reaction center of
PSII (as is prompt fluorescence). This is the case because delayed fluorescence is still seen
in algal mutants that lack the PSI reaction center P700 (see Section 12.5.1.1 for a
discussion of reaction centers and the two photochemical systems). Moreover, delayed
fluorescence in dark-adapted plants shows oscillations like those seen for 02 evolution
(described in Section 12.5.2.7) when a sequence of flashes is provided.

12.3.5. The Photosynthetic Unit

Plants contain large amounts of pigments, but only a small fraction of these pigments
function in the crucial photochemical steps of photosynthesis. Depending on the particular
photosynthetic organism, there can be from 50 to more than 1000 ChI (or BChl) and other
light-gathering pigments that serve as an antenna to funnel excitation energy into one
reaction center. This concept has been known for over 40 years from the experiments of
Emerson and Arnold, (63,64) who measured O2 evolution from the green alga Chlorella
exposed to brief but strong and repetitive flashes of light. They used very short flashes of
about lOJ..Ls (1 J..Ls = 10- 6 s) of red light that were absorbed by ChI, and were able to
excite every photochemical reaction center only once. They measured the amount of 02
evolved per flash as a function of the dark time between flashes. If the intervals between
flashes were suffIciently long, there was enough time for the dark reactions to process the
products made during the light reactions. Ifthe flashes were spaced too quickly, however,
the dark reactions were unable to keep pace, and photosynthesis, as measured by 02
evolution, slowed down. The dark time needed for the enzymes to react with the pho-
toproducts was found to be about 0.4 s at 1°C, and 0.02 s at 25°C.
When the intensity of the flashes was suffIciently high so that every ChI molecule
was excited at least once, then one 02 was evolved for every 2400 Chl excited. This result
was a surprise, since the authors had expected that under optimum conditions (optimum
dark reaction, and all centers turning over) there would have been one 02 for every Chl
present. Thus, the concept of the "photosynthetic unit" was born. The assemblage of
2400 ChI per reaction center enzyme was termed the photosynthetic unit (2400 Chll02).
In later experiments it was found that the minimum quantum requirements (number of
quanta needed to fIx one CO 2 or release one 02) was around 8. It now appears that 2400/8
or 300 ChI molecules cooperate to "process" one photon. This collection of 300 ChI
molecules with its assemblage of dark reactions may now be termed a photosynthetic unit
(300 ChI per reaction center).
In green plants one commonly sees photosynthetic unit sizes of 200-400 molecules
Photosynthesis 359

of antenna ChI for each reaction center, but the antenna for shade plants is often larger.
Photosynthetic bacteria have about 50-100 BChl depending on the growing conditions,
whereas green photosynthetic bacteria that can grow rapidly in dim light can have about
1000 molecules of BChl associated with each reaction center. In green plants that have
two photosystems, PSI and PSII (Section 12.5), it appears that each of these photosystems
has its own particular set of antenna pigments in addition to a larger antenna that can
serve, or partially serve, both reaction centers .
. Calculations show that, even in full sunlight, each ChI molecule in a leaf absorbs
only a few photons per second. The experiments of Emerson and Arnold(63,64) show how
well plants have adapted to this situation by employing a set of collector pigments to
capture and funnel energy to a single specialized ChI that does the photochemical conver-
sion. With this antenna arrangement to concentrate light energy, the dark reactions are
thus not limited by the rate at which photochemical conversions take place.

12.4. ANOXYGENIC PHOTOSYNTHESIS OF GREEN AND PURPLE

BACTERIA(65)

The purple and green bacteria have only one photosystem, and are unable to use
water as an electron donor and produce oxygen (anoxygenic photosynthesis) as do the
cyanobacteria, algae, and higher plants (Section 12.5). These bacteria use substances of
lower redox potential than water, such as H 2 S, H 2 , or simple organic compounds. These
organisms contain BChl in addition to other photosynthetic pigments. The pigment con-
tent of the photosynthetic bacteria can be regulated by the light intensity available during
growth. In low light these bacteria elaborate vast amounts of antenna pigments to better
harvest the few available quanta. In the dark, all of the photosynthetic bacteria can oxidize
stored polysaccharides to obtain energy. The two major groups of photosynthetic bacteria,
the purple and green bacteria, are each subdivided into two subgroups-the purple sulfur
and purple nonsulfur bacteria, and the green sulfur and green gliding bacteria.

12.4.1. Purple Bacteria

Some of the purple bacteria possess flagellar motility which, combined with pho-
totactic and chemotactic responses, allows them to optimize their growth conditions. The
purple bacteria contain BChl a and b in addition to a large variety of carotenoids such as
spirilloxanthin, lycopene, and spheroidene, which gives them a wide range of colors such
as yellow, brown, red, and purple. (66)
The photosynthetic reactions in the purple bacteria occur within intracytoplasmic
membranes that are folded in and are continuous with the cytoplasmic membranes.
Disruption of these bacteria causes these folded regions to break up and form vesicles
(about 30-100 nm in diameter) termed chromatophores.
The BChl a of purple bacteria has its main absorption band in the near-infrared in
vivo between 800 and 950 nm (Fig. 12-2A). Some purple bacteria contain BChl b, which
absorbs at somewhat longer wavelengths (800-1020 nm). All of the photosynthetic
pigments of bacteria (as well as those of algae and higher plants) are bound to specific
proteins that are embedded within the lipid bilayers of the photosynthetic membrane.
360 Chapter 12

Most of the BChl is associated with proteins that act as the light-harvesting antenna to
absorb and pass excitation to the small number of pigment molecules that serve as the
reaction center where primary photochemical reactions take place. In purple bacteria, the
cyclic flow of electrons is linked to proton movement that is coupled to ATP formation.
The enzymes of the Calvin-Benson-Bassham carbon reduction cycle or reductive
pentose phosphate cycle (Section 12.8.1) have been found in nearly all of the photo-
synthetic bacteria. These bacteria are capable of fixing CO 2 when they contain ATP and
NADH that has been obtained directly or indirectly from photosynthetic electron trans-
port.

12.4.1.1. Purple Sulfur Bacteria (Chromatiaceae)

These bacteria, as the name implies, oxidize sulfide or elemental sulfur to sulfate and
fix CO2 under anaerobic, obligately phototrophic (light-requiring) conditions. Molecular
hydrogen can also serve as an electron donor for photosynthetic CO2 fixation. Some of
these bacteria can grow in the absence of sulfide under anaerobic reducing conditions if
simple organic compounds are available. Some examples of purple sulfur bacteria are
Chromatium, Thiocapsa, Thiocystis, and species containing gas vacuoles such as Lam-
procystis and Thiodictyon.

12.4.1.2. Purple Nonsulfur Bacteria (Rhodospirillaceae)

These bacteria are unable to oxidize elemental sulfur to sulfate for phototropic CO 2
assimilation, but instead can use sulfide or thiosulfate, which can be oxidized to sulfate
without the intermediate production of sulfur. The purple nonsulfur bacteria are capable of
photoassimilating a large number of different organic compounds such as succinate,
malate, acetate, pyruvate, and dicarboxylic acids. Many of these organisms can also grow
phototrophic ally using H2 as their electron donor. Examples of purple nonsulfur bacteria
include Rhodospirillum rubrum, Rs. molischianum, Rhodopseudomonas spheroides, Rp.
capsulata, and Rp. viridis.
The great advantage of using photosynthetic bacteria for studies of photosynthesis is
that these organisms have only one photosystem, and purified reaction centers devoid of
secondary electron carriers and antenna pigments can be prepared that can perform the
initial photochemical charge separation reactions of photosynthesis with high efficiency.
With the development of spectroscopic techniques having subpicosecond resolution,
it has become possible to elucidate the initial photochemical events occurring within the
reaction center of photosynthetic bacteria. Upon absorption of a photon of light in the
reaction center of a purple bacterium such as Rp. sphaeroides, there is a decrease of BChl
a absorbance (bleaching) that is maximum near 870 nm. This bleaching results from the
initial charge transfer produced in the bacterial reaction center by the initial primary
photochemical event, which produces P870+ BPh - within a few picoseconds, where
P870 + represents the cationic free radical and BPh - is the anionic free radical of Bph.
The production of BPh - can also be detected spectrophotometrically. Whether BChl
reduction precedes BPh reduction is an open question. P870 apparently exists as a dimer
(BChlh of BChl (or "special pair"). The electron moves next, within several hundred
picoseconds, to a quinone molecule (menaquinone) QA" All of these reactions are almost
Photosynthesis 361

".5 ·1.5
Fig. 12-4. Electron transport pathways A Purple Bacteria B Green Bacteria

in the purple (e.g., Rhodopseudomonas

".0 .1.0
spheroides, A) and green (e.g., Chlo-
robium limicola, B) photosynthetic
~ ·0.5 '0.5
bacteria arranged to show the redox po-
tentials of the reacting components. The ~
upward arrows represent the raising of w
E o
an electron against the thermodynamic
0.5 0.5
gradient to a higher energy level upon
absorption of a photon of light by the ~tiDn Cyt~bc,
1.0 center complex
1.0
reaction center. P870, P840 reaction
center bacteriochlorophylls; BChl, bac-
teriochlorophyll; BPh, bacteriopheophytin; Q. quinone; cyt. cytochrome; FeSR. Rieske iron-sulfur protein; FeS,
bound iron-sulfur centers; Fd. ferredoxin; Fp. flavoprotein or ferredoxin-NADP reductase. (Courtesy of R. E.
Blankenship; see Ref. 78,)

temperature-independent and proceed with high efficiency, indicating that all the compo-
nents are held within close proximity to each other within the confines of the reaction
center.
After the initial reactions, the electron is transferred at a much slower rate (~100 j.Ls)
to a second quinone acceptor (ubiquinone) QB' which becomes reduced to its semiquinone
form (QB -). The absorption of a second photon and electron transfer sequence reduces
QB- to the fully reduced QB 2 - form. In bacteria, protons derived from inside the cell
then convert QB 2- to dihydroquinone (QBH2)' Electrons are subsequently removed from
QBH2 and used to reduce P870 + again by a sequence of reactions that are complex and
involve b- and c-type cytochromes, a ubiquinone, and an Fe-S protein (Fig. 12-4A). Only
the electrons from QBH2 are used to reduce P870 +; the H + remaining is released at the
exterior of the cell. Thus, the cyclic flow of electrons around P870 results in an increased
alkalinity inside the cell (as well as increased negative charge). Proton movement back
into the cell occurs via a coupled ATPase system that generates ATP. An analogous
situation occurs in plants, except that protons are pumped inside the thylakoid as a result
of the light reactions and pass to the exterior through the ATP-generating coupling factor
(Section 12.6.2). In the purple bacteria, the reduced NADH needed for CO 2 reduction is
produced by various pathways from the energy available in ATP or from the elec-
trochemical gradient.
Some species of purple bacteria, e.g., Rp. viridis, use BChl b molecules as their
primary donor (termed P960). A great step forward was recently made by H. Michel and
coworkers who prepared crystallized reaction centers from Rp. viridis(67) and obtained X-
ray analysis(68) of single crystals at a resolution of 3 A (l A = 10- 8 cm). These studies
revealed the locations of four BChl b, two BPh b, one nonheme iron, one menaquinone,
and four heme groups (cytochromes); a second quinone and ubiquinone being partially
lost during isolation. (69) Although the reaction center BChl of Rp. viridis is BChl b instead
of the usual BChl a, it is likely that the structural data of Michel and coworkers has
applicability to other bacteria. It may even have applicability to PSII of plants. (70) The
sequence of events thought to occur(68,69) in preparations of crystalline reaction cen-
ters(67) of Rp. viridis is diagrammed in Fig. 12-5. Here the primary electron donor P960, a
BChl b dimer (BChl b2), is suggested to pass an electron to BChl b to form BChl b - in 1
362 Chapter 12

~ @ rQ,:l
~ 100J.lS ~
\ 230ps
I BPh b I I BPh b I
120 ps Fig. 12-5. Electron transfer pathway and transfer times for the compo-
nents of crystallized reaction centers from the purple nonsulfur bacterium
ISChlbl ISChlbl
Rhodopseudomonas viridis. The location of the components is thought to
,-------<---,
/1ps reflect their relative positions in the reaction center. The reaction center
SChl b . B Chi b containing the dimer of BChl b is indicated as P960. BChl b, bac-
(P960) teriochlorophyll b; BPh b, bacteriopheophytin b; Fe, circled, represents a

1270 ns nonheme iron center; QA, menaquinone; QB, ubiquinone. Whether BChl
b is an intermediate between P960 and BPh b remains to be estab-
4Cyt lished.7 1 (Modified after Ref. 69 by permission.)

ps, and to BPh to form BPh - within 20 ps. The involvement of BChl b in this reaction has
been questioned, since some experiments show no evidence for the formation of BChl b-
prior to the formation of BPh - . (71) Q A reduction follows within 230 ps. Spectroscopic
evidence suggests that only the pathway proceeding to the right, as shown by arrows in
Fig. 12-5, occurs in the reaction center. A strongly bound quinone (QA) is located only
within the "right" branch of the reaction center structure. A more loosely bound quinone,
QB' seems to be located on the "left" side of the structure. X-ray analysis showed that the
closest distance between Q A and QB is 14 A.. The inorganic iron is located midway
between Q A and QB' The function of the BChl and BPh in the left side of the reaction
center is not clear. The oxidized special pair of P960 are reduced by another electron-
transporting chain containing four heme grpups of c-type cytochromes. The closest heme
is 21 A. away from the reaction center, and may account for the "slow" transfer time from
Cyt to P960 (270 ns) compared to the transfer time of a few ps from the special pair P960
to BPh.
The reaction center of Rp. viridis is composed of four different protein components,
one being a c-type cytochrome mentioned previously (molecular mass = 38 kDa) and the
other protein subunits designated as L (light), M (medium), and H (heavy), whose
molecular masses using SDS-PAGE (sodium dodecyl sulfate polyacrylamide gel elec-
trophoresis) are 24, 28, and 38 kDa, respectively. X-ray analysis has revealed rela-
tionships between the components of the reaction centers and these protein subunits. (69,70)
Land M are intertwined within the chromophore region, whereas H is on the opposite,
cytoplasmic side to that of the cytochromes.

12.4.2. Green Bacteria

Except for the gliding green bacteria (Section 12.4.2.2), the green bacteria are
nonmotile. In green bacteria the photosynthetic pigments, as well as most of the carot-
enoids, are located in extrinsic pigment-protein structures termed chlorosomes(72-74) that
are attached to the surface of the cytoplasmic membrane (Fig. 12-6). Green bacteria
contain BChl c, d, and e (sometimes called Chlorobium chlorophylls) in addition to BChl
a. As with the purple bacteria, the green bacteria regulate the amount of BChl and the
numbers of chlorosomes on the basis of the light intensity. The reaction centers located in
the cytoplasmic membranes contain specialized molecules of BChl, which act as electron
 ~
g
~
~
iii"
=

lorQe i ntromembrone~porticle$ o - prolein 11

(mostly IO - 14nm in d iometer,
reoction cenler-non-<:ryslo"iI'Ie
LH-Bchlo complexes?)

Fig. 12-6. A model of the chlorosome and the associated cytoplasmic membrane of a green photosynthetic bacterium (Chlorobium
limicola). (From Ref. 73, reprinted by permission.)

~
364 Chapter 12

donors when they receive energy transferred to them by the antenna carotenoids and by
BChl c, d, and e.

12 .4.2 .1. Green Sulfur Bacteria (Chlorobiaceae)

These nonmotile bacteria are strictly anaerobic and need both H2 S and light for their
growth. Most green sulfur bacteria can use molecular hydrogen while others can use
thiosulfate as an electron donor for CO 2 reduction. Some members of this group can also
use organic compounds such as acetate, butyrate, or pyruvate as electron donors. Unlike
the purple bacteria, the green bacteria seem to use a reductive carboxylic acid cycle
instead of the Calvin-Benson-Bassham cycle (Section 12.8.1) as the main method of
assimilating CO 2 .(75) In Chlorobium limicola and Prosthecochloris aestuarii, commonly
studied representatives of this group, the antenna size is around 1500 BChl molecules per
reaction center(76,77) instead of 50-100 as found in purple bacteria. This large antenna
may have hampered the isolation of pure reaction center preparations from green sulfur
bacteria.
The primary electron donor, P840, of the green sulfur bacteria is BChl a which, as in
purple bacteria, may also be a dimer. Photooxidation of this pigment produces a max-
imum decrease of absorbance at 840 nm. There is some evidence from ps-spectroscopy
that the primary electron acceptor is BChl a. BPh c may serve as the next acceptor and it,
in tum, appears to be oxidized by an Fe-S protein having a midpoint potential of about 550
my(78) (Fig. 12-4B). The green bacteria have been found to contain another Fe-S protein
functioning in electron transport that has a redox potential near 420 mY. There is a very
large variety of b- and c-type cytochromes, both membrane-bound and soluble, in the
green sulfur bacteria, and their participation in cyclic electron flow is not completely
understood. The green sulfur bacteria can reduce NAD + directly, analagous to the reduc-
tion ofNADP+ by PSI (Section 12.5.1.3). A comparison of Fig. 12-4 with Fig. 12-7 will
show that there are more similarities between the electron transport in green bacteria and
PSI than between these bacteria and the purple bacteria.
So far it has not been possible to observe light-induced ATP formation in the green
bacteria. However, a membrane-bound ATPase has been found(79) in Chlorobium, which
suggests that these bacteria may be able to use cyclic electron flow to generate ATP as in
the purple bacteria.

12.4.2.2. Green Gliding Bacteria (Chloroflexaceae)

The filamentous green bacterium Chloroflexus aurantiacus was discovered only in
1977 ,(80) it inhabits hot springs around the world where it lives at its optimum temperature
ranging between 50 and 60°C. Unlike the green sulfur bacteria, the green bacterium
Chloroflexus exhibits gliding motility much like that seen in some species of blue-green
algae (Oscillatoria) and in the flexibacteria.
Under anaerobic conditions, this bacterium grows best photosynthetically when
organic substrates such as acetate, lactate, pyruvate, glycerol, or glucose are available to
serve as electron donors. It may also be able to use H2 S. Under aerobic conditions, the
synthesis of BChl stops and the organism reverts to the respiratory metabolism of simple
organic substrates. It is unable to grow in the dark under anaerobic conditions. Chloroflex-
us seems to have a photosynthetic electron transport much like that found in purple
Photosynthesis 365

rr\U-------------"""--
-1.5 -1.5
p'
-1.0 -1.0
p. FeS x

~
0
-0.5
n~ \
~-A~ -0.5

0. "'"
G- '~Q cytb • __ ' .- 0
E
O
LU hv
0.5 ~ "'" "" "-~ 0.5

H2°_I O.E C.I

1.0 -z 1.0
'P680

Fig. 12-7. The light reactions and electron transport pathways for oxygenic photosynthesis (the so-called Z
scheme) showing excited state redox potentials. The upward arrows represent the light reactions taking place in
the reaction centers. GEC. oxygen-evolving complex; Z. donor to PSII; P680 • reaction center chlorophyll of
PSII; Ph. pheophytin acceptor of PSII; QA. QB. quinone compounds; cyt. cytochrome; FeSR. Rieske iron-sulfur
proteins; PC. plastocyanin; P700 • reaction center Chi of PSI; Ao (Chi a?). Al (quinonelike compound?). early
acceptors of PSI (quinone-chi compounds?); FeSx. FeSB. FeSA. bound iron-sulfur protein acceptors of PSI
where FeSB and FeSA may operate in parallel; Fd. ferredoxin. Fp. flavoprotein (ferredoxin-NADP reductase);
cyclic electron flow around PSI is indicated by the dashed line. A description of the pathways of electron flow in
the cyt b6 /f complex (box) are given in Ref. 170. (Adapted from Ref. 82 by permission.)

bacteria. It contains BChl c in addition to BChl a, and has its antenna pigments within
chlorosomes. Its primary electron donor, a form of BCh!, designated as P870, apparently
exists as a dimer as in other photosynthetic bacteria. Excitation of P870 reduces BPh a.
Two distinct species of menadione are reduced in sequence. Cytochromes of the c-type
link electron flow back to P870 to complete the cyclic electron flow. (78.81)

12.5. OXYGENIC PHOTOSYNTHESIS OF CYANOBACTERIA, ALGAE,

AND HIGHER PLANTS

It is possible that the joining together of an electron transport chain of the type found
in the green sulfur photosynthetic bacteria (which resembles photosystem I; Section
12.5.1) with the type found in the purple photosynthetic bacteria (which has similarities to
PSII; Section 12.5.2) by means of an iron-sulfur protein complex that contained quinones
and cytochromes (Section 12.5.1.4) produced organisms that could ultimately use water
as their oxidant to reduce CO 2 to the level of carbohydrate (Fig. 12-7). After further
evolution of the redox chemistry of the reaction center and of the water-oxidizing system,
such a union would provide the newly evolved organism with two light reactions that
could operate in a concerted way to elevate an electron from -!-0.8 V (the potential of the
H2 0/02 couple) to -0.4 V (the potential of (COiglucose) for a total span of 1.2 V.(82)

12.5.1. Photosystem 1(69)

As mentioned previously, this photosystem has homologies to the green sulfur photo-
synthetic bacteria. Photosystem I (PSI) is able to elevate an electron to a sufficiently low
366 Chapter 12

Appressed (grana) Membranes Unappressed (stromal) membranes

I
H+
'--..----' '--..----' '--..----' '- ¥ ./ ' - - . . - - - - '
Chi alb LHC Photosystem II Cyt IJsIf Photo system [ ATP synthase

Fig. 12-8. The physical organization of PSI and PSII and their associated antenna Chi, electron transport
components, and the ATP-synthase (CF[-CFo) as they may be held in the thylakoid membrane of a green plant.
FNR, ferredoxin-NADP+ reductase; LHC, light-harvesting complex; PQ, plastoquinone. The numbers refer to
the masses (kDa) of the proteins indicated. Other abbreviations used are given in Fig. 12-7. (Adapted by
permission from T. G. Dunahay and L. A. Staehelin.)

potential (-1.0 V) so that the reduced primary electron acceptors reduce iron-sulfur
proteins and ferredoxin which, in tum, reduce NADP+ to NADPH.
The reaction center "core" of PSI is a hydrophobic, membrane-bound pigment
protein complex composed of several polypeptides. Two of these polypeptides, having
molecular masses from 62 to 70 kDa (Fig. 12-8), are thought to contain P700 and the
acceptor compounds (for a review, see Refs. 83 and 84). A light-harvesting complex,
termed LHC-I, (85-87) contains about 40% of the total ChI of PSI. The LHC-I complex
contains some Chl b but much less than LHC-II (Section 12.7.2), which has a Chl alb
ratio of about 3.7 vs. 1.2 for LHC-1. The LHC-II complex contains subunits having
molecular masses of 21 and 23.5 kDa, respectively. Carotenoids are present in both the
core complex of PSI as well as in LHC-1.

12.5.1.1. P700(83)
The primary electron donor of PSI, P700, is a ChI molecule situated within a
specialized protein complex that comprises the reaction center. It was designated P700
because upon its oxidation by light it exhibits a maximum bleaching (absorbance de-
crease) in the light around 700 nm. (88) Detailed flash-induced difference spectra from 250
to 850 nm have been measured(89) that for spinach show negative peaks at 430 and 682 nm
in addition to the 700-nm peak. The differential absorption coefficient of P700 at 700 nm
is 64 mM-I. P700 has a reasonably low redox potential (+0.48 V), and is therefore
rather easy to oxidize chemically (e.g., with ferricyanide) as well as by light. A correla-
tion has been seen between oxidized P700 signals measured optically with those measured
by light-induced ESR (electron spin resonance) spectroscopy. The ESR signal corre-
sponds to a free radical centered at G = 2.0025, has a peak-to-peak width of about 7.5
gauss,(90) and can be seen at low temperature (e.g., 77°K).
Photosynthesis 367

Upon absorbing a quantum of light, P700 reduces an acceptor having a redox

potential of about - 1. 0 V (Fig. 12-7). It is uncertain as to whether P700 exists in the
reaction center as a monomer or a dimer, but most evidence suggests that in its oxidized
state it is in the monomeric form.

