Methods in Enzymology Volume 58 Cell Culture

Preface
Many of the problems that have caught the interest and imagination of
biochemists are studied best with cultured cells. We offer in this volume,
in a format familiar to investigators in biochemistry, the general techniques necessary for working with cells in culture and illustrate such
general methods with specific examples from the large variety of cells that
have been cultivated.
The tools and methods for cell culture are presented in Part I. Part II
provides a group of specialized techniques that are useful for many of the
applications that biochemists and other investigators with their widely
different approaches may require. Part III is concerned with specific
methods for specific cell types that have been chosen to represent the
wide range of cells that may now be prepared.
There is some duplication in the presentations. For example, portions
of certain methods are repeated in one or another form both in Part I and
Part III. We believe that this repetition is necessary to convey faithfully to
the reader a complete method of proven effectiveness. Additionally, we
hope that a heuristic effect will be achieved that will enable investigators
unfamiliar with cell culture to assess what is available and to predict what
might be most suitable for their own purposes.
WILLIAM B. JAKOBY
IRA H. PASTAN
xiii
Contributors to V o l u m e L V I I I
Article numbers are in parentheses following the names of contributors.
Affiliations listed are current.
RONALD T. ACTON (17, 18), Department of
ROBERT B. CAMPENOT (25), Department of
Microbiology and the Diabetes Research

and Training Center, University of
Alabama in Birmingham, University Station, Birmingham, Alabama 35294
DOLPH O. ADAMS (43), Department of
Pathology,
Duke Medical
Center,
Durham, North Carolina 27710
W. FRENCH ANDERSON (44), Laboratory of
Molecular Hematology, National Heart,
Lung, and Blood Institutes, National Institutes of Health, Bethesda, Maryland
20014
TSUKASA ASHIHARA (20), Department of
Pathology, Shiga Medical College,
Moriyama-cho, Moriyama City, Shiga
524, Japan
W. EMMETT BARKLEY (4), Building 13,
Room 2E47, National Institutes of
Health, Bethesda, Maryland 20014
PAUL A. BARSTAD (17), Department of Microbiology and the Diabetes Research and
Training Center, University of Alabama in
Birmingham, University Station, Birmingham, Alabama 35294
RENATO BASERGA (20), Department of
Pathology and Fels Research Institute,
Temple University School of Medicine,
Philadelphia, Pennsylvania 19140
MARK M. BASHOR (9), Letterman Army
Institute of Research, Presidio of San
Francisco, San : Francisco, California
94129
SHELBY L. BERGER (42), Section of Cellular
and Molecular Physiology, Laboratory of
Pathophysiology, National Cancer Institute, National Institutes of Health,
Bethesda, Maryland 20014
JANE BOTTENSTEIN (6), Department of Biology, University of California, San Diego, La Jolla, California 92093
NOEL BOUCK (24), Department of Microbiology, University of Illinois at the
Medical Center, Chicago, Illinois 60680
Neurobiology, Harvard Medical School,

Boston, Massachusetts 02115
WILLIAM CARLISLE (54), The Salk Institute,
P.O. Box 1809, San Diego, California
92112
P. COFF1NO (19), Departments of Medicine
and Microbiology, University of California, San Francisco, California 94143
LEwis L. CORIELL (3), Institute for Medical
Research, Copewood Street, Camden,
New Jersey 08103
ROBERT T. DELL'ORco (1), Biochemical Division, The S. R. Noble Foundation,
Route 1, Ardmore, Oklahoma 73401
GIAMPIERO DI MAYORCA (24), Department
of Microbiology, University of Illinois at
the Medical Center, Chicago, Illinois
60680
WILLIAM H. J. DOUGLAS (1,10), Department
of Anatomy, Tufts University School of
Medicine, Boston, Massachusetts 02115
CATHERINE DUFF (27), Genetics Department and Research Institute, The Hospital for Sick Children, Toronto, Ontario,
Canada
ROBERT M. FRIEDMAN (23), Laboratory of
Experimental Pathology, National Institute of Arthritis, Metabolism, and Digestive Diseases, National Institutes of
Health, Bethesda, Maryland 20014
T. V. GOPALAKRlSHNAN (44), Laboratory of
Molecular Hematology, National Heart,
Lung, and Blood Institutes, National Institutes of Health, Bethesda, Maryland
20014
J. W. GRAY (19), Biomedical Sciences,
Lawrence Livermore Laboratory, University of California, P. O. Box 808, Livermore, California 94143
MAURICE GREEN (36), Institute for Molecular Virology, St. Louis University
School of Medicine, St. Louis, Missouri
63110
ix
CONTRIBUTORS TO VOLUME LVIII
Department of
Pediatrics, Division of Medical Genetics,
Washington University School of Medicine, St. Louis, Missouri 63110
P. M. GULLINO (14), Laboratory of
RICHARD G. HAM (5), Department of Molecular, Cellular, and Developmental Biology, University of Colorado, Boulder,
Colorado 80309
EDWARD HAWROT (53), Department of
IZUMI HAYASHI (6), Department of Biology,
University of California, San Diego, La
Jolla, California 92093
W. FRED HINt (39), Department of Entomology, Ohio State University, Columbus, Ohio 43210
BHARATI HUKKU (13), The Child Research
Center of Michigan, Children's Hospital
of Michigan, Detroit, Michigan 48201
ERIC HUNTER (32), Department of Microbiology, The Medical Center, University of Alabama in Birmingham, Birmingham, Alabama 35294
SHARON HUTCHINGS (6), Department of Biology, University of California, San Diego, La Jolla, California 92093
ROGER H. KENNETT (28), Department of
Human Genetics, The Human Genetics
Cell Center, University of Pennsylvania
School of Medicine, Philadelphia,
Pennsylvania 19104
GEORGE KHOURY (34), Laboratory of DNA
Tumor Viruses, National Cancer Institute,
National Institutes of Health, Bethesda,
Maryland 20014
MICHAEL KLAGSSRUN (50), Departments of
Surgical Research and Biological Chemistry, Children's Hospital Medical Center,
Harvard Medical School, Boston, Massachusetts 02115
R. A. KNAZEK (14), Laboratory of
K. S. KOCH (47), The Salk Institute, Post
Office Box 1809, San Diego, California
92112
JEFFREY GRUBB (38),
IRWIN R. KONIGSBERG(45), Department of
Biology, University of Virginia, Charlottesville, Virginia 22901

CHING-Ju LAI (34), Laboratory of DNA
Tumor Viruses, National Cancer Institute,
Maryland 20014
H. L. LEFFERT (47), The Salklnstitute, Post
92112
DAVID W. LEVINE (15), The Cell Culture
Center, Massachusetts Institute of Technology, Cambridge, Massachusetts 02139
JAMES A. MCATEER (10), W. Alton Jones
Cell Science Center, Old Barn Road, Lake
Placid, New York 12946
DON B: McCLURE (6), Department of Biology, University of California, San Diego,
La Jolla, California 92093
GERALD J. MCGARRITV (2, 37), Institute for
Medical Research, Copewood Street,
Camden, New Jersey 08103
WALLACE L. MCKEEHAN (5), Department
of Molecular, Cellular, and Developmental Biology, University of Colorado,
Boulder, Colorado 80309
WILLIAM F. McLIMANS (16), Roswell Park
Memorial Cancer Institute, New York
State Department of Health, Buffalo,
New York 14263
HIDEO MASUI (6), Department of Biology,
JENNIE MATHER (6), Department of Biology, University of California, San Diego,
T. MORAN (47), The Salk Institute, Post
92112
TOSHIO MURASHIGE (41), Department of
Plant Science, University of California,
Riverside, California 92521
SUGAYUKI OHASA (6), Department of Biology, University of California, San Diego,
IRA PASTAN(30), Room B27, Building 37,National Institutes of Health, Bethesda,
Maryland 20014
MANFORD K. PATTERSON, JR. (11), The
Samuel Roberts Noble Foundation, Route
One, Ardmore, Oklahoma 73401
PAUL H. PATTERSON (53), Department of

JOHN PAWELEK (51), Department of Dermatology, Yale University, School of
Medicine, New Haven, Connecticut
06510
D. PERLMAN (7), School of Pharmacy, University of Wisconsin, Madison, Wisconsin
53706
WARD D. PETERSON, JR. (13), The ChildResearch Center of Michigan, Children's
Hospital of Michigan, Detroit, Michigan
48201
LOLA C. M. REID (12, 21), Department of
Molecular Pharmacology and Liver Research Center, Departments of Medicine
and Biochemistry, Albert Einstein College
of Medicine, Bronx, New York 10461
JOHN F. REYNOLDS (41), Department of
Plant Science, University of California,
Riverside, Riverside, California
ANGIE RIZZlNO (6), Department of Biology,
MARCOS ROJKIND (21), Department of Molecular Pharmacology and Liver Research Center, Departments of Medicine and Biochemistry, Albert Einstein
College of Medicine, Bronx, New York
10461
ALBERT W. RUESINK (29), Department of
Biology, Indiana University, Jordan Hall
138, Bloomington, Indiana 47401
MILTON H. SAIER, JR. (49), The Department
of Biology, The John Muir College, University of California, San Diego, La
GORDON SATO (6), Department of Biology,
BERNARD P. SCHIMMER (52), Banting and
Best Department of Medical Research,
University of Toronto, Toronto, Ontario
M5G IL6, Canada
DAVID SCHUBERT (54), The Salk Institute,
P.O. Box 1809, San Diego, California
92112
GINETTE SERRERO (6), Department of Biology, University of California, San Diego, La Jolla, California 92093
CHARLES J. SHERR (35), Laboratory of Viral
Carcinogenesis, National Cancer Insti-
xi
tute, National Institutes of Health,

SEUNG-IL SHIN (31), Department of Genetics, Albert Einstein College of Medicine,
Bronx, New York 10461
WILLIAM F. SIMPSON (13), The Child Research Center of Michigan, Children's
Hospital of Michigan, Detroit, Michigan
48201
WILLIAM S. SLY (38), Department of
Pediatrics, Division of Medical Genetics,
Washington University School of Medicine, St. Louis, Missouri 63110
RALPH E. SMITH (33), Departments of Microbiology and Immunology, Duke University Medical Center, Durham, North
Carolina 27710
GRETCHEN H. STEIN (22), Department of
Molecular, Cellular, and Developmental
Biology, University of Colorado, Boulder,
Colorado 80309
ARMEN H. TASHJIAN, JR. (46), Laboratory of
Toxicology, Harvard School of Public
Health, and Department of Pharmacology,
Harvard Medical School, Boston, Massachusetts 02115
MARY TAUB (49), The Department of Biology, The John Muir College, University of
California, San Diego, La Jolla, California 92093
WILLIAM G. THILLY (15), Genetic Toxicology Group, Department of Nutrition
and Food Science, Massachusetts Institute of Technology, Cambridge, Massachusetts 02139
E. BRAD THOMPSON (48), Laboratory of
Biochemistry, National Cancer Institute,
Maryland 20014
LARRY I . THOMPSON (26), Biomedical Sciences Division L-452, Lawrence Livermore Laboratory, University of California, Livermore, California 94550
GEORGE J. TODARO (35), Laboratory of
Viral Carcinogenesis, National Cancer
Institute, National Institutes of Health,
M, WILLIAMS (47), The Salk Institute, Post
O~ce Box 1809, San Diego, California
92112
KIM S. WISE (18), Department of Microbiology, and the Diabetes Research and
xii
Training Center, University of Alabama in

Birmingham, University Station, Birmingham, Alabama 35294
WILLIAM S. M. WOLD (36), Institute for
Molecular Virology, St. Louis University School of Medicine, St. Louis, Missouri 63110
KEN WOLF (8, 40), National Fisheries
Center-Leetown, Fish and Wildlife Service, Department of the Interior, Route
3, Box 41, Kearneyville, West Virginia
25430
RICHARD WOLFE (6), Department of Biology, University of California, San Diego,
RONALD G. WORTON (27), Genetics De-
partment and Research Institute, The

Hospital for Sick Children, Toronto, Ontario, Canada
REEN WO (6), Department of Biology, University of California, San Diego, La Jolla,
California 92093
ROSALIND YANISHEVSKY (22), Department
of Molecular, Cellular and Developmental Biology, University of Colorado,
Boulder, Colorado 80309
ROBERT K. ZWERNER (17, 18), Department
of Microbiology and the Diabetes Research and Training Center, University of
Alabama in Birmingham, University Station, Birmingham, Alabama 35294
[1]
ASPECTS
OF
TISSUE
CULTURE
LABORATORY
[1 ] P h y s i c a l A s p e c t s o f a T i s s u e C u l t u r e L a b o r a t o r y
By WILLIAM H. J. DOUGLAS and ROBERT T. DELL'ORCO

I. Introduction
The material included in this section is intended to present the basic
requirements necessary for the introduction of cell and tissue culture
techniques into a biochemistry laboratory. The information will be presented as the space and equipment needs for performing the routine operations that are necessary for cell culture production regardless of the size
of the proposed facility. These will be considered under four headings:
1. Cleaning and sterilization facilities
2. Media preparation and storage facilities
3. Work area for aseptic manipulation of cell cultures
4. Equipment for routine cell maintenance
These four topics are generally applicable to any type of proposed cell
culture; however, this presentation deals exclusively with the culture of
mammalian cells. While it will offer a suitable starting point, certain
modifications will be necessary for the cultivation of cells from other
sources, such as invertebrates and plants. More detailed information on
the requirements for these systems can be obtained in recently published
reviews. 1,2
Regardless of the cell system to be employed, the scope of the laboratory facilities will depend largely upon the role planned for cell culture
procedures in the individual investigative program. When only a minor
role is planned, a minimum of space will be dedicated to cell production
and support facilities. When a more active role is anticipated, however,
space requirements will be increased and more elaborate facilities may be
deemed necessary. Thus, the facilities could all be compressed into one
laboratory or separated into individual laboratories each performing only
one function. Whether or not a major involvement is planned, another
factor to be considered in overall space and equipment requirements is the
type of investigations that will be done. For example, cells grown for the
harvest of a biological product, such as a hormone, or for the purification
of a particular enzyme would require large quantities of cells and the
necessary space and equipment requirements for mass culture
o. L. Gamborgand L. R. Wetter, "Plant Tissue CultureMethods." PrairieReg. Lab., Nat.
Res. Counc. Can., Saskatoon, 1975. See also this volume [41].
2 K. Maramorosch, ed., "Invertebrate Tissue Culture: Research Applications." Academic
Press, New York, 1976. See also this volume [39].
METHODS IN ENZYMOLOGY, VOL. LVIII
Copyright ~ 1979 by Academic Press, Inc.

All rights of reproduction in any form reserved.
ISBN 0-12-181958-2
BASIC METHODS
[1]
capabilities. In contrast, most routine biochemical procedures, such as

e n z y m e assays, can be performed with relatively little cellular material
and proportionately less space and equipment. Therefore, the types of
specialized equipment that are necessary for particular programs can be
predetermined with some degree of accuracy.
Although some of the facilities needed for an adequate cell culture
laboratory can be used for nothing else, much of the required facilities and
equipment need not be dedicated exclusively to this purpose. By sharing
equipment through collaborative efforts, the initial investment necessary
to begin a tissue culture laboratory can be reduced. Central services and
already available resources such as sterility testing, glassware washing,
and animal handling areas should be utilized w h e n e v e r possible. Also, the
shared use of ancillary equipment, e.g., microscopes, pH meters, centrifuges, etc., within the same laboratory or with other laboratories should
be considered. H o w e v e r , certain precautions, to be detailed later, must be
taken when the sharing of facilities is contemplated.
It is hoped that this article will c o v e r most of the main requirements
for setting up and running a functional cell culture laboratory. Additional
information concerning not only laboratory set-up but also detailed cell
culture techniques can be found in several well-written books, a-6 These
texts should be referred to before introducing cell and tissue culture
technology into any laboratory.
II. Cleaning and Sterilization Facilities
Although this area of a cell culture laboratory remains critically important, some o f the impact of poor laboratory practices has been lost in
recent years due to the ready availability of sterile, disposable labware
and commercially prepared media and reagents. With the exception of
very small operations in which it is financially feasible to purchase all
materials in a prepackaged, disposable form, at least some cleaning and
sterilization of glassware is necessary in almost e v e r y laboratory. Because of this and because cells in culture can be nutritionally fastidious,
investigators should be aware of the care that must be taken in the proper
handling of glassware which not only comes into direct contact with the
z p. F. Kruse, Jr. and M. K. Patterson, Jr., eds., ' 'Tissue Culture: Methods and Applications."
Academic Press, New York, 1973.
4 j. Paul, "Cell and Tissue Culture," 5th ed. Churchill-Livingstone,Edinburgh and London,
1975.
5 R. C. Parker, "Methods of Tissue Culture," 3rd ed. Harper (Hoeber), New York, 1961.
G. Penso and D. Balducci, "Tissue Cultures in BiologicalResearch." Elsevier, Amsterdam,
1963.
[1]
ASPECTS OF A TISSUE C U L T U R E LABORATORY
cells but also with such things as pipettes which are used to transfer
culture media. This is of particular importance in connection with toxic
substances introduced into glassware by normal processing. While
everyone is aware of the problems associated with microbial contamination, little thought may be given to toxic organic products introduced
during the manufacture of certain items, inorganic residues from detergent washes, or contamination by metal ions sloughed from pipes. Therefore, proper procedures for cleaning and sterilization of glassware should
be carefully followed; several such procedures are available in the literature. 3''~-7 Also it is recommended that glassware used in cell culture procedures be employed exclusively for this purpose and not mixed with
glassware used for other purposes. This includes not only the culture
vessels themselves but flasks, pipettes, and other miscellaneous items.
This precaution insures that all material used in culturing techniques has
been subjected to the same vigorous cleaning and that diffficult-to-remove
reagents do not contaminate the culture systems.
A complete separation of the cleaning facilities from the preparation
and the aseptic areas is the ideal situation; however, because of space
limitations, the preparation area can be combined with the cleaning area if
the glassware is not routinely contaminated with viruses or bacteria. If at
all possible, the aseptic areas should be maintained in a location isolated
from the cleaning area. The general size of the cleanup area is largely
dependent upon the quantity of material to be handled, but a laboratory of
100 to 150 ft2 will accommodate the maximum amount of equipment that
would be needed. In laboratories where acid cleaning is to be employed,5.7
a fume hood with sufficient ventilation and safety features should be incorporated into the overall design.
In general the layout of the laboratory will be determined by the location of the sink. The sources of hot and cold tap water will dictate the
placement of the washing equipment, i.e., decontamination and soaking
buckets, water purification system, and pipette washer. If the volume of
glassware is sufficiently large, a built-in glass washer would be advantageous. Several commercially available models are acceptable; however, an
adequate supply of purified water is necessary for the final rinses. Water
of suitable purity for these final rinses can be obtained by a single glass
distillation, demineralization, or reverse osmosis. The choice of purification method depends on such factors as the condition of the untreated
source water, the quantity of water needed, and the amount of space
available for the necessary equipment. It may be practical to employ a
single water purification system for all laboratory needs if the system is
r F. M. Price and K. K. Sanford, Tissue Cult. Assoc. Man. 2, 379 (1976).
BASIC METHODS
[1]
capable of producing ultrapure, reagent-grade water. Such systems are

discussed in greater detail in Section III which deals with the media preparation area of the laboratory.
After final rinsing, glassware is ready for drying and prepared for
sterilization. Large, bulky items may be drained and dried at room temperature on a drying rack. Most glassware is dried at elevated temperatures in a drying oven. It is also convenient to use a specially designed
dryer for pipettes since large numbers of pipettes are frequently used in
cell culture procedures. While the drying ovens and the sterilizing equipment may be located in a separate room or area, we have found it more
convenient for the drying apparatus to be located close to the washing
area. After drying, all glass vessels should be covered with paper or
aluminum foil and stored in a covered area to prevent dust accumulation.
Pipettes should be plugged with cotton and immediately stored in drawers. For larger operations an apparatus for automatically plugging pipettes
is available. Adequate bench and storage space needs to be allotted for
the handling of glassware; a 10- to 15-foot bench area with overhead and
under-counter storage is sufficient for most medium-size laboratories.
Since glassware is seldom being washed and prepared for sterilization at
the same time, this bench area can be used for both functions.
Sterilization of glassware and other materials can be accomplished by
either dry heat with a sterilizing oven or with moist heat by autoclaving.
Both pieces of equipment are necessary, and their size and subsequent
location in the laboratory design depends upon the projected amount of
use. Because of obvious problems with heat and ventilation, it would be
better to locate this equipment in a separate room which is readily accessible to the cleanup area. If smaller units are suitable for the intended
traffic, they can be placed in the same room as the washing facilities. All
sterilizing units should be equipped with temperature-recording charts to
maintain a complete record of sterilizing time and temperature. Because
of loading and/or air circulation problems within any unit, certain articles
in any one load may not reach the desired temperature. It is, therefore, a
good practice to label each article with a heat-sensitive indicator tape
which changes when the proper temperature has been reached and maintained for the proper time.
Almost all glassware, except that containing rubber tubing connections, may be sterilized by dry heat. The method also is used for material
such as silicone grease which cannot be effectively sterilized by moist
heat. However, dry heat sterilization is time-consuming and more difficult
to control even when a forced-air circulation system maintains uniform
conditions within the oven. Because of this, most sterilization procedures
are carded out with moist heat by autoclaving.
[1]
Since autoclaving is the most commonly used method, a word of caution is necessary about the quality of steam used to supply the autoclave.
When the steam is heavily contaminated with impurities, they settle on
the surfaces during autoclaving and the advantages of careful washing and
rinsing procedures are lost. This may be a major problem where larger,
shared facilities are supplied by house steam. Whatever the situation,
however, it is recommended that any autoclave using house steam be
equipped with a filtering device to remove contaminating material. A
better solution to the problem is to use an autoclave that has provisions
for its own steam generation. The water for such a unit can be obtained
from a purified source thereby eliminating the contaminants at the source.
In addition to the major items of equipment mentioned in this section,
several minor ones have proven useful and should be considered when
outfitting the cleaning and sterilizing facilities. These include carts to
facilitate the transfer of articles between the different areas of the laboratory. These are almost a necessity when the different functional units are
very widely separated. Also, provisions should be made for disposal containers in the cleanup area. Ideally, these should be closed containers
which would serve as receptacles for used disposable labware, wrappings
from sterilized items, and the like. Other items are pipette jars for soaking
pipettes before washing, a liquid detergent dispenser, and an ultrasonic
cleaning bath for hard-to-clean glassware.
III. Media Preparation and Storage Facilities

As with the other functional units described in this article, it would be
ideal if a separate area were set aside exclusively for media and reagent
preparation. If laboratory space is available, a 100 to 150 ft2 room would
be adequate to handle the equipment and to provide the bench space for
the necessary operations. However, as noted in the previous section, this
area can be conveniently combined with that designated for cleanup and
sterilization. Although media and other reagents can be purchased as
sterile, ready-to-use material, most investigators formulate at least part of
what they use. The operations involved in preparing any reagent for use in
cell culture are extremely critical, and several things, including a suitable
water source, high-quality chemicals, good filtration equipment, and
proper storage facilities, are essential for the successful maintenance of
cell populations in vitro.
The major component of media and other reagents for the propagation
of cells in culture is water. Although completely chemically defined media
are not yet possible, it is necessary to know within reasonable limits what
BASIC METHODS
[1]
has been added to media. It is also necessary to eliminate toxic elements

of both an organic and an inorganic nature. For these reasons, only water
of the highest possible purity should be used to formulate cell culture
reagents. 8 The selection of the proper system for generating ultrapure
water is largely based on the quality of water coming into laboratory and
the quantity of water needed. In many cases it is feasible to employ a
relatively high output unit so that the pure product can be used for all
laboratory purposes. Even when the water entering the laboratory has
been treated by a central purification system, it is advisable to purify it
again before use in the reagents.
Treatment of tap water with a mixed-bed ion exchanger followed by
glass distillation or two to three successive glass distillations will supply
enough ultrapure water for reagent purposes for an average cell culture
operation. A second method utilizes water previously treated by a central
facility and is capable of producing enough reagent-grade water for all
laboratory purposes (final rinsing in glassware washers, autoclave uses,
and media preparation). The water is first filtered to remove particulate
material, and then it is passed over activated charcoal to absorb organic
contaminents, subjected to a mixed-bed ion-exchange resin, and, finally,
passed through a submicron filter. A third method can use tap water as its
primary source and provides high-purity water for all laboratory uses.
This system pre-filters the water and then, in successive steps, subjects it
to reverse osmosis, mixed-bed deionization, and submicron filtration. Regardless of the method chosen to produce ultrapure water, provisions
should be made for storage of limited quantities of the product water as
well as for periodic testing of the water. It is not advisable to store
reagent-grade water for long periods of time. However, when it is necessary to store water in a rapid turnover situation, tightly closed, borosilicate glass carboys are recommended. With respect to testing, conductivity is the easiest method for measuring product water, but it only indicates
the amount of dissolved ions with no indication of particulate or organic
contamination. Since it is usually impractical to test for the other contaminants, and since conductivity at least gives a guide to the efficiency of ion
exchange, it should be used routinely. Usually conductivity meters are
incorporated into most purification systems; if not, relatively inexpensive
units can be purchased for testing purposes.
With few exceptions, essentially all liquid reagents for cell culture are
sterilized by filtration. Several types of bacteriological filtering systems
s R. W. Pumper,in "TissueCulture: Methodsand Applications"(P. F. Kruse,Jr. and M. K.
Patterson, Jr., eds.), p. 674. AcademicPress, New York, 1973.
[l]
have been employed for cell culture purposes including porcelain, asbestos, sintered glass, and membrane filters. Because of their uniform
characteristics, ready availability, and superior filtering qualities, membrane filters with a 0.22 /zm pore size are most widely used for filter
sterilization in cell culture laboratories. These filters can be obtained in
sizes ranging from 13 to almost 300 mm in diameter with the subsequent
capability of sterilizing from a few milliliters to several liters of material.
The filter holders are constructed of noncorrosive material and, like the
filters, come in various sizes and configurations, some of which are prepackaged in a sterile disposable form. It is necessary to have a variety of
sizes on hand. The most useful are syringe adapters which take the 13 and
25 mm diameter filters for small volumes; the pressure filter holder for the
47 mm filter which is used for volumes up to 1 liter; and at least one of the
larger sizes for sterilizing liter quantities of media. All membrane filtration
should be accomplished under positive pressure, thereby reducing foaming, by use of either 95% air-5% CO2 or nitrogen. Larger quantities of
media may be placed in specially designed stainless-steel pressure tanks
and passed through the filter into a receiving vessel fitted with a filling bell
for dispensing into bottles for storage. The entire filtering assembly and
receiving vessel can be sterilized as a unit by autoclaving.
Both before and after cell culture media and related reagents are formulated and sterilized, specific types of storage facilities are required. At
room temperature, closed, dust-free space is necessary for the storage of
stable chemicals, pipettes and other glassware, and filtering apparatus.
Again, the size of this space depends upon the size of the operation and
the amount of formulation that is anticipated. In addition, a refrigerator,
- 2 0 freezer, and - 7 0 ultrafreezer space are required. The refrigerator is
used to store a medium once it has been sterilized. Filtered media are
usually dispensed into 100 or 500 ml bottles and are stable at refrigerator
temperatures for several weeks. The - 2 0 freezer is needed for the shortterm storage of labile biochemicals, stock solutions of amino acids and
vitamins, and serum. For longer term storage of these materials, a - 7 0
ultrafreezer is necessary.
In addition to the equipment that has been mentioned as being required
for a media preparation and storage area, several more items are recommended to complete the facility. These include an analytical balance, a
pH meter, an osmometer, a hot plate, a magnetic stirrer, and a full range
of glassware, including a good supply of bottles for media storage. It is
possible to use much of this equipment in other laboratory areas; but
certain instruments, e.g., the osmometer, should be maintained in the
media preparation area.
10
BASIC METHODS
[l]
IV. W o r k A r e a for Aseptic Manipulation of Cell Cultures
Many cell culture manipulations may be routinely conducted on a

clean bench in an area free from drafts and removed from heavy traffic
areas. For more rigorous work or when hazardous agents are used, many
different shields and hoods are available which provide the several degrees of protection required by the guidelines of the National Cancer
Institute (see this volume [4]). These protective devices provide a sheltered working area and should be equipped with a germicidal lamp and be
accessible to gas, vacuum, and electrical outlets. The work surface in a
hood must be constructed of stainless steel or Formica. These surfaces
will not readily retain dust, and they can be repeatedly cleaned with
disinfectants. A minimum of 9 ft 2 of work surface should be available in
the hood.
The use of filtered laminar air ventilation was introduced to help prevent airborn contaminants in work with cell cultures. Units are available
in a variety of configurations which range from whole-room units to models which can be conveniently placed on laboratory bench tops. The working area within a laminar air-flow hood is vented by a continuous displacement of air that has passed through a high-efficiency particle (HEPA)
filter. HEPA filters remove particulates (diameter greater than 0.3 /~m)
from the air and thus provide a clean environment for manipulation of cell
cultures. Filtered air enters the hood from the top and exits at the bottom.
Thus aerosols generated during handling of the cells are rapidly removed
from the area as they ride the airstream to the exit. Horizontal flow hoods
that dir6ct the air toward the operator should not be used for cell culture
work. Bunsen burners or hot plates must be avoided in laminar flow hoods
since the generation of thermal and convection currents will destroy the
laminar flow patterns and reduce the effectiveness of the unit. The performance of a laminar flow unit (HEPA filters, air balance, and air flow)
should be checked at least three times a year; a detailed description of
these procedures has recently been published. 9
In order to ensure a clean work surface within the hood, it is important
to establish and maintain daily cleaning procedures. The walls and work
surface of the hood should be washed with disinfectant (Chlorofcn, Zephiran chloride) at the beginning and end of each work day. Additionally, it is
important to clean the work surface with disinfectant after completing
work with each cell strain. If this practice is rigidly adhered to, the likelihood of cross-contamination of cell strains is greatly reduced.
Sterile materials and other items routinely used in the aseptic work
areas should be stored in close proximity to the tissue culture hood in
G. J. McGarrity, Tissue Cult. Assoc. 1, 181 (1975).
[l]
11
closed, dust-free cabinets. Sterile, reusable glassware should be replenished frequently and a schedule of resterilization of unused material
should be maintained. Those items that are to be used in a specific cell
culture procedure should be placed in the hood prior to work. This step is
particularly important when working in a laminar flow system. If the
materials are placed in the unit 10 to 15 min before work commences then
the laminar flow system has ample time to remove particulates from the
work area.
V. Equipment for Routine Cell Maintenance
Basic tissue culture operations require an incubator, microscope, cell
repository materials, and culture vessels in which to grow the cells.
A. Incubators
Incubators may range from temperature-regulated boxes to elaborate
units which control temperature, humidity, and carbon dioxide levels in
the atmosphere. The units should be maintained on emergency power
circuits in case of an electrical power failure and should be placed close to
the aseptic work area. If the buffering system in the tissue culture medium
requires equilibration with CO2, e.g., Earle's salts, one will want to use a
CO2 incubator. When selecting a CO2 incubator for purchase, consideration should be given to the type of air flow patterns that exist in the unit.
An even temperature distribution within the incubator is essential, and
experience indicates that horizontal air flow units provide a more uniform
distribution of both temperature and CO2. Some incubators are waterjacketed but this is not a necessity. When tissue culture medium containing Hanks' salt is used or when closed culture systems are employed, a
CO2 incubator is not necessary and one may simply use a unit that
provides a constant and even temperature distribution throughout the
chamber.
B. Microscopes
Microscopes are essential since it is important to examine cultures
daily to observe their morphology and rate of growth. Many investigators
grow cells, but do not take the time to become familiar with the
morphological features of their cultures. In the last few years many laboratories have encountered problems of contamination of their cell cultures
with other cell types (see this volume [2]). Frequently, cultures have
become contaminated with HeLa cells; these cells replicate rapidly and
12
BASIC METHODS
[1]
b e c o m e the dominant cell type in the culture. If an investigator is familiar

with the morphological appearance of the cells in use, it is easier to detect
the introduction o f a second cell type into the culture system. An inverted microscope equipped with phase-contrast optics is essential for
morphological analysis o f monolayer cultures and is ideally placed close
to the incubator facility. In addition, it is advantageous to have a dissecting microscope and a conventional, c o m p o u n d microscope. One may also
want to consider utilization of ultraviolet optics for fluorescence.
C. Cell Repository Materials
Techniques for long-term preservation o f cultured cells by storage in
liquid nitrogen are now available, and it is essential for cell culture work to
have access to a cryogenic storage facility. A cell repository permits one
to bank cultured cells at low population-doubling levels and to store these
cells in essentially the same state for a number o f years. The availability of
a cell repository also eliminates the need to constantly subculture cells in
order to maintain cell strains in the laboratory. (See this volume [3])
Monodisperse cell suspensions in tissue culture medium and a cryoprotective agent (glycerol or dimethylsulfoxide) may be frozen at a controlled rate. s For most mammalian cells a freezing rate of l-3/min is
used until a temperature o f - 7 0 is reached. The cells are then transferred
to liquid nitrogen storage at - 196, or they may be stored in liquid nitrogen
vapor phase at - 156. The choice of the type of controlled-rate freezer
that would be most practical for individual needs can be made from several commercially available models. Prior to controlled-rate freezing,
cells are placed in glass ampules or plastic screw-cap vials. When glass
ampules are used, the neck of the container is sealed by using an ampule
sealer and oxygen torch. Plastic vials are sealed by a screw-cap lid. However, this does not provide a tight seal, and the containers must not be
immersed in liquid nitrogen. Plastic vials are routinely stored in liquid
nitrogen vapor phase.
D. Culture Vessels
Many types of glass and plastic cell culture vessels are commercially
available for use in the cell culture laboratory. These generally are either
borosilicate, soda glass, or polystyrene plastic vessels. The polystyrene
plastic and soda glass containers are inexpensive and can be routinely
discarded after one use. The borosilicate vessels are more expensive, but
lo j. E. Shannonand M. L. Macy,in "Tissue Culture: Methods and Applications" (P. F. Kruse,
Jr. and M. K. Patterson, Jr., eds.), p. 712. Academic Press, New York, 1973.
[I]
13
they are resistant to heat and can be washed and sterilized repeatedly
without adverse effect on utility. The selection of vessel is usually based
partly on personal preference and on the availability of a properly
equipped and functioning glassware washing facility.
When borosilicate glassware is reused, care must be taken during the
washing procedure to remove all serum and cellular residues from the
glassware. In addition, washed glassware must be adequately rinsed to
insure complete removal of all traces of solvents and detergents. Because
of the problems encountered in washing and rinsing tissue culture
glassware, many laboratories routinely use commercially available,
sterile, disposable tissue culture plasticware for their monolayer culture
work. Sterile, polystyrene petri dishes are available in a variety of sizes
(35, 60, 100, and 150 mm diameter). It is important to order tissueculture-grade plastic petri dishes because cultured cells generally do not
attach to bacteriological-grade petri dishes. Various sizes of plastic tissue
culture flasks are also available (T-25, T-75, T-150 flasks). These vessels
provide 25, 75, or 150 cm 2 of surface area, respectively, for growing cells.
In addition, plastic tubes or multiwell plates can also be used as substrates
for growing monolayer cultures.
For many biochemical analyses a large number of cells are required.
The method selected for growing mass cultures is clearly related to the
particular cell strain or cell line to be used. If diploid, anchoragedependent cells are employed, then monolayer cell culture techniques are
required. In contrast, if heteroploid or other nonanchorage-dependent cell
types are selected then suspension culture techniques most probably can
be utilized. Culture systems that permit growth of large numbers of cells
with either monolayer or suspension culture techniques are commercially
available.
For large-scale growth of anchorage-dependent cells in monolayer culture, many laboratories use roller bottles. These culture vessels are available in a variety of sizes in both borosilicate glass or polystyrene plastic,
and they permit harvests of up to 2 107 viable cells from one 730-cm z
vessel. In 1970 Kruse 11 introduced a roller bottle perfusion apparatus that
permits harvest of 1.9 108 normal diploid cells or 2 109 hetcroploid
cells from a single 730 cm 2 borosilicate glass roller bottle. This high cell
density is achieved by the perfusion of fresh medium into the culture
vessels at programmed intervals. The system permits control of pH and
influent nutrient levels in a more precise manner than is possible in
monolayer cultures fed at 2 to 3 day intervals.
Polystyrene roller bottles containing a spiral of Melinex are also commercially available. The Melinex polyester sheet provides 8000 cm 2 of
11 p. F. Kruse, Jr., L. N. Keen, and W. L. Whittle, In Vitro 6, 75 (1970).
14
BASIC METHODS
[1]
culture surface, and from a single roller bottle 2 10a heteroploid cells
can be routinely harvested. 12
Suspension culture techniques are available for the large-scale production of nonanchorage-dependent cells. Cells can be maintained in suspension by use of a rotary shaker or by stirring. The shaker technique was
introduced by Earle et al. 13 and is particularly suited to studies requiring
replicate cultures. The cells are suspended in medium containing methylcellulose (15 cps) in Erlenmeyer or round-bottom flasks and placed in a
rotary shaker. Currently the most popular method for suspension cultures
uses a magnetic stirrer to keep the cells constantly agitated in tissue
culture medium supplemented with methylcellulose.14 A wide range of
sizes of spinner culture vessels are available. The stirring rate varies
depending on the cell type used, but it must be slow enough to avoid
foaming.
A recent modification of the spinner system is the spin filter culture
device. 15 This unit was developed to grow cells to high population densities and to provide for removal of drugs from the tissue culture medium at
accelerated rates. Medium is removed through a rotating filter, the design
of which prevents the accumulation of cells on or within the filter. The
volume of medium used in this unit can vary from 200--1000 ml, and cells
routinely can be grown to densities of 10s cells/ml.
Two other culture systems have recently been developed for specific
purposes, but they warrant mention because of their applicability to a
variety of research systems. The dual-rotary circumfusion system
developed by Rose permits maintenance of cells for periods of several
months in a differentiated, functional state in vitro.16 Such cells can be
readily examined throughout their in vitro culture period by highresolution microscopy, and replicate cultures are possible. The cells are
grown in Rose chambers under a cellophane membrane. In many instances the monolayer cultures are initiated from explants. The cells in
this culture system undergo only limited replication but remain in a viable
and differentiated state for several months in vitro. 18
The artificial capillary system introduced by Knazek and Gullino lr
permits cells to grow to tissue-like densities that often approximate the in
lz W. House, in "Tissue Culture: Methods and Applications" (P. F. Kruse, Jr. and M. K.
Patterson, Jr., eds.), p. 338. Academic Press, New York, 1973.
~s W. R. Earle, E. L. Schilling, J. C. Bryant, and V. J. Evans, J. Natl. Cancer Inst. 14, 1159
(1954).
~ R. S. Kuchler and D. J. Merchant, Proc. Soc. Exp. Biol. Med. 92, 803 (1956).
~5 p. Hemmelfarb, P. S. Thayer, and H. E. Martin, Science 164, 555 (1969).
~8G. G. Rose, in "Tissue Culture: Methods and Applications" (P. F. Kruse, Jr. and M. K.
~r R. A. Knazek and P. M. Gullino, in "Tissue Culture: Methods and Applications" (P. F.
Kruse, Jr. and M. K. Patterson, Jr., eds.), p. 321. Academic Press, New York, 1973. See
this volume [14].
[1]
15
ASPECTS OF A TISSUE CULTURE LABORATORY
vivo state. The artificial capillaries are semipermeable cellulose acetate or
polycarbonate membranes through which tissue culture media is constantly perfused. Diffusable cell products can be retrieved from the perfusate, and in several instances hormone-producing cells synthesize and
secrete at much greater rates in the artificial capillary system than comparable numbers of cells growing in monolayer culture.
VI. Conclusion
The material in this article, including the partial list of commercial
sources provided in the table, outlines in general terms the equipment and
A PARTIAL LIST OF SUPPLIERS OF CELL CULTURE MATERIALS
Media & Biologicals

Associated Biomedical Systems
872 Main St.
Buffalo, NY 14203
HEM Research, Inc.

5451 Randolph Rd.
Rockville, MD 20852
ICN Pharmaceuticals, Inc.
Life Science Group

Biofluids, Inc.
1146 Taft St.
Rockville, MD 20852
Cappel Laboratories, Inc.

Downingtown, PA 19335
Colorado Serum Company
4950 York St.
Denver, CO 80216
Connaught Laboratories Ltd.
1755 Steeles Ave. W.
Willowdale, Ontario
CANADA M2N 5T8
Difco Laboratories, Inc.
P. O. Box 1058A
Detroit, MI 48201
Flow Laboratories, Inc.
1710 Chapman Ave.
RockviUe, MD 20852
GIBCO
3175 Staley Rd.
Grand Island, NY 14072
26201 Miles Rd.

Cleveland, OH 44128
International Biological
481 S. Stonestreet Ave.
Rockville, MD 20852
International Scientific Industries, Inc.
P. O. Box 9
Cary, IL 60013
Irvine Scientific Sales Co., Inc.
P. O. Box 4492
Irvine, CA 92664
K. C. Biologicals, Inc.
P. O. Box 5441
Lenexa, KS 66215
Microbiological Associates, Inc.
4733 Bethesda Ave.
Bethesda, MD 20014
Miles Laboratories
Elkhart, IN 46514
(contmued)
16
BASIC METHODS
A PARTIAL LIST OF SUPPLIERS OF CELL CULTURE MATERIALS (continued)
Media & Biologicals (continued)

Mogul Corp.
GIBCO Diagnostics
Laboratory Park
Chagrin Park, OH 44022
North American Biologicals, Inc.
15960 NW 15th Ave.
Miami, FL 33169
Pacific Biological Co.
2400 Wright Ave.
Richmond, CA 94804
Reheis Biochemical Dept.
Armour Pharmaceutical Co.
P. O. Box 511
Kankakee, IL 60901
Plasticware
Cooke Laboratory Products
900 Slaters Lane
Alexandria, VA 22314
Coming Glass Works
Science Products Div.
Coming, NY 14830
Costar
Div. Data Packaging Corp.
205 Broadway
Cambridge, MA 02139
Lux Scientific Corp.

1157 Tourmaline Cr.
Newbury Park, CA 91320
Vangard International
Box 3112
Red Bank, NJ 07701
York Scientific Ltd.
Ogdensburg, NY 13669
Incubators
Forma Scientific
P. O. Box 649
Marietta, OH 45750
Hotpack Corp.
5086A Cottman Ave.
Philadelphia, PA 19135
Lab-Line Instruments, Inc.
Bloomingdale Ave.
Melrose Park, IL 60160
National Appliance Co.
Heinicke Instruments Co.
P. O. Box 23008
Portland, OR 97223
Percival Refrigeration & MFG. Co. Inc.
P. O. Box 249
Boone IA 50036
Wedco, Inc.
P. O. Box 223
Silver Spring, MD 20907
Falcon
Div. Becton-Dickinson
P. O. Box 243
Cockeysville, MD 20130
Filtration Systems
Lab-Tek Products
Div. Miles Laboratories
30 W. 475 N. Aurora Rd.
Naperville, IL 60540
Bioquest
Div. of Becton-Dickinson & Co.
P. O. Box 243
Cockeysville, MD 21030
Linbro Scientific
681 Dixwell Ave.
New Haven, CT. 06511
Gelman Instrument Co.

15 Chestnut St.
Sussex, NJ 07461
[1]
[1]
ASPECTS OF A TISSUE CULTURE LABORATORY
Filtration Systems (continued)

Millipore Corporation
Ashby Road
Bedford, MA 07130
Nalge Co., Div. Sybron Corp
Nalge Labware Div.
75 Panorama Creek Dr.
Rochester, NY 14602
Nuclepore Corp.
7035 Commerce Circle
Pleasanton, CA 94566
Water Systems
Barnstead Sybron Corp.
225 Rivermoor St.
Boston, MA 02132
Bellco Glass Co.
340 Edwardo Rd.
Vineland, NJ 08360
Coming Glass Works
Science Products Div.
Coming, NY 14830
Culligan USA
1 Culligan Parkway
Northbrook, IL 60062
Hydro Service & Supplies, Inc.
P. O. Box 2891
Durham, NC 27705
Ultrasciences, Inc.
2504 Gross Point Rd.
Evanston, IL 60201
Bausch & Lomb, Inc.

Scientific Instrument Div.
Depot Rd., R.D. # 6
Auburn, NY 13021
E. Leitz
Link Drive
Rockleigh, NJ 07647
Nikon, Inc.
Instrument Group
EPOI, 623 Stewart Ave.
Garden City, NY 11530
Olympus Corp.
Precision Instrument Div.
2 Nevada Dr.
New Hyde Park, NY 11040
Swift Instruments, Inc.
1190 N. 4th St.
San Jose, CA 95112
Wild Heerbrugg Instruments, Inc.
465 Smith St.
Farmingdale, NY 11735
Carl Zeiss, Inc.
444 Fifth Ave.
New York, NY 10018
Cryogenic Supplies
Forma Scientific
Box 649-T
Marietta, OH 45750
Kelvinator Commercial Products, Inc.
621 Quay St.
Manitowoc, WI 54220
Microscopes
Revco, Inc.
Scientific and Industrial Div.
1188 Memorial Dr.
West Columbia, SC 29169
American Optical Corp.

Scientific Instrument Div.
Sugar & Eggert Rds.
Buffalo, NY 14215
Union Carbide
Linde Division
270 Park Avenue
New York, NY 10017
17
18
BASIC METHODS
[9.]
space required to initiate a cell culture program in a biochemistry laboratory. Few specific recommendations were given because the absolute
requirements for any individual program will depend upon the existing
facilities, the projected scope of the program, as well as the type of cells
to be cultured and their intended use. Although the specifics will change
according to the individual circumstances, the basic principles and problems considered here will remain unchanged regardless of the situation.
Whether coverslip cultures or mass cultures are employed, or whether the
entire operation is contained in one room or a suite of laboratories, cleanliness of glassware, purity of water and other reagents, and maintenance
of sterility will ultimately determine the reliability of any cell culture
system. The information presented here is a guide to what is necessary to
achieve these basic goals.
[2] D e t e c t i o n o f C o n t a m i n a t i o n
By GERARD J. MCGARRITY
The presence of adventitious agents in cell cultures is incompatible
with the concept of standardized, defined systems. Presence of extraneous agents may produce either gross turbidity and rapid destruction of the
culture, or no turbidity and little to moderate cytopathic effects on the
culture. The latter type can remain undetected for prolonged periods, and
have profound effects on the cell culture and experimental results.
Contamination may originate in the tissue specimen used to initiate the
cell culture, in media, especially in bovine serum, or in the general environment. The contamination may be bacterial, mycoplasmal, fungal, viral, or cellular. Detection methods described here will effectively monitor
microbiological contamination in cell cultures and media. These should be
viewed as part of an overall quality-control program. Methods to prevent
and control contamination are described elsewhere. 1-4
The limitations of detection methods must be appreciated. The major
limitations for detection of bacterial, mycoplasmal, and fungal organisms
are sample size, level of contamination, type of growth media utilized,
and presence of antibiotics in the sample. Contamination in bovine sera
i G. J. McGarrity and L. L. Coriell, In Vitro 6, 257 (1971).
2 G. J. McGarrity, Tissue Cult. Assoc. Manual 1, 167 (1975).
3 G. J. McGarrity, Tissue Cult. Assoc. Manual 1, 181 (1975).
4 G. J. McGarrity, V. Vanaman, and J. Sarama, in "Mycoplasma Infection of Cell Cultures" (G. J. McGarrity, D. G. Murphy, and W. W. Nichols, eds.), p. 213. Plenum, New
York, 1978.
Copyright 1979 by Academic Press, Inc.

ISBN 0-12-181958-2
[2]
DETECTION OF CONTAMINATION
19
and biologicals is generally more difficult to detect because of the low

frequency of contamination (1% or less of a lot) and the low concentration
of organisms in contaminated units, 1-10 organisms per milliliter or less.
Difficulties in detecting contaminants in cell cultures are frequently due to
improper selection of microbiological media and presence of antibiotics in
the sample. Antibiotics can mask contamination and yield false negatives.
The United States Pharmacopeia states that there is a possibility that low
levels of contaminants may remain undetected even in properly designed
sterility tests?
Detection of Bacterial and Fungal Contaminants
Three factors influence the results of sterility tests. First, no single
microbiological medium will detect all possible contaminants. A compromise is necessary for practical quality-control testing. Second, the
sterility of the unit is based on testing a small portion of the unit. Third,
presence of antibiotics can yield false negative results.
When contamination is introduced into antibiotic free cell cultures, the
organisms typically will grow rapidly and produce gross turbidity in 18-24
hr. Antibiotics in the medium will kill sensitive organisms and select
resistant ones. Resistant organisms may produce rapid turbidity, e.g.,
pseudomonads and molds, or they may be more fastidious and slower
growing. In the latter case, they can remain undetected even when subjected to routine testing. In this laboratory, whenever fastidious organisms have been isolated from cell cultures, they have been in cultures
containing antibiotics. L forms, mycobacteria, anaerobes, species of
Hemophilus, fastidious diphtheroids, and yeast have been isolated from
cell cultures containing antibiotics.
A. Bovine Serum
A large number of bacterial, fungal, mycoplasmal, and viral agents
have been isolated from bovine serum. 4 In this Institute, samples of each
lot of bovine serum are tested before purchase. After purchase, each
bottle is tested for bacteria and fungi before use. Testing before purchase
includes bacterial, fungal, mycoplasmal, and growth promotion. For tests
before purchase, the serum supplier should be requested to supply a
sample of the lot. If an entire lot is purchased, three bottles can be requested. If less than a lot is purchased, a 100-200 ml aliquot can be
5 "The United States Pharmacopeia," 19th ed. Mack Publ. Co., Easton, Pennsylvania,
1975.
20
BASIC METHODS
[2]
requested. Bacterial and growth promotion tests are described below.

Tests for mycoplasma and viruses are described elsewhere in this article.
1. Testing for Bacteria and Fungi before Purchase

Prepare duplicate racks containing one each of the following tubes of
culture media.
a. Fluid thioglycolate broth (Bioquest, Inc., Cockeysville, Maryland),
5 ml/tube
b. Trypticase soy broth (Bioquest, Inc.) 5 ml/tube
c. Sabouraud dextrose broth (Bioquest, Inc.), 5 ml/tube
d. Sterile short disposable tube
e. Two blood agar plates
Aseptically open serum bottle in a mass air flow room or laminar air flow
cabinet. Carefully flame the top of the bottle. Using a sterile pipette,
remove a 10-ml aliquot for testing. Add 3.0 ml of serum to each of the
short disposable tubes. Add 0.5 ml to each of the culture tubes of fluid
thioglycolate broth, trypticase soy broth, and Sabouraud dextrose broth.
Inoculate 0.1 ml onto each blood agar plate. Incubate one rack and one
blood agar plate at 37 and the other set at 30. Check the tubes and plates
daily for 2 weeks for evidence of contamination.
If present, contamination is generally evident within 3-4 days. Contamination in the tubes of broth must be distinguished from crystal formation and precipitation. This can be done by making a transfer to another
tube of broth and blood agar plate and by making a gram stain on the broth
or a centrifuged pellet of the broth. If possible, contaminants should be
identified to aid in tracing possible sources. To prevent false positives,
introduction of environmental contaminants during the testing must be
minimized. One episode of bacterial contamination of cell cultures has
been traced to environmental contamination that occurred during sterility
tests. 4
2. Testing for Growth Promotion of Cell Cultures

Select two standard cell cultures of the type for which the serum is to
be used as a growth supplement.
Prepare culture medium as it will be used for these cell cultures and
add 20% of the test bovine serum. As a control, use medium with 20%
bovine serum of proven growth promotion ability.
Seed a T25 plastic flask with 500,000 cells of each cell culture and add
5 ml of culture medium.
Incubate at 37 for 1 week or until culture is confluent. Harvest the
cells by the usual technique and pass to three T25 flasks.
[2]
21
Repeat this procedure once a week for four passages and observe for
growth, morphology, full sheeting of cells after 1 week incubation, and for
evidence of toxicity, granularity, and vacuolization as compared to
growth of the same cells in a standard nontoxic serum.
By the end of 1 month data are available on sterility tests and growth
promotion and on whether the serum lot can be accepted or rejected.
Note: The above time and cell concentration references may have to be
modified for different cell cultures.
3. Testing Individual Bottles of Serum

Contaminated bottles of bovine serum have occasionally been detected in a lot which passed all sterility tests before purchase. For this
reason we require that users pretest each bottle of serum before it is added
to cell cultures media. This test is carried out in the same way as listed
under paragraph 1 above, and observed for 2 weeks before use.
B. Cell Culture Media

All purchased single-strength media, as well as media and solutions
formulated in-house from basic ingredients or powdered base, are aseptically opened and tested as in paragraph 1 above except that the broths are
incubated only at 37 and observed for 10 days.
C. Cell Cultures
Cell cultures free of antibiotics can be tested for bacterial and fungal
contamination by procedures described in paragraph 1 above except that
broths are incubated only at 37 and observed for 10 days.
Mycoplasmal Testing
Mycoplasmas are particularly well suited as infectants of cell cultures.
They grow to high concentrations in cell cultures, 106 to I08 colony fprming units (CFU) per milliliterof supernatant medium being representative;
they are resistant to many antibiotics; they usually produce no gross
turbidity in the cultures they infect; they are relatively difficultto detect;
and they have profound effects on the cultures they infect. Reviews of the
effects mycoplasmas have on cell cultures and other effects are
available.6.7
E. Stanbridge, Bacteriol. Rev. 35, 206 (1971).
r G. J. McGarrity, D. G. Murphy, and W. W. Nichols, eds., "Mycoplasma Infection of Cell
Cultures." Plenum, New York, 1978.
22
BASIC METHODS
[2]
M a n y techniques are available for detection of m y c o p l a s m a s . T h e s e

are b a s e d on microbiological culture, s'a fluorescence methods, l,H electron m i c r o s c o p y , TM and biochemical methods. 13-17 E a c h a p p r o a c h has certain limitations; the relative sensitivity of e a c h is unknown. Most o f the
data available on detection techniques have b e e n b a s e d on fibroblast cell
cultures. S o m e methods and criteria m a y h a v e to be modified when working with epithelial cells, l y m p h o c y t e s , endothelial cultures, and other cells
with specialized functions. Positive and negative controls are essential for
each test method. This m a y necessitate the use of an indicator cell culture
for some detection methods.
A. B o v i n e Serum
M y c o p l a s m a l testing is p e r f o r m e d on bovine serum before purchase.