12.5 .1.2. Primary Acceptor of Photosystem 1

A flash-induced absorbance decrease at 430 nm, termed P430, was attributed to the
reduction of the primary electron acceptor of PSI. (91) This substance may be a semi-
quinone molecule. It appears that an even earlier acceptor preceeds P430 and, judging
from its spectrum, it appears to be a ChI anion. The reduction of the ChI anion takes place
at cryogenic temperatures in less than 10 ps, (92-95) and P430 becomes reduced within
about 200 ps. Subsequent to the reduction of these primary acceptors, certain iron-sulfur
proteins act as tertiary acceptors and become reduced within about 200 ns.
The primary reactions of PSI share homologies with primary reactions of PSII as well
as with the green and purple photosynthetic bacteria. A common feature of the reaction
center in all these photosystems is the occurrence of ChI-like pigments in monomeric or
dimeric form acting as the primary donor as well as the primary acceptor. Apparently only
ChI a, BChl a, and in a few cases BChl b, function in this capacity as primary reaction
center pigments.

12.5 .1.3 . Ferredoxin and F erredoxin-NADP + Reductase

Ferredoxin is a type of nonheme iron-sulfur protein that has been shown to occur in
all organisms investigated so far, from obligate anaerobic bacteria to plants and animals.
These proteins serve as electron transfer agents for a number of different types of reac-
tions. (96)
Plant ferredoxin has a low molecular mass of 12 kDa and consists of a single
polypeptide chain containing a high percentage of acidic amino acids. There are many
similarities in the amino acid sequences of plant and bacterial ferredoxins, suggesting that
higher plants and photosynthetic bacteria had a common ancestor in the anaerobic bacte-
ria, such as the present day Clostridium.
Ferredoxin is reduced during electron transfer in PSI and it in tum can reduce
cytochrome in the Cyt b 6 /f complex (see below) or participate in the reduction of NADP +
to NADPH (Fig. 12-7). This reaction is the last step of photosynthetic electron transport,
and it requires reduced ferredoxin and a flavoprotein enzyme termed ferredoxin-NADP+
reductase (termed FNR or Fp). This flavin enzyme has been isolated from spinach and
crystallized.

12.5.1.4. The Cytochrome hlc Complex(97,98)

A protein complex containing both b- and c-type cytochromes and an FeS compound
is present in cyanobacteria and plants as well as in purple bacteria and, perhaps, in green
sulfur bacteria as well. In oxygenic (02-evolving) photosynthesis, this complex mediates
electron transport between PSII and PSI, and in purple bacteria the complex functions in
cyclic electron transport. This complex is similar to the well-known complex III, an
ubiquinol-cyt c oxidoreductase that functions in mitochondrial electron transport. (99)
368 Chapter 12

The cyt blc complex functioning in photosynthesis contains two molecules of cyt b 6
and one cyt c (which, in the case of green plants, is replaced by cyt t) and one FeS protein
(called the Rieske FeS protein after its discoverer). Considerable work has been done on
the protein composition as well as the source for the encoding of the subunits of this
complex.(l01,I02) The cyt b6 /f complex also functions as a link in cyclic electron flow
around PSI. The electrons produced by the photoreaction of PSI, if not used to reduce
NADP+, can be passed to this complex (Fig. 12-7). This cyclic electron flow is coupled
to proton translocation and can be used as a means of generating A')iP when the reduction
of NADP+ has ceased.
The cyt b 6 in this membrane-spanning protein seems to be located at the site nearer to
the "outside" of the membrane, while the FeS center and cyt f are located more toward
the inner side of the thylakoid membrane. The function of this protein is to oxidize the
plastoquinol (PQHz) reduced by PSII and reduced plastocyanin (Fig. 12-8). The mecha-
nism of the plastoquinol oxidation is currently under investigation, but a so-called Q-cycle
mechanism(103) seems to be a likely candidate.(I04-I06) A Q cycle brings about a sequen-
tial two-step oxidation of plastoquinol by the cyt b6 /f-FeS complex, giving rise to the
translocation of two protons across the membrane for every electron transferred to P700.
So far Q cycles have been more clearly shown to occur in photosynthetic bacteria and
mitochondria than in plants.

12.5.2. Photosystem II
The function of photosystem II (PSII) is to generate sufficient oxidizing power so
that water can be oxidized and plastoquinone reduced. This oxidation of water and the
concomitant evolution of 0z takes place in distinct protein domains that are tightly
coupled to other separate protein domains where the primary photochemistry of PSII
occurs (Fig. 12-8).
Considerable progress has been made recently on the protein composition of the PSII
complex.(l07,IOB) The central core of PSII contains a 47-kDa pigment-protein complex
that spans the membrane and apparently contains the photochemical machinery of PSII.
This protein holds an array of light-harvesting ChI a molecules and is suggested to contain
P680, the primary electron donor of PSII. This protein is also suggested to contain the
primary electron acceptors of PSII. In addition, the 47-kDa protein may also hold the
secondary electron plastoquinone acceptor Q A (which is apparently complexed with iron)
and the secondary electron donor termed Z (which appears to be a species of plastohydro-
quinol) (Fig. 12-7). The 47-kDa protein has a fluorescence band at 695 om at 77°K, and
shows a low temperature thermoluminescence band (the latter originating by a reversal of
charge separation).
The membrane-spanning 47-kDa protein has three hydrophilic proteins tightly asso-
ciated with it on the inner side of the thylakoid lumen, whose molecular masses are 33,
23, and 18 kDa, respectively. (107) There is a strong possibility that four Mn ions, required
for 0z evolution, may be held at the interface of the 47- and 34-kDa proteins, and that the
water-oxidizing reaction is mediated by the 34-kDa protein, perhaps at this interface. In
contrast, there are suggestions that the Mn ions are on the interface of the extrinsic 33-kDa
protein and an intrinsic 34-kDa protein (also called the Dz protein), and the two are
essential for O 2 evolution. The 23- and 18-kDa proteins may function in some way to
Photosynthesis 369

shield the 02-evolving complex from reductants other than water. The 23-kDa protein
may hold Ca2 + as well as CI- ions, and the 18-kDa protein may bind Cl- ions. These
ions apparently play an essential role in PSll, but more work is needed before a better
understanding of their functioning will be obtained.
Several other proteins are associated with the PSII complex: a 43-kDa protein having
about 20 Chl a molecules (the so-called core antenna of PSII, which has a flourescence
band at 685 nm at 77°K) , and a 29-kDa ChI protein having mostly ChI a and, in addition, a
small amount of ChI b. The light-harvesting ChI alb (LHC-I1) protein contains the bulk of
the pigments that are used for light absorption by PSII. Figure 12-8 portrays diagrammati-
cally the structure of the 02-evolving photosynthetic apparatus as it might be arranged in
the chloroplast thylakoid membrane. There is at present no general consensus on the
composition and identification of many of the polypeptide components. Many recent
studies are aimed at understanding the detailed topography of the proteins in thylakoid
membranes. (109)
Two intrinsic proteins having molecular masses of 32 kDa (lysine-poor) and 34 kDa
(lysine-rich), respectively, are thought to be closely associated with the 47-kDa reaction
center protein. Except for the hypothesis mentioned above, the function of the 34-kDa
protein (perhaps equivalent to the so-called D2 protein) is not well understood. It may,
however, playa role on the oxidizing side of PSII.(l1O-1l2) The 32-kDa protein (equiv-
alent to the D1 protein) contains the secondary plastoquinone acceptor of PSII (QB) and is
known as the herbicide-binding protein (QB) ,(1 13) since it is the site where potent inhib-
itors of photosynthesis such as DCMU [3-(3,4-dichlorophenyl)-I,I-dimethylurea] and
Atrazine bind to this protein and prevent normal electron flow from QA to QB (Section
12.5.2.3). It appears that this is also the protein where anions, particularly bicarbonate,
modulate electron flow. (114) PSII also contains two proteins having molecular masses of
about lO kDa, one of which appears to bind cyt b559 (Section 12.5.2.6).

12.5.2.1. P680
P680 has a high redox potential that has been estimated to be near 1.1 V(l15) (Fig.
12-7). For this reason it is difficult to oxidize P680 chemically, as can be done with P700.
Detailed light-induced difference spectra have been obtained(l16) for P680 that show
major negative bands at 430 and 674 nm. In the near-infrared, a small, broad, positive
peak occurs near 825 nm. An ESR signal attributed to oxidized P680 was seen(l17) that
had a g value of 2.0025 and M of 80. Although it has not been measured, it is likely
that, like the oxidation of P700, the oxidation of P680 takes place within a few pico-
seconds. It is not certain whether P680 exists as a monomer or a dimer.

12.5.2.2. Pheophytin
Pheophytin a (Ph a) has been considered to be the primary acceptor for the electron
provided by the photooxidation of P680,u18) Ph a appears to exist as a monomer on the
basis of its ESR (g = 2.0333, M = 130)(119) and ENDOR (electron-nuclear double
resonance) characteristics. (120) Ph a has been estimated to have a redox potential of -0.61
V. The function of Ph a was detected in experiments where Q A (Section 12.5.2.3) was
reduced chemically in the dark by adding dithionite (Na2S 20 4). Under this condition, the
370 Chapter 12

reversible photochemical activity (of Ph a) could still be observed if an electron donor

were also provided for P680.
The existence of an acceptor prior to Ph a has been proposed. (121) Wraight(l22)
pointed out that a species of ChI a may function as the primary acceptor in PSII. If so,
there would be a conservation of function in all photo systems from bacteria to higher
plants in that only Chl-type molecules (ChI or BChl) serve as the primary photochemical
reactants in photosynthesis. This idea awaits proof that BChl serves as the primary
acceptor in bacteria.

12.5.2.3. QA and QB
QA is the major electron acceptor between Ph and QB' the second quinone electron
acceptor of PSil. The redox potential of QA is around zero volts. There is evidence for
another form of QA having a lower redox potential (-0.25 V). (123) QA is apparently a
protein-bound Fe/quinone complex,(124) since its ESR signal is like that seen in bacterial
reaction centers. If QA is reduced, fluorescence is high and if it is oxidized, fluorescence
is low,025) i.e., fluorescence has a variable yield depending on the redox state of QA' A
more recent unproven concept for the variable fluorescence is that it is a rapid delayed
fluorescence generated upon charge recombination between reduced Ph and P680. (126)
A light-induced absorbance decrease at 550 nm has been used as a measure of
primary reactions in PSII (reduction of QA) and is apparently caused by a bandshift of
Ph. (127,128) QA reduction can also be followed as absorbance increases in the ultraviolet
region near 320 nm. (127) QA is reduced by Ph (or ChI) within 200-400 ps (for a review,
see Ref. 129).
The electron acceptor designated QB is a plastoquinone molecule attached to a 32-
kDa protein that is folded back and forth across the thylakoid membrane 5 or 7 times. This
QB protein accepts two electrons from QA within 600 fLS before it is reduced to QBH2'
Further electron flow involves an exchange of the QB H2 molecule from the plastoquinone
(PQ) pool (Section 12.5.2.3). Several anions (such as formate, acetate, and nitrite) inhibit
this reaction, but bicarbonate ion reverses this inhibition and restores efficient electron
flow from QA to QB and from QB to PQ.(114,129) The QB protein is known as the
herbicide-binding protein since many herbicides such as DCMU and Atrazine bind to this
protein and prevent normal electron flow from QA to QB'
QB requires two electrons for its reduction, whereupon it is apparently released, in its
protonated form, within 1 ms to the mobile PQ pool (see below). (130) It thus functions as a
two-electron gate(131-133) to transfer electrons in pairs to the large pool of plastoquinone
molecules. The reduction of QA by the reaction center is a single electron transfer process.
QB is a form of plastoquinone, since it shows a maximum light-induced absorbance
increase in the ultraviolet region at 320 nm, characteristic of reduced semiquinone, after
the first and third of a series of flashes. (134) Other electron carriers may also function
between Ph and quinones. (135)

12.5.2.4. Plastoquinone
Early experiments by Amesz(136) and by Rumberg et at. (137) showed that in intact
algae there was a large pool of PQ that was reduced by PSII and oxidized by PSI. The size
Photosynthesis 371

of this PQ pool is about 5-10 molecules per PSII reaction center, each capable of storing
two electrons (for a review, see Ref. 138). As noted earlier, doubly reduced QB after
protonation exchanges with a PQ pool (Fig. 12-8). The proton uptake from outside the
thylakoid is too slow (-60 ms) to account for this protonation. Thus, it is believed that the
protonation is derived from the H + obtained from the neighboring protein.

12.5.2.5. Donors to P680(l39)

The nature of the primary electron donor(s) to P680 has not yet been elucidated
completely. A donor, termed Z, is apparently located on the 47-kDa reaction center
polypeptide. Other possibilities cannot yet be disproven. There is also the possibility that
more than one species of Z donates electrons to PSII (for a review, see Ref. 129).
Interestingly, Z has been suggested to be a species of plastohydroquinol, whereas Q, the
primary acceptor (Section 12.5.2.3) of PSII, is the oxidized form of the same PQ. It has
been suggested that a cation quinone radical such as Z + would have a redox potential near
+0.95 V at physiological conditions, while an anion quinone such as Q would be near 0.0
V.(140) An ESR signal (termed signal II) that has the characteristics of an oxidized
semiquinone (perhaps complexed with two protons or a metal ligand to give it a high
redox potential) seems to be produced by a donor to P680. The ESR signal II has been
designated as signal IIvf (very fast), as that seen in intact samples, and signal IIf (fast) as
that seen in samples treated by washing with Tris buffer. (141) Z can transfer its electron to
photooxidized P680 in about 40 ns. (142)

12.5.2.6. Cytochrome b559(l39,143)

Cytochrome b559 undergoes low-temperature oxidation mediated by PSI!. Cyt b559
cannot be the primary donor to PSII; prior to freezing, this cyt can be chemically oxidized,
e.g., by the addition of potassium ferricyanide; then at low temperature, the C550 change
(Ph reduction; Section 12.5.2.2) can still be seen even though cyt b559 is oxidized. Cyt
b559 may function by mediating cyclic electron flow around PSII, which may act as a
protection mechanism for the photochemical system of PSII by preventing the accumula-
tion of the strong oxidizing equivalents that could induce photoinhibition.

12.5.2.7. Oxygen Evoiution Cl39 )

There has been progress recently in opening the black box that contains the mystery
of how plants can cleave water to release electrons, hydrogen, and 02 all under physiolog-
ical conditions. In order to release one molecule of 02' electrons must be removed from
two molecules of water. Since the absorption of one quantum produces one free electron
by the reaction center, it is necessary to repeat this process four times before 02 can be
released. It has long been known that a very short and intense flash of light given to plants
kept in the dark for a sufficiently long time yields no 02 evolution. loliot et ai. (144) and
Kok et ai. (145) provided the key to understanding the mechanism of 02 evolution. In their
experiments, sensitive polarographic techniques were used to study the 02 evolved from
dark-adapted algae illuminated with successive brief flashes of light strong enough to
excite every reaction center. In algae kept in the dark for a long period, the first flash
372 Chapter 12

.\ Yield at steady s~ate

~
O~
'0
1.0 I--I---\---I-.~.~.~
\.
/ . .
.................... .--•
"-

"0
0;
;;:

3 5 9 11 13 15 17 19 21 23
Flash number (n) Fig. 12-9. Relative yield of O2 per flash in
dark adapted spinach chloroplasts in relation to
the steady-state value. (Adapted from Ref. 144
H,O"""'>..
0, with permission of P. Joliot.j

yielded no 02 evolution, and the second flash only a negligible amount. The third flash,
however, yielded the highest amount, and the fourth flash somewhat less (Fig. 12-9). This
sequence continued with the next two flashes showing low yields of 02 evolution fol-
lowed by two flashes having high yields. Interestingly, the peaks of highest 02 yield
occurred after every fourth flash, except for the first peak of O 2 evolution that was seen
after the third flash. Kok et at. (145) proposed a model by which the oxygen-evolving
complex (OEC) can exist in several redox states, the so-called S states; the complex
accumulates 4+ charges from the reaction center of PSII. This complex in its fully
oxidized state (S4 +) oxidizes water releasing a molecule of 02 and returns the system to
the fully discharged state (SO). The intermediate oxidation states, S2+ and S3+, are
unstable and decay in the dark to the stable SI + state. Thus, in dark-adapted algae given
three flashes, O2 evolution is maximum, and continued flashes give maximum 02 evolu-
tion with a periodicity of four. This periodicity continues as a damped oscillation with a
period of four.
Kok's model (45 ) provides a useful framework to explain the mechanism of 02
evolution, and some new results have clarified the nature of the OEC. Part of the difficulty
of studying this system is that it is very labile: it is destroyed, for example, by mild
heating (a few minutes at 50°C), by washing with Tris buffer, or by UV irradiation.
Mn plays an important role in 02 evolution. (146-148) Removal of about four Mn
atoms/400 ChI from the native site leads to the complete inhibition of0 2 evolution. Mn is
not involved in the light reactions of PSII since Mn-deficient samples that can no longer
evolve O2 can still function photochemically if suitable chemical donors are provid-
ed.(l49,150) This result suggests that Mn plays a role in the enzyme involved in water
oxidation. (147)
The protein responsible for some of the binding of Mn in PSII has only recently been
suggested(l51 ,152) to be the extrinsic protein having a molecular mass of 33 kDa; it is still
possible that a contaminant protein was the nonbinding protein in these experiments. Mn
may also be associated with the 34-kDa protein (Fig. 12-8). This Mn protein may repre-
sent the enzyme that accumulates the four charges necessary for 02 evolution. However,
Photosynthesis 373

the possible role of the intrinsic 34-kDa protein in the 02 evolution mechanism needs to
be elucidated by further studies (see, e.g., Ref. 110). It has also been seen that Ca2 + and
CI- ions must be bound to some sites on the OEC in order to have maximum PSII
activity. (107)

12.6. PHOTOSYNTHETIC PHOSPHORYLATION(l53,154)

12.6.1. Noncyclic and Cyclic Photophosphorylation

The job of the two light reactions with their coupled electron transport in oxygenic
photosynthesis is completed with the formation of NADPH and the production of ATP.
These two substances are required in order to drive CO2 reduction reactions to form
carbohydrates. We have seen how electron transport leads, as its end result, to the
formation ofNADPH, but this flow yields ATP as well. The coupling of ATP formation
with the one-way transport of electrons to NADP has been termed "noncyclic" pho-
tophosphorylation. It has been shown that under certain experimental conditions ATP is
the only product formed in the light when adenosine diphosphate (ADP) and inorganic
phosphate (Pi) are added to the reaction mixture containing chloroplasts(l55,156) or chro-
matophores. (157) This process is known as "cyclic photophosphorylation."
The light and dark phases of phosphorylation can be separated. (158) After first
illuminating chloroplasts and then transferring them to a solution kept in the dark that
contains ADP, Pi, and Mg2 + , it is still possible to form some ATP. It is clear from this
type of experiment that something is formed in the light, with a lifetime of several
seconds, that can lead to the formation of ATP in the dark.

12.6.2. The Chemiosmotic Hypothesis(l59-161)

Mitchell(160,161) was honored with a Nobel Prize for his "chemiosmotic" hypothesis
whereby the electrochemical potential generated across the photosynthetic membrane as a
result of the light reactions and electron transport gives rise to movements of H + from the
outside to the inside of the membrane. The sum of the pH gradient (~pH), resulting from
the translocated H+ , and the membrane potential (~IjJ), resulting from the gradient of
electric potential across the membrane, is called the proton motive force (pmf).
The sequence of events according to the chemiosmotic hypothesis begins with the
02-evolving system located on the inside of the thylakoid, which releases one proton for
each electron extracted from water by PSII. The reduction of Q A by PSII requires the
transfer of one e-. QB' which acts as a two electron "gate," transfers 2e- to PQ, and
requires the uptake of 2H+ from outside the thylakoid. Oxidation of plastoquinone is
achieved by removal only of its electrons by the reactions of electron transport. The
protons released from plastoquinoI preferentially diffuse into the inside space of the
thylakoid. ATP formation results when the internal H+ ions pass out of the membrane
through the ATPase enzyme or "coupling factor" (CFo-CF 1 ATP-synthase; see Fig.
12-8) resulting in the dehydration of ADP and Pi to form ATP.(162-164) Only the CFo part
of the ATPase enzyme is embedded within the membrane. The major portion (termed
eF 1) is extrinsic to the membrane and contains five different subunits.(165)
374 Chapter 12

Boyer(l66) and Slater(167) provided a mechanism to explain how the coupling factor
may work. Boyer's model, often termed the "energy-linked binding change mecha-
nism," suggests that the coupling enzyme contains three reactive sites whose binding
constants for substrates and products change as a result of energy-dependent conforma-
tional changes induced by the energy of protons that pass out through the coupling
factor. (168) There is evidence for localized domains of H + in chloroplasts. Protons in-
volved in coupling are not always in equilibrium with the protons in the surrounding bulk
water phase enclosed by the thylakoid.(l69-171)
There are many results that favor the Mitchell hypothesis. For example, it is known
that in order to have ATP formation it is necessary to have a membrane that maintains a
separation between an inner and an outer space so that differences in ion concentration can
be achieved. Water movement as a result of such ion transport gives rise to volume
changes in chloroplasts that can be detected as large light-scattering changes. Interesting-
ly, illuminated chloroplasts cause the pH of their bathing medium to rise, presumably as a
result of H+ being taken up by the illuminated chloroplasts. This H+ disappearance is
reversible. The pH of the external medium decreases again (H + goes out) after the
chloroplasts are put in the dark. The separation of the light and dark phases of phos-
phorylation can be explained by the generation of a high internal H + concentration
produced by the light reactions. In the dark, these H + pass out through the coupling factor
giving rise to ATP formation.
Even more interesting, and perhaps providing more support for Mitchell's hypoth-
esis, are experiments(l72) showing that the action of light could be substituted for by a
dark soaking of chloroplasts in a medium at pH 4, so that the concentration of H + inside
the chloroplasts was increased. Raising the pH to a value of about 8 in the presence of
ADP and Pi also gave ATP formation.
The Mitchell hypothesis explains the experimental observation that electron transport
is coupled to phosphorylation in that the rate of electron transport is controlled by the rate
at which ATP formation can take place. If phosphorylation is "uncoupled," then electron
transport can proceed at a much faster rate. Many different types of substances act as
uncouplers. Uncouplers act according to the chemiosmotic hypothesis by permitting the
passive movements of H+ and other ions so that the pH and/or the electrochemical
gradients are dissipated. The uncoupling action of ammonia or of an amine such as
methylamine, apparently, results from the ability of these compounds to diffuse freely
through the membrane and combine with the H + on the inside space of the thylakoid to
form NHt or the positive amine ion.
There are a number of antibiotic-like compounds, termed ionophores, that act as ion-
translocating substances. (173 .174) The antibiotic nigericin allows the exchange of H + for
K + and other alkali metal cations, or vice versa, thus affecting phosphorylation but not
diminishing the electric charge across the membrane that results from the unequal ion
concentration on the inside and outside spaces. Addition of nigericin and K + to chlo-
roplasts allows the H + inside the thylakoid to be replaced by K +. Another antibiotic,
gramicidin, and the compound carbonylcyanide m-chlorophenylhydrazone (CCCP) appar-
ently make "holes" in the membrane, allowing free diffusion of H+ in and out, thus
abolishing the charge across the membrane and uncoupling phosphorylation. The antibiot-
ic valinomycin causes membranes to be permeable to ions such as K + or NHt , but not to
H+. Thus, a membrane potential will either be formed or dissipated depending on the
direction of electron flow.
Photosynthesis 375

12.7. STRUCTURAL ORGANIZATION OF PHOTOSYNTHESIS(l75,176)

12.7.1. The Chloroplast

Except for photosynthetic bacteria and blue-green algae, the higher plants and other
algae have their photosynthetic apparatus localized in deftnite structures termed ch~o­
roplasts. Chloroplasts vary widely in shape, depending on the plant species, from the cup-
shaped structures in Chlorella, to spiral bodies in Spirogyra, to star-shaped structures in
other algae. Spinach chloroplasts are flattened spheres with a diameter of about 5-7 /-Lm.
A chloroplast is surrounded by a continuous double membrane. The outer membrane
is connected to the endoplasmic reticulum. Apparently the inner part of the double
membrane is not connected to other membranes of the cell but is involved with the
formation of the extensive lamellar system found inside the chloroplasts. This envelope
acts as a selective barrier to the passage of metabolites in and out of the chloroplast.

12.7.2. Lamellae, Grana, and the Stroma(l77,178)

A characteristic internal feature of higher plant chloroplasts is the occurrence of a
complex structural arrangement of lamellae (or thylakoids). Frequently, the lamellae are
closely layered one upon the other to form structures termed grana, which look somewhat
like stacks of pancakes. Portions of the lamellae may extend out from the grana stacks into
the surrounding space, called the stroma, which is composed of a proteinaceous matrix.
The light reactions and coupled electron transport are localized in the lamellae and grana
structures. The products of the light reactions, ATP and NADPH, diffuse out of these
structures into the stroma where the enzymes of CO 2 fixation are located.
Chloroplasts can be prepared with and without their surrounding double membranes.
The so-called class I chloroplasts, with intact surrounding membranes, retain their soluble
stroma components and can fix CO 2 at high rates. However, the class II chloroplasts,
which lack surrounding membranes, lose their soluble components from the stroma and
require the readdition of these lost components in order to fix CO 2 ,
The stroma also contains ribosomes and DNA, which are involved in chloroplast
regulation and replication. Starch grains are also commonly seen in the stroma region, as
are globules containing the lipophilic chloroplast quinones, such as plastoquinone, vi-
tamin K 1 , and both a-tocopherol and a-tocopherylquinones. These globules do not playa
direct role in photosynthesis but appear to serve as a storage pool for these substances.
Since grana stacks are found in almost all higher plant chloroplasts, it would seem
that they play some significant role in these plants. All the factors that lead to grana stack
formation are not understood, and chloroplasts lacking grana show all the photochemical
activities thus far measured in chloroplasts containing grana. For example, the blue-green
algae carry on complete photosynthesis, including 02 evolution, yet have no grana stacks,
but only have unstacked lamellae extending throughout the chloroplasts. Studies compar-
ing photochemical activities have shown that PSI appears early in the development of
chloroplast lamellae. The appearance of PSII activity seems to be correlated with the later
formation of grana.
The grana and stroma lamellae of chloroplasts of higher plants, such as spinach, can
be separated from each other by mechanically disrupting the chloroplasts. Differential
centrifugation yields a dense fraction containing almost pure grana and a less dense
376 Chapter 12

fraction consisting of intergrana lamellae (or stroma lamellae). The intergrana lamellae
have a high ChI alb ratio (about 5-6), very low ChI a fluorescence, and a high concentra-
tion of P700. This fraction does not evolve 02 or show other reactions that are associated
with PSII.
In contrast to the less dense intergrana lamellae that contain mostly PSI, the denser
grana fraction contains both PSI and PSII. This fraction has a low ChI alb ratio (2.5), has
a high ChI a fluorescence, evolves 02' and has cyt bss9 and P680.
There are at least five supramolecular protein complexes held within the thylakoid
membranes: the PSI and PSII pigment-protein complex, the cyt blc complex, the cou-
pling factor, and the light-harvesting ChI-protein complex. Most of the latter complex,
termed the ChI alb light-harvesting complex (LHC-II), belongs to PSII and contains
nearly all of the ChI b. There is evidence that PSI has specific ChI alb proteins also (LHC-
I; Section 12.5.1).
The supramolecular complexes are embedded in the lipid bilayers that compose the
thylakoid membranes and are composed of about 50% lipid. About 80% of the lipids of
higher plant thylakoids is composed of neutral galactolipids monogalactosyldiacylglycerol
(MGDG) and digalactosyldiacylglycerol (DGDG) whose acyl chains are composed of
linolenic acid, a fatty acid having a high degree of unsaturation which is very fluid at
physiological temperatures.o 79)
As stated earlier, most of the psn complexes and the PSII light-harvesting complex-
es are located in the appressed grana membranes with only about 10-20% in the stroma-
exposed membranes. (177 .178) It is considered likely that the PSII units in the stroma
lamellae are different from those in the grana region. Most of the PSI complex as well as
the coupling factor complex is found on the stroma thylakoids. The cyt b/c complex, by
contrast, seems to be rather evenly distributed between stroma and grana thylakoids.
This uneven distribution of the photosystems between grana and stroma suggests that
there must be some mobile substance capable of transferring electrons from PSII to PSI.
Plastoquinone has been suggested to be one of the two mobile electron carriers that can
shuttle electrons from PSII to PSI by migrating within the fluid matrix of the thylakoid
membrane. Plastocyanin is the other carrier that plays the role of an electron shuttle. Since
most of the light-harvesting chlorophylls of PSII are separated from PSI, it would seem
that extra light energy absorbed by PSII cannot be made available to PSI. The redistribu-
tion (or "spillover") of energy from PSII to PSI has been postulated(l80) to be brought
about by a mechanism involving a reversible phosphorylation of the LHC-II, which
increases their negative charge causing them to move out of the grana toward the PSI-
containing stromal regions, thereby allowing more of the light absorbed by PSII to reach
PSI. Many differentially pigmented plants are able to redistribute their absorbed quanta
between PSII and PSI. However, not all plants use a mechanism involving phosphoryla-
tion of LHC-II. In the red and blue-green algae, a different mechanism involving cyclic
electron flow in PSI is apparently involved (for a review, see Ref. 181).

12.7.3. Granal and Agranal Chloroplasts

Some plants that possess the so-called C4 -dicarboxylic acid pathway for CO 2 fixation
(Section 12.8.2) have two different types of chloroplasts, depending on their location in
the plant. In leaves of a C4 plant, e.g., maize (Zea mays), the cells that surround the
Photosynthesis 377

vascular bundles contain almost no grana but only stroma lamellae, while chloroplasts
contained in the mesophyll regions have extensive grana as well as stroma lamellae.
Initially it was suggested that the chloroplasts obtained from vascular bundles that have no
grana would have no PSII activity, by analogy to intergrana lamellae of spinach, and that
the mesophyll chloroplasts of maize would contain both PSI and PSIL A good deal of
evidence now suggests that the agranal bundle sheath chloroplasts do have high PSI
activity and are somewhat deficient in PSII, having perhaps only about 40% of the activity
seen in mesophyll chloroplasts. These results suggest that a direct correlation cannot be
made between the extent of grana stacking and the level of PSII activity.

12.8. CARBON DIOXIDE FIXATION(l82,183)

12.8.1. The Calvin-Benson-Bassham or C3 Carbon Cycle

CO2 fixation requires the NADPH and ATP that is formed by the light reactions of
photosynthesis, and proceeds via a number of dark reactions. In general, there are two
major types of CO 2 fixation: the 3-carbon (C 3 ) and the 4-carbon (C4 ) pathways.
The C3 pathway, or reductive pentose phosphate (RPP) cycle, is also termed the
Calvin-Benson-Bassham cycle after its discoverers. Melvin Calvin was awarded the
Nobel Prize for chemistry in 1961 for this work that elucidated the pathway of carbon in
photosynthesis. The crucial reaction of the C3 pathway is the addition of CO 2 to the 5-
carbon sugar ribulose bisphosphate (RuBP) to form two molecules of phosphoglyceric
acid (PGA). The reducing power of NADPH along with ATP is used to change the acid
group in PGA into the aldehyde group of the 3-carbon sugar phosphoglyceraldehyde
(triose phosphate). Energy storage of photosynthesis is achieved with the formation of this
3-carbon sugar. The primary carboxylase in all photoautotrophs is ribulose-l ,5-bisphos-
phate carboxylase (RuBPCase). Since this enzyme also acts as an oxygenase, it is often

°
abbreviated RuBISCO, where RuBIS stands for ribulose bisphosphate, C for carboxylase,
and for oxygenase. This ubiquitous protein accounts for about a quarter of the protein
content of plants, and is thus the most abundant protein on earth.
The other steps of the Calvin-Benson-Bassham cycle serve to regenerate the initial
CO2 acceptor, RuBP, via a complex series of reactions involving 3-, 4-, 5-, 6-, and 7-
carbon sugar phosphates. Although sugars and carbohydrates are formed as end products
of photosynthesis, other compounds can be formed as well. Substances such as fatty
acids, lipids, amino acids, and organic acids can also be synthesized in photosynthetic
CO 2 fixation. The formation of these substances seems to be partially under the control of
environmental factors such as light intensity and the concentrations of CO 2 and 02' A
better understanding of how these factors interact may lead to the ability to control growth
conditions so as to produce plants that have the desired amount of sugars, lipids, or
proteins.

12.8.2. The C4 Cycle(l83-186)

There are a number of plants widely scattered throughout the plant kingdom that
seem to thrive in hot and arid environments. The vascular bundles in these plants are
378 Chapter 12

Bundle Sheath

Fig. 12-10. The mechanism used by C4 plants

to concentrate CO 2 in the bundle sheath cells
(C02 )bs above the concentration found in the
Leak CO. + O. -"'1t-'\1\1\J\J'-+-<"
atmosphere surrounding the mesophyll cells
(COZ)i' Loss of CO 2 is retarded by the re-
sistance to diffusion (R) presented by the cell
wall. PEPCase, phosphoenol pyruvate carbox-
ylase. (From Ref. 186, reprinted by permis-
sion.)

tightly surrounded by the so-called bundle sheath cells containing chloroplasts that can be
distinguished from mesophyll cells located outside and around the bundle sheath cells
(Fig. 12-10). In these plants the initial fixation of CO 2 does not occur via condensation
with RuBP, as in the Calvin-Benson-Bassham cycle, but via an intermediate cycle
termed the C4 cycle by which CO 2 first combines with the 3-carbon acid phosphoenolpy-
ruvate (PEP) to form the 4-carbon acids, malic and aspartic in mesophyll chloroplasts
(Fig. 12-10). These reactions proceed via the intermediate 4-carbon oxaloacetic acid and
are catalyzed by the enzyme PEP carboxylase. The entire pathway is often called the
Hatch-Slack-Kortschak pathway, after its discoverers. The 4-carbon acids, malic and
aspartic, cannot be converted into sugars, however, until they are decarboxylated (lose the
CO 2 fixed previously) to form the 3-carbon pyruvic acid. By reaction with ATP formed
photosynthetically, the pyruvic acid is again converted into PEP that can again accept a
CO2 molecule. Essentially what this cycle accomplishes is to pump CO2 from the outside
and release it into the inside of the plant. Once inside the plant, the released CO 2 reacts in
the usual manner with RuBP via the Calvin-Benson-Bassham cycle in the bundle sheath
chloroplasts. This may seem like a useless endeavor for the plant especially when it is
considered that an extra ATP must be used to operate this "C0 2 pump." Plants having
the C3 cycle require 3 ATP and 2 NADPH molecules, but C4 plants require 4 molecules of
ATP and 2 of NADPH. As mentioned in Section 12.7.3, C4 plants have a characteristic
anatomy and two quite distinct types of chloroplasts.
It appears that the CO 2 pump of C4 plants is located in the mesophyll cells and
functions efficiently by trapping any CO 2 that diffuses through the stomatal pores on the
epidermis of the leaf. This trapped CO 2 is then released into the interior bundle sheath
cells where it is fixed via the Calvin-Benson-Bassham cycle. The CO 2 pump operates
efficiently to keep the concentration of free CO 2 in the mesophyll cell so low that CO 2
tends to diffuse from the atmosphere, where it is present in concentrations of about 300
ppm (0.03%), into the inside space of the leaf even though the stomates may be almost
closed. The ability to obtain CO 2 even with stomates partially closed means that less
water will be lost, a vital requirement for plant survival in climates that are both hot and
dry. It thus appears that it is worthwhile for the plant to spend the extra ATP needed to
Photosynthesis 379

operate the CO 2 pump, since it enables it to collect sufficient CO 2 but not to lose
substantial quantities of water in the process. In any case, this extra ATP may be rather
easily obtained by cyclic photophosphorylation, since plants living in arid regions usually
receive an ample amount of light.
The ability of C4 plants to photosynthesize and grow in arid regions has important
practical consequences for agricultural production in these regions. Until now, most of the
agricultural effort has been directed to developing crop plants that are suited to temperate
regions. In the future, more effort will undoubtedly have to be devoted to developing crop
plants that are well suited to grow in the earth's vast arid regions. Plants with the C4
pathway may play an important role in this undertaking, since they have higher rates of
photosynthesis at high temperatures and high light intensities compared to C3 plants.

12.8.3. Crassulacean Acid Metabolism (CAM)(187)

Plants having crassulacean acid metabolism (CAM) fix CO 2 in a way somewhat like
that of C4 plants, except that C4 plants use separate compartments for the initial CO2
fixation and the subsequent Calvin-Benson-Bassham cycle, whereas the CAM plants fix
CO 2 into malic acid and store it in a vacuole during the night. The Calvin-Benson-
Bassham cycle operates during the day using the CO 2 released upon enzymatic decarbox-
ylation of malic acid. Thus, the C4 plants separate CO 2 fixation and the Calvin-Benson-
Bassham cycle spatially, whereas CAM plants separate them temporally. Typically CAM
plants, such as those in the family Crassulaceae from which the name is derived, are
adapted to arid environments where water conservation is essential. (188) Typically CAM
plants are from regions (such as South Africa) where the environment has alternating wet
periods and hot, dry periods. The CAM plants are protected by a thick cuticle and a
reduced number of stomates. CAM plants do not avoid photorespiration (see next sec-
tion), but the CO 2 that is produced by photorespiration is not lost, since it is prevented
from escaping from the plant during the day because stomates are mainly closed during
this time.

12.8.4. Photorespiration(189, 190)

In addition to CO2 as the substrate for RuBPCase, 02 may act as an alternate

°
substrate in which case there is the production of phosphoglycolate rather than phos-
phoglyceric acid. Thus, the enzyme is also called RuBIS CO where stands for oxy-
genase, as noted earlier. Photorespiration is an energy-consuming process, and is seen
mainly in C 3 plants and competes with the fixation of CO 2 , Although photorespiration
appears to be a wasteful process that C 3 plants must endure because of the high 02
concentration in the atmosphere, there is evidence that photorespiration may protect plants
from photooxidative damage. (191) If ways were found to eliminate photorespiration, then
it might be possible to produce a considerable increase in crop productivity. C4 plants
such as maize, sorghum, and other important crop plants as well as CAM plants do not
suffer from photorespiration, since the CO 2 is sequestered within the plant where it can
compete effectively with O2 for RuBP. As was noted earlier, the C4 plants are superior to
C3 plants under full-sun, high-temperature, and dry conditions, but C 3 plants are better
under conditions of low light and cool temperatures.
380 Chapter 12

12.9. PHOTOSYNTHESIS AS A SOURCE OF ENERGyo7,192-194)

There is increasing interest in the possibility of using photosynthetically converted

light energy as a means of expanding our energy supply. As has been pointed out by
Hall,(193) the magnitude of energy stored by photosynthesis (about 10 times the world's
annual use of energy) is not generally appreciated. Only about 10% of agricultural
production is used for food. "Waste" products from agriculture and forestry represent an
enormous energy potential. Stalks and chaff from grain crops, sawdust, wood chips, and
paper contain energy trapped by the photosynthetic process. Wood chips can be converted
by pyrolysis into charcoal and numerous valuable hydrocarbon products. Plant material
can be fermented into methane, converted into wood alcohol, or burned efficiently to
release energy.
It has been suggested(95 ) that the sugar (as well as the cellulose) produced photo-
synthetically by cane or sugar beet plants could be economically converted into ethyl
alcohol with almost no energy loss. Alcohol can be added to gasoline to reduce the
demand for petroleum. It may be economically feasible to harvest hydrocarbons directly
from plants rather than obtaining them indirectly, such as by the conversion of carbohy-
drates into ethanol. The rubber tree (Hevea) or guayule (Parthenium), for example, might
serve as a direct source of photosynthetically produced hydrocarbons that could substitute
for petroleum hydrocarbons currently being used for chemicals and for starting materials
for many synthetic reactions. Some algae, such as Dunaliella, inhabit waters of extremely
high salinity, such as the Dead Sea, and produce large quantities of the valuable com-
pounds glycerol and j3-carotene. (196)
In principle, it is also possible to use solar energy trapped by bacteria or hydrogen-
adapted algae to make hydrogen, a nonpolluting and easily transportable fuel. The anaer-
obic purple photosynthetic bacteria do not evolve 02' and since they cannot oxidize water
they must depend on a large variety of organic or inorganic substances for their metabolic
needs. The photosynthetic production of H2 requires that the appropriate bacteria be
supplied with an organic (or inorganic) electron donor, the right amounts of amino acids
(or limiting NHt), and that they be in a condition where there is a high activity of energy
conversion in relation to the cell's biosynthetic needs. Under these conditions, the bacteria
regulate themselves to produce H2 by energy-dependent means. Of course, any large-
scale system based on bacterial H2 production will also require large amounts of inexpen-
sive and easily available electron donors. This requirement may impose a severe limita-
tion on the economic feasibility of this type of system.
The most readily available electron donor is, of course, water, but only higher plants
and algae can extract electrons from water with the operation of their two-pigment
systems. We have seen that PSI can reduce compounds that have redox potentials near
-1.0 V (Section 12.5.1). Since at the physiological pH of 7 the E~ of the hydrogen
electrode is about -0.42 V, there is sufficient potential to produce molecular hydrogen.
But algae do not normally produce H2 with the reductant generated by PSI unless they
have been hydrogen-adapted. Hydrogen adaptation is achieved by keeping algae such as
Chiorella in the dark for 10-12 h under nitrogen or hydrogen. Hydrogen production can
apparently be seen after appropriate adaptation conditions in a large number of variously
pigmented algae, including the marine red and brown algae, in addition to the green algae.
The adaptation procedure apparently activates a hydrogenase enzyme, but this en-
Photosynthesis 381

zyme is inactivated if 02 is present. The photoproduction of hydrogen is mediated only by

PSI, and the 02-evolving system of PSII is not operative under anaerobic conditions. The
substrate for H2 production may be a product of anaerobic metabolism that accumulates
during the adaptation period at a very slow rate in the algal cell. This slow formation of
the substrate for H2 production may be the cause of the low rate of H2 evolution that
usually amounts to only about 10% of the maximum rate of 02 produced photosyntheti-
cally.
To utilize the ability of algae to oxidize water, release 02' and at the same time
produce hydrogen requires considerable reengineering of the photosynthetic machinery. A
separation of incompatible reactions has already been achieved by certain blue-green
algae that fix nitrogen in their resting cells (heterocysts). These special cells provide
physical protection of the nitrogen-reducing enzymes against the destructive effects of 02
by removing it quickly by means of a high respiratory activity in the heterocysts.
It may be possible to isolate from the living cell the various subsystems of photo-
synthesis such as the photochemical reaction centers, the 02-evolving system, and the H2-
evolving system, and use them in vitro. (197,198) For example, it might be possible to
absorb photosynthetic components on appropriate supports so that the products of one
system would be brought close enough to be used by the next enzyme system, and so on

°
until the desired products could be obtained. In order t6 avoid inactivation of a hydro-
genase enzyme by 2, it may be possible to separate across opposite sides of a synthetic
membrane the reactions yielding 02 from those producing H2. This may be accomplished,
for example, by a system employing microencapsulation, but at present our knowledge
about the separation of components of the photosynthetic system is rudimentary at best.
A laboratory scale production of H2 has been demonstrated by coupling the reducing
power produced by chloroplasts with an algal or bacterial hydrogenase. A problem with
this system is that, again, 02 production inactivates the hydrogenase system. This prob-
lem may be alleviated by removing the O2 from the reaction mixture, e.g., with the
addition of glucose and glucose oxidase. Unfortunately, the yield of H2 from these
composite systems is disappointingly low (only a few percent of the average rate of
normal photosynthesis), and these systems are very labile and expensive.
It is clear that the energy stored and the organic matter made available from photo-
synthesis are enormous. Many economic obstacles currently hinder the utilization of this
vast potential, and hopefully these problems can be overcome. Continued research di-
rected toward obtaining a better fundamental understanding of the process will aid in our
efforts to tap this vast energy source.

12.10. LOOKING AHEAD

It may seem at first that photosynthesis has been so intensively investigated that little
remains to be done. However, a closer look into almost any ofthe topics covered here will
show that many rewarding results await future investigators.
A great step forward has recently been made in our understanding of the structure of
the bacterial reaction center with the crystallization of isolated reaction centers from
Rhodopseudomonas viridis. This achievement will allow the development of a better
understanding of how the primary energy conversion of photosynthesis takes place.
 382 Chapter 12

Moreover, since these reaction centers have features similar to those of green plants and
algae, which can cleave water, it is likely that these purified centers can serve as models
for further study of plant reaction centers.
The recent availability of PSII membrane preparations retaining high rates of 0z
evolution have allowed the isolation and the possible identification of the function of
certain membrane proteins that are part of the Oz-evolving complex. The function of
several of the other proteins known to be associated with PSII remains to be elucidated, as
does the role of calcium and chloride ions in the reaction centers of PSII. The manganese-
containing enzyme that functions eventually to oxidize water remains to be isolated and
characterized.
PSII in chloroplasts of higher plants and algae is apparently not homogeneous, but
seems to be heterogeneous on both the oxidizing and reducing sides of the PSII reaction
center. This heterogeneity of PSII comes mainly from measurements of the yield and
. induction kinetics of ChI a fluorescence from PSII, and is currently not completely
understood (for a review, see Ref. 199).
The role of cyt b559 remains an enigma. Some evidence suggests that this cyto-
chrome functions to prevent excess quanta from damaging PSII by mediating a cyclic flow
of electrons around PSII. This idea has yet to be proved conclusive, however.
Electron transport in photosynthesis requires the transfer of electrons and hydrogen
ions across the thylakoid membranes. This flow of electrons is mediated by three integral
membrane-spanning protein complexes, PSI, PSII, and the Cyt b6 /f complex. Future
research is needed to reveal the detailed topography of these proteins, which can be
obtained from determinations of their amino acid sequences, as well as from computer
analysis to predict the folding patterns of the proteins from hydrophobicity measurements
(hydropathy plots). The unraveling of the details of light-driven proton gradient forma-
tion(ZOO) as well as the gaining of a better understanding of the molecular mechanism of
photophosphorylation(ZOI) provide many challenges for future investigators.
The way in which proteins are contained within the lipid matrix of the thylakoid
membrane is not yet clearly understood. The polar lipid monogalactosyldiacylglycerol
(MGDG) constitutes about half of the dry weight of the thylakoid membrane, and yet this
lipid does not form the bilayer configuration that is thought to be important in all biolog-
ical membranes. This major component of the thylakoid membrane may provide the
necessary fluid environment to support the dynamic movement of intermediates involved
in electron transport, and may as well serve to anchor the large intrinsic proteins within
the thylakoid bilayer. More work is needed to elucidate why MGDG is such a major
component of the thylakoid membrane, and how it interacts with proteins and other lipids.
These are just a few of the problems that remain to be answered. As is usually the
case, the answering of one question leads to many more. We have a long way to go before
we understand nature's elaborate process of converting solar energy into stable com-
pounds that are vitally important to all life.