I f m y c o p l a s m a s are present, they are likely to be in low concentrations.
Therefore, test samples are p e r f o r m e d with large volumes, following the
p r o c e d u r e of Barile and K e r n . TM
1. M e d i a
Base m e d i u m is p r e p a r e d in bulk and stored in 75-ml amounts until

use. Base m e d i u m consists of m y c o p l a s m a broth b a s e (Bioquest, Inc.)
s u p p l e m e n t e d with 1 g dextrose, 1 g L-arginine m o n o h y d r o c h l o r i d e (Eastman K o d a k Co., R o c h e s t e r , N e w York), and 1 ml of 0.2% phenol red
in 1000 ml distilled water. After autoclaving and cooling, the m e d i u m is
dispensed in 75-ml aliquots. Add 5 ml y e a s t extract (Flow L a b o r a t o r i e s ,
Rockville, Maryland), 20 ml horse s e r u m (Flow Laboratories), and 1 ml of
a 0.2% D N A stock solution (Thymic D N A , Miles L a b s , K a n k a k e e , Illinois). Broth tubes are dispensed in 5-ml aliquots in s c r e w - c a p p e d tubes
and refrigerated until used. I f m y c o p l a s m a testing will be done on a lima G. J. McGarrity, Tissue Cult. Assoc. Man. 1, 133 (1975).
9 M. F. Barile, H. E. Hopps,M. W. Grabowski,D. B. Riggs, and R. A. DelGiudice, Ann.
N . E Acad. Sci. 225, 251 (1973).
10T. R. Chen, Exp. Cell Res. 104, 255 (197")).
11 R. A. DelGiudice and H. E. Hopps, in "Mycoplasmal Infection of Cell Cultures" (G. J.
McGarrity, D. G. Murphy, and W. W. Nichols, eds.), p. 57. Plenum, New York, 1978.
12D. M. Phillips, in "Mycoplasmal Infection of Cell Cultures" (G. J. McGarrity, D. G.
Murphy, and W. W. Nichols, eds.), p. 105. Plenum, New York, 1978.
is E. M. Levine, Methods Cell Biol. 8, 229 (1974).
14 E. L. Schneider, E. J. Stanbridge, and C. J. Epstein, Exp. Cell Res. 84, 311 (1974).
15 A. G. Perez, J. H. Kim, A. S. Gelbard and B. Djordjevic, Exp. CellRes. 70, 301 (1972).
16 E. M. Levine, L. Thomas,D. McGregor,L. Hayflick, and H. Eagle, Proc. Natl. Acad.
Sci. U.S.A. 60, 583 (1968).
lr G. J. Todaro, S. A. Aaronson, and E. Rands, Exp. Cell Res. 65, 256 (1971).
la M. F. Barile and J. Kern, Proc. Soc. Exp. Biol. Med. 1311,432 (1971).
[2]
23
ited basis, the volume of medium prepared at one time can be reduced
accordingly.
For agar medium, 9 g of Noble agar (Difco Labs, Detroit, Michigan)
are added per liter of base medium, autoclaved, cooled, and dispensed in
75-ml aliquots in containers that will conveniently hold 100 ml. These
bottles are stored in the refrigerator until use. For preparation of agar
plates, heat the base medium to 96 to dissolve the agar. Place the bottles
of base medium, horse serum, yeast extract, and DNA into a water bath
at 50. Allow all components to equilibrate at 50. Add 5 ml yeast extract,
20 ml horse serum, and 1 ml of a 0.2% DNA solution to 75-ml aliquots.
Dispense into 60 x 15 mm Petri dishes (Falcon Plastics, Oxnard, California), approximately 10 ml per plate. Final mixing of media components
above 50 can result in cloudy plates that are more difficult to read. Mixing
and dispensing should be done quickly since agar will solidify at 45. Agar
plates are refrigerated and used within 1-2 weeks. Wrapping stacks of
plates in aluminum foil or plastic will reduce dehydration. The gel strength
of different lots of Noble agar may vary. The lowest concentration that
yields a gel should be used. All media should be autoclaved as soon after
mixing as pos'sible. Final pH should be 7.2 ___ 0.2.
2. Testing Serum
Testing serum for mycoplasma is performed by inoculation of 25 ml of
tl~e serum under test into each of two bottles containing 100 ml of mycoplasma broth and incubation at 37 under aerobic and anaerobic conditions.
Anareobic atmospheres can be generated easily in a Gas Pak system
(Bioquest, Inc.) Aliquots of 0.1 ml are subcultured from the broths to agar
plates after 4 and 7 days incubation. Agar plates are examined microscopically (x 100) for colonies for at least 3 weeks. Aerobic plates should be
taped to reduce drying.
3. Testing Cell Cultures
a. Monolayer Cultures. Monolayer cultures should be at least 60%
confluent for testing and be in antibiotic-free medium. Since proteolytic
enzymes may have an adverse effect on mycoplasma, cells should be
scraped from the container surface.
Remove and discard all but 3 ml of the antibiotic-free medium. Scrape
some cells from the monolayer surface with a rubber policeman or glass
rod. Place the suspension of cells and spent medium in a Wasserman or
tube of similar size. Cap. To test the suspension cultures, gently shake
flask and remove 3 ml for assay.
24
BASIC METHODS
[2]
Inoculate a mycoplasma broth tube with 0.5 ml and mycoplasma agar

plate with 0.1 ml of the test specimen. Incubate the plate and broth
anaerobically in a Gas Pak or equivalent system. The cap on the broth
tube should be loose and the plate unsealed. Monitor the anaerobic state
with a methylene blue oxidation-reduction indicator (Bioquest, Inc.) or
equivalent. After 7 days of incubation inoculate an agar plate with 0.1 ml
of the inoculated broth.
Microscopically examine plates at least weekly for presence of typical
mycoplasma colonies using 100 magnification. Colonial morphology
may vary among different species, especially on primary isolation. However, the "fried egg" appearance will predominate.
Keep plates 3 weeks before recording negative results. Mycoplasma
colonies must be distinguished from crystal formation or "pseudocolonies." Mycoplasmal isolates can be identified by the growth inhibition
test 19 or by epi-fluorescence. 2
Microbiological culturing may be difficult to set up unless a microbiology lab or microbiologist is available. Quality control of various components should be instituted. It should be recognized that microbiological
culture will not detect all strains of Mycoplasma hyorhinis.
b. Limitations of Microbiological Culture for Mycoplasmas. Hopps et
al. Zl have shown that certain strains ofM. hyorhinis, a frequent cell culture
isolate, will not propagate on cell-free media. DelGiudice and Hopps H
have shown that 62% of M. hyorhinis cell culture isolates did not propagate on agar. To detect these strains and to obtain a rapid test result, we
inoculate 0.1 ml of the test specimen into a plastic Leighton tube containing a plastic coverslip (Costar Industries, Cambridge, Massachusetts) that
has been inoculated 1-2 days previously with 20,000 3T-6 mouse embryo
fibroblast cells as an indicator culture. This system is incubated for 4
days. At the end of that time, the plastic coverslip is removed, cut with
sterile scissors, and two fluorescent stains are performed as enumerated in
sections c and d, below. These are an immunofluorescent sera specific for
M. hyorhinis and Hoechst fluorochrome compound 33258. The Hoechst
stain binds to DNA, staining both cell nuclei and, if present, mycoplasmal
DNA.
c. Immunofluorescent Staining for M. hyorhinis. The procedure is essentially that reported by DelGiudice and Hopps. H Coverslips are rinsed
twice in phosphate-buffered saline (PBS), pH 7.4, fixed in acetone for 2
min, rinsed in PBS, and stained with immuno-fluorescent antisera specific
for M. hyorhinis, strain BTS-7. Stock antisera and antigens for preparing
19 W. A. Clyde, Jr., J. lmmunol. 92, 958 (1964).
R. A. DelGiudice, N. Robillard, and T. R. Carski, J. Bacteriol. 93, 1205 (1966).
21 H. E. Hopps, B. C. Meyer, M. F. Barile, and R. A. DelGiudice, Ann. N.Y. Acad. Sci.
225, 265 (1973).
20
[2]
25
working solutions are available from Research Resources Branch, National Institute of Allergy and Infectious Diseases, NIH, Bethesda, Maryland 20014. The stain is allowed to incubate for 30 min at room temperature and is rinsed with PBS. Buffered glycerol (Bioquest, Inc.) is used as
a mounting medium.
d. DNA Stain. This procedure has been reported by Chen. TM We have
incorporated the modifications of DelGiudice and Hopps into our procedures.ll
Prepare stock solution of stain to a concentration of 50 ~g/ml in distilled water. This is stored at 4 in light-protected bottles. Thimersol
(1:10,000) or similar disinfectant is added to retard microbial growth.
Working solutions are prepared fresh for each application by dilution in
potassium phospate-citric acid, pH 5.5, to make a final stain concentration of 0.05 ug/ml.
Keep the coverslip submerged in 1 ml of medium and add 1 ml of
fixative, acetic acid methanol (1:3). Aspirate after 2 min. Air dry and stain
for 10 min; wash twice with distilled water, and mount in buffered glycerol
(Bioquest, Inc.).
e. Microscopy. Preparations can be viewed with either transmitted or
incident illumination. Incident illumination has been used in this laboratory (Leitz Orthoplan plus Ploem Illumination). In immunofluorescent
preparations, individual M. hyorhinis organisms will fluoresce. This is
particulary noticeable on plasma membranes of the cultured cells. With
the DNA stain, appearance of mycoplasma and other prokaryotic organisms can be clearly observed as extranuclear fluorescent particles.
Controlled studies have indicated that the DNA stain is effective in
detecting mycoplasmas11 although standardization, proper controls, and
confirmation testing by other techniques are necessary.
f. Biochemical Tests. A variety of biochemical tests are available.
These are based on properties of mycoplasmas not shared by mammalian
cells in culture. Depending on the capabilities and facilities, these procedures might be more suitable than the techniques described above. The
limitations of each test method must be acknowledged. Appropriate controls must be included in each test.
'g. Uridine Phosphorylase Assay. This procedure was developed by
Levine and is based on the presence in all mycoplasma13 tested of uridine
phosphorylase activity.22 Although mammalian tissues in vivo have this
activity, cells in culture do not. Activity is detected by paper chromatography.
22 E. M. Levine and B. G. Becker, in "Mycoplasma Infection of Cell Cultures" (G. J.
McGarrity, D. G. Murphy, and W. W. Nichols, eds.), p. 87. Plenum, New York, 1978.
26
BASIC METHODS
[2]
A total of 2 x 107 cultured cells are used and are lysed with 3 ml of
phosphate--Triton buffer, za Levine states that as few as 3 x 106 cells in 0.5
ml buffer can be used. [14C]uridine (40 /zCi//xmol) is added as the nucleoside substrate, and the mixture is incubated at 37 for 180 min. After
incubation, uridine is separated from uracil at 37 by paper chromatography. The solvent system consists of 70 ml boric acid (4%, ca. 0.65 M),
1 ml concentrated ammonium hydroxide, and 430 ml n-butanol. Some
phase separation occurs at room temperature but is cleared by warming to
37 and shaking.
Samples are obtained at 30 and 180 min. Uridine and uracil spots are
located on the developed chromatogram with a short-wave UV light. The
radioactivity in these spots is determined. The conversion rate at 180 min
in uninfected cells is generally lower than 10%; mycoplasma-infected cells
have much higher rates, typically 20-90%.
h. Uridine/Uracil Uptake. This procedure was developed by Schneider
and Stanbridge, 14,24 and based on the marked alterations in the incorporations of free bases and nucleosides into the nucleic acid of
mycoplasma-infected cells. Mammalian cells incorporate relatively large
amounts of uridine and only negligible amounts of uracil. Mycoplasma
can incorporate uracil. Mycoplasmal uridine phosphorylase can convert
uridine to uracil, preventing uptake of uridine by cultured cells. The net
effect is that the ratio of uptake of uridine to uptake of uracil is
significantly reduced in mycoplasma-infected cells.
Cells to be tested are inoculated into each of six plastic 25 cm 2 flasks,
2.5 x 105 cells per flask. After 24 hr incubation at 37, three flasks are
inoculated with [3H]uridine, 5 /zCi/ml; three flasks are inoculated with
[3H]uracil, 5 /zCi/ml. After 18-24 hr additional incubation at 37, the
labeled nucleic acids are extracted and counted. Cells are removed from
the flask by trypsinization and pelleted at 200 g for 10 min in a refrigerated
centrifuge. The pellet is washed twice with 1 ml of phosphate-buffered
saline (PBS, pH 7.4) free of calcium and magnesium. Centrifuge at 200 g
for 10 min. Decant and suspend the pellet in 1 ml 5% trichloracetic acid
(TCA). Precipitate the nucleic acids by placing the tube in ice for 30 min.
Centrifuge at 900 g for 15 rain. Decant supernatant liquid and add 1 ml of
23 The phosphate-Triton buffer is made from three stock solutions. A 0.5 M solution of
Na~HPO4 is prepared and adjusted to pH 8.1 with 1 N HC1 (approximately 2.9 ml/100 ml).
Triton X-100 (Rohm and Haas, Inc., Philadelphia, Pennsylvania) is prepared as a 10% (v/v)
solution in water. Nonradioactive uracil is prepared in a 0.01 M water solution for use as a
tracer in the chromatography system. All solutions are stored at 4; Na2HPO4 must be
warmed before use. The final buffer is prepared as follows: 10 ml Na2NPO4, 10 ml uracil, 5
ml Triton X-100, 75 ml water. The final buffer is stored at 4.
24 E. L. Schneider and E. J. Stanbridge, Methods Cell Biol. 10, 277 (1975).
[2]
27
perchloric acid (PCA). Incubate in 90 water bath for 15 min to hydrolyze

the nucleic acids. Remove and retain the supernatant fluid. Add 0.1 ml of
each of the PCA supernatants to separate vials containing Bray scintillation fluid or equivalent. Count in a liquid scintillation counter.
Ratio of uptake of uridine to uracil is determined by dividing the mean
of uridine incorporation by the mean of uracil incorporation. According to
Stanbridge and Schneider, ratios of 1000 for cell lines and 400 for human
skin fibroblasts are considered negative; values below 100 have always
been found to indicate mycoplasma. 14"z4An alternate method is available
using double labeling. 25
i. Other Biochemical Procedures. An adenosine phosphorylase assay
was developed similar to uridine phosphorylase by Hatanaka et al. 26
This is similar to the uridine phosphorylase described above, although
mycoplasma strains have been encountered that have lacked the
enzyme.
Todaro et al. developed a rapid detection method using density gradient separation of tritiated nucleic acids. 17 Cell cultures are incubated for
18-20 hr with either tritiated uridine or uracil, supernatant fluids are precipitated with ammonium sulfate, layered on a 15-60% linear sucrose
density gradient, and centrifuged at 40,000 rpm in an SW-41 rotor for 90
min. The gradient is collected dropwise, precipitated in trichloracetic
acid, and counted in a scintillation counter. Mycoplasmal infected cultures yield a clearly defined peak at 1.22-1.24 gm/cmZ; clean cultures lack
this peak. This technique has not been extensively used with different cell
cultures and mycoplasma species.
Mycoplasma ribosomal RNA can be distinguished from mammalian
ribosomal RNA by electrophoretic mobilities) 6 Polyacrylamide gel electrophoresis has been used to detect mycoplasma 16 S and 23 S RNA in
infected cultures. Levine has noted that some mycoplasmal strains will
not incorporate uridine tracers, leading to false negatives.
Autoradiography has been used to detect mycoplasma infection) 5 The
difference between infected and noninfected cultures can be striking. Uninfected cultures will incorporate tritiated thymidine only over the nuclear
areas. Infected cultures, however, will incorporate thymidine (cleaved to
thymine by mycoplasma phosphorylase) around the plasma membrane of
cells, giving the appearance of cytoplasmic grains. It has been reported
that certain strains of mycoplasma cannot incorporate precursors and
may result in false negativesY 2
25 E. J. Stanbridge and E. L. Schneider, Tissue Cult. Assoc. Manual 2, 371 (1976).
2~ M. Hatanaka, R. A. DelGiudice, and C. Long, Proc. Natl. Acad. Sci. U.S.A. 72, 1401
(1975).
28
BASIC METHODS
[2]
B. Electron Microscopy
Since the concentration of mycoplasmas in infected cultures is usually
on the order of 106 to 108 per milliliter,electron microscopy can be used as
a method of detection.12 True infection must be distinguished from artifacts that can render diagnosis difficult.Use of an indicator cell culture
known to be free of mycoplasma can minimize this difficulty.
Although transmission electron microscopy (TEM) is generally more
available, sample preparation is more prolonged. In the examination of
sections, only one plane through a cell is examined. Therefore, multiple
sections are required. Because of the geometry of sectioning and the level
of mycoplasma adsorbed to cells, false negatives are possible.
Scanning electron microscopy (SEM) is more suitable for routine
screening. Van Diggelen et al. 27 and Phillips 12 have used this to detect
mycoplasma infection. Cells are grown on glass coverslips for 3-4 days
and fixed in 2.5% glutaraldehyde buffered at pH 7.4 with 0.2 M Collidine.
Cells are dehydrated in ethanol series to absolute alcohol and transferred
to acetone. Coverslips are critical point dried with liquid CO2 in a Sorvall
Critical Point Drying System, coated with gold using an Edwards 306
coater, and viewed in a scanning electron microscope.
Cultures examined by SEM are usually so heavily infected that interpretation is easy. Mycoplasma species vary in their ability to cytadsorb
onto cultured cells. In cultures infected with M. hyorhinis, most mycoplasma organisms are cytadsorbed onto cells. Little cytadsorption occurs
in A-9 mouse cells infected with Acholeplasma laidlawii, with most of the
mycoplasmas attached to the coverslip.
SEM has worked well in examination of fibroblast cultures. Some
difficulty has been observed in interpretation of lymphocyte cultures
grown in suspension (D. M. Phillips and G. J. McGarrity, unpublished
results). As with other test systems, use of an indicator cell culture may
alleviate some problems in interpretation.
Virus Testing
Bovine serum has been shown to contain several bovine viruses .28.29 It
is not always feasible for individual laboratories to screen for them unless
specialized facilities are available. The viruses most commonly isolated
from bovine sera are bovine virus, diarrhea virus, infectious bovine
rhinotracheitis virus, parainfluenze virus, and bovine herpes virus.
2T O. P. Van Diggelen, S. Shin, and D. M. Phillips, Exp. Cell Res. 106, 191 (1977).
2s C. W. Molander, A. Paley,C. W. Boone, A. J. Kniazeff, and D. T. Imagawa, In Vitro 4,
148 (1968).
29 C. W. Molander, A. J. Kniazeff,C. W. Boone, A. Paley, and D. T. Imagawa, In Vitro 7,
168 (1972).
[3]
PRESERVATION, STORAGE, AND SHIPMENT
29
Many suppliers of serum test their product for presence of bovine

viruses. Requests can be made for specific test methods and results. Procedures are available for concentration of pooled sera, 29 and these preparations can be viewed under electron microscopy and/or inoculated in
embryonic bovine tracheal cells grown in virus-free serum. These cultures can be compared to uninoculated controls for cytopathic effect and
can be viewed by electron microscopy. Alternately, preparations can be
histologically stained and, if desired, stained with appropriate fluorescent
antibodies.
Comments
Each laboratory must decide which quality-control procedure is appropriate and feasible for its own use. However, a minimum of testing of
serum for sterility and growth promotion and assay of serum and cell
cultures for mycoplasma are essential. Culturing for bacteria and fungi is
relatively easy. Mycoplasma assays are somewhat difficult to perform,
and most laboratories use two methods. Appropriate controls are essential in each test. Testing for viruses is more complex and, to the best of
our knowledge, not a routine in most laboratories.
Special note should be taken of bovine serum as an experimental variable. In addition to the organisms isolated from bovine serum, considerable variation exists in concentrations of chemicals in serum. 3 In this
study, the concentration of free fatty acids was the only variable that
affected growth of human fibroblasts although other factors would be
expected to affect the outcome of biochemical studies.
In addition to the detection techniques described here, publications
are available listing other aspects of quality control. 1-4 Techniques are
available for monitoring contamination by HeLa and other cell types? 1
All of these measures are relatively easy and highly cost-effective in the
establishment of axenic, standardized cell cultures and experimental and
diagnostic procedures.
.~0 C. W. Boone,N. Mantel,T. D. Caruso, E. Kazam, and R. E. Stevenson, In Vitro 7, 174
(1972).
3~ W. A. Nelson-Rees and R. R. Flandermeyer, Science 195, 1343 (1977).
[3]
Preservation, Storage, and Shipment

By LEWIS L. CORIELL
Mammalian cells serially cultivated in vitro are in an artificial environment which differs in many known and unknown respects from the normal
habitat in vivo. No culture medium can duplicate the in vivo environment,

ISBN 0-12-181958-2
30
BASIC METHODS
[a]
and we have learned by experience that cells taken from different organs
require different culture media for growth and/or preservation of specialized function. Some cell types inevitably die out in any given culture
medium, and the cells which persist and continue to proliferate also lose
or gain characteristics which were not apparent in vivo. Examples of
changes frequently observed include: microbial contamination, chromosome abnormalities, loss of antigens, loss of ability to proliferate, and
death of the cell culture. Many schemes have been employed with varied
success to avoid or delay these undesired changes. Use of primary cell
cultures and the use of organ cultures are examples which function well
for selected short-term studies. Incubation of cell cultures at reduced
temperature slows the metabolic rate and the frequency of refeeding and
subculture. Storage of cell cultures in the frozen state following the general procedures used commercially for storage of bovine semen 1 was
reported by Scherer and Hoogasian in 19542 and by Swim et al. in 1958.3
These techniques used glycerol as an adjuvant with storage in Dry Ice at
- 7 0 . Viable cell cultures could be recovered for many months, but there
was gradual loss of viability when stored at - 7 0 . Storage of frozen cell
cultures in liquid nitrogen ( - 196) is the accepted procedure at this time,
and when properly carried out it seems to be compatible with prolonged
preservation of viability and other characteristics of mammalian cell
cultures.4-6
Without the presence of an adjuvant, freezing is lethal to most mammalian cells. Damage is caused by mechanical injury by ice crystals,
concentration of electrolytes, dehydration, pH changes, denaturation of
proteins, and other factors not well understood. These lethal effects are
minimized by: (1) adding an adjuvant such as glycerol or dimethylsulfoxide (DMSO) which lowers the freezing point; (2) a slow cooling rate
which permits water to move out of the cells before it freezes; (3) storage
at a temperature below -130 which retards the growth of ice crystals;
and (4) rapid warming at the time of recovery so that the frozen cell
culture passes rapidly through the temperature zone between - 5 0 - 0
where most cell damage is believed to occur.
Different cell strains may vary widely in their ability to withstand
physical and chemical insults during the freezing and thawing process.
C. Polge, A. U. Smith, and A. S. Parks, Nature 164, 666 (1949).

z W. F. Scherer and A. Hoogasian, Proc. Soc. Exp. Biol. Med. 87, 480 (1954).
3 H. E. Swim, R. F. Hall, and R. F. Parker, Cancer Res. 18, 711 (1958).
4 H. T. Meryman, Science 124, 515 (1956).
5 H. T. Meryman, Fed. Proc., Fed. Am. Soc. Exp. Biol. 22, 81 (1963).
e j. Nagington and R. I. Greaves, Nature (London) 194, 933 (1962).
[3]
31
Epithelial cells of h u m a n skin survive best in 20-30% glycerol adjuvant,

w h e r e a s fibroblasts in the same tissue survive best in 10% glycerolY We
h a v e frozen and r e c o v e r e d mosquito cell cultures with high viability in 1%
glycerol. 8 T h e optimal cooling rate for m o u s e e m b r y o cells is l/min, a for
m a r r o w s t e m cells 2-3/min, ' for y e a s t cells 10/min, ' ' and for h u m a n red
blood cells in Polyvinylpyrrolidone (PVP) the optimal freezing rate is 300/
min. '2 T h e s e e x t r e m e e x a m p l e s are mentioned here to caution the reader
that no single suspending m e d i u m and p r o c e d u r e will be ideal for processing
and cryogenic storage of all cell cultures. H o w e v e r , m o s t investigators have
had excellent success in preserving m a n y types of m a m m a l i a n cells with
5 - 1 0 % glycerol or dimethylsulfoxide as the adjuvant and a cooling rate of
1-1&/min. The p r o c e d u r e s r e c o m m e n d e d in this article are designed for
p r e s e r v a t i o n o f h u m a n fibroblast and l y m p h o c y t e cell cultures. T h e y are
equally satisfactory for m a n y other h u m a n and animal normal and malignant cell cultures, but if they p r o v e less than satisfactory for a given cell
culture, the reader should titrate each variable to arrive at the optimal
p r o c e d u r e for the cell u n d e r study.
Preservation
Nutrition
Cell cultures survive cryogenic storage best if they are frozen in a good
state of nutrition and maximal viability. This is achieved as follows:
Select cells in the logarithmic phase of growth and r e m o v e the culture
medium. Assess sterility by culture on a blood agar plate. Replace with
fresh m e d i u m and harvest 24 hr later for storage in liquid nitrogen. The
cell culture m e d i u m should be the one which best supports growth and
p r e s e r v a t i o n o f desired characteristics of the cell. Cell cultures grown in
synthetic m e d i u m without s e r u m have been successfully frozen and recovered. 'a'14 H o w e v e r , m o s t workers e m p l o y a culture m e d i u m for preservation containing 5 - 2 5 % fetal bovine s e r u m which provides unknown
7 B. Athreya, E. Grimes, H. Lehr, A. Greene, and L. Coriell, Cryobiology 5, 262 (1969).
8 L. Coriell, unpublished.
D. G. Wittinghorn, S. P. Liebo, and P. Mazur, Science 178, 411 (1972).
,0 E Mazur and J. J. Schmidt, Cryobiology 5, 1 (1966).
" P. Mazur and J. Schmidt, Cryobiology 5, 1 (1968).
,2 A. P. Rinfret, Fed. Proc., Fed. Am. Soc. Exp. Biol. 22, 94 (1963).
,3 V. Evans, H. Montes de Oca,J. C. Bryant,E. L. Schilling, and J. E. Shannon, J. Natl.
Cancer Inst. 29, 749 (1962).
,4 C. Waymouth and D. Varnum, Tissue Cult. Assoc. 2, 311 (1976).
32
BASIC METHODS
[3]
growth factors and acts as an adjuvant for protection against damage

during cryogenic storage.
Harvest
Suspension cell cultures are counted to determine the number of viable cells, centrifuged at 800-1000 g for 10 min, and suspended in inactivated fetal bovine serum (56 for 15 min) containing 5% dimethylsulfoxide. Dilute to 10 million cells/ml, test for sterility, dispense into
ampules, and process as described below.
For monolayer cultures each flask is inspected under the binocular
microscope before harvest. Any flasks containing cells of abnormal appearance are discarded. It is desirable to release the cells from the
monolayer surface with a minimum of chemical and physical trauma. Add
to each T75 flask, 5 ml of 0.02% ethylenediaminetetraacetic acid (EDTA)
and let stand for 8 min. Remove the liquid and discard it. Add 3 ml of
EDTA in dilute trypsin 15 and observe under the binocular miscroscope
until the cell sheet starts to lift from the surface around the edges. Tap the
flask gently to loosen all the cells. This process usually requires 1 to 7 min.
Stop the action of trypsin-EDTA by adding 3 ml of growth medium containing 20% fetal calf serum. Transfer these cells to a common pool in a
centrifuge bottle that is chilled in a cracked-ice bath. Mix thoroughly,
count in a haemocytometer, and dilute with growth medium to 5 105
viable cells per milliliter as determined by trypan blue exclusion. Culture
aliquots on a blood agar plate, trypticase soy broth, Sabouraud dextrose
broth, fluid'thioglycolate broth, mycoplasma agar and broth, and by one
of the indirect methods to detect Mycoplasma hyorhinis (see this volume
[2]). Sediment the cells in a refrigerated centrifuge at 800-1000 g, discard
the supernatant, and resuspend in complete culture medium with 20% fetal
calf serum plus 10% sterile glycerol. Dispense the chilled cell suspension
into 1.2 ml borosilicate glass ampules with an automatic syringe. The
ampules should be previously marked with printer's ink and sterilized in
the dry air oven to insure a label that will not come off when stored in
liquid nitrogen. The ampules must be closed in an oxygen flame with a
pulled seal. To detect leaks, ampules are placed on aluminum canes m and
tested by immersion in a cylinder of alcohol stained with methylene blue
and chilled to 4. If a pinhole leak exists the contents of an ampule will be
stained blue and should be discarded. Sealed ampules should be kept at 4
and frozen as soon as possible. Significant loss of viability of WI-38 cells
was observed when stored at 4 for more than 6 hr. t7
15 T. Puck, J. Exp. Med. 1114, 6t5 (1956).
m Available from McCracken & Sons, Inc., 636 N 13 Street, Philadelphia, Pennsylvania,
19123.
17 A. Greene, B. H. Athreya, H. B. Lehr, and L. L. Coriell, Cryobiology 6, 552 (1970).
[3]
33
Freezing
Place the sealed ampules on labeled canes and position in a controlled
cooling rate apparatus. Set the controls to achieve a cooling rate of 1-2
per minute. When the heat of fusion is released at about - 10 the cooling
rate should be increased briefly to exchange the heat of fusion and to
reestablish the cooling rate. When the temperature reaches -25 the cooling rate can be increased to 5-10 per minute. When the temperature of
the specimen reaches - 100 the ampules can be transferred quickly to a
liquid nitrogen refrigerator for storage in the vapor or in the liquid phase.
Success with cell cultures has been achieved by many different slow
freezing techniques. 1'17'18'19 When large numbers of ampules are frozen
conditions and cooling rate settings must be adjusted to provide the desired cooling rates. The optimal cooling rate and procedures should be
determined for each cell type. Mouse embryos, for example, survive
poorly if cooled faster than 1 per minute 9 whereas many fibroblast or
epithelial cell cultures tolerate a faster cooling rate TM. As a general guide to
the adequacy of the whole process, the trypan blue viability of the cell
culture after recovery from liquid nitrogen storage should not be more
than 5% below the viability determined before storage. If it consistently
exceeds 10% loss of viability every step in the process should be critically
evaluated.
Storage
Permanent storage should be in liquid nitrogen whether liquid or vapor
phase. This will ensure a temperature well below -150 and prevent ice
crystal growth and enzyme activity. A few investigators have described
changes in cells stored in liquid nitrogen, 2-22 but most investigators have
not been able to observe detectable loss or gain of properties when cells
are properly stored in it. Twenty-two cell cultures stored in liquid nitrogen
in our laboratory have been recovered periodically and observed for viability as determined by trypan blue exclusion, plating efficiency, and cell
yield in milk dilution bottles and roller tubes after 7 days of incubation.23
No significant change in these parameters has been observed after storage
in liquid nitrogen for up to 12 years. 23'24
Is V. Perry, C. Kraemer, and J. L. Martin, Tissue Cult. Assoc. Man. 1, 119 (1975).
19 R. Moklebust, N. Diez, and I. Goetz, Tissue Cult. Assoc. Man. 3, 671 (1977).
2o W. P. Peterson and C. Stulberg, Cryobiology 1, 80 (1965).
21 H. T. Meryman, "Cryobiology," p. 65. Academic Press, New York, 1966.
22 L. Berman, M. P. McLeod, and E. P. Powsner, Lab. Invest. 14, 231 (1965).
23 A. E. Greene, B. H. Athreya, H. B. Lehr, and L. L. Coriell, Proc. Soc. Exp. Biol. Med.
124, 1302 (1967).
24 A. E. Greene, M. Manduka, and L. L. Coriell, Cryobiology 12, 583 (1975).
34
BASIC METHODS
[3]
Shipment
In design of a standard procedure for shipment of cell cultures consideration must be given to (1) safe delivery of viable cells, (2) cost, and (3)
regulations imposed by government and carders.
Shipment of Frozen Ampules

Frozen ampules of cell cultures may be shipped in small liquid nitrogen thermos refrigerators, or they may be packed in Dry Ice. Such packages are heavy, costly to ship, and must be returned. Furthermore, they
require special handling, and if delayed in route there is danger of evaporation of the refrigerant. After safe arrival at the destination in a frozen
state, there are still hazards. After delivery the specimen must be placed
in fresh Dry Ice or liquid nitrogen until it is recovered. If stored in a
refrigerator (4) or at - 2 0 , or even at - 5 0 , cells will be damaged. When
recovered the ampule must be warmed in a water bath at 37 and immediately cultured in an ideal medium. Because of the above problems routine
shipment of frozen ampules is impractical, except for special situations or
when a courier can accompany a shipment.
Shipment of Monolayer Cell Cultures

It is recommended that frozen cell cultures be thawed, subcultured,
and shipped as an actively growing culture. The procedure is as follows:
After rapid thawing of the frozen ampule place the contents in a T25
plastic tissue culture flask containing 5 ml of culture medium with 20%
fetal bovine serum. Next day remove the medium and feed. In 4-5 days
there should be a confluent sheet of cells. Remove the medium, fill the
flask with fresh medium, tape the screw cap in place, package, and ship
by air mail special delivery. This procedure also gives the sending laboratory the opportunity to verify, examine, and/or test each culture before it
is shipped, and it has resulted in better than 95% success in delivery of
viable cell cultures in over 5000 shipments to domestic and foreign laboratories. Upon receipt at the destination the flask should be placed in an
incubator at 37 overnight to permit recovery from trauma, exposure, and
shaking which may have dislodged some cells during transit. Next day,
open the flask and remove and save the excess culture fluid for use when
the flask is subcultured.
Regulations
Most cell cultures are free of microbial contaminants and etiologic
agents and therefore can be shipped without meeting any special stan-
[3]
PRESERVATION,
STORAGE, AND SHIPMENT
35
BIOMEDICAL
MATERIAL
I
NOTICE TO CARRIER
This package coatalm LESS THAN 50 ml OF AN ETIOLOGIC AGENT, N.O.S., is
packaged and labeled in accordance with the U.S. Public Health Service Interstate
Quarantine Regulations (42 CFR, Section 72.25 (c) (1) and (4)). and MEETS ALL
REQUIREMENTSFOR SHIPMENT BY MAIL AND ON PASSENGERAIRCRAFT.
This shipment is EXEMPTED FROM ATA RESTRICTED ARTICLES TARIFF
6-D (see General Requirements 388 ( d ( l ) ) and from DOT HAZARDOUS MATERIALS REGULATIONS (see 49 CFR, Section I73.386(d)(3)). SHIPPER'S
CERTIFICATES, SHIPPING PAPERS, AND OTHER DOCUMENTATION OR
LABELING ARE NOT REQUIRED.
Date
Signature o f Shipper
Institute f o r Medical Research
Copewood Street
Address
Camden, New Jersey 08103
FIG. 1. T h e s e labels m a y be obtained from a n u m b e r of c o m p a n i e s including Marion

Manufacturing Co., Atlanta, Georgia, and L a b e l m a s t e r , 6001 N. Clark St., Chicago, Illinois.
dards as to volume, labels, and packaging except that they arrive safely
and in an undamaged condition. However, some cell cultures may be
contaminated and some do contain etiologic agents, e.g., lymphocyte cell
lines transformed with Epstein-Barr virus. If known etiologic agents are in
the cell culture a package and labeling procedure for shipment of cell
cultures that will meet all requirements for etiologic agents should be
followed. 25 The packaging requirements for etiologic agents are as
follows:
T h e y shall be placed in securely closed, watertight primary receptacles which shall be
enclosed in a durable, watertight s e c o n d a r y packaging. Several primary receptacles m a y be
enclosed in a single s e c o n d a r y packaging, providing that the total v o l u m e of all the primary
receptacles so enclosed does not e x c e e d 50 ml. The space at the top, b o t t o m and sides
b e t w e e n the primary receptacles and s e c o n d a r y packagings shall contain sufficient nonparticulate absorbent material s u c h as cotton wool, to absorb completely the contents of the
~5 U. S. Public Health Service, " I n t e r s t a t e Quarantine Regulations" (42 CFR, Sect. 72.25).
U S P H S , W a s h i n g t o n , D.C., 1972.
36
BASIC
METHODS
[4]
primary receptacle in case of breakage or leakage. Each set of primary receptacle and
secondary packaging shall then be enclosed in an outer packaging constructed of corrugated
fiberboard, wood, or other material of equivalent strength. The maximum amount of
etiologic agents that may be carried in any one package is 50 milliliters.
Packages containing an etiologic agent must resist breaking or leakage of
the contents including: (1) a water spray test for 30 min, (2) a free drop test
through a distance of 30 feet, and (3) a puncture test of a 3.2-cm steel
cylinder weighing 7 kg when dropped 1 m onto the exposed surface of the
package. The outside of the package should display the two labels shown
in Fig. 1 if it contains etiologic agents.
Packaging Recommendations
The T25 plastic flask is filled to the top with complete culture medium,
and the screw cap is tightened firmly and anchored in place with masking
tape. The flask is placed in a liquid-tight polyethylene envelope # 101. 26
The loose space around the flask is filled with absorbent cotton, and this
envelope is placed in a larger liquid-tight polyethylene envelope #302 and
sealed; this is then placed in a tongue and grooved Styrofoam utility
mailer #3731-3227 and sealed with Fiberglas tape. The outer c a n o n is of
heavy corrugated cardboard sealed with vinyl sealing tape. The completed package, ready for mailing, weighs approximately 5 ounces and has
sufficient insulation to withstand subzero temperatures without freezing
the contents.
26 Briton liquid tight bags distributed by Medical Associates Int., Inc., P.O. Box 123, Topeka, Kansas 66601.
27Utility Mailer #3731-32 obtained from Cole Plamer, 7425 Noah Park Ave., Chicago,
Illinois 60648.
[4] S a f e t y C o n s i d e r a t i o n s
in the Cell Culture Laboratory
By W. EMMETT BARKLEY
The basic tenet of safety in research that involves potentially hazardous organisms is strict adherence to good laboratory practice. This tenet
demands an awareness of the possible risks associated with the research
materials that are handled, knowledge of mechanisms by which exposure
may occur, use of procedures and techniques that reduce the potential for
exposure, and continuous vigilance to guard against compromise and error. The cell culture w o r k e r is not unfamiliar with these basic principles,
although the primary intent o f their application has been the protection of
METHODS IN E N Z Y M O ~ Y ,
VOL. LVIII

ISBN 0-12-181958-2
[4]
SAFETY CONSIDERATIONS
37
the cell culture rather than the cell culture investigator. Indeed, the experienced cell culture worker is aware that cell cultures are susceptible to
contamination; knows the potential sources of contamination and the
means by which contaminants can be introduced into cell cultures; is
proficient in the use of procedures and techniques for preventing cell
culture contamination; and well understands that a relaxation in good
laboratory practice often results in cell culture contamination. Fortunately, the methods of the successful investigator are also applicable to
protecting the individual from potential hazards that may be associated
with research involving cell cultures.
Biohazards and Cell Cultures
Cell cultures represent a potential biohazard because of their capacity
to infect the laboratory worker with unrecognized viruses. The
significance of this potential biohazard is emphasized by the few but serious cases of laboratory-acquired infections associated with the preparation and handling of primary monkey cell cultures. The most notable
examples involve infections with Herpesvirus simiae (B virus) and Marburg virus. 1
No laboratory-acquired infections have been associated with the use
of continuous cell culture lines that are assumed to be free of infectious
virus. These cell lines, however, should not be considered free from
hazard since they may harbor latent viruses. For example, transformed
cell lines may spontaneously produce viruses with oncogenic potential in
animals, and human lymphoid cell lines may harbour Epstein-Barr
virus.2,3 Lymphoid cell lines may also represent a unique potential hazard
if they are obtained from persons who plan to maintain them in culture. In
the event of accidental ingestion or injection, these cells may not be
rejected since they would possess common histocompatibility antigens. If
the cells were transformed in culture, they may have the capacity to
create transformed foci in the cell culture worker which may subsequently progress to true malignancy)
Cell cultures used in virological studies should be assessed to present
the same degree of hazard as the infectious virus under study. In such
studies, actual hazards can be assessed easily and appropriate safeguards
R. N. Hull, in "Biohazards in Biological Research" (A. Heilman, M. N. Oxman, and R.
Pollack, eds.), p. 3. Cold Spring Harbor Lab., Cold Spring Harbor, New York, 1973.
2 G. J. Todaro, in "Biohazards in Biological Research" (A. Hellman, M. N. Oxman, and R.
Pollack, eds.), p. 114. Cold Spring Harbor Lab., Cold Spring Harbor, New York, 1973.
3 j. A. Schneider, in "Biohazards in Biological Research" (A. Hellman, M. N. Oxman, and
R. Pollack, eds.), p. 191. Cold Spring Harbor Lab., Cold Spring Harbor, New York, 1973.
38
BASIC METHODS
[4]
established. W h e r e infectious viruses are not intentionally handled, an

assumption of hazard should be the basis for establishing safeguards.
Potential for Exposure
Analysis of c o m p r e h e n s i v e surveys of laboratory-acquired infections
reveal that recognized accidents are not the cause for m o s t infections
a m o n g l a b o r a t o r y workers 4-6 Indeed, f e w e r than 20% of d o c u m e n t e d
infections can be attributed to accidental contact, ingestion, or injection
with infectious materials. 6 T h e actual causes for m o s t laboratory-acquired
infections are not known. H o w e v e r , the knowledge that m o s t microbiological practices create aerosols suggests that inhalation of undetected aerosols m a y contribute significantly to occupational illness among
laboratory w o r k e r s who handle infectious materials. 7 T h e cell culture
w o r k e r should be a w a r e of the possibility o f inadvertent inhalation exposures during the conduct of cell culture work. This potential for e x p o s u r e
should be considered in the selection of cell culture procedures.
R e c o m m e n d e d Safety Practices
Cell culture workers are encouraged to b e c o m e familiar with the safe
practices used in infectious disease research. Excellent review articles on
good l a b o r a t o r y practices in infectious disease laboratories h a v e been
published, s-~2 This information will be o f great value to those planning to
conduct research with potentially hazardous viruses. H e r e , emphasis will
be placed on four topic areas o f good l a b o r a t o r y practice that pertain
specifically to the cell culture technique.
Good Laboratory Practice
P i p e t t i n g . G o o d pipetting practice requires the use o f pipetting aids to
p r e v e n t e x p o s u r e s by ingestion and careful technique to reduce the poten-
4 S. E. Sulkin and R. M. Pike, Am. J. Public Health 41,719 (1951).

5 S. E. Sulkin, Bacteriol. Rev. 25, 203 (1961).
6 R. M. Pike, Health Lab. Sci. 13, 105 (1976).
7 R. L. Dirnmick, W. Vogl, and M. A. Chatigny, in "Biohazards in Biological Research"
(A. Hellman, M. N. Oxman, and R. Pollack, eds.), p. 246. Cold Spring Harbor Lab., Cold
Spring Harbor, New York, 1973.
s A. G. Wedum, Public Health Rep. 79, 619 (1964).
a G. B. Phillips, J. Chem. Educ. 42, Part One, A43; ibid. Part Two, All7 (1965).
10 H. M. Darlow, Methods Microbiol." 1, 169 (1969).
11 M. A. Chatigny and D. I. Clinger, in "An Introduction to Experimental Aerobiology" (L.
Dimmick and A. B. kers, eds.), p. 194. Wiley, New York, 1969.
~2 M. A. Chatigny, Adv. Appl. Microbiol. 3, 131 (1961).
[4]
39
tial for creating aerosols. There are a variety of pipetting aids available
that range from simple bulb- and piston-actuated systems to more sophisticated devices which contain their own vacuum pumps. In selecting a
pipetting aid, consideration should be given to the kind of pipetting procedure that is to be performed, the ease of operation, and the accuracy of
delivery that is needed. It is likely that several pipetting aids will be
necessary; no universal aid is available that is appropriate for all pipetting
procedures.
Pipetting technique that reduces the potential for creating aerosols
should be developed. Pipettes should be plugged with cotton. Rapid mixing of liquids by alternate suction and expulsion through a pipette should
be avoided. No material should be forcibly expelled out of a pipette, and
air should not be bubbled through liquids with a pipette. Pipettes that do
not require expulsion of the last drop are preferable to other types. Care
should be taken to avoid accidentally dropping cultures from the pipette.
When pipetting infectious materials, place a disinfectant-soaked towel on
the working surface. Liquids from pipettes should be discharged as close
as possible to the fluid or agar level of the receiving vessel, or the contents
should be allowed to run down the wall of the receiving vessel. Dropping
of the contents from a height into vessels should be avoided. Contaminated pipettes should be carefully placed in a pan containing enough suitable disinfectant to allow complete immersion.
Hygienic Practice. The cell culture worker should avoid eating, drinking, and smoking in the laboratory. Hands should be washed after handling cell cultures and before leaving the cell culture laboratory. To avoid
contact contamination, the investigator should develop the habit of keeping hands away from mouth, nose, eyes, and face. Laboratory gowns or
other suitable clothing should be worn when handling cell cultures; the
same clothing should not be worn to eating areas or outside of the
laboratory,
Decontamination. Work surfaces should be disinfected with a suitable
chemical disinfectant following the completion of procedures involving
cell cultures. Cell culture wastes which may be contaminated with virus
should be disinfected with a suitable chemical disinfectant or sterilized by
autoclaving before disposal. Contaminated items such as glassware and
laboratory equipment should be disinfected or autoclaved before washing,
reuse, or disposal. Contaminated materials that are to be autoclaved or
incinerated at a site away from the laboratory should be placed in a
durable, leak-proof container that is closed before removal from the
laboratory.
Control of Aerosols. Most techniques used in a cell culture laboratory
have the capacity to create aerosols. Therefore careful attention should be
40
BASIC METHODS
[4]
given to measures that can reduce the potential for inhalation exposures
to aerosols that are produced by laboratory operations. Although the
extent of aerosol production can often be controlled by good technique,
efforts to eliminate exposure require the use of containment equipment.
Such special containment equipment for use in preventing the creation
of aerosols is available. For example, capped safety centrifuge cups and
sealed centrifuge heads can prevent the production of aerosols while centrifuging.
The most important measure for controlling aerosols is the use of
ventilated safety cabinets for containment of operations that produce considerable aerosols. The cell culture worker is encouraged to use this containment equipment for operations that involve grinding, blending,
homogenizing, and resuspending tissues or cell cultures. This should be
standard practice if the cell culture materials are suspected of being contaminated with pathogenic microorganisms or viruses. Specific recommendations for the effective use of this equipment are presented in the
following section.
Ventilated Safety Cabinets

The most versatile ventilated safety cabinet available to the cell culture worker is the Class II biological safety cabinet. This cabinet is commonly known as a laminar flow safety cabinet. Its versatility is derived
from the protection afforded to both the cell culture and the investigator.
Cell cultur6s within this cabinet are protected from airborne contamination by an airflow of uniform velocity that has been filtered by highefficiency particulate air (HEPA) filters before reaching the work space.
The HEPA filters used are virtually impenetrable by microorganisms and
viruses. The cabinet provides protection to the cell culture worker by
preventing, under normal operating conditions, airborne cells or viruses
released during experimental activity from escaping the cabinet and entering the laboratory. This protection is achieved by an air barrier at the front
opening of the cabinet and the filtration capability of the exhaust air
HEPA filters. It must be emphasized that the protection features of the
Class II cabinet are only applicable to airborne contamination; protection
against exposures by ingestion, injection, and direct contact must be
achieved by standard good laboratory practice.
The effectiveness of laminar flow safety cabinets in providing protection
for the cell culture and those working with them, will depend on the
performance capability of the cabinet, the location of the cabinet within
the laboratory, and certain operational practices. The National Sanitation
Foundation (NSF), Ann Arbor, Michigan, has developed a national stan-
[4]
41
dard for Class II cabinets. 1~ The selection of a Class II cabinet that has
been certified by NSF provides an assurance of performance capability.
The performance capability, however, should be validated by measuring
airflow and filter efficiency once the cabinet has been installed in the
laboratory.
Cabinets should be located in an area of the laboratory away from
doorways, supply air diffusers, and spaces of high activity. This is done to
reduce the adverse effects of room air currents that could compromise the
operation of the cabinet. Generally, the best location for the cabinet is
along a side wall at a position farthest from the laboratory door.
The operational practices listed below are recommended when using
laminar flow safety cabinets. Their use will provide excellent protection
for cell cultures against contamination. These practices are most important, however, when the cell cultures are experimentally infected or contaminated with viruses.
1. Cabinet users should be encouraged to wear long-sleeved gowns
with knit cuffs and gloves. This reduces the shedding of skin flora into the
work area of the cabinet and protects the hands and arms from contact
contamination.
2. The work surface of the cabinet should be decontaminated with a
chemical disinfectant before placing equipment and material into the
cabinet.
3. The cabinet should not be overloaded. Ideally, everything needed
for the complete cell culture procedure should be placed in the cabinet
before starting work so that nothing passes in or out through the air
barrier until the procedure is completed.
4. Nothing should be placed over the exhaust grills located at the front
and rear of the work surface. This practice will help maintain uniform
airflow within the cabinet.
5. The arrangement of equipment and material on the work surface is
an important consideration. Contaminated items should be segregated
from clean ones and should be located so that they never have to be
passed over clean items. Discard trays should be located to the rear of the
cabinet.
6. After all equipment and material has been placed in the cabinet, the
cabinet user should wait a few minutes before beginning work. This time
period allows the cabinet air to remove any contaminants that may have
been introduced into the cabinet during the loading process.
7. Always use aseptic techniques when performing cell culture work
within the cabinet. This is important in preventing contact contamination.
13 "National Sanitation Foundation Standard No. 49 for Class II Biohazard Cabinetry."
Natl. Sanit. Found., Ann Arbor, Michigan, 1976.
42
BASIC METHODS
[4]
8. When working with pathogenic viruses, all contaminated items

should be contained or disinfected before they are removed from the
cabinet. Trays of discarded pipettes and glassware should be covered
before they are removed from the cabinet.
9. After all items have been removed from the cabinet, the cabinet
should be allowed to operate for a few minutes to assure the removal of
airborne contaminants. The work surface should then be decontaminated
with a chemical disinfectant.
Special Considerations and Recommendations
Lymphoid and Virus-Transformed Human Cell Lines

The Human Genetic Mt]tant Cell Repository operated by the Institute
for Medical Research, Camden, New Jersey, under sponsorship of the
National Institute of General Medical Sciences has recommended
minimum safety guidelines for working with lymphoid and virustransformed human cell lines.14 In addition to emphasizing the importance
of good laboratory practice, the guidelines encourage the creation of
biosafety committees to oversee work involving human cell lines, assign
responsibilities to the laboratory supervisor for preparing safety protocols, recommend certain medical surveillance and screening procedures, and limit access to the cell culture laboratory to authorized
individuals. The concern addressed by these guidelines is the potential
hazard, that may be associated with the release of Epstein-Barr virus or a
virus whose genome may have been integrated in the DNA of the
transformed cell.
Cell culture workers who handle long-term lymphoid lines or their
derivatives are advised to have serum samples collected and stored annually. Anti-EB virus titers should be obtained on these samples. Serum
samples should also be obtained and evaluated following accidental exposure to these cell cultures.
The guidelines endorse the use of ventilated safety cabinets in all cell
culture work. Individuals who obtain long-term lymphoid lines from the
repository are required to use ventilated safety cabinets as well as to
adhere to the principles of the recommended guidelines.
Recommendations of the Medical Research Council, Canada, for Special

Facility Safeguards in Cell Culture Work
The Medical Research Council, Canada, published in February 1977,
"Guidelines for the Handling of Recombinant DNA Molecules and Ani~4 "The Human Genetic Mutant Cell Repository." Inst. Med. Res., Camden, New Jersey,
1977.
[4]
43
mal Viruses and Cells. ''15 These guidelines recommend that good laboratory practice be complemented by special facility safeguards for certain
cell culture work. The guidelines require that cell culture work involving
(1) the co-cultivation or fusion of human cells with cells of nonhuman
mammals and (2) short-term nonhuman primate cells be confined to segregated rooms or cubicles that have exhaust air ventilation. The exhaust air
from these laboratories is to be discharged outdoors. However, recirculation of air is allowed providing that the air is treated by filtration with
HEPA filters. These requirements have been established to prevent the
dissemination of cells and viruses, which the exhaust air may contain, to
uncontrolled areas of. the laboratory facility in the event that they are
accidentally released within the segregated laboratory area.
The facility precautions are based on a possibility of hazard that is
considered to be greater than that associated with most cell culture work.
For co-cultivation and fusion studies with human cell lines, the principal
concern is the possibility of unmasking a virus with oncogenic potential
for man. The hazard concern for short-term nonhuman primate cells is the
evidence that these cells may contain Herpesvirus simiae (B virus).
Cell Cultures and Infectious Viruses

When cell cultures are used with, or are known to contain, an infectious virus, the safety precautions selected for the cell culture work
should be based on recommended safety guidelines for the infectious
virus. Safety guidelines have been developed for viruses pathogenic to
man and for viruses known to cause cancer in animals. The Center for
Disease Control has classified pathogenic viruses on the basis of hazard
and has recommended minimal safety conditions for their control. ~6 The
National Cancer Institute has published safety standards for research
involving oncogenic viruses, lr The investigator will find these documents
useful in establishing specific safety protocols for work with cell cultures
and infectious viruses.
~ "Guidelines for the Handling of Recombinant DNA Molecules and Animal Viruses and
Cells." Med. Res. Counc., Ottawa, Canada, 1977.
~n "Classification of Etiological Agents." DHEW, PHS, CDC, Office of Biosafety, Washington, D.C., 1974.
~7 "National Cancer Institute Safety Standards for Research Involving Oncogenic Viruses."
DHEW Publ. No. (NIH) 75-900. DHEW, PHS, NIH, Washington, D.C., 1974.
44
BASIC
[5]
Media
By RICHARD G.
METHODS
and Growth
[5]
Requirements
I-IAM and WALLACE L. MCKEEHAN
Introduction
This brief review o f culture media and their applications is addressed
primarily to the use of cell culture as an experimental tool in fields of
r e s e a r c h other than analysis of cellular growth requirements. Our emphasis is on the t y p e s of m e d i a available, and on principles involved in selection of the m e d i u m that is best suited for the e x p e r i m e n t that is to be done.
Environmental requirements for cellular multiplication and theoretical aspects of m e d i u m d e v e l o p m e n t are reviewed in greater detail elsewhere. 1-4
T y p e s of Cultured Cells and Culture Systems
The choice of which culture m e d i u m to use and what supplement (if
any) to add to it is strongly influenced by the type of cell that is to be
grown and by the w a y that cell is to be grown. Thus, it will be helpful to
begin with a brief outline of the types of cultured cells and culture s y s t e m s
that can be e m p l o y e d in various kinds of research.
Classification of Cells
The following criteria are often e m p l o y e d to describe cells:
Karyotype: Are the c h r o m o s o m e s fully normal for the species both in

n u m b e r and m o r p h o l o g y ? I f so, the cell is considered diploid (or in special
c i r c u m s t a n c e s haploid or tetraploid). I f not, it is considered to be aneuploid or heteroploid.
Multiplication potential: Does the cell cease to multiply after a limited
n u m b e r of divisions in culture (finite life span), or is it a p e r m a n e n t line
capable of continuous long-term multiplication?
Anchorage dependence: Does the cell require attachment to a substrate
for multiplication, or will it also multiply in suspension culture? This is
1 R. G. Ham and W. L. McKeehan, In Vitro 14, 11 (1978).
2 R. G. Ham and W. L. McKeehan, in "Nutritional Requirements of Cultured Cells" (H.
Katsuta, ed.). Jpn. Sci. Soc. Press, Tokyo (in press).
3 R. G. Ham, C. Waymouth, and P. J. Chapple, eds., "The Growth Requirements of
Vertebrate Cells in Vitro.,' Cambridge Univ. Press, London and New York (in press).
4 G. H. Rothblat and V. J. Cristofalo, eds., "Growth, Nutrition and Metabolism of Cells in
Culture," 3 vols. Academic Press, New York, 1972-1977.

All fights of reproduction in any form reserved.
ISBN 0-12-181958-2
[5]
MEDIA AND GROWTH REQUIREMENTS
45
often tested by suspending individual cells in soft agar and determining

whether they will form colonies.
Density-dependent inhibition of multiplication: Do the cells arrest their
cell cycles in G~ (or Go) when the culture becomes crowded, or will they
continue to multiply and pile up on one another?
Malignancy: Will the cells form an invasive malignant tumor when
injected into appropriate test animals (e.g., " n u d e " mice; this volume
[311)?
Differentiation: Does the cell exhibit tissue-specific differentiated properties in culture?
Exact classification of cultured cells is difficult because many of them
tend to blend into a continous spectrum, rather than fitting into sharply
defined classes. At the extremes of the spectrum, the fully normal cell is
usually thought of as being diploid, finite in life span, anchorage dependent, density inhibited, and not malignant, while the classical permanent
cell line is aneuploid, capable of indefinite multiplication, not anchorage
dependent, not inhibited by density, and malignant. Unfortunately, these
differences do not always occur as a set. There are apparently "normal"
cells that are not anchorage dependent, e.g., chondrocytes and various
blood-related cells, and others that do not exhibit density-dependent inhibition, e.g., keratinocytes. There are also permanent lines, such as 3T3,
that are aneuploid and capable of indefinite multiplication, but are
nevertheless anchorage dependent, density inhibited, and not malignant
by most tests. Such cells can be "transformed" by oncogenic viruses to
yield malignant lines that are no longer anchorage dependent or density
inhibited (this volume [30]). Because of this, they are sometimes referred
to as "normal," but it is better, in view of their aneuploidy and absence of
a finite life span in culture, to refer to them as "nontransformed" permanent lines. Ideally, the term "normal" should only be used to describe
cells that do not differ in any major way from the cells that occur in intact
animals. Note that the term "nontransformed" is used to refer to normal
cells with finite life spans as well as to the permanent lines that lack
transformed properties.
Relationship between Cell Type and Growth Requirements

Different types of cells have different growth requirements. Although
fully normal cells and nontransformed lines are beginning to receive more
attention, fully transformed permanent lines have dominated previous
studies of cellular growth requirements and, at present, are the only types
that can be grown in "defined" media consisting of low-molecular-weight
nutrients without protein supplements. (Quotation marks are used about
46
BASIC METHODS
[S]
the term "defined," since in most cases growth in such media appears to
be dependent on trace impurities such as inorganic trace elements that are
not included in their formulas, as will be discussed in greater detail
below.)
The growth requirements of normal cells and nontransformed lines are
not as well characterized, and at the present time, it is necessary to use
poorly characterized supplements of serum proteins or other biological
materials to obtain satisfactory multiplication of such cells. The data are
still incomplete, but it currently appears highly probable that most such
cells have specific requirements for protein growth factors that cannot
easily be replaced by improvements in the low-molecular-weight portion
of the culture medium or in other aspects of the culture system (this
volume [6]).
Types of Culture Systems
The major distinctions among culture systems are based on cellular
population density and on whether the cells are grown attached to a
substrate or in suspension. When the cell density is so low that the descendants of single cells form discrete colonies (clones), the procedures
arc described as "clonal." The term "cloning" should be used only for
procedures in which a new culture is established from the progeny of a
single cell. All other experiments involving formation of colonies from
single cells are referred to as "clonal growth." Experiments involving
cloning or clonal growth are normally done either with the cells attached
to a solid substrate under a liquid medium, or else with the cells suspended in a semisolid medium, e.g., in soft agar.
Dense cultures of cells attached to a substrate arc often called
"monolayer" cultures. Cellular yield in monolayer cultures is limited by
the surface available on which cells can grow. This can be increased by
use of roller bottles, which are rotated in such a way that their entire inner
surfaces are bathed with medium (this volume [16] and [17]), or by use of
capillary perfusion systems (this volume [14]) in which medium is circulated through tightly packed artificial capillaries, or by use of suspensions
of particulate microcarriers on which the cells can attach and multiply
(this volume [15] and [16]).
True suspension cultures, in which the cells are not attached to a
substrate, are generally done with continually stirred or shaken culture
vessels of various sizes, ranging from small spinner flasks with magnetic
stirrers to huge industrial-scale fermentors (this volume [16] and [17]).
Suspension cultures offer the special advantages of case of handling of
large numbers of cells and ability to harvest cells without the use of
[5]
47
enzymes. For certain types of cells that do not readily attach to each other
or to substrates, suspension cultures without agitation are also possible.
Both the cellular population density and whether the cells are attached or
suspended have significant effects on cellular growth requirements, as is
discussed below.
Cellular Population Density. Early in the history of cell culture, it was
observed that large populations of cells would often multiply under conditions where single isolated cells would not. Sanford and co-workers 5 isolated single cells in capillary tubes with small volumes of medium and
demonstrated that the critical variable was the volume of medium per cell
and that growth was due to "conditioning" of the medium. Subsequently,
Puck and Marcus 6 used a feeder layer of nonmultiplying heavily irradiated
cells to condition the medium in a petri dish so that it would support
multiplication of single isolated nonirradiated cells to form macroscopically visible colonies. (See this volume [21]).
At least three separate mechanisms are involved in "conditioning" of
culture media: (1) Inhibitory materials that bind to the cells are neutralized without reaching a dose per cell that is large enough to block
growth. (2) Low-molecular-weight metabolic intermediates that diffuse
readily out of cells accumulate in the extracellular medium in amounts
which, by equilibration, maintain adequate intracellular levels for biosynthesis and metabolism. Such substances, which must be for biosynthesis
and metabolism. Such substances, which must be supplied from an
exogenous source when the volume of medium per cell is too large for
effective conditioning, are referred to as "population-dependent" growth
requirements. The most common examples are carbon dioxide, pyruvate,
and the so-called "nonessential" amino acids, particularly serine. (3)
Macromolecules synthesized by the cells sometimes must build up to a
critical level before growth will occur. In addition to cases where a particular type of cell produces the macromolecules needed for its own multiplication, there are numerous cases in which macromolecules released
by one type of cell are beneficial for another.
Density-Dependent Inhibition. Most types of cells stop or slow their
multiplication at a characteristic population density in any particular
medium. The effect tends to be more pronounced with nontransformed
cells than with those that have undergone a malignant transformation. The
exact mechanisms involved are not fully worked out and remain quite
controversial. However, it is likely that for most types of cells, some
combination of cell contact, localized or generalized depletion of nutrients
K. K. Sanford, W. R. Earle, and G. D. Likely, J. Natl. Cancer Inst. 9, 229 (1948).
6 T. T. Puck and P. I. Marcus, Proc. Natl Acad. Sci. U.S.A. 41, 432 (1955).
48
BASIC METHODS
[5]
or macromolecular growth factors, and the production of cell-generated

inhibitors (chalones) is involved. The composition of the basal medium
and the amount of serum added to it both affect the cellular population
density at which density-dependent inhibition (sometimes called "contact
inhibition") occurs.
One popular assay for multiplication-promoting factors from serum
measures stimulation of DNA synthesis in cultures of nontransformed
cells (either normal diploid fibroblasts or permanent lines similar to 3T3)
that have ceased to grow in a low serum medium due to densitydependent inhibition. Caution must be exercised in interpreting data from
such an assay, however, since it measures only the first factor to become
limiting in crowded cultures, and usually does not detect all of the requirements of the cells for sustained multiplication.
Exhaustion of Media. Two other phenomena related to cellular population density are depletion of nutrients and accumulation of waste products. For clonal growth, these are normally not serious problems, but in
more crowded monolayer and suspension cultures they must be considered. In such cultures, measurements should be made to determine that
heavily metabolized nutrients such as glucose and some of the amino
acids are not seriously depleted, and that products of cellular metabolism,
particularly acids, have not accumulated to a level that is inhibitory to
further growth. The simplest test for both is generally the addition of fresh
medium. Depletion of oxygen can also be a serious problem, particularly
in large-scale suspension cultures where the surface-to-volume ratio is not
favorable (see also this volume [16]). Continuous perfusion with fresh
medium and controlled gassing with oxygen are desirable for optimum
multiplication of crowded cultures, but caution must be exercised with the
oxygen, since too much oxygen is often highly inhibitory to cultured cells.
Suspension Cultures. Apart from cellular population density, the second major difference in culture systems is whether the cells are grown in
suspension or attached to a substrate. Among cells that are not anchorage
dependent and can be grown either attached or suspended, the culture
method that is utilized affects at least two classes of growth requirements.
The first set is related to specific requirements for attachment. Media for
suspension culture generally have very low levels of calcium, to minimize
attachment of cells to the culture vessel and to each other. Media for
substrate-attached growth generally contain substantially more divalent
cations, and many nontransformed cells require quite high concentrations
of calcium for optimal growth. The second effect of substrate attachment
is that it tends to improve cellular efficiency of nutrient uptake, possibly
because of the larger exposed surface of the attached and flattened cells.
Cells that can be grown either anchored or suspended tend to require
[5]
49
higher nutrient concentrations and/or more serum protein when grown in

suspension than when grown on a substrate. In addition, suspension culture media generally have enhanced buffering to minimize acidification by
the dense cultures that they must accommodate (see also this volume
[16]).
Modifications of the Culture System that Minimize the Requirement for