12.11. REFERENCES

I. C. Sybesma (ed.), Advances in Photosynthesis Research, Vois. 1-4, Martinus Nijhoff/Dr. W. Junk, The
Hague (1984).
Photosynthesis 383

2. G. Akoyunoglou (ed.), Photosynthesis, Vols. 1-6, Balaban International Science Services, Philadelphia
(1981).
3. D. O. Hall, J. Coombs, and T. W. Goodwin (eds.), Proceedings of the Fourth International Congress on
Photosynthesis, The Biochemical Society, London (1978).
4. M. Avron (ed.), Proceedings of the Third International Congress on Photosyllthesis, Vols. 1-3, Elsevier,
Amsterdam (1975).
5. G. Forti, M. Avron, and A. Melandri (eds.), Proceedings of the Second International Congress on
Photosynthesis Research, Dr. W. Junk, The Hague (1972).
6. H. Metzner (ed.), Progress in Photosynthesis Research, Vols. 1-3, International University of Biological
Sciences, Tiibingen (1969).
7. R. K. Clayton, Molecular Physics in Photosynthesis, Blaisdell, New York (1965).
8. L. P. Vernon and G. R. Seely (eds.), The Chlorophylls, Academic Press, New York (1966).
9. E. Rabinowitch and Govindjee, Photosynthesis, Wiley, New York (1969).
10. O. V. S. Heath, The Physiological Aspects of Photosynthesis, Stanford University Press, Stanford (1969).
11. R. K. Clayton, Light and Living Matter, Vol. 1, The Physical Part, McGraw-Hill, New York (1970).
12. R. P. F. Gregory, Biochemistry of Photosynthesis, Wiley-Interscience, London (1971).
13. R. K. Clayton, Photosynthesis: How Light Is Converted to Chemical Energy, Addison-Wesley, New York
(1974).
14. Govindjee and R. Govindjee, The absorption of light in photosynthesis, Sci. Am. 231,68-82 (1974).
15. Govindjee (ed.), Bioenergetics of Photosynthesis, Academic Press, New York (1975).
16. J. Barber (ed.), Primary Processes of Photosynthesis, Elsevier, Amsterdam (1977).
17. A. Mitsui, S. Miyachi, A. San Pietro, and S. Tamura (eds.), Biological Solar Energy Conversion,
Academic Press, New York (1977).
18. R. K. Clayton, Photosynthesis: Physical Mechanisms and Chemical Patterns, Cambridge University
Press, Cambridge (1980).
19. D. O. Hall and K. K. Rao, Photosynthesis, 3rd ed., Edward Arnold, New York (1981).
20. M. D. Hatch and N. K. Boardman (eds.), The Biochemistry of Plants, Vol. 8, Photosyllthesis, Academic
Press, New York (1981).
21. Govindjee (ed.), Photosynthesis: Energy Conversion by Plants and Bacteria, Vol. 1, Academic Press,
New York (1982).
22. Govindjee (ed.), Photosynthesis: Development, Carbon Metabolism, and Plant Productivity, Vol. 2,
Academic Press, New York (1982).
23. S. M. Danks, E. H. Evans, and P. A. Whittaker, Photosynthetic Systems: Structure, Function and
Assembly, Wiley, Chichester (1983).
24. C. H. Foyer, Photosynthesis, Wiley, New York (1984).
25. Govindjee, J. Amesz, and D. C. Fork (eds.), Light Emissionfrom Plallts and Bacteria, Academic Press,
Orlando (1986).
26. E. Broda, The Evolution of the Energetic Process, Pergamon Press, Oxford, (1975).
27. E. Broda and G. A. Peschek, Did respiration or photosynthesis come first? J. Theor. Bioi. 81,201-212
(1979).
28. A. I. Oparin, The origin of life and origin of enzymes, Adv. Enzymol. 27, 347-380 (1965).
29. J. P. Ferris and D. A. Usher, Origins oflife, in Biochemistry (G. Zubay, ed.), pp. 1190-1241, Addison-
Wesley, Reading, Mass. (1983).
30. R. E. Dickerson, Chemical evolution and the origin of life, Sci. Am. 239, 70-85 (1978).
31. C. R. Woese, A proposal concerning the origin of life on the planet earth, J. Mol. BioI. 13, 95-101
(1979).
32. c. R. Woese, The Origin of Life (J. S. Head, ed.), Carolina Biology Readers, Carolina Biological Supply
Co., Burlington (1984).
33. E. Gantt, Phycobilisomes, Light-harvesting pigment complexes, Bioscience 25, 781-788 (1975).
34. E. Gantt, Phycobilisomes, Annu. Rev. Plant Physiol. 32, 327-347 (1981).
35. A. N. Glazer, Phycobilisome, A macromolecular complex optimized for light energy transfer, Biochim.
Biophys. Acta 768, 29-51 (1984).
36. C. B. van Niel, The present status of the comparative study of photosynthesis, Annu. Rev. Plant Physiol.
13, 1-26 (1962).
384 Chapter 12

37. R. Radmer and O. Ollinger, Isotopic composition of photosynthetic O2 flash yields in the presence of
H180 and H1803-, FEBS Lett. 10,57-61 (1980).
38. A. Stemler and R. Radmer, Source of photosynthetic oxygen in bicarbonate-stimulated Hill reaction,
Science 190, 457-458 (1975).
39. A. W. D. Larkum and J. Barrett, Light harvesting processes in algae, in: Advances in Botanical Research,
Vol. 10 (H. W. Woolhouse, ed.), pp. 1-219, Academic Press, New York (1983).
40. T. W. Goodwin (ed.), Chemistry and Biochemistry of Plant Pigments, Academic Press, London (1976).
41. T. W. Goodwin (ed.), Chemistry and Biochemistry of Plant Pigments, Vols. 1 and 2, Academic Press,
London (1976).
42. J. J. Katz, L. L. Shipman, and J. R. Norris, Structure and function of photoreaction-centre chlorophyll, in:
Chlorophyll Organization and Energy Transfer in Photosynthesis, Ciba Found. Symp. 61 (new series),
pp. 1-40, Excerpta Medica, Amsterdam (1979).
43. H. H. Strain and W. A. Svec, Extraction, separation, estimation, and isolation of the chlorophylls, in: The
Chlorophylls (L. P. Vernon and G. R. Seely, eds.), pp. 21-66, Academic Press, New York (1966).
44. J. H. C. Smith and C. S. French, The major and accessory pigments in photosynthesis, Annu. Rev. Plant
Physiol. 14, 181-224 (1963).
45. Govindjee and B. Z. Braun, Light absorption, emission, and photosynthesis, in: Algal Physiology and
Biochemistry (W. D. P. Stewart, ed.), pp. 346-390, University of California Press, Berkeley (1974).
46. J. M. Anderson and J. Barrett, Chlorophyll-protein complexes of brown algae: P700 reaction centre and
light-harvesting complexes, in Chlorophyll Organization and Energy Transfer in Photosynthesis, Ciba
Found. Symp. (new series), pp. 81-104, Excerpta Medica, Amsterdam (1979).
47. D. Siefermann-Harms, Carotenoids in photosystem I. Location in photosynthetic membranes and light
harvesting function, Biochim. Biophys. Acta 811, 325-355 (1985).
48. K. Sauer, Primary events and the trapping of energy, in: Bioenergetics of Photosynthesis (Govindjee, ed.),
pp. 115-181, Academic Press, New York (1975).
49. L. N. M. Duysens, Transfer of excitation energy in photosynthesis, Ph.D. Thesis, Univ. of Utrecht
(1952).
50. L. N. M. Duysens, Photosynthesis, in: Progress in Biophysics and Molecular Biology, Vol. 14 (J. A. V.
Butler and H. E. Huxley, eds.), pp. 1-104, Pergamon Press, Oxford (1964).
51. G. Hoch and R. S. Knox, Primary processes in photosynthesis, Photophysiology 3, 225-251 (1968).
52. R. S. Knox, Excitation energy transfer and migration: Theoretical considerations, in: Bioenergetics of
Photosynthesis (Govindjee, ed.), pp. 183-221, Academic Press, New York (1975).
53. T. Forster, Zwischenmolekulare Energiewanderung und Fluoreszenz, Ann. Physik. 2, 55-75 (1948).
54. T. Forster, Delocalized excitation and excitation transfer, in: Modern Quantum Chemistry, Part Ill, Action
of Light and Organic Molecules (0. Sinanoglu, ed.), pp. 93-137, Academic Press, New York (1965).
55. R. van Grondelle and J. Amesz, Excitation energy transfer in photosynthetic systems, in: Light Emis-
sionfrom Plants and Bacteria (Govindjee, J. Amesz, and D. C. Fork, eds.), Academic Press, Orlando
(1986).
56. G. R. Seely and J. S. Connolly, Fluorescence of photosynthetic pigments in vitro, in: Light Emission from
Plants and Bacteria (Govindjee, J. Amesz, and D. C. Fork, eds.), Academic Press, Orlando (1986).
57. G. Hoch and R. S. Knox, Primary processes in photosynthesis, in: Photophysiology, Vol. 3 (A. C. Giese,
ed.), pp. 225-251, Academic Press, New York (1968).
58. R. S. Knox, Theory and modeling of excitation de localization and trapping, in: Encyclopedia of Plant
Physiology, Vol. 19, new series, (L. A. Staehelin and C. J. Arntzen, eds.), pp. 286-298, Springer-
Verlag, Berlin (1986).
59. J. Lavorel, Luminescence, in: Bioenergetics of Photosynthesis (Govindjee, ed.), pp. 223-317, Academic
Press, New York (1975).
60. S. Malkin, Delayed luminescence, in: Primary Process of Photosynthesis (J. Barber, ed.), pp. 349-431,
Elsevier, Amsterdam (1977).
61. Govindjee and P. A. Jursinic, Photosynthesis and fast changes in light emission by green plants, in:
Photochem. Ph%biol. Rev., Vol. 4 (K. C. Smith, ed.), pp. 125-205, Plenum Press, New York (1979).
62. J. Lavorel, J. Lavergne, and A. L. Etienne, A reflection on several problems of luminescence in photo-
synthetic systems, Photochem. Photobiophys. 3, 287-314 (1982).
63. R. Emerson and W. Arnold, A separation ofthe reactions in photosynthesis by means of intermittent light,
1. Gen. Physiol. 15, 391-420 (1932).
Photosynthesis 385

64. R. Emerson and W. Arnold, The photochemical reaction in photosynthesis, J. Gen. Physiol. 16, 191-205
(1932).
65. R. K. Clayton and W. R. Sistrom (eds.), The Photosynthetic Bacteria, Plenum Press, New York (1978).
66. N. Pfennig, General physiology and ecology of photosynthetic bacteria, in: The Photosynthetic Bacteria
(R. K. Clayton and W. R. Sistrom, eds.), pp. 3-18, Plenum Press, New York (1978).
67. H. Michel, Three-dimensional crystals of a membrane protein complex. The photosynthetic reaction
centre from Rhodopseudomonas viridis, J. Mol. BioI. 158, 567-572 (1982).
68. J. Deisenhofer, D. Epp, E. Miki, R. Huber, and H. Michel, X-ray structure analysis of a membrane
protein complex. Electron density map at 3A. resolution and a model of the chromatophores of the
photosynthetic reaction center from Rhodopseudomonas viridis, J. Mol. Bioi. 180, 385-398 (1984).
69. J. Deisenhofer, H. Michel, and R. Huber, The structural basis of photosynthetic light reactions in bacteria,
Trends in Biochem. Sci. 10, 243-248 (1985).
70. J. Deisenhofer, O. Epp, K. Miki, R. Huber, and H. Michel, Structure of the protein subunits in the
photosynthetic reaction centre of Rhodopseudomonas viridis at 3A. resolution, Nature 318, 618-624
(1985).
71. N. W. Woodbury, M. Becker, D. Middendorf, and W. W. Parson, Picosecond kinetics of the initial
photochemical electron-transfer reaction in bacterial photosynthetic reaction centers, Biochemistry 24,
7516-7521 (1985).
72. J. M. Olson, Chlorophyll organization in green photosynthetic bacteria, Biochim. Biophys. Acta 594,33-
51 (1981).
73. L. A. Staehlin, J. R. Golecki, and G. Drews, Supramolecular organization of chlorosomes (Chlorobium
vesicles) and of their membrane attachment sites in Chlorobium limicola, Biochim. Biophys. Acta 589,
30-45 (1980).
74. T. Wechsler, F. Suter, R. C. Fuller, and H. Zuber, The complete amino acid sequence of the bac-
teriochlorophyll c binding polypeptide from chlorosomes of the green photosynthetic bacterium Chlo-
roflexus aurantiacus, FEBS Lett. 181, 173-178 (1985).
75. B. B. Buchanan, P. Schiirmann, and K. T. Shanmugam, Role of the reductive carboxylic acid cycle in a
photosynthetic bacterium lacking ribulose 1,5-diphosphate carboxylase, Biochim. Biophys. Acta 283,
136-145 (1972).
76. C. F. Fowler, N. A. Nugent, and R. C. Fuller, The isolation and characterization of a photochemically
active complex from Chloropseudomonas ethylica, Proc. Natl. Acad. Sci. USA 68, 2278-2282 (1971).
77. B. K. Pierson and R. W. Castenholz, Photosynthetic apparatus and cell membrane of the green bacteria,
in: The Photosynthetic Bacteria (R. K. Clayton and W. R. Sistrom, eds.), pp. 179-197, New York
(1978).
78. R. E. Blankenship, Electron transport in green photosynthetic bacteria, Photosyn. Res. 6, 317-333
(1985).
79. D. D. Burns and M. Midgley, Localization and possible role of an adenosine triphosphatase in Chlorobium
thiosulfatophi/um, Eur. J. Biochem. 67, 323-333 (1976).
80. B. K. Pierson and R. W. Castenholz, A phototrophic gliding filamentous bacterium of hot springs,
Chloroflexus aurantiacus, gen. and sp. nov., Arch. Microbiol. 100,5-24 (1974).
81. 1. Amesz, Electron transport in green bacteria, in: Advances in Photosynthesis Research, Vol. 1 (C.
Sybesma, ed.), pp. 621-628, Martinus NijhofflDr. W. Junk, The Hague (1984).
82. R. E. Blankenship and R. C. Prince, Excited-state redox potentials and the Z scheme of photosynthesis,
Trends Biochem. Sci. 10, 382-383 (1985).
83. A. W. Rutherford and P. Heathcote, Primary photochemistry in photo system I, Photosyn. Res. 6, 295-
316 (1985).
84. J. Amesz, Photosynthesis: Structure of the membrane and membrane proteins, Prog. Bot. 47, 87-104
(1985).
85. 1. E. Mullet, J. J. Burke, and C. J. Arntzen, Chlorophyll proteins of photosystem I, Plant Physiol. 65,
814-822 (1980).
86. P. Haworth, J. L. Watson, and C. Arntzen, The detection, isolation and characterization of a light-
harvesting complex which is specifically associated with photosystem I. Biochim. Biophys. Acta 724,
151-158 (1983).
87. J. M. Anderson, A chlorophyll alb-protein complex of photosystem I, Photobiochem. Photobiophys. 8,
221-228 (1984).
386 Chapter 12

88. B. Kok, On the reversible absorption change at 705 mIL in photosynthetic organisms, Biochim. Biophys.
Acta 22, 399-401 (1956).
89. T. Hiyama and B. Ke, Difference spectra and extinction coefficients of P700, Biochim. Biophys. Acta
267, 160-172 (1972).
90. H. Beinert, B. Kok, and G. Hoch, The light induced electron paramagnetic resonance signal of pho-
tocatalyst P700, Biochem. Biophys. Res. Commun. 7, 209-212 (1962).
91. T. Hiyama and B. Ke, A new photosynthetic pigment "P430": Its possible role as the primary electron
acceptor of photosystem I, Proc. Natl. Acad. Sci. USA 68, 1010-1013 (1971).
92. l. M. Fenton, M. J. Pellin, Govindjee, and K. l. Kaufmann, Primary photochemistry of the reaction center
of photosystem II, FEBS Lett. 100, 1-4 (1979).
93. K. Kanogawa, A. Namiki, N. Nakashima, K. Yoshihara, and I. Ikegarni, Picosecond transient behavior of
reaction-center particles of photosystem I isolated from spinach chloroplasts. Energy and electron transfer
upon single and multiple photon excitation, Photochem. Photobiol. 34, 511-516 (1981).
94. V. A. Shuvalov, B. Ke, and E. Dolan, Kinetic and spectral properties of the intermediary electron
acceptor Al in photosystem I, FEBS Lett. 100, 5-8 (1979).
95. V. A. Shuvalov, A. V. Klevanik, A. V. Sharkov, P. G. Kryukov, and B. Ke, Picosecond spectroscopy of
photosystem I reaction centers, FEBS Lett. 107, 313-316 (1979).
96. D. O. Hall, R. Cammack, and K. K. Rao, Non-haem iron proteins, in: Iron in Biochemistry and Medicine
(A. Jocob and M. Worwood, eds.), Academic Press, New York (1974).
97. G. Hauska, E. Hurt, N. Gabellini, and W. Lockau, Comparative aspects of quinol-cytochrome c/plasto-
cyanin oxidoreductases, Biochim. Biophys. Acta 726,97-133 (1983).
98. J. Bennett, The simple art of making photosynthetic complexes, Trends in Biochem. Sci. 6, VI-VII
(1981).
99. l. S. Rieske, Composition, structure, and function of complex III of the respiratory chain, Biochim.
Biophys. Acta 456, 195-247 (1976).
100. l. Alt, P. Westhoff, B. B. Sears, N. Nelson, E. Hurt, G. Hauska, and R. G. Herrmann, Genes and
transcripts for the polypeptides of the cytochrome bdf complex from spinach thylakoids, EMBO J. 2,
979-986 (1983).
101. W. Heinemeyer, 1. Alt, and R. G. Herrmann, Nucleotide sequence of the clustered genes for apocyto-
chrome b6 and subunit 4 of the cytochrome b/f complex in spinach plastid chromosome, Curro Gen. 8,
543-549 (1984).
102. J. Alt and R. G. Herrmann, Nucleotide sequence of the gene for preapocytochrome f in the spinach plastid
chromosome, Curro Gen. 8,551-557 (1984).
103. P. Mitchell, Possible molecular mechanisms of the protonmotive function of cytochrome systems, J.
Theor. Bioi. 62, 327-367 (1976).
104. M. A. Selak and l. Whitmarsh, Kinetics of the electrogenic step and cytochrome b6 and f redox changes in
chloroplasts, FEBS Lett. 150, 286-292 (1982).
105. P. loliot and A. loliot, Proton pumping and electron transfer in the cytochrome b/f complex of algae,
Biochim. Biophys. Acta 849, 211-222 (1986).
106. P. R. Rich, Mechanisms of quinol oxidation in photosynthesis, Photosyn. Res. 6, 335-348 (1985).
107. N. Murata and M. Miyao, Extrinsic membrane proteins in the photosynthetic oxygen-evolving complex,
Trends Biochem. Sci. 10, 122-124 (1985).
108. D. M. Ghanotakis and C. F. Yocum, Polypeptides of photosystem II and their role in oxygen evolution,
Photosyn. Res. 7,97-114 (1985).
109. W. A. Cramer, W. R. Widger, R. G. Herrmann, and A. Trebst, Topography and function of thylakoid
membrane proteins, Trends Biochem. Sci. 10, 125-129 (1985).
110. T. Kambara and Govindjee, Molecular mechanism of water oxidation in photosynthesis based on the
functioning of manganese in two different environments, Proc. Natl. Acad. Sci. USA 82, 6119-6123
(1985).
111. J. G. Metz and D. Miles, Use of a nuclear mutant of maize to identify components of photosystem II,
Biochim. Biophys. Acta 681,95-102 (1982).
112. J. G. Metz and M. Siebert, Presence in photosystem II core complexes of a 34-kilodalton polypeptide
required for water photolysis, Plant Physiol. 76, 829-832 (1984).
113. D. J. Kyle, The 32000 dalton <2B protein of photosystem II, Photochem. Photobiol. 41, 107-116 (1985).
114. J. J. Eaton-Rye, D. J. Blubaugh, and Govindjee, Action of bicarbonate on photosynthetic electron
Photosynthesis 387

transport in the presence or absence of inhibitory anions, in: Ion Interactions in Energy Transfer Bio-
membranes (G. C. Papageorgiou, J. Barber, and S. Papa, eds.), pp. 263-278, Plenum Press, New York
(1986).
115. P. Jursinic and Govindjee, Temperature dependence of delayed light emission in the 6 to 340 microsecond
range after a single flash in chloroplasts, Photochem. Photobiol. 26,617-628 (1977).
116. H. J. van Gorkom, M. P. J. Pulles, and J. S. C. Wessels, Light-induced changes of absorbance and
electron spin resonance in small photosystem II particles. Biochim. Biophys. Acta 408,331-339 (1975).
117. H. J. van Gorkom, J. J. Tamminga, and J. Haveman, Primary reactions, plastoquinone and fluorescence
yield in subchloroplast fragments prepared with deoxycholate, Biochim. Biophys. Acta 347, 417-438
(1974).
118. V. V. Klimov and A. A. Krasnovsky, Pheophytin as the primary electron acceptor in photosystem 2
reaction centers, Photosynthetica 15, 592-609 (1981).
119. V. V. Klimov, E. Dolan, and B. Ke, EPR properties of an intermediary electron acceptor (pheophytin) in
photosystem II reaction centers at cyogenic temperatures, FEBS Lett. 112, 97-100 (1980).
120. J. Fajer, M. S. Davis, A. Forman, V. V. Klimov, E. Dolan, and B. Ke, Primary electron acceptors in
plant photosynthesis, J. Am. Chem. Soc. 102,7143-7145 (1980).
121. A. W. Rutherford, EPR evidence for an acceptor functioning in photosystem II when the pheophytin
acceptor is reduced, Biochem. Biophys. Res. Commun. 102, 1065-1070 (1981).
122. c. A. Wraight, Current attitudes in photosynthesis research, in: Photosynthesis: Energy Conversion by
Plallts and Bacteria, Vol. 1 (Govindjee, ed.), pp. 17-61, Academic Press, New York (1982).
123. W. F. J. Vermaas and Govindjee, The acceptor side of photosystem II in photosynthesis, Photochem.
Photobiol. 34, 775-793 (1981).
124. J. H. A. Nugent, B. A. Diner, and M. C. W. Evans, Direct detection of the electron acceptor of
photosynthesis II, FEBS Lett. 124, 241-244 (1981).
125. L. N. M. Duysens and H. E. Sweers, Mechanism of two photochemical reactions in algae as studied by
means of fluorescence, in: Studies on Microalgae and Photosyllthetic Bacteria, pp. 353-372, University
of Tokyo Press (1963).
126. V. V. Klimov, S. I. Allakhverdiev, and V. Z. Pashchenko, Measurement of the activation energy and
lifetime of fluorescence of photosystem 2 chlorophyll, Dokl. Akad. Nauk. SSSR 242, 1204-1207 (1978).
127. H. J. van Gorkom, Identification of the reduced primary electron acceptor of photosystem II as a bound
semiquinone anion, Biochim. Biophys. Acta 347, 439-442 (1974).
128. V. V. Klimov, A. V. Klevanik, V. A. Shuvalov, and A. A. Krasnovsky, Reduction ofpheophytin in the
primary light reaction of photosystem II, FEBS Lett. 82, 183-186 (1977).
129. Govindjee, Photosystem II: The oxygen evolving system of photosynthesis, in: Advances in Photo-
synthesis Research, Vol. I (C. Sybesma, ed.), pp. 227-238, Martinus Nijhoff/Dr. W. Junk, The Hague
(1984).
130. W. A. Cramer and A. R. Crofts, Electron and proton transport, in: Photosynthesis: Energy Conversion by
Plallts and Bacteria, Vol. I, (Govindjee, ed.), pp. 387-467, Academic Press, New York (1982).
131. B. R. Velthuys, Electron-dependent competition between plastoquinone and inhibitors for binding to
photosystem II, FEBS Lett. 126,277-281 (1981).
132. B. R. Velthuys and J. Amesz, Charge accumulation at the reducing side of system 2 of photosynthesis,
Biochim. Biophys. Acta 333, 85-94 (1974).
133. B. Bouges-Bocquet, Electron transfer between the two photosystems in spinach chloroplasts, Biochim.
Biophys. Acta 314, 250-256 (1973).
134. M. P. 1. Pulles, H. J. van Gorkom, and J. G. Willemsen, Absorbance changes due·to the charge-
accumulating species in system 2 of photosynthesis, Biochim. Biophys. Acta 449, 536-540 (1976).
135. M. C. W. Evans, Y. E. Atkinson, and R. C. Ford, Redox characterization ofthephotosystem II electron
acceptors. Evidence for two electron carriers between pheophytin and Q, Biochim. Biophys. Acta 806,
247-254 (1985).
136. J. Amesz, Spectrophotometric evidence for the participation of a quinone in photosynthesis of intact blue-
green algae, Biochim. Biophys. Acta 79, 257-265 (1964).
137. B. Rumberg, P. Schmidt-Mende, and H. T. Witt, Different demonstrations of the coupling of two light
reactions in photosynthesis, Nature 201, 466-468 (1964).
138. J. Amesz and L. N. M. Duysens, Primary and associated reactions of system II, in: Primary Processes of
Photosynthesis, Vol. 2 (J. Barber, ed.), pp. 149-185, Elsevier, Amsterdam (1977).
388 Chapter 12