Serum Protein
Treatment of the Culture Surface with Basic Polymers. For some types
of anchorage-dependent cells, the nature of the culture surface has a
major effect on cellular growth and the requirement for serum proteins.
Glass has a negative surface charge, as does typical cell culture plastic,
which is produced by selective displacement of protons from polystyrene
to generate a net negative surface charge. Recent experience in this laboratory has shown that growth of normal and nontransformed cells with
minimal amounts of serum protein can be improved substantially by coating the culture surface with a positively charged polymer. 7 Essentially any
positively charged polymer will work. The following procedure uses
poly-D-lysine to avoid possible interference with amino acid balance
studies and possible introduction of impurities carried by natural
polymers such as histone or protamine, which are also active.
Dissolve 0.10 mg/ml poly-D-lysine" HBr (tool wt 30,000-70,000) in
distilled water and sterilize by passage through Millipore type GS (0.22
/zm) filter membrane. Pipette 0.5 ml of this solution into each 60 x 15 mm
plastic tissue culture petri dish. Rock the dish gently to be certain that the
solution spreads uniformly over the entire surface. After 5 min, remove
the solution as completely as possible with a Pasteur pipette attached to
an aspirator. Add 1.5 ml of sterile water to the dish, rock it gently, and
completely remove the water by aspiration. Care must be exercised to
remove all solutions completely, since free polylysine tends to inhibit
growth when added directly to the culture medium. The dishes can be used
immediately, or allowed to dry and used at a later time. It is also possible
to do the coating without sterile precautions and then sterilize the coated
dishes with ultraviolet (UV) irradiation.
Minim&ing Damage During Subculturing. In cases where stock cultures of cells are maintained as attached monolayers, the procedures used
to remove the cells from that substrate and dilute them into new subcultures or into clonal growth experiments have a profound effect not only on
viability, but also on growth requirements of the surviving cells. The
7 W. L. McKeehan and R. G. Ham, J. Cell Biol. 71, 727 (1976).
50
BASIC M E T H O D S
[5]
requirement for serum protein of human diploid fibroblasts and chicken

embryo fibroblasts, as well as other nontransformed anchorage-dependent
cells, can be reduced significantly by use of a subculturing procedure
designed to minimize mechanical damage to cells and the internalization
of trypsin by pinocytosis. 8
Essential features of that procedure include use of the lowest possible
concentration of trypsin, use of the smallest possible volume of trypsin
solution, exposure of the cells to the trypsin solution for the shortest
possible time, and performance of all operations at ice-bath temperatures.
The minimum concentration of trypsin and the minimum time of exposure
that c[m be used depend on the type of cell, the cellular population density, the amount of serum, in the medium the monolayer was grown in,
and the length of time since the last subculturing. Since trypsin solutions
are subject to self-digestion, it is desirable to divide a freshly prepared
trypsin solution into small aliquots and store each frozen until it is to be
used. In addition, it is desirable to measure enzymic activity of each new
trypsin solution and to dilute it such that the cells are always exposed to
the same number of trypsin activity units. 8 The following procedure is
recommended for human fetal lung fibroblasts (WI-38, MRC-5, etc.)
grown to a confluent monolayer in medium MCDB 1040 supplemented
with 1.0 mg/ml dialyzed fetal bovine serum protein (equivalent to about
2% serum). The volumes that are given are for a 25 cm 2 culture flask, and
should be increased or decreased accordingly for culture vessels with
different surface areas.
Chill all solutions and media that are to be used in an ice bath. Place
the culture flask containing cells to be subcultured on ice for 15 min.
Remove the medium and rinse the cellular monolayer twice with 0.5 ml
aliquots of solution I (4.0 mM glucose; 3.0 mM KCI; 122 mM NaCl; 1.0
mM Na2HPO4; 3.3 pA/phenol red; 30 mM HEPES-NaOH, final pH of
solution 7.6). Cover the cells with 2.0 ml of twice-crystallized trypsin at a
concentration of 100/zg/ml in solution I. (This solution should have an
activity of about 3.6 units/ml measured under the actual trypsinization
conditions, s) Immediately remove as much as possible of the trypsin solution. Close the flask tightly to prevent evaporation and leave it on ice for
about 5 min with only a thin film of trypsin solution covering the cells.
Progress of the digestion may be checked by brief examination with an
inverted microscope, but the flask should not be removed from the ice
long enough to warm appreciably. When the cells are partially loosened,
add 5 ml of chilled serum-free culture medium to the flask and shake
s W. L. McKeehan, Cell Biol. Int. Rep. 1, 335 (1977).
a W. L. McKeehan, K. A. McKeehan, S. L. Hammond, and R. G. Ham, In Vitro 13, 399
(1977).
[5]
51
gently to suspend the cells. Break up clumps by gentle pipetting and dilute
the cells as needed for subculture. If a hemocytometer count is made for
clonal growth, the suspended cells should be kept on ice during that
procedure. The cells are not warmed above ice-bath temperature until
they are placed in their final culture medium and returned to the cell
culture incubator. As long as they are maintained in the cold, the cells do
not round up, but rather remain elongated and irregular in shape. All
mechanical procedures involving the cells are done as gently as possible.
For other types of cells, and for cells with different culture histories, it
is necessary to vary the amount of trypsin and time of exposure. For
example, for human-foreskin fibroblasts, previously grown under the
same conditions as the human fetal lung cells described above, about 500
/zg/ml crystalline trypsin will be needed, and the digestion time must be
increased to 15 min.
Cellular Growth Requirements
Many different variables combine to determine whether or not cells
multiply in vitro. We have already considered the fact that different types
of cells have different growth requirements, and the fact that many aspects of the culture system that have nothing directly to do with medium
composition affect growth requirements, including cellular population density, substrate attachment or its absence, the nature of the substrate, and
the way the cells are handled during subculturing. In this section, we will
consider in sequence the following aspects of cellular growth
requirements:
1. The medium must supply all essential nutrients. These include all
raw materials needed for the synthesis of new cells, substrates for energy
metabolism, vitamins and trace minerals whose function is primarily catalytic, and bulk inorganic ions whose functions are both catalytic and
physiological.
2. Physiological parameters, such as temperature, pH, osmolality, and
redox potential, must be kept within acceptable limits.
3. The culture system must be free from toxic or inhibitory effects,
including those due to excess amounts of essential components.
4. Precise quantitative adjustments, including balance relationships
among the components of the culture system, are very important.
5. Serum, which is frequently added to defined basal media to stimulate multiplication, interacts with virtually every other variable in the
culture system, and it also serves as a source of macromolecular growth
factors that are essential for multiplication of many, but not all, types of
cells in currently available basal media.
52
BASIC M E T H O D S
[S]
6. Cells multiply only when all of their requirements have been

satisfied. A holistic approach is very helpful in dealing with the many
different variables that influence multiplication and the complex interactions that occur among them.
Qualitative Requirements for Low-Molecular-Weight Nutrients

Typical cell culture media contain a mixture of defined lowmolecular-weight nutrients dissolved in a buffered physiological saline
solution. The term "nutrient" is generally restricted to substances that
enter cells and are utilized, either as substrates for biosynthesis or
metabolism, or else as catalysts in those processes, Most media also
contain a few nonnutrient components, such as phenol red as a pH indicator, and, sometimes, HEPES as a buffer: The following classes of nutrients are found in typical culture media.
Bulk Ions. The bulk ions may be required primarily for physiological
roles, such as maintenance of membrane potentials and osmotic balance,
rather than as nutrients in the narrow sense of the term, but they are
included here in view of their universal requirement and the fact that
most, if not all, also have co-factor roles in various enzymic reactions.
Cultured cells require sodium, potassium, calcium, magnesium, chloride,
and phosphate for multiplication. Bicarbonate (or carbon dioxide) is also
required for a number of biosynthetic reactions, and an exogenous source
must be provided when the cellular population is sparse. However, when
the volume of medium per cell is small, e.g. in crowded cultures, or when
loss of carbon dioxide by diffusion is inhibited, the cellular requirement
can be satisfied entirely by metabolically generated carbon dioxide.
Trace Elements. Inorganic trace elements tend to be ubiquitous contaminants of the chemicals used in preparation of media, and most socalled "defined" media list in their formulas few, if any, of the trace
elements known to be essential for whole animal nutrition. Limited data
suggest that growth in such media is dependent on trace elements that are
present as contaminants. 10-~2Cultured cells clearly require iron, zinc, and
selenium, and there are data suggesting requirements for copper, manganese, molybdenum, and vanadium, although considerable background
growth occurs without deliberate addition of those elements, presumably
because of contaminationY 3
10 W. G. Hamilton and R. G. Ham, In Vitro 13, 537 (1977).
n j. A. Thomas and M. J. Johnson, J. Natl. Cancer Inst. 39, 337 (1967).
n T. Takaoka and H. Katsuta, Exp. Cell Res. 67, 295 (1971).
13 W. L. McKeehan, W. G. Hamilton, and R. G. Ham, Proc. Natl. Acad. Sci. U.S.A. 73,
2023 (1976).
[5]
53
It is likely that additional elements whose requirements are currently

masked by contaminants will prove to be essential for cellular multiplication in the future. Until all such requirements are known, there is a constant danger of growth failure in previously adequate media due to increased purity of chemicals. This has already happened due to selenium
deficiency in the case of medium F12. TM Cultured cells grown in proteinfree media or in media supplemented with minimal amounts of exhaustively dialyzed serum hold great promise for future studies of trace element requirements and metabolism, particularly in the case of humans,
where direct studies involving long-term dietary depletion are not a realistic alternative.
Sugars. All commonly used cell culture media contain a six-carbon
sugar as a source of metabolic energy and reduced carbon for biosynthesis. Glucose is the most common form, although in some media it is
partially or fully replaced by galactose, which is metabolized more slowly
and leads to less buildup of acid in the culture medium. There are reports
that a mixture of pyruvate and ribose will replace glucose in some systems, but this mixture is not widely used.
Amino Acids. Thirteen amino acids, arginine, cyst(e)ine, glutamine,
histidine, isoleucine, leucine, lysine, methionine, phenylalanine,
threonine, tryptophan, tyrosine, and valine, are generally considered essential for cultured cells. Under specialized conditions, some types of
cells can synthesize enough glutamine to satisfy their needs, and in rare
cases, sufficient cystine can be made from methionine to support multiplication. Among the "nonessential" amino acids, serine is frequently required on a population-dependent basis for growth of sparse cultures.
Asparagine is required by certain malignant cells, including a number of
lines of leukemia. Proline is required by the widely used mutant Chinese
hamster ovary line, CHO, and may be beneficial on a populationdependent basis for other types of cells. Glycine is sometimes essential in
cases of borderline folic acid deficiency, or if folic acid uptake or
metabolism is inefficient, as has been reported for several types of primary cultures. All 20 of the amino acids that are involved in protein
synthesis are often included in culture media to reduce cellular biosynthetic loads, and it is sometimes possible to manipulate experimental conditions to make growth dependent on normally nonessential amino acids
such as glutamic acid, aspartic acid, and alanine. The quantitative requirement for essential amino acids becomes larger when nonessential
amino acids are not provided, since in such cases a portion of the essential
amino acids must be degraded to provide a nitrogen source for synthesis
of the nonessential amino acids. Hydroxyproline is included in the formulas of a number of older media, but there is no convincing evidence
54
BASIC METHODS
[5]
that it is actually required. The tripeptide glutathione is also sometimes

included, but again evidence for its requirement is not convincing.
Vitamins. Cultured cells generally require the B vitamin: biotin, folic
acid, niacinamide, pantothenic acid, pyridoxine, riboflavin, and thiamin.
Vitamin B12 is probably also required, but it is often difficult to achieve a
clean background, t~-Lipoic acid (thioctic acid) is sometimes added to
culture media, but there is little evidence that it is actually needed. Coenzymes can generally be substituted for the parent vitamins, but, with the
possible exception of folinic acid, they appear to offer no special advantage. Ascorbic acid may be beneficial, particularly to cells that synthesize
collagen in culture, but its instability to oxidation makes it very difficult to
work with, and there continue to be conflicting reports about its effectiveness. The fat-soluble vitamins appear not to be essential, at least for the
types of cells that have been carefully studied in serum-free media. The
possible role of fat-soluble vitamins in multiplication of normal cells is still
clouded by the requirement of such cells for serum proteins, which could
be serving as carriers of fat-soluble vitamins.
Choline and Inositol. Both choline and meso-inositol are required by
most types of cultured cells, although some Chinese hamster ovary lines
show no response to inositol, even at clonal density in completely synthetic media. These compounds are often listed as "vitamins" in the
formulas of culture media, but their metabolic roles are clearly as substrates, rather than as catalysts.
Other Organic Nutrients. The organic nutrients listed above (glucose,
13 "essential" amino acids, 7 or 8 B vitamins, choline, and meso-inositol)
constitute the apparent minimal qualitative requirements for organic nutrients for multiplication of a number of highly adapted permanent cell
lines. These are the only nutrients in certain "minimal" media, such as
the widely used Minimum Essential Medium of Eagle (Tables I and II,
see text p. 84 for main description of Tables I and II). However, a number
of additional requirements can be demonstrated for various other cultures.
A purine source (adenine or hypoxanthine) together with thymidine is
often beneficial, particularly in cases where folic acid is in short supply or
used inefficiently. Under some conditions uridine may also be beneficial.
Some types of cells appear to have specific requirements for
polyunsaturated fatty acids, and in other cases, mixtures of mono- and
polyunsaturated fatty acids, e.g., oleic plus linoleic, are reported to be
beneficial. Cholesterol is also reported to be beneficial in certain cases.
Free lipids present two major problems in aqueous culture media--they
are rather insoluble and they are quite toxic. It currently appears that the
use of membranous vesicles (liposomes) generated by sonication of phos-
[5]
55
phatidylcholine (lecithin) suspensions 14 may provide a soluble and relatively nontoxic source of lipids for cells grown in adqueous media.15,16
Human diploid fibroblasts and chicken embryo fibroblasts both appear to
have definite requirements for lipids.l'2
Putrescine or other polyamines are required for growth of Chinese
hamster ovary cells in protein-free media and are beneficial to certain
other cells. Many types of cultured cells exhibit a population-dependent
requirement for pyruvate or other small o~-oxo acids. The requirement is
relatively nonspecific and may reflect the need for a substrate that can be
used to oxidize NADH, rather than the need for a specific metabolic
intermediate. A variety of other organic compounds have been incorporated into some of the more complex media on the basis that they might be
beneficial, and it probably does no harm to include them. However, this
philosophy can be a serious barrier to the development of serum-free
media since substances that do not appear to be harmful in the presence of
large amounts of serum sometimes prove to be quite inhibitory in media
with lower concentrations of protein.
Physiological Requirements
In addition to satisfying nutrient requirements, the culture environment must have physicochemical properties that fall within physiologically acceptable limits for cellular survival and multiplication. Important
parameters include the following:
Temperature. Cellular multiplication rate typically increases with temperature until a limiting value is reached, above which further increases in
temperature rapidly become inhibitory. Most laboratories do not have
enough incubators available to perform properly controlled experiments
on the effects of temperature on cellular multiplication, and except for
studies involving temperature-sensitive mutations of cells or viruses, little
attention is usually given to this variable. However, different species of
warm-blooded animals do have different body temperatures, and certain
tissues and organs such as skin and testes are normally maintained at
temperatures below the overall body temperature. Therefore, studies of
temperature optima could prove to be very profitable in certain cases.
Typically, cultures of cells from warm-blooded animals are grown at 37
and cultures from cold-blooded animals at temperatures near the upper
limit of optimum body temperature for the intact animal.
24 G. Poste, D. Papahadjopoulos, and W. J. Vail, Methods Cell Biol. 14, 33 (1976).
15 N. N. Iscove and F. Melchers, J. Exp. Med. 147, 923 (1978).
le W. L. McKeehan and R. G. Ham, Abstracts 1978 TCA Meeting, In Vitro 14, 353 (1978).
56
BASIC M E T H O D S
o_
_ r,-~
- ~ ~
"o
~
[S]
~o
_~
.'~
~,~o.,o~.~-~
~'~"~ ~ 2 , ~
~ ~
"
~ ~ - ~ .~'~
~~ ~.~ B~
,,
~.~
I
m
ZZr~
J
~z
Z
t-
.I
0
~,
~
~
~":
.~
,~
g
S
~-.
_~i
.~-~
.~
~.0
~r.~
i~
[5]
MEDIA
AND
GROWTH
REQUIREMENTS
.~
.~'=o
"~-
=,.,
57
.~
~- ~
~..
.=_
4-
-F
,,~
.~ ~
~
~ ~
.
..~ ~ ~
~.-~
4-
'~
"5
b~
4-
,..1 . l
F-
~-
.1
ZZ
.=
E
E
[,,-,
F-
["-
.4
=..~
's
.=
",~
~-
58
BASIC
METHODS
[S]
~.
o_~i ~ ~ ~
-
~-~
~
~.~
.o ~,
~o.~
.~oo i . ~
...~
~ '~
~ ' ~ .~
x:
~ r~
~o~
.~
+
6t~
. ~
6
t~
mt
t~
6
Z
a~
[5]
MEDIA
.'
AND
.'
GROWTH
"~
~'
'~
E
.=.~
59
REQUIREMENTS
,a
"~
"El "=
.~
.~
.=
=
._=
.=_ ~
~a
~
e
=~
,~
.-
....
.=
._=
=
_
~.=
~=-~
g
o'=
.~,.= ~
).~-~
~:~
++l
.) . 1 Z
<
~=
~:'a
~.=~.~
= ~'~'~
I+
,.az
~ - ~
.=_
.=_
m
r,.
E.= ~ ~.~-
..=
E
..~
~ ; ~ ~.~'~'~
r~L)
.)
L)
L)
x.t
.~-
. ~. ~ ~,=~. ~
BASIC METHODS
60
.~
-~~
.:
~:
.~
o '~-
.~ ~ - ~
~.,~~-.-~ , .~~, , ~
~'0~
=~-.=
. a . ~
~ r ~
~,~,~
~ ~..~
~ 4~d,<~
~,,~
~.~.e~-~_~a'~z=
~ ' 4 ~
.~
..~.~
[S]
.~
4~
",4~,,-;~
~:~ ~
",~,4-q
.-
~.~ g~"
=,,
o~
"
[5]
&
f~zo~
~ =
.....
~o
~ ~
~-~
II
'i
2~
"==g=
~'~
o
-~ > > -
~
=
61
62
BASIC METHODS
[S]
oo
~
m
6
o o o o o ~ o ~ o ~
t-
in
m
m
0
0
0
~ o o o o o o o o o ~ o o
[5]
MEDIA
AND
GROWTH
63
REQUIREMENTS
;<
,z= .-
II
~111111
I I
IIIIIII
I I
I I il
~.~
ii
~lll~li
I I I
itlitiliili
.r-
o
~
o o o o o o
.<
],<
I I I
~ ~
I ~
I~
I ~ ~
I I
64
BASIC METHODS
~<
[5]
IIl,lllll::~lllllll,llllllll
tttltt
tii
t~t~titttlttltlttttt
i i ~t~l~t
it~ttt~
i ~ iii
i i iiIi
ii
t~ltti~ttttitltttt
tltu~,l
I,Illl,ll,l~
~,~,~
o(
u~
it
. . . .
Illll
I III
I I I II~l
~4
I ~ l t ~ t ~ ~
~ ~ = = ~
IS]
MEDIA
AND
GROWTH
REQUIREMENTS
65
~~
~o~
.e
~eg
.N
o.
~q
m
~
I~
Illllll~l~
rqoi
I~11
~,~
IIIIIIIIIIIIIIII
Im~m~m~m
~
. . . . . . .
Illllllllllll
II I
IIIllllllllll
II I
IIIIIll~lllllllllllllllllll
"0
66
BASIC METHODS
[S]
r~
III
c~
IIIIII
IIIIIIIIII
IIIIII
e~
I III
II Ill
llll~illlllll
II
III
lllIlllllllll
I II~l
II III
III
I~IIIIII~II
I IIII
IIIIIIIII
llllll~Itlllllillllllrltli
IrllII~II~IIIIIIII~II1~tl
r--
~S
,~
~ o
....
~.~'~
~'
~ ~ .
[5]
MEDIA A N D GROWTH REQUIREMENTS
67
Illlllll
e4
e,i ,.-:
I I
II
~l
iI
IIIIIIif
II
II
IIIIIIPI
r,i
~1
r4
I I I
III
III
~1
II
III
eq
~ o 4
r I I
~1
II
z~
II
IIII
I I
Ir
lit
68
BASIC METHODS
I-~o
[S]
. . . . . . . .
t ~
II~lll~lllllllll~llllllllll
I I I I I I I I I I I I I I t I I~ t I ~ t I I I r I
6
I I I I~
I I I I I I I I I~
I I I I I
I~
Illlllllllllllllll~llllllll
-.I
",
'~
II,lll
Jl
llll,lll
llllt,ll
[5]
MEDIA A N D GROWTH REQUIREMENTS
69
q
b~
III
>- I"O
III
II
II
eq
-6
-r"
III
II
I II
IIII
III
III
I I
III
I~111
I I
[.h
r-O
L)
o~
o
z
L~
IIIIIIIIII1~1111
70
BASIC METHODS
I I
I I
II
II
[S]
cD
-:
-:1
II
I
m
,..r.
I I
IM
I~
[5]
~:
:..~
"~
=o
~.~
.~ . .~ . ~
.~
:~
o.
o_
. ~
.-
o.o~
71
=o
!ill
ilil
il!iiiii
ii
~=
.-
.~_
~.
72
BASIC METHODS
IS]
pH. The pH range that supports optimum cellular multiplication is

generally quite narrow and varies somewhat with the type of cell. For
example, the optimum range for clonal growth of human diploid fetal lung
fibroblasts (WI-38) with minimal amounts of serum protein is pH 7.157.45, with some growth from pH 6.8-7.8. The optimum for chicken embryo fibroblasts under similar conditions is pH 6.95-7.30, with some
growth from pH 6.4-7.7. Permanent lines of transformed cells are often
much more tolerant of acidic conditions than diploid cells.
Carbon Dioxide Tension. The population-dependent requirement of cultured cells for carbon dioxide and/or bicarbonate ion as substrates for
biosynthesis has already been described above. Since bicarbonate and
carbonic acid are in a rapid equilibrium relationship determined by the
pH, and since carbonic acid equilibrates with water and gaseous-phase
carbon dioxide, it is impossible to maintain an adequate concentration of
carbonic acid and bicarbonate in the culture medium without also maintaining an increased partial pressure of carbon dioxide in the gaseous
phase. For monolayer or suspension cultures this can be accomplished by
limiting diffusion of metabolic carbon dioxide out of the culture vessel,
although there is danger of over-acidification due to excess accumulation
of carbon dioxide. For clonal cultures, and for crowded cultures where
better pH control is desired, the culture chamber is gassed with a mixture
of humidified carbon dioxide in air, and the culture flasks or dishes are not
tightly closed. Most culture media are designed for use with 5% carbon
dioxide, which is close to the value found in body fluids, and some are
used with as much as 10%. Recent studies in our laboratory indicate that a
concentration oT 2% is sufficient, and possibly even superior, for clonal
growth of sensitive diploid cells with low levels of serum protein.
Buffering. The necessity of using carbon dioxide in the incubation
chamber creates additional problems of pH control. A bicarbonatecarbon dioxide buffer will provide excellent pH control as long as the
cultures are left in the incubator, but in the absence of supplementary
buffering, media that contain bicarbonate become alkaline very rapidly
due to loss of COE when removed from the incubator. Current practice is
moving toward the use of organic buffers, such as ot-glycerolphosphate or
HEPES, at concentrations sufficient to reduce the pH change on removal
of cultures from the incubator to a few tenths of a pH unit. We currently
add 30 mM HEPES to media that are to be used with 2% carbon dioxide.
Sufficient sodium hydroxide is added to bring the pH to the desired value
under incubation conditions.
Since both temperature and CO2 tension affect the final pH of the
medium, it is important to measure pH under actual incubation conditions. However, after determination of the amount of pH change that
occurs when a particular medium is placed in the incubator, it is possible
[5]
73
to adjust the pH at room temperature and in air so that it will shift to the
correct value in the incubator. It is unnecessary to include bicarbonate in
the formulation of a medium whose pH is adjusted in this manner, since
bicarbonate is formed from dissolved carbon dioxide as the buffered
medium equilibrates in the incubator. One other potential area of concern
is possible inhibitory effects of large quantities of artificial buffers. In the
case of HEPES, there does not appear to be a problem up to at least 50
mM, provided that appropriate osmotic corrections are made.
Osmolality. The optimum range of osmotic pressures for cellular
growth is quite narrow and varies with the type of cell and the species.
For human fetal lung fibroblasts, optimum clonal growth occurs between
about 250 and 325 mosmol/kg, with some growth from about 200-375
mosmol/kg. For chicken embryo fibroblasts, the optimum is about 275325 mosmol/kg, and the range of suboptimum growth is about the same as
for the human cells. Cells from amphibia require significantly lower osmolality than avian or mammalian cells, and this is often achieved by
diluting standard culture media with distilled water. Cells from parts of
the body such as the kidneys that are exposed to differing osmolalities
may also have different requirements.
The narrowness of the optimum range of osmolality makes it necessary
to adjust the concentration of sodium chloride when major additions are
made to a culture medium, such as the use of 30-50 mM HEPES buffer (plus
enough NaOH to bring it to a physiological pH). Osmotic considerations
may also be important when media are used with high concentrations of
carbon dioxide. With an atmosphere of 10% carbon dioxide and a pH of
7.4, the combined concentrations of carbonic acid and bicarbonate ion
generated by carbon dioxide dissolving in a buffered medium add up to
about 30 mM. (See also this volume [16] for a discussion of osmolarity.)
Humidification of Incubators. Since rapid equilibration between the
culture medium and the gaseous carbon dioxide-air mixture in the culture
chamber is frequently important, culture dishes usually cannot be sealed
tightly. The narrow limit of acceptable osmolality requires that essentially
all evaporation be prevented during the time that the dishes are in the
incubator, which may be 2 weeks or longer for some clonal growth experiments. Limitation of evaporation is accomplished by maintaining the
relative humidity very close to saturation in the incubation chamber. This
can be a difficult task in a dry climate. One of the specific requirements is
that the water that is used for humidification be kept at the same temperature as the rest of the incubator. Each time that the incubator door is
opened, dry air enters and cools the water by evaporation. If there is
inadequate heat transfer, the water may remain cool for a prolonged
period and fail to humidify the incubation chamber adequately to prevent
evaporation and maintain the osmolality of the culture medium. Similarly,
74
BASIC METHODS
[S]
if there is a cool spot anywhere in the incubator, such as a poorly insulated door, it will serve as a site of condensation and prevent the partial
pressure of water vapor from reaching the needed level. To avoid these
problems, there must not be any surface in the incubator, including that of
the water used for humidification, that is more than a few tenths of a
degree centrigrade below the temperature of the medium in the culture
dishes.
One of the best ways to monitor humidification is to measure the
osmolality of control dishes of medium at the beginning and the end of the
incubation period for the experiments being performed. If there is a
significant increase in osmolality, the humidity is not high enough and
evaporation is occurring. Excessive humidification that leads to condensation on the culture dishes is also undesirable, since it increases the risk of
microbial contamination.
Oxygen Tension. A minimal amount of oxygen is essential for the multiplication of most types of cells in culture. However, the partial pressure
of oxygen in normal body fluids is significantly less than that of air. Systematic studies of the effects of oxygen tension on cultured cells are
technically difficult due to diffusion gradients, leakage, and release of
dissolved oxygen from plastic culture vessels. However, there are numerous reports that growth of cultured cells can be improved by reducing the
percentage of oxygen in the gaseous phase to between 1 and 10%. The
trace element, selenium, is a component of the enzyme glutathione peroxidase, which destroys metabolically generated peroxides. The inhibitory
effects of excess oxygen are much more severe when cells are marginally
deficient in selenium than when adequate amounts are supplied.
There have been a number of attempts to correlate cellular growth
with redox potential measured by inserting electrodes into the culture
medium, but the results are difficult to interpret because a cell growth
medium is not an equilibrium system. In fact, for cellular growth to occur
at all, the cell must be able to catalyze an energy-yielding reaction among
components of the culture system (normally glucose and oxygen). Cysteine and various other reducing agents are present in large amounts in
some of the older culture media, but there is little evidence that the
reducing conditions that they were intended to generate are actually
needed, and in at least some cases, very high concentrations of cysteine
are clearly inhibitory.
Freedom from Toxicity

The requirement that the culture system must be free from toxic or
inhibitory effects might seem so obvious that it would need no discussion.
[5]
75
However, there are many subtle types of toxic and inhibitory effects that
are not immediately obvious. A sufficiently large excess of any component
of the culture system is inhibitory, even if only for osmotic reasons, and
many essential nutrients become inhibitory at surprisingly low levels.
Contaminants in the chemicals or water used in medium preparation can
be toxic. Membrane-type sterilizing filters may introduce substantial
amounts of detergents into culture media. Because of this, detergent-free
filters should always be specified. Short-wavelength visible light interacts
with certain components of culture media (riboflavin, tyrosine, tryptophan) to generate toxic photoproducts. ~7 Therefore, it is necessary to
keep exposure of media, to ordinary laboratory fluorescent lights at a
minimum. Toxicity can sometimes be generated by interactions among
components of the culture medium that are not individually toxic. Thus,
for example, bovine serum cannot be used in media that contain spermine
or spermidine because its high content of spermine oxidase will generate
toxic oxidation products from these polyamines. Putrescine is safe to use
with bovine serum, however, as it is not affected by the enzyme. In
certain cases, inhibitory contaminants may alter responses to other components of the culture system. For example, the presence of traces of
cadmium may increase substantially the amount of zinc needed for optimal multiplication. Similarly, in certain cases, a substantial portion of the
requirement for serum protein may be to overcome inhibitory effects.
Quantitative Requirements
The full significance of quantitative requirements for cellular multiplication is just beginning to be realized. For classically studied permanent
cell lines, precise quantitative adjustments do not seem to be very important, although there is some evidence that balance relationships affect
growth. TM For normal cells that are now receiving increasing study, however, such adjustments are proving to be critically important.l'2.9
For every required component in a culture medium, a three-part
growth response curve, similar in principle to the idealized example
shown in Fig. 1, can be obtained. In the first region on the left, multiplication is roughly proportional to the amount of the component added. In the
second region, in the center, the growth response is "saturated," and the
response curve exhibits a "plateau," within which growth is virtually
independent of concentration of the component. Finally, on the right,
above a critical threshold, the component becomes inhibitory and the
growth response is negative.
a7R. J. Wang,In Vitro 12, 19 (1976).
as R. G. Ham, In Vitro 10, 119 (1974).
76
BASIC METHODS
[5]
Plateau of
Optimum Growth
i-
Stimulation?
o
o.
\-
~lnhibition
Midpoint of Plateau
on Semilog Plot
is Selected as
"Optimum"
Concentration
o
o
/ . . . . . . . I , , , : , , i.d
,
10-2
i 0 -t
, , li,.,l
i0 o
* I
tl..I
i0 i
,~'~.,,I
......
I ,.
,
102
103
Nutrient Concentration ( relati,ve to optimum)

FIG. 1. Idealized clonal growth response illustrating the method used to select the "optimum" concentration of a nutrient for inclusion in a culture medium. From R. G. Ham and
W. L. McKeehan, In Vitro 14, 11 (1978).
Response curves of this type can be used to establish an "optimum"

concentration for each component of the medium. Generally, it is desirable to select a concentration that is as far removed as possible both from
deficiency and from inhibitory effects. This is accomplished by selecting
the midpoint of the plateau of the growth response on a semilog plot, as
illustrated in Fig. 1.
We have recently systematically adjusted the concentrations of all
components of the medium to optimum concentrations for clonal growth
of several different types of normal cells with minimal amounts of serum
protein. This was done sequentially at progressively lower concentrations
of serum protein until no further reduction could be achieved. 1,2,9 The
result was a very substantial decrease in the amount of serum protein
needed, 19 as well as a significant improvement in growth. Parallel studies
with normal human, mouse, chicken, and duck fibroblast-like cells have
shown that each has a characteristic pattern of quantitative requirements,
and each grows best in a medium that has been optimized specifically for
its own needs. 1,2,9,2 For example, medium MCDB 105, which was optimized for human diploid fibroblasts and medium MCDB 202, which was
optimized for chicken embryo fibroblasts, both contain the same 56 components, but they differ in the quantities that were found to be optimal for
~9 W. L. McKeehan, D. P. Genereux, and R. G. Ham, Biochem. Biophys. Res. Commun.
80, 1013 (1978).
2o W. L. McKeehan, K. A. McKeehan, and R. G. Ham, in "The Growth Requirements of
Vertebrate Cells in Vitro" (R. G. Ham, C. Waymouth, and P. J. Chapple, eds.). Cambridge Univ. Press, London and New York (in press).
[5]
77
clonal growth of 26 of those 56 components (Table II). None of the "standard" media developed over the years for permanent established lines
will support clonal growth of these cells with comparably small amounts
of serum protein.
Another aspect of quantitative requirements that must be considered is the balance among interrelated components of the culture system.
For example, in addition to osmotic requirements and absolute requirements for sodium and potassium, the Na+/K + ratio appears to be important, with different optimum values for different types of cells. Another
set of balance relationships that appears to be very important occurs
among the amino acids.18 The ratios of amino acids in media optimized for
various normal and nontransformed cells differ significantly. (Compare
MCDB 105,202, and 401 in Table II.)
Serum and Its Relationship to Other Growth Requirements
The reason that serum is such an effective supplement for promoting
cellular multiplication appears to be that it contains a large number of
different growth-promoting activities in a physiologically balanced blend.
The multitude of different growth-promoting roles that serum is capable of
playing can perhaps best be appreciated in terms of the many different
ways in which the requirement for serum can be partially replaced. In a
more detailed review, we have listed and discussed 24 different ways in
which serum could conceivably contribute to cellular multiplication.2
Only a few of the more significant ways are described here.
Whole serum contains most of the low-molecular-weight nutrients
needed for cellular multiplication, and at one time the culture of permanent cell lines such as Mouse L and HeLa was routinely done in mixtures
of serum, embryo extract, and saline, with no defined organic nutrients
added to the medium other than glucose. Dialyzed serum retains many
small molecular nutrients in bound form, and it also possesses many
other growth-promoting activities. Large amounts of serum protein allow
cells to multiply with concentrations of essential nutrients that are inadequate for multiplication with smaller amounts of serum protein.
Dialyzed serum is also able to reduce the inhibitory effects both of contaminants and of essential nutrients that are present in excess amounts. In
some cases, serum macromolecules may buffer toxic nutrients by binding
them and releasing them in small amounts as their free concentration in
the medium is reduced by cellular metabolism. Serum protein neutralizes
trypsin and other proteases and also makes cells better able to overcome
the damage done to them by harsh subculturing procedures. Serum has an
undefined effect on the interaction between cells and their substrate,
78
BASIC METHODS
[S]
which can be replaced by coating the substrate with polylysine. Serum

provides carder proteins for water-insoluble substances, such as lipids,
that appear to be required by normal cells. Serum may also provide peptide hormones or hormone-like growth factors, or other proteins that are
specifically required for cellular multiplication (see also this volume [6]).
The preceding is only a minimal outline of the very large number of
different roles that serum appears to perform in supporting the multiplication of various types of cells in various types of basal media. The number
of such roles that are required appears to vary greatly from one type of
cell to another. As mentioned earlier, there are some types of very highly
adapted cells that can be grown without serum protein in a number of
different defined media. There are also a few other cell types that can be
grown without protein supplementation, but only in media that have been
tailored precisely to satisfy their individual needs. 1,21
Appropriate adjustments in the many different variables that serum
interacts with greatly reduce the amount of serum protein that is needed
for growth of a variety o f " n o r m a l " cells, but it currently appears unlikely
that this approach will make it possible to eliminate completely the requirement of such cells for macromolecular growth factors. Requirements
in the milligram per milliliter range have proven easy to replace, but for all
of the diploid cells that we have studied, requirements in the microgram
per milliliter range are not replaceable by any combination of polylysinecoated dishes, improved subculturing, and qualitative and quantitative
optimization of the basal medium used to assay for the multiplicationpromoting activity. Preliminary fractionation studies have reduced the
total amount of active material needed to a few micrograms per milliliter
in the case of human diploid fibroblasts, and the number of components
that can be detected in the active fraction suggests that the ultimate requirements for individual active components will be in the nanogram per
milliliter range when purification is completed. 19
Although the ~complete set of growth requirements has not been fully
worked out for most types of cells, there are numerous reports in the
literature that peptide hormones or hormone-like growth factors are
needed in the nanogram or microgram per milliliter range for multiplication of a variety of types of cells. A detailed discussion of these factors is
beyond the scope of this article, but they have recently been reviewed by
Gospodarowicz and Moran, zz and a number of them are discussed in a
recent symposium volume. 3 They range from well-characterized peptide
hormones like insulin to specialized growth factors, such as epidermal
growth factor, fibroblast growth factor, and the somatomedins, and on to
21R. G. Ham, Proc. Natl. Acad. Sci. U.S.A. 53, 288 (1965).
2~ D. Gospodarowiez and J. S. Moran, Annu. Rev. Biochem. 45, 531 (1976).
[5]
79
factors that have not yet been fully characterized chemically, such as
"platelet" factor and "multiplicaton stimulating activity" from conditioned medium. Gordon Sato's research group has recently reported
complete replacement of the serum requirements of several permanent
cell lines with mixtures of hormones and peptide growth factors z3 (see also
this volume [6]).
Some of the peptide growth factors, such as epidermal growth factor
and fibroblast growth factor, are beginning to be produced commercially.
Such factors will undoubtedly play increasingly important roles in the
study of cellular growth requirements during the next few years, as will
numerous factors that are currently under study, but not yet fully purified
or characterized. It currently seems reasonable to anticipate that growth
of certain normal diploid cells i n " defined" media supplemented only with
nanogram per milliliter quantities of purified macromolecular factors will
become possible within a few years.
Holistic Approach to Cellular Growth Requirements

There are so many individual requirements for multiplication of cells
in culture that it is difficult to gain a clear perspective of the total set of
conditions that must be satisfied before cells will multiply. The problem is
further complicated by a multitude of complex interactions that occur
among the individual variables, including quantitative balance relationships, the existence of alternate methods for satisfying many of the individual requirements, and the ability of serum to interact with virtually
every other variable in the entire set. Because of these interactions, we
have found it desirable to adopt a holistic view of cellular growth requirements, in which all variables that affect cellular multiplication in any
way are viewed as members of a single complex interacting set, which is
so constituted that a change in any one of its variables can potentially
have an effect on the growth response to any other variable in the entire
set. 1 Systematic study of the effects of individual variables requires
specifically that all interactions be recognized and that all other significant
variables be held constant while the effects of each single variable or
interacting group on cellular multiplication are being studied.
Preliminary Decisions to Be Made before Choosing a Medium

The choice of which medium to use will normally be one of the last
decisions made before an experiment is undertaken. Preliminary decisions
23 I
Hayashi, J. Lamer, and G. Sato,
In Vitro 14, 23 (1978).
80
BASIC METHODS
[5]
related to the experimental objective that first must be made include: (1)
what kind of cell to use; (2) what type of culture (monolayer, suspension,
clonal, etc.) to use; and (3) what degree of chemical definition is needed.
Choosing Which Cell to Use

In most cases the first decision to be made will be what cell type to use
in the experiment. That choice is ultimately determined by the goals of the
experiment. Important considerations include: (1) What differentiated
properties are needed in the cell that is selected? (2) Does it matter which
species is used? (3) Does the experiment require the use of normal diploid
cells or can it be done just as well with permanent lines of transformed
cells whose growth requirements are simpler and better understood?
Differentiated Properties. For some types of research that directly involve differentiated functions, highly specialized differentiated cells must
be used. In such cases, the primary consideration is whether to use a
normal diploid culture or a permanent line of differentiated tumor cells.
The tumor lines will generally grow in simpler media and be easier to
handle, but are also less likely to show fully normal regulation of expression of their differentiated properties. For experiments that do not directly
involve differentiation, there is usually a considerable latitude in the type
of cell that will allow the experimental objective. In such cases, it is
desirable to select a cell whose growth requirements are relatively well
understood and that will grow well in the type of culture and with the
degree of chemical definition needed for the experiment.
Species. In certain cases the experimental objectives will dictate cells
from a particular species. Examples include the use of human cells for
medically related studies and the use in culture of the same species that
was used in a previous study that did not involve culture. Since species
has a significant effect on quantitative growth requirements and balance
relationships, particularly for normal cells, it is important to employ a
medium that will work well with the specific species. In cases where
species is relatively unimportant to the experimental objectives, it is desirable to select a species whose growth requirements are relatively well
understood and whose growth properties in culture are compatible with
the experiment to be performed.
Normal versus Transformed. Strong arguments can often be mustered
on both sides of the question of whether to use a normal diploid cell or a
transformed permanent cell line. The permanent lines, and particularly
those that can be grown without serum, offer obvious advantages in
terms of ease of handling and more precisely defined culture systems.
However, at the same time, they are grossly abnormal cells that have
[5]
81
undergone extensive adaptation and chromosomal evolution in culture.

They have generally lost essentially all growth regulation, and except for
differentiated tumor lines, they exhibit few, if any, differentiated properties. If the experimental variable that is being studied can be shown not to
differ significantly between permanent cell lines and normal cells in the
intact animal, it may be worthwhile to employ the permanent lines and the
more highly defined media that accompany them. However, such a choice
should be made cautiously, and the results that are obtained from such a
study should ultimately be verified with unaltered normal cells.
Choosing a Culture System
The choice of a culture system is influenced both by the type of cell
that has been selected and by the experimental objectives. The major
considerations involved are:
1. Is the actual experiment to be done in cell culture, or is cell culture
simply being used to grow cells for use in other types of experimentation?
2. How many cells are needed?
3. Is it important to be able to evaluate the responses of individual
cells?
4. Are the cells to be grown attached or suspended?
The first three questions are all interrelated. If cell culture is to provide
cells that are then harvested and used for other experiments, it is likely
that reasonably large numbers of cells will be needed. In such cases, it
generally will be desirable to use monolayer or suspension cultures to
obtain the required number of cells with a minimum of effort. For very
large yields, specialized systems such as roller bottles and capillary perfusion are useful. In all cases where high cellular yield is necessary, it is
important to select media that will support dense growth and to renew the
media either by periodic replacement or by continuous perfusion as
needed for optimal growth. In cases where cellular responses are being
measured in culture, the number of cells needed depends on the sensitivity of the assay procedure. Large populations may be needed to obtain
enough material for chemical analysis. However, it may be more important to look at the responses of individual cells in clonal growth assays. In
the latter instance, it is important to select a medium that satisfies all of
the population-dependent requirements of isolated cells. Such media generally have lower concentrations of nutrients to avoid inhibitory effects,
and are not particularly well suited for dense cultures.
The option of using either attached or suspended cultures is available
only when a cell type that is not anchorage-dependent is selected. In most
cases, suspension culture is used only as a convenient means of growing
82
BASIC METHODS
[5]
large numbers of cells, rather than for performing actual experiments. If

suspension culture is selected, it is usually necessary either to select a
medium designed for suspension culture or to modify a conventional
medium by reducing its calcium content. Many of the media designed
specifically for suspension culture have particularly high concentrations of
nutrients to avoid possible depletion by the dense cellular populations that
are involved.
How Much Chemical Definition Is Needed

Because all chemicals contain some amount of impurity, there is no
such thipg as a medium that is truly chemically defined. However, the
options available for cell culture range from "defined" media prepared
from the best available grades of chemicals and water to media that are
supplemented with large amounts of grossly undefined additives such as
whole serum and embryo extract. The decision of what degree of chemical definition to seek for a particular experiment is influenced both by the
type of cell that is used and by the experimental objecti~ces.
At present, all normal cells require at least some serum protein for
sustained multiplication, and many whose requirements are not yet well
characterized require large amounts of whole serum. Most permanent
lines can be grown with small amounts of serum protein, and, at least in
dense cultures, many can be grown in synthetic media without any supplementation. 24'25 In a few cases, clonal growth without serum supplementation is also possible.I
When serum does not interfere in any way with the objectives of the
experiments that are to be performed, it is often simpler to use serumcontaining media than to try to replace the serum. Such an approach will
usually work when the only objective is to grow large numbers of a particular cell. However, when cellular metabolism or interactions between
the cells and their environment are being studied, experimental results are
likely to be influenced by components of the undefined and not strictly
reproducible supplements.
Where the macromolecular components of serum do not interfere, but
a reasonably defined small molecular environment is needed, it is often
convenient to use dialyzed serum as a supplement. Although dialyzed
serum is frequently referred to as "serum protein" in this article, it is
important to remember that it is actually a complex mixture that also
contains substantial amounts of lipids and carbohydrates in the form of
24 K. Higuchi, Adv. Appl. Microbiol. 16, 111 (1973).
2s H. Katsuta and T. Takaoka, Methods in Cell Biol. 6, 1 (1973).
[5]
83
lipoproteins and glycoproteins, as well as an assortment of bound small

molecules, all of which can interfere in some types of experiments. For
example, with cells that can be grown without serum protein, small
amounts of serum protein effectively replace the requirement for biotin. It
is not clear at this time whether the effect is due to bound biotin, or to
end-product substitution of fatty acids from lipoproteins, whose synthesis
would otherwise require biotin. Similar results are also observed for most
trace elements and for certain amino acids. For example, tryptophan
binds strongly to serum albumin, and cysteine reversibly forms - - S - - S - linkages with serum proteins, which can sometimes be r~luced to release
enough cysteine for cellular multiplication. If dialyzed serum is employed,
it is often convenient to prepare large standardized batches by dialysis
against distilled water, followed by lyophilization and storage as a dry
powder at - 20. 13
As has been described above, it is frequently possible to reduce the
amount of serum protein needed for cellular multiplication by employing
media and culture conditions that have been optimized for the particular
type of cell that is being grown. In some instances, still further reduction
of the amounts of undefined additives can be achieved by use of hormones 23 or partially purified growth factors ~9in place of dialyzed serum.
Whenever the experimental objectives are compatible with the use of
permanent cell lines, there are definite advantages, such as freedom from
interfering substances, in using a defined medium that does not require
protein supplementation. However, if such a system is employed, it is
important to avoid the introduction of contaminants from all other sources
including the cellular inoculum. Cells that have been previously grown as
stock monolayer or suspension cultures in a richly supplemented medium
are likely to introduce large amounts of undefined substances into the
"defined" experimental medium. Even if the cells are washed thoroughly,
such substances may be carried over intracellularly and perhaps even
released into the medium by death of a portion of the inoculum in amounts
sufficient to influence experimental results. One of the special advantages
of the clonal growth technique is that it minimizes the effects of such
carry-over by using only a very small inoculum (typically 100 cells in 5 ml
of medium) and by requiring that each cell undergo repeated cycles of
multiplication, which rapidly dilute out any intracellular carry-over
effects.
Finally, as alluded to earlier, serious questions remain about the purity
of "defined" media. In cases where contaminants are a serious problem,
such as trace element research, it may prove necessary to prepare the
"defined" media from specially purified chemicals and water in a care-
84
BASIC METHODS
[S]
fully decontaminated laboratory to achieve the required freedom from

contamination. In such cases, it may also prove necessary to add extra
nutrients to achieve growth comparable to that in the less-pure medium.
Culture Media and Which One to Use
The research literature describes hundreds of different media for
growth of vertebrate cells in culture. Not all are widely used, but there is a
sufficient number to cause confusion to the newcomer to cell culture.
About a dozen a~e so generally used that they are available from virtually
every commercial supplier, and many others are available from some of
the larger companies. In addition, there are a number of newer and less
well known media that are not currently available commercially, but
which may be very useful in certain types of research. All of the most
widely used media plus a number of the less common ones that are of
potential interest are listed in Table I. The chemical composition of a
number of these media is summarized in Table II. [See pp. 56-71 for
Tables I and II.]
Brief History of Commonly Used Media

The development of modern culture media can be divided into three
somewhat overlapping eras, which are reflected in the organization of
Table I.
1. The first era (part A of Table I) occurred prior to the general
availability of permanent cell lines, and employed cells that had only been
in culture a short time to measure growth (or survival) responses. Of this
group the only medium still used to a significant extent is medium 199 of
Morgan, Morton and Parker, which was originally developed for prolonged survival of chicken embryo heart fibroblasts without serum. When
supplemented with serum, it supports quite good growth of many types of
cells, and despite its age, it remains the medium of choice in some cases.
2. The second era, which began in the early 1950s (and is still continuing to some extent), was based on the use of permanent cell lines to
measure growth responses (parts B, C, D, E, and F of Table I). Two lines,
mouse L and HeLa, dominated the rapid progress in medium design made
in the late 1950s. With only a few exceptions, the media that currently
enjoy the most general usage were either developed through use of mouse
L or H e L a to measure growth responses, or else through use of other
permanent lines that had first been adapted to grow in media originally
developed for mouse L or HeLa. Media from that era include formulas
such as the widely used Eagle's basal and minimum essential media that
[5]
85
were designed for use with dialyzed serum, as well as a number of media
that support growth of permanent lines without serum supplementation.
The latter have been reviewed in detail by Higuchi 24 and by Katsuta and
Takaoka. z~
3. The third era (part G of Table I) has been under way for many
years, but is just now beginning to pick up momentum. It represents a
return to cells that are far more " n o r m a l " in their properties for the
measurement of growth responses. The classical permanent lines generally underwent extensive nutritional adaptation during the process of becoming established in culture, and media developed to satisfy their needs
generally do not support growth of normal cells without heavy serum
supplementation. Current research on cell growth requirements is focusing increasingly on normal cells and on nontransformed lines whose
growth requirements are more like those of normal cells. At the present
time, the growth requirements of normal cells are not yet fully understood, and the limited number of improved media for such cells that have
been described in the literature are just beginning to become available
commercially.
Media Designed for Use with Serum or Serum Protein

It is difficult to organize a discussion of media and their applications,
since many are currently used for purposes very different from the one for
which they were originally developed, and since certain popular media
are used for widely divergent purposes. The organization, of Table I is
based on the original purpose for which each medium was developed, and
an attempt has been made to indicate other uses in the " c o m m e n t s "
column. The discussion that follows is somewhat loosely organized, but
progresses in general from serum-supplemented media to "defined"
media, and within each type of medium, from transformed permanent
lines to normal cells.
Eagle's Media. The most widely used of all cell culture media, Eagle' s
minimum essential medium (MEM), was designed specifically for
monolayer growth of mouse L and H e L a cells with modest amounts of
dialyzed serum, as was its predecessor, Eagle's basal medium (BME). z6
Both of these media are simple and easy to prepare (Table II), and both
support good monolayer growth of many different types of cells when
~6There are actually several "basal" media described by Eagle that differfrom one another
in minor details. The Tissue Culture Association recommends use of the term "Eagle's
Basal Medium" only to describe the formulafor HeLa cells given in H. Eagle, Science
122, 501 (1955). That formula has been summarized in Table II.
86
BASIC M E T H O D S
[5]
supplemented with whole or dialyzed serum. MEM is the result of careful

quantitative balancing and can be used with an optional supplement of
nonessential amino acids and pyruvate for clonal growth. 27 Modified versions for suspension culture are also available from commercial suppliers.
Dulbecco's Medium. There are innumerable modifications of Eagle's
media. Perhaps the best known is that of Dulbecco. This medium has
evolved over the years, and there is a good deal of confusion about various versions of it. In an attempt to clarify the situation, the Tissue Culture
Association has designated the formula summarized in Table II as the
official version, z8 and has requested that all others be designated
"modified." DME contains 4 times the amino acids and vitamins of one of
the early versions of BME, and it also has a few added components (Table
II). The official version contains 1 mg/ml of glucose, but for some types of
cells, a variant with 4.5 mg/ml of glucose is superior. Dulbecco's modified
Eagle's medium should be abbreviated " D M E " and not " D M E M , " since
the latter leads to confusion with Eagle's MEM; DME was actually
developed from BME rather than from MEM. DME is widely used with
whole or dialyzed serum for nontransformed cultures, particularly of
mouse and chicken cells.
CIonal Growth. A number of media were developed specifically for
clonal growth of transformed cells, including McCoy's medium 5A for
Walker carcinosarcoma 256 cells, and Ham's FI0 and F12 for Chinese
hamster lines. All three media have been used widely for primary cultures
of differentiated cells from various species including rats, rabbits, and
chickens. In many cases, the concentrations of amino acids and pyruvate
in F10 and F12 are doubled, as first proposed by Hayden Coon. Medium
F12 was originally developed for protein-free growth, but it no longer
works well without serum protein unless selenium is added (see below).
Fischer's Medium. References to " F i s c h e r ' s " medium are potentially
ambiguous. Albert Fischer was one of the pioneers in medium development (Table IA), but his media are now primarily of historical interest.
More recently, Glenn Fischer and co-workers developed a medium for
growth of mouse leukemic cells in suspension culture that is rather widely
used (No. 10, Table IB). That medium is currently offered by several
commercial suppliers simply as "Fischer's medium."
Media Used without Carbon Dioxide. Several media have been
developed with special buffering for use without carbon dioxide gassing.
2 H. Eagle, Science 130, 432 (1959).
28 H. Morton, In Vitro 6, 89 (1970). This article, available as a monograph from the Tissue
Culture Association, contains r e c o m m e n d e d formulations for a variety o f commercially
prepared culture media.
[S]
87
The most widely used of these is medium L-15 of Leibovitz, which substitutes galactose for glucose and contains high concentrations of the basic
amino acids. This formula has been used with serum for a variety of cells,
both permanent lines and diploid primary cultures. A second such
medium is the CMRL 1415-ATM modification of CMRL 1415, which was
designed for primary cultures of mouse embryo cells. It is also useful for a
variety of other types of cells, and growth can sometimes be improved by
reducing its unusually high concentration of cysteine.
Media for Normal Cells. As described above and in Table I, many of
the media developed originally for established lines are widely used with
serum for growth of normal diploid cells and permanent lines of nontransformed cells. In addition, there are a number of media that have been
developed specifically for such cells (parts A and G of Table I). With the
exception of Medium 199 and similar media from the era of development
that preceded the availability of permanent cell lines, all of these formulas
are modifications of previous media for permanent lines. However, some,
such as the MCDB series, represent complete reworkings of the formulas
to satisfy the needs of the nontransformed cells.
A number of the formulas listed in Table IG were developed primarily
for survival and maintenance of function of normal cells without serum
supplements, and will be discussed below under the defined medium heading. Dulbecco's DME and Richter's IMEM-ZO are both modifications of
original formulations by Eagle. As already discussed, DME is one of the
most widely used media for primary cultures of mouse and chicken cells.
IMEM-ZO has been reported to support normal rat and human cell
growth, including clonal growth with relatively low concentrations of
serum proteins. CMRL 1415 is used with serum for growth of mouse cells,
and CMRL 1969 for human and monkey cells. F12K is a further modification of Coon's modification of F12, and it has been reported to work well
for a number of types of differentiated rat cells.
The MCDB series of media is the result of a process of qualitative and
quantitative optimization for clonal growth with minimal amounts of
serum protein. In each case, they will support clonal growth of diploid
fibroblast-like cells from the species they were designed for (MCDB 104,
105, human; MCDB 202, chicken; MCDB 401, mouse; MCDB 501, duck)
with far less dialyzed serum than conventional media. The composition of
MCDB 202 is such that it will also work reasonably well with cells from a
variety of other species. These media are specifically recommended for
clonal growth of their respective species, although they will not necessarily work for all types of cells from a particular species. They also can be
used for monolayer growth with modest amounts of dialyzed serum. At
the time of this writing, only MCDB 104 is available commercially.
88
BASIC METHODS
[5]
Media Used with Purified Proteins. Table IC lists a number of cases in

which proteins of varying purity have been used in place of serum. These
range from rather crude fetuin and serum albumin used in the development of medium F10 to mixtures of hormones and hormone-like factors
used by Gordon Sato and his co-workers to replace serum protein, z3
Recent media from Waymouth's laboratory for differentiated cells (Table
IG) are also supplemented with hormones. Several protein-supplemented
"serumless" media are listed in Table IC, including formulas by
Neumann and Tytell, by Jenkin and co-workers, by Higuchi, and by
Yamane and co-workers. These appear to support good growth of a
number of types of cells and may be of value when whole serum cannot be
employed. However, none have gained widespread usage, and they are
either not available commercially or else are listed only by a few
companies.
"Defined" Media
A number of media have been developed that support growth of permanent lines without the use of serum or other undefined supplements.
The probable role of contaminants in such growth makes it questionable
to call those media "defined," but in the absence of a suitable alternative,
we will employ that term with implied reservations. The following sections consider in order defined media for monolayer growth, suspension
culture, and clonal growth of permanent lines, and then maintenance
without growth of normal cells.
Defined Media for Monolayer Cultures. During the late 1950s and early
1960s several series of defined media for monolayer cultures of mouse L
cells were developed (Table ID). Media NCTC 135, CMRL 1066, and MB
752/1 are all widely used and readily available from commercial suppliers.
As can be seen from Table II, Waymouth' s MB 752/1 is somewhat simpler
than the other two, and on that basis it is probably preferable for use in
situations where all three support growth equally well. Newer formulations from Waymouth's laboratory with trace element supplements (MD
705/1 and MAB 87/3) are available from some suppliers. All have been
used with serum for growth of a variety of types of cells that do not grow
in them in the absence of serum.
A fourth series of defined media has been used for many years by
Katsuta and Takaoka and their co-workers in Japan, although it has never
achieved generalized usage in other parts of the world. DM 120, DM 145,
and the more recent DM 160 are all simple and seemingly reliable formulas
that appear to deserve more widespread usage than they have received.
There are relatively few reports of the growth of differentiated tumor
[5]
89
lines in defined media. One such report is growth of the C6 glial tumor line
by Donta in medium 924 MAP/l, developed originally by Waymouth. A
second is the growth of monolayers of C1300 mouse neuroblastoma cells
in medium MCDB 411, which has been optimized qualitatively and quantitatively for growth of the C 1300 neuroblastoma, and will support multiplication without delay or apparent adaptation of cells previously grown
exclusively in serum-containing media. It can also be used without supplementation to initiate cultures directly from transplantable C1300
tumors that have not previously been grown in culture, again without
delay or the need for adaptation. Another set of media developed in the
same manner are F12 and its more recent replacement, MCDB 301, which
are discussed below under clonal growth, although they also support
monolayer growth of Chinese hamster ovary lines. F12 and MCDB 301
have an unusually high concentration of zinc, which should be reduced if
they are used for other types of cells, such as mouse L. 21
Defined Media for Suspension Culture. A number of completely synthetic media for suspension culture have been described in the literature,
including several designed specifically for high cellular yield, but none
have become widely used. Suspension cultures in defined media require
very close attention to details, and it appears that in most cases, the
addition of serum does not interfere sufficiently with experimental objectives to justify the use of defined media. Media that are available for
defined suspension cultures include minor modifications of NCTC 109 and
MB 752/1, and various special-purpose formulas (Table IE). Birch and Pirt
adjusted concentrations of nutrients for efficient utilization and high yields
in suspension cultures of LS cells. Ling et al. used unusually high concentrations of nutrients to develop a medium (7C's) that would yield a maximum number of cells. Nagle and Brown developed a suspension culture
medium that is free of glutamine, autoclavable, and effective for cells from
many different species. However, none are widely used or generally available commercially. RPMI 1640 and several other formulas developed by
Moore and his co-workers for lymphoblastic cells in suspension culture
are commercially available. They are usually used with serum, but will
also support growth of some lines without supplementation.
Defined Media for Clonal Growth. Clonal growth in defined media has
not been widely pursued, but there are a few cases in which it is possible.
Medium F12 was developed for clonal growth of Chinese hamster ovary
and lung lines without protein supplementation, but later failed to support
such growth. Recent studies have shown that the main problem was a
requirement for selenium, which was apparently provided as a contaminant in the original studies.I The current medium of choice for such growth
is MCDB 301, which corrects for the selenium deficiency and also incor-
90
BASIC METHODS
[S]
porates other improvements. However, it is not yet commercially available. Commercially prepared F12 can be used with a selenium supplement, but in many cases it will also be necessary to add linoleic acid,
which is extremely labile to oxidation and is often not present in an active
form in commercial F12, including dry powder formulations.
Defined Media for Normal Cells. A number of defined media have been
developed specifically for maintenance of normal cells. Although little or
no growth occurs without serum supplementation, the cells remain fully
viable, and in some cases they continue along developmental pathways
during prolonged periods in culture. When supplemented with serum,
these media generally support excellent growth of normal cells. The original assay~used for development of medium 199 was long-term survival of
chicken embryo heart fibroblasts in the absence of serum supplementation. When supplemented with serum, medium 199 supports good growth
of many types of cells, and it is still widely used. In our own laboratory,
we have recently found it to be the best starting point among commercially available media for the development of a medium that is specifically
optimized for human epidermal keratinocytes.2a
Medium CMRL 1415 was developed for survival of mouse embryo
fibroblasts and for their growth when supplemented with partially purified
serum fractions. Our experience suggests that clonal growth of mouse
embryo fibroblasts with dialyzed serum can be improved by omitting the
very high level of cysteine in CMRL 1415 and using only the lower level of
cystine,a However, even better growth can be obtained with medium
MCDB 401 plus dialyzed serum. Medium MCDB 401 and most of the other
MCDB media, which were developed for growth of normal or nontransformed cells with small amounts of dialyzed serum, will also support
prolonged survival of their respective cells, even at clonal density, without
protein supplementation.1
Several special-purpose defined media for differentiated cells have appeared in the recent literature, including media E and G of Williams for rat
liver cells; the very complex defined medium of Parsa for maintenance
and differentiation of pancreatic rudiments; medium SM-20 of Halle and
Wollenberger for beating heart cells; and modifications of Waymouth's
media for liver and pancreatic cells (Table IG).
Making the Choice among Media