139. Govindjee, T. Kambara, and W. Coleman, The electron donor side of photosystem II: The oxygen
evolving complex, Photochem. Photobio!' 42, 187-210 (1985).
140. P. M. Wood and D. S. Bendall, The reduction of plastocyanin by plastoquinol-l in the presence of
chloroplasts, Eur. 1. Biochem. 61, 337-344 (1976).
141. G. T. Babcock, R. E. Blankenship, and K. Sauer, Reaction kinetics for positive charge accumulation on
the water side of chloroplast photosystem II, FEBS Lett. 61, 286-289 (1976).
142. J. A. van Best and P. Mathis, Kinetics of reduction of the oxidized primary electron donor of photosystem
II in spinach chloroplasts and in Chlorella cells in the microsecond and nanosecond time ranges following
flash excitation, Biochim. Biophys. Acta 503, 178-188 (1978).
143. J. M. Briantais, C. Vernotte, M. Miyao, N. Murata, and M. Picaud, Relationship between O2 evolution
and cytochrome b-559 high-potential form in photosystem II particles, Biochim. Biophys. Acta 808, 348-
351 (1985).
144. P. Joliot, G. Barbieri, and R. Chabaud, Un nouveau modele des centres photochimiques du systeme II,
Photochem. Photobiol. 10, 309-329 (1969).
145. B. Kok, B. Forbush, and M. McGloin, Cooperation of charges in photosynthetic O2 evolution. I. A linear
four step mechanism, Photochem. Photobiol. 11, 457-475 (1969).
146. G. M. Cheniae, Photosystem II and O2 evolution, Annu. Rev. Plant Physiol. 21,467-498 (1970).
147. J. Amesz, The role of manganese in photosynthetic oxygen evolution, Biochim. Biophys. Acta 726, 1-12
(1983).
148. G. C. Dismukes, The metal centers of the photosynthetic oxygen-evolving complex, Photochem. Pho-
tobio!. 43, 99-115 (1986).
149. G. M. Cheniae and I. F. Martin, Sites of function of manganese within photosystem II. Roles in O2
evolution and system II, Biochim. Biophys. Acta 197, 219-239 (1970).
150. P. Homann, Effects of manganese on the fluorescence of chloroplasts, Biochem. Biophys. Res. Commun.
33, 229-234 (1968).
151. J. Barber, Has the mangano-protein of the water splitting reactiop. of photosynthesis been isolated? Trends
Biochem. Sci. 9, 79-80 (1984).
152. D. A. Abramowicz, T. K. Raab, and G. C. Dismukes, Manganese proteins from spinach thylakoids and
02-evolving photosystem-I1 enriched membranes (OEPII): Their role in oxygen evolution, in Advances in
Photosynthesis Research, Vol. 1 (C. Sybesma, ed.), pp. 349-353, Martinus NijhofflDr. W. Junk, The
Hague (1984).
153. R. E. McCarty, Photosynthetic phosphorylation by chloroplasts of higher plants, in: Photochemical and
Photobiological Reviews, Vol. 5 (K. C. Smith, ed.), pp. 1-48, Academic Press, New York (1980).
154. D. R. Ort and B. A. Melandri, Mechanism of ATP synthesis, in: Photosynthesis, Energy Conversion by
Plants and Bacteria, Vol. 1 (Govindjee, ed.), pp. 537-587, Academic Press, New York (1982).
155. D. I. Arnon, M. B. Allen, and F. R. Whatley, Photosynthesis by isolated chloroplasts, Nature 174, 394-
396 (1954).
156. D. I. Arnon, The discovery of photosynthetic phosphorylation, Trends Biochem. Sci. 9,258-262 (1984).
157. A. Frenkel, Light induced phosphorylation by cell-free preparations of photosynthetic bacteria, 1. Am.
Chem. Soc. 76, 5568-5569 (1954).
158. G. Hind and A. T. Jagendorf, Separation of light and dark stages in phosphorylation, Proc. Natl. Acad.
Sci. USA 49, 715-722 (1963).
159. J. Barber (ed.), Electron Transport and Photophosphorylation, Elsevier, Amsterdam (1982).
160. P. Mitchell, Coupling of phosphorylation to electron and hydrogen transfer by a chemi-osmotic type of
mechanism, Nature 191, 144-148 (1961).
161. P. Mitchell, Chemiosmotic coupling in oxidative and photosynthetic phosphorylation, Bioi. Rev. 41, 445-
502 (1966).
162. G. G. Hammes, Mechanisms of ATP synthesis and coupled proton transport: Studies with purified
chloroplast coupling factor, Trends Biochem. Sci. 8, 131-134 (1983).
163. R. E. McCarty and C. M. Nalin, Structure, mechanism, and regulation of the chloroplast H+-ATPase
(CF1-CFo), in: Encyclopedia of Plant Physiology, new series, Vol. 19 (L. A. Staehelin and C. J. Arntzen,
eds.), pp. 576-583, Springer-Verlag, Berlin (1986).
164. H. Strotmann, Evaluation of results on nucleotide-binding sites of the ATPase complex, in: Encyclopedia
of Plant Physiology, new series, Vol. 19 (L. A. Staehelin and C. J. Arntzen, eds.), pp. 584-594,
Springer-Verlag, Berlin (1986).
Photosynthesis 389

165. H. Strottmann and S. Bickel-Sandkotter, Structure, function, and regulation of chloroplast ATPase, Annu.
Rev. Plant Physiol. 35, 97-120 (1984).
166. P. D. Boyer, Coupling mechanisms in capture, transmission and use of energy, Annu. Rev. Biochem. 46,
957-966 (1977).
167. E. C. Slater, Mechanism of oxidative phosphorylation, Annu. Rev. Biochem. 46, 1015-1026 (1977).
168. P. D. Boyer and W. E. Kohlbrenner, The present status of the binding-change mechanism and its relation
to ATP formation by chloroplasts, in: Energy Coupling in Photosynthesis (B. R. Selman and S. Selman-
Reimer, eds.), pp. 231-240, Elsevier, Amsterdam (1981).
169. D. R. Ort and B. A. Me1andri, Mechanism of ATP synthesis, in: Photosynthesis: Energy Conversion by
Plants and Bacteria, Vol. 1 (Govindjee, ed.), pp. 537-587, Academic Press, London (1982).
170. B. A. Me1andri and G. Venturoli, Local and delocalized interactions in energy coupling, in: Encyclopedia
of Plant Physiology, new series, Vol. 19 (L. A. Staehlin and C. J. Arntzen, eds.), pp. 560-569, Springer-
Verlag, Berlin (1986).
171. R. A. Dilley, Proton-membrane interactions in chloroplast thylakoids, in: Encyclopedia of Plant Phys-
iology, new series, Vol. 19 (L. A. Staehelin and C. J. Arntzen, eds.), pp. 570-575, Springer-Verlag,
Berlin (1986).
172. A. T. Jagendorf and E. Uribe, ATP formation caused by acid-base transition of spinach chloroplasts,
Proc. Natl. Acad. Sci. USA 55, 170-177 (1966).
173. R. A. Dilley, Coupling of ion and electron flow in chloroplasts, Curro Top. Bioenerg. 4,237-271 (1971).
174. H. Baltscheffsky and M. Baltscheffsky, Energy conversion reactions in bacterial photosynthesis, Curro
Top. Bioenergy. 4, 273-325 (1971).
175. C. J. Arntzen and J. M. Briantais, Chloroplast structure and function, in: Bioenergetics of Photosynthesis
(Govindjee, ed.), pp. 51-l13, Academic Press, New York (1975).
176. S. Kaplan and C. J. Arntzen, Photosynthetic membrane structure and function, in: Photosynthesis:
Energy, Conversion by Plants and Bacteria, Vol. I (Govindjee, ed.), pp. 65-151, Academic Press, New
York (1982).
177. J. M. Andersson, Consequences of spatial separation of photosystem 1 and 2 in thylakoid membranes of
higher plant chloroplasts, FEBS Leu. 124, 1-10 (1981).
178. J. M. Andersson and B. Andersson, The architecture of photosynthetic membranes: lateral and transverse
organization, Trends Biochem. Sci. 7,288-292 (1982).
179. C. Hitchcock and B. W. Nichols, Plant Lipid Biochemistry, Academic Press, London, New York (1971).
180. J. Bennett, K. E. Steinback, and C. J. Arntzen, Chloroplast phosphoproteins: Regulation of excitation
energy transfer by phosphorylation of thylakoid membrane proteins. Proc. Natl. Acad. Sci. USA 77,
5253-5257 (1980).
181. D. C. Fork and K. Satoh, The control by state transitions of the distribution of excitation energy in
photosynthesis, Annu. Rev. Plant Physiol. 37, 335-361 (1986).
182. J. A. Bassham, Kinetic studies of photosynthetic carbon reduction cycle, Annu. Rev. Plant Physiol. 15,
101-120 (1964).
183. J. A. 'Bassham and B, B. Buchanan, Carbon dioxide fIxation pathways in plants and bacteria, in:
Photosynthesis,' Development, Carbon Metabolism, and Plant Productivity, Vol. 2 (Govindjee, ed.), pp.
141-189, Academic Press, New York (1982).
184. M. D. Hatch and C. K Slack, Photosynthetic CO 2 fIxation pathways, Annu. Rev. Plant Physiol. 21, 141-
162 (1970).
185. O. Bjorkman and J. Berry, High effIciency photosynthesis, Sci. Am. 229, 80-93 (1973).
186. J. A. Berry and G. Farquhar, The CO 2 concentrating function of C4 photosynthesis. A biochemical model,
in: Proc. Fourth Int. Congr. on Photosynthesis (D. O. Hall, J. Coombs, and T. W. Goodwin, eds.), pp.
119-131, The Biochemical Society, London (1978).
187. M. Kluge, Crassulacean acid metabolism, in: Photosynthesis: Development, Carbon Metabolism, and
Plant Productivity, Vol. 2 (Govindjee, ed.), pp. 141-189, Academic Press, New York (1982).
188. C. B. Osmond, Crassulacean acid metabolism: A curiosity in context, Annu. Rev. Plant Physiol. 29, 379-
414 (1978).
189. W. L. Ogren and R. Chollet, Photorespiration, in: Photosynthesis: Development, Carbon Metabolism, and
Plant Productivity, Vol. 2 (Govindjee, ed.), pp. 191-230, Academic Press, New York (1982).
190. I. Zelitsch, Photosynthesis, Photorespiration, and Plant Productivity, Academic Press, New York, Lon-
don (1971).
390 Chapter 12

191. C. B. Osmond and O. Bjorkman, Simultaneous measurements of oxygen effects on net photosynthesis and
glycolate-metabolism in C3 and C4 species of Atriplex, Carnegie Inst. Wash. Year Book 71, 141-148
(1972).
192. J. S. Connolly (ed.), Photochemical Conversion and Storage of Solar Energy, Academic Press, New York
(1981).
193. D. O. Hall, Photosynthesis for Energy, in: Advances in Photosynthesis Research. Vol. 2 (C. Sybesma,
ed.), pp. 727-740, Martinus Nijhoff/Dr. W. Junk, The Hague (1984).
194. L. N. M. Duysens, Specially developed photosynthetic organisms as future possibility for large-scale low-
cost conversion of solar radiation into useful high-grade energy, in: Advances in Photosynthesis, Vol. 2
(c. Sybesma, ed.), pp. 741-744, Martinus Nijhoff/Dr. W. Junk, The Hague (1984).
195. M. Calvin, Solar energy by photosynthesis, Science 184, 375-381 (1974).
196. M. Avron, The outdoor cultivation of the halotolerant alga Dunaliella: A model for biosolar energy
utilization for useful chemical products, in: Advances in Photosynthesis Research, Vol. 2 (C. Sybesma,
ed.), pp. 745-753, Martinus NijhofflDr. W. Junk, The Hague (1984).
197. S. G. Boxer, Model reactions in photosynthesis, Biochim. Biophys. Acta 725, 265-292 (1983).
198. J. Barber (ed.), Photosynthesis in Relation to Model Systems, Topics in Photosynthesis, Vol. 3, Elsevier,
Amsterdam (1979).
199. M. T. Black, T. H. Brearley, and P. Horton, Heterogeneity in chloroplast photosystem II, Photosyn. Res.
8, 193-207 (1986).
200. R. C. Prince, Redox-driven proton gradients, Bioscience 35, 22-26 (1985).
201. P. D. Boyer, B. Chance, L. Ernster, P. Mitchell, E. Racker, and E. C. Slayter, Oxidative phosphorylation
and photophosphorylation. Annu. Rev. Biochem. 46, 955-1026 (1977).
 13
Bioluminescence

13.1. Introduction ..................................................................... 391

13 .1.1. Occurrence ............................................................... 392
13.1.2. Light Organs and Organization ............................................... 393
13 .1. 3. Origin and Function of Bioluminescence ....................................... 394
13.1.4. Mechanism ............................................................... 396
13.2. Chemiluminescence Reactions ....................................................... 397
13.3. Generalizations about Bioluminescence Reactions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 400
13.3.1. Activation ................................................................ 400
13.3.2. Oxygenation .............................................................. 400
13.3.3. Excitation ................................................................ 401
13.304. Turnover ................................................................. 401
1304. Firefly Bioluminescence ........................................................... 401
13 04.1. Chemical Mechanism .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 402
13 04.2. The Light Emitters ......................................................... 402
13.5. Cypridina and Coelenterate Bioluminescence .......................................... 404
13.5.1. Chemical Mechanism. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 404
13.5.2. Enzymology .............................................................. 405
13.5.3. The Light Emitters ......................................................... 406
13.504. Sensitized Bioluminescence. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 407
13.5.5. Control .................................................................. 408
13.6. Bacterial Bioluminescence. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 409
13.6.1. Chemical Mechanism. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 410
13.6.2. Enzymology .............................................................. 411
13.6.3. Lumazine Protein .......................................................... 411
13.7. Low-Level Luminescence, Biological Chemiluminescence ................................ 413
13.8. Applications of Bioluminescence .................................................... 414
13.8.1. Ultrasensitive Analysis. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 414
13.8.2. Recombinant DNA Techniques ...... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 416
13.9. References .......................................................................417

13.1. INTRODUCTION

For most people, bioluminescence is represented by the flash of the ftrefly or the "phos-
phorescence" that frequently occurs on agitating the surface of ocean water. Because of
the ease of obtaining material, the firefly's bioluminescence reaction was one of the first
to receive detailed biochemical study, with the result that this system is the archetype of a

John Lee • Bioluminescence Laboratory, School of Chemical Sciences, Department of Bio-

chemistry, University of Georgia, Athens, Georgia 30602.

391
392 Chapter 13

variety of enzymatic processes that produce light in the many bioluminescent organisms,
ranging from the marine bacteria to the large luminous beetles from South America. What
is usually understood by the term bioluminescence is a cold-light emission of high effi-
ciency that has a biological function for the organism concerned, although what this
function is is still conjectural in many cases. In contrast to bioluminescence, a great
number of biological processes have been observed to be accompanied by the emission of
a very low level of light, so-called biological chemiluminescence. This observation and
the fact that bioluminescence is so widespread among many phyla (although rarely are
many members of a phylum bioluminescent) has led to the suggestion that low-level light
emission arose very early in evolution, and that the efficient bioluminescence was a
secondary adaptation that enabled some species to compete more effectively in their
biological niches.

13.1.1. Occurrence
More than half of all phyla in the animal kingdom contain members that are bio-
luminescent. (I) Many bioluminescent organisms have a worldwide distribution and some
are found only in certain areas of the world. Table 13-1 lists the organisms that have· been
well studied.
In the first half of this century, E. N. Harvey of Princeton University made a lifelong

TABLE 13·1. Properties of Bioluminescent Organisms

Bioluminescence
maxima range
Type and examples in vivo AB(nm) Occurrence
Bacteria, Photobacterium, Vibrio 472-545 Marine
Fungi 525-550 Terrestrial
Protozoans, dinoflagellates 470-480 Marine
Coelenterates 455-508 Marine
Hydrozoan jellyfish (Aequorea)
Ctenophores (Mnemiopsis) 480-505 Marine
Anthozoans (Renilla) 505-510 Marine
Annelids
Polychaete worms 465-520 Marine
Earthworms 500-570 USA, Australasia
Molluscs
Limpet (Latia) 535 Freshwater, New
Zealand
Bivalve (Pholas) 490 Mediterranean
Squids 416-540 Marine
Crustaceans (Cypridina) 465 Marine, Japan
Euphausiids and decapods 455-475 Marine
Millipedes 496 USA
Insects
Firefly 552-582 Terres trial
Glow-worm (Arachnocampa) 488 Australasia
Railroad worm (Phrixothrix) 560,630 Brazil
Fish 455-507 Marine
Bioluminescence 393

PH YLUM CNIDARIA

Fig. 13-1. Some bioluminescent coelenterates.

PHYLUM CT£NOPHORA
Left to right, Cnidaria: Hydrozoa (two types),
Scyphozoa, and Anthozoa (Renilla); Ctenophora:
Tentaculata (Pleurobrachia and Mnemiopsis) and
Nuda. [From W. W. Ward, Energy Transfer Pro-
cesses in Bioluminescence, Photochem. Pho-
tobiol. Rev. 4, 1-57, (1979); see this reference for
a full description.]

study of bioluminescence, especially of its biological aspects. Over 40 years ago, he

wrote:
Apparently, there is no rhyme or reason in the distribution of luminescence throughout the plant
or animal kingdom. It is as if the various groups had been written on a blackboard and a handful
of damp sand cast over the names. Where each grain of sand strikes, a luminous species
appears .... Several large groups contain no luminous forms whatever. It is an extraordinary
fact that one species in a genus may be luminous and another closely allied species contain no
trace of luminosity. (2)

Bioluminescence occurs in many terrestrial forms but is most common in the sea,
particularly in the deep ocean where the majority of species are luminescent.(l) On land,
in addition to the ftrefly, there exist luminous fungi ("foxfire"), glowworms, freshwater
snails (Latia) found only in New Zealand, earthworms, and many insects including the
"railroad worm" found in South America. In the ocean, protozoa are mainly responsible
for the so-called phosphorescence, but the coelenterates contain the most bioluminescent
species, i.e., the soft corals (Anthozoa), jellyftsh (Hydrozoa), and comb-jellies (Cteno-
phora). Some representatives from the coelenterates are illustrated in Fig. 13-1.
There are many types of luminous ftsh some of which derive their luminescence from
a culture of symbiotic bacteria. There are marine worms (Chaetopterus, Balanoglossis,
Odontosyllis), a clam (Pholas), crustacea (Cypridina or Vargula), squid (Watasenia),
shrimp (Holophorus), and echinoderms (sea stars and sea urchins). Bioluminescence is
not found in any organisms higher than fish.

13.1.2. Light Organs and Organization

In most cases, the bioluminescence is manufactured within specialized cells called
photocytes. Many higher vertebrates have a highly organized collection of these cells
making up a light organ or photophore. Harvey observed that' 'there seems to have been
no development of luminosity along direct evolutionary lines, although a more or less
definite series of gradations with increasing structural complexity may be traced among
the forms with highly developed luminous organs. "(2)
394 Chapter 13

In those fish that utilize the bioluminescence from symbiotic luminous bacteria, the
photophore is like a culture vessel. Many other fish use a different bioluminescence
system to produce light involving either intra- or intercellular processes. There also exist a
number of highly specialized accessory structures to the light organ, in some cases
reaching a degree of complexity comparable to, and in fact resembling, the vertebrate eye.

13.1.3. Origin and Function of Bioluminescence(3-5)

The biological purpose of bioluminescence is to be seen, so it could not have been
until the development of primitive visual processes that evolutionary pressure favored the
diversion of valuable metabolic energy into light production. Energy conservation is
perhaps the reason that bioluminescence systems are found to be very efficient. One
proposal for the origin of bioluminescence is that the first bioluminescence system devel-
oped out of a biological chemiluminescence reaction involving an oxygenase enzyme
acting on a hydrocarbon. This mechanism is a free-radical one. The accumulation of a
suitably fluorescent metabolite, e.g., the highly fluorescent proteins that are found in the
bioluminescent bacteria and coelenterates, would then provide a means to sensitize or
amplify this light production at a minimal energy cost. We will see shortly how such a
mechanism operates in the bacterial and coelenterate bioluminescences. Higher on the
evolutionary scale, selection was made of reactions that would produce highly fluorescent
product molecules directly in their excited singlet state, and with everything done on only
one enzyme. This is the mechanism of the firefly, Cypridina. and other higher forms.
It is a reasonable (though controversial) conjecture that the bacteria were the first
organisms to "discover" bioluminescence and a niche in which it could be of advantage.
This is based on the facts that bacteria lie lowest on the evolutionary scale, they can
mutate and evolve rapidly if benefit is to be derived from it, and they already possess the
substrates for bioluminescence. Although luciferase, the enzyme needed to generate the
bioluminescence, is unique to the bioluminescent bacteria, it has been postulated that
luciferase might also have a terminal oxidase function in a respiratory pathway operative
at a very low oxygen level, i.e., one approaching anaerobiosis.
The alternative view held by most scientists in the field is that bioluminescence arose
independently in many organisms. This is because the property is spread over such a
diversity of organisms that it seems a relatively recent event in evolution, and the lu-
ciferins often have very different chemical structures, even among closely related orga-
nisms. Additionally, there is no structural property of the luciferases that would suggest a
common lineage.
The biological aspect of bioluminescence function that has received the most study
and is best understood is that of the firefly system. (6) The flash is a sexual signaling and is
made up of a series of short flashes (Fig. 13-2), a "Morse code" that is characteristic of
the species. The flying male flashes and may be answered after a definite time interval,
about 2 s, by the female. In one exceptional though interesting case, the female of one
species is insectivorous, feeding on the males of other species and luring its prey to it by
imitating the female response appropriate to the prey species.
There is also a visually discernible difference in the color of bioluminescence be-
tween certain species of firefly. This appears to be an adaptation since the yellow bio-
luminescence (Fig. 13-3c) is from fireflies active at dusk, the yellow light contrasting
Bioluminescence 395

Fig. 13-2. Flash sequence from live fireflies; the vertical direction rep-
resents relative light intensity. (Top) Photinus gracilobus, (middle)Pho- ~ ~
tinus meianuris, (lower) Photinus nothus. [Redrawn from H. H. Seliger ~-------'
and W. D. McElroy, Light: Physical and Biological Action. Academic o~-~---7:--~--!;-2-~
Press, New York (1965).] TIME (5)

against the green background of foliage. The night-active fireflies emit at shorter wave-
lengths (Fig. 13-3a), the green light more efficiently matching the visual sensitivity of
their species. (7)
It has recently been shown that starting with a mixed population of luminous and
nonluminous dinoflagellates and their predators under controlled laboratory conditions,
the luminous types show an increase in survival over nonluminous types. The luminous
flash probably deters the predator, conferring survivorship on the species that possess this
property. Other escape flashes are evident in shrimp, squid, and the Cypridina, but have
received little controlled study.
Some luminous fish have a light organ that houses a symbiotic culture of luminous
bacteria. Some fish (the "flashlight" fish) have even adapted a lid to move over the light
organ to create a flashing of the otherwise continuous emission. Other fish appear to
contain a luminous system that is the same as in Cypridina, i.e., the luciferins and
luciferases cross-react. The luminescence probably serves as a mating signal, or as a lure.
In the Australasian glowworm, which is the larva of a dipteran (fly), in contrast to the
firefly, which is a coleopteran (beetle), the luminescence has a lure function. The larva
hangs in the middle of a web and emits a blue glow. Small, winged insects are attracted to
the glow, ensnared in the web, and devoured.
Bioluminescence, therefore, appears to be a biological chemiluminescence that has
been adapted and refined to function in ways advantageous for all species that possess it.

100

>- 80
f-
Vi
Z
W
~ 60
w
>
~ 40
--'
W
Ct:
Fig. 13-3. Firefly bioluminescence spectra in vivo. (a)
Photurus pennsylvanicus; (b) Photinus pyralis; (c) Pyro-
phorus plagiothalamus. [Redrawn from H. H. Seliger and
W. D. McElroy, Light: Physical and Biological Action, 500 600
Academic Press, New York (1965).] WAVELENGTH (nm)
396 Chapter 13

Seliger summarized the biological functions of bioluminescence by the "four P's":

predation, protection, prenuptial (ftrefly communication), and perfidy (the insectivorous
ftrefly).