The final choice of a medium for a specific series of experiments is
usually based on two pragmatic considerations: (1) Will it yield the data that
29 D. M. Peehl and R. G. Ham, Abstracts 1978 TCA Meeting, In Vitro 14, 352 (1978).
a0 R. G. Ham and P. M. Sullivan, in "Cell Culture and its Application" (R. T. Acton and
J. D. Lynn, eds.), p. 533. Academic Press, New York, 1977.
[5]
91
are needed? (2) Is its use feasible on the scale that is needed? The former
is determined by trying the medium, often in competition with other
media, and the latter by practical considerations such as how hard the
medium is to prepare and whether or not it is commercially available.
For laboratories not equipped to prepare their own media, lack of
commercial availability can be a serious obstacle to trying new or unusual
media. It is possible to obtain custom-made batches of media, but the
companies generally require a rather large minimum order to recover their
expenses, and they do not have testing programs for determining whether
the batch that they prepare actually measures up to the claims made about
it in the literature. (Lack of adequate testing can also be a problem with
standard media. Often media are tested only for sterility and ability to
support monolayer growth with whole serum. Special-purpose media,
such as those designed for clonal growth with dialyzed serum, are frequently not tested to see if they will actually do what they are supposed
to. If problems arise, it is desirable to find out from the supplier exactly
what testing has been done.)
For the novice, it is generally desirable to begin with media and culture
conditions that previously have been reported to support growth of the
cell in question. Cell culture is a complex art with many different things
that can go wrong. It is therefore highly desirable to begin with conditions
that are known to work for the cell, and to wait until the initial problems of
getting it to grow have been resolved before attempting innovations. In
many instances, if the original conditions work well and are compatible
with the planned experiment, no further experimentation with media may
be necessary.
If previously reported culture systems are not adequate, or if the cell
type has not previously been cultured from the species that is to be used,
it will be necessary to begin a program of systematic testing of media.
Generally it is desirable to begin by attempting to obtain good growth with
whole serum, and then to proceed as far as needed toward chemical
definition. A generalized outline for that process is given in Table III, and
the details are presented in several other papers. 1,z,9"2
Species is a major factor in determining which medium to use, but
tissue specificity can also have a major effect. Media that tend to have
rather broad species applicability include 199, MEM, DME, F12K, and
MCDB 202. For human and monkey cells, 199, BME, MEM, L15, 5a,
RPMI 1640, CMRL 1969, MCDB 104, and MCDB 202 can be tried. For rat
and rabbit, MEM, 5a, F12, F12K, and MCDB 104 may work. For nontransformed mouse cells, DME, CMRL 1415, MCDB 202, MCDB 401, and
some of the newer Waymouth formulations can be tried. For chicken
cells, 199, DME, F12K, and MCDB 202 are possibilities. Media for certain
invertebrate cells are discussed in this volume [39]. Other formulations
92
BASIC METHODS
[S]
TABLE III
GENERALIZED SEQUENCE OF STEPS FOR MINIMIZING SERUM REQUIREMENTS AND FOR
DEVELOPING DEFINED MEDIA a
1. Obtain growth in vitro of the type of cell of interest by any means necessary.
2. Modify media, supplements, and culture techniques as needed to obtain clonal growth.
3. Survey readily available media and supplements and select the combination that supports the best clonal growth.
4. Replace all undefined supplements, e.g., serum, with dialyzed supplements plus ultrafiltrates, hydrolysates, etc., if they are needed.
5. Identify all low-molecular-weight nutrients needed for clonal growth in thepresence of
dialyzed supplements and add them to the synthetic portion of the medium.
6. Reduce the amount of the dialyzed supplement(s) to a level that supports less than
optimum growth.
7. a. Adjust all components of the synthetic portion of the medium sequentially to experimentally determined optimum concentrations for clonal growth with minimal
amounts of serum protein, beginning with the one found to be most limiting in a
preliminary survey.
b. Systematically test low-molecular-weight compounds considered likely to have
growth-promoting activity, e.g., trace elements and nutrients known to be required
by other types of cells or organisms.
c. Test complex low-molecular-weight mixtures such as hydrolysates, extracts, and
ultrafiltrates for growth-promoting activity.
d. Attempt to separate low-molecular-weight growth-promoting activity from the
dialyzed supplements through the use of dissociating conditions. If activity is detected in steps 7c or 7d, return to step 5.
e. Attempt to improve growth through refinement of the culture techniques (e.g., use of
treated dishes, trypsinization at low temperature, etc.).
8. Whenever growth is improved significantly, reduce the concentration of the dialyzed
supplement until it is again growth-limiting, and repeat all parts of step 7.
9. Keep repeating steps 7 and 8 until no further benefits can be obtained.
10. Isolate and characterize all macromolecular growth-promoting activities that continue to
be required under the "optimum" conditions developed in step 9.
11. Identify all "contaminants" both in the synthetic medium and in the "purified" macromolecular factors, and if any affect growth, add them to the synthetic medium at their
experimentally determined optimum concentrations.
12. Verify experimentally that each component of the "final" medium is at its optimum
concentration, and make any changes that may be needed as a result of adjustments
made in steps l0 and 11.
After all requirements for multiplication of isolated cells have been identified and optimized, it may be necessary to make additional adjustments for monolayer growth,
e.g., by adjusting nutrients to the upper end of their optimum plateaus to avoid depletion and adding factors needed to reduce the effects of density-dependent inhibition.
may work in many instances. In cases of extreme difficulty, it is probably

desirable to test virtually every medium that can be obtained easily. In
addition, it is often helpful to test combinations of media. For example, a
mixture of 199 and F12 is superior to either alone for clonal growth of
[5]
93
chicken embryo fibroblasts, and that combination formed the starting

point for the development of MCDB 202.
Future Prospects
Looking ahead, the prospects are that overall progress in the understanding of cellular growth requirements and the development of new
media may be rather slow. We are currently at the brink of a complete
understanding of the growth requirements of a very limited number of
normal diploid cells. However, the number of laboratories actively enganged in these studies is sufficiently small so that, at our present pace,
many years will be required to gain a broad general understanding of
cellular growth requirements.
Hopefully, an expanding awareness of how much we still need to learn
about the growth requirements of normal cells, together with a growing
realization of the urgency for gaining such knowledge, will provide the
stimulus to draw many more qualified investigators into the field and
accelerate our progress. However, until then, cell culture will remain in
many respects an art rather than an exact science. Any investigator who is
unwilling to work with an undefined system, or with one of the very
limited number of cell types whose requirements are beginning to be
understood, will find it necessary to do pioneering work on the growth
requirements of his cell before he can undertake the study that he ultimately plans to do with that cell.
Acknowledgments
Special thanks are due to Susan Jenningsfor assemblingthe data for the tables and to
Karen Brownfor typingthe manuscriptand tables; also to Dr. RaymondKahnfor use of a
computer printout that provided much of the raw data for Table II. Developmentof the
MCDB series of mediawas supportedby contract223-74-1156fromthe Bureauof Biologics,
U.S. Foodand Drug Administration,and by grants AG00310and CA15305fromthe National
Institutes of Health.
94
BASIC METHODS
[6]
[6] T h e G r o w t h o f C e l l s in S e r u m - F r e e
Hormone-Supplemented Media
B y JANE BOTTENSTEIN, IZUMI HAYASHI, SHARON HUTCHINGS, HIDEO
MASUI, JENNIE MATHER, DON n . MCCLURE, SUGAYUKI OHASA, ANGIE

RIZZINO, GORDON SATO, GINETTE SERRERO, RICHARD WOLFE, a n d
REEN WU
Introduction (G. Sato)
A series of experiments from our laboratory indicated that hormonal
depletion of the serum component of the medium is necessary to demonstrate hormone-dependent growth of cells in culture.l-3 From these observations, it was proposed that the main role of serum in cell culture is to
provide complexes of hormones and that it should be possible to substitute hormones for serum in cell culture media. 4 Experimental support for
this hypothesis was first provided in the case of GH3, HeLa, and B H K
cells and extended to other cell lines. 5-7 In this article we present the
practical details for growing a number of cell lines in serum-free
hormone-supplemented medium.
In the absence of serum, greater than usual care must be taken in
preparation of the synthetic portion of the medium. The water used in
preparation of our medium is distilled in glass stills in three steps. In the
first, deionized water is distilled and collected. It is placed in the boiler of
a second still to which is added approximately 10 g of potassium permanganate and 1 ml of concentrated sulfuric acid per liter. The distillate from
the second still is freshly distilled from a third still just prior to the preparation of the medium. Careful preparation of water is essential for consistent results with srum-free medium.
Serum and even dialyzed serum can mask nutritional requirements of
1j. L. Clark, K. L. Jones, D. Gospodarowicz, and G. H. Sato, Nature (London), New Biol.
236, 180 (1972).
2 H. A. Armelin, K. Nishikawa, and G. H. Sato, in "Control of Proliferation and Animal
Cells" (B. Clarkson and R. Baserga, eds.), Cold Spring Harbor Lab., Cold Spring Harbor,
New York, 1974.
3 K. Nishikawa, H. A. Armelin, and G. Sato,Proc. Natl. Acad. Sci. U.S.A. 72, 483 (1975).
4G. H. Sato, Biochem. Actions Horm. 3, 391 (1975).
5 I. Hayashi and G. Sato, Nature (I~ondon) 259, 132 (1976).
6 j. Mather, R. Wu, and G. Sato, in "Growth Requirements of Vertebrate Cells in Vitro"
(R. G. Ham and C. Waymouth, eds.). Cambridge Univ. Press, London and New York. (in
press).
r S. E. Hutchings and G. H. Sato, Proc. Natl. Acad. Sci. U.S.A. 75, 901 (1978).
METHODS IN ENZYMOI.X)GY, VOL. LVIII

ISBN 0-12-181958-2
[6]
SERUM-FREE HORMONE-SUPPLEMENTED MEDIA
95
cells in culture. For this reason we use one of the more complex media,
Ham's F12, s as the basic synthetic medium. We also supplement the
medium with the trace elements recommended by Ham. a Ham's F12 medium, supplemented with trace elements, is sufficiently complex to satisfy
the nutritional requirements of cells in the absence of serum. However,
the concentration of the components is often not optimal. This is remedied
by using mixtures of F12 and DME. 1'11
Commercially available powdered media are adequate for serum-free
work although each batch must be checked for suitability.
Serum also serves the function of a trypsin inhibitor in conventional
tissue culture procedures. In the absence of serum, we use soybean
trypsin inhibitor to neutralize the trypsin after subculture. After trypsinization, the cells are suspended in serum-free F12 medium containing
soybean trypsin inhibitor (1 mg/ml), centrifuged, and resuspended in
serum-free F12 for plating.
Hormones, growth factors, and transferrin are maintained as 100
stocks and added to the medium just prior to plating the cells. Stock
cultures are usually kept in a 1:1 mixture of Ham's F12 and Dulbecco's
modified Eagle's medium (DME), supplemented with sodium bicarbonate
(1.2 g/liter), N-2-hydroxyethylpiperazine-n '-2-ethanesulfonic acid, pH 7.2
(HEPES) (15 mM), penicillin (192 units/ml), streptomycin (200 /zg/ml),
ampicillin (25 ttg/ml), 5% (v/v) horse serum, and 2.5% (v/v) fetal calf
serum. The cells are grown on plastic tissue culture dishes in a humidified
atmosphere of air containing 5% CO2.
In those cases in which subculture involves trypsinization, two techniques for growing cells in serum-free medium are used. The first involves
the transfer of cells by trypsinization from stock medium containing
serum, directly to serum-free medium. In the second procedure, cells are
first plated into serum-containing medium and, after a day, the medium is
replaced with serum-free medium after washing the cells with serum-free
medium. In those cases in which this procedure is followed, the initial
incubation in serum-containing medium may be essential. There are two
likely explanations for the requirement for serum pretreatment: (1) The
serum may be necessary for the repair of trypsinization damage after
subculture; or (2) the residual serum left after its removal, even after one
or more washes, may be furnishing unknown growth factors. The latter
possibility seems unlikely in view of the extensive growth which is possi8 R. G. Ham, Proc, Natl. Acad. Sci. U.S.A. 58, 288 (1965).
aW. L. McKeehan, W. G. Hamilton, and R. G. Ham, Proc. Natl. Acad. Sci. U.S.A. 72,
2023 (1976).
10 H. Eagle, Science 122, 501 (1955).
11 j. D. Smith, G. Freeman, M. Vogt, and R. Dulbecco, Virology 12, 185 (1960).
96
BASIC METHODS
[6]
ble after transfer to serum-free medium. In any case, we intend to settle

this question by growing these cells on beads 12 (see also this volume
[15]). This system of culture allows the subculture of cells without
trypsinization.
In the course of defining the requirements of Ce glial cells, an active
factor was found in the rat submaxillary salivary gland (see section on C6
cells). The crude extract of salivary gland can be maximally active at
concentrations as low as 2 ~g protein/ml of medium; the purification of
this material is still in progress in our laboratory. In order to avoid any
implication of its possible physiological role, we have named this factor
rat submaxillary gimmel factor or gimmel. Gimmel is of great importance
because it has been found to be required for the growth in serum-free
medium of Balb 3T3 and Swiss 3T3 cells (see sectior)s on Balb and Swiss
3T3 cells). In addition, it is possible that a factor other than MSA is in
conditioned Buffalo rat liver cultures and is responsible for the growth
stimulation of rabbit aortic intimal cells (see section on intimal cells). The
preliminary evidence supporting this notion is that the growth activity for
intimal cells and the MSA activity (stimulation of DNA synthesis in
serum-deprived 3T3 cells) do not copurify on Dowex columns. These two
examples illustrate how the serum-free approach readily lends itself to the
discovery of novel factors of possible physiological significance.
We have also included an example of hormone binding to cells grown
in serum-free, hormone-supplemented media (see section on [125I]EGF
binding to HeLa cells). The significance of this section is that the phenomenon of hormone internalization, degradation, and " d o w n " regulation of
receptors does not seem to occur in defined media. Since EGF
is required for growth of these cells, one can tentatively conclude that
internalization and " d o w n " regulation of receptors is not necessary
for the mitogenic activity. If this generalization holds up in future experimentation, it represents an unexpected technical advantage in doing
experiments in serum-free hormone-supplemented media.
The cell lines used in the study are rat neuroblastoma/3 GH3,'4 HeLa,,5
Balb 3T3,16 rat ovarian cells (RF),17 B 16 melanoma M2R,18 mouse testicu12 D. W. Levine, J. S. Wong, D. I. C. Wang, and W. G. Thilly, Somatic Cell Genet. 3, 149
(1977).
13 D. Schubert, S. Heinemann, W. Carlisle, H. Tarikas, B. Kimes, J. Patrick, J. Steinbach,
W. Culp, and B. Brandt, Nature (London) 249, 224 (1974).
1~ y. Yasumura, A. H. Tashjian, Jr., and G. H. Sato, Science 154, 1186 (1966).
15 G. O. Gey, W. D. Coffman, and M. T. Kubicek, Cancer Res. 12, 264 (1952).
16 S. A. Aaronson and G. J. Todaro, J. Cell. Physiol. 72, 141 (1968).
17 H. Masui, S. Ohasa, and S. Mackensen, in preparation.
is j. Mather and G. H. Sato, (S. Federoff, ed.). in "Cell Tissue and Organ Cultures in
Neurobiology," p. 619. Academic Press, New York, 1978.
[6]
97
lar cells TM4, TM rat glioma C6,2 mouse embryonal carcinoma, 21,z~ Swiss
3T3, 23 and a line of cells isolated from rabbit aortic intima. 24
Note
Bovine crystalline insulin, human transferrin, and steroid hormones

are purchased from either Sigma or Calbiochem. Epidermal growth factor
(EGF) and fibroblast growth factor (FGF) are purchased from Collaborative Research, Waltham, Massachusetts. A gift of a somatomedin preparation from Dr. Knut Uthne, Kabi Co., Stockholm, Sweden made it
possible to start this work. Gifts of EGF from Dr. Stanley Cohen, Vanderbilt University, and FGF from Dr. Denis Gospodarowicz, University
of California Medical School, San Francisco, were helpful in the early
stages of this work. Highly purified pituitary hormones and somatomedins
were generously provided by Dr. Harold Papkoff of the University of
California, San Francisco, and Dr. Judson Van Wyk, University of North
Carolina, Chapel Hill, respectively. Partially purified pituitary hormones
are obtained from the National Institutes of Health Hormone Distribution
Program . We are very grateful to Dr. Brian Kimes of the NCI. Without his
timely assistance, this work would not have been possible. This work was
supported by grants from the National Science Foundation, NIH GM
17019, NIH CA 19731, NIH GM 17702, and NCI CA 18885.
Neuroblastoma Ceils in Serum-Free Medium (J. Bottenstein)
Cell Culture in the Presence of Serum
The B104 rat neuroblastoma 13 stock cultures are maintained in a 1:1

mixture of Ham's F12 (F12) and Dulbecco's modified Eagle's medium
(DME) supplemented with 5% (v/v) horse serum, 2.5% (v/v) fetal calf
serum, NaHCO3 (1.2 g/liter), HEPES buffer (15 mM), penicillin (40 mg/
liter), ampicillin (8 mg/liter) and streptomycin (90 mg/liter). Stocks are
grown in 25 cm 2 tissue culture flasks at 37 in a humidified atmosphere
containing 5% CO2 and are subcultured every 4-6 days. After an initial
3-ml wash with PBS, 2 ml of 0.1% trypsin-0.9 mM EDTA in PBS for 3 min
19 j. Mather, unpublished results.
2o p. Benda, J. Lightbody, G. Sato, L. Levine, and W. Sweet, Science 161, 370 (1968).
zl H. Jakob, T. Boon, J. Galliard, J.-F. Nicolas, and F. Jacob, Ann. Microbiol. (Paris) 124b,
269 (1973).
22 K. Artzt, P. Dubois, D. Bennett, H. Condamine, C. Babinet, and F. Jacob, Proc. Natl.
Acad. Sci. U.S.A. 70, 2988 (1973).
23 H. Green and O. Kehinde, Cell 1, 113 (1974).
24 V. Buonassisi and L. Ozzello, Cancer Res. 33, 874 (1973).
98
BASIC M E T H O D S
[6]
TABLE I
EFFECTS OF GROWTH FACTORS ON BI04 CELLS a
Cell number 103
Addition
None
4F
4F + NasSeOa
1 0 % f e t a l calf serum
Day 3
179
493
573
1307
_+ 19
-+ 11
-+ 14
-+ 3 0
Day 5
78
24
1856
6395
Growth rate (hr)
-+ 18
--- 5
_ 97
--- 186
--28
21
a Data are expressed as the mean -+ SD of triplicate samples. Initial inoculum is 7 x 104.
at 37 is used to detach the cells. To inactivate trypsin, 2 ml o f medium

(vide supra) are added and the cell pellet obtained by centrifugation is
diluted with medium to the appropriate plating density, usually
7-14 104 cells/flask.
Cell Culture in the Absence of Serum

Cells are subcultured as above and plated in medium (vide supra) at a
density o f 7 104 cells/60-mm dish. After 18 hr this medium is removed,
cells are washed twice with 3 ml of serum-free medium, and 3 ml of
serum-free experimental medium are added. The four factors (4F) are:
bovine insulin (5 tzg/ml), human transferrin (5/zg/ml), putrescine dihydrochloride (100/zM), and progesterone (20 nM). The media are changed on
day 3 only. Table I shows that with 4F alone, cell number increases about
3-fold at day 3, but is markedly reduced by day 5. The addition of 30 n M
selenium (Na2SeO3) to 4F has a slight stimulatory effect at day 3 and
allows the cells to continue growing at the same rate. Thus, in a serumfree condition with 4F and Na2SeO2, the growth is 75% o f that when 10%
fetal calf serum is present.
G H a Cells in a Defined M e d i u m (I. Hayashi)
G H 3 is a clonal line derived from a spontaneous, transplantable rat
pituitary tumor.14 The cells produce growth hormone and prolactin and
respond to thyroid hormones for growth. 25 Stock cells are maintained in
D M E supplemented with 12.5% fetal calf serum. In our experience, it is
important that cells be kept in medium containing at least 10% serum to
s5 H. H. Samuels, J. S. Tsai, and R. Cintron, Science 181, 1253 (1973).
[6]
SERUM-FREE
HORMONE-SUPPLEMENTED
MEDIA
99
get good hormone response. The stocks are subcultured about once a
week.
For growing GHa cells in serum-free condition, exponentially growing
stock cells are trypsinized (0.1% trypsin in PBS) and then treated with
soybean trypsin inhibitor (0.1% of the inhibitor in PBS). The treated cells
are washed by centrifugation, resuspended in serum-free F12 medium,
and inoculated into hormone-containing, serum-free F12 medium at 0.5
105 (35-mm dishes) or 1.0 10~(60-mm dishes) per plate. Unlike the other
lines described here, we have found that GH3 cells prefer F12 medium to
any other media, or the combination of F12 and DME. These cells have
not responded to hormones under serum-free conditions unless F12 was
used. The hormones and their final concentration in the medium that are
required for growth of GH3 cells in the serum-free condition are as follows: 3,3',5-triiodothyronine (3 10-11 M), thyrotropin-releasing hormone (1 ng/ml), parathyroid hormone (0.5 ng/ml), transferrin (5 /zg/ml),
insulin (5 ttg/ml), fibroblast growth factor (1 ng/ml), and somatomedin C (1
ng/ml). 26 When these hormones are added to the serum-free F12 medium,
the cells not only grow as well as in serum-supplemented medium, but
also can be maintained for a long period by serial propagation. Requirement for parathyroid hormone, which has the least effect on short-term
growth of these cells, will become obvious in such long-term culture in
serum-free condition. For long-term growth in serum-free, hormonesupplemented medium, trace metals such as selenium and cadium are
added at the final concentration of 50nM and 0.5/JM, respectively. Some
of the other trace metals, especially manganese, are strongly inhibitory to
GH3 cells and, therefore, are not used for these cells.
One of the immediate interests in defining hormone requirements of a
cell line and growing it in a serum-free condition is the question of the
universality of the requirement. For instance, will the hormone combination that allows the growth of GH3 cells also allow the growth of normal
rat pituitary cells which produce growth hormone and/or prolactin? To
examine these possibilities, we made primary cultures from young normal
rat pituitaries. The pituitaries were washed several times in serum-free
medium, minced in the presence of trypsin inhibitor, and plated in serumfree F12 medium supplemented with the hormones mentioned above.
Comparison of such cultures with those in serum-supplemented (10% fetal
calf serum) F12 medium showed that there is at least a significant suppression of fibroblast growth in serum-free, hormone supplemented media. At
present, it is too early to determine whether the cells in serum-free,
hormone-supplemented medium are the normal equivalent of GH3 cells
which produce growth hormone and/or prolactin, but the results indicate
~ I. Hayashi, J. Larner, and G. Sato, In Vitro 14, 23 (1978).
100
BASIC METHODS
[6]
that, in the future, hormone-supplemented serum-free medium could be

used to select specific cell types from mixed populations of cells in primary cultures.
Single-CeU Plating of HeLa Cells in Serum-free, Hormone-Supplemented
Medium (S. Hutchings)
The cell strain, designated HeLa-S, was obtai0ed from Dr. John Holland of the University of California, San Diego, La Jolla, California.
Stock cultures are maintained in a l: 1 mixture of F12 and DME supplemented with serum as described in the introduction.
Stock cultures in the exponential phase of growth a r e trypsinized
(0.1% trypsin with 0.03% EDTA in PBS), suspended in serum-free F12,
SFF12, containing 0.1% soybean trypsin inhibitor, and centrifuged. The
cell pellet is resuspended in SFF12 and plated at a low density (about
1,000 cells) in 100 mm plastic dishes containing 10 ml of medium. The
medium is always warmed and equilibrated with the atmosphere in a COs
incubator before use for either the primary plating or medium changes.
The defined medium consists of F12 medium containing trace elements, 9
to which is added insulin (5 /zg/ml), transferrin (0.5 ~g/ml), EGF (50
ng/ml), FGF (100 ng/ml), and hydrocortisone (100 rnM). At day 5, the
medium is changed, and at day 9 the cells are fixed in methanol and
stained with 0.1% crystal violet. The plating efficiency in the defined
media is approximately 50% of that in serum-containing media. Omission
of any one of the above-mentioned components results in reduced plating
efficiency.
The most stringent test of the adequacy of a medium is its ability to
support colonial growth from single cells at low plating density. Serumfree, hormone-supplemented media support such growth.
Normal Rat Follicular Cells in Serum-Free Medium (H. Masui)
A follicular cell line (RF-1) has been established in culture from the
ovaries of normal Fisher rats. The medium used for establishing these
cells was a mixture of DME and F12 medium (1:1) supplemented with
2.5% fetal calf serum and 12.5% horse serum. Stock cultures of RF-1 cells
are maintained in the medium. RF-1 cells have characteristics of normal
cells, such as diploid chromosome number (2n I= 42), density-dependent
inhibition of growth, and no tumorigenicity in Fisher rats or nude mice.
RF-1 cells produce several steroids from exogenous cholesterol; the main
product seems to be testosterone, lr
When RF-1 cells are grown in a mixture of DME and F12 medium (1:1)
[6]
101
supplemented with various concentrations of fetal calf serum, the rate of

cell growth increases with the serum concentration and reaches a maximum between 3 and 6%. When either DME or F12 medium is used instead
of the mixture, cells do not grow even in the presence of serum.
To grow RF-1 in serum-free media, cells are trypsinized and plated in
the DME/F12 mixture supplemented with 2.5% fetal calf serum and 12.5%
horse serum. The cells are kept in this serum-containing medium for at
least 6 hr. Thereafter, they are washed with serum-free medium and the
medium is replaced with DME/F12 (1:1) supplemented with bovine insulin
(2/zg/ml), hydrocortisone (4 ng/ml), triiodothyronine (3 x 10-l M), and
transferrin (5/zg/ml). Usually 9 x 104 cells are plated per 60-ram dish and
increase to 3-4 x 104 cells after 5 days of growth. During the exponential
phase of growth, the doubling time is about 26 hr which is comparable to
the doubling time in serum-supplemented media.
Growth of Balb/c Mouse Fibroblasts in Serum-Free M e d i u m (D. McClure)
Serum-free medium supplemented with insulin, fibroblast growth factor, transferrin, and a crude extract prepared from female rat submaxillary gland, rat submaxillary gimmel factor, supports continuous division
cycles of Balb/c 3T3 mouse fibroblasts washed free of serum. Under these
serum-free conditions, the growth rate and saturation density are similar
to those observed for the same cells cultured in DME containing 10% fetal
calf serum. Morphologically, these cells appear more epithelioid than
their serum-grown counterparts.
Trypsinized stock cultures of logarithically growing Balb/c 3T3 cells,
maintained at an early passage number, are subcultured at 5 104 cells
per dish in DME plus 10% fetal calf serum and incubated ovemight at 37
in a humidified atmosphere containing 5% CO2. On day 1, after washing
each culture twice with serum-free DME, the medium is changed to
serum-free medium, consisting of three parts DME and one part F12, plus
trace elements. Hormone additions are made directly into each culture
dish from freshly prepared concentrated stock solutions. The final concentrations of insulin, transferrin, hydrocortisone, EGF, and NIH-LH ( a
source of FGF) are 2/xg/ml, 5/xg/ml, 10-7 M, 50 ng/ml, and 1 /xg/ml,
respectively. Rat submaxillary gimmel factor is added at a concentration
of 2 /zg/ml. The preparation of gimmel factor is described in the section
on C6 glial cells.
Balb/c 3T3 fibroblasts, washed free of serum and incubated in serumfree media supplemented with insulin, transferrin, hydrocortisone,
NIH-LH (FGF), EGF, and a crude preparation of rat submaxillary gland,
grow from a low cell density (3 x 103 cells/cm2) to form a confluent
102
aASIC METHODS
[6]
monolayer (7 x 104/cm2) with an approximate generation time of 20 hr.

Hormone-supplemented medium without the addition of rat submaxillary
gimmel factor is sufficient to support only two rounds of replication after
which the cells cease to divide at an appreciable rate. Addition of as little
as 2/~g/ml of crude gimmel factor is sufficient to support continuing cell
growth. By omitting each individual component from the above mixture of
hormones and growth factors it can be further shown that only insulin,
transferrin, NIH-LH, and rat submaxillary extract are necessary for
Balb/c 3T3 cells growth in the complete absence of serum.
Others have succeeded in growing Balb/c 3T3 cells, maintained in low
serum, by the addition of hormones and polypeptide growth factors to the
culture medium.27'2s Here, we demonstrate that Balb/c 3T3 fibroblasts
exhibit continuing growth in hormone-supplemented medium in the complete absence of serum. We suggest that rat submaxillary gimmel factor
replaces the residual serum requirement hitherto found necessary for the
growth of 3T3 cells in totally defined medium.
Melanoma Cells in Defined Medium (J. Mather)
A clonal line of melanoma, designated M2, was isolated from the B16
transplantable tumo~s obtained from the Jackson Laboratory, Bar Harbor, Maine. This line metastasizes in 50% of the animals in which cells
have been injected subcutaneously. A group of hormones has been shown
to replace the serum requirement for growth of these cells in culture. Cells
are grown in F12/DME (1:1) supplemented with sodium selenite (10-7M)
and the following hormones: insulin (5/zg/ml), transferrin (5/zg/ml), ovine
follicle-stimulating hormone (purified FSH) (40 /zg/ml), luteinizing
hormone-releasing hormone (LRH) (10 ng/ml), nerve growth factor (NGF)
(10 ng/ml), and either testosterone (5 nM) or progesterone (1 n M ) . 6'1s The
cells can be traffsferred by trypsinization directly into the serum-flee
medium.
Transferrin seems to be acting primarily, but not exclusively, as an
iron transport protein in culture since 75% of the growth stimulation seen
with transferrin can be obtained by increasing the ferrous iron concentration in the medium. However, the remainder of the stimulation cannot be
accounted for solely be increased iron transport. Insulin has been shown
to stimulate glucose incorporation into both glycogen and fatty acids in
these cells and appears to be acting in a manner consistent with our
understanding of insulin action in classic target tissues. Although the cells
27 D. Gospodarowicz and J. Moran, Proc. Natl. Acad. Sci. U.s.A. 71, 4584 (1974).
2s H. Armelin, Proc. Natl. Acad. Sci. U.S.A. 70, 2702 (1973).
[6]
103
are stimulated by either progesterone or testosterone, the increase in cell

number with testosterone is greater than that seen with equivalent
amounts of progesterone and the effects do not seem to be additive. When
radioactive hormone was given and the products analyzed by thin-layer
chromatography greater than 99% of the testosterone migrated as testosterone after 24 hr in culture. In a similar experiment three peaks were
obtained from [~H]progesterone, none of which comigrated with testosterone; the products have not been identified. The results indicate that
progesterone does not affect the cells by being metabolized to testosterone nor do the two hormones act via a common steroid metabolic
product.
The metastatic cells have been injected subcutaneously into C57 BL/6
mice. After 2 weeks, when no metastases are visible and only a small
tumor is formed, the mouse was sacrificed and the lungs, lymph nodes,
and spleen removed, minced, and plated into medium supplemented with
the hormones which will support the growth of melanoma cells. After 10
days, the only cells surviving were small colonies of melanoma cells,
presumably arising from microscopic metastatic foci in the organs plated.
Thus, the hormone-supplemented medium can be used to select for a
desired cell type in primary cultures.
Testicular Cells in Defined Medium (J. Mather)
An epithelial cell line, TM4) 9 was isolated from normal immature (llday) Balb/c mouse testes. This line has been cloned 3 times and has been
in culture for over 2 years. The doubling time is 16 hr. The cells do not
form tumors in Balb/c-nu/nu mice.
The cell line can be grown in F12/DME (l: 1) supplemented with a
mixture of hormones and growth factors. These cells are growth stimulated by insulin (5/zg/ml), transferrin (5 ~g/ml), epidermal growth factor
(3 ng/ml), FSH (0.5 ftg/ml), somatomedin C (1 ng/ml), ovine growth
hormone (6.5 pU/ml), and retinoic acid (50 ng/ml). Under hormonesupplemented conditions, growth rates are obtained equivalent to those
seen in the same medium supplemented with 5% fetal calf serum.
Primary cultures of Sertoli cells from immature testes can be grown in
the above hormone-supplemented medium. Survival times are increased
significantly over those of primary cultures plated in serum-free medium
(6 weeks as opposed to 5-10 days) with no fibroblast overgrowth.
Rat Glial C~ Cells in Serum-free Medium (S. Ohasa)
The maintenance of stock cultures is carried out as follows: rat glial C~
cells in stock culture bottles (100 20 mm) are incubated with 0.1%
104
BASIC M E T H O D S
[6]
trypsin solution (in PBS containing 0.9 mM EDTA) at room temperature

for 2 min and suspended in 10 ml of a mixture of DME/F12 (1:1) supplemented with 6% fetal calf serum. The medium is pipetted up and down
with a 10-ml sterile pipette in order to detach the cells from the culture
dish. Cells are diluted to the desired cell density and plated. Usually, 5
1@ cells per 100-mm plate are inoculated and cultured in DME/F12 with
6% fetal calf serum at 37 under a mixture of air containing 5% CO2. This
process is repeated every 5 days.
For growth in serum-free medium, the cells in stock cultures are detached by trypsinization, suspended in l0 ml of 1:1 DME/F12 with 6%
serum, and then inoculated into a 60-mm culture dish. After incubating the
cells in 5 ml of the serum-supplemented medium for 1 day the cells are
washed 3 times with 3 ml DME/F12 in order to remove the added serum
and then cultured in 5 ml of serum-free DME/F12 supplemented with
transferrin (5 /~g/ml), insulin (2 /~g/ml), NIH-LH (0.4 ~g/ml), and an
extract of rat submaxillary gland (5-10/~g/ml).
The activity of rat submaxillary gland extract was discovered in experiments with CO glial cells. The Co system is currently being used as the
assay in the purification of the active factor, gimmel factor. Gimmel factor
is prepared as follows: Submaxillary glands (1 g) from a young, male or
female Fisher rat are minced, suspended in 5 ml of PBS containing 0.2 M
sucrose, and homogenized with a glass/Teflon homogenizer. The homogenate is centrifuged at 27,000 g for 30 rain, and the clear supernatant solution is collected to give a submaxillary gland extract. The sample is diluted with PBS to yield a 1 mg/ml protein solution, sterilized by filtration,
and then used for cell culture.
T A B L E II
EFFECTS OF HORMONES ON THE C6 CELL GROWTH a
Addition
N o n e (control)
S e r u m (6%)
N I H - L H (0.4/zg/ml)
F G F (20 ng/ml)
Ins (2/xg/ml)
T r a n s (5/~g/ml)
G i m m e l (10 #,g/ml)
Ins + trans + FGF*
Ins + trans + FGF* + glmmel
Cell no. x 10 -5
4.7
48.2
10.3
8.6
12.2
8.4
15.0
36.8
53 8
-+ 0.9
_+ 0.5
+ 0.1
0.4
-+ 0.7
_+ 0.1
-- 0.6
-+ 0.3
0.7
Relative cell no.

1.0
10.3
2.2
1.8
2.6
1.8
3.2
7.8
11.4
a (26 cells (8 x 104) were plated into a 6 0 - m m culture dish, incubated in 5 ml D M E / F 1 2

with 6% FCS, w a s h e d 3 times with 3 ml DME/F12, and then cultured for 4 days in
5 ml D M E / F 1 2 with the indicated factors. T h e a d d e d factors were s e r u m (FCS), F G F
or N I H - L H (indicated as FGF*), Ins (insulin), Trans (transferrin), a n d gimmel (submaxillary gland extract) w h i c h was prepared from F i s h e r rat submaxillary gland. T h e
indicated cell n u m b e r per 6 0 - m m culture dish is an average o f 2 determinations
standard error.
[6]
SERUM-FREE
HORMONE-SUPPLEMENTED
, MEDIA
105
C6 cell growth is stimulated by insulin, FGF or NIH-LH, transferrin,

and rat submaxillary gimmel factor when they are added individually. The
growth rate is the greatest when all four factors are present resulting in an
approximate doubling time of 15 hr. Rat submaxillary gimmel factor cannot be replaced by purified EGF or N G F which are present in the submaxillary gland, suggesting that gimmel factor is distinct from them.
Embryonal Carcinoma Cells in Serum-Free Medium (A. Rizzino)
Two embryonal carcinoma cell lines, PCC.4 aza-1 '1 and F9, 22 have
been cultured in the absence of serum for over 70 and 20 generations, respectively.29 The cultures were kept at 37 under a mixture of air containing 5% COs. The cells are subcultured (when grown in serum or in serumfree media) by removing the spent medium and replacing it with fresh
serum-free medium (see below). This is repeated once more for the purpose of washing the cells. The medium is then pipetted up and down over
the cells with a sterile Pasteur pipette in order to detach the cells from the
culture dish. The cells are diluted to the desired cell density and plated.
Usually 3-6 x 10 ~ cells are plated per 25 cm 2 " T " flask. In the case of F9
cells, trypsin (0.1%)-EDTA (0.9 raM) in PBS can also be used to detach
the cells from the culture dish; this is unnecessary unless a single cell
suspension is desired. When trypsin-EDTA is used, 3 ml are added to a 25
cm 2 " T " flask long enough to detach the cells after which an additional 3
ml of serum-free medium are added. The cells are collected by centrifugation at room temperature; trypsin inhibitor is unnecessary. Once the cells
reach high density, about 2 l0 s cells per flask, they may be subcultured.
The serum-free medium used for embryonal carcinoma cells is refer~ed
to as EM-1. This medium is composed of F12 supplemented with fetuin
(500/zg/ml, Calbiochem-Pedersen type), bovine insulin (10/zg/ml), human
transferrin (5/zg/ml), and 10/zM/3-mercaptoethanol. For optimal growth
the medium should be changed daily; this is more important for PCC.4
aza-1 than for F9 cells. The medium (EM-1) is prepared every day from the
appropriate stock solutions. DME cannot be used to replace F12.
Mouse Swiss 3T3 Fibroblasts, Clone 3T3-L1 in a Serum-Free Medium
(G. Serrero)
Clone 3T3-L1 of mouse Swiss 3T3 fibroblasts, isolated in Dr. Howard
Green's laboratory, has the property of differentiating into adipocytes
when the cultures are confluent. Attempts have been made to grow these
cells in a serum-deprived medium supplemented with hormones and other
growth factors in order to study differentiation into adipocytes.
The cells are routinely cultured in DME supplemented with 10~ fetal
calf serum. Cells are plated at a density of 2000 cells/cm~ and grown to
~9 A. Rizzino and G. Sato, Proc. Natl. Acad. Sci. U.S.A., 75, 1844 (1978).
106
BASIC METHODS
[6]
conflttency. The medium is changed every other day with cultures maintained at 37 in a humid atmosphere containing 5% CO2. The cells can be
grown in a serum-free medium consisting of 75% DME and 25% F12
enriched with a mixture of trace elements; 9 insulin (0.5 /Lg/ml); NIHLH-B9, a biological standard of bovine luteinizing hormone (2/~g/ml); and
rat submaxillary gimmel factor prepared as described in the section on
C6 cells.
In this medium, the cells have the same morphology as those cultured
in the presence of 10% fetal calf serum. In serum-rich medium, they divide
with a generation time of 20 hr and reach a saturation density of 75 x 103
cells/cm2; in hormone-supplemented medium, their generation time is 28
hr and they reach a saturation density of approximately 36 10 a cells/cm ~.
The efficiency of this serum-free medium for supporting the growth and
viability of 3T3-L1 depends on the presence of gimmel; its omission results
in an 80% inhibition of growth.
The nature of the factor responsible for the multiplication of 3T3 is
under study. Experiments show that purified EGF is unable to replace
gimmel in supporting the growth of 3T3. The omission of NIH-LH-B9,
insulin, or transferrin results in growth inhibition of 26%, 40%, and 72%,
respectively. In the work described above, the cells were plated and
incubated with DME/10% fetal calf serum for 15 hr after trypsinization
before being cultured in serum-free medium. This "serum-incubation"
appears to be necessary for the cells to recover from the trypsinization
and to spread on the dishes, a process that seems important in the response of the cells to the hormones and growth factors. Attempts to
eliminate the step of serum incubation are in progress. Addition of serum
fractions such as fetuin, and Cohn fractions I and IV, seem to improve
greatly the plating efficiency and the spreading of 3T3-L1 in a serum-free
medium just after trypsinization.
Another clone of mouse Swiss 3T3, 3T3-C2, that does not differentiate
into adipocytes also can be grown in the absence of serum in a medium
similar to that established for 3T3-L1.
Rabbit Intimal Cell Line in a Serum-Free Medium (G. Serrero)
The cell line discussed in this section was established by Dr. Vincenzo
Buonassisi from cells isolated by trypsinization of the subendothelial portion of the intima of rabbit aorta. Cells are routinely cultured in F12
medium supplemented with 10% fetal calf serum and kept at 37 in a
humidified atmosphere containing 5% COs. The cells have a fibroblast-like
appearance in a subconfluent stage and become enlarged and oriented in a
parallel fashion in the culture dish when they reach confluency. The cells
grow in a serum-free medium that consists of F12 enriched with trace
[6]
SERUM-FREE HORMONE-SUPPLEMENTED
MEDIA
107
elements, 9 insulin (10/zg/ml), transferrin (5/~g/ml), NIH-LH-B9 (2/zg/ml),

progesterone (1 nM), growth hormone (50 ng/ml), and 5% of a serum-free
medium conditioned by growing cells. The medium is conditioned by
Buffalo rat liver cells (BRL clone 3A, kindly supplied by Dr. Peter
Nissley). BRL cells produce and release into their culture medium a
growth factor known as "multiplication stimulating activity" (MSA) 3-32
that promotes DNA synthesis in chick embryo fibroblasts and 3T3 cells.
However, preliminary experiments suggest that the factor stimulating the
multiplication of the intimal cell line in serum-free medium is not MSA.
Characterization of the responsible factor is in progress.
In the medium described above, the intimal cells have similar doubling
times (24 hr) and the same morphology (the cells become elongated and
oriented when they reach confluency) as do cells cultivated in the presence of 10% fetal calf serum. The cells have been grown continuously and
transferred in the defined medium. They have been kept for 2 months in
the absence of serum and have been passed 10 times in the synthetic
medium without loss of their growth properties.
Subconfluent or confluent cells are washed flee of medium with PBS
and detached from the plates with 1 ml of trypsin (1 mg/ml) per 100-mm
dish. The excess is immediately removed by aspiration so that only a thin
film remains on the plate. After 2 min at 37, the reaction is stopped by
adding 1 ml soybean trypsin inhibitor (1 mg/ml). Cells are then detached
and washed free of trypsin and trypsin inhibitor by suspension into
serum-free F12. Cells are inoculated at a density of 2 x 103/cm 2. Incubation with serum is not necessary for spreading the cells on the dishes and
the start of optimal growth.
In the absence of BRL conditioned medium, the cells cannot be transferred continuously in the defined medium; their growth drops after the
second subculture. If insulin and transferrin are also omitted, growth is
reduced by 70-90% and there are drastic modifications of cell
morphology.
[125I]-Labeled EGF Binding to HeLa-S in a Defined System (R. A. Wolfe
and R. Wu)
Many cell lines exhibit marked growth responses to EGF. 3~'34 Furthermore, ['25I]EGF retains this activity in a bioassay, and its binding to
a0 N. C. Dulak and H. M. Temin, J. Cell. Physiol. 81, 161 (1973).
3, N. C. Dulak and H. M. Temin, J. Cell. Physiol. 81, 153 (1973).
32 G. L. Smith and H. M. Temin, J. Cell. Physiol. 84, 181 (1974).
33 D. Gospodarowicz and J. S. Moran, Am. Rev. Biochem. 45, 531 (1976).
34 G. Sato and L. Reid, in "Biochemistry and Mode of Action of Hormones" (H. V.
Rickenberg, ed.). Univ. Park Press, Baltimore, Mar~' m d (1978).
108
BASIC M E T H O D S
600
e) 500
'f
..~---
[6]
"~ck
~ 400
0
",.
300
200
100
60
120
180
240
MINUTES
FIo. 1 Time course of 1zsI-EGF binding. Six hundred microliters of standard binding
mixture [SF-4X(-EGF), 8 ng/ml 125I-EGF] are applied to the dishes at 37 for various times.
n , cells grown in the hormone-supplemented medium; , cells grown in 10% FCS.
human fibroblasts has been shown to be both specific and saturable. 35 We

have labeled EGF via the chloramine T technique a6 and developed a
miniaturized binding assay that does not require the presenqe of serum,
albumin, or other undefined components.
[125I]EGF binds to HeLa-S cells that are grown in 10% fetal calf serum
in a manner consistent with other reports. There are approximately
140,000 specific binding sites per cell, and the apparent dissociation constants are in the range of 1-7 10 -1, depending upon growth and assay
conditions. At 37, binding is maximal after 45-60 min and, upon continued incubation, decreases with an apparent half-life of 1-1.5 hr. This
must be due to a decrease in the number of binding sites available since
the ligand remains capable of binding quantitatively to fresh cells after a
9-hr incubation period. We have suggested several hypotheses: the
[~25I]EGF-receptor complex may be internalized and degraded with the
labeled fragments released into the media; the receptor could be masked
by a membrane component or by a conformational change; or the iodinated component of the EGF could be removed by a protease leaving the
receptor occupied by the unlabeled residue.
35 G. Carpenter and S. Cohen, J. Cell Biol. 71, 159 (1976).
36 W. M. Hunter and F. C. Greenwood, Nature (London) 194, 495 (1962).
[6]
SERUM-FREE HORMONE-SUPPLEMENTED
MEDIA
109
Cells grown in the serum-free system described below do n o t exhibit

this decrease in apparent binding capacity upon exposure to the ligand.
Furthermore, the addition of a very small amount of serum to the assay
mixture, comparable to the amount of serum resisting the prewash, does
not alter this phenomenon. Thus, this decrease in apparent binding capacity is due to a residual effect of culturing the cells in the presence of large
amounts of many undefined materials; i.e., serum. It is quite possible that
serum proteases adsorbed by the cells are responsible for this effect. It
must also be stressed that EGF is a potent mitogen in this system, decreasing the doubling time from 40 to 20 hr.6 Therefore, the decrease in
apparent binding capacity, whatever the mechanism, is not involved with
the growth-promoting activity of EGF in this system.
The serum requirement of HeLa-S cells can be replaced by the addition of several hormones and transport proteins to the media. 7 The cells
grow in F12 supplemented with insulin (5 /~g/ml), transferrin (5 /~g/ml),
FGF (50 ng/ml), hydrocortisone (50 nM), and EGF (50 ng/ml). Stock
cultures are maintained in this "completely supplemented" media and in
media supplemented with all of the above factors except EGF, subsequently referred to as SF-4X (minus EGF). Cells are plated at a density
such that they will be nearly confluent 5 days after plating (250-300,000/
35 mm dish). The media is changed after 2 and 4 days of growth and the
binding assay performed on the fifth day.
All washes include a second aspiration of medium from a tilted dish.
The cells are washed twice with 2 ml of F12 before the assay is performed
in SF-4X ( - E G F ) . After adding 0.6 ml of the assay mixture (containing
approximately 4 ng labeled [~25I]EGF), the dishes are incubated at 37. The
assay mixture is then removed, and the plates are washed 4 times with 2
ml of cold PB S. After trypsinization in 1 ml of trypsin solution (1 mg/ml of
trypsin and 0.9 mM EDTA) is complete, 0.6 ml of the solution is used for
determination of radioactivity and 0.2 ml is used to determine the cell
number. Nonspecific binding is assayed in the presence of 2.5 /.tg unlabeled EGF.
110
BASIC
METHODS
[7]
[7] U s e o f A n t i b i o t i c s in Cell C u l t u r e M e d i a
By D. PERLMAN
Like Death and Taxes, contamination or the threat o f contamination
is always with us and we need all the weapons to combat them.1
Introduction
Bacterial, yeast, and fungal contamination are hazards to those studying various phases of metabolism and growth of mammalian cellsin vitro.
Although we have used antibiotics for more than 30 years to eliminate or
suppress unwanted microbial contaminants, many who add them to tissue
cultures for this purpose are unfamiliar with the products used, the limitations for their use, and the practical value of this strategy.
The basic requirements are:
1. The antibiotic must eliminate the microbial contaminant. (Bactericidal compounds are preferred over bacteriostatic.)
2. The antibiotic must not inhibit growth and metabolism of the
mammalian cells in tissue culture.
3. The antibiotic must provide "protection" for the complete experimental period.
4. The antibiotic should not affect any ultimate use intended for the
mammalian cells, e.g., virus production or the preparation of antigens.
5. The antibiotic should be nontoxic and safe as far as handling by
laboratory personnel is concerned.
6. The antibiotic should be compatible with other components of the
culture media.
7. The antibiotic should be inexpensive and should not contain excipients, e.g., buffers, which affect cell growth or metabolism adversely.
Materials
Most of the antibiotics considered useful in coping with microbial

contamination in mammalian cells in tissue culture may be obtained by:
(1) direct purchase from a local pharmacy (clinical, community, or hospital); (2) purchase from a biological supply house; or (3) purchase from the
manufacturer.
1 L. L. Coriell, Natl. Cancer Inst., Monog. 7, 33 (1962).
Copyright 19/9 by Academic Press, Inc.

ISBN 0-12-181958-2
[7]
ANTIBIOTICS IN CELL CULTURE MEDIA
111
Those available from the local pharmacy are products that are usually
intended for parenteral administration in the therapy of acute infectious
diseases. They frequently contain buffers, organic acids, and other excipients which aid their usefulness as therapeutic dosage forms, but might
lead to problems in mammalian tissue culture, e.g., the presence of ascorbic acid in the tetracycline formulation. The purity of the antibiotic in
these preparations is usually very high, and the manufacturer frequently
can supply information on the quantity and identity of the impurities.
However, the presence of buffers may result in the potency of the material
on a weight basis being as low as 10% that of the pure material. The
parenteral dosage form usually is sterile and the container fabricated of
materials such that there is minimal carry-over of plasticizers and other
minor chemical contaminants. The capsular dosage form usually is
"semisterile" and the filler and other excipients are likely to be reasonably water soluble. This formulation should be dissolved in the basal
medium and then sterilized by filtration through Millipore or equivalent
filters; asbestos filters should not be used since the antibiotic may react
with the filter.
The antibiotics available from biological and chemical supply houses
include some primarily intended for use in mammalian culture media and
others which are sold on a "user's risk basis." The former should have
been tested for cytotoxicity in mammalian cell cultures, and the vendor
should have data to support claims of low or minimal cytotoxicity; each
manufacturing batch should have its own documentation. The latter materials are sometimes of uncertain origin and not infrequently contain
cytotoxic contaminants and/or microbial contaminants. The user should
be warned to evaluate these materials for cytotoxicity over a period of
time before adding them to a critical experiment. This evaluation might
include study of the interaction of the antibiotic (or mixture o f antibiotics)
with constituents of the medium and other factors.
Several antibiotics may be obtained directly from the manufacturer.
These are usually in the form of nonsterile powders, and the administrative difficulties in obtaining them frequently outweigh the advantages of
obtaining them through this route. Most of the manufacturers are reluctant to sell a few grams of material, and most laboratories are not interested in purchasing kilogram amounts. The result is often a gift of a few
grams, and no guarantees are involved with regard to potency, presence
of microbial and/or chemical contaminants, or cytotoxicity.
Some of the antibiotics which may be useful in the elimination or at
least suppression of contaminating microorganisms from mammalian cell
cultures are listed in Table I together with information on the useful concentration and the length of time protection might be expected under
112
BASIC METHODS
[7]
TABLE I
SOME ANTIBIOTICS USEFUL FOR ELIMINATION OR SUPPRESSION OF CONTAMINATING
MICROORGANISMS FROM MAMMALIAN CELL CULTURES
Antibiotic
Amoxicillin
Amphotericin B
(as deoxycholate
complex)
Amphotericin B
methyl ester
Ampicillin
Carbenicillin
Cephalothin
Chloramphenicol
7-Chlortetracycline
6-Demethyl-7chlortetracycline
Dihydrostreptomycin
Doxycycline
Erythromycin
Gentamycin
5-Hydroxytetracycline
Kanamycin
Lincomycin
Neomycin
Nystatin
Antimicrobial spectrum
Gram-positive and gramnegative bacteria
Fungi and yeasts
Fungi and yeasts
Recommended
concentration
in #g/ml
mediaa
Stability in
media at
37 days b
100
2.5
1.0

Gram-positive and gramnegative bacteria,
especially pseudomonads
Gram-negative bacteria
100
100
100

Gram-positive bacteria and
mycoplasma
Gram-positive and gramnegative bacteria and
mycoplasma
mycoplasma
Gram-positive bacteria
Fungi and yeasts
10
100
20
100
50
100
100
50
4
5
50
[7]
113

TABLE I (continued)
Antibiotic
Paromomycin
Polymyxin B
Penicillin G
Penicillin V
Tetracycline
Streptomycin
Tylosin
Viomycin
Antimicrobial spectrum
Gram-negative bacteria
mycoplasma
Gram-positive bacteria and
mycoplasma
Recommended
coricentration
in/xg/ml
media"
Stability in
media at
37 days ~'
100
50
100
100
10
5
3
3
4
100
100
50
Concentration effective in controlling "light" infection and at the same time not cytotoxic to L~29 and KB cells in serum-containing media.
b Length of time in which at least 10% of initial activity could be demonstrated in
incubation at 37 up to 5 days in serum-containing media.
normal circumstances. Since most are more cytotoxic in the absence of

serum, the r e c o m m e n d e d concentrations should be reduced by (or more)
if serum-free media are used.
Commercial sources for some of the antibiotics listed in Table I are
mentioned in Table II. All of these preparations are likely to contain some
excipients, and the user is advised to check the label on the vial or the
package insert to learn the nature of these materials (as some of them may
have some effect on the mammalian cell growth and/or metabolism). Unfortunately, the nature o f these added excipients that will vary from one
manufacturer to the next, and each source should be treated as unique.
Although a few antibiotics are heat or light stable, most should be
considered as labile, and all preparations should be refrigerated until
needed. Once the dry p o w d e r has been dissolved in water it may be frozen
until used. The exceptions to this latter procedure are nystatin and amphotericin B. The former is water insoluble and once suspended in a buffer
will form aggregates of considerable size. Amphotericin B is available as a
colloidal dispersion (containing sodium deoxycholate); upon refrigeration
the colloid may form an amorphous gel.
114
BASIC METHODS
[7]
TABLE II
SOME COMMERCIALLY AVAILABLE ANTIBIOTIC PREPARATIONS WHICH MAY BE USED
AS TISSUECULTUREMEDIA SUPPLEMENTS
Antibiotic
Amoxicillin
Amphotericin B
Ampicillin
Carbenicillin
Cephalothin
Chloramphenicol
7-Chlotetracycline
6-Demethyl-7-chlortetracycfine
Dihydrostreptomycin
Doxycycline
Erythromycin
Gentamb~cin
5-Hydroxytetracycline
Kanamycin
Lincomycin
Neomycin
Nystatin
Penicillin G
potassium
Penicillin V
potassium
Commercial source
Beecham Laboratories
Roche
E.R. Squibb and Sons
Bristol Laboratories
Ayerst Laboratories
Pfizer Laboratories
E. R. Squibb and Sons
Wyeth Laboratories
Beecham Laboratories
Pfizer Laboratories
Eli Lilly Company
Park, Davis and Company
Lederle Laboratories
Eli Lilly Company
Pfizer Laboratories
Pfizer Laboratories
Abbott Laboratories
Eli Lilly Company
Upjohn Company
Schering Corporation
Pfizer Laboratories
Upjohn Company
S.B. Penick Company
Pfizer Laboratories
Upjohn Company
Pfizer Laboratories
Wyeth Laboratories
Abbott Laboratories
Eli Lilly Company
Wyeth Laboratories
Comments
Take capsules and dissolve
contents and sterile filter
Available as power for
reconstitution; contains
deoxycholate and buffer
Use injectable formulation

note buffer composition
Use injectable formulation;
note composition oflbuffer
Distributed as sterile
solution
solution
Distributed as sterile powder

for resuspension
and not~ composition of
buffer
and note composition of
buffer
[7]
115
TABLE II (Continued)
Antibiotic
Commercial source
Paromomycin
Parke, Davis and Company
Polymyxin B
Streptomycin
Tylosin
Burroughs Wellcome Company

Eli Lilly Company
Merck Sharp and Dohme
Pfizer Laboratories
Pfizer Laboratories
Rachelle Laboratories
Eli Lilly Company
Viomycin
Pfizer Laboratories
Tetracycline
Comments
and note composition of
buffer
solution
Some Strategic Considerations

A n y procedure evaluating the use of antibiotics in preventing or
eliminating microbial contaminants from mammalian cell cultures must
include:
1. A study of the cytotoxicity of the candidate antibiotic to a series of
mammalian cell lines growing in a series of different media;
2. A determination o f the stability of the antibiotic in the inoculated
and uninoculated tissue culture media incubated at 37 in tissue culture
vessels for up to 5 days; and
3. An experimental series where mammalian cells are deliberately
contaminated with bacteria, yeasts, fungi, and/or m y c o p l a s m a and there
is a determination of how effective the antibiotic is in eliminating (not
suppressing) the contaminating organisms. These "purified" mammalian
cell cultures have to be tested for microbial contamination after four to six
transfers in antibiotic-free media.
Since many o f the contaminating microorganisms likely to be a problem in mammalian cell cultures are also likely to be antibiotic resistant,
any antibiotic evaluated in step 2 above should not d e c o m p o s e very rapidly in tissue culture media when incubated at 37 . (Some of the practical
parameters are summarized in our earlier publications. 2-4)
2 D. Perlman and S. A. Brindle, Antimicrob. Agents Chemother. p. 458 (1964).
3 D. Perlman, N. A. Giuffre, and S. A. Brindle,Proc. Soc. Exp. Biol. Med. 105, 1015(1961).
4 D. Perlman, S. B. Rahman, and J. B. Semar, Appl. Microbiol. 15, 82 (1967).
116
BASIC
METHODS
[8]
Our studies have shown that many antibiotics may be fairly well tolerated by mammalian cells and chick tibroblasts growing in various media.
For practical purposes we recommend benzylpenicillin (penicillin G) and
dihydrostreptomycin for elimination or suppression of bacteria, and gentamicin or tylosin to prevent growth of mycoplasmas. Amphotericin B is
fairly useful in controlling fungal or yeast contamination but we question
whether it should be used on a continuous basis since it is somewhat toxic
to most cells.
If the bacterial contaminants turn out to be resistant to the combination of penicillin with dihydrostreptomycin, treatment of the contaminated mammalian cell cultures with tetracycline or chloramphenicol may
be one approach to rescuing the cell line. Cross-resistance among antibiotics should be given some consideration when resistant contaminating organisms are the enemy: organisms resistant to penicillin G are likely to be
resistant to ampicillin, penicillin V, and p~rhaps the cephalosporins,
cephalothin and cephalexin. Organisms resistant to dihydrostreptomycin
may be resistant to kanamycin, gentamy~in,lneomycin, paromomycin,
and ribostamycin.
[8] L a b o r a t o r y M a n a g e m e n t
of Cell Cultures
By KEN WOLF
Introduction
Animal cell or tissue cultures--widely used in different scientific
disciplines--are biological systems that are seldom completely predictable or amenable to absolute control. Accordingly, cell culture has some
attributes of a science while other characteristics resemble those of an art.
As such, some practitioners are consistently more adept or successful
than others. However, anyone who has had more than passing experience
with cell cultures soon learns that things can and do go awry; in fact,
catastrophes sometimes happen.
Contamination is an ever-present threat; bacteria, yeasts, and molds
can flourish in tissue culture media. The presence of such microbial contaminants, fortunately, is usually quickly and readily visible. Moreover,
in many situations, antibiotics are used to minimize the risk of contamination. Cell cultures may also become contaminated but show little or no
visible signs of a problem; virus is one example of nonapparent infection,
but mycoplasmas are by far the most frequent culprits. Detection of
mycoplasma contamination with any degree of confidence is a challenging
feat. A variety of methods have been developed for this purpose and for

ISBN 0-12-181958-2
[8]
LABORATORY MANAGEMENT OF CELL CULTURES
117
the control of mycoplasma and other microbiological contamination, and

they are reviewed in this volume [2].
In addition to microbial contamination, cultures of cell lines can become adulterated, i.e., cell lines can become mixed. Different human and
animal cell lines, for example, have been found to be partly or even wholly
HeLa cells. One wonders about the validity of the conclusions when the
cells were not what they were thought to be. This subject is treated, in
part, in this volume [13].
Culture media, serum, and components of media can be toxic and
jeopardize the cells' performance and, at the extreme, even their continued existence. Precautions to be taken in assuring quality control of
serum for cell culture have been covered in this volume [2]. Additional
problems are encountered when laboratory glassware has harmful residues or when plastic tissue culture ware leaks or has not been uniformly
coated to permit cell attachment. Major disasters or catastrophes happen
when there is critical equipment failure, e.g., when freezers malfunction
or incubators overheat and cultures are killed.
Disciplined and systematic management of cell culture activities will
minimize the number of things that can go wrong and, more importantly,
will greatly reduce the impact on operations when problems do occur.
There are ways by which the real catastrophe can be avoided or completely negated.
The following system of management employs a dichotomous approach, redundancy wherever possible, quality control, safekeeping measures, and careful record-keeping. The system is probably of greatest
advantage to continuing or long-term operations, but some of its features
lend themselves to short-term work as well. It is especially appropriate for
modest laboratoriesmthose with limited budgets and personnel--and in
the long run it will actually save time.
Sequence of Steps in Managing Cell Cultures
1. Determine whether or not there is actually a need to adopt a system
of management. If, for instance, short-term work is planned and quality
cells of the kind to be used are readily available--such as from the American Type Culture Collection--it may be most practical and economical to
purchase certified cells, use them for the particular application, and simply discard them when work has been finished.
2. If it is determined that management method is needed, the quality
of starting stock should be appropriate, i.e., American Type Culture Collection Certified Cell Lines or the equivalent. Mongrel cultures should be
avoided. Propagation of the cells should be divided into stock cultures that
118
BASIC METHODS
[8]
are grown to maintain the lineage under the best possible conditions and
working cultures that are to be used for the actual work applications. Propagation of stock cultures must be accorded the highest work priority and
be carded out with strict aseptic technique and without antibiotics.
3. The second element of dichotomy of the system is in carrying two
sets of stock cultures. To minimize the risk of culture loss due to equipment failure, the two sets should be housed in different incubators. Each
set of cultures should have its own pretested, reserved, single-use portions of frozen, antibiotic-free media; separate records; and separate
schedules for handling. Ideally, the division of risk and responsibility
should be between two persons, but the principle can be employed by an
individual who does the work as two independent activities.
4. Each stock culture should be maintained on two different lots of
appropriate medium. Stock culture medium stored at - 2 0 or lower remains useful for several years, and single-use portions minimize risk of
contamination and obviate formation of precipitates from repeated
freeze-thaw cycles. In practice, stock cultures can be kept at reduced
temperature and may require handling only once a month and, in some
cases, only several times a year.
5. Serum should be obtained from a supplier who certifies it to be free
of mycoplasma. Several different samples are obtained prior to purchase
and tested to determine which gives the best plating efficiency for the cells
to be used. For this test, a suspension containing about 200 cells is added
to duplicate or triplicate 25 cm 2 flasks containing 5 ml of medium with at
least a 20% level of the several samples of serum to be tested. Incubation
is allowed to proceed for 2-3 weeks, after which the flasks are drained,
rinsed with balanced salt solution, stained with 1% aqueous or alcoholic
crystal violet, wasl~ed, and the cell colonies (clones) counted. That serum
giving the most clones should be purchased. Additional measures for
quality control of serum are considered in this volume [2]. Serum should
be stored at - 2 0 or lower and may safely be kept for at least a year
without noticeable loss of potency.
6. Medium and other solutions are prepared only with water of
tissue-culture quality. A freshly prepared borosilicate glass distillate of
deionized water is usually satisfactory. Preferred storage is under sterile
conditions in borosilicate glass. For the small user, USP water for injection is suitable; it is available from pharmacies and hospital supply houses
in liter bottles.
7. Locally prepared media or solutions must be thoroughly tested for
sterility before use. Samples of at least 5% are suggested as is incubation
for 7-10 days at the usual incubation temperature and at 37 . Details are
presented in this volume [2].
[9]
DISPERSION
AND DISRUPTION
OF T I S S U E S
119
8. To minimize adulteration of one cell line with another, only one cell
line should be opened and handled at any given time. All cultures must be
identified at the time of handling.
9. Stock cultures should be propagated in multiples. Because plastic
vessels permit slow gas exchange, 16 x 125 glass tubes with rubber-lined
screw caps, small flasks, or bottles are advantageous since they assure gas
retention, provide multiple cultures in reduced space, and are easily examined microscopically.
10. Stock cultures are to be handled with minimal frequency. To accomplish this use reduced seeding levels and low-temperature incubation.
I1. Maintain detailed records of medium and solutions that are prepared. Include relevant data on all components and notations of when,
how, and by whom cells are handled.
12. Stock cultures should be tested at least once a year and preferably
twice for mycoplasma and other possible contaminants.
13. Replace regular working cultures annually with newly expanded
stock cultures.
14. Attempts should be made to disperse the cells from stock cultures
with an autoclaved solution of EDTA that does not contain trypsin. Trypsin
solutions are not sterilized, and they represent a possible source of culture
contamination. Accordingly, for greatest security, stock cultures should
be dispersed with an autoclaved solution of EDTA.
15. Frozen stocks of unique cell populations or those that are epecially
valuable should be maintained in liquid nitrogen and at more than one
facility. The American Type Culture Collection, Rockville, Maryland
20852 offers such a safe-deposit service.
[9] D i s p e r s i o n a n d Disruption of Tissues

By MARK M. BASHOR
Initial studies in tissue culture carried out near the beginning of this
century involved the explantation of fragments of tissue from an organism
into an artificial in vitro environment. Propagation of such explants usually
involved aseptically cutting or trimming the original explant to reduce its
size so that appropriate gas and nutrient diffusion could occur, and so that
multiple cultures could be obtained. From these pioneering efforts have
developed the closely related disciplines of cell, tissue, and organ culture.
i T h e opinions or assertions contained herein are t h e private views o f the author and are not
to be c o n s t r u e d as offacial or as reflecting the views o f t h e D e p a r t m e n t o f the A r m y or the
D e p a r t m e n t o f Defense. Citation o f trade n a m e s in this report does not constitute an
offacial e n d o r s e m e n t or approval o f the u s e o f s u c h items.
METHODS IN ENZYMOLtX~Y, VOL. LVIII

ISBN 0-12-181958-2
120
BASIC METHODS
[9]
This article will present materials and methods generally applicable to

the preparation of primary cell cultures from organs or tissues of warmblooded animals.
Methods used for preparing cell suspensions from intact tissue may be
broadly classified as (1) mechanical (cutting, mincing, sheafing, sieving);
(2) chemical (usually omission of divalent cations, with or without the
addition of chelating agents); and (3) enzymic (digestion with trypsin,
collagenase, pronase, hyaluronidase). Most techniques are, however,
combinations of two or all three categories. The choice of procedures
depends upon the nature and amount of available tissue, the intended use
of the cultures, and perhaps precedents described in the literature. However, in all instances the methods should be designed to minimize physical
and chemical trauma to the isolated cells in every way possible.
Balanced Salt Solutions and Growth Media
The concept of a balanced salt solution (BSS) is generally credited to
the elegant studies of Sydney Ringer TM published between 1880 and 1895.
The first such solution to be designed specifically for mammalian cells was
by Tyrode 2 in 1910, and many others have since been developed. These
solutions serve several important functions in cell culture. They are useful
for washing tissues and cells, as vehicles for treating cells and tissues with
various agents, and they provide the cells with water and inorganic ions,
while maintaining a physiological pH (7.2-7.6) and osmotic pressure.
Several BSS formulations also include glucose as a source of energy.
Some of the more popular formulations are presented in Table I. Earle's 3
and Hank's 4 BSS and Dulbecco and Vogt's 5 phosphate-buffered saline
(PBS) are the most commonly used. These are all available commercially
as sterile solutions and in some instances as sterile concentrates or dry
powders. If they are to be formulated in the laboratory it is advised that
the solutions which contain bicarbonate be filter-sterilized by pressure
rather than vacuum filtration; the use of 95% air/5% CO2 as pressurant is
also advised. Due to COs volatility it is sometimes useful to prepare the
bicarbonate as a separate sterile solution (7.5%) and add it to the BSS just
prior to use.
The choice of BSS to be used during the preparation of cell suspensions depends to a certain extent upon the procedures which will be used.
If all steps are to be performed under an air atmosphere, then a BSS
,a S. Ringer, J. Physiol. (London) 18, 425 (1895).
M. V. Tyrode, Arch. Int. Pharmacodyn. Ther. 20, 205 (1910).
a W. R. Earle, J. Natl. Cancer Inst. 4, 165 (1943).
4 j. W. Hanks and R. E. Wallace, Proc. Soc. Exp. Biol. Med. 71, 196 (1949).
5 R. Dulbecco and M. Vogt, J. Exp. Med. 99, 167 (1954).
[9]
DISPERSION
AND
DISRUPTION
OF
TISSUES
121
,O
"O
;>..
e-.
'
m
t.
:::t
~.
--
--.
~t
',D
O~
%
z
-/
[-
0
r~
~D
<
t::
e'~
O~ ON
"O
mm
,.--:.~
m
J
,.
~
tt3
,.-2,
0
Z
["
oe~
O",
~ . ,eeq q~
t..
~mo
~ _ ~ : ~ , . ~
<
..,:
~.~
o ~
122
BASIC METHODS
[9]
which is not buffered with bicarbonate should be used, since loss of CO2
will cause the solution to become too alkaline. On the other hand, if some
or all steps will be done under an air/COs atmosphere, then a BSS using a
bicarbonate buffer will have to be employed to prevent acidification of the
medium by dissolved COs. Phenol red may be included, usually at a
concentration of 20-50 mg/liter, as a visual indicator of pH. Glucose may
be included or omitted, depending upon the time elapsing between extirpation of the tissue and the final dispersion of the cells in complete growth
medium. The final consideration in the choice of BSS composition is the
inclusion or exclusion of the divalent cations calcium (Ca) and magnesium
(Mg). As early as 1900, it was recognized that these ions increased the
stability of the intracellular matrix, and many investigators have confirmed this in tissue culture experiments. For this reason several of the
BSS are also commercially available (or are prepared in the laboratory) as
calcium-and magnesium-free solutions (CMF-BSS). The further addition
of chelators, or the readdition of Ca or Mg when certain enzymes are used
as dispersing agents, is discussed below.
It is not the intent of this section to present an extensive discussion of
growth media, as this is covered elsewhere in this volume [5]. Many
formulations are available commercially as sterile solutions and in some
instances as sterile concentrates or dry powders. Most such media were
originally formulated for a particular cell type, such as RPMI-1640 for
lymphocytes or Dulbecco's modified Eagle's medium (DMEM) for fibroblasts. In almost all instances these media are further supplemented with
an animal serum or sera (horse, calf, fetal calf, human; 2-20%). Regardless of the type or composition of the growth medium, or the nature of the
supplements which are added to it, all represent no more than an elaboration of nutrients, hormones, or "growth factors" to a BSS. Occasionally
these media, usually without a serum supplementation, are used in place
of BSS during the intermediate stages of tissue dispersal. More frequently,
however, they are used only as a fnal rinse, immediately prior to suspension of the cells in the complete (serum-supplemented) growth medium
and plating in the culture vessels as described in a later section.
The use and choice of antibiotics is discussed seperately from several
points of view in this volume [2] [3] and [7].
Procurement and Preparation of the Tissue
The source or sources of tissue to be cultured will vary with the design
and needs of a particular laboratory. Two precautions should be observed
in all cases insofar as is possible. The tissue must be kept moist, and a
minimal amount of time should elapse between extirpation of the tissue
and initiation of culture. Under ideal circumstances the culture laboratory
[9]
DISPERSION AND DISRUPTION OF TISSUES
123
will be situated adjacent to the surgical rooms. Recognizing that this ideal
is seldom realized, the following suggestions are made. Samples obtained
from a hospital operating room may be placed by the surgeon in sterile
dishes containing a BSS that one supplied. Once this is accomplished, an
assistant can move them to the culture laboratory. When a considerable
time may elapse between excision and culture, it may be advisable to
place the tissue in a BSS containing glucose or in a complete culture
medium. If extensive transportation is involved, as between a
slaughterhouse and the laboratory, chilling with ice may be required. For
small samples a 100-150 mm sterile, disposable tissue culture dish may be
employed. These are conveniently sealed for transportation by stretching
a length of Parafilm " M " (American Can Co., Neenah, Wisconsin) around
the circumference two or three times. The closest situation to ideal is
when tissues are removed from laboratory animals, in which case the
excision of tissue can be accomplished readily in or very near the culture
laboratory. The animal is anesthetized or sacrificed, the area of incision
shaved (if necessary), sterilized by liberal application of 70% ethanol, and
the sterile field isolated by draping the animal with sterile cloth or gauze.
The incision is then made, the tissue located, removed aseptically, and
placed immediately in sterile BSS or complete medium. The instruments
should be chosen and the surgery accomplished in such a way as to
minimize mechanical trauma to the tissue.
When the tissue is of fetal origin, the uterus is first removed in toto to a
suitable sterile dish, the fetuses dissected out, decapitated, transferred to
clean sterile dishes, and the tissue(s) then removed. Fresh sterile BSS is
used whenever the material needs to be moistened or rinsed free of blood
or small pieces of debris; excess BSS can be aspirated using sterile Pasteur pipettes connected by tubing to a vacuum flask. Small surgical instruments which are used repeatedly may be cleaned and sterilized readily by keeping at hand a beaker of detergent solution and a small brush (a
toothbrush is adequate), one or two beakers of distilled water for rinsing,
and a container in which the instruments can be immersed in 70% ethanol.
Before use, instruments are removed and dried by evaporation or flaming
(carefully).
Dispersion of the Tissue
It is only necessary to consider the extreme differences in tissue architecture exhibited by tissues such as brain, heart, skeletal muscle, liver,
lung, and kidney to appreciate that there is no "standard" procedure for
preparing cell suspensions from these or any other tissues. Perhaps as we
learn more about the nature of the cell surface and cell-cell interactions
we will be able to devise better, more uniform methods for disrupting the
124
BASIC METHODS
[9]
intracellular matrix. In spite of the diversity of these structures, there are

certain steps in the dispersion process which are common to virtually all
procedures. After discussing these, specific examples and exceptions will
be considered.
Mincing the Tissue

It is necessary to render the tissue into pieces of sufficiently small size
so that the solution or solutions used for dispersion can easily penetrate
the matrix. The size of the tissue fragments is generally of the order of 1-3
mm cubes. Except where very large pieces or amounts of tissue must be
processed, the use of scissors should be avoided due to compression
injury to the cells adjacent to the cut. Tissue samples that are obtained
from surgery or biopsy, from adult laboratory animals, and particularly
tissue of fetal origin are obtained in amounts of only a few grams, and
frequently much less. These amounts may be minced conveniently in
sterile 100-mm dishes containing medium or BSS. This may be accomplished by passing two very sharp cataract knives or scalpels 6 smoothly in
opposite directions through the tissue. Care should be taken to accomplish clean cuts through the tissue and to avoid tearing the material. Once
the tissue is minced, the fragments can be transferred with a large-bore
pipette to the sterile vessels in which the cells will be dispersed.
Selecting and Preparing Dispersant Solutions and Treating the Tissue

As mentioned, mechanical, chemical, and enzymic methods are used
to prepare cell suspensions from tissue. Mechanical methods include
homogenizing (very harsh and probably should be avoided), shaking on a
gyrorotary shaker, vortexing, triturating with fine-bore or wide-bore
pipettes, and forcing the minced tissue through a nylon or stainless-steel
mesh; the last is frequently used for preparing lymphocytes from spleen,
thymus, and lymph nodes. 7-9 Typically, however, a procedure will use an
enzymic and/or chemical treatment to weaken the intracellular matrix and
a mild mechanical treatment to release the cells. The use of trypsin for
preparing cell suspensions was first reported by Rous and Jones in 1916. l
Since then the use of hog or bovine pancreatic trypsin preparations has
6 A useful scalpel for this is the Bard-Parker No. 11 blade (disposable) in the No. 7 handle
(Becton, Dickinson and Co., Waltham, Mass.). These are sold by most general laboratory
supply companies.
7 H. R. Hendricks and I. L. Eestermans, J. lrnmunol. Methods 12, 345 (1976).
O. Braendstrup, V. Andersen, and O. Werdelin, Cell. lmmunol. 25, 207 (1976).
9 M. Ohishi and K. Onoue, Cell. Imrnunol. 18, 220 (1975).
10 p. Rous and F. A. Jones, J. Exp. Med. 23, 555 (1916).
[9]
125
become very popular. Several other enzymes have also been used, notably pronase, u-13 collagenase,14'15 hyaluronidase,14-17 elastase, 15'~7 pancreatin, 17,1s papain, 17 and deoxyribonuclease.17.18 It should be mentioned
that virtually all of these enzymes normally contain several activities.
Crude trypsin may contain, in addition to trypsin, chymotrypsin, elastase, ribonuclease, deoxyribonuclease, and amylase. This may be of
considerable importance in designing a protocol for preparing cell suspensions, since in some instances the efficacy of a trypsin-induced dissociation has been attributed to the presence of contaminating activities and
substitution of more highly purified crystalline trypsin has proven less
effective or even harmful. On the other hand, some investigators have
found that crude trypsin contains toxic or harmful substances, whereas
the highly purified enzyme is suitable. A statement of the rationale for
including deoxyribonuclease in some dispersing media may be instructive
at this point with regard to "contaminating" enzyme activities. Essner et
al. 19 found that clumps of a rat ascites hepatoma could be dispersed by
several treatments. Disaggregation by trypsin or chymotrypsin proceeded
in two phases, the first being dissolution of the clumps into single cells, the
second phase being the agglutination of freed cells by a mucous-like
coagulum which formed. Auerbach and Grobstein, 2 studying dissociation
and reaggregation of embryonic mouse tissues, found that certain lots of
trypsin produced a "gummy" material which enmeshed the cells and cell
clumps, preventing complete disaggregation. This material was not
formed when crude pancreatin was included in their trypsin solutions.
SteinbergTM subsequently observed the appearance of a "slimy" material
which trapped cells that had been obtained by trypsin dissociation of heart
and liver cells from 5-day chick embryos. This material was shown to be
DNA which was released from damaged cells, hydrated, and uncoiled to
form a viscous material which readily enmeshed the cells. Steinberg
found that inclusion of crystalline deoxyribonuclease in the dissociation
medium would completely prevent the appearance of the gel.
In recent years better preparations of several enzymes have become
available for tissue culture work. As an example, Worthington Biochemiu D. Weinstein, Exp. Cell Res. 43, 234 (1966).
12 j. C. Sullivan and I. A. Schafer, Exp. Cell Res. 43, 676 (1966).
13 j. F. Foley and B. Aftonomos, J. Cell. Physiol. 75, 159 (1970).
14 p. O. Seglen, Exp. Cell Res. 76, 25 (1973).
P. O. Seglen, Exp. Cell Res. 82, 391 (1973).
16 L. M. J. Rinaldini, Exp. Cell Res. 16, 477 (1959).
17 L. L. Wiseman arid W. R. Hammond, J. Exp. Zool. 197, 429 (1976).
is M. S. Steinberg, Exp. Cell Res. 30, 257 (1963).
19 E. Essner, H. Sato, and M. Belkin, Exp. Cell Res. 7, 430 (1954).
so R. Auerbach and C. Grobstein, Exp. Cell Res. 15, 384 (1958).
126
BASIC METHODS
[9]
cal Corporation (Freehold, New Jersey) produces four types of crude

collagenase. Type I, having a "normal balance" of enzyme activities, is
recommended for fat cells and adrenal tissue; Type II, high in clostripain
activity, is recommended for liver, bone, thyroid, heart, and salivary
tissue; Type III, low in proteolytic activity, is recommended for mammary tissue; Type IV has low tryptic activity and is recommended for
pancreatic islets.
Just as there is no "standard" procedure for tissue dissociation, so too
is there no "standard" buffer for dispersing tissues. Generally, the solutions are prepared in BSS or CMF-BSS rather than a culture medium.
The choice between bicarbonate and nonbicarbonate buffered BSS has
been discussed. The concentrations of trypsin most frequently used are
125-250/zg/ml. When highly purified crystalline trypsin preparations are
used, these concentrations may be-reduced to 10--50/zg/ml. If, in addition
to CMF-BSS, a chelator is desired, EDTA zl or EGTA 22have been used at
concentrations of approximately 20 /~g/ml. Rinaldini, 2a in an extensive
survey of enzymes suitable for preparing dispersions of embryonic chick
cells, used Tyrode's CMF-BSS with 100 /~g/ml sodium citrate as a
chelator. Use of the potassium chelator, tetraphenylboron, should probably be avoided. 24"25
Collagenase is usually used at a concentration of 0.1-0.3 mg/ml, and
sometimes in conjunction with trypsin (0.25 nag collagenase and 0.1 nag
trypsin per milliliter). Pronase has been used at concentrations of 1-2
mg/ml. Deoxyribonuclese is sometimes added, as discussed above, in
concentrations ranging from 1-50 t~g/ml. It should be remembered that
collagenase requires the presence of Ca ions whereas DNase requires Mg
ions.
With the exception of pronase and some collagenase preparations, all
of these enzymes are readily dissolved in BSS or CMF-BSS and may be
sterilized by filtration through 0.2 /xm sterile filters. Pronase and collagenase may require centrifugation to remove insoluble particles prior to
filtration. The solutions may be stored at -10 for 3-6 months.
Once a solution or solutions have been chosen for dispersing the tissue,
there remains the choice of method(s) for treating the minced tissue.
Again there is an endless list of variables and variations, but fortunately
only a few basic principles. The procedure of Dulbecco and Vogt5 for
21 EDTA, ethylenediaminetetraacetic acid.
22 IEGTA, ethyleneglycol bis(/3-aminoethyl ether) N,N'-tetraacetic acid.
23~L. M. Rinaldini, Exp. Cell Res. 16, 477 (1959).
24 S. R. Hiller and J. M. Brown, Exp. Cell Res. 65, 246 (1971).
2s '.p.R. Kerkof, S. Smith, H. T. Gagn~, D. R. Pitelka, and S. Abraham, Exp. Cell Res. 58,
445 (1969).
[9]
127
preparing cell suspensions of monkey kidney (used in the production of

polio vaccines) still serves as a model of general steps for dispersing
tissue. Tissue from the kidneys is minced (as described above) and transferred to a 250-ml flask or centrifuge bottle, washed several times in PBS,
and suspended in PBS containing 0.25% trypsin prewarmed to 37. After
10-min incubation at 37, this solution is removed and discarded,z6 This is
replaced with 20 ml of fresh, prewarmed trypsin and the fragments agitated by pipetting back and forth 10--15 times with an automatic pipette, z7
After allowing the undigested fragments to settle, the turbid supernatant is
transferred to a centrifuge bottle. Warm trypsin is again added, the fragments agitated as before, and the released cells collected. This is repeated
several times until the fragments lose their brownish color and become
clumps of whitish connective tissue. The combined supernatant fluids are
washed 3 times by centtifugation (3 min at 300 g) and resuspension in PBS.
After the last centtifugation, the packed cells are suspended in 10-20 times
their volume of complete culture medium, the cells are enumerated by
hemocytometer count, and diluted to contain 3 l& cells/ml before plating in culture vessels.
Although agitation of the tissue fragments in dispersant solution is still
frequently accomplished by trituration with large-bore and small-bore
pipettes, Rappaportzs modified the procedure of Dulbecco and Vogt and
introduced special "trypsinizing flasks." These are Erlenmeyer flasks
modified for use with a magnetic stirrer by indenting the sides at tight
angles to the bottom at four equally spaced points around the circumference to produce maximum turbulence and mixing. 2s A constriction in the
neck just below a side arm permits trapping of fragments in the bottom of
the flask while pouring the suspended cells from the side arm. Recognizing that trypsin-treated cells are unusually fragile, Rappaport also designed a similar apparatus that allowed continuous addition of fresh trypsin while removing cells as they were suspended. Several other designs
are commercially availableaa such as that devised by Shipman and Smitha
for continuous flow of dispersant. Alternatives to both repetitive pipetting
and magnetic stirring are swirling on a gyrorotary shaker or gentle
vortexing.
26 This initial treatment is used in many other procedures and serves to remove further blood
cells and cells injured during the preparation of tissue fragments.
27 Pipette fillers such as the Bel-Art H-Pumps (A. H. Thomas Co., Philadelphia, Pennsylvania), operated by a thumbwheel, used with sterile cotton-plugged pipettes, is a convenient aid. An automatic pipette filler such as the Drummond Pipet-Aid (Bellco Glass Co.,
Inc., Vineland, New Jersey) is also useful.
~s C. Rappaport, Bull. W. H. 0. 14, 147 (1956).
29 Bellco Glass Co., Inc., Vineland, New Jersey 08360.
3o C. Shipman, Jr. and D. F. Smith, Appl. Microbiol. 23, 188 (1972).
128
BASIC METHODS
[9]
After suspending the cells with an enzyme solution it may be necessary

to reduce or stop the enzyme action by immediate dilution into serumcontaining BSS or growth medium prior to centrifugation and washing
procedures; serum contains potent trypsin inhibitors.
Separation of released cells from undigested fragments, connective
tissue, and other debris can be accomplished by allowing the fragments to
settle (as in the procedure of Dulbecco and Vogt), by very brief centrifugation (15-30 sec at 300 g), by filtration through fine mesh nylon netting,zl
stainless-steel screen, a2 or several layers of washed cheesecloth or gauze.
The filtered cells are transferred to sterile centrifuge tubes z3 and pelleted
by centrifugation for 5-10 min at 300-600 g. The supernatant fluid can be
drawn off without disturbing the cell pellet by placing a Pasteur pipette
attached to a vacuum flask at the top of the tube and slowly inverting the
tube as the medium is drawn off. Cells are resuspended in fresh BBS or
medium by very gentle swirling or vortexing, and the centrifugation is
repeated. Washing may be repeated once or twice more before suspending
the cells in complete growth medium, with the cell density determined by
hemocytometer count or with an electronic particle counter. The cells are
diluted as necessary to a final density of 105-106 cells/ml, inoculated into
the culture vessels, and placed in the incubator.
Alternative Methodologies
While the gequence that has been described is applicable to most tissues, there are interesting and very useful deviations from the general
outline. Some of these may be used for dispersing tissues other than those
described below.
Organ Perfusion: Adult R a t L&er
Adult rat liver is an organ which is particularly amenable to mechanical perfusion, both in situ and in vitro. Seglen 14,15has described a method
for preparihg hepatocyte suspensions from adult rat liver which has been
widely used and occasionally modified. The procedure is unique in that
the dispersant solutions are perfused through the liver rather than treating
minced tissue. The method also involves sequential treatment of the tissue
with different solutions. The method as used by Williams et ai. a4 is briefly
31Availablefrom Tetko, MontereyPark, California91754.
32Availablefrom PerforatedProducts Inc., Brookline,Massachusetts02146.
oa Disposable, sterileplastic centrifugetubes, with caps, are availablein 15- and 40-ml sizes
from several manufacturersof tissue culture plasticware.
34G. M. Williams,E. Bermudez,and D. Scaramuzzino,In Vitro 13, 890 (1977).
[9]
129
stated as follows. The liver of an adult rat is perfused for 4 min with
Hanks's CMF-BSS buffered with 0.05 M HEPES 35 and containing 0.5
mM EGTA. 2z After this treatment the liver is perfused for 10-12 min with a
coilagenase solution (100 units/ml) in Williams' s Medium E a6 buffered with
0.05 M HEPES. The cells are combed into fresh collagenase solution,
centrifuged (50 g for 4 min), and resuspended in complete growth medium
before inoculating into culture vessels.
Sequential Enzyme Treatment: Mouse Mammary Gland

It is sometimes possible and advantageous to use a sequential
enzyme-treatment procedure for preparing cell suspensions. A good example is a method described by Wiepjes and Prop07 for preparing epithelial cells from mouse mammary glands. Virgin mice, 9-10 weeks old, are
sacrificed and the skin and mammary glands are removed. The glands are
dissected free, minced, and the fragments treated for 15 min at 36 with a
CMF-BSS containing glucose. Gentle agitation is provided by a gyrorotary shaker. The floating tissue fragments are transferred to a solution of
0.125% collagenase, 0.1% hyaluronidase, 0.4% demineralized bovine
serum abumin in CMF-BSS and incubated for 45 min. A subsequent
incubation on the gyrorotary shaker for 60 min in 1.25% pronase dissolved
in Medium 199 completely disrupts the tissue. Twice the volume of serum
is added, and the ceils are sedimented by centrifugation (2 min at 300 g).
The supernatant liquid and floating fat cells are removed and discarded.
The pellet is gently suspended in complete culture medium containing
approximately 1 /~g/ml DNase, filtered, washed twice with complete culture medium, and plated in culture dishes.
Enzymic Dispersion after Culture: Rat Pineal Gland

It is not always necessary to prepare cell dispersions from minced
tissue immediately after excision and fragmentation. The pieces of tissue
can be rinsed with several changes of BSS, transferred directly to culture
dishes containing complete growth medium, and placed in the incubator,
changing the medium as necessary. During this time of culture the
explants will attach to the bottom of the dish and cells may begin to grow
out of the explant and onto the surface of the dish surrounding the
explant. Should the explants tend to float, a smaller volume of medium
may be used, or the explants may be held against the bottom of the dish by
35HEPES, N-2-Hydroxyethylpiperazine-N'-2-ethanesulfonicacid.
36 G. M. Williams and J. M. Gunn, Exp. Cell Res. 89, 139 (1974).
37 G. J. WiePies and F. J. A. Prop, Exp. Cell Res. 61, 451 (1970).
130
BASIC METHODS
[9]
overlaying a piece of washed, sterilized dialysis tubing or a sterile filter

soaked with the medium. At the discretion of the investigator the medium
is drawn off, the attached fragments and cells rinsed with BSS, treated
with a dispersing agent, and suspended in fresh medium. At this stage,
cells can be separated from remaining fragments by centrifugation or filtration, or they may again be cultured together.
Nathanson et aL as have used such a method for preparing cell cultures
from adult rat and avian pineal glands. Essentially their method involves
removing and mincing the isolated glands into 10--15 fragments. The
explants from 2-4 glands are then transferred to 100-mm tissue culture
dishes and incubated at 37 in a modified Ham's F-12 medium containing
10% fet~ calf serum and antibiotics. After 2-4 weeks of culture, the dishes
are rinsed With Hanks's BSS and incubated for 10 min at 37 with 10 ml of
0.25% trypsin. Gentle trituration with a pipette is used to suspend the cells
and fragments which are then transferred to Erlenmeyer flasks and incubated an additional 10 min on an incubator gyrorotary shaker at 37. This
suspension is again gently pipetted, transferred to centrifuge tubes containing an equal volume of ice-cold complete culture medium, and centrifuged at 200 g for 3 min. The pellet is then resuspended in fresh cold
complete medium, and the remaining clumps of cells are removed by
centrifugation (50 g for 1 min). The supematant is centrifuged again (200 g
for 3 min) to pellet the single cells which are then suspended in fresh
medium and plated at a density of 2 x 104 to 10 ~ cells per 100-mm dish.
Mechanical (Nonenzymic) Dispersal of Tissue: Chick Embryo Skeletal
Muscle
The use of mechanical forces to prepare suspensions of lymphocytes
from spleen and thymus 7-9 has been mentioned. The procedures cited rely
on forcing the tissue through a fine-mesh nylon or stainless-steel screen.
An entirely different mechanical procedure has been successful for preparing chick embryo skeletal muscle cultures. 39"4The procedure relies on the
dispersion of cells from the tissue fragments by the shearing forces of
circularly flowing medium. The swirling action is provided by suspending
the fragments in medium in a test tube which is "vortexed" on a suitable
laboratory mixer 41for a minute or less. The cells are then filtered through
nylon cloth, diluted, and placed in culture vessels. The authors state that
a8 M. A. Nathanson, S. Brinkley, and S. R. Hilfer, In Vitro 13, 843 (1977).
a9 A. I. Caplan, J. Embryol. Exp. Morphol. 36, 175 (1976).
40 j. C. BuUaro and D. H. Brookman, In Vitro 12, 564 (1976).
41 Vortex-Genie, Scientific Industries, Inc., Bohemia, New York 11716; Super-Mixer, LabLine Instruments, Inc., Melrose Park, Illinois 60160.
[9]
131
the advantages are that the method is more rapid than conventional trypsinization; that the cells are released directly into the growth medium,
myoblasts having a much shorter recovery period before fusion occurs;
and that a lower level of fibroblast contamination is achieved.
Similar procedures have been used for other types of muscle4~ and for
neural tissue, 4a and they may prove useful for other tissues as well.
Determining Cell Yield and Viability
Judging a procedure for the preparation of cell suspensions from living
tissues as successful or.unsuccessful depends to a certain extent upon the
needs and requirements of the investigator and the purpose for which the
cell cultures are being established. Important considerations in making
such a judgement are: the adequacy of the cell yield, i.e., the number of
cells obtained per gram or milligram of tissue; the viability of the cells;
and the ability of the cells tO retain in culture the properties and functions
expected of them.
The cell yield is determined by counting the number of ceils in a
known volume of the final suspension and calculating the total number of
cells obtained from the original mass of tissue (See this volume [11]).
The most commonly used criterion for cell viability is based on the
assumption that viable cells will exclude certain dyes such as trypan blue,
whereas nonviable cells will take up the dye (See this volume [13]). Other
criteria for viability include measures of the ability of the cells to attach to
the surface of the culture vessel, i.e., the fraction of cells inoculated into a
vessel that will attach to the surface. Occasionally, the ability of the cells
to take up or incorporate a radioactive compound may be useful. The
ability of cells to survive normal culture conditions or clonal growth conditions is a further criteria of viability. The integrity of the cell membrane
and the general appearance of the suspended cells should be monitored by
phase-contrast microscopy. If possible, the ultrastructure of the cells
should be examined by scanning and transmission electron microscopy.
42 S. M. Heywood, A. S. Havaranis, and H. Herrmann, J. Cell. Physiol. 82, 319 (1973).

43 R. E. Mains and P. H. Patterson, J. Cell Biol. 59, 329 (1973).
132
BASIC METHODS
[10]
[10]
Monolayer Culture Techniques
B y JAMES A . M C A T E E R a n d W I L L I A M
H. J.
DOUGLAS
Introduction
Monolayer cell cultures are frequently established from single cell
suspensions prepared by the enzymic dissociatibn of organ fragments.
The cell preparation is inoculated into a culture vessel containing fluid
medium and incubated in a controlled atmosphere. Viable cells settle and
attach to the substrate within several hours. The resultan t primary culture
is a mixed cell population which contains many of the cell types present in
the tissue of origin. In order to obtain monolayer cultures consisting of a
single cell type, it is necessary to isolate that cell from the primary preparation. Cell separation and isolation methods such as density gradient
centrifugation, ~ electrophoresis, 2 or affinity column separation a can be
applied to the initial cell suspension prior to culture. Alternatively, the
cell population can be enriched using in vitro methods that select cell types
based on their attachment or growth characteristics. For example, fibroblasts and macrophages attach to culture surfaces more rapidly than other
cells and thus they can be removed by selective adherence techniques. 4
Fibroblasts replicate faster than most cell types. In mixed cell cultures
they tend to overgrow the population and obscure other cells of interest.
Dilution plating at clonal density 5 is often used to separate cell types so
that colonies which develop may be physically isolated and then subcultured as pure populations.
The cells routinely studied in monolayer culture can be derived in the
laboratory from primary cultures, or obtained from commercial suppliers
and nonprofit organizations such as the American Type Culture Collection 6 and Institute for Medial Research. 7 The most commonly studied,
well-characterized cell lines include the WI-38 and IMR-90 (human diploid, fibroblastic), 3T3 (mouse embryonic, fibroblastic), and HeLa (hu1 T. G. Pretlow II, Int. Pathol. 16, 42 (1975).
2 R. C. Boltz, Jr., P. Todd, M. J. Streibel, and M. K. Louie, Prep. Biochern. 3, 383 (1973).
a S. F. Schlossman and L. Hudson, J. lmmunol. 110, 313 (1973).
4 F. H. Kasten, in "Tissue Culture: Methods and Applications" (P. F. Kruse, Jr. and M. K.
5 R. D. Cahn, H. G. Coon, and M. B. Cahn, in "Methods in Developmental Biology" (F. H.
Wilt and N. K. Wessels, eds.), p. 493. Crowell (Collier), New York, 1967.
The American Type Culture Collection. Rockville, Maryland, 20852.
r The Human Genetic Mutant Cell Repository. Inst. Med. Res., Camden, New Jersey,
08103.
METHODS IN ENZYMOLOGY,VOL.LVIII
Copyright1979by AcademicPress, Inc.

All rightsof reproductionin any formreserved.
ISBN 0-12-181958-2
[10]
MONOLAYER CULTURE TECHNIQUES
133
man cervical adenocarcinoma, epithelioid) lines. Numerous other cell

lines from many vertebrate, insect, and plant species have been established, characterized, and catalogued, and specific techniques for them
are presented throughout this volume.
Basic monolayer cell culture techniques are well established and accessible. Publications are available that detail culture procedures s and
techniques courses are sponsored by organizations such as the Tissue
Culture Association. 9
The information presented in this article provides background, resources, and practical procedures for the investigator interested in establishing a research program involving monolayer cell culture.
Properties of Monolayer Cultures
Most cell populations maintained in monolayer culture form multiple
cell layers and in the case of smooth muscle, form large, grossly visible
colonies. Only a few cell types, e.g., vascular endothelial and 3T3 cells,
grow as true monolayers.
Cell types which require attachment to a rigid substrate in order to
replicate in vitro are classified as anchorage-dependent cells. These include most, but not all, cells maintained in monolayer culture. Many
abnormal cell types, e.g., transformed and neoplastic cells, may grow well
in fluid suspension culture or in a soft agar matrix.
The growth kinetics of cells in monolayer culture follow a characteristic pattern. Following seeding the cells undergo a quiesceht period (lag
phase) during which there is no cell division. The duration of the lag phase
is dependent, in part, upon the cell type, seeding density, media composition, and previous handling of the cells. The cells then enter a log phase of
growth in which there is an exponential increase in cell number. During
the log phase the cells exhibit their highest metabolic activity. When
culture conditions will no longer support cell division the population enters a stationary phase, during which the cell number remains constant.
The attainment of stationary phase is largely dependent upon nutritional
factors within the system and can occur before or after confluenc'y is
reached.
The term "confluent" describes a cultured population which occupies
all available growth surface. If nutritional conditions are adequate, a confluent cell population may continue to replicate in log phase. In practice,
however, it is not wise to maintain cells beyond the point of confluency
8 Tissue Culture Association Manual: Techniques, Methods and Procedures for Cell, Tissue and Organ Culture. Tissue Cult. Assoc., Rockville,Maryland,20852.
9 Course Secretary, W. AltonJones Cell Science Center, Lake Placid, New York.
134
BASIC METHODS
[10]
since essential metabolites are rapidly depleted at high cell density. Some
cell types will not replicate beyond confluency and are described as being
contact inhibited. "Contact inhibition" is a widely misused term which
encompasses the phenomena of contact inhibition of movement TM and contact inhibition of replication. 11 These are separate, probably unrelated,
aspects of cell behavior. Contact inhibition of movement describes the
well-documented condition that occurs when the directional movement of
a cell changes upon contact with another cell. Contact inhibition of replication is the cessation of growth due to physical crowding of cells at
confluency. This occurrence is not well documented and in most cases
may be a misinterpretation of a well-known cellular response to nutrient
depletion more correctly described as a density-dependent inhibition of
growth, lz
Cultured cells can become " t r a n s f o r m e d . " Cellular transformation is
a spontaneous or induced change, e.g., viral transformation, in the
characteristics of a cell type expressed as an abnormal karyotype, a
change in morphology, or alteration in cell behavior or growth potential.
Spontaneous cellular transformation often occurs with extended time in
culture. Since a transformed cell population will no longer respond
characteristically to experimental conditions, the age of the cells at the
onset of transformation is an important parameter. The age of a culture is
usually expressed as either its passage number or number of population
doublings. Passage number refers to the number of times the cells have
been subcultured since the time of explantation. This is not a precise
estimate of cell age for it is affected by numerous variables including: the
number of cells replanted at each passage, cell survival after inoculation,
and the cell density (i.e., confluency, semiconfluency) at subcultivation.
The number of population doublings is a more critical estimate of cell age
since it accounts for the time required for the cells to double in number.
The number of population doublings a culture has undergone is calculated
by knowing the number of cells originally planted and the cell number at
the next subcultivation.
Normal diploid cells in vitro have a finite lifespan, while transformed
cells apparently possess unlimited proliferative potential. The process of
in vitro cellular aging is considered to be a deterioration of the cellular
processes that are necessary to support continued replication. As a cultured cell population ages its population doubling time progressively increases until proliferation eventually ceases. During the process of senelo M. Abercrombieand J. E. M. Heaysman,Exp. Cell Res. 6, 293 (1954).
11L. N. Castor, J. Cell. Physiol. 72, 161 (1968).
12M. G. P. Stoker and H. Rubin, Nature (London) 215, 171 (1967).
[10]
135
sence, cells undergo alterations such as chromosomal aberrations and

changes in membrane structure and permeability.
Environmental Factors in Cell Culture
The physical culture environment should be well defined and precisely
controlled. Numerous factors influence the growth of cultured cells including temperature, pH, osmolality, humidity, and the composition of
the overlying gas phase. The control of these variables is essential for
interpretation of cellular response and requires an understanding of not
only the cellular requirement for each factor, but the interactions between
factors. The overall culture milieu must be constructed with the knowledge that cells in v i t r o continuously alter their environment. Cells utilize
nutrients and thus deplete the media of certain constituents. They also
produce metabolites and "waste products," which can interact with medium components to degrade growth-promoting factors or alter pH. The
range of proper incubation temperature is usually quite narrow, and for
most cells from homeothermic animals it is between 35-37 . Cells can
survive for extended periods at a lower than optimal temperature but do
not tolerate high temperatures. Lower temperatures depress cellular metabolism, usually without deleterious effect. Cultures remain viable at
room temperature for periods of up to several days, long enough to allow
shipment through the postal system of living cells that have been properly
packaged. Cultures derived from poikilothermic animals are normally incubated at lower temperatures. Cells from cold-water fishes TM are cultured
at 15-20 , warm-water fishes at 22-27 , and insect cells TM have been cultured in a wide temperature range, 23-37 . (See this volume [39] and [40]).
Optimum pH of the medium is usually within the range 7.2-7.5, though
some cells will survive at the extremes, pH 6.6-7.8. The conditions which
support maximum growth in turn range _+0.5 units. The determination of
proper pH within a culture system is dependent upon both the cell type
and the specific physiological parameter being assessed. For example, a
certain cell type may exhibit its greatest secretory activity at a pH different from that which supports the most rapid growth.
The maintenance of pH involves the interaction of several factors,
most importantly the buffer system and the overlying gas phase. Cells in
culture have a metabolic requirement for both bicarbonate ion and CO2.
Some cells produce sufficient CO2 as a metabolic by-product to satisfy this
la K. Wolfand M. C. Quimby,Tissue Cult. Assoc. Man. 2, 445 (1967).
~4j. L. Vaughn,in "Invertebrate Tissue Culture" (C. Vago, ed.), Vol. 1, p. 3. Academic
Press, New York, 1971.
136
BASIC METHODS
[10]
requirement, however, bicarbonate must be added to the medium. Many

media are buffered with some type of bicarbonate buffer system. These
contain bicarbonate in excess of that needed by the cells, and require CO2
from the gas phase to activate the buffer and establish the bicarbonate ~ CO2 equilibrium. The pCOz required to equilibrate the buffer is
dependent upon the bicarbonate concentration, but usually it falls in the
range of 12-38 mm Hg. With most commercially available tissue culture
media this requires an atmosphere of approximately 5% CO2. Media can
be constructed with organic buffers which do not require equilibration
with COs. 15 These buffers make it possible to culture within closed systems in which the Oz requirement is easily met by the ambient air (gas
phase) within the vessel. Recently, however, the use of organic buffers
has been criticized, Reports indicate that certain organic buffers may be
toxic to cultured cells. 16
Cultured cells require oxygen. A pO2 of 15 to 75 mm Hg is usually
adequate for monolayer culture. Open culture systems (Petri dishes, or
flasks with loosened caps) are usually incubated in a 95% air-5% COs
mixture. Higher oxygen concentrations can be detrimental, and 02 levels
only slightly higher than ambient have been shown to be toxic to some
cultured cells. 17
A very high relative humidity (>98%) must be maintained when working with open culture systems. Since cells in monolayer are submerged in
fluid medium there is no danger of drying at the cell surface, although
evaporation from the medium must be avoided. The humidity of the incubator should be in equilibrium with the gas phase immediately overlying the media.
Other environmental factors are known to influence monolayer culture. Visible light has been shown to affect cells in culture. Direct metabolic effects are known, as well as light-induced production of toxic compounds in some media.18 Cells should be cultured in the dark and should
be exposed to room lighting as little as possible. Vibration in the laboratory may also be detrimental to certain systems. Cells in clonal growth are
particularly sensitive to vibration.
The chemical composition of the culture medium is of obvious importance, not only from a nutritional standpoint (detailed in this volume [5]
but with regard to osmolality. Media osmolality for mammalian cell
15 N. E. Good, G. D. Winget, W. Winter, T. N. Connolly, S. Izawa, and R. M. M. Sing,h,
Biochemistry $, 467 (1966).
16 H. Eagle, Science 174, 500 (1971).
17 A. Mizrahi, G. V. Vosseller, Y. Vagi, and G. E. Moore, Proc. Soc. Exp. Biol. Med. 139,
118 (1972).
is R. J. Wang, In Vitro 12, 19 (1976).
[10]
137
culture is usually held in the range of 280-300 mosmol/kg. The optimal

value may depend upon the specific cell type.
Substrates for Monolayer Culture
Cells in monolayer culture require a substrate for attachment. Ideally
this surface should be nontoxic, biologically inert, and optically transparent. It must allow cell attachment and cell movement required for replication and migration during growth.
The interaction between cell and substrate is an important factor in the
status of the culture system. If the two are incompatible, the culture will
not survive. Cells may not adhere at all to some surfaces while other
substrates may allow attachment but inhibit growth.
A variety of materials have been used successfully as culture substrates. Cells can be grown on cellophane, carbon or collagen-coated glass
and plastic, Teflon, silicone rubber, polycarbonate and cellulose ester
(Millipore) filters, and many other surfaces. Routinely most laboratories
use commercially available borosilicate glass or specially treated polystyrene plastic culture ware. Plastic tissue culture products are more popular because of the problems encountered with washing and rinsing
glassware. The plastics used for monolayer culture are specially treated
during manufacture to provide a hydrophilic surface which promotes both
cell attachment and growth. These products are designated "culture
grade" as opposed to "bacteriological grade" vessels which will not support the growth of cultured cells. Under certain culture conditions even
culture grade plastics are not suitable for cell growth. When diploid cells
are seeded at clonal density in medium containing a low serum concentration plating efficiency is poor. Pretreatment of the plastic culture vessel
with polylysine permits efficient clonal growth. TM (See also [5]).
Various types of culture vessels are available for specific applications.
Those designed for routine monolayer culture are made with a uniform,
transparent growth surface to facilitate their use with an inverted microscope. These include the petri dish, multi-well plate, screw-cap culture
flask (T-flask), and Leighton tube.
Monolayer culture systems used primarily in industry for the mass
propagation of anchorage-dependent cells are designed to provide the
maximum surface area for growth. The roller bottle 2 is used for this
purpose. The system consists of a cylindrical borosiUcate glass or
polycarbonate plastic bottle which rests horizontally on a roller device
!9 w. L. McKeehanand R. G. Ham, Cell Biol. 71, 727 (1976).
20W. L. Whittleand P. F. Kruse, Jr., in "TissueCulture: Methodsand Applicatiolas"(P. F.
Kruse, Jr. and M. K. Patterson, Jr., eds.), p. 327. AcademicPress, New York, 1973.
138
BASIC METHODS
[10]
within an incubator or controlled-temperature room. The bottle is partially filled with medium, and as it turns the cells adherent to its walls are
alternately bathed in the fluid medium and exposed to the gas phase.
Routine Cell Culture Procedures

Subculturing or passaging cells involves detaching the cells from their
substrate and transferring them to new culture vessels. This is done when
cells have utilized the growth surface available to them or have reached a
population density which suppresses their growth. Since the replicative
capacity v f a cultured cell population declines when confluency is
reached, cells are often passaged at "semiconfluency," when they are still
in log-phase growth.
Various methods can be used to remove cells from the culture surface.
Mechanical means such as scraping with a silicone rubber spatula removes the cells in clumps or irregular patches. This is somewhat disruptive to the cells and usually results in a poor recovery. Dissociation with a
proteolytic enzyme such as trypsin, 21 collagenase, 22 or pronase 23 gives a
better cell yield and much higher plating efficiency. Enzymes are usually
prepared in a Ca2+-Mg 2+ free balanced salt solution. They are used separately, in combination, or with the addition of a divalent cation chelating
agent such as EDTA. Enzyme concentrations of 0.05-0.5% are used, and
crude enzymes are usually preferred. (See also this volume [9]).
Enzymic dissociation damages cells, especially the cell surface. Thus
it is desirable to minimize the duration of contact between the cells and
enzyme solution. It is also wise to avoid harsh physical manipulation of
the cells while in the presence of enzymes, although gentle pipetting is
usually necessary to prepare a monodisperse suspension.
The following protpcol outlines one procedure for passaging cells in
monolayer culture.
Dissociation
1. Remove and discard the culture medium.

2. Wash the cell sheet thoroughly with warm Moscona's saline24; discard the fluid.
21C. Shipman, Jr., in "Tissue Culture: Methods and Applications" (P. F. Kruse, Jr. and
M. K. Patterson, Jr., eds.), p. 5. Academic Press, New York, 1973.
22 S. R. Hilfer, in "Tissue Culture: Methods and Applications" (P. F. Kruse, Jr. and M. K.
2a R. B. L. Gwatkin, in "Tissue Culture: Methods and Applications" (P. F. Kruse, Jr. and
M. K. Patterson, Jr., eds.), p. 3. Academic Press, New York, 1973.
24 A. A. Moscona, Exp. Cell Res. 3, 535 (1952).
[10]
139
3. Add warm enzyme solution (collagenase 0.1%, trypsin 0.1%, chicken serum 1% in Moscona's saline) (1 ml/5 cm ~ culture surface). After 1
min withdraw the solution, leaving only a very thin layer of fluid coveting
the cells.
4. Observe the cells with an inverted phase-contrast microscope; note
distribution and predominant cell morphology.
5. Cap the flask tightly and incubate at 37 for 3-5 min.
6. Again observe the cells with a phase microscope; look for freefloating cells and attached cells which have pulled in their processes and
are rounding up. If the majority of cells do not appear to be detached
return the flask to the incubator.
7. After several minutes of incubation hold the flask horizontally and
give its side a sharp rap with the hand. This will help detach loosely
adherent cells. Often the cells loosen in sheets and these can be seen to
drift across the substrate surface when the flask is tipped at an angle.
Observe the cells by phase microscopy. Repeat if necessary.
8. Add several milliliters of culture medium to the flask. Use a
pipette to gently triturate the cells.
Seeding
9. A viable cell count 25 (dye exclusion) is then performed on the suspension so that other flasks can be seeded at a known cell density.
10. Cell dilutions are made and aliquots seeded to new culture vessels
containing warm (equilibrated) medium.
11. The newly seeded culture vessels are incubated undisturbed to
facilitate cell attachment.
Morphological Evaluation of Monolayer Cultures
Many cultured cells possess distinctive morphological characteristics
observable by phase-contrast microscopy. This enables the investigator
to identify specific cell types in mixed cultures and to carry out procedures such as clonal isolations. Routine observations by phase microscopy are essential for proper assessment of a cell population. The morphology of cultured cells may change in response to slight alterations in
culture conditions. An abnormal morphological change in the cell population is often the first indication that a problem exists within the system.
Therefore, it is useful to frequently monitor culture morphology and to
keep a record of the observations. A Polaroid-type camera back affixed to
an inverted microscope is very useful for this purpose.
25 H. J. Phillips, in '*Tissue Culture: Methods and Applications" (P. F. Krusc, Jr. and M. K.
140
BASIC METHODS
[10]
Fixed and stained cell preparations of various types are used for morphological analysis of monolayer cultures. Clonal plates are prepared for
gross visual examination by fixing the cultures in situ (buffered 2.5%
glutaraldehyde) and then staining directly with 0.5% crystal violet. For
light microscopical analysis of monolayer cultures/n situ cells can be
grown on coverslips, fixed with Bouin's solution, stained with hematoxylin and eosin, and mounted on microscope slides. 26 This provides good
resolution, and cell structure is well preserved.
Critical morphological analysis of cultured cells by high-resolution
light microscopy and transmission electron microscopy necessitates embedding and sectioning of the cultures. The cells need not be removed
from their substrate for processing, and all steps through embedding can
be carded out within the culture vessel. Since monolayer cultures are only
one to several cell layers in thickness, fixation and dehydration are rapid,
and infiltration with resin is uniform. Convefitional preparation methods
can be used with the precaution of avoiding solvents that might dissolve
the tissue culture vessel. Processing cells in situ does not hinder sectioning
since resin will infiltrate between the cells and substrate allowing them to
be separated following polymerization. 27 The tissue block can then be
oriented for enface or transverse sectioning. Certain types of resins may
work better with specific brands of plastic culture ware. One reliable
combination is the use of an Epon mixture 28 with Lux or Coming plastic
surfaces. Processing cells grown in glass dishes is more difficult because
polymerized resin does not readily separate from the dish. Cells grown on
glass coverslips are easier to work with. If the coverslip is embedded so
that resin contacts only the side which bears cells the coverslip can be
separated from the polymerized plastic by immersing them repeatedly in
liquid nitrogen and then boiling water.
The preparation of cultured cells for scanning electron microscopy
presents no unusual difficulties. Processing by the critical-point drying
method is recommended. 29 One necessary precaution is to assure that the
cells are grown on a substrate that can be conveniently placed in the
chamber of the critical-point dryer. Alternatively, pieces of the culture
surface must be cut from the culture vessel without damaging the cells.
The cells may then be fixed, critically point dried, and coated by routine
methods.
26 j. Fogh and J. A. Sykes, In Vitro 7, 206 (1972).
27 W. H. J. Douglas, E. P. Dougherty, and G. W. Phillips, ~ssue Cult. Assoc. Man. 3, 581
(1977).
2s C. C. Haudenschild, R. S. Cotran, M. A. Gimbrone, Jr., and J. Folkman, J. Ultrastruct.
Res. 50, 22 (1975).
29 K. R. Porter, D. Kelley, and P. M. Andrews, Proc. Annu. Stereoscan Syrup., 5th, 1972 p.
1 (1972).
[11]
MEASUREMENT
OF G R O W T H
AND VIABILITY
141
[11 ] M e a s u r e m e n t o f G r o w t h a n d V i a b i l i t y o f Cells in C u l t u r e
By MANFORD K. PATTERSON,JR.
A variety of techniques has been proposed for the quantitation and

measurement of viability of cells grown in tissue culture. Only a few have
emerged as being suitable for routine purposes. Since the preparation of
tissues for cell culture or cells for subculture is metabolically destructive,
the use of cell count alone for quantitation may be misleading. A desired
adjunct for growth studies, therefore, is a determination of the viable cell
population.
Cell culture measurements can be divided into four major catagories:
(1) visual methods which employ the use of light microscopy and devices
commonly used in hematology; (2) chemical methods which employ
commonly used analytical biochemical procedures adapted to tissue culture; (3) electronic systems using flow-through cells or apertures for measurement of incorporated dyes or cell numbers; and (4) an array of miscellaneous procedures.
Cell Preparation
The basic adaptations of analytical biochemical procedures for tissue
culture material have been for increased sensitivity because of the limited
amount of tissue available and to accomplish in situ dissolution of the
cells. The latter has been necessary to avoid the cellular alterations
caused by the hydrolytic enzymes used to dissociate or detach cells although, under certain circumstances, cells can be detached I or dissociated 2 by nonenzymic procedures.
In the procedures to be described it should be noted that often more
than one measurement can be made on a single preparation of cells. For
example, cells prepared for direct nuclei counting or for nucleic acid
analysis also can be used for protein determination.
The handling of cells prior to analysis is extremely important for a
variety of reasons, namely: (1) The exogenous growth media may contain
components that are being assayed in the cell, e.g., proteins that adsorb to
cells. (2) Leakage of cellular components may occur if cellular integrity is
lost at this point. (3) Loss of cells may result from severe mechanical
handling. (4) Use of hydrolytic enzymes may lead to the loss of measurai M. K. Patterson, Jr., in " T i s s u e Culture: M e t h o d s and A p p l i c a t i o n s " (P. F. K r u s e , Jr. and
M. K. Patterson, Jr., eds.), p. 192. A c a d e m i c Press, N e w York, 1973.
2 C. W a y m o u t h , In Vitro 10, 97 (1974).
METHODS IN ENZYMOLOGY,VOL. LVIII
Copyright 1979by AcademicPress, Inc.