13.1.4. Mechanism
The color of the emitted light ranges from the red of the railroad worm through the
deep blue characteristic of most of the marine creatures. Thus, the ftrst questions that
might arise in the mind of an investigator are: What is the chemical reaction and the
identity of the molecule responsible for the light emission? Why are the colors different?
Are all the bioluminescence systems the same or similar in mechanism?
It is relatively straightforward to Pleasure the spectral emission of the biolumines-
cence from an organism in vivo, and this is usually characterized by the wavelength
position of the spectral emission maximum, which we will designate by the symbol "B'
Table 13-1 shows that many types of organism have a range of "B; often the "B is found to
be species-dependent.
However, it is considerably more difftcult to obtain extracts from an organism that in
solution will react to produce the same bioluminescence. Extensive efforts are usually
needed to obtain the chemically pure components. If the pure bioluminescence reaction
has high efftciency, as measured by the quantum yield of bioluminescence (see Section
13.2), then the steps leading to light emission can be characterized in a chemical sense.
Since a chemiluminescence reaction is one that proceeds with sufficient release of energy
to produce a molecule in the system in an electronically excited state, bioluminescence
can be viewed as enzyme-catalyzed chemiluminescence.
To return to the first of the above questions, it was discovered by Boyle in the 17th
century that air was required for the bioluminescence of bacteria and fungi, and it is now
known that oxygen in some form is involved in all bioluminescence systems. At the end of
the last century, Dubois found that two components could be extracted from the firefly
light organ, one with hot water and the other with cold water. The cold-water extract,
which was heat-labile, he named luciferase, and the hot-water extract, luciferin. He made
a similar observation with the clam Pholas but found that a cross-reaction between clam
luciferin and ftrefly luciferase, and vice versa, did not occur. (2) Sometimes it has been
found that the luciferins (designated LH2 in Table 13-2) and the luciferases ofbiolumines-
cent members of a class do cross-react; this is also occasionally observed between classes,
and even in more distant relationships. The ability to cross-react or not is a consequence of
the chemical structure of the luciferin, which is quite different, say, between the clam and
ftrefly, and therefore the luciferin from the one is not a substrate for the luciferase of the
other species.
In Table 13-2 the properties of some of the better studied bioluminescence systems
are given. It needs to be emphasized here that the symbol LH2 is used for the organic
molecule substrate in the reaction. It is a generic symbol and we will see later that the
chemical structure of this molecule is usually different between the types of biolumines-
cent organisms. Table 13-2 also shows the remarkable biochemical variability from one
type of reaction to the next. Most exhibit the classical enzyme (luciferase)-substrate
reaction, but some have a precharged system called a "photoprotein," which will be
explained later. Some systems only involve luciferase and luciferin with 02; others use
several proteins. Some luciferases are small proteins, whereas others are large, multi-
Bioluminescence 397

TABLE 13-2. Properties of Bioluminescence Systems

Proteins involved
Source Substrates (M,.; subunits)
Bacteria Luciferase (76,000; 0., (3)
Lumazine protein (21,200)
Dinoflagellates Luciferase (130,000)
LH2 -binding protein
Coelenterates: Hydrozoa, Ctenophores Photoprotein (21,000)
Green-fluorescent protein (30,000)
Anthozoa Luciferase (35,000)
Green-fluorescent protein (52,000; 0.2)
LH 2-binding protein (18,500)
Crustaceans: Cypridina LH z, O2 Luciferase (68,000; «6)
Firefly Lh2 , ATP, O 2 Luciferase (60,000)

subunit enzymes. Also, each luciferase is specific for each type of organism, and no
relatedness has been observed between the luciferases.
Thus, the biochemistry of bioluminescence is analogous to the biological potpourri
remarked on by Harvey, i.e., each system has found its own pathway to the [mal product,
light emission. Many believe that this argues for bioluminescence being a case of con-
vergent evolution. However, what has emerged in recent years is that, in contrast to the
diversity of its biology and biochemistry, the subject is unified by its chemical mecha-
nism. To expose this common mechanistic thread that runs through each system, we will
first need to consider some basic physical principles, and then consider in depth several of
the well-studied bioluminescence mechanisms.

13.2. CHEMILUMINESCENCE REACTIONS(8)

Many chemical reactions may proceed with the release of sufficient free energy to
excite one of the participants into an excited electronic state, but visible light does not
result unless certain conditions are met. In the gas phase at low pressure, the rate of
energy release on bond formation can be faster than the rate of dissipation of excess
energy by collisions with the surrounding molecules. What is called "nonadiabatic cross-
ing" to the excited state manifold of products is a likely occurrence, with the consequent
generation of visible or infrared chemiluminescence.
In solution, the time scales for collision with solvent molecules and for atomic
motions are the same. In other words, the vibrational interaction with the medium is so
predominant that fast cooling by collision with solvent molecules prevents any highly
energetic or "hot" species from being formed as the result of reactions involving the
rearrangement of atoms. Therefore, in 1939, Weiss proposed that the only process fast
enough to produce chemiluminescence excitation in solution is electron transfer, and this
idea has received much experimental support. (9) An example is the one-electron oxidation
of a donor molecule (D) by an acceptor (A) where there is a large redox energy difference
between the pair. On placing two electrodes into a solution of D and A, electron transfer
pairs of opposite charge can be generated under very precisely controlled conditions, and
their mutual annihilation produces "electrochemiluminescence" if there is sufficient
398 Chapter 13

energy to place one of the products in its excited electronic state. There now exists a
detailed understanding of electrochemiluminescence reactions, and a good example is the
electron transfer between the radical anion of diphenylanthracene, DPA" - , and its radical
cation, DPA" +, where about 10% of the products appear in the excited state DPA*):

DPA"+ + DPA"- ~ 0.2 DPA* + 1.8 DPA (13-1)

The asterisk is used here to mean that a compound is in an excited singlet state, the same
state from which fluorescence occurs.
Bioluminescence reactions are of course not electrochemical, and so we must look
elsewhere for chemical reactions that can be used as bioluminescence models. The appro-
priate model reactions are the oxygen oxidations of certain organic compounds that give
good yields of chemiluminescence, and it is well established that some of these proceed
through the formation of an intermediate cyclic peroxide, often a dioxetanone (I), which
contains a high chemical energy content by virtue of its highly strained ring. The decom-
position of dioxetanones, however, usually generates the product ketone more in its triplet
than singlet state:
(I )

Rtf
R

R>= °
R
+ CO 2 (13-2)

A triplet product does not yield chemiluminescence, but rather its energy is degraded to
heat.
Therefore, it was proposed by Schuster and by McCapra (see Ref. 8) thatan electron
transfer is the necessary excitation step in the chemiluminescent breakdown of dioxetanes
or other cyclic peroxides that give excited singlets in substantial yield. The best evidence
for this is in the chemiluminescence of diphenoyl peroxide (II), as catalyzed by the
presence of an easily oxidizable and fluorescent "activator" such as rubrene. The mecha-
nism has been shown to be recombination of a rearranged radical ion pair within a solvent
cage:

II + Rubrene _
~
o
~06-;-
~
Rubrene

j -co, (13-3)

~
o o ~o + Rubrene
*
-
~07
0 "'0
Rubrenet
Bioluminescence 399

Yields of the excited activator up to 10% have been reported. This mechanism has been
dubbed "chemically induced electron exchange luminescence" (CIEEL). It is believed
that the decomposition of dioxetanes yields excited states by a similar mechanism, but this
is not so firmly established.
Prior to knowledge of the chemical mechanism, it is often useful to categorize
chemiluminescence reactions on the basis of spectra. If the emitting molecule is a product
of the reaction, the process is called "direct chemiluminescence":

A + B~ C* (13-4)
(13-5)

The spectral distribution of direct chemiluminescence corresponds to the fluorescence of

C [Eq. (13-5)]. The excitation energy, however, may end up on another molecule D,
which, although present in the solution, is neither a reactant nor a product of the chem-
istry. This is called "indirect" or "sensitized" chemiluminescence by analogy to sen-
sitized photochemistry and sensitized fluorescence:

A +B+ D~ C + D* (13-6)
(13-7)

The chemiluminescence emission in this case is the same as the fluorescence of D. D is

called the chemiluminescence "sensitizer" to correspond to the photochemical sensitizer
in the reverse of Eqs. (13-7) and (13-6). In both cases the sensitizer is the component
unchanged by the chemiluminescent or photochemical process.
There are some cases of sensitized chemiluminescence in which the nature of the
primary excited species C is established or strongly inferred, and the excitation of D is by
one of the energy transfer processes discussed in Chapter 1. Electron transfer interactions
between excited molecules are also possibilities. In most cases the identity of C is not
established; the most outstanding example is the reaction of oxalyl chloride and hydrogen
peroxide in the presence of a dye, which is one of the most efficient chemiluminescence
reactions known. This is the reaction used in the commercially available "light-stick."
The quantum yield is an important characteristic of chemiluminescence reactions,
just as it is for photochemical reactions (Chapter 2). In order to preserve the Second Law
of Photochemistry, i.e., the correspondence between one photon and one molecule, we
define the chemiluminescence quantum yield with respect to one of the chemical partici-
pants, e.g., molecule A in Eq. (13-4), <Pc(A), as the number of photons emitted divided
by the number of A reacted. Note that this is the inverse of the definition of quantum yield
when photons are entering the system. For a bioluminescence quantum yield, we use the
symbol <PB(A), or <PB(C) if the quantum yield is with respect to the number of molecules
of C produced, etc.
The <Pc for direct chemiluminescence is the product of two efficiencies, one for the
fluorescence of the eruitter C, <PF(C) [Eq. (13-5)], and the other <PE' the efficiency for
putting the product into its excited state [Eq. (13-4)]. We can then write:

(13-8)

since, as written, Eq. (13-4) has unit stoichiometry.

400 Chapter 13

We have already seen in Chapter 1 how a number of competing processes reduce <PF
below unity. For reasons that are not understood, <PE is found to be much less than unity
for most chemiluminescence reactions in solution, but in the gas phase, and also surpris-
ingly for some bioluminescence reactions for which reliable information is at hand, <PE
can be very nearly unity. Thus, any mechanism proposed for excitation in biolumines-
cence has to account for this important observation.
The <PF obviously cannot exceed unity, and this means that <PF(C) > <pdC). In other
words, any proposed emitter, as a necessary qualification, has to have a fluorescence
quantum yield that is greater than the chemiluminescence quantum yield. Additionally,
the parameters describing its fluorescence spectral distribution, i.e., the position of the
maximum AF , the width at half-height, etc., have to be the same as for the chemilumines-
cence spectrum.

13.3. GENERALIZATIONS ABOUT BIOLUMINESCENCE REACTIONS(lO,ll)

The term bioluminescence system will be used to describe the combined sequence of
reactions used by an organism to generate light. The present state of knowledge of the
details of six or more bioluminescence systems, the chemical structure of the luciferins
(LHz), and the characteristics of the luciferase enzymes (E) suggest a general sequence of
reactions that lead to the emission of light. Usually some preliminary steps occur to
"prepare" the luciferin for the chemiluminescence reaction, but all the steps have to be
untangled from the total system before the excitation step can be isolated for study. Some
bioluminescence systems use all of the general sequence of reactions and some omit
certain steps.

13.3.1. Activation
After the initial binding of the substrates to luciferase, the first step in the sequence is
the activation of luciferin,

E: LHz + A ~ E: LHzB + C (13-9)

where A is a cofactor and C is some product, e.g., adenosine triphosphate (ATP) and
pyrophosphate, respectively, in the case of firefly bioluminescence. We use the symbol
E: to designate the substrate interaction with the enzyme, usually known to be
noncovalent.

13.3.2. Oxygenation
The second step in our generalized scheme is oxygenation;

E: LHzB + 0z ~ E: LBHOOH (13-10)

The Cypridina bioluminescence system leaves out the activation step; oxygenation of its
luciferin is its first reaction.
Bioluminescence 401

13.3.3. Excitation
The peroxide breaks down via a mechanism described above to produce some excited
molecule:

E:LBHOOH + M ~ E:LBO* + P + M (13-II)

where M is another cofactor that may be needed to set off this process, like Ca2 + in the
bioluminescent jellyfish extract (aequorin) reaction (see below), and P is other products,
like H20, or CO 2 for the firefly. Light emission results from deexcitation of E: LBO*.

13.3.4. Turnover
Finally, the organism might want to use the enzyme again, i.e., the luciferase is
released ready to react again:

E: LBO ~ E + LO + B (13-12)

Apparently the firefly does not need to do this in vivo, since it has all the luciferase it
needs for a lifetime of one-time reactions. In vitro, firefly luciferase and most of the other
types of luciferase can be made to tum over.
It is clear that to understand a bioluminescence reaction in chemical terms, the nature
of the oxygenated species produced by Eq. (13-10) must first be established, and then the
excitation process [Eq. (13-II)] can be proposed on the basis of model chemilumines-
cence reactions. In the following sections we shall examine the three bioluminescence
systems that have received the most extensive studies, and show what sequence of
reactions is involved, and where understanding of the excitation reaction has progressed to
up to this time.

13.4. FIREFLY BIOLUMINESCENCE(12)

Firefly luciferin (LH 2 ) is D( -)-2-(6' -hydroxy-2' -benzothiazolyl)-.:l2-thiazoline-4-car-

boxylic acid (III):
m
o
rtJrNHNI'OH
HO~S S

The reference mark (#) indicates the optically active center, and this turns out to be the
"business end" of the molecule. The first step in the sequence is the activation, by a
luciferase-catalyzed reaction, of the carbonyl group on LH2 with ATP to form the lu-
ciferyl adenosine monophosphate derivative, LH2AMP, with the release of pyrophos-
phate, PPj:

E:LH2 + ATP~ E:LH 2AMP + PPj (13-13)

402 Chapter 13

13.4.1. Chemical Mechanism

Oxygenation proceeds in several steps, the first possibly being Eq. (13-14):

E : LH2AMP + O2 - ? E : LHOOHAMP (13-14)

anp there is ample but indirect evidence that this hydroperoxide cyclizes to the diox-
etanone (IV). This has been suggested to decompose by a CIEEL mechanism:

_oJOc:H::0o- [orfX>==<'r']
0-0

TIl
~ S S

j (13-15)

COz +
(A>-(fJ- [O'>==<'j' co; 1 0'" ~ s s

:2:

Here the excitation step is an intramolecular electron transfer, the benzothiazolyl

portion of the molecule playing the role of the easily oxidizable fluorescent activator
required for the CIEEL mechanism.
The intermediacy of the dioxetanone (IV) has been inferred from mass spectrometric
or radioactivity analysis of the enzymatic reaction products with heavy-atom or radioac-
tive labels.

(13-16)

A cyclic peroxide intermediate leads to the product CO 2 containing one atom of 180.
It took many years before these labeling results were agreed upon because the experiments
are very difficult to perform. (13)

13.4.2. The Light Emitters

The <PB(LH 2) has been measured to be almost unity, the highest found for any
bioluminescence system, and the quantum yields for oxygen consumption <h(02) and
Bioluminescence 403

cOz production <h(CO z) are about the same. We can infer that <PE is unity, but unfortu-
nately the product V [Eq. (13-15)] is too unstable to have its fluorescence measured
accurately.
The in vivo and in vitro bioluminescence emission spectra are the same, but their
maxima AB are species-dependent. Figure 13-3 shows three bioluminescence spectra
representing the range of fIrefly bioluminescence encountered, AB = 552-575 nm. The in
vitro spectra show the same species dependence, and since all these reactions use the same
synthetic fIrefly luciferin, the spectral differences must originate in some property of the
luciferase. It has been suggested that the binding site on the enzyme must differ slightly
from one species of luciferase to the next, e.g., in electrostatic charge or dielectric
constant. This produces what is analogous to a "solvent" effect on the fluorescence
spectrum of the emitting product V (Chapter 2).
Firefly luciferin can be made to chemiluminesce if it is reacted with oxygen in an
aprotic solvent, i.e., one like dimethylsulfoxide that has no available protons for hydrogen
bonding. If the solution is kept very dry and made very basic, the chemiluminescence is
yellow-green, with practically the same emission spectrum as the bioluminescence. In a
weaker base, a red emission of lower quantum yield is found. The yellow-green emission
(AC = 562 nm) comes from VI, a dianion that in solution may add a proton at a rate in
competition with the radiative rate to form the red-emitting species (AC = 615 nm) V.(l4)

CJOrH1j*
H+U (13-17)

CJO(~17
Thus, the yellow bioluminescence is believed to come from VI. In the active site on
luciferase, the protonation reaction VI ~ V is hindered unless the pH is lowered below 6
(Fig. 13-4). Raising the temperature or adding heavy metal ions (Znz+ or Cdz +) also
induces the red emission band, and these are all conditions that must relieve the hindrance
caused by luciferase on the protonation reaction [Eq. (13-17)]. (15)
Before we go on to consider the other bioluminescence systems, let us fIrst summa-
rize what we have learned about the firefly reaction. A single protein, fIrefly luciferase,
possesses all the required catalytic properties to generate yellow light: activation, oxy-
genation, and excitation. In addition, the protonation of the emitter is slowed to the point
that it does not compete with the emission rate (about 108 s - 1). Firefly luciferin was the
first one to have its structure fully determined. The luciferin is activated by reaction with
ATP. The bioluminescence reaction products are HzO, COz, and a molecule whose
fluorescence spectral properties correspond to the bioluminescence. The fluorescence of
product V also has an acid-base relationship between two emitting species, as does the
chemiluminescence of luciferin in dimethylsulfoxide. Luciferases purified from different
firefly species give slightly different emissions, ranging from yellow through orange,
404 Chapter 13

suggesting that the product environment on each enzyme is not the same. We shall
encounter these same or modified properties when we deal with the bioluminescence
systems from the lower orders.

13.5. CYPRIDINA AND COELENTERATE BIOLUMINESCENCE(10,16)

A substantial body of chemical knowledge is now available for many marine bio-
luminescence systems. Many of them use an imidazopyrazine luciferin (VII):

O~

d
O
:L~Y
N
I H
NH2
+
NH2

H
:wI

HO
d~XO~
O~O
0
I I OH

]X

Although the mechanism of oxidation is the same, the substituent groups (R) differ between
Cypridina (VIII) and "coelenterate-type" (IX) luciferin (also called "coelenterazine"). A
remarkable fact about these bioluminescence systems that use imidazopyrazine luciferin is
that they have achieved their biological adaptations by certain changes at the molecular
level that we can easily study and understand.
We should further contrast these bioluminescence systems with the firefly in that the
latter is only found in insects that are terrestial, whereas systems utilizing the im-
idazopyrazine luciferin are found in seven phyla (out of a recognized 30 total animal
phyla) all in the marine environment. Firefly luciferase alone contains all the functions
required for the bioluminescence system on one protein, but other systems, the coelenter-
ates in particular, utilize a variable number of enzymes and proteins with an additional
variety of packaging.

13.5.1. Chemical Mechanism

Spectral data to be discussed later, and multilabeling experiments similar to those
alluded to in the firefly discussion, have now established the oxidation mechanism of Eq.
(13-18):
Bioluminescence 405

~O
0~R3 O)----=(R3 O~R3
N N -H+ 02
)0
.. JIN.&X
N
RfNXR R f . X :2 R, N R2
'H 2

IOL:{
(R~NXRZoy). (R~N-OyJ
(13-18)

R3
N NH L 7NXN +CO Z E
N
fNX -
I RZ R, N RZ

xn· XI· X

In basic aprotic solvents a chemiluminescence reaction occurs with the natural lu-
ciferin or with synthetic analogs, e.g., Rl = p-hydroxyphenyl, phenyl, or H; R2 = methyl
or H; R3 = methyl. With Cypridina luciferase only the natural substituents give bio-
luminescence, but the coelenterate luciferases appear to be less restrictive. The chem-
iluminescence spectra are green to yellow, and are distinct from the bioluminescence
spectra, which range from blue to green depending on the substituents used.

13.5.2. Enzymology
Th~ enzymology of the Cypridina system is the simpler. The system is named from
its primary source, a small ostracod, Cypridina hilgendorfii, a crustacean that is particu-
larly abundant in Japanese waters. The system is also found in many species of luminous
fish. When stimulated, the ostracod secretes the luciferin and luciferase from separate
glands into the surrounding seawater, so it can be said that the in vivo and in vitro
reactions are the same. There are three reactions that constitute this system, and the one
enzyme luciferase catalyzes them all: oxygenation, excitation, and release for turnover:

E + LH2 ~ E:LO* + CO2 (13-19)

The coelenterates are a diverse group of organisms made up of two main phyla,
Cnidaria and Ctenophora. Within Cnidaria are the classes Anthozoa, such as the soft
coral, Renilla reniformis ("sea pansy"), Hydrozoa, e.g., the jellyfish Aequorea ae-
quorea, and Scyphozoa. The bioluminescence systems from Renilla and Aequorea are the
ones that have received the most study. The bioluminescence systems of all of the
coelenterates involve at least three of the general reactions of Eqs. (13-19) and (13-12).
Although the chemical mechanism is apparently the same for all of them, they differ
among the classes of the Cnidaria in the steps of the activation reaction, and among the
phyla as to the nature of the excitation step. In fact these systems now require us to
introduce more complications into the simple picture represented by Eqs. (13-9) to
406 Chapter 13

(13-12). Their study also revealed, for the ftrst time, the phenomenon of "sensitized
bioluminescence" [cf. Eqs. (13-6) and (13-7)].
The ftrst step in anthozoan bioluminescence is activation:

(13-20)

Luciferyl sulfate, LH2S (LH2 is IX), and a sulfokinase, Es ' have been obtained from
Renilla. The activating reaction [Eq. (13-20)] requires the coenzyme 3' ,5'-diphospho-
adenosine, PAP, and liberates the corresponding sulfate, PAPS. Free luciferin is rapidly
autooxidized in solution, but in the cell it is protected by a binding protein, EB , from
which it is released by the addition of Ca2+ .

(13-21)

It is thought that this Ca2+ -binding step controls the light flash in the whole animal.
The next steps for coelenterate luciferin are oxygenation and excitation, both on the
anthozoan luciferase, E.

+ O2 ~ E: LHOOH
E: LH2 (13-22)
E:LHOOH~ E:LO* + CO 2 (13-23)

The product of oxygenation is not detectable, as it quickly breaks down to the excited final
product, E: LO* .
There is a remarkable difference between the Anthozoa and the Hydrozoa in this last
step, so much so that the protein involved in the latter is not called luciferase but
"photoprotein," specifically "aequorin," "mnemiopsin," etc., depending on its source.
What is extracted from the jellyftsh is the oxygenated species EA : LHOOH, the photopro-
tein (aequorin). The Hydrozoa do not use the activation or binding reactions [Eqs. (13-20)
and (13-21)]. Adding Ca2+ to photoproteins triggers the bioluminescence:

EA : LHOOH + Ca2+ ~ EA : LO* + CO2 (13-24)

The stable oxygenated complex may be regenerated by adding LH2 (IX) to apoaequorin.

13.5.3. The Light Emitters

For the reaction of Cypridina luciferin in basic aprotic solvents, the chemilumines-
cence emission is from the monoanion XI, as the R2 guanidino group is not positively
charged under these basic conditions. For luciferin of structure IX, the emitting product is
a dianion or trianion by virtue of dissociation of the RI or R3 to the phenolates. These
chemiluminescence spectra have maxima at much longer wavelengths (Ae > 500 nm)
than the bioluminescence spectra.
The in vitro reaction of vrn with Cypridina luciferase is very efficient, <VB = 0.3,
and the blue bioluminescence is a broad featureless spectrum with AB at 460 nm. This
spectrum is the same as the fluorescence of XII bound to luciferase. In aqueous solution
Bioluminescence 407

100

)- 80
I-
iJi
z
w
Fig. 13-4. Effect of pH on the firefly bio- ~ 60
luminescence spectra in vitro. The spectrum
W
with a maximum at 616 nm is normalized to >
unity, but in absolute terms its contribution is ~ 40
...J
only one-half that of the band having its max- W
0::
imum at 562 nm. The pH values are indicated 20
on each spectrum. [Redrawn from H. H. Se-
liger and W. D. McElroy, Light: Physical and
BiologicalAction, Academic Press, New York 500 600 700
(1965).] WAVELENGTH (nm)

free XII is not fluorescent. Thus, the binding site on Cypridina luciferase must be one that
allows rapid protonation of the initially formed product to produce XII* , and that inhibits
the nonradiative loss of its excited state energy.
The in vitro bioluminescence of coelenterate luciferin has a relatively low efficiency,
<PB = 0.07. The emission is broad with hB = 480 nm (Fig. 13-4). However, the fluores-
cence of the neutral species (XlI) derived from natural coelenterate luciferin has hF = 402
nm in aqueous solution. From a number of studies with luciferin analogs it has been
concluded that, in contrast to the Cypridina reaction, the monoanion (IX) is the emitter of
coelenterate bioluminescence. In basic aprotic solvent, IX has hF = 480 nm and <PF =
0.07. The active site of coelenterate luciferase must then hinder the excited protonation of
the initially formed Xl to XII; a similar property has been described for firefly luciferase.
Whether protonation of the excited product is slowed or not is a mechanism for adaptation
that would tune the bioluminescence spectrum to satisfy some biological function. For
example, a spectral shift might satisfy a requirement for optimal transmission of the
bioluminescence signal in the marine optical milieu, or for a better match to the spectral
sensitivity of the target organism, etc.