ISBN 0-12-181958-2
142
BASIC METHODS
[11]
TABLE I
BALANCED SALT SOLUTIONS a
CaCI2
CaCI2 2 H20
KC1
KH2PO4
MgCI~ 6 H20
MgSO4 7 H20
NaCl
NaHCOa
Na~HPO4 7 H20
NaH~PO4 - H20
Glucose
Phenol red
Dulbecco b
Earlec
Geyd
Hankse
0. l0g
0.20
0.17g
0.14#
0.20
0.20
0.I0g
0.40
0.37
0.03
0.21
0.07
7.00
2.27
0.226
8.0
0.20
6.80
2.20
2.16
0.14
1.00
0.01#
i.00
Puck Ff
Puck Gs
0.40
0.06
0.016
0.285
0.083
0.016
0.40
0.150
0.20
8.00
0.33
0.09
0.154
7.40
1.20
0.29
0.154
8.0
1.00
0.01#
1.10
0.0012
1.10
0.0012
0.29
a Usually prepared as a x 10 solution and diluted prior to filtration. Cell suspensions are
generally washed with calcium- and magnesium-free solutions. Concentrations given
are g/liter, x 1 solution.
bSterilize by autoclaving; R. Dulbecco and M. Vogt, J. Exp. Med. 99, 167 (1954).
cSterilize by filtration; J. C. Bryant, Tissue Cult. Assoc. 1, 185 (1975).
d Sterilize by filtration; G. O. Gey and M. K. Gey, Am. J. Cancer 27, 45 (1936).
e Sterilize by filtration; J. H. Hanks, Tissue Cult. Assoc. 1, 3 (1975).
rSterilize by filtration; T. T. Puck, S. J. Ciecura, and A. Robinson, J. Exp. Med. 108,
945 (1958).
v Prepare as separate solution and 'mix prior to use; may be omitted if solution is to be
used for wash.
b l e m a t e r i a l s . P r o p e r h a n d l i n g o f c u l t u r e s r e q u i r e s c a r e f u l a t t e n t i o n to p H ,
t e m p e r a t u r e , a n d o s m o t i c p r e s s u r e d u r i n g w a s h i n g , a n d a b a l a n c e bet w e e n a d e q u a t e r e m o v a l o f g r o w t h m e d i a w i t h o u t loss o f cells.
Choice of washing solutions appears to be a simple matter of prefere n c e . T h e b a s i c f o r m u l a s f o r t h e m o s t c o m m o n l y u s e d are g i v e n i n T a b l e I.
I n g e n e r a l t h e c h o i c e s h o u l d b e d i c t a t e d b y t h e salt s o l u t i o n u s e d as the
b a s e for the g r o w t h m e d i u m . M o r e o v e r , c a l c i u m - f r e e salt s o l u t i o n s are
g e n e r a l l y p r e f e r a b l e f o r all cell s u s p e n s i o n s s i n c e this d i s c o u r a g e s cell
a g g r e g a t i o n . A c o m p r e h e n s i v e listing o f b a l a n c e d salt s o l u t i o n s h a s b e e n
published, z
Growth Measurements
Visual Methods
T h e m o s t c o m m o n l y u s e d m e a s u r e m e n t o f g r o w t h is d i r e c t e n u m e r a t i o n o f cells e m p l o y i n g a h e m o c y t o m e t e r . T h i s m e t h o d is b e s t a p p l i e d
n c. Waymouth, in "Cell Biology" (P. L. Altman and D. D. Katz, eds.), Vol. I, p. 61.
FASEB, Bethesda, Maryland, 1976.
[11]
MEASUREMENT OF GROWTH AND VIABILITY
143
when only a few samples are to be counted or quantitative information on

cell viability is desired. If numerous cultures are to be counted on a
routine basis, electronic enumeration should be considered.
HEMOCYTOMETER C O U N T I N G 4
Reagents and Equipment

Balanced salt solution-see Table I
Hemocytometer-double chamber with Neubauer rulings
Microscope-with 16-ram objective and x 10 ocular
Hand counter--tally register
Procedure. Cell suspensions (direct or prepared from tissue or
monolayers) are diluted with balanced salt solution to contain approximately 300,000-500,000 cells/ml. Taking care not to overfill, the cell suspension is added to both chamber sides of the hemocytometer using a
pipette or dropper. The chamber coverslip should be firmly in place.
Any number of squares in the hemocytometer can be counted. Reproducibility is the key factor so that sufficient squares should be counted to
obtain a count statistically representative of the population. Each large
square (there are nine on each chamber side) represents an area of 1 mm 2
and a depth of 0.1 mm, i.e., a volume of 0.1 mm s. As an example, in 10
1-mm 3 squares (the four comer and center squares of each chamber side),
200 total cells are counted. The original cell suspension had been diluted
1:10; therefore, 200 cells x 10 (dilution) x 1000 (mm a in 1 cm a) equals 2
106 cells/ml in the original cell suspension.
This procedure is also applicable to the direct enumeration of cell
nuclei and the dye exclusion test for viable cells. Some cells, e.g., human
diploid cells, are difficult to count directly, and a staining procedure is
incorporated to enhance visibility of the nuclei.
NUCLEI COUNTING
Reagents
Citric acid-O.1 M (1.9212 g/100 ml distilled water) to which a few
crystals of thymol are added
Crystal violet----0.2% dissolved in 0.1 M citric acid
Balanced salt solution--see Table I
Procedure. Monolayer cultures are rinsed 3 times with balanced salt
solution (this is optional depending upon media and subsequent use, e.g.,
if citric acid extract is to be used also for protein determination) and
4 M. Absher, in "Tissue Culture: Methods and Applications" (P. F. Kruse, Jr. and M. K.
144
BASIC METHODS
[1 1]
drained. Two milliliters of 0.1 M citric acid are added to a T-255 flask (5.0
ml for T-75) and allowed to stand at room temperature. The time can vary
but a minimum of 10 min has been found adequate for disruption of all
cells that have been tested. Frequent shaking of the flask during incubation can aid in disrupting the cells but is not always necessary, since the
final suspension is prepared by scraping the flask surface.
Note: Protein analysis may be performed on this extract by adding
Lowry's reagent directly to an aliquot (see "Chemical Methods").
A 0.1-ml aliquot is diluted to 1.0 ml with 0.1 M citric acid and 1.0 ml of
the crystal violet solution added. The resulting suspensions of muclei have
been found stable for up to 1 week at 4 .
The suspended nuclei are counted in a hemocytometer; calculations
are made as described.
Chemical Methods
The relationship of protein and nucleic acid biosynthesis to the growth
of cells has led to the measurement of th~se components as an expression
of cell growth. However, variations in their content occur, and a linear
relationship is not always found. The validity of this type of assay should
be established by comparison with the more direct visual methods.
PROTEIN DETERMINATION
Oyama and Eagle s were the first to adapt the Lowry 7 assay for protein
to measurement of cells in tissue culture. Subsequent studies have shown
wide variations in ratio of protein per cell values (Table II).
Reagents
Lowry solution A---20 g Na~CO3 and 4 g N a O H pellets dissolved to
give a final volume of 1 liter (store in plastic bottle)
Lowry solution B--Equal parts of 1% CUSO4.5 H~O and 2.7%
sodium potassium tartrate
Lowry solution C m l 0 0 ml solution A, 2 ml solution B
Phenol reagent (Folin and Ciocalteau)ml N (usually purchased as a 2
N solution)
Human serum albumin, crystallized--Stock solution containing 200
/zg/ml
s Tissue culture flasks are available in various sizes constructed of glass or plastic. The
"25" represents 25 cm ~ surface area.
e V. I. Oyama and H. Eagle, Proc. Soc. Exp. Biol. Med. 91, 305 (1956).
r O. H. Lowry, N. J. Rosebrough, A. L. Farr, and R. S. Randall, J. Biol. Chem. 193, 265
(1951).
[11]
145
TABLE II
PROTEIN CONTENT OF CELLS IN TISSUE CULTURE
Days after plating
3
References
480
500
450
500
370
330
350
330
b
o
930
810
890
1010
850
960
920
780
1010
740
640
790
940
700
320
360
220
175
800
710
300
350
180
150
400
720
320
410
175
150
390
400
310
400
175
100
872
860
787
678
498
588
545
446
Cell line
L-5 (monolayer)
L-5 (suspension)
HeLa
Human leukemia
Intestinal epithelium
WI-38
Jensen sarcoma
WISH
HEp-2
Chang (suspension)
HeLa (suspension)
CF-3
IMR-91
Human fetal lung
20-29 doublings
30- 39 doublings
40-49 doublings
50- 59 doublings
320 a
350
550
500
pg/cell
1150
1100
350
390
400
230
c
c
c
410
390
320
350
a
a
d
d
e
s
s
g
490
480
883
1287
Italicized values are extrapolated from curves; data given in picograms per cell.
b A. Tsuboi, T. Kurotsu, and T. Terasima, Exp. Cell Res. 103, 257 (1976).
c V. I. Oyama and H. Eagle, Proc. Soc. Exp. Biol. Med. 91, 305 (1956).
d E. Miedema and P. F. Kruse, Jr., Biochem. Biophys. Res. Commun. 20, 528 (1965).
e p. Voipe and T. Eremenko-Volpe, Cur. J. Biochem. 12, 195 (1970).
i R. T. Dell'Orco, unpublished.
g E. L. Schneider and S. S. Shorr, Cell, 6, 179 (1975); days not specified; cell cultures
either in logarithmic growth (<1.3 104 cells/cm~) or when the monolayer reached
light confluency (1.3-2.6 104 cells/cm2).
Procedure. Cell monolayers or suspensions are washed 3 times with

balanced salt solutions and allowed to drain by inversion. Sufficient
Lowry solution C is added to cover the cell layer, e.g., 5 ml for a T-25
flask, and it is then incubated. Incubation conditions should be sufficient
to dissolve the cellular material. Generally, 37 for 1 hr is sufficient or
temperatures up to 60 may be used for 5 min. (Following dissolution the
solutions may be stored at 4 for several days.) The intensity of the color
produced by the reaction of the alkaline copper-tartrate solution C is
indicative of the aliquot required to obtain a final readable color. If, for
example, no color is detectable in the 5-ml solution in the T-25 flask, then
146
BASIC METHODS
[11]
2- to 3-ml aliquots are required; a deep blue color suggests a 1:2 or 1:10
dilution with solution C. A final volume of 5 ml of sample in solution C
plus 1 ml distilled water or standard (0-200/zg) receives 0.5 ml of phenol
reagent, jetted to obtain rapid admixture. A digest from cell-free controls
is best used as a blank. Thirty minutes after addition of the phenol reagent, the absorbance of the samples is measured with a spectrophotometer at 660 nm. Color is stable for at least 2 hr. Absorbance is compared
with the absorbance values of standards in the range of 0-200/zg.
Limitations. A large number of compounds including organic buffers,
sucrose, s glycerol, s,9 Tris, ~ and certain reducing agents u interfere with
the Lowry protein method, and special sample treatment may be necessary when such reagents are used in cultures.
D N A DETERMINATION
As with protein, the DNA content per cell depends upon the stage of
growth and the phase of the cell cycle.n-Is It is therefore advisable to
equate the results of DNA determination with other methods of evaluation. The procedures in general use for tissues are adapted to cells derived
from cell culture systems. Several critical steps exist in the estimation of
nucleic acid, TM the majority of which concern extraction and hydrolysis.
Since one is dealing with small amounts of tissue, special care must be
exercised if quantitation is to be achieved. The disintegration of the cell
sample is a critical step in obtaining an accurate determination of nucleic
acids because of the widespread presence of nucleases. Two general procedures exist: (1) removal of the cells from the solid matrix, or (2) in situ
extraction. An example of the former uses a freeze-thaw procedure which
separates a particulate fraction containing the cellular DNA and a soluble
fraction containing cytoplasmic enzymes TM prior to removal of interfering
substance by cold acid precipitation of the nucleic acids. An example of in
situ extraction can be found in the procedure to be described (cf. also
Setaro and MorleylZ).
s A. Bensadoun and D. Weinstein, Anal. Biochem. 70, 241 (1976).
g P. Bl'fimel and W. Uecker, Anal. Biochem. 76, 524 (1976).
to R. Rej and A. H. Richards, Anal. Biochem. 62, 240 (1974).
11 F. Higuchi and F. Yoshida, Anal. Biochem. 77, 542 (1977).
lz E. L. Schneider and S. S. Shorr, Cell 6, 179 (1975).
la A. Leyva, Jr. and W. N. Kelley, Anal. Biochem. 62, 173 (1974).
14 V. J. Cristofalo and D. Kritchevsky, Med. Exp. 19, 313 (1969).
t5 R. Yanishevsky, M. L. Mendelsohn, B. H. Mayall, and V. J. Cristofalo, J. Cell. Physiol.
84, 165 (1974).
16 H. N. Munro and A. Fleck, Methods Biochem. Anal. 14, 113 (1966).
1 F. Setaro and C. G. D. Morley, Anal. Biochem. 71, 313 (1976).
[11]
147
Methods of analysis of hydrolyzed extracts for DNA vary widely.

Most have used the colorimetric reagent, diphenylamine, which is sensitive to about 10/zg or approximately l0 s cells. One method, based on the
fact that an antibiotic, mithrarnycin, binds to DNA and fluoresces at 540
nm in direct proportion to the DNA present, reportedly can detect as little
as 0.5/zg or 5 104 cells. TM Other microfluorometric methods using antibiotics, 19ethidium bromide, ~-22 and D-aminobenzoic acid 17,2z,24have been
reported. Table III gives the literature values for DNA content of cells
grown in tissue culture.
Reagents
Perchloric acid, 1.0N; dilute 8.62 m170-72% reagent grade to 100 ml
Sodium hydroxide, 0.3 N; 1.2 g/100 ml
Diphenylamine--dissolve 1.5 g in 100 ml glacial acetic acid; add 1.5
ml concentrated H2SO4; mix fresh and store in dark until used
Acetaldehyde, 16 mg/ml; dilute 0.5 ml to 25 ml with distilled water;
store at 4
Burton reagent--just before use add 0.1 ml of acetaldehyde solution
to each 20 ml of diphenylamine reagent required
DNA standard--sodium desoxynucleate, dissolve 0.2 mg/ml of 5 mM
NaOH; store at 4; mix equal volume with 0.5N perchloric acid for
a final concentration of 100/.~g/ml
Extraction Procedure (for T-25flask). Wash cell layer 3 times with balanced salt solution. Wash once with 5 ml cold 0.2N perchloric acid. Add 5
ml cold 0 . 2 N perchloric acid and let stand 10 min at 4; decant. Add 2 ml
of 0.3N N a O H and let stand 1 hr at 37; transfer to 15-ml centrifuge tube
and cool in ice. Add 2 ml of cold 1.0N perchloric acid and let stand 10 min
at 4 to precipitate DNA and protein; centrifuge (supernatant contains
RNA fraction). Wash the pellet with 2 ml cold 0.2 N perchloric acid;
centrifuge. Resuspend pellet in 2 ml of 0.5 N perchloric acid and heat in
water bath for 15 min at 90 to hydrolyze DNA. Centrifuge and retain
supernatant liquid. Wash the pellet with 2 ml 0.5 N perchloric acid; centrifuge and combine supernatant fluid with that obtained in the immedila B. T. Hill and S. Whatley, FEBS Lett. 56, 20 (1975).
19 B. T. Hill, Anal. Biochem. 70, 635 (1976).
z0 U. Karsten and A. Wollenberger, Anal. Biochem. 46, 135 (1972).
21 U. Karsten, Anal. Biochem. 77, 464 (1977).
22 M. J. Blackburn, T. M. Andrews, and R. W. E. Watts, Anal. Biochem. 51, 1 (1973).
2a W. Y. Fujimoto, J. Teague, and R. H. Williams, In Vitro 13, 237 (1977).
24 H. L. Hosick, Tissue Cult. Assoc. Man. 1, 45 (1975).
148
aASXC METHODS
[1 1]
TABLE III
DNA CONTENT OF CELLS IN TISSUE CULTURE
Cells
DNA (pg/cell)
Method
Reference
WI-38 passage 20-29

30-39
40-49
50-59
Human fibroblast---day 0
2
5
WI-38 passage 19-54
Human fibroblast
9.6 +- 3.7
9.7 _+ 2.8
10.6 -+ 3.0
8.4 _ 2.5
7.4
9.8
7.5
7.94 ___ 0.55
9.0 + 0.4
8.3 -+ 0.45
8.5
8.0 -+ 0.1
7.5 _+ 0.4
7.8 +- 0.4
7.5 -+ 0.27
8.62 -+ 0.67
13.7
Colorimetric
Colorihaetric
Human iymphocytes
Granulocytes
Granulocytic leukemic
3T3 mouse
Colorimetric
Fluorometric
Fluorometric
Fluorometric
Colorimetric
Absorbance
Fluorometric
Fluorometric
Absorbance
Absorbance
d
:
g
E. L. Schneider and S. S. Shorr, Cell 6, 179 (1975).

b A. Leyva, Jr. and W. N. Kelley, Anal. Biochem. 62, 173 (1974).
c V. J. Cristofalo and D. Kritchevsky, Med Exp. 19, 313 (1969).
d B. T. Hill, Anal. Biochem. 70, 635 (1976).
e B. T. Hill and S. Whatley, FEBS Lett. 56, 20 (1975).
s U. Karsten, Anal. Biochem. 77, 464 (1977).
g U. Karsten and A. Wollenberger, Anal. Biochem. 46, 135 (1972).
h W. Y. Fujimoto, J. Teague, and R. H. Williams, In Vitro 13, 237 (1977).
G. T. Rudkin, D. A. Hungerford, and P. C. Nowell, Science 144, 1229 (1964).
J H. Morimoto, P. A. Ferchmin, and E. L. Bennett, Anal. Biochem. 62, 436 (1974).
ately preceding step. (Note: The pellet may be dissolved in 0.3 N NaOH
and used for protein determination.)
Analytical Procedure (Colorimetric-Burton). Standard curve: Prepare
duplicate standards of 5, 10, 25, and 50/zg of D N A per tube by adding
0.05, 0.10, 0.25, and 0.50 ml of working standard (100/~g/ml) to individual
tubes and adding 0 . 5 N perchloric acid to a final volume of 2.0 ml. Prepare
a standard blank of 2.0 ml 0.5 N perchloric acid. To 2 ml of tissue extract
and tubes containing the D N A standards, add 4.0 ml of Burton's reagent.
Incubate 16-20 hr at room temperature. Read absorbance at 600 nm and
calculate by comparing with the absorbance of the D N A standards.
Recent modifications suggest substitution of paraldehyde for acetaldehyde and its inclusion with diphenylamine to produce a more stable
[11]
149
reagentZS; a shortened incubation period (4 hr) reportedly results in an

increase in sensitivity when acetaldehyde is omitted and the incubation is
conducted with 10,000-15,000 lx of light. 26
Miscellaneous Enumeration Methods

ELECTRONIC
For routine purposes, especially when many samples are involved,

blood cell counters are used. 2r-29 There are several commercial models
available that also yield information about the size of cells. The more
recently developed flow systems for automated cytology (FMF or flow
microfluorometry z) have been applied to measurement of cell number
and viability, zl Simultaneous measurement of both D N A and protein has
been accomplished, zz
HEMATOCRIT
Although a rapid and simple method has been proposed by

Waymouth, 33 it is seldom used. The method correlated with hemocytometer counts up to 3 106 cells.
IN SITU
Efforts to enumerate cells in situ have been directed mostly toward the
use of individual frames from time-lapse cinematography: a4,a~ Others
have enumerated cells in a microscopic field, ao-as These methods are limited to low cell counts.
25 G. M. Richards, Anal. Biochem. 57, 369 (1974).
26 j. R. Decalionne and J. C. Weyns, Anal. Biochem. 74, 448 (1976).
27 M. Harris, in "Tissue Culture: Methods and Applications" (P. F. Kruse, Jr. and M. K.
zs W. F. Daly, in "Tissue Culture: Methods and Applications" (P. F. Kruse, Jr. and M. K.
29 M. E. Kaighn and D. J. Merchant, Tissue Cult. Assoc. Man. 3, 649 (1977).
30 H. A. Crissman, P. F. Mullaney, and J. A. Steinkamp, Methods Cell Biol. 9, 179 (1975).
31 W. P. Drake, P. C. Ungaro, and M. R. Mardiney, Transplantation 14, 127 (1972).
a2 H. A. Crissman, M. S. Oka, and J. A. Steinkamp, J. Histochem. Cytochem. 24, 64 (1976).
aa C. Waymouth, J. Natl. Cancer Inst. 17, 305 (1956).
a4 L. N. Castor, Exp. Cell Res. 68, 17 (1971).
as G. Froeze, Exp. Cell Res. 65, 297 (1971).
36 C. W. Jamieson, D. Russin, E. Benes, C. W. DeWitt, and J. H. Wallace, Nature (London)
222, 284 (1969).
37 G. Martinez-Lopez and L. M. Black, In Vitro 9, 1 (1973).
as G. F. -Y. Lee and D. L. Engelhardt, J. Cell. Physiol. 92, 293 (1977).
150
BASIC METHODS
[11]
OTHER
Included here are turbidometry a9 and dry weight 4 measurement.
Expression of Growth Data

Measurement of growth rate implies a change in cell number during a
specified period of time. In most instances the data can be expressed
simply as cell number; in others it is advantageous to express the data in
such a manner that cell doubling time or percent of doubling is easily
determined. Expression of cell number as a function of logz is used if
such is the case. Cell number can easily be converted to log2 by multiplication of the common logarithm of the cell count by the factor 3.3219. For
example:
Cell count at time 0 = 2 105
Cell count at 48 hr = 1.3 l0 s
(a) log2 at time 0 = (5.3010) (3.3219) = 17.6094
(b) log2 at 48 hr = (6.1139) (3.3219) = 20.3080
(c) number of doublings in 48 hr (b - a) = 2.6996
Doubling time of culture population, 48 + c = 17.78 hr
Graphical expression of log2 (cell count) can be made on linear graph
paper if the above calculations are used or with log graph paper if direct
count is plotted.
Measurements of Viability
Numerous criteria are used to determine cell viability. A survey by
Malinin and Perry 41 listed 40 viability assay methods. These may be
grouped into six broad categories: (1) survival and growth in tissue culture, (2) membrane integrity, (3) metabolite incorporation, (4) enzyme
spectrum, (5) trafisplantation potential, and (6) structural alteration,
chemical composition, and electroconductivity.
The ability of cells to withstand the rigors of dispensing agents,
changes in environment, freezing, and thawing is indicative of cell integrity but is not sufficiently quantitative for description. Beating heart cells
39 C. Peraino and W. J. Eisler, Jr., in "Tissue Culture: Methods and Applications" (P. F.
Kruse, Jr. and M. K. Patterson, Jr., eds.), p. 351. Academic Press, New York, 1973.
40 D. J. Merchant, R. H. Kahn, and W. H. Murphy, in "Handbook of Cell and Organ
Culture," p. 160. Burgess, Minneapolis, Minnesota, 1964.
~1 T. I. Malinin and V. P. Perry, Cryobiology 4, 104 (1967).
[11]
151
are strong indicators of viability but give no representation of the survival

of the cell inoculum.
Although cloning, 4z assaying for "cultivable" cells, 43 trypsin digestion, 44 and uptake of isotope 45 have been used to determine cell viability,
all are time-consuming. Most commonly used have been the metabolic
stains and dyes whose uptake is dependent upon cellular membrane integrity or whose metabolism is indicative of enzyme activity. Of the latter,
the tetrazolium salts are the most common. 44"46These are water-soluble,
colorless substances which, in contact with cellular reductases, are reduced to insoluble chromogens. Following incubation (up to 6 hr) of cells
in the presence of tetrazolium salts the appearance of discrete granules in
the cytoplasm is indicative of a viable cell.
For routine purposes " d y e exclusion" tests are preferred. Schrek 4r
was the first to suggest that an intact cell membrane was necessary for the
exclusion of certain dyes. Dyes which have been used for this purpose
include safranin, 4s eosin, 43'44'47'49 Congo red, 5 erythrocin, 51 nigrosin, 52
trypan blue, 43"44"51"53"54and alcian blue. 55 The last was compared to the
other six stains for use following gluteraldehyde fixation and found to be
the only one that was reproducible before and after fixation. Similar resuits were obtained with cells fixed for up to 1 week. No differences were
noted among the dyes when applied to unfixed cells using established
procedures. For routine purposes the most commonly used procedure is
that with trypan blue and a visual count of the unstained "live" cells.
Since trypan blue absorbs effectively at 633 nm, the wavelength of the red
Helium-Neon laser, the laser used in certain of the flow-automated cytology systems, electronic enumeration of live-dead cell ratios is possible. 31
42 K. B. MacDonald and W. R. Bruce, Exp. Cell Res. 50, 471 (1968).
42 j. R. Tennant, Transplantation 2, 685 (1964).
44 j. M. Hoskins, G. G. Meynell, and F. K. Sanders, Exp. Cell. Res. 11, 297 (1956).
45 B. T. Mossman, Tissue Cult. Assoc. 3, 663 (1977).
4e F. H. Straus, N. D. Cheronis, and E. Straus, Science 108, 113 (1948).
47 R. Schrek, Am. J. Cancer 28, 389 (1936).
42 j. Paul, in "Cell and Tissue Culture," p. 284. Williams & Wilkins, Baltimore, Maryland,
1961.
49 j. H. Hanks and J. H. Wallace, Proc. Soc. Exp. Biol. Med. 98, 188 (1958).
50 C. F. Geschickter, Stain Technol. 5, 49 (1936).
5~ H. J. Phillips, in "Tissue Culture: Methods and Applications" (P. F. Kruse, Jr. and M. K.
~2 j. p. Kaltenbach, M. H. Kaltenbach, and W. B. Lyons, Exp. Cell Res. 15, 112 (1958).
53 W. Sawicki, J. Kieler, and P. Briand, Stain Technol. 42, 143 (1967).
54 S. Tolnai, Tissue Cult. Assoc. Man. 1, 37 (1975).
55 D. K. Yip and N. Auersperg, In Vitro 7, 323 (1972).
152
BASIC METHODS
[12]
Acridine orange 56 and a fluorescent derivative of disulfonic acid 57 have

also been used as a vital stain.
Reagents and Equipment

Trypan blue (vital stain, C.I. 23850)-400 mg added to 90 ml water
containing 810 mg NaC1, 60 mg K2HPO4, and 50 mg methyl
p-hydroxybenzoate. Heat mixture to boiling, cool, and adjust pH
to 7.2-7.3 with 1N N a O H (approx. 8 drops). Adjust to final volume
of 100 ml.
Hemocytometer
Procedure. Prepare cell suspension as described for cell count. In a

test tube mix 0.9 ml of cell suspension and 0.1 ml trypan blue. After 5 min,
add to hemocytometer. Count blue stained cells as nonviable.
Note: Trypan blue has a greater affinity for serum proteins than for
cellular protein. 5 If the background shows heavy staining the concentration of dye should be increased or cells should be centrifuged and resuspended prior to staining. Timing of exposure of cells to dye can be critical
in that cells continue to take up stain. If this appears to be the case, cells
should be removed from the staining solution prior to counting. ~
56 M. R. Melamed, L. R. Adams, A. Zimring, J. G. Murnick, and K. Mayer, Am. J. Clin.
Pathol. 57, 95 (1972).
5~ I. Katz, Tissue Cult. Assoc. Man. 1, 41 (1975).
[12] C l o n i n g
By LOLA C. M. REID
A. Introduction
A cloned cell population is one derived from a single parental cell.
Cloned populations provide the experimenter with the distinct advantage
of minimal genetic variability in the model system under study. Methods
for isolating single cells and propagating them into a population of cells
are called cloning, and the majority of them fall into one of three categories of techniques:
1. Dilution Plating. Suspensions of cells are diluted with a sufficient
volume of medium to permit addition of single cells to dishes.
2. Cloning of Anchorage-Dependent Cells. Dilute concentrations of cells
are seeded onto one of several possible substrates. The cells are allowed to
attach and to grow into colonies of cells. The colonies of cells are then
individually subcultured and transferred to other dishes.
Copyright 1979by Academic Press, Inc.

All rights of reproductionin any form reserved.
ISBN 0-12-181958-2
[12]
CLONING
153
3. Cloning of Anchorage-Independent Cells. Dilute concentrations of

cells are seeded into medium solidified by agar or agarose. Cells capable
of growing in suspension in the agar-medium form morulae-shaped colonies. These colonies are transferred individually to culture dishes.
All cloning procedures require careful consideration of factors affecting the survival and growth of isolated cells. In the development of cloning techniques it was discovered that cells leak essential nutrients into the
medium and secrete chemical messengers that are required for survival
and growth. 1-4 In cell culture jargon, the cells are said to "condition the
medium," and medium containing these nutrients and chemical messengers is referred to as "conditioned medium." Because an isolated cell in a
large volume of medium cannot condition the medium adequately, the
ability of the isolated cell to grow in the given medium is a measure of the
cell's autonomy to the factors in conditioned medium. A quantitative
measure of this is provided by the plating efficiency, the number of colonies of cells (each colony deriving from one cell) that form in plates
seeded with dilute concentrations of cells (usually 100-200), calculated as
the number of colonies, over the number of cells seeded, multiplied by
100. Many established cell cultures and most primary cell cultures have
low plating efficiencies, often less than 1% and as low as 0.001%. Since low
plating efficiencies are due to the cells' nutritional needs in low cell densities, special cloning media have been developed that are enriched for
amino acids and vitamins. 5-1z The chemical messengers found in conditioned media are provided by feeder layers of cells or by supplementation of the cloning medium with old media from confluent cell populations.
The techniques described include several of the most routinely used
methods from each of the three categories of cloning techniques, as well
as methods for preparing feeder layers and cloning media supplemented
with conditioned media. The capillary technique originated by Sanford 13
and used to produce the first clonal cell cultures is effective but tedious
i H. Eagle, Science 122, 501 (1955).
z H. Eagle, Science 130, 432 (1959).
3 K. Sanford, L. Dupr~e, and A. Covalesky, Exp. Cell Res. 31, 345 (1963).
4 R. G.Ham, Exp. Cell Res. 29, 515 (1963).
5 R. G. Ham, Biochem. Biophys. Res. Commun. 14, 34 (1964).
6 R. G. Ham, Proc. Natl. Acad. Sci. U.S.A. 53, 288 (1965).
7 R. G. Ham, In Vitro 10, 119 (1974).
s W. McKeehan, S. Hamilton, and R. G. Ham, Proc. Natl. Acad. Sci. U.S.A. 73, 2023
(1976).
W. McKeehan, K. McKeehan, S. Hammon, and R. Ham, In Vitro 13, 399 (1977).
10 S. Iwakata, and J. Grace, N . E State J. Med. 64, 2279 (1964).
11 R. Parker, Spec. Publ., N.Y. Acad. Sci. 5, 503 (1957).
12 j. Morgan, J. Morton, and R. Parker, Proc. Soc. Exp. Biol. Med. 73, 1 (1950).
13 K. K. Sanford, W. R. Earle, and G. D. Likely, J. Natl. Cancer Inst. 9, 229 (194'8).
i54
BASIC METHODS
[12]
even in the simplified form.~4 It has been superceded by easier methods

and consequently will not be presented here. Methods for cloning plant
cells are essentially analogous to those for mammalian cells (see this
volume [41]). Guides to specific nutritional requirements and other modifications necessary for plant cells ls,le (see this volume [41]) and for cells
from invertebrates (see this volume [39]) are available.
B. Equipment and Supplies Needed for All the Cloning Procedures
Equipment
Desk-top centrifuge
Inverted phase microscope with a total magnification of x 50 to x 100
Incubator, maintained at 37 with a water-saturated atmosphere containing 5% CO2 in air
Biogaard or laminar flow hood. The hood minimizes contamination of
cultures. If a hood is not available, one can do reasonably well with
an inexpensive, plastic-sheltered area in which an ultraviolet
light is placed to minimize pathogens.
Hemocytometer for counting cells. Cell counting also can be done
with a Coulter counter if it is available.
Bunsen burner
Supplies
Plastic culture dishes and flasks can be obtained from a number of
suppliers. However, the plating efficiency and growth of cells may
vary somewhat on the plastics from different commercial suppliers.
Therefore, one should select plastics from one company and use
them exclusively for all experiments in which data are to be compared.
Pipettes:
Sterile Pasteur pipettes (7" and 9"). Some pipettes should be
plugged with cotton prior to sterilization.
Sterile 1- and 5-ml volumetric glass or plastic pipettes
Sterile test tubes holding 10 ml
Basal media: See Table I for a listing of possible cloning media.
Serum: Fetal bovine serum is most commonly used in cloning since it
has high concentrations of growth factors and pregnancy hor14K. K. Sanford, in "Tissue Culture: Methods and Applications" (P. F. Kruse, Jr. and
M. K. Patterson, eds.), pp. 237-241. Academic Press, New York, 1973.
15 A. C. Hildebrandt, in "Tissue Culture: Methods and Applications" (P. F. Kruse, Jr. and
16 p. K. Dougall, in"Tissue Culture: Methods and Applications"(P. F. Kruse, Jr. and M. K.
Patterson, eds.), pp. 261-264. Academic Press, New York, 1973.
[12]
CLONING
155
TABLE I
CLONING MEDIAa
Name
Eagle's Minimal
Essential Medium (MEM)
Ham's Cloning Media
F10
F12
F12M
F12K
MCDB104
RPMI
CMRL- 1066
Medium 199
Use
References
Established cell lines
2~
Primary cultures or freshly explanted cells
5-9
4
WI-38 and other fibroblast-llke cells

Lymphocytes and leucocytes
Embryonic cells
Virus-infected cells
6
a
9
20
~1
n
a The compositions of the media listed can be found in this volume [5] and in the
specific references noted. Media should be freshly prepared every 2-3 weeks and
stored at 4. Longer storage results in the loss of labile components such as
glutamine. Selection of a medium for a call population to be cloned should be
made after assaying the plating efficiency and growth behavior of the cells in the
medium. Ham's F12K medium, in particular, has proven to be an excellent cloning
medium for many cell types and is used routinely in many laboratories.
mones. However, if cells grow ideally in medium supplemented with

another serum type, it should be used preferentially. For some cell
lines serum components may be replaced with specific mixtures of
hormones and growth factors. These techniques are presented in
this volume [6] and have been summarized elsewhere. 17Serum used
should be heat-inactivated for 30 min at 56.
Trypan blue solutionm0.1% solution in phosphate buffered saline
(PBS)
Enzyme solutions--one of the following:
Trypsin--0.1% trypsin in PBS (Gibco or Sigma)
Collagenase--O. 1% collagenase in PBS (Sigma). There is usually a
significant amount of proteolytic activity present in the collagenase purchased from most commercial suppliers.
Pronase--0.01% in PBS. A 1% stock solution is prepared and
stirred for several hours at 4 centrifuged, and the supernatant
fluid sterilized by Millipore filtration. The stock solution is diluted
to the final concentration prior to use. Particularly effective for
fibroblast-like cells; pronase is not as good for epithelial cells.
70% Ethanol solution
27 G. Sato, and L. Reid, in "Biochemistry and Mode of Action of Hormones"(H. V. Rickenberg, ed.), Chapter 7. Academic Press, New York, 1978.
156
aASl METHODS
[12]
C. Dilution Plating Techniques

In brief, the techniques include the preparation of single cell suspensions from monolayer cultures or from solid tissue pieces in an appropriate cloning media; dilution of the cell suspension to a concentration of 10
cells per milliliter of medium; addition of 0.1 ml of medium to a microtest
well; incubation of the cells for several weeks to permit growth of the
colonies; and transfer of individual colonies to tissue culture dishes or
flasks.
The original dilution plating techniques were aimed at adding a single
cell to a tissue culture plate or to a test tube. With the realization of the
cells' need to adjust the medium with "conditioning factors," the volume
of fluid was reduced to approximately 0.1 ml per cell by adding the cell to
a small capillary la,14 or to a microdrop under paraffin. TM More recently, the
technique has been modified by the introduction of microtest plates (FalconPlastics) containing 96 wells, each well holding up to 0.4 ml. One can
add a single cell in 0.1 ml of medium per well and thereby provide the
individual cell with its own culture dish and with a sufficiently small
amount of fluid to minimize the conditioning factor problem.19.~
Supplies Necessary for Dilution Plating Techniques

Falcon Microtest Plates
Tissue culture plates (35 mm diameter)
Cloning medium supplemented with 10% serum (or, where feasible,
specific hormones and growth factors)
Enzyme solution (trypsin is adequate for most cell types)
Procedures
1. The cell cultures to be cloned should be recently subcultured, be in
an active state of growth, and show no signs of ill health.
2. Cloning of freshly explanted normal or neoplastic tissue is difficult
and almost always requires use of feeder layers or conditioned media
(techniques presented in subsequent sections). Tissue or tumor to be
cloned should be sterilely dissected from the animal.
3. Single cell suspensions are prepared from monolayer cultures as
follows. Remove the medium from the culture plates and rinse the plates
twice with PBS. Add 1 ml of the trypsin solution to a 60-mm plate (2-3 ml
for larger plates or flasks) of cells and incubate the plates at 37 for 5-15
is A. Lwoff,R. Dulbecco, M. Vogt, and M. Lwoff,Virology 1, 128 (1955).
1~j. A. Robb, Science 170, 857 (1970).
~oj. A. Robb, in "Tissue Culture: Methods and Applications"(p. F. Kruse, Jr. and M. K.
Patterson, eds.), pp. 270-274. AcademicPress, New York, 1973.
[12]
CLONING
157
min. The plates should be watched carefully to ensure that the time of
contact with trypsin is minimized. As soon as the majority of cells are
rounded and mostly detached from the plates, squirt the ceils with the
conditioned medium supplemented with serum. Most serum contains a
trypsin inhibitor (chicken serum is an exception). By squirting the cells
with the medium, trypsin is inactivated and detachment of the cells from
the plates is completed. To completely inactivate trypsin requires a 9:1
dilution of the trypsin solution with serum-supplemented medium. One
can also remove trypsin by centrifuging the cells, removing the supernatant fluid, and resuspending the ceils in the cloning medium.
Single cell suspensions are prepared from solid tissues, either normal
or neoplastic 42 (see this volume [9]), as follows. The dissected tissue is
minced finely with scissors sterilized by dipping them in 95% ethanol.
Treat the mince with 0.1% collagenase (Sigma) in PBS at a ratio of 10 ml
of enzyme solution to 1 ml of mince. Enzymic digestion should take place
in a shaker bath at 37. After 15 min of enzymic treatment, allow the
chunks of tissue to settle to the bottom of the test tube, and pipette the
supernatant fluid, containing suspended cells, into another test tube. Centrifuge the cells in a desk-top centrifuge at 900 rpm. Resuspend the pellet
in cloning medium. The chunks and tissue mince can be repeatedly
treated with collagenase solution until they are totally disaggregated.
4. Count the cells by the trypan blue exclusion assay in order to
determine the number of viable cells (see this volume [11]!.
5. Dilute the cell suspensions, whether from the monolayer culture or
from the solid tissue dispersion, to a concentration of 10 cells/ml.
6. Use 0.1 ml of the medium containing the cells per well. Repeatedly
agitate the cell suspension to ensure homogeneous dispersion of the cells
throughout the medium. For rapid inoculation of cells into microtiter
wells, a Hamilton repeating dispenser (Hamilton Company, Whittier, California, # PB600-1) is useful. 2
7. Carefully observe each well and score those containing only one
cell.
8. Incubate the cultures in 37 incubators with 5% CO2 in air and ~vith
95% relative humidity.
9. During the growth of the colonies, it should be unnecessary to
change the medium. However, if colony formation is unusually slow, and
a medium change is required, the spent medium can be gently aspirated
from the cultures, and fresh medium added. Leave a residue of fluid over
the cells while changing medium to prevent desiccation of the cells.
10. When colonies of 500-600 cells have formed, usually in 2-3
weeks, remove the medium from the wells. Rinse twice with phosphate
buffered saline. Add 0.1 ml of trypsin solution per well. Allow the cells to
158
BASIC METHODS
[12]
round, and then add cloning medium with 10% serum to suspend the cells.
Transfer the cells to an appropriate tissue culture dish (35-60-mm plate)
or flask.
D. Cloning of Anchorage-Dependent Cells
The techniques in this category are slight variations of the ones described above in "Dilution Plating Techniques." The preparation and
dilution of single cell suspensions are the same as described there. The
cells are seeded onto 60-ram tissue culture dishes or on 60-ram petri dishes
coated with either collagen or fibrin. By seeding only 100-200 cells per
plate, one can obtain large colonies of cells that are sufficiently separated
to permit isolation of individual colonies.
The technique can be used only for cells that require anchorage and
spreading on a substrate in order to grow. This includes most normal cells
as well as many established cell cultures.
Supplies Necessary for Cloning of Anchorage-dependent Cells

Sterilized cloning rings: metal cylinders that are 6 mm in diameter, 12
mm high, and have 1-mm-thick walls
Sterilized Dow-Corning stopcock grease: Layer the grease on a glass
petri dish and sterilize. The grease may be kept sterile by taping on
the lid and keeping the dish in the hood under a UV lamp.
Forceps
60,mm tissue culture dishes or, for collagen- or fibrin-coated substrates, 60-ram petri dishes.
35-ram tissue culture dishes
Procedures
1. Cells vary as to the substrates to which they will anchor. Established cells lines will attach to plastic coated with polycations. However,
freshly explanted tissues usually prefer either collagen-coated plates or
fibrin-coated plates. Preparation of collagen-coated plates is given elsewhere in this volume [21]. Fibrin-coated plates are prepared as follows:
Prepare a solution of cloning medium containing 0.12 units of thrombin
per 100 ml of medium. Prepare a solution of 250 mg bovine fibrinogen, 800
mg sodium chloride, and 25 nag sodium citrate per liter of glass-distilled
water. Sterilize by filtration. Add 1 ml of the fibrinogen solution and 4 ml
of the medium containing thrombin to a tissue culture dish and mix rapidly
with a sterile Pasteur pipette. A clear gel will form within several minutes.
This technique was first introduced by Schindle& 1 who used it to replace
21 R. Schindler, M. Day, and G. A. Fischer, Cancer Res. 19, 47 (1959).
[12]
CLONING
159
agar cloning (described in the next section). Schindler suspended the cells
in fibrin and permitted them to form colonies shaped like morulae. However, one can also attach the cells to the surface of the fibrin and observe
colony formation.
2. Prepare single cell suspensions from monolayer cultures or from
tissues by the methods given in "Dilution Plating Techniques."
3. Depending on the plating efficiency of the cells, select an appropriate concentration of them to yield ultimately 1-10 colonies per plate. For
example, with a plating efficiency of 5-10%, one should plate 100 cells per
plate.
4. Dilute the cell suspensions to an appropriate density so as to add
the correct number of cells per 60-mm dish. The 60-mm dishes hold about
5 ml of medium.
5. The plates are incubated at 37 in 5% CO2 in air and with 95%
relative humidity.
6. The following day, single cells that are sufficiently isolated to permit easy cloning are marked by encircling them on the plastic with a
grease pencil or felt pen.
7. The cells are allowed to grow for several weeks into colonies containing approximately 500-1000 cells.
8. Remove the medium from the plates and rinse them twice with
PBS. Leave a residue of the PBS over the cells to prevent their drying.
9. With forceps sterilized by dipping them into 70% ethanol, dip previously sterilized cloning rings into the petri dish containing stopcock
grease. Place the grease-coated end down against the plastic so that the
ring encircles a colony of cells and so that the greased end forms a seal
with the plastic.
10. Add enzyme solution, e.g., 0.1% trypsin, to each of the rings.
Usually 1-2 drops of solution will fill the rings.
11. Incubate the cultures for 5-15 min at 37, observing the cultures
periodically with a phase microscope. When the cultures are rounded and
ready to detach, use a 9" Pasteur pipette and pipette the solution gently
up and down to complete the detachment process.
12. Pipette the solution of suspended cells into a 35-mm tissue culture
dish or 35-mm petri dish coated with collagen or fibrin. Add 2 ml of
cloning medium per plate.
E. Cloning of Cells that Are Anchorage-Independent for Growth
Most cells require anchorage to and spreading on a substratc for
growth. However, some cells have acquired the ability to grow in suspension cultures. Factors contributing to this cellular capability include viral
transformation (SV40, polyoma), infection of cells with mycoplasma, and
160
BASIC METHODS
[12]
malignant transformation. 22-25 Puck et al. 26 showed that cells which

are anchorage-independent for growth can be cloned in a medium solidified by low concentrations of agar. The method has been expanded into
a family of techniques for cloning of cells in semisolid media. The version
given below is that of Macpherson 27 who developed methods in which a
base layer of agar-medium is overlaid with a thin layer of agar-medium
containing the cells to be cloned. By layering the cells at the surface of the
base layer, they can be observed readily with an inverted phase
microscope.
Agar contains acidic and sulfated polysaccharides that are inhibitory
to most cells but not to many vitally or malignantly transformed cells. The
selective inhibition of the agar-medium is due both to the inability of cells
to anchor to the agar matrix and to the inhibitory polyanions within the
matrix. Growth of cells in agar-medium has become an assay for viral
and/or malignant transformation, since growth of the cells in agar is significantly correlated with tumorigenicity of the cells in immunologically
suppressed hosts such as the athymic nude mice. 25"2s'29 To reduce the
toxicity of the agar, the polyanions were reduced yielding a matrix referred to as agarose. Agarose is used routinely to clone cell types that are
anchorage independent but will not grow in agar due to its toxicity. As
noted in Section D, one may also use fibrin gels for cloning cells in suspension. However, cells attach to fibrin even though they do not spread on it
and so are got strictly anchorage-independent. 3 Normal cells, as well as
transformed ones, will form colonies when embedded within a fibrin
matrix.
Supplies Necessary for Cloning Anchorage-Independent Cells

Agar: prepare a 1.25% agar stock from Difco Bacto-agar. Preparative
details are given below.
Cloning medium: prepare a double-strength stock solution. Sterilize
by filtration.
Tryptose phosphate broth: Sterilize by autoclaving.
22 I. Macpherson, and L. Montagnier, Virology 23, 291 (1964).
23 p. H. Black, Virology 28, 760 (1966).
24 I. Macpherson, J. Cell Sci. 1, 145 (1966).
25 V. Freedman, and S. Shin, Cell 3, 355 (1974).
ze T. Puck, P. I. Marcus, and S. J. Cieciura, J. Exp. Med. 103, 273 (1956).
27 I. Macpherson, in "Tissue Culture: Methods and Applications" (P. F. Kruse, Jr. and
zs H. P. Klinger, S. Shin, and V. H. Freedman, Cytogenet. Cell Genet. 17, 185 (1976).
zg C. D. Stiles, W. Desmond, L. Chuman, G. Sato, and M. Saier, Jr., Cancer Res. 36, 3300
(1976).
ao L. Reid, C. Stiles, M. Rindler, and M. Saier, submitted for publication.
[12]
CLONING
161
60-mm petri dishes

35-mm tissue culture dishes
Procedures
1. Agar stock. Add 12.5 g of Difco agar to 800 ml of boiling glassdistilled water and heat until the agar dissolves. Make up to 1 liter with
water. Sterilize by autoclaving 80-ml portions of the agar solution in
screw-capped bottles of 250-ml capacity. Autoclave under low pressure
and with the screw caps loosened. Allow the bottled solutions to come to
room temperature. Then tighten the bottle caps. Store at room temperature.
2. Melt the agar stock solution by placing the bottle in boiling water
until the agar liquefies. Cool for several minutes and place in a water bath
at 44 so that the hot water is above the level of the agar.
3. Prepare the complete agar medium by mixing 80 ml of the cloning
medium (double strength) with 20 ml tryptose broth and 20 ml serum in
the bottle with the agar. Mix by swirling.
4. Pipette 1-2 ml of medium into 60-ram petri dishes or tissue culture
dishes and swirl the plate to evenly coat the solution over the bottom of
the dish. Allow the base layers to solidify (10-15 min).
5. Prepare a cell suspension as described under "Dilution Plating
Techniques."
6. Dilute the cell suspension with cloning medium to a concentration
that is 3-fold that of the final, desired concentration. Mix 1 volume of the
cell suspension with 2 volumes of the complete agar solution.
7. Spread the cells over the base layer and swirl the plate to assure
even spreading.
8. Incubate the cultures at 37 in an atmosphere of 5% CO2 in air and
with 95% relative humidity.
9. Allow the cells to form colonies of 500-1000 cells.
10. Colonies are transferred by sucking them from the agar with a
Pasteur pipette and transferring to a sterile test tube containing 1 ml
of medium. The colony of cells is pipetted to dissociate it from the
agar.
11. The cell suspension is transferred to a 35-mm dish to which is
added 2 ml of cloning medium.
F. Preparation of Cloning Media, Feeder Layers, and Conditioned Media
A listing of some of the media used for cloning various cell types is
provided in Table I. All of them employ a basal medium enriched in
162
BASIC METHODS
[12]
particular nutrients, especially those, such as pyruvate and nohessential

amino acids, that are lost from the cells into the medium.
Feeder Layers
The low plating efficiency of some cells, especially freshly explanted
cells, prevents or seriously complicates the cloning procedure. As noted,
the low plating efficiency is due in part to the loss of essential nutrients
from the isolated cells. However, in the special cloning media listed in
Table I these nutrients are provided. The other factors, available in highdensity cultures of cells but not in low density cell cultures, are "chemical
messengers" secreted by tho ceils and regulating survival and/or growth.
The chemical messengers playing roles in colony formation are referred to
as Colony Stimulating Activity factors (CSA). To provide such factors, it is
necessary to co-culture the cells to be cloned with a feeder layer of the
same or another ceil type. Many established cell lines requiting feeder
layers under clonal conditions can use feeder layers of the same cell type.
Freshly explanted epithelial cells, whether from normal or neoplastic tissue, often do better when cloned over feeder layers of mesenchymal ceils
(L. C. M. Reid, unpublished data). Many of the CSA factors have been or
are being purified in a number of laboratories. Several of the bettercharacterized CSA are listed in Table II. al-a7 Except for some anaplastic
cell lines, supplementation with only the CSA factors is inadequate for
colony formation, suggesting that other factors are necessary. Corroborating this interpretation are the findings by Sato and his colleagues
that multiple hormones, growth factors, and transfer factors are essential
as supplements to basal media to attain growth of many ceil types 17 (see
this volume [6]).
Supplies for Preparing Feeder Layers or Conditioned Medium

Microtiter wells (Falcon)
35-mm dishes
60-mm dishes
Cloning medium
31 N.
3~ G.
33 T.
a4 T.
a3 G.
38 T.
27 R.
Dulak and H. Temin, J. Cell. Physiol. 81, 161 (1973).

Smith and H. Temin, J. Cell. Physiol. 84, 181 (1974).
Chen, J. Mealey, Jr., and R. L. Campbell, J. Natl. Cancer Inst. 55, 1275 (1975).
Bradley and M. Sumner, Exp. Biol. Med. Sci. 46, 607 (1968).
Price, E. McCullough, and J. Till, Blood 42, 341 (1973).
Landau and L. Sachs, Proc. Natl. Acad. Sci. U.S.A. 68, 2540 (1971).
Stanley and P. Heard, J. Biol. Chem. 252, 4305 (1977).
[12]
CLONING
163
.8
"
.~
"
<
<
<
c~
'~
..~
r,
0
Z
Z
o
c~
0
Z
0
,.d
0
r..)
0
"~V-
~ra
co
0
Z
<
~o
i ~- i ~
<
m
r,..)
,~
.~ ..~
z
"~
ell
o -~
rj
o .=
r..)
164
BASIC METHODS
[13]
Procedures
1. Feeder layers may be prepared from any cell population desired.

Single cell suspensions are obtained from monolayer cultures or from
tissues as described for methods for dilution plating.
2. The suspensions are added to whatever type of dish will be used in
the cloning procedure. For microtiter dishes, add 50-100 cells per well;
for 35-mm dishes, add 104 cells; for 60-mm dishes, add l&-10 e cells per
plate.
3. Incubate the cells at 37 with 5% CO2 in aA air atmosphere and with
95% relative humidity. Allow the cells to attach to the plates and grow for
1 or 2 days.
4. Irradiate the plates with 4000-5000 rad (cobalt-60 or cesium-135) to
eliminate all mitotic activity. Mitomycin C has been used to mitotically
kill the feeder layer cells; it is used less frequently now, since it has been
found to adversely affect the cells being cloned.
5. The method of addition of the cells depends on the cloning technique that is used:
a . Dilution technique: The single cells are added directly to wells
containing confluent feeder layer.
b. Anchorage-dependent technique: Cells to be cloned may be added
directly to the feeder layer. For those cells that grow best on collagen or
fibrin substrates, polymerize the collagen or fibrin over the feeder layer.
The cells to be cloned are then added to the surface of the substrate.
c. Anchorage-independent technique: As in (b), the agar or agarose
matrix is poured as an overlay onto the feeder layer. Cells to be cloned are
then added to the surface of the base agar or agarose layer.
6. For conditioned medium, l-2-day-old medium from confluent
feeder layer cultures is removed and mixed at a 3:1 ratio with fresh
medium. The medium is used as the cloning medium and changed every
4 - 5 days.
[13] C e l l C u l t u r e C h a r a c t e r i z a t i o n : M o n i t o r i n g for C e l l
Identification 1
By W A R D D.
PETERSON, JR., WILLIAM
F.
SIMPSON,
and
BHARATI HUKKU
Techniques and cell culture support systems now available enable any
bioscience laboratory to successfully propagate animal cells in vitro.
While the relative ease with which cell cultures can now be started or
Supported in part by National Cancer Institute Contract NO1 CP 3-3333.

ISBN 0-I2-181958-2
[13]
CELL CULTURE CHARACTERIZATION
165
maintained is a great advantage, it frequently happens that a requirement

for one cell line or strain in a particular research inquiry is followed by a
need for a second, third, and fourth cell line. Soon, either by acquisition
or initiation, the number of cell lines escalates. An escalation in the number of cell lines in a laboratory beyond one should be accompanied by the
institution of cell identification and monitoring regimens.
The need for cell line monitoring arises from more than just numbers
of cell lines. Cells are passaged, cocultivated, transformed, and/or transplanted for numerous investigative purposes. In each of these operations,
there is a need to know the status of the cells in vitro for at least these
reasons: to annotate changes in cells upon passage; to determine the
species of the predominant cell population in a cocultivated culture; to
confirm the fact of cell transformation; or to establish whether transplantation resulted in a tumor of cell or host origin. Beyond that, cell monitoring apprehends such technical mishaps as mislabeling or inadvertent mixing of cells that may occur in each of the aforementioned operations.
Monitoring also is necessary when cultures are exchanged between
laboratories. Cultures often are sent out to another investigator without
being newly recharacterized, and even more frequently they are received
and worked with by the recipient investigator who also fails to examine
them for identity. In this way, subsequent conforming or diverging results
may perpetuate misinformation in very costly and time-wasting ways.
Numerous reports during the past 25 years have documented instances of cross-contamination between cell cultures. The subject has
been reviewed more than once, 2'3 and it has been recognized by the establishment of repositories for characterized cell populations. 4,5 More recent
publications indicate that interspecies and intraspecies contamination of
cell cultures may be as high as 20-30%. 6,r While these recent reports may
represent findings from a somewhat biased sample, in that at least a portion of the cell lines were sent for examination because of suspicions by
the submitting investigator, it is just as likely that there are many crosscontaminated cultures about which no suspicions are held. In short, the
problem of cross-contamination is a substantive and persisting problem
that may afflict any laboratory using cell cultures. Monitoring cell cqltures
for identity is a necessary but often neglected adjunct control in the
laboratory.
2 C. S. Stulberg, in "Contamination in Tissue Culture" (J. Fogh, ed.), p. 1. Academic Press,
New York, 1973.
3 p. p. Ludovici and N. R. Holmgren, Methods Cell Biol., 6, 143 (1973).
4 American Type Culture Collection, Rockville, Maryland.
5 Human Genetic Mutant Cell Repository. Institute for Medical Research, Camden, New
Jersey.
8 C. S. Stulberg, W. D. Peterson, Jr., and W. F. Simpson, Am. J. Hematol. 1, 237 (1976).
7 W. A. Nelson-Rees and R. R. Flandermeyer, Science 195, 1343 (1977).
166
BASIC METHODS
[13]
Selected for description here are three approaches to cell monitoring

that utilize genetically stable phenotypic expressions of cells. Cells may
be identified by isozymes, chromosomes, or species-specific antigens, and
of course by information derived in combination from all three of these
different cell markers. Each of the procedures to be described yields rapid
results and reasonable certainty of accuracy when properly controlled. As
presented, each method has limitations. However, each can be extended
to give rather precise intraspecific information about a cell line, as will be
briefly indicated in each section. Each method also has certain virtues that
will enable a laboratory to quickly adopt one or more of them because of
compatibility with the laboratory's research direction and available
equipment.
I. Monitoring of Cells by Isozymes
Extraction of enzymes from cells, and comparison of electrophoretic
mobilities of enzymes on zymograms, comprise a convenient method for
determining species. By comparing electrophoretic mobilities of only two
or three enzymes, species can be readily determined. Further, by choosing isozymes polymorphic in a species, intraspecies characterizations can
also be carded out as was demonstrated by Gartler, s who chose glucose
6-phosphate dehydrogenase and phosphoglucomutase to examine a series
of human cell lines. He demonstrated contamination of many human cell
lines by the cell line HeLa,9 on the evidence that the dehydrogenase Type
A and mutase Type11 and Types1, characteristic of HeLa cells, were also
found in cultures whose presumptive donors could not possess those
forms of the two isozymes. For that reason, and since human cells are
cultured extensively, a procedure for determining glucose 6-phosphate
dehydrogenase mobility is described here. 10.~ The finding of Type A mobility may be helpful in raising suspicions about a human cell line. 12Also
described is the determination of lactate dehydrogenase isozyme
mobilities. ~3,14Both isoenzymes are generally abundant in cells, and the
s S. M. Gartler, Natl. Cancer Inst., Monogr. 36, 167 (1967).
9 W. F. Scherer, J. T. Syverton, and G. O. Gey, J. Exp. Med. 97, 695 (1953).
~0 M. C. Rattazzi, L. F. Bernini, G. Fiovelli, and P. M. Mannucci, Nature (London) 213, 79
(1967).
H W. D. Peterson, Jr., C. S. Stulberg, N. K. Swanborg, and A. R. Robinson, Proc. Soc.
Exp. Biol. Med. 128, 772 (1968).
n W. A. Nelson-Rees and R. R. Flandcrmeyer, Science 191, 96 (1976).
13 E. S. Vesell, J. Philip, and A. G. Beam, J. Exp. Med. 116, 797 (1962).
~4 F. Montes De Oca, M. L. Macy, and J. E. Shannon, Proc. Soc. Exp. Biol. Med. 132, 462
(1969).
[13]
167
procedures used are readily carded out. Use of both isozymes give sufficient information to determine most cell species. 15
A. Preparation of Isozyme-containing Extracts
Reagents
Trypsin-EDTA solution:
1:300 Trypsin, crude, 2.5 g/1
EDTA (ethylenediaminetetraacetic acid), 0.2 g/1
Glucose, 1.0 g/1
NaC1, 1.0 g/1
NaHCO3, 1.89 g/1
Procedure
1. Grow 5 x 106 cells in appropriate medium and disperse with
trypsin-EDTA if it is an adherent cell line. When cells are monodispersed,
stop the action of trypsin-EDTA by adding growth medium and collect
the cells in a conical centrifuge tube.
2. Centrifuge the cell suspension at 1500 rpm for 10 min, remove the
supernate, and wash the cells 3 times with 10 volumes of 0.9% NaC1.
3. After the last washing, resuspend the cells in an equal volume of
0.9% NaC1. Freeze the cell suspension by placing the centrifuge tube in a
methanol---dry ice solution. When frozen remove the tube and thaw the
cell suspension slowly at room temperature. Repeat the procedure 3
times. This rapid freeze-thaw procedure will disrupt cells.
4. Centrifuge the suspension at 3000 rpm for 10 min and draw up the
supernatant liquid in a Pasteur pipette. Place the sample in a closable 12 x
75 mm tube for storage at - 2 0 . The isozymes are stable for several weeks
when stored in this manner.
B. Electrophoresis of Sample
Reagents
Glucose 6-phosphate dehydrogenase buffer, 0.075 M Tris-citric acid,
pH 7.5:
Tris, 9.08 g/1
EDTA, 1.49 g/1
Citric acid, 15.75 g/1
Add 20 ml of citric acid solution to 1 liter of Tris-EDTA solution.
15j. E. Shannon and M. L. Macy, in "Tissue Culture: Methods and Application" (P. F.
Kruse, Jr. and M. K. Patterson, Jr., eds.), p. 804. AcademicPress, New York, 1973.
168
BASIC METHODS
[13]
Lactate dehydrogenase buffer, 0.05 M barbital, pH 8.8:

Sodium barbital, 10.31 g/1
EDTA, 0.35 g/1
Adjust to pH 8.8 with 1 N NC1.
Procedure
1. Pour the buffer solution suitable for electrophoresis of the isozyme
into all chambers of the electrophoretic apparatus to appropriate volume.
Inexpensive low-voltage electrophoretic equipment TM is commercially
available from a number of manufacturers.
2. Thaw samples and prepare for application to electrophoretic support medium. Inexpensive electrophoretic support media are commercially available as acetate strips, cellulose acetate gel strips, or agarose gel
film. Each requires a somewhat different procedure for application. A
convenient system is that of the agarose gel film17 with eight preformed
slots to which 1-/zl samples can be added by syringe. The application is
simple, and eight samples can be run at a time for comparison of electrophoretic mobility. After application, the film is inverted so that the
agarose side is down, and the sample origin is placed at the cathodal side
of the box. Wicking of the agarose to the buffer is accomplished by filter
paper wicks to complete the circuit across the substrate. Power is turned
on, and the sample is electrophoresed at 6 mA (50-150 V) for 90 min at
room temperature.
3. After. electrophoresis, remove the film preparatory to staining.
Empty the electrophoresis chamber of buffer to restore ionic balance
throughout the buffer solution before using it in a subsequent electrophoresis run.
C. Staining of Isoenzymes
Reagents
Glucose 6-phosphate dehydrogenase stain:
Amount/
solution
NADP
Glucose &phosphate
Phenazine methosulfate (PMS)
Nitro blue tetrazolium (NBT)
Cobaltous chloride
Tris. HC1 (pH 8.6)
10 mg/ml
25 mg/ml
2 mg/ml
2 mg/ml
0.5 M
0.075 M
Proportion
in stain
0.4
0.6
1.0
1.0
0.4
5.0
ml
ml
ml
ml
rnl
ml
toj. K. Turner model 310-211 with electrophoresis chamber 310-120 is used in this
laboratory.
~7 Universal electrophoresis film, agarose; Coming ACI, Palo Alto, California.
[13]
169
PMS solution is made fresh each day and should not be exposed to
light. NAD, glucose 6-phosphate, and NBT solutions are stable at 4
for 1 week. The stain is bright light yellow and will not react if the
color darkens. Prepare stain and incubate for 15 min at 37 immediately
before use.
L D H stain:
Amount/
solution
Proportion
in stain
NAD
10 mg
10 mg
Sodium lactate
1M
1.0 ml
Nitro blue tetrazolium
1 mg/ml
3.0 ml
Phenazine methosulfate
1 mg/ml
0.3 ml
Barbital buffer, pH 8.8
0.5 M
1.0 ml
Stain does not require preincubation before use.
Procedure
1. Prepare stain as indicated. Pour into a clean shallow staining dish.
Remove electrophoresis film, and place agarose side directly onto stain so
that stain comes in contact with area to which isoenzymes have migrated.
2. Incubate the staining dish in a humidified chamber at 37. Color
should be observed within 5 min, and the staining reaction can be developed for 20 min. The staining reaction should be followed every few
minutes so that overstaining, with consequent blurring of reaction bands,
does not occur.
3. When the staining reaction is satisfactory, remove the electrophoresis film from the staining dish and place it in 1:40 formaldehyde
for 10 min to stop the reaction.
4. Comparative electrophoretic mobilities can be determined immediately by removing the film from the formaldehyde solution, shaking the
film to remove moisture, and overlaying it on lined or graph paper. By
measuring the distance from the origin to the reactive enzyme site, and by
noting the patterns, comparison of enzyme mobilities with known and test
isoenzyme extracts can be carried out.
5. A permanent record for the electrophoresis run can be made by
drying the agarose film at 37 overnight or at 60 for 2 hr. The stained film
serves as a record itself, or it may be photographed. To prevent scratching
or dislodging of the electrophoresis film with time, an adherent clear
plastic sheet may be placed over the reactive side of the film.
Comment. The procedure presented above utilizes inexpensive
equipment, offers simplicity of operation, and gives rapid results. The
entire procedure will take 2-3 hr, and eight different samples can be
accommodated. The resolution is not as great as that achieved by starch
gel or acrylamide substrates, nor is its application to other enzyme sys-
170
BASIC METHODS
[13]
tems as facile when direct visual staining is utilized. If other electrophoretic systems are already in use, it will require very little additional work to
develop appropriate cell monitoring regimens.
O'Brien et al. is have shown that appropriate selection of additional
isoenzymes polymorphic in a species can be used to develop a genetic
signature for each cell line. Seven isozymes were used to study 30 human
cell lines. They showed that the enzyme profiles obtained distinguished
between 19 human glucose 6-phosphate dehydrogenase Type B cell lines
and 5 dehydrogenase Type A cell lines, demonstrated that two dehydrogenase Type B cell lines thought to be different were actually the same
line, anti demonstrated that three cell lines suspected of being HeLa cells
by other criteria were confirmed as such since all three had a genetic
signature identical to that of HeLa. The probability of two different cell
lines having the same signature with the seven isozymes by chance alone
was calculated to be less than 5%. Extension of the concept seems possible with regard to other species as a means for intraspecies identification
of individual cell lines.
II. Monitoring of Cells by Chromosomal Examination
Examination of chromosomes is the definitive method for determining
cell species, and if carried out in a thorough way it is one of the definitive
methods for individual cell line identification. The methodology for this is
a fairly complicated and time-consuming task--so much so that cytogenetics tends to be completely avoided by those not in that area of study. In
fact, however, it is not difficult to quickly obtain basic information on cell
cultures by chromosomal examination. Metaphase preparations can be
prepared, stained, and examined in a short time. In terms of cell culture
monitoring, these procedures are extremely useful for determining the
current status of a cell culture. Familiarity with the chromosomal complement of the cells cultured in a laboratory will enable one to quickly
denote similarities or differences in a cell line with passage and time in
culture. On the basis of such findings, a decision to obtain more expert
analysis can be made (see also [27]).
A. Preparation of Cell Metaphases
Reagents
Colcemid (N-deacetyl-N-methylcolchicine) solution:

Colcemid, TM 100 mg
is S. J. O'Brien, G. Kleiner, R. Olson, and J. E. Shannon, Science 19S, 1345 (1977).
1, CIBA Pharmaceutical Company, Summit, New Jersey. Also available as prepared solution from other companies.
[13]
171
95% ethanol, 5.3 g

Propylene glycol, 10 g
Na2HOP4.12 HsO, 6.2 g
NaH2PO4.2 H20, 0.15 g
Distilled water, q.s. 100 ml
Dissolve Colcemid in ethanol and add propylene glycol. Add phosphate buffer to the solution of dissolved Colcemid. Sterilize by autoclaving at 10 lb/in 2for 30 min. Disperse in 1-ml aliquots and store at 4. For use,
dilute 1 ml in 24 ml of Hank's balanced salt solution (HBSS).
Fixative solution:
Absolute methanol, 3 volumes
Glacial acetic acid, 1 volume
Procedure
1. Establish culture with actively dividing cells as in logarithmic

growth phase. Adherent cells are usually satisfactory 2 days after transfer, when cells cover about of the flask surface. Suspension cultures are
satisfactory the day following feeding.
2. Add to the culture one drop of diluted Colcemid solution per 25 ml
of growth medium to arrest the dividing cells in metaphase. Following a
1-3 hr incubation period, depending on the rate at which the cells multiply, the Colcemid-containing medium is removed, and 5 ml are placed in a
centrifuge tube. Five milliliters of trypsin-EDTA solution are added to the
culture flask to disperse an adherent monolayer of cells.
3. When the cells are dispersed, pipette and transfer them to the centrifuge tube with Colcemid medium. Trypsin-EDTA action on the cells is
stopped. Centrifuge the cell suspension at 1500 rpm for 10 min. Remove
all but 1 ml of supernatant from the cell pellet. Resuspend the cells by
gentle repeated pipetting in a manner to avoid foam formation.
4. To 1 ml of cell suspension, add 4 ml 9 f distilled water and suspend
the cells in this hypotonic solution for 8-10 min at room temperature. For
human lymphoblast cultures, better preparations are obtained by treatment with hypotonic 37.5 mM KC1 for 18 min. The time in hypotonic
solution is critical for obtaining good metaphase preparations, and that
given here is only representative. Since each cell culture may require
somewhat different hypotonic treatment periods, precise timing can only
be determined by trial.
5. Add about 0.7 ml of fixative solution drop by drop, while gently
agitating the tube to obtain good mixing. Centrifuge the suspension at
500-800 rpm for 5 min.
6. Remove the supernatant from the cell pellet and add 0.2 ml of
fixative solution (approximately 3-4 times the volume of the pellet).
172
BASIC METHODS
[13]
7. Resuspend the cells by gently bubbling air through a Pasteur

pipette, and add sufficient fixative solution to bring the cell suspension
volume to about 5 ml. Leave the cells in fixative solution for 10 min, and
centrifuge the cell suspension as before.
8. Remove the supernatant, and resuspend the cells in 0.5 ml of fixative solution.
9. Take up a small portion of the suspension in a Pasteur pipette. From
a distance of 30-40 cm above a horizontally held, cold, wet slide, drop
two drops of cell suspension at different locations, on the slide. Blow once
down the length of the slide which is held 5-10 cm from the mouth, in
order to spread the cells on the slide and to remove excess fixative. Then
dry the slide on a hot plate at 60 for 2 min, or air dry by placing the slide
narrow dimension down against a rack for about 30 min.
10. Examine the slide with a 10 objective for cell nuclei and for
chromosomes spread in metaphase plates. If there are judged to be too
many cells for the matephases to be well spread, that is, the metaphases
are overlapped, dilute the suspension appropriately with fixative solution
and make another trial slide.
11. Once a satisfactory trial slide is obtained, additional slides are
made as needed in the same manner. Slides are marked for identity and
dried as before.
B. S t a i n i n g a n d E v a l u a t i o n of M e t a p h a s e s
Reagents
Giemsa stain 2:
Giemsa powder, 7g
Glycerol, 482 ml
Methanol (absolute), 462 ml
Mix Giemsa powder and glycerol. Place solution in a 60 oven for 2 hr.
Allow solution to cool and add methanol. Filter through Whatman #1
paper as used.
Giemsa buffer, pH 7.0:
Na2HPO4 (anhydrous), 9.5 g/1
NaH2PO4H20, 9.2 g/1
Add 61.1 ml of dibasic phosphate solution and 38.9 ml of monobasic
phosphate solution to 900 ml of distilled water. The working solution of
z0 Harleco Company, Philadelphia, Pennsylvania. Also available as prepared working solution from other companies.
[13]
173
Giemsa stain is prepared by adding Giemsa stain to Giemsa buffer in a

ratio of 1:25.
Procedure
1. Place slides in a glass staining jar containing 1 N HCI prewarmed in

a 56 water bath and hydrolyze for 11 min.
2. Wash slides in the staining jar with running tap water for 10 min.
Then rinse with 1% Giemsa buffer in distilled water to neutralize any
remaining acid.
3. Transfer the slides to a staining jar and stain for 11 min with the
working dilution of Giemsa stain.
4. Remove the slides and dip them into 1% Giemsa buffer in distilled
water to rinse.
5. Shake the rinsed slides to remove excess water, wipe the back of
the slide with paper toweling, and firmly blot the metaphase side with
bibulous paper. Place the slide on a hot plate at 60 for 2 min, or allow the
slide to dry overnight at room temperature.
6. When slides are dry, add one drop ofPennount 21 at each end of a 22
40 mm #1 coverslip lying on paper toweling. Invert the slide, metaphase side down, and place on the coverslip. Squeeze excess Permount out
from between the coverslip and slide by pressing firmly on the slide. Do not
change the position of the slide with respect to the coverslip while pressing
down. Allow Permount to dry for 15 min prior to examination.
7. Examine slides microscopically by first using a 16 objective to
locate well spread metaphases. Change to a 100 oil immersion objective
to count chromosomes in the metaphase plate. A sample of at least 15
metaphases should be counted to determine the chromosomal modal
number. At the same time chromosomes are evaluated morphologically
for species characteristics. In the beginning, it will be convenient to make
a photographic record of several metaphases of a preparation to compare
chromosome morphology with published karyotypes. 22 With experience,
counting and sorting out of chromosomes can be done microscopically in
a short time. The total amount of actual time in preparation and examination of metaphases approximates 4 hr.
Comment. It is certain that chromosomes will be visible if the above
method for obtaining stained metaphases is followed. From these, an
evaluation of the status of the cell culture can be made to ascertain the
species of most cell cultures, degree of ploidy, and notation of abnor21 Fisher Scientific Company, Fairlawn, New Jersey.
22 T. C. Hsu and K. Benirschke, " A n Atlas of Mammalian Chromosomes," Vols. 1-10.
Springer-Verlag, Berlin and New York, 1967-1977.
174
BASIC METHODS
[13]
malities. With only a relatively small investment in time and materials

much information can be obtained about a cell culture in this way.
However, a chromosome examination in the manner described will
not fully define a cell culture to the extent now possible using various
chromosome banding techniques 23-2s (see also this volume [27]). Useful
intraspecies chromosome markers, such as the human Y chromosome, 26
H e L a or other human tumor cell marker chromosomesfl 7-29 or a number
of chromosome markers found in cell lines of other species a9 can only be
identified with certainty from banded preparations. Even some interspecies identifications require chromosome banding. For example,
banding of the number 2 chromosomes distinquishes between the
karyotypes of rhesus monkey and baboon. 3t Banding techniques strongly
enhance monitoring and cell identification capability.
III. Monitoring of Cells by Species-Specific Antigen- Antibody Reaction
Membrane antigens that are species specific can be used as a marker of
cell identity. Antisera prepared against membranes can be used to determine species of cells in culture by several serologic methods. The method
described here employs fluorescence as an indicator of a positive specific
reaction. It is a useful method because the proportion of different species
of cells in mixed culture populations can be determined.
A. Preparation of Species-Specific Antiserum
1. Satisfactory antigen sources can be secured by using red blood
cells, primary cell strains initiated for the purpose, or well-characterized
cell lines as are obtainable from the American Type Culture Collection. 4
Cells can be grown at a rate sufficient to satisfy the inoculation schedule.
23 T. Casperson, L. Zech, and C. Johansson, Exp. Cell Res. 62, 490 (1970).
24 M. Seabright, Lancet 2, 971 (1971).
25 H. A. Lubs, W. H. McKenzie, S. R. Patil, and S. Merrick, Methods Cell Biol. 6, p. 345
(1973).
2e W. D. Peterson, Jr., W. F. Simpson, P. E. Ecklund, and C. S. Stulberg, Nature (London),
N e w Biol. 2,12, 22 (1973).
27 O. J. Miller, D. A. Miller, P. W. Allerdice, V. G. Dev, and M. S. Grewal, Cytogenetics 10,
338 (1971).
2s W. A. Nelson-Rees, R. R. Flandermeyer, and P. K. Hawthorne, Science 184, 1093 (1974).
29 W. A. Nelson-Rees, R. R. Flandermeyer, and P. K. Hawthorne, Int. J. Cancer 16, 74
(1975).
30 W. A. Nelson-Rees, in "Proceedings of Workshop on Cell Substrates for Vaccine Production" (J. Petrieciani, chin.), p. 33. NIH, Bethosda, Maryland, 1976.
31 D. S. Markarjan, I . A . Gvaramia, and N.V. Shonia, Mature. Chromosomes Newsl.
17, 5 (1976).
[13]
175
However, a convenient procedure is to grow the cells to approximately 6

107 cells, disperse the cells by trypsin-EDTA solution, and collect and
centrifuge them into a pellet.
2. Resuspend the cells in medium containing 10% serum; count the
cells, and adjust the cell concentration of part of the suspension to 3 106
cells/ml, sufficient for five ampules. The remainder, adjusted to 6 106
cells/ml, is distributed to at least seven ampules. The cell suspensions are
then frozen at - 2 0 .
3. When ready to begin the immunization procedure, rapidly thaw
sufficient ampules each day of inoculation to have sufficient cells according to the following schedule: 3 x 105 cells on day 1; 6 105 cells on day 4;
1.5 106 cells on day 7; 3 x 106 cells on day 11; and 6 10s cells on days
14, 17, and 21. Cells are centrifuged and washed 3 times with 10 volumes
of HBSS and resuspended in 1 ml of HBSS for each dose. Two rabbits are
injected in the marginal ear vein with the same cell density.
4. Bleed rabbits by cardiac puncture 7 days following the last inoculation. Forty to eighty milliliters of blood are obtained, which are allowed to
clot at room temperature for 2 hr.
5. Separate the serum from the clot, and pool the sera collected from
each animal. Store sera in 5-ml portions at - 2 0 .
6. For antiserum of higher specificity, or for antiserum against rabbit
cells, inoculation of guinea pigs is satisfactory. Six guinea pigs each are
injected intraperitoneally with 1 ml of 6 106 washed cells and subcutaneously in four sites with 0.25 ml of 1.5 106 washed cells suspended
in Freund's complete adjuvant. 32 The intraperitoneal injection is repeated
21 days later, and the animals are bled by cardiac puncture on day 28. The
serum is processed and stored as indicated above.
B. Flourescent Labeling of Species-Specific Antiserum s3
Reagent
0.5 M carbonate-bicarbonate buffer, pH 9.0:

A. Na2CO3, 53 g/1
B. NaHCO3, 42 g/1
Mix 1 part A with 4 parts B.
Procedure
1. Thaw 10 ml of antiserum and add 20 ml of 27% sodium sulphate.

Incubate the solution for 16 hr at 37. Gamma globulins will precipitate
32Difco Laboratories, Detroit, Michigan.
33C. S. Stulberg, W. F. Simpson, and L. Berman, Proc. Soc. Exp. Biol. Med. 108, 343
(1961).
176
BASIC METHODS
[13]
during this period. Centrifuge the suspension at 3000 rpm for 30 min and
remove the supernatant liquid.
2. Add 3 ml of distilled water to the pellet of gamma globulins to
dissolve them and transfer the solution to dialysis tubing. Tightly knot the
dialysis tubing and dialyze against two changes of 1 liter of distilled water
for 30 min each. Dialyze further against six changes of 0.9% NaC1 during
an 18-hr period.
3. After removing the protein solution from dialysis tubing, sample 0.1
ml to determine the protein concentration; the Biuret method a4 is satisfactory. On the basis of the protein assay, weigh sufficient fluorescein
isothiocyanate (FITC) to have 0.025 mg FITC/mg protein. Mix this with
nine parts by weight of Celite 501 powder, a5
4. The dialyzate is adjusted to pH 9.0 by the addition of 0.5 M
carbonate-bicarbonate buffer (1 volume to 9 volumes dialyzate), and the
FITC-Celite 501 mixture is added slowly with gentle agitation of the
suspension. Continue to gently shake the suspension for 10 min during
which time the FITC becomes bound to gamma globulin. Centrifuge the
suspension at 1500 rpm for 5 min and remove the supernate containing
FITC-coupled globulin.
5. Because there is free FITC in the protein solution the preparation is
charged onto a Sephadex G-25 (fine) column (15 1 cm). Collect the
effluent from the column until it appears colorless. Add an additional 10
ml of saline to the column to remove remaining globulin. Store the
coupled globulin in 0.5-ml portions in ampules at - 2 0 .
C. Titering and Use of Coupled Antiserum
Reagents
Phosphate buffered saline (PBS) pH 7.5:
NaH2PO,H20, 1.38 g/1
NaCI, 9.0 g/1
Adjust to pH 7.5 with 1 N NaOH.
Procedure
1. Grow approximately 3 x 10n cells of the same cell line or a line of
the same species as that against which antiserum was prepared. Disperse
the cells by trypsin-EDTA, if an adherent cell line, and collect the
monodispersed cells. Neutralize the dispersing agent activity by adding 5
ml of growth medium containing serum and centrifuge the cell suspension
s4 j. R. M a r r a c k a n d H. H o c h , J. Clin. Pathol. 2, 161 (1949).
s5 J o h n s Manville Co., Waterville, Ohio.
[13]
177
at 1500 rpm for 5 min. Wash the pelleted cells 3 times with 10 volumes of
PBS. Adjust the concentration of cells to 3 x 106 cells/ml.
2. Thaw an ampule of antiserum and prepare a series of 2-fold dilutions, from undiluted to 1:128 inclusive of antiserum in PBS. Place 0.1 ml
of each dilution of antiserum in a 12 75 mm tube and add 0.1 ml of cell
suspension to each tube.
3. Place the eight tubes on a low-speed rotary shaker to mix the cellantiserum suspension thoroughly for 30 min at room temperature.
4. Add 1.0 ml of PBS to each tube, shake, and centrifuge at 3000 rpm
for 2 min. Remove the supernatant and wash the cell-antiserum mixture
again in the same manner. Remove the supernatant and resuspend the
cells in 2 drops of PBS.
5. Place a large drop of the resuspended cells on a glass slide and
carefully float a 22 x 30 mm #1 coverslip on the drop so as to avoid
trapping air bubbles or crushing the cells in suspension.
6. Place the slide on the stage of a fluorescent microscope and observe
the preparation with a x 16 or 25 objective by dark-field illumination.
When cells are located, switch to fluorescent illumination. Cells reacting
specifically have a halo of peripheral bright green fluorescence delimiting
the cell membrane. Nonreactive cells are difficult to discern because of a
failure to show reaction. Dead cells have bright green fluorescence across
the entire cell. The brightness of a specific reaction will quickly fade (30
sec) if fluorescent illumination continues, so a new microscopic field
should be located by changing back to dark-field illumination and repeating the process.
7. Examine each slide in the dilution series as described to qualitatively determine the dilution at which brightness is less intense. The working dilution of antiserum is of that dilution.
8. Test the working dilution of the antiserum on other species of cells
maintained in the laboratory and compare its reaction with that against
the homologous species of cells. If nonspecific cross-reactions do occur,
absorption with 1 volume of the cells of the cross-reacting species and 9
volumes of the working dilution of antiserum will usually eliminate the
cross-reaction without significantly diminishing the homologous reaction.
Alternatively, use of a higher dilution of antiserum will frequently take
care of the problem.
9. For routine monitoring cells should be tested within 4 hr of harvesting. The procedure works only with living cells. It is not well adapted to
fixed cell preparations or tissue sections. Excess working dilution antiserum can be refrozen and used one additional time. After that, a fresh
working dilution antiserum should be prepared for use. The undiluted
178
BASIC METHODS
[14]
coupled antiserum is stable in frozen storage for several years. After that
period of time the antiserum may require titering.
Comment. T h e procedure described is not difficult to perform. Results
can be obtained in 2-3 hr, and the test can be activated for use as needed.
The only reagent that need be prepared for each day of test is PBS.
Antisera are stable for a long period of time, and they can be used for
other serologic tests of cell identity, including cytotoxicity, hemagglutination, and mixed agglutination. 3e-3a Emphasis is placed on a direct immunofluorescent procedure because, unlike the other methods, it can be
used to follow cocultivation experiments involving cells of two different
species, as well as detecting early evidence of cross-contamination. As
few as one cell of one species can be detected in a mixture of 10,000 cells
of another species. 39 The indirect immunofluorescent procedure can also
be used.
The limitations of the method derive mostly from the limitations of
specificity of antisera that can be prepared. Cross-reactions between
closely related species occur. For example, chimpanzee, orangutan, and
human cells react to substantially equal degree with human antiserum.
Rhesus monkey, African green monkey, and baboon cells react similarly
with rhesus antiserum, so one should carefully consider the monitoring
capability of this procedure in relation to the species of cells being utilized
in the laboratory. However, the method can be adapted for use in intraspecies cell characterization. For example, using reciprocally absorbed
antisera, specific human T cell and B cell antisera have been developed to
analyze T and B lymphoblast cell lines by indirect immunofluorescence. 4
3e A. E. Greene, L. L. Coriell, and J. Charney, J. Natl. Cancer Inst. 32, 779 (1964).
3T K. G. Brand and J. T. Syverton, J. Natl. Cancer Inst. 28, 147 (1962).
3e D. Franks, B. W. Gurner, R. R. A. Coombs, and R. Stevenson,Exp. CellRes. 608(1963).
39 W. F. Simpson and C. S. Stulberg, Nature (London) 189, 616 (1963).
4o j. Kaplan and W. D. Peterson, Jr., Clin. Immunol. lmmunopathol. 8, 530 (1977).
[14]
T i s s u e C u l t u r e o n Artificial C a p i l l a r i e s
By P. M. GULLINO and R. A. KNAZEK
Survival and growth of cell populations in vivo are supported by a

vascular network that delivers nutrients and removes products of cellular
metabolism. The capillary culture unit (CCU) described here consists of a
network of artificial capillaries that simulates the in vivo vascular matrix.
Employment of this technique permits cells to grow and to reach a density
characteristic of solid tissue. In contrast to standard in vitro culture proMETHODS IN ENZYMOLOGY, VOL. LVIII

ISBN 0-12-181958-2
[14]
TISSUE CULTURE ON ARTIFICIAL C~PILLARIES
179
cedures, the CCU preserves the pericellular microenvironment and simulates the physiological conditions in which cells function in vh,o.
Principles
Under standard tissue culture procedures, cells are immersed in a pool
of medium containing essential nutrients and metabolic products. The
concentration of both constituents changes as the cell population grows,
thus influencing cell survival and function. In the CCU the culture medium flows within the capillaries and its composition can be kept constant.
The cell population grows within the extracapillary space, and nutrients
and metabolites are exchanged by diffusion from the perfusate within the
capillaries. A more complex form of CCU embodies a double network of
capillaries whereby differential pressure between the two capillary sets
superimposes extracapillary convection upon diffusion.
In neoplastic tissues, anoxic necrosis was observed in vh, o when the
distrance between the capillary wall and the cells was about 150/zm or
more.1 The CCU was built on the assumption that within a rapidly growing population sufficient oxygen tension could not be maintained at distances greater than 150/xm from the capillary wall. If a bundle of artificial
capillaries of about 300 /~m in outside diameter are tightly packed in
parallel, most of the extracapillary space should be within the oxygen
diffusion distance of 150/xm.
Construction of Culture Units
The standard culture unit consists of polymeric membranes in the
shape of tubes bundled together within a transparent cylinder (Fig. 1A,B).
We refer to these tube-shaped membranes as "capillaries" although their
diameter and wall thickness are much greater than an in vivo capillary.
Several types of artificial capillaries made of polysulfone, acrylic
copolymers, or cellulose acetate are available commercially, as is the
complete CCU. 2 Such units, individualized to suit specific needs, can be
constructed with a variety of available capillaries, as follows:
A bundle of 100-300 capillaries, having inner and outer diameters of
- 2 0 0 and - 3 5 0 ~m, respectively, are arranged in parallel and pulled
tightly into a glass or plastic shell about 1 cm in diameter and about l0 cm
in length. The bundle is closed with a suture tie at both ends of the shell
and immersed in a mixture of 12 g liquid silicone rubber (General Electric
1 R. H. Tomlinson and L. H. Gray, Br. J. Cancer 9, 539 (1955).
2 Amicon Corporation, Lexington, Massachusetts; Gulf South Research, New Orleans,
Louisiana.
180
BASIC METHODS
[14]
FIG. I. (A) Capillary culture perfusion circuit: (a) tissue culture medium reservoir; (b)
peristaltic pump; (c) silastic tubing coil for oxygenation of perfusate; (d) capillary culture
unit; (e) syringes for loading ceils onto capillaries. (B) Schematic presentation of capillary
culture unit: (a) loading ports; (b) cutaway view shows capillaries encased in polymeric
potting material (stippled area). (C) Histologic cross-section of human mammary carcinoma
cells (BT20) cultured for 4 weeks within a culture unit. Note the extracapillary space filled
with cells that also penetrate the porous outer structure of the capillary walls. Inner diameter
of capillary is 200/xm.
RTV-11) a n d 8 g M e d i c a l F l u i d 360 ( D o w C o m i n g C o r p . , M i d l a n d , M i c h i gan) t h a t h a d b e e n f o r m e d b y c a t a l y s i s w i t h a p p r o x i m a t e l y 40 m g stann o u s a c e t a t e ( T e n n e c o , N u o c u r e 28). A n e p o x y r e s i n c a n b e u s e d in p l a c e
of the silicone rubber mixture. After solidifying overnight, the excess
[14]
TISSUE CULTURE ON ARTIFICIAL CAPILLARIES
181
polymer is trimmed flush to the shell to expose the lumen of the capillaries. The process is then repeated for the opposite end. The resultant
seal must be tight so that fluid pumped through one end of the CCU will
flow only within the capillaries while the.extracapillary space, in which
the cells are growing, remains undisturbed. With this type of arrangement
the exchange between inner and extracapillary spaces occurs mostly by
diffusion through the capillary walls.
In the more complex CCU, convective currents are generated within
the extracapillary space. The capillaries are intertwined in a braid-like
fashion rather than arranged in a parallel array. This unit consists of a
central part where the capillaries form a single bundle and two extremities
where the capillaries diverge into two bundles, one designated as the
"artero-venous" (A-V) circuit and the other as the "lymphatic" circuit.
The A-V circuit is filled with medium circulating in a closed recirculating
loop to which pressures up to 300 mm Hg can be applied. The lymphatic
system is filled with medium circulating at the same rate but at atmospheric pressure. Steady convective currents, therefore, move within the
extracapillary space from the artero-venous to the lymphatic networks.
The CCU with lymphatic drainage is not available commercially.
Perfusion Circuit
a. Assembly
Under standard conditions the CCU is operated at 37 in 95%

humidified air + 5% CO2. In this environment the perfusion circuit requires four components: a reservoir for the medium, an oxygenator, a
pump, and the CCU interconnected by silicone rubber tubing.
Initial experiments were performed using a fiat membrane oxygenatora
which predisposed the perfusion circuit to leaks and contamination. Subsequent work showed that the oxygenator could be replaced by 2-3 m of
silastic tubing (0.125 in. OD, 0.063 in. ID) (Dow Coming Corp.) inserted
between the reservoir and the CCU. Since silastic is very permeable to
gases, that length of tubing is sufficient to maintain a physiological pH and
p 02 in the perfusate flowing at 5 ml/min.
The pump is an extremely important component of the circuit since it
must function for several weeks without interruption. The roller-type
pump (Holter Co., Extracorporeal Medical Specialties, King of Prussia,
Pennsylvania) proved to be the most efficient. It has the advantage of not
being in direct contact with the medium, thereby minimizing contamina3 T. Kolobow, W. Zapol, and J. Marcus, in "Organ Perfusion and Preservation" (J. C.
Norman, ed.), p. 155. Appleton, New York, 1968.
182
BASIC METHODS
[14]
tion, but the disadvantage of wearing out the wall of the pumping chambers. Replacement of this chamber every 10 days is advisable. Another
problem is the liberation of microscopic fragments from the inner wall of
the pumping chamber into the perfusing medium. After several weeks of
uninterrupted activity, occlusion of the capillaries may occur with irreparable damage to the culture. 4
A CCU with lymphatic drainage requires two circuits, assembled as
described above, with an additional device to increase the pressure of the
medium circulating within the A-V circuit. This may be attained by
pressuring the A-V reservoir with an air-CO2 mixture. The lymphatic
circuit is connected to a second reservoir, and the medium is circulated at
normal atmospheric pressure. An equal flow rate is usually maintained in
both lymphatic and A-V circuits.
It is convenient, although not indispensable, to insert a filter in the
perfusion circuit downstream from the pump with an optional bypass. In
this manner, sampling sites can be inserted into the circuit.
The recommended sterilization of the CCU is in ethylene oxice for 10
hr followed by extensive aeration; however, certain types of CCUs may
be autoclaved, or the unit may be perfused in a 3% formaldehyde solution
at 37 overnight. 4 The remainder of the perfusion circuit may be autoclaved. All components are then assembled in an aseptic environment. The
assembled circuit must be perfused with sterile, pyrogen-fi-ee water for at
least 2 and preferably 7 days with at least three changes of water prior to
replacing the perfusate with complete medium. The objective of the preliminary flushing is to eliminate residues that are toxic to the cells.
The sterilized circuit should always be handled under rigid aseptic
conditions; one should replace the reservoir in a sterile hood while wearing sterile gloves, and the stopper of the reservoir should be covered by
aluminum foil to be replaced with each change of the medium.
Seeding of the CCU is done with approximately 5 106 cells obtained
either by dissociating tissues or from an established culture. The cell
suspension is injected through one shell port while a slight vacuum is
drawn by a syringe on the second port (Fig. 1A). The fluid must be
transferred slowly to deposit, rather than channel, cells or cell clumps
between the capillaries.
A perfusion flow rate of 0.5-1.0 ml/min is maintained at the beginning
of the culture and then gradually increased as the culture grows. When the
cells fill more than 25% of the extracapiUary space, a perfusion volume of
10 ml/min must be maintained to supply adequate amounts of oxygen and
nutrients. Under these conditions the recirculating medium becomes depleted very rapidly and frequent changes of the reservoir are required.
4 C. F. W. Wolf and B. E. Munkelt, Trans. Am. Soc. Artif. Intern. Organs 21, 16 (1975).
[14]
TISSUE CULTURE ON ARTIFICIAL CAPILLARIES
183
Any culture medium used for a specific cell type, grown in standard in
vitro cultures, can be used for CCU perfusion. In this system the capacity
to supply oxygen to the culture is severely limited by the low solubility of
oxygen in the medium. To improve delivery, hyperbaric conditions may
be employed. Fluorocarbons (3M Corp., St. Paul, Minnesota) or erythrocyte suspensions may also be used as oxygen carders, although they are
not necessary under standard conditions.
Cell growth can be detected by gross observation of the CCU. During
the second week after seeding, a white haze appears and gradually
obscures the capillaries. The amount of tissue present within a CCU can
be estimated by measuring the total DNA content, but this requires destruction of the culture. A less precise estimation can be made by measuring the glucose and oxygen consumption of the growing cell population in
the CCU. In repetitive cultures of the same cell population, the correlation between both metabolic parameters and DNA content of the CCU is
reasonably reliable.
Histologic sections through the capillary bundle can be obtained by
flushing the perfusion circuit for 1-2 rain with warm buffered saline and
then circulating a fixative solution, such as 10% formaldehyde, through
the capillaries for several hours. The extracapillary medium is then replaced with a 2% agarose solution at 37 by slowly injecting the solution
through one shell port while withdrawing it from the other until agarose
has penetrated the whole bundle. After solidification the shell is removed
and the capillary bundle can be processed for histologic examination (Fig.
1C).
Utilization of the Culture Units
The major theoretical reason for developing the CCU was to reproduce tissue-like structures in vitro. In the CCU a high cell density
(100 l0 e cells/era3) is obtained, much higher than under standard in vitro
culture conditions; the pericellular microenvironment is preserved; the
exchange of metabolites occurs through a capillary-like network; and the
same culture can be studied for several weeks. The importance of these
parameters in regard to the biological properties of cell populations has
already been evaluated under a variety of conditions. Prolactin and
growth hormone production were studied in rat and human cells of pituitary origin. 5 Secretory stimulation of human/3 cells was studied, 6,7 and
5 R. A. Knazek and J. S. Skyler, Proc. Int. Syrup. Growth Horm. Rela. Pept., 3rd, 1975 p.
386 0976).
6 W. L. Chick, A. A. Like, and V. Lauds, Science 187, 847 0975).
7 R. A. Knazek, in "Pancreatic Beta Cell Culture" (E. yon Wziselewski and W. L. Chick,
eds.), p. 29. Excerpta Med. Found., Amsterdam, 1977.
184
BASIC METHODS
[15]
CCU units were used in a successful attempt to reproduce an "artificial

pancreas."s-~ Bilirubin conjugation was observed with hepatic cells cultivated in CCU units, and in vivo transplants were also attempted. 4
Chorionic gonadotropins were produced in large amounts from human
choriocarcinomas, 11-1a and carcinoembryonic antigen was also harvested
from cells (LS-174) isolated from a human colon carcinoma. 13 Cells maintained in culture under standard procedures have shown two fairly general
characteristics, i.e., loss of functional differentiation and shedding of cell
components. Both events represent a severe limitation in correlating in
vitro and in vivo conditions; the CCU represents a step toward bridging
this gap.
s W. J. Tze, F. C. Wong, L. M. Chen, and S. O'Young, Nature (London) 264, 466 (1976).
9 A. M. Sun and H. G. Macmorine, Diabetes 25, Suppl. 1,339 (1976).
10 W. L Chick, A. A. Like, V. Lauris, P. M. Galletti, P. D. Richardson, G. Panol, T. W. Mix,
and C. K. Colton, Trans. Am. Soc. Artif. Intern. Organs 21, 8 (1975).
11 R. A. Knazek, P. M. Gullino, P. O. Kohler, and R. L. Dedrick, Science 178, 65 (1972).
1~ R. A. Knazek, P. O. Kohler, and P. M. Gullino, Exp. Cell Res. 84, 251 (1974).
la L. P. Rutzky, J. T. Tomita, M. A. Calenoff, and B. D. Kahan, In Vitro 13, 191 (1977).
[15] M i c r o c a r r i e r C u l t u r e : A H o m o g e n e o u s E n v i r o n m e n t
Studies of Cellular Biochemistry
for
By WILLIAM G. THILLY and DAVID W. LEVINE
Microcarrier culture is the growth or maintenance of anchoragedependent cells on small beads suspended in a stirred tank. The primary
conception of this technology was made by A. L. van Wezel of the
Netherlands. 1 He and his colleagues were responsible for several important demonstrations of feasibility. ~-4 Other laboratories have confirmed
their observations. 5~ Van Wezel and others noted a marked toxicity of the
original microcarriers which was ascribed variously to binding of essential
nutrients or unidentified toxic contaminants associated with manufacture.
However, recent studies have revealed a simple parameter, surface
1 A. L. van Wezel, Nature (London) 216, 64 (1967).
A. L. van Wezel, Prog. Immunobiol. Stand. 5, 187 (1972).
3 A. L. van Wezel, in "Tissue Culture: Methods and Applications" (P. F. Kruse, Jr. and
M. K. Patterson, eds.), p. 372. Academic Press, New York, 1973.
4 D. van Hemert, D. G. Kilburn, and A. L. van Wezel, Biotechnol. Bioeng. 11,875 (1969).
s C. Horng and W. McLimans, Biotechnol. Bioeng. 17, 713 (1975).
6 R. E. Spier and J. P. Whiteside, Biotechnol. Bioeng. 18, 659 (1976).
M E T H O D S IN ENZYMOLOGY, VOL. LVIII
Copyright ~ 1979 by Academic Press, Inc.

ISBN 0-12-181958-2
[15]
MICROCARRIER CULTURE
185
charge density, which, when optimized, permits the facile attachment and
growth of animal cells on microcarriers without any of the toxic effects
previously noted. 7-9
The advantages of microcarrier culture are based on provision of a
large surface area in relation to the volume of the growth vessel used. This
may be understood by comparing roller bottles to microcarriers in terms
of the surface area afforded. A standard 500 cm z roller bottle requires 100
ml of nutrient medium. One milliliter provides 5.0 cm 2 of growth surface.
In our microcarfier cultures, 1 ml of medium provides a minimum of 30
cm 2 of growth surface, and the opportunity exists for increasing this number by 2 or 3 times. Thus, for most cell types, a 5-liter spinner flask would
replace a facility using 300 roller bottles. While the simple increase in
surface area is important, other advantages are offered as a result of the
suspended, stirred configuration of microcarrier culture. Possibly the
most important advantage is the provision of a homogeneous environment
for the cell population, in which the time-dependent changes in microenvironment which occur in culture plates and minimally agitated roller bottles are reduced by simple mixing of the cellular microenvironment and
the bulk liquid of the culture. We suspect that students of cell physiology
are performing many studies without knowledge of microenvironmental
changes or their effects on cellular behavior. Microcarrier culture permits
the investigator to increase the rate of exchange between the cell surface
and the bulk liquid. Thus, the properties of the cellular macro- and microenvironments can approach identity. The bulk properties (pH, redox
potential, concentrations of excreta and nutrients, etc.) become controllable by the experimenter with the facility afforded in bacterial, yeast, or
suspension-adapted animal cell culture. When needs arise for amounts of
cellular material in the range of 10-100 g, scale-up can be accomplished
easily using very simple equipment.
Source of Microcarriers
Microcarriers are obtained from Flow Laboratories, Inc., Rockville,
Maryland. A usable small-scale procedure for their synthesis is based on
that of Hartman TM in which diethylaminoethylchloride. HC1 is reacted
7 D. W. Levine, D. I. C. Wang, and W. G. Thilly, in "Cell Culture and Its Applications"
(R. T. Acton and J. D. Lynn, eds.), pp. 191-216. Academic Press, New York, 1977.
8 D. W. Levine, J. S. Wong, D. I. C. Wang, and W. G. Thilly, Somatic Cell Genet. 3, No. 2,
149-155 (1977).
9 D. W. Levine, D. I. C. Wang, and W. G. Thilly, Biotechnol. Bioeng. (submitted for
publication).
10 M. Hartmann, U.S. Patent 1,777,970 (1930).
186
aASIC METHODS
[15]
with the glucose of cotton fiber under alkaline conditions as previously

described. 8
Uncharged, unhydrated beads of cross-linked dextran are obtained
commercially from Sigma Chemical Co., St. Louis, Missouri, as the gel
chromatography support Sephadex G50M. The commercial product is
sieved, and the dry bead fraction having diameters between 50 and 75/~m
is used for subsequent steps. It should be noted that the commercial
products variously described as DEAE-Sephadex and marketed both by
Pharmacia Fine Chemicals, Piscataway, New Jersey, or Sigma Chemical
Company, have been found unsuitable for use as microcarriers in our
hands. The limitations of these products have been extensively discussed
elsewhere. 1-s
Diethylaminoethylchloride. HC1 ( D E A E . HCI) is obtainable from
Sigma Chemical Company. Before use, it is twice recrystallized from
reagent-grade methylene chloride. Sodium hydroxide used in the reaction
must also be reagent-grade material.
One gram of dry sieved beads is hydrated with 10 ml of distilled water.
Ten milliliters of distilled water which contain 10 mmol DEAE HC1 and
15 mmol NaOH are added to the hydrated beads. The suspension is agitated briskly at 60 for 60 min. Microcarders are separated from the
reaction mixture by filtration and are washed with 1 liter of distilled water.
The reaction is essentially complete in 60 min and binds 2.0 mmol of
DEAE 1o 1 g of dry cross-linked dextran beads. This corresponds to a
microcarrier with an exchange capacity of 2.0 meq/g cross-linked dextran.
To produce a carder with a lower degree of DEAE binding, similar reaction conditions are employed but reaction periods of less than 1 hr are
used. For example, to produce carders having an exchange capacity of
1.0 meq/g cross-linked dextran, the above reaction mixture is agitated on
a shaking water bath for 3 min at 60.
The synthetic product has been found to result in excellent culture
productivity for a number of different cell types: primary or secondary
cultures, fibroblasts, or epithelial cells. The charge density of 2.0 meq/g
dry dextran bead is optimal for growth of secondary chick fibroblasts, s It
is probable that other cell types will require slightly different charge densities to achieve optimal results. Our experience to date indicates that a
range of 0.5-2.5 meq/g dry dextran bead may be profitably explored when
a new cell type is approached.
After synthesis, microcarders should be titrated to determine their
exact charge capacity as part of a quality-control procedure. The carders
formed by the reaction of 1 g of dry cross-linked dextran with
DEAE HC1 are thoroughly washed with 0.1 N HCI (4 50 ml) to allow
saturation of all the exchange sites with chloride ions. The carders are
[15]
187
rinsed with 10-4N HC1 (4 x 50 ml) to remove the unbound chloride. The
beads are rinsed with aqueous sodium sulfate (10% w/w), and the effluent
is collected. This procedure replaces bound chloride ions with sulfates,
releasing the chlorides into the effluent solution. The chloride is titrated
with 1 M silver nitrate in the presence of dilute potassium chromate as an
indicator. The carders are washed with 30-ml portions of 10% sodium
sulfate solution until the amount of released chloride in the effluent is less
than 0.1 mM. Typically, three to four washes are required. After titration,
the carders are again washed thoroughly with water (500 ml) and then
with phosphate-buffered saline (PBS). The carders are suspended in 100
ml of PBS and autoclaved.
Initiation of the Culture
The microcarrier culture bottle is simply a glass or plastic vessel containing a suspended stirrer in which the provision is made to avoid grinding surface contacts with the liquid culture. Simple bottles which are
particularly convenient for microcarrier work are available from Wilbur
Scientific, Boston, Massachusetts, in the volume range of 50-1000 ml. As
mentioned, microcarriers are sterilized by autoclaving a suspension of
physiological saline buffered at pH 7.
The culture is begun by the combination of microcarriers, e.g., 5 mg
dry weight beads per milliliter culture medium, with the desired volume of
culture medium and the introduction of the cells to be used (freshly trypsinized or otherwise collected) as a suspension into the culture bottle. The
bottle is gassed to the desired COs concentration if necessary and placed
on a magnetic stirrer motor (50-90 rpm) and allowed to incubate at 37.
This process is indeed as simple as it must seem to the reader acquainted
with cell culture practice.
Observation of the Culture
The use of the microcarriers at 5 mg dry weight dextran per milliliter
will provide about 40,000 beads/ml. Seeding at 300,000 cells/ml assures
distribution of cells to all carriers. Within a few minutes, most cells will be
attached to microcarriers as illustrated for secondary chick fibroblasts in
Fig. 1A. Samples of stirred cultures are simply withdrawn by pipetting
using nonwettable pipettes to avoid adherence of beads to hydrophilic
surfaces. Plastic pipette tips (Rainin Instrument Company, Brighton,
Massachusetts) are the simplest solution to bead adherence. Samples are
observed directly under a microscope. In a few hours, cells will have
attained their normal culture morphology as illustrated in Fig. lB. Growth
188
BASIC METHODS
[15]
FIG. 1. Attachment and growth of secondary chick embryo fibroblasts on microcarriers.

(A) Thirty minutes after seeding. Five grams per liter microcarriers, 2.0 meq/g dextran.
Initial concentration of 300,000 cells/ml and approximately 40,000 beads/ml. Only 10% of
seeded cells can be found in the supernatant liquid 30 min after seeding under stirred
conditions. (B) Twelve hours after seeding. All cells have assumed typical fibroblastic morphology. (C) Forty-eight hours after seeding. Cell number has increased almost 4 times
relative to seeding density. (D) One hundred twenty hours after seeding. Beads are covered
with a confluent cell layer.
is easily observed microscopically as seen in Fig. 1C. Confluence is attained as on other culture surfaces (Fig. 1D).
Cell counting is easily performed using a hemocytometer and the
technique for enumeration of nuclei reported by Sanford et al. 11 A 1-ml
u K. K. Sanford, W. R. Earle, V. S. Evans, J. K. Waltz, and J. E. Shannon, J. Natl. Cancer
Inst. 11, 773 (1951).
[15]
189
sample of the culture is centrifuged and resuspended in 1 ml ofO.1 M citric

acid containing O.1% crystal violet, in which it is allowed to incubate at
37 for I hr,pipetted up and down gently, and loaded onto a hemocytometer. The hypotonic solution leads to cellular but not nuclear lysis, and the
crystal violet stains the nuclei a dark purple to permit easy enumeration.
The beads are too large to enter the hemocytometer counting chamber
and do not interfere with the'count.
The counting procedure of removing cells from carders by trypsinization, sieving to remove beads, and counting cells electronically should
also work easily, but we have not yet diverted sufficient attention to
develop a specific protocol.
Figure 2 shows the growth curve for secondary chick fibroblasts
shown in Fig. 1A-D. Saturation is attained by the fifth day at a cell
concentration of 4 x 10~ cells/ml. Within 3 days of reaching confluence,
cell death and detachment are evident. A 50% replacement of medium is
made 48 hr after culture initiation.
Similar results, short Or nonexistent lag times, rigorous growth to confluence, and high final cell densities have been observed for a variety of
cell types, both fibroblastic or epithelial. Figure 3 shows the appearance
of (A) Chinese hamster ovary (CHO) cells in exponential growth prior to
attaining confluence and (B) HeLa cells at approximately the same growth
i
40
SECONDARY CHICK EMBRYO
i~
FIBROBLASTS
50
'o
x
.j 20
_$ i,
T I M E (DAYS)
FIG. 2. Growth curve of secondary chick embryo fibroblasts. Initial seeding at 300,000
cells/ml with 5 g/fiter microcarriers at 2 meq/g dextran. After an initial lag phase (typically
shorter than observed in this experiment) exponential growth to about 2 10 cells/ml is
followed by linear growth to saturation between 3-4 x l0 e cells/ml. Use of 10 g/liter microcarriers with timed partial medium changes will bring chick embryo fibroblast concentrations to 1.0-1.2 l0 T cells/ml.
190
BASIC METHODS
[15]
A
8
FIG. 3. Appearance of epithelial cell lines on microcarriers. Cultures at 5 g/liter microcarriers, 2.0 meq/gdextran. Both cu.lturesare in exponentialgrowth phase. (A) Chinese
hamster ovary cells; (B) HeLa cells.
state. CHO, HeLa, murine cl 1, and human embryonic lung fibroblasts
(HEL299) have demonstrated vigorou s growth without any refeeding
after seeding. However, some cell types such as African green monkey
kidney (AGMK) or human foreskin fibroblasts (FS-4) have not demonstrated equally vigorous growth (Fig. 4).
To date, no dependence on medium type or formulation has been
observed. At this time, it seems that media suitable for growth of cells in
roller bottles or petri dishes are equally suitable when used with the
microcarriers described in this article.
Harvesting
The mode of harvesting will vary depending on individual research
needs. The simple washing of the microcarriers with buffered saline, trypsin exposure, and separation of cells from beads by differential sedimentation rate or by sieving is easy and effective. Increased shear can be
achieved in a number of ways such as increased impeller velocity, pumping through a small orifice, 6 or other strategems.
Some workers have asked about removing cells without trypsin for
biochemical studies in which loss of external cellular components is undesirable. We have been unable to devise a means of preventing such losses
i n harvest, but note that considerable cellular material remains on microcarriers and standard glass or plastic dishes when either trypsinization
or strictly mechanical means are used to remove cells from the attachment substrate. An alternate approach to this problem might be the em-
[15]
i
40
OHILo
/~cl I
3C
echo
i.i11!. 2
191
I-1FS- 4
g
~" 20
N
J
0
TIME (DAYS)
FIG. 4. Growth curves for various cell types. All were seeded at 2-3 x 105 ceUs/mlin
100-mlcultures, 5 gm/litermicrocarriers at 2.0 meq/gdextran. HeLa--human epithelial-like,
aneuploid; CHO--hamster epithelial-like,pseudodiploid; cl 1--murine epithelial-like,aneuploid; FS-4--human fibroblast, diploid; AGMK--African green monkey, epithelial-like,
diploid.
ployment of a synthetic growth surface of which a known component
would be severed by enzymic treatment, leaving the cell surface or its
associated exudate unaltered.
Viral Studies
Viruses of interest to investigators may remain physically within a
cell, be budded continuously from the cell surface, or be expressed into
the liquid support medium by cell lysis. In the case of a cell-contained
virus, e.g., Marek's disease, the cells simply can be stripped from the
beads as in cell harvest and the cell suspension used in biochemical studies or as the basis of a vaccine.
Some budding viruses, e.g., RNA tumor viruses, present a special
challenge to virologists in that loss of infectious units with time is often a
problem necessitating frequent or continuous harvesting. Microcarrier
culture offers a ready solution to these problems. The motor drive of a
microcarrier culture, and a pump connecting the culture to a refrigerated
collection bottle, can be connected to a timer. The culture is grown to an
appropriate cell density, infection is accomplished, and when viral expression commences, the stirring motor is turned off, the microcarriers
192
BASIC METHODS
[15]
allowed to settle for a few minutes, the supernatant pumped into the
collection bottle, fresh medium pumped into the culture, and stirring
begun again. This cycle is repeated at intervals appropriate to the cells
and viruses involved.
Viruses expressed into the culture medium lytically are handled similarly. Stirring ceases, and the supernatant liquid above the settled beads is
pumped into a receiving container.
One consideration in the use of microcarrier culture is that the viral
growth cycles may not be the same in these microcarrier suspension
cultures as they are on roller bottles or plates, possibly because of differences in microenvironmental conditions. Thus, preliminary experiments
optimizing for time of infection, latent period, and harvest are necessary.
Viruses on microcarriers that have been studied in this manner include
Sindbis, polio, Moloney murine leukemia, and vesicular stomatitis. 12
Rabies virus has been produced in excellent yield, ~a and a number of
commercially important veterinary vaccine viruses are apparently produced in good yield.
Interferon and Other Cellular Products
Theoretically, interferon or other secretory products could be produced in a manner directly analogous to the budding viruses on a semicontinuous basis. Initial interferon studies ~4 have found that using the same
conditions of cell type, growth media, and superinduction, the interferon
production on a per-cell basis was the same for cells in microcarrier
culture or in roller bottles.
There seem to be no obvious reasons to expect that other cellular
products such as peptide hormones or secretory enzymes could not be
made or studied to advantage through the use of microcarrier systems. It
should be noted, however, that the microcarriers described in this article
are porous, positively charged hydrogels. Some products having a low
molecular weight or a negative charge may be expected to associate with
the microcarriers and possibly require special elution procedures.
Mitotic Cell Selection
One limitation on study of the mammalian cell division cycle has been
the difficulty of obtaining a sufficient number of synchronously dividing
1~D. J. Giard, W. G. Thilly, D. I. C. Wang, and D. W. Levine,Appl. Environ. Microbiol. 34,
668-672 (1977).
18 B. Mered, personal communication.
14 D. J. Giard, D. H. Loeb, W. G. Thilly, D. I. C. Wang, and D. W. Levine, Biotechnol.
Bioeng. (submitted for publication).
[15]
193
cells. One way of obtaining cells of sufficient synchrony is to selectively

detach the relatively nonadherent mitotic cells from their attachment substrate. This method, devised by Terasima and Tolmach, 15 yields suspensions of cells, some 80-95% of which are mitotic. At this writing, we have
extended this concept to microcarriers. Microcarrier cultures of Chinese
hamster ovary cells in exponential growth (-106 cells/ml, 5 mg/ml microcarriers) are simply agitated more vigorously by increasing the spinner
speed for 10 min after 2 hr of growth in the presence of 0.08 ~g/ml
colcemid. About 85% of the cells freed into suspension are mitotic as
measured microscopically, and the increase in cell number observed soon
after plating agrees with this estimate. No doubt other cell lines will lend
themselves to this approach. Since 1-2% of the total population has been
recovered in our experiments with CHO cells, 100 ml of culture yield
between 1 and 2 l0 s mitotic cells. Of course, scale-up is possible.
Biochemical Engineering
Microcarriers appear to offer a ready means for large-scale use of
anchorage-dependent animal cells. The special concerns of such usage are
beyond the scope of this article, which is intended to offer the basic details
of this approach for the consideration of cell biologists and biochemists
who may be hampered by the difficulty and expense of present means of
cell culture. However, engineering questions are discussed in a separate
publication. 9
Comments
Our collective knowledge of the nutritional substrates for growth of
bacteria, yeasts, and higher plant and animal cells has advanced considerably beyond our knowledge of how living cells use--and why they
sometimes require--attachment substrates for cellular growth toward division or for growth in a particular physiological mode. Our present understanding may be likened to that of the early nutritional biochemists
who knew many of the nutrients that were necessary for the growth of
bacteria or young animals, but did not yet comprehend the specific enzymic systems which used the nutrients as substrates or required the
various vitamins as catalysts. In this presentation of methodology, which
is su~cient for the growth of anchorage-dependent animal cells, we recognize that the underlying molecular mechanisms for the specific requirements are essentially unknown. Perhaps the definition of these requirements will lead to hypotheses useful in the design of experiments that can
~5 T. Terasima and J. Tolmach, Exp. Cell Res. 30, 344 (1963).
194
BASIC METHODS
[16]
result in better understanding of a cell's relation to its attachment

substrate.
Of course, investigators have been growing cells on solid substrates
for some 75 years. Blood clots, collagen, glass, treated plastics, various
cross-linked proteins, and other materials, including stainless steel, have
been applied with varying degrees of success. The petri dish, the T-flask,
and the roller bottle used by cell researchers today are primarily made of
polystyrene treated with ozone via plasma discharge exposure and catty a
net negative charge of approximately 0.3-0.5 meq/cm2, TM creating a wettable surface, but it is not clear that hydrophilicity alone is both necessary
and sufficient for cell attachment and growth. Wettable surfaces can, of
course, be created with positively charged substituents, as discussed in
this article. Our ignorance of specific mechanisms prevents us from defining which specific microenvironmental parameters are necessary for cell
attachment and growth.
Acknowledgments
The research reported in this chapter was supported by National Science Foundation
Grant #77-15463PCM and National Cancer Institute Grant #NIH-2-ROI-CA-15010-04, and
it was performedin the M.I.T. Cell CultureCenter. We gratefullyacknowledgethe technical
assistance of John Y. Ng and Charles L. Crespi in the preparation of growth curves and
photography, whichare originalwith this article.
le N. C. Maroudas,J. Theor. Biol. 49, 417 (1975).
[16] M a s s C u l t u r e o f M a m m a l i a n C e l l s
By WILLIAM F. MCLIMANS
Introduction
The purpose of this article is to provide a guide to the mass culture of
mammalian cells for those investigators who require milligram or gram
quantities of cells for biochemical studies. Thus, some points may appear
controversial or even offensive to the "cell culturist" anxious to protect
his domain. The "state of the art" and need are such, however, that an
effort must be attempted, even at that risk. The rewards are great for
those who would exercise control over their own source of cells.
The approach is complicated by the multiple configurations of equipment and methods that are available. Basically, however, there are only
two effective mass-scale systems; one employs the suspension culture of
established cell lines, and the other and newer technique, is that of direct

ISBN 0-12-181958-2
[16]
MASS C U L T U R E OF M A M M A L I A N CELLS
195
growth in suspension of primary cells that are anchored to small "beads"

or microcarriers. Since both systems have been extensively employed in
our laboratories, we will present our operative methods. These are methods that may not be optimal, superior, or even equal to others, yet they
are methods that we know "first hand" to be reliable. It should be emphasized that, as a biologic system, there is no universal culture media, no
universal procedure, and certainly no universal cell system.
No matter which system is used, detailed attention must be directed
toward the "tooling up" proceduresmthe preparative operations, i.e.,
glassware, media preparation, reagent selection, water quality, etc. Accordingly, these prerequisite, operative procedures will be dealt with in
the first section, prior t/~ discussing the actual operations of primary vs.
cell line mass culture configurations. Little attention will be directed toward equipment configurations and/or applications, since these have been
covered in other reports, such as that of Nyiri.1
Preparative
Operations
Water
Extensive experience has demonstrated conclusively that the single

most impo~ant constituent of the culture media is proper-quality water. It
is recommended that wide-mouth bottles of sterile, pyrogen-free water, as
purchased for intravenous use in man (a), 2 be employed in the direct
formulation of all culture media and/or reagents, as directly added to the
culture environment. This entails an expense of about $0.66/liter of
media, 3 a minor cost, compared to the uniformity of culture results. In
large systemsml00 liters or more--this cost might be prohibitive. In such
cases, triple glass-distilled water may be used, but this requires care in
preparation as well as periodic checks to certify the absence of pyrogens
prior to use for media formulation. Pyrogen checks are particularly important following installation of new stills, after cleaning or repair of stills,
and in the instance of stills which have not been used for a long period.
On a day-to-day basis, stills should be drained each evening and
"steamed" out prior to use each morning.
Triple glass-distilled water should be employed as the final rinse for all
glassware, as well as for the preparation of various noncritical reagents
and solutions. In no instance should the distilled water be collected in
L. K. Nyiri, Biotechnol. Bioeng. Symp. 3, 31--40 (1972).
z The italicized lower case letters in parentheses refer to vendors whose names are summarized in Table I.
3 Prices are estimates as o f January 1978.
196
BASIC METHODS
[16]
TABLE I
SOURCES: MATERIALS AND EQUIPMENT
Letter ref.
//
Source
Travenol Laboratories, Inc., 1 Baxter Parkway, Deerfield, Illinois 60015
Difco Laboratories, Detroit, Michigan
Flow Laboratories, Inc., 1710 Chapman Avenue, Rockville, Maryland
20852
Grand Island Biological Company (GIBCO), 3175 Staley Road, Grand
Island, New York 14072
Microbiological Associates, 5221 River Road, Bethesda, Maryland 20016
Colorado Serum Company, 4950 York Street, Denver, Colorado 80216
Fiske Associates, Inc., Bethel, Connecticut
Cryogenic Supply Co., Inc., Quinby Park, Hamburg, New York 14075
The Matheson Co., Inc., East Rutherford, New Jersey
Armour Pharmaceuticals, Inc., P.O. Box 511, Kankakee, Illinois 60901
Bellco Glass, Inc., 340 Edrudo Rd., Vineland, New Jersey 08360
Ace Glassware, Inc., Vineland, New Jersey / 54 Muddy St., (P.O. Box
425), Ludlow, Massachusetts 01056
Virtis Co., Route 208, Gardiner, New York 12525
New Brunswick Scientific Co., Inc., 1130 Somerset St., New Brunswick,
New Jersey 08903
Fermentation Design, Inc., Div. of New Brunswick Scientific Co.,
Bethlehem, Pennsylvania 18017
anything other than P y r e x glass or plastic containers specifically r e s e r v e d

for this purpose. Distilled w a t e r which has b e e n stored without sterilization for m o r e than 6 hr should not be used.
Chemicals
Biologic grade or U S P chemical reagents should be used in the formulation o f all tissue culture preparations. Certification of the source and
purity o f the reagents should be furnished b y the vendor. A certificate of
analysis o f amino acids and vitamin mixtures should be obtained prior to
p u r c h a s e o f bulk lots.
T h e use o f c o m m e r c i a l l y p r e p a r e d liquid culture m e d i a is not recomm e n d e d for the following critical reasons. One has no control o v e r the
quality o f the w a t e r c o m p o n e n t . T h e formulation structure is too rigid to
permit critical adjustment o f p H , osmolarity, and glucose concentration at
various levels. S o m e m e d i a c o m p o n e n t s , such as penicillin and glutamine,
are labile and thus undergo continuous changes in concentration as a
[16]
MASS CULTURE OF MAMMALIAN CELLS
197
result of degradation. 4 We have found the preweighed, dry powder, media

mixes without salts, glucose, antibiotics, and/or bicarbonate, in 1-, 5-, 10-,
or lO0-1iter batches to be ideal for our investigations (b, c, d).
Serum 5
An important variable of cell culture media is the serum component. 6,r

Commercially available serum is usually procured by random bleeding of
slaughter-house animals (b, c, d, e). Relatively little control is exercised,
over the selection of donors, as based on the animal history, i.e., exposure
to infectious agents, toxic grasses (bishydroxycoumarin), pesticides, etc.
The problem can be resolved to a degree by obtaining test aliquots from a
large serum lot. Following the demonstration of efficacy of the serum
aliquot in supporting cell growth and generation time, the entire lot may
be reserved for future use. This has proven to be a satisfactory procedure
in many laboratories. More costly, but superior, is the commercial production of sera from animal herds maintained by the producer with absolute control over the history and exposure of animals to such things as
antibiotic-containing feed and diethystilbestrol (f).
4 G. L. Tritsch and G. E. Moore, Exp. Cell Res. 28, 360-364 (1962).
s Careful certification of the serum requirements to the vendor may be made in terms of the
following typical specifications: (1) All sera should be prepared from blood collected under
aseptic conditions. (2) Periodic inspection of bleeding procedures should be made by the
vendor. (3) Insofar as possible, efforts should be made to preclude the bleeding of animals
which are: suspect of having a disease or derived from a herd known-to contain any sick
animals; known to have been treated with drugs or vaccines for 30 days prior to bleeding;
known to have had access to or to have been exposed to antibiotic-containing feed,
pesticides, or insect sprays for 30 days prior to bleeding. The vendor should furnish a
statement as to the feasibility and the contemplated degree, if any, of his compliance with
this specification. (4) Preservatives, decolorizing agents, or foreign materials should not be
added to sera. (5) Sera should be immediately frozen, stored, and delivered, in the frozen
state, without prior heat inactivation. (6) Immunoelectrophoresis may be performed to
verify source and normal protein distribution. (7) Physical and chemical specifications of
serum may include a protein concentration of at least 3.0%; hemoglobin content not above
25 mg/100 ml (Mollison); total solids not below 5.5%; and nonprotein nitrogen not greater
than 0.7 mg/ml. The specific gravity should not be less than 1.02, with a pH range within
the limits of 7.0-7.5. The serum should not be turbid, should be low in lipids, and should
be totally free of visible surface "fatty ring." (8) The final serum product should be
critically tested to insure sterility. The sera should be demonstrated to be PPLO and
virus-free, noncytotoxic, and able to support the growth of primary and cell line tissue
cultures.
s K. Honn, A. Singley, and W. Chavin,Proc. Soc. Exp.IBiol. Med. 149, 344-347 (1975).
7 S. Fedoroff, V. J. Evans, H. E. Hopps, K. K. Sanford, and C. W. Boone,In Vitro 1, No.
3, 161-167 (1972).
198
BASIC METHODS
[16]
TABLE II
STOCK SALT SOLUTIONS
Stock
number
Stock
reagent
Final (x l)
(mg/liter)
1
2
3
4
5
6
NaCI
CaCI~ (anhyd.)
KC1
MgSO4 7 H~O
NaH2PO4H20
Phenol red
6460
I00
400
97.7
580
2
Glucose
Gentamycin
Mycostatin
10
NaHCO3
11
NaCI (mos M)
Stock (x 10)
(g/liter)
64.6
10.0
40.0
9.77
58.0
20.0 mg
Stock
(ml/liter)
90a
I0~
10
10
10
2
Glucose
1~4 g/liter
Stock of 100 mg/ml or 100 g/liter in water.
Antibiotic
50 mg/ml stock: use 0.5 ml/liter = 0.025 mg/ml =
25 tzg/ml.
500,000 ~g/vial + 10 ml H20 = 50,000 Izg/ml stock:
use 2.0 ml stock/liter for final cone. of 100 mg/ml.
Buffer adjust stock
5 g/100 ml or 5%
Osmolarity adjust stock
1 mg/ml
a Stock number 1: NaCI is added at lower level than that normally used so as to permit
osmolarity adjustment later.
b In some suspension-type cultures the deletion of Mg~+ and Ca2+ ions is recommended
in order to reduce the tendency of some cell types to clump.
Media Preparation
W e p r e p a r e c u l t u r e m e d i a a s f o l l o w s . T h e s t o c k salt s o l u t i o n s , a s
d e t a i l e d in T a b l e I I , a r e p r e p a r e d in a v o l u m e o f 1 l i t e r w i t h d i s t i l l e d w a t e r
as t h e d i l u e n t . T h e s t o c k s o l u t i o n s a r e e i t h e r f r e s h l y p r e p a r e d a n d u s e d
immediately, or stored as sterile solutions. Solutions are sterilized by
filtration in o r d e r t o a v o i d p o s s i b l e c o n t a m i n a t i o n o f t h e s t o c k s o l u t i o n s
during autoclaving with steam containing volatile boiler preservatives, s
S t e r i l i z a t i o n b y a u t o c l a v i n g in u n i t s fitted f o r s t e a m , a s g e n e r a t e d f r o m
d i s t i l l e d w a t e r , is a c c e p t a b l e . C a r e m u s t b e e x e r c i s e d d u r i n g t h e filtration
t o w a s h t h e filter b y d i s c a r d i n g t h e first p o r t i o n o f fluid p a s s i n g t h r o u g h t h e
filter; t h i s p r e c a u t i o n will r e m o v e f i l t e r - b o n d i n g m a t e r i a l , d e t e r g e n t , o r
w a t e r t h a t m a y h a v e b e e n r e t a i n e d in t h e filter.a
a G. E. Gifford, J. Bacteriol. 80, No. 2, 278-279 (1960).
R. D. Cahn, Science 155, 195-197 (1967).
[16]
199
We have found McCoy's 5A to be satisfactory basic media. Extensive

modifications in the salt and sugar levels are frequently and readily made.
Thus, the sugar concentration may be reduced to 100 mg/100 ml and the
bicarbonate and/or osmolarity levels may be varied.
Media pH
A problem in the preparation of tissue culture media is the uncontrolled variability of the final pH under the actual experimental conditions. This, particularly in the instance of bicarbonate buffer systems,
results from: variation in the bicarbonate levels, as found in various serum
components; the intrinsic buffer capacity of the media mixture; variation
in the actual concentration of the CO2 overlay (see section on culture gas);
and the adjustment of pH at room temperature during media preparation
whereas the cells and medium are actually maintained at 36.5 . Thus, if
medium is prepared at 25 with the pH adjusted to 7.30, the actual pH on
incubation at 36.5 will be 7.40. For CO~oicarbonate systems in this range
a differential of 0.087 pH units per degree C increase is observed.
The critical importance of pH shifts to cellular function has been described in detail. 1 For several years, we have used a very simple "tonometer" for pH adjustment that consistently has yielded culture media
poised at any preset or desired pH level (Fig. 1). The tonometer consists
of a 100-ml water-jacketed flask, via which the contained medium is held
at the desired culture temperature--36.5 . A Teflon-coated spin bar is
used to facilitate equilibration. The upper configuration of the tonometer
possesses two ports for the direct flow and circulation of the selected
CO2-air mixture over the stirred flask contents. Normally, the same cylinder of gas is used for the tonometer as for the actual cultures. A third
indwelling port is provided as a thermometer well. Finally, a silicone
stopper is added to both seal the chamber and allow insertion of a combination pH electrode the tip of which is centered in the middle of the
medium. A vaccine stoppered port penetrates the stopper and permits
direct addition via a tuberculin syringe (2-5 ml) of 5% NaHCOa without
loss of equilibration gas.
The bicarbonate is added step by step, being sure that the equilibration
level has been reached at each step before the next addition. Extrapolation from the 50-100 ml tonometer volume to the actual batch volume
yields the correct amount of 5% NaHCO3 required for any desired pH, at
any incubation temperature, with any gas mixture, despite variations in
the buffer capacity of the medium. Its prime disadvantage is that it re10S.
Gailiani, W. F. McLimans, A. Nussbaum, F. Robinson, and O. Roholt, In Vitro 12,