13.5.4. Sensitized Bioluminescence

There is a remarkable difference between the in vivo bioluminescence spectra from
the Ctenophores and some of the Cnidaria. The latter have a narrow structured emission,
hB = 509 nm, identical to the fluorescence of a "green-fluorescent protein" (GFP) that
can be purified from extracts of these organisms. The in vivo and in vitro biolumines-
cences and this fluorescence are compared in Fig. 13-5. The green-fluorescent protein is
very fluorescent, <PF = 0.8, and has no luciferase activity. If present in the in vitro
reaction at a concentration of more than 10- 5 M, the bioluminescence emission corre-
sponds exactly to the green fluorescence and also to the in vivo hB' and the <PB is enhanced
to over 0.2. Thus, it is evident that the green-fluorescent protein sensitizes the biolumines-
cence.
The Ctenophores, on the other hand, have a broad bioluminescence emission, hB =
408 Chapter 13

1.0
z Fig. 13-5. Absorption (left full line) and
o
ill 0.8 fluorescence (right full line) spectra of the
~ green-fluorescent protein from Renilla. The
w dashed line is the in vitro bioluminescence
5 0 .6
w with Renilla luciferase and a synthetic lu-
u
~ 0.4 ciferin. On addition of green-fluorescent
tIl
a:: protein (1ILM), the relative bioluminescence
~ 0.2 yield and spectrum change to match that of
tIl
<! the right full line. [Redrawn from W. W.
Ward, Energy transfer processes in bio-
luminescence, Photochem. Photobiol. Rev.
WAVELENGTH,nm 4, 1-57 (1979).]

490 nm, similar to the in vitro spectrum with luciferin (IX). No green-fluorescent protein
is found in the Ctenophores or the Scyphozoans.
The mechanism of the sensitization process is unknown. Dipole-dipole energy trans-
fer has been suggested [Eq. (13-25)], but it is also possible that the green-fluorescent
protein functions as a fluorescent activator in a CIEEL mechanism [see Eq. (13-3)].

EA:LO* + GFP~ EA:LO + GFP* (13-25)

The fluorophore of the green-fluorescent protein is XIII. Its structure suggests that it
is related to the luciferin. The biosynthesis of all these molecules (VIII, IX, and XIII) is
probably cyclization of natural amino acids whose side chains can be recognized in the R
substituents.

XIII

13.5.5. Control(17)
If Caz + is added to a mixture of Renilla luciferin bound to ~ [Eq. (13-21)] and
luciferase, the bioluminescence intensity (Fig. 13-6, soluble) rises in a few tenths of a
second and falls very slowly, unlike the in vivo flash (Fig. 13-6) obtained on stimulating
the animal. This discrepancy is explained by the observation that the light-giving organ,
or photocyte, of Renilla is made up of much smaller structures called "lumisomes,"
membrane-enclosed vesicles about 0.2 fA-m in diameter, containing EB , LH z , E, and GFP.
On adding Caz + to a preparation of the lumisomes, a flash much more like the in vivo
bioluminescence (Fig. 13-6) is observed. It is suggested that the proteins are packaged
within the lumisome so that the transient concentration of Caz + is controlled in such a
way as to produce the in vivo bioluminescence kinetics.
Bioluminescence 409

Fig. 13-6. Light flash from the Renilla bioluminescence system on the addition
of Ca2 + at zero time. [Redrawn from 1. M. Anderson and M. J. Cormier,
Lumisomes, the cellular site of bioluminescence in coelenterates, 1. Bioi. Chern. o I 2
248, 2937-2943 (I973).J TIME (5)

Thus, we see in organisms that lie lower on the evolutionary scale than the firefly, a
bioluminescence system involving a multienzyme package. Adaptations of the bio-
luminescence spectrum are also evident and are understood at the molecular level. These
adaptations involve excited state protonation kinetics as well as energy transfer to a
sensitizing acceptor molecule. In the firefly system, all the reactions have been "pack-
aged" onto the same macromolecule, i.e., activation, light reaction, and spectral adapta-
tion. We shall now look at bioluminescence in a lower form of life, the bacteria.

13.6. BACTERIAL BIOLUMINESCENCE(l8,19)

The luminous bacteria are mostly found in the ocean, both free living and symbiotic
with certain fish. Most of the marine types are classified into two genera, Photobacterium
and Vibrio, and these are the ones that have received the most biochemical study. It can be
said that bacterial bioluminescence is one of the longest studied mechanisms in bio-
chemistry, dating from the 17th century, as mentioned in the Introduction, when Boyle
observed what we now know as the oxygen requirement. In the bioluminescence field,
this system is the one that has received the most attention, probably because the mecha-
nism defies elucidation and remains a challenge for future research. An advantage in the
study of bacterial bioluminescence is that the cells can be grown in kilogram quantities by
routine fermentation procedures, the yield of luciferase is high, and the luciferase is
relatively easy to purify.
410 Chapter 13

~OH
I Hb
)CI N1;N
yO
~ NH
N
o

13.6.1. Chemical Mechanism

Light emission can be obtained in vitro if reduced riboflavin monophosphate (i.e.,
flavin mononucleotide, FMNH2, XIV) is added to a solution of luciferase and a long-
chain aliphatic aldehyde. In vivo, tetradecanal is probably the aldehyde "luciferin," and
the flavin is not free but bound to a protein, probably luciferase itself. Two known
products of the in vitro reaction are FMN (XVI) and an aliphatic acid of the same chain
length as the aldehyde.
No CO2 can be detected in the products, so a dioxetanone intermediate is not
considered to occur. Instead the intermediate species of high energy content [Eq. (13-22)]
is believed to be some type of strained peroxide. Indeed, if aldehyde is omitted from the
reaction, a flavin-oxygen-Iuciferase complex is formed that has a distinct absorption
spectrum, "-A = 370 nm, and it can be stabilized for as long as 24 h. Because many
dihydroflavin derivatives have similar absorption spectra, it was proposed that this com-
plex was a dihydroflavin hydroperoxide substituted at the 4a position (XV in Scheme 4).
This idea has now been proven by measuring the 13C NMR of the complex using synthetic
FMNH2 containing i3C at specific positions around its isoalloxazine ring. The resonance
from the 4a C is the only one that shows a distinct chemical shift in the NMR of the
complex (Fig. 13-7), and this can be interpreted as due to a highly strained peroxide
substitution at this position. (20)
In all the bioluminescence systems this complex is the only oxygenated species to be
directly characterized in this way. Furthermore, in other flavoenzyme oxidation reactions
this same species has been proposed to be a key hydroxylating species, but the luciferase
complex is the only one stable enough for NMR characterization. The luciferase-bound

Fig. 13-7. Carbon-13 nuclear magnetic resonance spectra of flavin

enriched at the 2 and 4a positions with l3C, in the presence of
bacterial luciferase. The top is for FMNH2 and the middle is after
adding O2, and this represents the intermediate II. Only the reso-
nance from the 4a-C is significantly affected. The lowest spectrum is
after full oxidation to FMN. The unlabeled bands are the contribu-
160 140 120 100 80 tions from the natural abundance l3C in the luciferase molecule.
PPM(S) (Redrawn from Ref. 20.)
Bioluminescence 411

species XV is called "intermediate II." It reacts with aldehydes to produce biolumines-

cence possibly by the CIEEL mechanism.

13.6.2. Enzymology
Within the cell we can consider activation to be the generation of FMNH2 coupled to
the oxidation of NADH by an oxidoreductase, ER :

~:NADH + FMN + H+ ---? ~:NAD+ + FMNH2 (13-26)

The luminous bacteria contain very high concentrations of luciferase, and thus the
FMNH2 is rapidly shuttled into the oxygenation reaction before it has a chance to be
oxidized directly and uselessly by oxygen:

(13-27)

Although not as stable, this luciferase-bound peroxide is very analogous to the photopro-
teins (e.g., aequorin) discussed in the last section.

13.6.3. Lumazine Protein(21)

The in vitro reaction of bacterialluciferase yields a broad emission spectrum with ""B
about 495 nm. However, for Photobacterium cells in vivo the spectrum is distinctly
different with ""B around 475 nm (Fig. 13-8). This is due to the sensitization of the
reaction by a second protein identified in these cells and named "lumazine protein"
because the ligand is 6,7-dimethyl-8-ribityllumazine (XVII). This lumazine derivative is a
natural metabolite, the precursor to riboflavin in procaryotes.
If lumazine protein is included in the in vitro reaction, the bioluminescence spectrum

'-'--'-'--'-~--'-'--'-'-~I
P pnosphorelJm

>-
!::
(f)
z
~
w
....
~

Fig. 13-8. (top) Shift of the in vitro bioluminescence z

o
(dashed curve) on including lumazine protein in the reac- ~ ~~~--+--+--+-~~~~--+~
tion (reformed from the apoprotein, 25 11M and the ~
w
lumazine derivative, 811M). (lower) Fluorescence of the
lumazine derivative (dashed curve), lumazine protein w
>
(dots), and the reconstituted hololumazine protein (solid i=
<t
line). [From J. Lee, L. A. Carreira, R. Gast, R. M. Irwin, ...J
W
P. Koka, E. D. Small, and A. J. W. G. Visser, Properties a:
of a lumazine protein from the bioluminescent bacterium
Photobacterium phosphoreum, in: Bioluminescence and
Chemiluminescence (M. A. DeLuca and W. D. McElroy, 400 600

eds.), pp. 103-112, Academic Press, New York (1981).] WAVELENGTH (nm)
412 Chapter 13

is blue-shifted, and if the lumazine protein concentration is at least 10 j.LM, the spectrum
becomes identical to that of the in vivo bioluminescence, which is also the same as the
fluorescence of lumazine protein itself, AF = 475 nm, thus proving the sensitizer function
of lumazine protein. The mechanism of sensitization is not known, but it is obviously not
the same energy transfer process as in the coelenterate reaction [Eq. (13-25)], where the in
vitro bioluminescence emitter is proposed to be the energy donor with GFP the acceptor.
The important differences are that the bioluminescence spectral shift caused by lumazine
protein is to a higher energy, not a lower one (Fig. 13-8), and there is only a small overlap
between the absorption spectrum of lumazine protein and the in vitro bioluminescence
spectrum. Furthermore, the inclusion of lumazine protein can enhance the <l>B up to three
times with values as high as <l>B = 0.25 being observed.
The FMN product is fluorescent, but its spectrum with AF = 535 nm disqualifies it
from being the origin of the bioluminescence emission. On the other hand, in the course of
the bioluminescence reaction another highly fluorescent but transient species has been
detected. It has a fluorescence spectrum identical to the bioluminescence spectrum and its
concentration bears a kinetic relationship to the bioluminescence intensity (Fig. 13-9). A
number of reduced compounds related to FMNH2 also produce bioluminescence with
luciferase and aldehyde, but with a shifted bioluminescence spectrum, within the range
475-560 nm, depending on the compound. These reactions also show a fluorescent
transient with a spectrum corresponding to the shifted bioluminescence spectrum, proving
that this transient is indeed the in vitro bioluminescence emitter and that its structure,
presently unknown, must be derived from the reduced substrate used.
To account for the complex kinetics of the in vitro bioluminescence, it has been
proposed that the reaction is sensitized by the fluorescent transient, but this still remains to
be thoroughly established. Again, it is an attractive possibility to consider that a CIEEL
process occurs with either the fluorescent transient or lumazine protein, if present, as
fluorescent activators.
Bioluminescent bacteria produce a continuous light emission at a level controlled
only by the metabolic status of the organism, e.g., oxygen concentration, carbon sources,
and the like. In the brightest bacteria a large fraction of the total metabolic energy is
allocated to the generation of bioluminescence. How bioluminescence has evolved in the
procaryotes is a fascinating questionP2) The small molecules involved are common:
FMNH2, the lumazine derivative, and the fatty aldehydes. Flavin peroxide intermediates
are believed to occur in many flavoenzyme reactions. The lumazine apoprotein could be

- Fluorescence
EXCitation --- Bioluminescence

1.0
>-
f-
iii Fig. 13-9. Excitation (left curve) and fluorescence spectra
zw of the fluorescent transient fonned in the bacterial bio-
f-
~ luminescence reaction. This fluorescence and the bio-
luminescence spectra (dashed curve) are identical. [From I.
B. C. Matheson and J. Lee, Kinetics of bacterial bio-
300 400 500 luminescence and the fluorescent transient, Photochem.
WAVELENGTH (nm) Photobiol. 38,231-240 (1983).]
Bioluminescence 413

the binding protein for the lumazine metabolite. Oxidoreductase is also ubiquitous in the
procaryotes. The bioluminescent bacteria seem to distinguish themselves from other
procaryotes primarily by the presence of luciferase. By the application of DNA recombi-
nant techniques it may soon become possible to trace the lineage of the bacterial lu-
ciferases and answer the question of how and why the bacteria "invented" this enzyme.
No new principles have come out of studies of the many other bioluminescence
systems. Some of these systems use H 20 2 instead of 02' and a few others have had their
luciferins characterized: they are aldehydes in the earthworm and snail systems, and a
tetrapyrolle in the dinoflagellates. Sensitization of the bioluminescence is found in the
shrimp system and probably the earthworm.

13.7. LOW-LEVEL LUMINESCENCE, BIOLOGICAL

CHEMILUMINESCENCE(23,24)

Although we have been discussing the terms light emission, chemiluminescence, and
quantum yield, we have not yet considered the important device that has made the
accurate measurement of these quantities possible. This is the photomultiplier tube (Chap-
ter 1), a vacuum tube device containing a photosensitive cathode coating on the inside of
the front window, which emits an electron on absorption of a photon, the electron then
being attracted by a voltage gradient to a series of electroemissive dynodes. These
dynodes, made of a Ag!Mg alloy, have a low work function, and when one electron hits
the surface of the dynode it will knock off another two or more electrons. These are
attracted by a positive voltage to the next dynode where they multiply again, and so on
down the line of 10 or more dynodes until out of the anode comes a burst of 106 or more
electrons, a negative pulse of current, corresponding to that initial impinging photon.
The photomultiplier serves as the "eye" of a television camera, and was developed
primarily for this purpose and for scintillation counting in nuclear physics in the early
1940s. Since that time, photomultipliers have been developed to be more and more
sensitive to light, and to have a wide spectral sensitivity. At the present time it is possible
to make a photomultiplier device that can detect less than 100 photons! s coming from a
chemical reaction contained in a volume of about 1 cm3 . This means that light emission
would be measurable for a fast reaction with a <Pc of below 10- 17. It has been observed
that a great many exothermic chemical reactions, and a number of biological systems, do
indeed emit light at a very low level.
The human eye is also a very good light detector, comparable in sensitivity, but not
accuracy, to the photomultiplier. Therefore, weak light emission from a number of
chemical reactions has been known for hundreds of years. Many investigations, particu-
larly by groups of Russian workers during the 1930s, have shown that a variety of living
systems give weak visible and ultraviolet luminescence. They used Geiger tubes for
detection, and this luminescence is sometimes referred to as mitogenetic or Gurvich
radiation, after its discoverer. This luminescence is usually associated with rapidly grow-
ing or respiring cells, e.g., onion root tips, dividing yeast cells, white blood cells (leuko-
cytes), liver mitochondria and microsomes, contracting skeletal and heart muscle, etc.
In order to distinguish this low-level luminescence from what we have been discuss-
ing up to now, we shall use the term' 'biological chemiluminescence." Unlike most cases
414 Chapter 13

of bioluminescence, where the light serves some definite biological function, such as
communication in the fireflies, biological chemiluminescence has no apparent advantage
to the organism and may just be the result of a very minor energy wastage through an
inevitable Maxwell-Boltzmann probability of populating an excited state of a product
molecule in a very exergonic reaction. It could be, on the other hand, that it is a very
efficient chemiluminescent reaction, but occurring to only a minor extent in the total
reaction. Oxygen and radical intermediates appear to be involved in biological chemilumi-
nescence.
Bioluminescence and biological chemiluminescence can be better compared in terms
of the light emission from the functional biological unit, i.e., from a single bacterium or
the organelle of a higher organism. Luminous bacteria each typically emit 103 photons!s,
while a lumisome from Renilla will produce a flash of peak intensity about 102 photons! s,
and a single dinoflagellate, which probably contains thousands of organelles, when stimu-
lated will give off more than 108 photons. In contrast, the rapidly respiring mitochondria
or leukocytes undergoing phagocytosis have an average light level far less than 1 photon!s
per cell. Actually, the distinction between bioluminescence and biological chemilumines-
cence is blurred by the fact that the so-called dark mutants of luminous bacteria are only
dark because they give off 1O,000-fold less light than the wild-type strain, i.e., at levels
that could be regarded as that of biological chemiluminescence.
Not much is known about the mechanism of biological chemiluminescence, but since
the oxidation of lipids and other unsaturated hydrocarbons gives rise to a chemilumines-
cence of low efficiency, it is proposed that such a process is probably the basis for
biological chemiluminescence. This luminescence requires molecular oxygen and can
often be stimulated by the addition of hydrogen peroxide. A mechanism based on the
chain oxidation of a hydrocarbon is

R' + O2 - ROO' (13-28)

ROO' + RH - ROOH + R' (13-29)
ROO' + ROO' - RO* + RO + O2 (13-30)

The RO* is an excited ketone, and the fluorescence yields of these are extremely low,
resulting in the very low overall yield of light. However, this can be enhanced up to 106
times by adding certain fluorescent substances.

13.8. APPLICATIONS OF BIOLUMINESCENCE(25)

13.8.1. Ultrasensitive Analysis

Some efficient chemiluminescence reactions are widely used as emergency light
sources and as novelties. The use of bioluminescence reactions in this way, however, is
impractical due to the cost of preparation and the difficulty of storing enzymes over long
periods. However, bioluminescence reactions are finding increasing application in the
analysis of biological materials. The detection and assay of ATP by the firefly reaction,
for instance, is more sensitive than any other method. Often a substance may be assayed
Bioluminescence 415

indirectly by coupling it via some enzyme to a bioluminescence cofactor like ATP or

NADPH. Together with bacterialluciferase the latter has wide applicability since many
assays in the clinical laboratory are presently made by NAD(P)H coupling, and the
luciferase technique can improve sensitivity by many orders of magnitude.
Bioluminescence-linked immunoassays also offer advantages over alternative tech-
niques in sensitivity, stability, simplicity of measurement, and avoidance of the use of
radioisotopes with their accompanying hazards and disposal problems. For a biolumines-
cent immunoassay an antibody (Ab) has to be produced for the substance to be measured
(the hapten, H), and also the hapten has to be covalently attached to a component of the
bioluminescence system, e.g., a protein (E). The assay reaction is

E-H:Ab + H~ H:Ab + E-H (13-31)

Only the free E-H will have activity in the bioluminescence reaction. A potential advan-
tage of these systems is in the development of homogeneous assays where the complete
assay is achieved without separation steps. This goal is being sought using sensitized
bioluminescence systems, i:e., where the acceptor protein (A-H) also has to be freed for
the correct bioluminescence to be generated.
The use of the firefly system to assay ATP is the most widely used bioluminescence
method. We will describe this one in detail to show how these procedures are done. For
very low amounts of ATP, the firefly assay is practically the only available technique. If a
limiting quantity of ATP is added to a solution of firefly luciferin and luciferase, a flash of
light is produced, which, if detected by a photomultiplier linked to a current meter and
recorder, produces a curve like that shown in Fig. 13-10. A straight line results when total
photons emitted in the first 15 s is plotted against the amount of ATP added, and this is the
calibration line. For an unknown sample, the ATP content can be calculated from its light
yield. This method has been used to find out how much ATP is present in nerve cells,
plant seeds, and bacteria, and modification of the procedure by enzyme' coupling enables
the assay of other important cofactors like AMP, cyclic AMP, and pyrophosphate. En-
zymes that use ATP can be determined, e.g., creatine phosphokinase. One of the most
important applications has been to the assay of biomass via the ATP content, to which it is

Fig. 13-10. Light flash from the firefly bioluminescence in

vitro reaction obtained by injecting ATP into a mixture of
luciferase and luciferin. The vertical direction represents rela-
tive light intensity. For 10 - 9 g of ATP the flash maximum 5
(Io) has an intensity of about 10 10 photons/so TIME (5)
416 Chapter 13

directly related. In a hospital laboratory, for example, a blood or urine specimen can be
rapidly measured for infectious levels of bacteria in this way.
Whole cells of bioluminescent bacteria are used in a method for determining toxic
materials in foods and stream water, since the light emission is quenched by many
poisons. Also, dark mutants of these bacteria can be designed that will back-mutate in
response to specific mutagens, and this has been proposed as a replacement for the Ames
test for carcinogens. Also under this subject of whole-cell luminescence, some clinical
laboratories are using the biological chemiluminescence associated with leukocytes to
monitor phagocytic competence.
The aequorin reaction is able to detect Ca2 + concentrations down to 10- 12 M. Many
physiological processes involve rapid changes in Ca2 + concentration, and the aequorin
reaction has therefore found use in studying such things as Ca 2 + ion transients in mus-
cular contraction. In this context, an application of a different type can be performed that
is highly dependent on bioluminescence, i.e., coupling an image intensifier tube to a
microscope to observe the localization and time responses of light emission in a bio-
luminescent organism. For nonluminescent ones microinjection of aequorin allows obser-
vation of Ca2 + concentration fluxes in single organisms. By this method nerve impulses
across single dinoflagellates and waves of Ca2 + concentration changes on fertilization of
single eggs have been visualized.

13.8.2. Recombinant DNA Techniques

The techniques of modem molecular biology are providing new approaches to solv-
ing problems in the bioluminescence field just as they are in most of the biological
sciences. The primary sequence of aequorin has been determined by cloning as well as by
chemical methods. However, the luciferases from bacteria and firefly were resistant to
having their sequence done by chemical means, and so their primary structures have had
to be inferred from their DNA base sequences. Recombinant DNA methods are also
making available several of these proteins in much larger quantities than have hitherto
been extractable from natural sources. Also as with other proteins, specific amino acid
alterations can be introduced as a means for highlighting those residues that influence the
bioluminescence activity.
While there are no unique applications of recombinant DNA methods to biolumines-
cence problems at this time, it is in the reverse direction that some novel developments are
occurring. Bacterialluciferase has been used as a marker for detecting which colonies of
treated E. coli have been transformed to contain a certain foreign gene. The two genes for
the luciferase subunits, called LuxA and LuxB, are inserted into the transforming DNA,
between the promoter of the foreign gene and the gene itself. Cells that are transformed
will manufacture both luciferase and the foreign gene product when the promoter is turned
on, and will be luminescent. The method is rapid and sensitive. (26)
In another application of bacterialluciferase, the expression of nitrogenase genes in
Rhizobium during stages of its infection of plant cells, has been followed by fusing the Lux
genes after the nitrogenase promoter, and detecting luminescence. Recently, both the Lux
and the firefly genes have been inserted into the genomes of plant cells themselves. Since
the production of luciferase can measured by its bioluminescence yield, then in these
transformed plants the bioluminescence might provide the ability to rapidly and relatively
precisely measure the rates of gene expression. (27)
Bioluminescence 417

13.9. REFERENCES

1. E. N. Harvey, Bioluminescence, Academic Press, New York (1952).

2. E. N. Harvey, Living Light, Princeton University Press, Princeton, N.J. (1940).
3. H. H. Seliger, Origin of bioluminescence, Photochem. Photobiol. 21, 335-361 (1975).
4. J. Buck, Functions and Evolutions of Bioluminescence, in: Bioluminescence inAction (P. J. Herring, ed.),
pp. 419-460, Academic Press, New York (1978).
5. J. W. Hastings, Biological diversity, chemical mechanisms, and the evolutionary origins of bioluminescent
systems, J. Mol. Evol. 19, 309-321 (1983).
6. J. E. Lloyd, Bioluminescence and Communication, in: How Insects Communicate (T. A. Sebeok, ed.), pp.
164-183, Indiana University Press, Bloomington (1977).
7. H. H. Seliger, A. B. Lall, 1. E. Lloyd, and W. H. Biggley, The colors of firefly bioluminescence. I.
Optimization model. II. Experimental evidence for the optimization model, Photochem. Photobiol. 36,
673-688 (1982).
8. K.-D. Gundermann and F. McCapra, Chemiluminescence in Organic Chemistry, Springer-Verlag, Berlin
(1987).
9. L. R. Faulkner and R. S. Glass, Electrochemiluminescence, in: Chemical and Biological Generation of
Excited States (W. Adam and G. Cilento, eds.), pp. 191-227, Academic Press, New York (1982).
10. W. W. Ward, General Aspects of Bioluminescence, in: Chemi- and Bioluminescence (J. G. Burr, ed.), pp ..
321-358, Marcel Dekker, New York (1985).
11. P. J. Herring (ed.), Bioluminescence in Action, Academic Press, New York (1978).
12. M. A. DeLuca and W. D. McElroy, Purification and properties of firefly luciferase, Methods Enzymol. 57,
3-15 (1978).
13. O. Shimomura, Mechanism of bioluminescence, in: Chemical and Biological Generation of Excited States
(W. Adam and G. Cilento, eds.), pp. 249-276, Academic Press, New York (1982).
14. E. H. White, J. D. Miano, C. J. Watkins, and E. J. Breaux, Chemically produced excited states, Angew.
Chern. (Engl.) 13, 229-243 (1974).
15. L. J. Bowie, R. Irwin, M. Loken, M. DeLuca, and L. Brand, Excited-state proton transfer and the
mechanism of action of frrefly luciferase, Biochemistry 12, 1852-1857 (1973).
16. F. H. Johnson and O. Shimomura, Introduction to the bioluminescence of Medusae, with special reference
to the photoprotein aequorin, Methods Enzymol. 57, 271-291 (1978).
17. M. J. Cormier, K. Hori, and J. M. Anderson, Bioluminescence in Coelenterates, Biochim. Biophys. Acta
346, 137-164 (1974).
18. M. M. Ziegler and T. O. Baldwin, Biochemistry of Bacterial Bioluminescence, in: Current Topics in
Bioenergetics, Vol. 12 (R. D. Sanadi, ed.), pp. 66-113, Academic Press, New York (1980).
19. J. Lee, The mechanism of.bacterial bioluminescence, in: Chemi- and Bioluminescence (J. G. Burr, ed.), pp.
401-437, Marcel Dekker, New York (1985).
20. J. Vervoort, F. Muller, J. Lee, W. A. M. Van der Berg, and C. T. W. Moonen, Identification of the true
carbon-13 nuclear magnetic resonance spectrum of the stable intermediate II in bacterial luciferase, Bio-
chemistry 25, 8062-8067 (1987).
21. A. J. W. G. Visser and J. Lee, Excited state probes in bacterial bioluminescence, in: Excited State Probes in
Biochemistry and Biology (A. G. Szabo and L. Masotti, eds.), Plenum Press, New York (1988).
22. H. H. Seliger, The evolution of bioluminescence in bacteria, Photochem. Photobiol. 45,291-297 (1987).
23. R. C. Allen, Biochemiexcitation: Chemiluminescence and the study of biological oxygenation reactions, in:
Chemical and Biological Generation of Excited States (W. Adam and G. Cilento, eds.), pp. 309-344,
Academic Press, New York (1982).
24. D. Slawinska and J. Slawinski, Low level luminescence from biological objects; and, Applications of
bioluminescence and low-level luminescence from biological objects, in: Chemi- and Bioluminescence (J.
G. Burr, ed.), pp. 495-531, 533-601, Marcel Dekker, New York (1985).
25. L. J. Kricka, P. E. Stanley, G. H. G. Thorpe, and T. P. Whitehead (eds.), Analytical Applications of
Bioluminescence and Chemiluminescence, Academic Press, Orlando (1984).
26. J. Englebrecht, M. Simon, and M. Silverman, Measuring gene expression with light, Science 227, 1345-
1347 (1985).
27. D. J. O'Kane, W. L. Lingle, J. E. Wampler, M. Legocki, R. P. Legocki, and A. A. Szalay, Visualization
of bioluminescence as a marker of gene expression in rhizobium-infected soybean root nodules, Plant Mol.
Bioi. 10, 387-399 (1988).
Index

Absorbance, 34 Bioluminescence (cont' d)

Acridine orange, 87 biological chemiluminescence, 413
Action spectra bioluminescent
bacteria killing, 43 organisms, 392
carotenogenesis, 294 systems, 397
circadian rhythm, 201 chemiluminescence, 397
conidiation, 297 CIEEL mechanism, 399,402,411, 412
defined, 42 coelenterate, 404
erythema (skin), 159 Cypridina, 404
extraretinal photoreception, 220 firefly, 401
germination, 278 gene transfer, 416
inhibition of hypocotyl elongation, 282 generalizations about, 400
mutagenesis, 126 light organs, 393
photoaccumulation, 326 mechanism, 396
photokinesis, 319 occurrence, 392
photophobic response, 319, 329 origin and function, 394
phototaxis, 319, 326 Blackbody radiation, 6
positive phototropic curvature (oat), 294 Blue-light mediated morphogenesis
prevention of flowering, 278 events under blue-light control, 292
retardation of cleavage, 44 photoreceptor( s), 294
reversing germination, 278
vision, 3, 258 Carbonyl compounds, 54
(6-4)-Adducts, 71, 128 Carcinogenesis (skin), 166
Aequorin, 401, 416 [3-Carotene, 186, 338, 352
Anaerobic photosensitization, 84 Carotinoids, 353
AP endonuclease, 123, 124 Chemical restoration, 119
Aquatic ecosystems, 140 Chemiluminescence, 397,413
Aschoff's rule, 223 Chlorophyll, 87, 351
Auxotroph, 128 Chloroplast, 333, 375
Chromophore (defined), 5
Beer's law, 34, 55, 114, 142 CIEEL mechanism, 399,402,411,412
Bilirubin, 183 Circadian rhythms
Bimolecular reaction, 52 action spectra, 201
Biological chemiluminescence, 413 Aplysia,218
Biological clock, 208 Aschoff's rule, 223
Biological consequences of stratospheric ozone biochemical rhythms, 202
depletion cellular mechanisms, 201
animals, 151 clock mutants, 206
plants, 148 defmed,196
Bioluminescence drug sensitivity, 210
applications of, 414 entrainment to
bacteria, 409 artificial light, 194

419
420 Index

Circadian rhythms (cont'd) DNA strand breaks, 72

entrainment to (cont' d) 7-Dehydrocholesterol, 51
environmental cycles, 197 Dihydrocytosine, 70
free-running rhythms, 195 Dihydrothymine, 68
general features, 194 Dipole-dipole interaction, 32
genetic control, 204 Dipole moment, 29
house sparrow, 223 Dose modification factor, 116
humans, 208
inhibitors of, 203 Early receptor potential, 259
jet lag, 208 Electromagnetic
location of circadian clocks, 206 spectrum, 3
locomotry activity, 195, 221 theory of Maxwell, 4
melatonin production, 218 Electron spin, 25
organization in multicellular organisms, 206 Electroretinogram, 258
photoreceptor for, 200 Energy transfer
pineal gland, 218 dipole-dipole, 32
sensitivity of mice to endotoxin, 210 Forster-type, 32, 60
stopping the clock with bright light, 224 resonance excitation, 32
temperature compensation, 196 triplet-triplet, 33
Circular dichroism (CD), 40 Entrainment (circadian rhythms), 194, 197
Clock (circadian), 196, 206, 221 Environmental photobiology
Crassulacean acid metabolism, 379 aquatic ecosystems, 140
Cyclobutane-type dimers biological consequences of stratospheric ozone
absorption spectrum, 69 depletion, 147
defmition, 68 biologically effective UV-B, 139
monomerization, 68 evolution of terrestrial organisms, 138
structure, 49, 69 protection from solar UV radiation
Cysteine, 49 animals, 145
5-S-Cysteinyl-6-hydrothymine, 49 aquatic organisms, 145
Cytochrome, 367, 371 terrestrial plants, 143
solar radiation, 136
Do, 114, 115 stratospheric ozone, 138
D37 , 114, 115 vegetation canopies, 141
Dq ,115 Enzyme inactivation
DNA-DNA cross-links, 72 kinetics, 64
DNA photochemistry theories, 62
DNA-DNA cross-links, 72 Eosin Y, 87
DNA-protein cross-links, 49, 68, 70, 72 Error-prone DNA repair, 127, 128
strand breaks, 72 Evolution
DNA-protein cross-links, 49, 68, 70, 72 stratospheric ozone (effects on), 138
DNA repair terrestrial organisms, 138
daughter-strand gaps, 124 Extrapolation number, 115
double-strand breaks, 124 Excision repair
excision repair, 120 base excision, 123
genes of E. coli, 117 nucleotide excision, 120
mismatch repair, 130 Excited states
photoreactivation, 119 atomic orbitals, 27
postreplication repair, 123 energy levels, 27
skin (human), 162 molecular orbitals, 26
spontaneous mutagenesis, 130 pi orbital, 27, 54
UV radiation mutagenesis, 126 production, 25
DNA replication sigma orbital, 27
mismatch repair, 130 Extraretinal photoreception
spontaneous mutagenesis, 130 Aschoff's rule, 223
UV radiation mutagenesis, 127 circadian rhythms, 218, 221, 223
Index 421

Extraretinal photoreception (cont' d) Humans (cont'd)

house sparrow, 221 skin [see Skin (human)]
invertebrates, 217 sunburn, 158
mammals, 227 Hydrated electron, 58, 60, 61
photoperiodism, 224 6-Hydroxy-5-hydrouracil, 50, 67
reproductive system, 224 5-Hydroxymethyl uracil, 68
sites for, 227 8-Hydroxypsoralen, 88
vertebrates, 218 Hypericin, 87
Eye Hypersensitivity, 187
light damage, 266
responses to light, 255 Inducible responses
role in circadian rhythms, 222 adaptive response, 118
structure and function, 231 heat shock system, 119
UV radiation effects on, 189 oxidation system, 119
vision, 231ff. SOS system, 117
Internal conversion, 30
Ferredoxin, 367 Intersystem crossing, 30
Finsen (Neils), 179 Isopsoralen, 88
Firefly, 401
First law of photochemistry, 5, 47, 80 Jablonski diagram, 30, 51
First-order reaction, 31, 54 Jaundice (neonatal), 183
Flash photolysis, 35, 52, 58, 59, 61, 62 Jet lag, 208
F1avinoid, 137, 140, 144, 296
Fluence, 54 Lambert-Bouguer-Beer law, 34, 55, 114, 142
Fluorescence Lamella, 375
defined, 31 Lasers
depolarization, 40 average power, 13
lifetime, 38 defined, 10
polarization, 39 frequency doubling, 11
quenching, 32 medical uses, 24
spectra, 36 mode-locking, 13
F1uorophore (defined), 32 peak power, 13
5-Formyluracil, 68 properties of, 11
Forster mechanism, 32, 52, 60, 356 Q switching, 13
Fovea, 234, 238 schematic diagram, 11
Frequency doubling, 11 Laws of photochemistry
Furanocoumarin, 84, 104 first, 5, 47, 80
second,399
LIAC, 294
Genetic control
Lifetimes
circadian rhythms, 204
singlet state, 52
DNA repair, 117
triplet state, 52
Glycosylases, 118, 123, 124
Light-induced absorbance change (LIAC), 294
Goeckerman regimen, 181
Light scattering, 22
Grana, 375
Light sources
arc lamps, 6, 7
Hematoporphyrin, 87, 182 fluorescent lamps, 6, 8
Herpes simplex infection, 182 incandescent lamps, 6
Heteroaddition reactions, 66 lasers, 6, 10
Humans photochemical lamps, 6
circadian rhythms, 208 Light-tissue interactions, 22
DNA repair, 162 Liquid-holding recovery, 75
jet lag, 208 Luciferase, 396
photomedicine, 155ff. Luciferin, 396
phototherapy, 179 Lumazine protein, 411
422 Index

Lumiflavin, 50 Pauli exclusion principle, 25

Lupus erythematosus, 185 Phosphorescence
defmed,31
Melatonin production, 218
spectra, 36
Melanogenesis, 163
Photoaccumulation, 310
Methionine, 84
Photoacoustic spectroscopy, 41
8-Methylpsoralen, 88
Photochemistry
Methyltransferase, 1I8
amino acids, 58, 63
5-Methyoxypsoralen, 186
bimolecular reaction, 52
8-Methoxypsoralen, 88, 181, 186
carbonyl compounds, 54
Methylene blue, 87
chain reactions, 54
Mismatch DNA repair, 130
cholesterol, 50
Molar extinction coefficient
cycloaddition, 49
amino acids, 63
7-dehydrocholesterol, 51
defined, 34, 56, 63
deoxyribose, 66
peptide bonds, 63
DNA,72
Molecular orbitals
fundamentals, 47
defined, 26
isomerization, 51, 183, 185
ethylene, 28, 29
kinetics, 54
formaldehyde, 28, 29
linear addition, 49
Mutagenesis
nucleic acids, 65
assays, 127
overall reactions, 49
errors in
photofragmentation, 50
DNA repair, 127
photohydration, 50, 66
DNA replication, 127
photooxidation, 50
genetic control, 129
photorearrangement, 51
kinetics, 128
polymerization, 54
mismatch repair, 30
primary reactions, 51
postreplication repair, 124
proteins, 57, 60
spontaneous, 130
purines, 66
types of mutations, 127
pyrimidines, 66
UV radiation, I 26ff.
quantitation, 75
Mutations (types of)
retinal, 51
back, 128
riboflavin, 50
base substitution, 127
RNA,74
frameshift, 127
unimolecular reaction, 52
missense, 127
uracil, 50
nonsense, 127
Photochromic adaptation, 298
transition, 127
Photodetectors
transversion, 127
properties of, 21
suppressor, 128
quantum, 20
n (extrapolation number), 115 thermal, 20
Near-UV radiation, 75, 1I9 Photodinesis, 310, 322
Photodispersal, 310
Optical activity, 40 Photodynamic action, 80
Optical components Photohydration, 50, 66
filters, 15 Photoimmunology
lenses, 13 delayed hypersensitivity, 187
mirrors, 13 immediate hypersensitivity, 187
monochromators, 15 Photokinesis, 308
optical fibers, 15 Photon (defined), 4
polarizers, 18 Photomedicine
Optical density, 34 effects of sunlight on the eye, 189
Optical rotatory dispersion (ORO), 40 effects of sunlight on skin
Ozone (see Stratospheric ozone) acute, 158
Index 423

Photomedicine (cont' d) Photosensitization (cont'd)

effects of sunlight on skin (cont' d) evolution and, 104
adverse, 171 humans, 100
beneficial, 178 lipids, 88
chronic, 164 mechanisms (diagnostic techniques), 85
optical properties of skin, 156 natural photosensitivity, 10 I
photoimmunology, 187 nucleic acids and derivatives, 89
photoprotection, 184 photocarcinogenesis, 100
phototherapy, 179 photodynamic action, 80
Photometric units, 12 plants, 97
Photomorphogenesis proteins, 90
blue-light mediated, 291 radicals, 82
diversity of, 274 singlet oxygen, 83, 85
light as a source of information, 273 type I (radical) mechanism, 82
photochromic adaptation, 298 type II (singlet oxygen) mechanism, 83
phytochrome mediated, (see Phytochrome) types of photosensitizers, 86-88
UV-radiation mediated, 296 viruses, 92
Photomovement Photosynthesis
cells, 315, 318 accessory pigments, 354
Chlamydomonas, 325 anoxygenic, 359
chloroplasts, 333 antenna molecules, 356
cyanobacteria, 319 as an energy source, 380
ecological significance, 341 brown algae, 354
Halobacterium halobium, 323 C3 carbon cycle, 377
methods, 311 C4 carbon cycle, 377
organelles, 332 carbon dioxide fixation, 377
photoaccumulation, 310 cyanobacteria, 354
photodinesis, 310, 322 cytochrome, 371
photodispersal, 310 delayed fluorescence, 357
photokinesis, 308 diatoms, 354
photophobic response dinoflagellates, 354
defined, 309 electron transport pathways, 361, 365
enhancers, 331 green algae, 355
inhibitors, 331 green bacteria, 353, 361, 362
photosensory transduction mechanisms, 317 higher plants, 355
phototaxis, 309 light harvesting, 350, 355
phototropism, 310, 316, 337 overview, 349 '
polarotropism, 310, 333 oxygen evolution, 371
primary photoreactions, 320, 324, 328, 330 oxygenic, 365
terminology, 307 P680, 369
Stentor coeruleus, 329 pheophytin, 369
Photoperiodism, 224 phosphorylation, 373
Photophobic response, 309 photochemical conversion, 355
Photoprotection, 75 photorespiration, 379
Photoreception photosynthetic unit, 358
extraretinal, 215 photosystem I, 365
phototropism, 337 photosystem II, 368
vision, 231 ff. pigments, 351
Photoresponse (defined), 308 plastoquinone, 370
Photosensitization purple bacteria, 353, 359
amino acids, 90 QA and QB, 370
anaerobic reactions, 84 reaction center, 357
animals, 98 red algae, 354
biological applications of, 101 structural organization, 375
cells, 93 Z-scheme, 365
424 Indel

Photosynthetically active radiation (PAR), 138 Race tube culture, 200

Phototaxis, 309 Radiative transitions, 25
Phototechnology Radicals, 59, 82
actinometry, 21 Radiometric units, 12
light sources, 5 Recovery phenomena
light-tissue interactions, 22 liqnid holding recovery, 115
optical components, 13 medium-dependent resistance, 116
photodetectors, 20 minimal medium recovery, 116
radiometry, 21 plateau phase recovery, 116
Phototherapy, 179 Resonance excitation transfer, 32
Phototropism, 310, 316, 337 Retina
Phycoerythrobilin, 352 light damage to, 266
Phycomyces, 316, 337 structure, 231, 236
Phytochrome Retinal, 51
differences between P, and Prp 288 Rhodopsin, 244, 245, 249, 252, 254, 255
discovery, 278 Riboflavin, 50, 87
events under phytochrome control, 280 Rose bengal, 87
mode of action, 290
photochemical reactions, 286 Second law of photochemistry, 399
photoequilibria, 283 Second order reaction, 54
properties of, 285 Shuttle vectors, 128
Pineal gland, 188, 218 Singlet oxygen
Polarotropism, 310, 333 defined, 83
Polarization, 18, 39 lifetime, 85
Polymorphous light eruptions, 185 light emission, 85
Primary photochemical reactions quenching, 85
excited state electron transfer, 53 trapping, 85
hydrogen atom abstraction, 54 type II mechanism, 83
photoionization, 53 SI units, 2
production of free radicals, 53 Skin (human)
singlet-singlet energy transfer, 52 carcinogenesis, 166
triplet-triplet energy transfer, 52 diseases, 179
Prototroph, 128 DNA repair, 162
Psoralen, 84, 88, 181, 186 effects of sunlight
Psoriasis acute, 158
description of, 180 adverse, 171
Goeckerman regimen, 181 beneficial, 178
PUVA therapy, 181, 186 chronic, 164
Pulse radiolysis, 58 melanogenesis, 163
PUVA, 181, 186 microscopic anatomy, 160
Pyronine Y, 87 optical properties, 156
photosensitizing agents, 185
Quanta (defined), 4 quick-tanning preparations, 186
Quantum mechanics, 27 structure, 157
Quantum yield sunburn, 158
amino acids, 63 Solar urticaria, 185
bioluminescence, 399 Solar radiation
defmed, 31, 48, 63 biological effectiveness, 138
inactivation, 114 effects on humans, 155ff.
proteins, 64 environmental photobiology, 135ff.
thymine, 68 filtration by vegetation canopies, 141
uracil, 68 how organisms protect themselves
Quenching animals, 143
fluorescence, 32 aquatic organisms, 143
singlet oxygen, 85 plants, 143
Quick-tarming preparations, 186 penetration into water, 140
Index 425

Solar radiation (cont'd) Spore photoproduct, 70

photomedicine, 155ff. Stokes shift, 36
spectrum, 3, 136 Stratospheric ozone
transmission through leaf epidennis, 149 depletion (biological consequences), 147
Spectrophotometers fonnation, 138
absorption, 35 seasonal variation, 138
luminescence, 37 spectrum, 173
Spectroscopy Stroma, 375
absorption, 34 Suicide enzyme, 118
action, 42 Sunburn, 158
Lambert-Bouguer-Beer law, 34 Sunscreens, 184
luminescence, 35 Superoxide radical anion, 58
optical activity, 40 Survival curves
photoacoustic, 41 alpha-beta model, 114
Spectrum dose modification factor, 116
action, 42 E. coli B, 112
atomic, 27 E. coli K-12, 113
benzyl radical, 59 parameters, 115
bioluminescence, 395, 411, 412 target theory, 114
black body radiation, 6
carbazole, 36 Target theory, 114
a-chymotrypsin, 61 Tautomeric shifts (DNA bases), 130
deuterium lamp, 7 Thiopyronin, 87
electromagnetic, 3 4-Thiouridine, 75
epidennal transmittance, 137 Thymine, 49
filter transmission, 16 Thymine dimer, 49
flavin (nuclear magnetic resonance), 410 Thymine glycol, 67
flavinoid, 137 5-Thyminyl-5,6-dihydrothymine, 70
fluorescent lamps, 9 Time scales in photophysics, photochemistry, and
gennicidal lamp, 8 photobiology, 48
green fluorescent protein from Renilla, 408 Transition
halogen lamp, 7 dipole, 30
hydrated electron, 59 moment, 29
mercury lamp, 7 Translesion DNA synthesis, 129
nucleic acid, 137 Transmittance, 34
ozone, 137 Triplet-triplet
phenylalanine, 57 annihilation, 53
photoacoustic, 41 energy transfer, 70
photoflood lamp, 3 Tumor therapy, 182
photopic vision, 3 Type r (radical) reaction, 82
photosynthetic organisms, 353 Type II (singlet oxygen) reaction, 83
protein, 137
retinal (II-cis), 249 Ubiquinone, 361
rhodopsin, 249 Unimolecular reaction, 52
Schiff bases, 248 Units
shadelight (under a broadleaf canopy), 283 photometric, 12
sunlight, 3, 137, 283 radiometric, 12
thymine, 69 Sr,2
thymine dimer, 69 Uracil, 50
tryptophan, 57 UV-A, 5, 59, 80, 84, 149, 156,296
tryptophan radical, 59 UV-B, 5, 80, 137, 138, 156,296
tyrosine, 57 UV-C, 5, 51, 80
tyrosine radical, 59 UV radiation effects
xenon lamp, 7 animals, 145
Speed of light, 4 beneficial effects, 51, 178
Spontaneous mutagenesis, 130 destruction of oxygen evolution by plants, 372
426 Index

UV radiation effects (cont'd) eye (cont'd)

DNA repair, 119 fovea, 234, 238
environmental photobiology, I 35ff. internal transmitter hypothesis, 262
eyes, 189 mechanism of visual excitation, 262
humans, I 55ff. overview, 231
inducible responses, 117 patch clamp, 262
mutagenesis, 126 recording of photoreceptor responses, 260
photomorphogenesis, 296 retina, 231, 236, 266
plants, 143 rhodopsin, 244, 245, 249, 252, 254, 255
recovery phenomena, 115 rods, 237, 253, 262
skin, 155 rod outer segments (ROS), 241
solar spectrum, 3, 136 visual cycle, 257
survival curves, 112 visual pigments, 246, 248, 251
UV -radition-mediated morphogenesis vitamin A, 243, 246
events under UV radiation control, 296 Vitamin A, 243, 246
photoreceptor(s), 297 Vitamin 0 3 ,51, 178
Vitiligo, 179
Vibrational relaxation, 30
Vision Xanthophylls, 354
bleaching intermediates of rhodopsin, 255 Xeroderma pigmentosum, 122, 163, 166-168, 170
calcium model, 263 X radiation
cGMP model, 264 chemical restoration, 119
cones, 237, 253 diffraction (rod outer segment), 243
dichroism of outer segments, 250 medium-dependent resistance, 116
disk membrane, 240, 243, 250 mutagenesis, 128, 129
early receptor potential, 259 plateau phase recovery, 116
electroretinogram, 258 repair of
eye DNA double-strand breaks,. 126
light damage, 266 potentially lethal damage, 116
responses to light, 255
structure and function, 223 Z-scheme, 365