363-372 (1976).
200
BASIC METHODS
[16]
PH
Probe
Il
Thermometer Well
~ _ ~ ~ " ~
Bar"
Fro. 1. Schematic o f the tonometer.
quires about 2 hr to complete the titration and "careless overshoot" can

create problems.
Osmolarity Adjustment
After adjustment of the pH, medium osmolarity is measured with an
appropriate osmometer. Our instrument (g) has a precision of _-_1.0
mosM/kg water. The balance of NaC1 required to achieve the desired
osmolarity is calculated in the following manner.
1. Desired or normal osmolarity of healthy persons = 291.8 -+ 2.6
mosM/kg water. 11
2. Measure the osmolarity of experimental media and calculate the
amount of a special stock NaC1 (1 mg/ml) which must be added to achieve
the desired osmolarity. Thus: 1 mg NaC1/ml = 1 ml stock (mosM) =
32 mosM increase. Therefore:
D-O
32
=X
11 G. H. Hobika, J. L. Evers, and R. Oliveros, J. Med. 7, No. 1, 13-31 (1976).
[16]
MASS C U L T U R E OF M A M M A L I A N CELLS
201
where D(mosM) = Desired mosM

O(mosM) = Observed mosM
X = ml of stock NaC1 (mosM) to be added per milliliter of
medium.
Contamination Control
A vast effort has been expended in pursuit of knowledge relating to the

cultured mammalian cell. Much of it may be of little real value, since the
studies were conducted on cells of questionable origin, cells contaminated
with microbial agents, or indeed comparative studies of different cell
lines, later proven to be all of one common origin. Therefore, great care
and time must be expended in the constant check and monitoring of the
culture system for contaminants, be they bacterial, viral, fungal, or other
mammalian cells being carded in the laboratory.
Critical and mandatory control procedures encompass (1) adherence
to strict aseptic techniques, as practiced in any bacteriology laboratory,
as well as (2) certifying the source of the " s e e d " culture in regards to a
detailed history of the cell line, histologic characteristics, growth curve,
banded marker chromosomes, and glucose 6-phosphate dehydrogenase
profile. The extent of prior attempts to isolate .and/or demonstrate microbial agents should be certified as well. Certainly, aliquots of the culture
should be submitted periodically to experts in the above noted fields for a
characterization check.
Periodically, antibiotic-free media should be prepared and incubated
in culture flasks to check the efficiency of all operational procedures from
media preparation to "cell-less" cultures. This insures "in-house" control of techniques. Ideally, all cultures should be antibiotic-free, but this
cannot be achieved in many of the facilities currently available to the
biochemist.
The problem is of sufficient importance that methods for its evaluation
are covered in detail in this volume [2] and [12].
Culture Gas
While the major goal remains the simplification of the tissue culture
"tool," meaningful studies demand that the mammalian cells be cultured
under physiologic conditions that are nontoxic and reproducible. In these
terms, one of the most important parameters of that environment is the
culture gas. Thus, for example, culture buffers operative without bicarbonate utterly fail to provide sufficient CO2 and, therefore, must'depend on
202
BASIC METHODS
[16]
TABLE III
GAS CONTAMINANTS IN COMMERCIAL BOTTLED GAS
Cylinders,
research grade
(ppm)
O~
Carbon monoxide
Nitrous oxide
Hydrocarbons (as methane)
C02
<50
<l
16
Estimated level
in gas overlay of
cell culture (ppm)
5% CO2 < 2.5 ppm
20% O~ < 0.2 ppm
20% O~ < 3.2 ppm
the cells to provide this gas for the environment. TM Indeed, the buffer
capacity of the CO~/bicarbonate system is perhaps the least important of
its many metabolic control configurations. 12Accordingly, in our laboratories we routinely use the CO~oicarbonate buffer system.
Specialty gas mixtures, medical or biologic grade, are available for use
as culture gas. However, wide variations may exist in the actual pC02
levels of commercial gas mixtures. Such variation can be extensive if
certified gas mixtures are not obtained. In our laboratories, infrared analyses of the actual CO2 content of 27 tanks of uncertified 5% COs in air
showed a mean variation of 0.9% from the tank value noted by the
supplier. The actual range of variation was from 0.1-3.3% with an overall
span of 3.2%. Very low values, 2-3%, may be encountered if the tanks
have not been properly mixed by rolling.
Gaseous contaminants may include particulate matter, microbial
agents, and oil. These are easily removed by in-line filtration. Other contaminants such as hydrocarbons, nitrous oxide, and carbon monoxide are
of concern as toxic or potentially mutagenic agents (Table III).
It is recommended that specialty gas mixtures be procured with clear
specifications as to the permissible level of contamination and variation of
component concentration, i.e., less then 3 ppm moisture, less than 2 ppm
CO, etc. Certification by lot to lot analyses is much less expensive than
tank to tank and just as effective.
Suspension Culture--Cell Lines
A number of investigators have demonstrated that established mammalian cell lines proliferate in agitated fluid suspensions. Initial tech~z W. F. McLimans, in "Growth, Nutrition and Metabolism of Cells in Culture" (G. H.
Rothblat and V. J. Cristofalo, eds.). Vol. 1, Chapter 5, pp. 137-162. Academic Press, New
York, 1972.
[16]
203
niques were introduced through the imaginative efforts of Earle et al., 1~

who reported successful growth of the L strain (No. 929) fibroblast utilizing a conventional rotary shaker equipped with special flasks and media.
Confirming studies with similar techniques were conducted by Kuchler
and Merchant.14 Prior to Earle, Owens et al. 15had demonstrated that the
deBruyn mouse lymphosarcoma MB (T-86157) could be cultured readily
in "tumbling tubes." Despite the fact that only small volumes were used
in this Latterinvestigation, it did suggest that the ability of cells to proliferate in submerged culture may not be restricted to one specific type of cell
or method, an observation since amply confirmed. Thus, a monkey kidney
cell strain could be propogated in roller tubes rotated around their horizontal axis at 40-50 rpm. 16 Concurrently, good growth of ascites tumor
cells was demonstrated in hexagon-shaped roller tubes. 17 Cherry and
Hull TM found that a fluid suspension contained in a round-bottom flask and
agitated by a suspended magnetic stirrer permitted growth of the LLMC
strain. Brown 19 obtained good growth of mammalian cells employing a
"wrist shaker," and McLimans et al. reported satisfactory growth of a
number of cell lines in spinner culture (25-100 ml),2-2~ in the New
Brunswick fermentor (1.5-5 liters)) 3 and in a stainless-steel, waterjacketed, impeller-agitated fermentor (20 liters). 24 Additionally, each system was used effectively for the production of viral particles and
complement-fixing antigen) 5 as well as for growth of anterior pituitary
cells of human origin. ~6
la W. R. Earle, J. C. Bryant, and E. L. Schilling,Ann. N. Y. Acad. Sci. 58, 1000-1011 (1954).
14 R. J. Kuchler and D. J. Merchant, Proc. Soc. Exp. Biol. Med. 92, 803-810 (1956).
15 O. von H. Owens, M. K. Gey, and G. O. Gey, Proc. Am. Assoc. CancerRes. 1, 41 (1953).
~e A. F. Graham and L. Siminovitch, Proc. Soc. Exp. Biol. Med. 89, 326-327 (1955).
17 A. K. Powell, Cancer Res. Campaign, Ann. Rep. 32, 125 (1954).
is W. Cherry and R. N. Hull, Tissue Cult. Assoc. Meet., 1956 p. 9 (1956).
lg A. Brown and F. M. Hardy, "Growth of Venezoelan Equine Encephalomyletis Virus in
Monolayer and Fluid Suspension Cultures of ' L ' Cells, "Maryland Soc. Am. Bacteriol.,
Ft. Detrick, Frederick, 1957.
30 W. F. McLimans, E. V. Davis, F. L. Glover, and G. W. Rake, J. Immunol. 79, No. 5,
425-433 (1957).
31 F. Giardinello, W. F. McLimans, and G. Rake, Appl. Microbiol. 6, No. 1, 30-33 (1958).
zz E. V. Davis, F. Glover, and W. F. McLimans, Proc. Soc. Exp. Biol. Med. 97, 454-456
(1958).
23 W. F. McLimans, F. E. Giardinello, E. V. Davis, C. J. Kucera, and G. W. Rake, J.
Bacteriol. 74, No. 6, 768-774 (1957).
24 D. W. Ziegler, E. V. Davis, W. J. Thomas, and W. F. McLimans, Appl. Microbiol. 6, No.
5, 305-310 (1958).
25 M. T. Suggs, H. L. Casey, D. D. Sligh, A. R. Fodor, and W. F. McLimans, J. Bacteriol.
82, No. 5,789-791 (1961).
28 W. F. McLimans and S. Gailani, in preparation.
204
BASIC METHODS
[16]
A summary of the literature, 27 as reported from nine laboratories,

revealed that 33 cell lines, derived from 14 tissues of human origin, have
been propagated in suspension-type systems. Additionally, at least 11 cell
lines originally established from monkeys, rabbits, mice, and rats have
been propagated in the submerged culture system. Failure to obtain satisfactory growth has been reported in the instance of monkey kidney lines,
LLC-Mk and LLc-MK, and the diploid line WI-38.
Accordingly, most established cell lines can be grown as discrete units
in suspension-type systems employing a variety of conditions, equipment,
and tissue culture media. Subsequent to the original reports of scale-up to
5-20 liter cultures about 15 years ago, 23'24 mass cultivation of such cells
was reported by several laboratories in 5-200 liter systems. 2s
Selection of Mass Culture Systems

Mass suspension cell culture systems are based on the "state of the
art," as borrowed from the bacterial physiologist. The simplest configurations of vessels range from the 30-1000 ml spinner flasks to a 5-15 liter
glass fermentor, or to a 20-200 liter stainless-steel fermentor. The material
presented here is for the most part limited to glass systems of less than 20
liters volume, with a configuration such that they can be operational in the
ordinary laboratory.
Selection of the appropriate size of a fermentor system entails careful
evaluation of the critical needs of the investigator, cost, available reagent
supplies, washroom facility, sterile rooms, hoods, etc. As a rule of thumb,
one calculates the grams of cells, wet weight, required per week and
assumes a biweekly harvest of of the fermentor volume or a weekly
harvest equal to the volume of the unit. The maximum cell population
attainable, while a function of the cell and its environment, is determined
in the " b a t c h " system by the rate of lactate production and the decrease
in pH with concomitant definition of the time interval between medium
changes. Usually this will be found to be at the peak of 1-2 10n cells/ml.
Since 1 x 106 cells equals approximately 1 nag (wet weight), then the
production of 1 g of cells per week would require a harvest twice each
week of 500 mg or 500 x 106 cells. Since the harvest should not exceed
of the fermentor volume, a fermentor volume of 1000 ml will yield a
harvest population of 1 106 cells/ml, a total of 500 mg of cells. Thus, the
27 Data based on a literature survey made by the author in 1964. The list of successful
cultures has, of course, been considerably expanded since that time.
2a G. E. Moore, J. W. Kullen, H. A. Franklin, and N. Kinsley, in "Germ-Free Biology:
Experimental and Clinical Aspects'"(E. A. Mirand and N. Back, eds.),Vol. 3, pp.343-356.
Plenum, New York, 1969.
[16]
205
cultures may be operated continually with the desired 1 or 2 g yield of

cells per week.
The cost of the above operation, exclusive of personnel and/or capital
equipment, will be in the range of $13.74/liter of media or $6.13/g (wet
weight) of cells. The simplest and safest procedure in this example would
be to use two or more 500-ml spinner cultures; thus, accidental microbial
contamination of one vessel would not long curtail the ongoing
investigations.
If much larger quantities of cells are required--10-100 g/week--then
conventional types of fermentors must be used for "batch" operation.
This may involve fermentor volumes of 10-100 liters. Moore e t al. z8 successfully operated, on a batch basis, a commercial, stainless-steel soup
kettle modified for the routine propagation of lymphoid-type cell lines.
One should consider multiple carboy-type spinner systems prior to investment of large sums in more sophisticated fermentors.
Since yield of cells is the prime objective, the most efficient large-scale
fermentors are those that operate on a continuous or chemostat-type principle. Such systems, while requiring extensive instrumentation, allow cell
populations of 6-10 l06 cells/ml (Table IV). This not only produces
savings in cost of medium, but also reduces enormously the size of the
fermentor necessary for the same weekly yield as a "batch" process.
Another major advantage is the more uniform yield of cells than with the
"batch" system.
Basic considerations for operation of a suspension cell culture include
the following: (1) the minimal population of cells per milliliter required to
initiate a culture. This varies with cell lines and may be as low as 25
10a/ml; routinely we use 2-6 105 cells/ml. (2) Calcium and magnesium
levels may be reduced, if excessive clumping occurs. (3) Although spinner
agitation speeds are a function of vessel size and impellor design, speed
should generally be kept low, between 50-150 rpm, in order to avoid
excessive shear forces. (4) The inactivated serum, i.e., heated to 56 for
30 min, is routinely used at a concentration of 10% although lower levels
may be acceptable. (5) The incubation temperature is normally 36.5 . Care
must be taken to insure lack of excessive localized heat transfer from the
matnetic drive assembly.
The simplest procedure for monitoring the growth response of the cells
is the direct counting technique (see this volume [10]). Good cultures
should display readily cellular viabilities of at least 80% or better. Anything less than this suggests an unsatisfactory environment. At times,
viability and state of culture may be markedly improved by trypsinization
of the entire cellular contents and transfer to a new spinner.
Morphologic assessment of suspension cultures can be made best by
206
BASIC METHODS
[16]
TABLE IV
A COMPARISON OF COSTS (LEss LABOR) OF BATCH VS. CONTINUOUS PROCESS OF
PRODUCTION OF MAMMALIAN CELLS AS REFLECTED BY CELL POPULATION,
FERMENTOR VOLUME, AND SERUM CONCENTRATION
IN CULTURE MEDIUM a
Batch b
Fermentor volume (liters)
Maximum cell Population (x 106/ml)
Day of first harvest
Number of harvest days
Total cells harvested (x 109)
Grams cell yield/week (g)
Total media used (liters)
Total serum (10%) used (liters)
Cost of serum-free media
(at $1.46/liter)
Cost of serum
Total media costc
Cost/gram of cells ($)
Continuous a
1
1.4
2-3
11
3.85
2.24
6.5
0.65
$8.54
10
1.4
2-3
I1
38.5
22.5
65.0
6.5
$85.41
10
10
5-6
8
800
467
120
12
$157.68
5.20
13.74
6.13
52.00
137.41
6.11
96.00
253.68
0.54
a All calculations are based on 24-hour generation time for 12 days following establishment of the culture on day 0.
Batch system calculations based on harvest of culture every day.
c Media cost based on 66d/liter for Baxter Water, 78.5d/liter for dry mix, 10d/liter for
salts and antibiotics, and $8.00/liter for serum at concentration of 10%, i.e., price
levels obtaining January 1978.
a Continuous systems are calculated to reflect conditions at the same cell population
level as batch and under conditions wherein cell population was maintained at about
x7 this level. Such levels are reafistic with some cell lines. We have operated units
at 8-10 x 10s cells/ml, while Earle has reported short-term propagation at levels as
high as 20 108 cells/ml.
establishing conventional monolayer subcultures and by periodically det e r m i n i n g t h e efficiency o f p l a t i n g w i t h a t t e n t i o n t o h i s t o l o g i c t y p e s o f
colonies.
In "batch-type" systems, the culture pH should be checked daily and
efforts m a d e t o m a i n t a i n it a b o v e 6.9. T h i s is a c h i e v e d b e s t b y t h e d a l l y
a d d i t i o n o f f r e s h m e d i a in a n a m o u n t g e n e r a l l y e q u a l t o 1 0 - 2 0 % o f t h e
t o t a l v o l u m e . E x c e p t f o r v e r y h i g h c e l l p o p u l a t i o n s , t h i s p r o c e d u r e will
help to maintain growth and circumvent possible exhaustion of a substrate
such as glucose.
Suspension Culture--Primary Cells
A critique of suspension culture systems must include observations
r e l a t i n g t o t h e m o r p h o l o g i c a n d b i o c h e m i c a l s i m i l a r i t y o f e s t a b l i s h e d cell
[16]
207
lines: abnormal chromosome patterns, malignancy (real or imagined), as

well as the undefined culture parameters, i.e., cell dissociation, culture
surface, gas, and nutritional environment. It is therefore important to
attempt to improve the definition and control of the culture environment
so that primary cells, representative of those found in the tissue of origin,
may be maintained in vitro with meaningful and reproducible growth and
function.
One may define an ideal culture system for the mass growth of mammalian cells as " a closed-system suspension culture configuration, capable of supporting the direct growth of normal, primary cells with longterm maintenance of specialized cellular function; as a system achieved
with cells that are minimally exposed to foreign substances or environments in their transition from the in vivo to the in vitro state; as a system
achieved with cells that are grown to high population densities, under
conditions such that the state of the culture may be critically assessed and
controlled via daily monitoring; and finally as a system with an operational efficiency such that the cost and availability of media components
are held at realistic levels."
One approach could well be to achieve growth of primary cells directly
in a homogeneous suspension system in which the definition and control
of the culture environment is easily established--a steady-state culture
system. During the last decade, a great deal of experimental work concerned with the environmental control of mammalian cells in suspensiontype cultures has been reported. However, in all instances the investigations were seriously limited by being restricted in application to established cell lines.
As a consequence, the reports from van Wezel and his associates z9-3~
that mammalian cells could be grown on the surface of a microcarrier,
such as DEAE-Sephadex beads, while being stirred in a suspension system were of great interest. Accordingly, a serious evaluation of the potential of microcarrier suspension for direct culture of primary cells from
differentiated tissue was made in our laboratories. It is here used as a
prototype to define and illustrate the appropriate techniques, although
these methods are undergoing rapid change as new " b e a d s " become
available. 32 The technique is also discussed in this volume [15].
Calf anterior pituitary cells (CAP) were employed as the differentiated
cell type. They were successfully cultivated in the microcarrier suspenz~ A. L. van Wezel, Nature .(London) 216, 64-65 (1967).
a0 A. L. van Wezel, in "Tissue Culture: Methods and Application" (P. F. Kruase, Jr. and
M. K. Patterson, Jr., eds.), Chapter 2, pp. 372-377. Academic Press, New York, 1973.
31 p. van Hemert, D. G. Kilburn, and A. L. van Wezel, Biotechnol. Bioeng. 11, 875-885
(1969).
a2 C-b. Horng and W. F. McLimans, Biotechnol. Bioeng. 17, 713-732 (1975).
208
BASIC METHODS
[16]
sion system. Details regarding the type of cell growth, kinetics of cell
proliferation, inhibitory effect of high bead concentrations, and growthenhancing activity by a cell factor have been presented, a2 The reported
experimental work appears to have established the necessary background
for a direct approach to the long-term, steady-state culture of primary
differentiated cells at high population densities in suspension culture.
The calf anterior pituitary gland, immediately after removal, was isolated, decapsulated, and soaked in a trypsin mediummMcCoy's 5A containing 6.250 IU/ml of Tryptar (j'). After transport to the laboratory, the
tissue was minced and trypsinized in a stirred 50-ml spinner flask containing 15 ml o f " t r y p t a r " medium. After 6 hr of trypsinization, the cells were
readily dissociated with a Pasteur pipette. Typically, from one pituitary
gland, 2-7 107 cells were obtained with a viability above 95%. In other
instances, e.g., mammary and prostate, cultures were initiated better directly from a fine tissue mince mixed with the prepared " b e a d s " ; trypsin
appeared to be deleterious to cells derived from these organs.
McCoy's 5A medium supplemented with 10% heat-inactivated fetal
calf serum was used as growth medium for the pituitary cell cultures. The
media was adjusted to a pH of 7.4 with a gas overlay of 2% COs in air for
normal tissue and 5% COs for tumor tissue. The osmolarity was controlled
at 291-315 _+ 10 mosM.
DEAE-Sephadex was, in the main, used as a microcarrier, although
other beads are available, a3 Prior to use, the beads are sequentially
washed with 0.5 N NaOH, distilled water, 0.5 N HC1, distilled water, and
a phosphate buffer solution, and sterilized by autoclaving at 121 for 20
min. They are stored at 4 . Just prior to use, the phosphate buffer is
replaced with culture medium.
Suspension cultures are carried out, for example, in a spinner flask of
50-100 ml capacity (k,l). The cultures are incubated at 36.5 with a magnetic spin bar adjusted to a speed of about 100 rpm. The spinner flasks are
continuously gassed via an overlay of 2% COs in air. The usual culture
volume is 30 ml. The pH of the culture fluid is monitored daily. If the pH
decreases below 7.0, the culture is usually fed with 10-20 ml of fresh
growth medium. Every 3-4 days, or daily, depending on the activity of the
culture, I ml of the cell/bead suspension is removed and mixed with 3 ml
of phosphate buffer solution contained in a petri dish. The percentage of
beads with cells growing on their surface is checked by microscopic observation. Subsequently, the beads with attached cells are washed with a
phosphate buffer solution and exposed to 4 ml of 0.1 M citric acid while
being incubated at 37 for 1 hr. Following the incubation, the released
aa D. W. Levine, J. S. Wong, D. I. C. Wang, and W. G. Thilly, Somatic Cell Genet. 3, No. 2,
149-155 (1977).
[16]
209
nuclei are washed from the bead/cell surface with the citric acid solution,
stained with gentian violet, and counted in a hemocytometer or by an
electronic counter; method for counting nuclei follows the procedure originally developed by Sanford et al. 34
Interestingly, "naked beads" added to a monolayer culture will pick
up and anchor cells. Conversely, "beads" with anchored cells when transferred from suspension to a flat culture surface will initiate monolayertype clones and monolayers.
For monitoring the morphology and cell types, petri dish cultures are
established from a dilution of the microcarrier suspension cultures. Feeding is done by replacing the acidic spent medium with 3 ml of fresh growth
medium at 2--4 day intervals. Quantitation of colonies, as well as histologic
evaluation, is made after 14 days of incubation in a COz chamber or bell
jar incubator. This constitutes a critical control procedure!
The DEAE-Sephadex bead is transparent in growth medium. Thus,
the morphology of the attached cells can be fixed by the usual methods
and monitored directly via the light or electron microscope.
As noted, petri dish cultures are employed for preliminary assessment
of cell growth under various conditions. Immediately after inoculation,
cells are observed to attach to the bead surface and form a round mass for
about 2 days. Thereafter, the cells appear to spread out and begin to
proliferate. By the 4th day, a partial monolayer of cell growth on the bead
surface can be observed. After 2 weeks of incubation, nearly all of the
beads are completely covered by cell layers. It appears that cells may
move from one bead to an adjacent bead, leading eventually to the formation of bead clumps with cell layers serving as a bridge; cells are predominantly of epithelial morphology when attached to the bead surface. Electron microscopy of cells from these cultures shows the presence of active
mitochondria, endoplasmic reticulum, and tight cell junctions. Some
cytoplasmic projections have been noted between bead and cell, which
probably indicates an active anchorage site, rather than passive physicalchemical attachment of the cells on the bead surface.
Suspension cultures in spinner flasks have been routinely successful.
Within certain limits of bead concentration, cells proliferated in the suspension system with growth on the order of magnitude of a 10-fold cell
population increase in 8 days. Contrary to the expectation that higher
bead concentration should provide greater surface area for cell growth
and a higher incidence of cell-bead collision and attachment, it appears
that with a constant cell inoculum, higher bead concentrations may inhibit
cell growth. This inhibitory effect, as would be expected, can be reduced
84 K. K. Sanford, W. R. Earle, V. J. Evans, H. K. Waltz, and J. E. Shannon, J. Natl. Cancer
Inst. 11, 773-795 (1950-1951).
210
by using proportionately higher quantities of inoculum. There might be a
critical cell-to-bead ratio below which the cells will not show sustained net
growth. By using a rough estimation of 1.6 105 beads/ml in the bead
bed, this ratio was calculated to be about 1 : 5 for the calf anterior pituitary
cells.
In our hands, direct primary cell culture with the microcarrier suspension system and batch feeding can routinely yield a cell population of
about 5 10e cells/ml. However, with such a dense population, nutrient
feeding becomes the limiting factor. Even feeding 3 times a day, one
cannot keep up with the extreme shifts in pH. To circumvent this problem, we ~ave devised a continuous-feed culture system with automatic
pH monitoring and feedback loop, as based on use of a modified "SpinFilter" (rn). Other systems may be equally applicable (n,o).
Employing the "Spin-Filter," a 500-ml microcarrier (DEAE) culture
of primary anterior pituitary cells was continuously maintained, at a high
cell population (4-7 10e cells/ml), for 57 days with excellent control of
the culture parameters. Repetitive subcultures to petri dishes were routinely successful.
Cell debris does not cause significant problems in healthy cultures.
However, freely suspended cells do appear. Whether or not these represent newly divided cells as released from bead surfaces remains to be
established. At times, freely suspended cells may reach a population as
high as 3 x 105/ml and have a viability above 85%. If such a cell population is truly being continuously released from the beads, it might well be
possible to consider these methods for synchronized cell culture studies.
Other systems and configurations for cell culture have been demonstrated, i.e., capillary tubes, 3s plastic or cellular sheets or tubes, 3e multiplate fermentors, 37 and even columns of Sephadex. a8 They are not considered he~-e, since they all appear to suffer from difficulty in quantitating
growth rates and responses. There is also a marked tendency for formation of clumps or cellular masses. Such clumps may, in turn, serve as foci
of necrosis and transformation. Vital to any system is the facility of meaningful monitoring techniques with which to assess the true kinetic or
histogenic profile of the culture. This, in turn, must depend on obtaining
truly representative aliquots.
It is the author's opinion that the microcarrier technique will, in time,
prove to be the best tool for this task.
s5 R. A. Knazek, E M. Gullino, P. O. Kohler, and R. I. Dedrick, Science 178, 65-67 (1972).
See also this volume [15].
se M. D. Jensen, D. F. H. Wallach, and P. Sherwood, J. Theor. Biol. 56, 443-458 (1976).
s7 R. Weiss and J. B. Schleicher, Biotechnol. Bioeng. 10, 601-616 (1968).
aa B. Bergrahm, Biotechnol. Bioeng. 10, 247-251 (1968).
[17]
PROPAGATION
AND
SCALING-UP
OF
CULTURES
211
Acknowledgment
I acknowledge the invaluable and tireless assistance of Chi-byi Horng, Barbara Kwasniewski, Angeline Wasielewski, Frances O. Robinson, Robert Zeigel, Elizabeth Repasky,
W. "Sam" McLimans, Peter Hasenpusch, and William Beers in the investigative phases that
form the bases of this article. I thank Ann M. Gannon for the editing and other assistance.
The United States Public Health Service supported with these grants-in-aid various facets of
this work: 5 R01-ES00030-05, 89013-01, NIH-NCI-G-72-3866, 1 R0"IGM22532-01, and 5 P,26
CA 1865203.
[17] Propagation
and Scaling-Up
of Suspension
Cultures
By RONALD T. ACTON, PAUL A. BARSTAD, and ROBERT K. ZWERNER
I. Introduction
One m a j o r consideration in cell culture is w h e t h e r a given cell line
grows in suspension or m u s t attach to a matrix. Initial attempts at growing
cells in vitro dealt mainly with anchorage-dependent types. Following the
early demonstrations 1~ that cells would multiply in suspension cultures,
m a n y investigators recognized the potential of this a p p r o a c h and sought
to increase the efficiency o f the process (see Perlman et al. 3-6 for review).
F r o m the beginning animal cell suspension culture w a s appealing due to
the anticipation o f utilizing fermentation techniques similar to those used
in cultivating microorganisms. Although the a p p r o a c h is similar in principle the difference in growth p a r a m e t e r s b e t w e e n p r o k a r y o t i c and e u k a r y o tic cells p o s e s unique problems.
An a p p r o a c h will be described here for the growth of cells in suspension culture f r o m 1 ml up to 200 liters, e Although the method was initially
derived for the propagation o f l y m p h o c y t e s , 7 it has been found amenable
to m a s t o c y t o m a , h e p a t o m a , L cells, and various n e r v o u s tissue derived
cell lines. Since cell culture requires a high level of attention to detail,
lo. v. H. Owen, M. K. Gey, and G. O. Gey, Ann. N.Y. Acad. Sci. 58, 1039(1954).
2W. R. Earle, R. L. Schilling, J. C. Bryant, and V. J. Evans, J. Natl. Cancerlnst. 14, 1159
(1954).
3 D. Perlman, Proc. Biochern. 2, 42 (1967).
4 G. E. Moore, Methods Cancer Res. 5, 423 (1970).
5 R. C. Telling and P. J. Radlett, Adv. Appl. Microbiol. 13, 91 (1970).
6R. T. Acton, P. A. Barstad, R. M. Cox, R. K. Zwerncr, K. S. Wise, and J. D. Lynn, in
"Cell Culture and Its Application" (R. Acton and J. Lynn, eds.), p. 129. Academic Press,
New York, 1977.
7 R. K. Zwerner, C. Runyan, R. M. Cox, J. D. Lynn, and R. T. Acton,Biotechnol. Bioeng.
17, 629 (1975).

ISBN 0-12-181958-2
212
BASIC METHODS
[17]
even the more mundane aspects of the approach are described. Experience has taught that inability to grow a particular cell line frequently lies
in the failure of an individual to adhere precisely to the prescribed steps
rather than in the method itself.
II. Cell Culture
1. Recovery of Cells
The initial procedure for culturing continuous lines depends on the
state in which cells are received. Cells may be transported either as viable
cultures or in the frozen state. If cells are received as a growing culture
one simply determines density and viability and places them in fresh
medium, ff received in the frozen state cells }nay be either stored in liquid
nitrogen or placed directly into culture. The ampule containing the frozen
cells is thawed by immediately immersing in a 45 water bath. The ampule
should be agitated to insure that the thawing process is completed in less
than 1 rain. The ampule is immersed in 70% ethanol to reduce the possibility of contamination and transferred to a vertical flow laminar hood. The
neck of the ampule is scored with a small sterile file and the cells transferred to a sterile 15-ml polystyrene screw-cap disposable centrifuge tube
(Coming #25310) using a sterile Pasteur pipette. Ten volumes of culture
medium at 4 are added and the tube is capped and inverted several times.
The cells are pelleted by centrifugation at 400 g for 5 rain. It is important
to conduct these procedures immediately after thawing the cells in order
to remove the cryoprotective agent as quickly as possible. Regardless of
the designated concentration of cells contained in the ampule the cell
density and viability should be determined at this time. The cells are then
apportioned to sterile 25 cm 2 or 75 cm 2 polystyrene tissue culture flasks
(Coming #25100 or 25110, respectively), and the cell concentration is
adjusted to approximately 1 10n cells/ml with medium at 37. An aliquot
is evaluated for microbial contamination. The flasks with slightly loose
caps are then placed in a humidified CO2 incubator at 37. Each day
thereafter the cells are enumerated and evaluated for viability, and the
cell concentration is adjusted by adding new medium. When the culture
volume allows, an aliquot of cells should be frozen and stored in liquid
nitrogen to serve as a source of future stocks and reference point. In the
event the cells grow slowly special attention must be given to include
centrifugation every other day and resuspension in fresh medium, assessing the use of other media or nutrient supplementation and variation
in cell density. 7
[17]
PROPAGATION AND SCALING-UP OF CULTURES
213
2. Optimizing Cell Growth

When cells appear to be growing well attempts should be made to
optimize the rate and density of growth. Maintaining pH within a narrow
range and sustaining the cells in the exponential stage of growth greatly
enhances the yield of cells or cell products per unit volume of medium. In
addition other factors influence the efficacy of cell production in vitro.
A. MEDIA
Several excellent media are available for the growth of established cell
lines. RPMI-1640 supplemented with horse serum (HS) or fetal calf serum
(FCS) (Flow Laboratories, Inc.) has been demonstrated to sustain a wide
variety of cells in culture and is routinely used in our laboratory. While
certain cell lines may have unique nutritional requirements for growth,
most established cell lines grow well in several media and serum supplements, s For example, the BW5147 cell line was obtained from the Cell
Distribution Center, Salk Institute, in Dulbecco's Eagle's Medium supplemented with 10% FCS (Flow Laboratories, Inc.). It readily adapted to
RPMI-1640 supplemented with either 10% FCS or 10% HS. The cell line
grew nearly as well in Minimal Essential Medium (MEM) supplemented
with insulin or zinc and 2% HS as in RPMI-1640 with 10% HS. Subsequently, it was demonstrated that BW5147 grew as well in RPMI1640 + 2% HS. 9 By trial and error and evaluation of growth curves one
should be able to derive an economical culture medium which generates
the maximum number of cells per unit volume in the shortest possible
time.
B . VESSEL CONFIGURATION
One aspect of cell culture frequently overlooked in suspension culture

is the size and configuration of the spinner flask. When cells are brought
from the frozen state to active culture they are first grown in a tissue
culture flask in a humidified COz incubator. This provides acceptable pH,
CO2, and oxygen condkions for growth. Frequently cells exhibit reduced
levels of growth when transferred to a spinner flask due to the dimensions
of the vessel. 7 We routinely overlay the spinner flask immediately after
inoculation of the seed culture with 5% CO2 + 95% air after which the cap
is tightly replaced; thus, the mixture of gases in the head space of the
a j. L. Williams, T. H. Stanton, and R. M. Wolcott, Tissue Antigens 6, 35 (1975). See also
this volume [5].
P. A. Barstad, R. K. Zwerner, and R. T. Acton, Protides Biol. Fluids, Proc. Colloq. 25,
637 (1978).
214
BASIC METHODS
[17]
vessel must be sufficient to sustain growth until the cells are evaluated
again. As one progresses to larger volumes the amount of oxygen available to the cells becomes a limiting factor. For volumes between 50 ml and
6 liters the Bellco-type spinner flasks (Bellco Glass, Inc.) offer the largest
volume above the culture and provide the best system for suspension
culture. For volumes above 6 liters the configuration of vessels from most
manufacturers are such that a gas mixture must be added continuously to
obtain optimal growth. For some cell lines a fermentation-type vessel in
which pH and dissolved oxygen are continuously monitored and controlled is more effective. These vessels are available in sizes from 1 liter to
14 liters with various optional instrumentation (New Brunswick Scientific
Co., Inc., Edison, New Jersey).
C. DEFINING GROWTH PARAMETERS
In most suspension culture operations cells are propagated under

semicontinuous culture conditions. This mode of operation represents
essentially a "feast-famine" situation in which cells are inoculated into
fresh medium at a relatively low density and overlayed with a gas mixture
of CO2 and air. Growth continues until a build-up of metabolic products or
a depletion of nutrients slows or stops cell growth. In order to efficiently
produce cells in vitro, the factors limiting cell growth must be understood
and ultimately controlled. An examination of the events which occur in a
theoretical, semicontinuous suspension culture of mammalian cells indicates which parameters can be controlled to optimize growth. A growth
curve of mammalian cells in suspension culture is a semilogafithmic sigmoidal plot of cell density (cells per milliliter of culture fluid) as a function
of time. Figure 1 is a theoretical mammalian cell growth curve. Segment 1
of the curve is called the "lag phase" of growth. Cells in this phase
undergo little or no division; i.e., the slope of the growth curve for the lag
phase is approximately 0. In segment 2 the slope of the curve changes
from 0 to a positive value, as the cells enter the terminal lag phase. In this
phase cell division has begun and its rate is constantly increasing. When
the slope of the curve reaches a maximum value (segment 3), the cells are
in the exponential phase of growth and are undergoing division at the
maximum rate. The cells continue growing at this rate until they enter the
"early stationary phase," segment 4. In this phase the cells' division rate
decreases. This decrease may be due to nutrient depletion or accumulation of deleterious waste products. The cells will eventually enter the
"stationary phase" of growth, segment 5, where the slope of the curve
once again approaches 0. If the cells remain in the stationary phase too
long, division ceases and a net decrease in the cell number results.
[17]
PROPAGATION
AND
SCALING-UP
OF
CULTURES
215
iJ
8
6
E
i
c
2
o
8
\ \
10 5
I
0
I
24
I
48
I
72
I
96
I
120
Culture Time (Hours)

FIG.
1.
Theoretical growth curve o f mammalian cells in suspension culture.
Figure 2 illustrates that mammalian cells grown in suspension culture

do indeed perform as just described. This figure defines the growth properties of the murine lymphoblastoid cell line $49.1 in a 4-liter spinner
flask. In this experiment the culture was started at a concentration of
approximately 4.8 105 cells/ml. For 24 hr after inoculation the cells
divided very slowly, as expected of cells in the lag phase of growth. Soon
after this interval, the cells entered the exponential phase of growth where
an average doubling time of 17 hr was observed. At the maximum density
of 2.3 x 10e cells/ml, cell division ceased and a reduction in cell number
was observed over the next 14 hr. Thus, the cells had entered the stationary phase of the growth curve (segment 5, Fig. 1). Cells taken from the
stationary phase and reinoculated into fresh medium generally fail to divide until 2~ ~8 hr have elapsed. 7 However, if cells are placed in fresh
medium from the exponential or early stationary phase of growth, they
continue to divide at a rapid rate and do not enter a lag phase.
216
BASIC METHODS
[17]
I0i
8
6
x_.x~.x..x ~
o
.)
~- 106
o
i'
~
"'x ~"x"x.
)zo~,
7.5
"x''x''x''x'~
6.s
0
(.)
~4
~8
7,z
9,e
Culture Timo (Hours)
FIG. 2. Growth of $49.1 in a 4-liter spinner flask (Wheaton Scientific). The medium was
RPMI-1640 supplemented with 10% HS. The culture was overlayed with a mixture of 5%
COz + 95% air and capped immediately.
A set of conditions has been established which proves useful for the
growth of cell lines in suspension culture. 6,1," When culturing mammalian cells according to this scheme, cells are maintained in concentrations
between late exponential and early stationary phases of growth by the
addition of fresh media.
In practice one may establish a growth curve and continuous culture
conditions by the following method:
1. Begin cultures at several initial densities starting at some concentration, e.g., 10~ cells/ml, using rapidly dividing cells. The optimal inoculation point is the lowest initial concentration that results in a very short or
no lag phase.
~op. A. Barstad, S. L. Henley, R. M. Cox, J. D. Lynn, and R. T. Acton, Proc. Soc. Exp.
Biol. Med. 155, 296 (1977).
, R . T. Acton and J. D. Lynn, Adv. Biochem. Eng. 7, 85 (1977).
[17]
217
i#
8
~i
'
,
'
4OOml
8C'Oml
260_mI 6.~0~I
6000 ~
i
1
1 1200Oral
15Oral 200m
Culture Time (doys)
FIG. 3. Scaling-up the growth of $49.1 from a tissue culture flask to growth in a 12-fiter
culture vessel. Growth from the 150-ml stage to the 2600-ml stage was conducted in Bellco
spinner flasks. Growth from 6.5-liter to 12-liter was conducted in a 14-liter New Brunswick
fermentor (New Brunswick Scientific Co.).
2. Begin cultures at the optimal inoculation point.

3. Monitor the slope of the exponential phase.
4. When the slope decreases to 50% of the maximum value, add fresh
media to dilute cells to the optimal inoculation density once more.
By adhering to these guidelines it is possible to culture cells in suspension in a reproducible fashion as shown in Fig. 3.12 The $49.1 cell line was
rapidly scaled up from a tissue culture flask to a 12-liter suspension culture. After an initial adjustment to the 250-ml spinner flask, the cells were
maintained in the exponential phase of growth with doubling times averaging 17 hr. For most of the 10-day culture period, maximum cell densities
ranged between 2-3 x 106 cells/ml.
3. Evaluation of Culture Status

Throughout the culture period several parameters can be utilized to
evaluate progress. The number of cells per unit volume and viability
should be determined at least every 24 hr. At times it may be informative
12j. D. Lynn and R. T. Acton, Biotechnol. Bioeng. 17, 659 (1975).
218
BASIC METHODS
[17]
to determine cell size distribution as well as to evaluate some phenotypic

properties if this latter capability is available. These parameters can be
measured in the following manner.
A . DETERMINATION OF C E L L DENSITY AND VIABILITY
Upon removal from the magnetic stirrer and before the cells settle, 0.5
ml of the culture is aseptically removed by use of a sterile 1-ml pipette and
placed in a 12 x 75 mm RTU culture tube (BD #7813 Becton Dickenson
and Co. Rutherford, New Jersey). A 100-/.d aliquot is mixed with 100/zl of
a solution of 0.4% trypan blue in normal saline (#525 Grand Island Biological Co.; Grand Island, New York). This high concentration of trypan blue
is necessary to counteract the affinity of the dye for proteins in solution. TM
An alternative dye is erythrosine B which has a greater alfnity for nonviable cells than the soluble proteins in the culture fluid. ~3 Cells are then
counted in a standard 0.1-mm deep hemocytometer bright-lipe counting
chamber (American Optical Corp. Buffalo, New York). By incorporating
one of the two vital stains into the diluting fluid as described the viability
of cells can be obtained simultaneously during cell enumeration. In the
hemocytometer viable cells will appear as bright circles whereas dead
cells appear as dark circles, often with irregular edges.
B . PHENOTYPIC ANALYSIS
Several cell surface components have been shown to vary in expression during various stages of the cell cycle."-ls Although it is impractical
to maintain cells in suspension culture in synchronous growth, this condition is approachable by close attention to the culture procedures previously described. If the main interest, for example, is cell surface components, it is important to evaluate the particular property of interest during
growth in order to determine the optimal time to harvest cells for maximal
yield. For example, Thy-1 alloantigen expression on murine lymphoblastoid cells is lowest during the lag and stationary phases of growth, T M and
maximal expression occurs when cells are maintained in the exponential
phase of growth. One method to quantitate the amount of cell surface
component on a cell is the absorption of antiserum as detected by the
~aH. J. Phillips, in "Tissue Culture Methods and Application" (P. Krase, Jr. and M. Patterson, Jr., eds.), p. 406. Academic Press, New York, 1973.
~4R. K. Zwerner and R. T. Acton, J. Exp. Med. 142, 378 (1975).
~s M. Cikes, J. Natl. Cancer Inst. 45, 979 (1970).
~6M. Cikes and G. Klein, J. Natl. Cancer Inst. 49, 1599 (1972).
~R. A. Lerner, M. B. S. OIdstone, and N. R. Cooper, Proc. Natl. Acad. Sci. U.S.A. 68,
2584 (1971).
~aM. A. Pellegrino, S. Ferrone, P. G. Natali, A. Pellegrino, and R. A. Reisfeld, J. Immunol.
108, 573 (1972).
[17]
219
cytotoxicity assay. 19--21By comparing the ability of cells to absorb a standard antiserum during various stages of the growth curve, one can determine the exact stage for harvest that yields the highest concentration of a
particular cell surface component. 7.22
As previously indicated, there are a number of factors that affect the
rate of cell growth. In the initial phase of optimizing cell growth one
should determine which culture medium produces the maximum number
of cells per unit volume in the shortest possible time. However, it is also
important to determine if the culture medium chosen also yields the highest quantity of the cell ~omponent under study. The type of serum supplement used in culturing cells can influence the expression of routine
lymphoblastoid cell surface components. For example, the L251A cell
line had a 15-fold greater expression of the routine thymus leukemia antigen (TL) when grown in medium supplemented with 10% fetal calf serum
instead of 10% horse serum.~4 Although it is currently difficult to ascertain
at this point what factors in serum affect the expression of cell surface
components, it is important to evaluate the culture medium selected with
regard to the particular phenotypic property of interest.
4. Quality Control
A . MICROBIAL C O N T A M I N A T I O N
Contamination by microorganisms is the most frequent problem in cell

culture. It is important throughout the maintenance of cells in culture to
routinely monitor the sterility of media as well as the culture. 23 We have
found it convenient to evaluate cultures and media by inoculation into
thioglycolate broth and streaking onto blood agar plates (Grand Island
Biological Co.) that are held at 37 . In addition, the cultures are streaked
onto a tube of Sabouraud's dextrose agar (Grand Island Biological Co.)
and held at room temperature as well as 37 . Most contaminants are
detected by these procedures within 24--48 hr.
Mycoplasmal contamination is more difficult to access but can significantly alter the phenotype or growth properties of cells. 24 Some of these
19p. Gofer and P. O'Gorman, Transplant. Bull. 3, 142 (1956).
20 E. A. Boise, M. Mijazawa, T. Aoki, and L. J. Old,Proc. R. Soc. London, Ser. B 170, 175
(1968).
21 H. Wigzell, Transplantation 3, 423 (1967).
22M. A. Pellcgrino, S. Ferrone, and A. Pellegrino, Proc. Soc. Exp. Biol. Med. 139, 434
(1972).
23 "Difco Manual," 9th ed. Difco Laboratories, Detroit, Michigan, 1977.
24M. F. Barile, in "Cell Culture and Its Application" (R. Acton and J. Lynn, eds.), p. 291.
220
BASIC METHODS
[17]
organisms can be detected by direct culture procedures in which aliquots

of the cell culture are inoculated into mycoplasma broth and agar (Flow
Laboratories, Inc., Mycoplasma Isolation Kit #3006800). Cells also can
become contaminated with nonculturable strains of mycoplasma which
dictate the use of indirect procedures to detect their presence. These
include binding of fluorescent antibodies specific for various strains of
mycoplasma, uracil/uridine ratios, and DNA staining procedures, as well
as a variety of other less commonly used approaches. 24
B. KARYOLOGY OF CELLS
When a cell line is to be utilized for long periods of time, a periodic

analysis of cell surface markers and karyotype should be performed.
When several cell lines are being carded in the laboratory at one time, it is
important to guard against cross-contamination of the cell lines. In addition to cell surface phenotype analysis, another criteria for monitoring cell
populations is karyotypic analysis. 25 The fact that the chromosome constitutions of cells differ between species as well as certain cell lines allows
one a convenient means of monitoring cross-contamination. There are
available a number of procedures for preparation and analysis of cell
karyotypes.25
C. F R E E Z I N G AND STORAGE
After the optimal growth parameters for a given cell line have been
established, aliquots of the culture should be frozen and cryopreserved.
Aliquots of cells are taken while in exponential growth, centrifuged at
400g for 10 min, and resuspended at 4 in either fetal calf serum or horse
serum containing 5% dimethylsulfoxide (Fisher Scientific Co.) to a concentration of 1-5 10* cells/ml in sterile, 2-ml, prescored cryule ampules
(Wheaton 200 Brand #651486, Wheaton Scientific). The tip of the ampule
is sealed in a hot flame by use of the pull-seal method. The ampules are
frozen at a controlled rate of l/min in a BF-4-1 freezing chamber regulated by a Linde BF-6 controller (Union Carbide Corp., Linde Division).
If a controlled-rate freezing device is not available the ampules can be
placed in a small Styrofoam box lined with cotton. The lid is taped into
place, and the carton is placed in a - 7 0 freezer overnight. After freezing,
the ampules are stored in liquid nitrogen. The next day, an ampule or two
of the frozen cells are placed in culture to evaluate the freezing process by
determining viability, growth properties, and possible microbial contamination. Periodically throughout the maintenance of a given stock of cells,
z5T. C. Hsu, in "Tissue Culture Methods and Application" (P. Krase, Jr. and M. Patterson,
Jr., eds.), p. 764. Academic Press, New York, 1973.
[18]
HARVESTING
PRODUCTS
OF
CELL
GROWTH
221
aliquots should be frozen for future reference points. It is also important

to maintain numerous vials of each cell line under cyropreservation, preferably in multiple locations, to protect against loss of the line should
contamination o r other accidents occur depleting the cultured stock.
5. Summary
The general approach discussed here is one that should serve as a
guide for the culture of almost any type of cell in suspension culture.
Following the initiation of cells in culture, it is advised that optimal conditions for growth be determined. In order to maintain the integrity of the
cells and to protect the time invested in their growth it is also mandatory
that one effect adequate quality-control measures to assure that cells are
maintained in a healthy state, free from contamination, and that aliquots
are stored for future reference and use. By adherence to these relatively
simple rules, one will find that cells in culture can be a most valuable tool
in molecular biology.
Acknowledgments
This work was supported by U.S. Public Health Service Grants CA 15338, CA 18609,
and CA 13148 from the NCI, IM33C from the American Cancer Society, GB-43575X from
the Human Cell Biology section of the NSF and the Diabetes Trust Fund. The work of
R. T. A. was done during the tenure of an Established Investigatorship of the American
Heart Association. P. A. B. was supported in part by National Service Award F32-AI 05423
from the NIAID. R. K. Z. was supported in part by National Research Service Award
T32-GM 07561 from the NIGMS.
[18]
H a r v e s t i n g t h e P r o d u c t s o f Cell G r o w t h
By ROBERT K. Z W E R N E R , K I M S. W I S E , and RONALD T. ACTION
Rapid and efficient harvesting of cultured cells and their products is a

major problem associated with biological investigations of cells. Depending on the particular needs of an investigator this process may be far more
costly and time consuming than growing cells. The process described here
is designed to provide subcellular organelles and products such as virus
and mycoplasma in the spent medium by harvesting rapidly and in high
yield. 1,2
1 M. J. Crumpton and D. Snary, Contemp. Top. Mol. Immunol. 3, 27 (1974).
2 D. F. H. Wallach and P. Sun Uln, Biochim. Biophys. Acta 30, 211 (1973).

ISBN 0-12-181958-2
222
BASIC METHODS
[18]
I. Acquisition of SubceUular OrganeUes

Figure 1 summarizes the product acquisition scheme. Cells are separated from the medium by centrifugation at 650 g for 20 rain in a Beckman
J6 centrifuge utilizing l-liter screw-cap polyproplylene centrifuge bottles
(Beckman Instruments, Inc., No. 878564) in a JS-4.2 SW rotor. The spent
medium is decanted and stored at 4 or processed further as described
below. The noted speed of centrifugation is effective in forming a solid but
not tightly packed cell pellet thereby permitting decantation of the culture
medium with minimal cell loss. Centrifugation at greater gravitational
forces packs cells too firmly, causing problems in their resuspension, and
CELL SUSPENSION
8~ @ 2.5 x 10~cells/ml
Centrifuge @ 650 1- 20 rain
BeCkman J6, JS-4.2 rotor
CELL SLURRY
CELL SUPERNATANT
I Wash in 10 mM Tris-0.15 M NaCI pH 72

Centrifuge @ 650 9-20 min
Cenfhfuge @ 6.0(X)E-30 min

I Sorva RC.2, GSA rotor
SUPERNATANT
CELL DEBRIS
CELLS
Re~uspendto 1 x 10= cells/ml
in Tris-NaCI
SUPERNATANT
Disrupt
1) Continuous flow zonal centnfugation
Beckman L5-50, CF-32 rotor

I
(30,000 rpm)
SUPERNATANT
2} Elute in 10-ml fractions
I
I
with 55% sucrose
p113-1.18 fractions
pooted and diluted
with STE buffer
POOLED-VIRUS
CONTAINING FRACTIONS
Centrifuge @ 100.(~0 g-1 hr
I Beckman L5-50. SW 27 rotor
PELLET
SUPERNATANT
1) Resuspend,n STE buffer

Centrifuge @ 100.0000-90 rain
through 5-40~ KT gradient
BeCkman L5-50, SW 27 rotor
2) ColleCt 1,115-1 17 band

3) Dilute with STE buffer
Centrifuge 100,000 p-1 hr
Beckman L5-50 SW 40 rotor
SUPERNATANT
Cm~trifuge@ 4809-20 mm
Soi~'alIRC-2, SS-34 rotor
NUCLEI AND
NONDISRUPTED
CELLS
I Resuspend Jn Tr~s-NaCI
Dounce homogenize
Centrifuge @ 480 g-20 rain
I
SUPERNATANT
NUCLEAR
POOL
PELLET
i
I
Cenlrifuge @ 4000B-20 mtn
SorvallRC-2, SS-34 rotor
SUPERNATANT
SUPERNATANT
POOL
MITOCHONDRIAL
Resuspend in TrlS-NaCI
Dounce homogenize
Centrifuge (~ 4000 g-20 rain
MITOCHONDRIAL
PELLET
Centrifuge @ 20.0001-30 mm
IISorva#RC-2,
SS-34 rotor
I
I
SUPERNATANT
I
MEMBRANE
PELLET
MuLV
PELLET
FIG. 1. Acquisition scheme for products of cell growth.
[18]
HARVESTING PRODUCTS OF CELL GROWTH
IOC
8C
223
I00
8O
6c
t~
.o
~ 4c
40
2o
2O
,~o
2~o
36o
,~o
Cell Disruplion Pressure(Ib/in=)
FIG. 2. The continuous and controlled-rate disruption o f E L A murine ]ymphoblastoid cell

line. A suspension of 1 x 108 cells/ml was pumped at a rate of approximately 2 liters/hr over
a range of back pressures on the disrupting valve.
may damage the cells due to shearing effects. The cells are resuspended in
one-half the original volume of Tris-NaCl buffer (10 mM Tris chloride and
0.15 M NaCI at pH 7.4) and centrifuged as before to remove residual
culture medium. Following this procedure 95% or more of the cells in the
starting suspension culture can be harvested with a viability greater than
95% as determined by trypan blue exclusion. The cells are resuspended in
Tris-NaCl buffer to a concentration of 1 x 108 cells/ml.
To obtain subcellular organelles, cells must be disrupted in a manner
preserving the integrity of these structures and allowing effective separation into pure fractions. We have found the Stansted Cell Disrupter
equipped with a model AO 612 air-driven hydraulic pump and a model 716
disrupting valve (Energy Service Co., Washington, D.C.) an excellent
instrument for this purpose. This instrument provides a means for the
continuous and controllable disruption of a wide variety of cell types.l'3'4
Disruption is caused by the shearing effect of the fluid as it passes under
pressure through an orifice whose dimensions are controlled by air
pressure on a piston. For each cell line a calibration curve must be determined to ascertain a disrupting pressure that results in the highest yield of
the desired product. Figure 2 represents a typical controlled disruption
profile for the murine lymphoblastoid cell line EL4. Disruption is measured by the severity of total cellular disintegration and cell viability as
a B. M. Wright, A. J. Eduardo, and V. E. Jones, J. lmmunol. Methods 4, 281, (1974).
4 R. T. Acton, P. A. Barstad, R. M. Cox, R. K. Zwerner, K. S. Wise, and J. D. Lynn, in
"Cell Culture and Its Application" (R. T. Acton and J. D. Lynn, eds.), p. 129. Academic
Press, New York, 1977.
224
BASIC METHODS
[18]
determined by trypan blue exclusion. The cell suspension was pumped

through the orifice at pressures varying between 50 and 375 psi. At higher
pressures there was increased total disruption as evidenced by a reduction
in the number of cells discernable under phase microscopy. With increased pressure, a large proportion of the remaining cells took up trypan
blue, indicating membrane damage. The optimal yield of plasma membranes from several murine lymphoblastoid cell lines was obtained under
conditions that result in 50% reduction in cell number with the remaining
cells being less than 10% viable. For the EL4 line the cell disruption
pressure of choice as observed from the plot in Fig. 2 was found to be 200
psi.
Following disruption of the cells, separation of the subcellular components is accomplished by differential centrifugation in 50-ml capped centrifuge tubes (Fisher Scientific Co., Cat. #5-529C) using a Sorvall RC-2
centrifuge with a SS-34 head. Nondisrupted cells and nuclei are first removed by centrifugation at 480 g for 20 min. The resulting pellet is washed
twice by resuspending in Tris-NaC1 buffer with a Dounce homogenizer
with a loose-fitting Teflon pestle. This step disperses clumped material
and releases entrapped subcellular organelles from nondisrupted cells.
The initial and subsequent wash supernatant fluids are combined and
centrifuged at 4000 g for 20 min. The resulting pellet is washed twice in
Tris-NaC1 buffer, with homogenization, and a mitochondrial pellet generated. The supernatant fluids from this step are combined and centrifuged
at 20,000 g for 30 min resulting in a microsomal pellet containing plasma
membrane. Should one desire, the latter supernatant can be subjected to
centrifugation at 100,000 g for 90 min in an L5-50 Beckman centrifuge with
a Beckman Type 45 fixed-angle titanium rotor, resulting in a ribosomal
pellet. The final supernatant liquid, containing a variety of soluble enzymes, is also available for further purification procedures.
The interests of our laboratory are directed mainly toward lymphocyte plasma membrane cell surface components which are utilized as an
indication of the efficiency for the procedures described. As can be seen in
Table I, this fractionation procedure results in a 45% yield of plasma
membrane as judged by analyzing the Thy-1 differentiation alloantigen, a
T-lymphocyte cell surface marker; 5 purification is almost 10-fold. The
uniformity of this membrane preparation may be assessed by isopycnic
sedimentation techniques. This may be accomplished by resuspending the
microsomal pellet in STE buffer (0.15 M NaC1, 1 mM ethylene diaminetetraacetic acid, 0.01 M Tris chloride; pH 7.4) and subjecting the sample
to isopycnic centrifugation on linear gradients of 5-40% (w/w) potassium
tartrate in STE buffer, using a Beckman SW27 swinging-bucket rotor in a
5 R. K. Zwerner, P. A. Barstad, and R. T. Acton, J. Exp. Med. 146, 986 (1977).
[18]
225
TABLE 1
DISTRIBUTION OF THY-I.1 ACTIVITY IN FRACTIONSOF DISRUPTED BW5147 CELLS
Fraction
Disrupted cells
400 g Supernatant
400 g Pellet
4000 g Supernatant
4000 g Pellet
20,000 g Supernatant
20,000 g Pellet
Total
protein
(mg)
Total
activity
(Thy-1 units)
Protein
(%)
Purification
Yield
(%)
2296
1019
1085
800
95
796
105
45.6
38.5
16.3
29.0
7.4
6.7
21.9
-44
47
35
4
35
5
-1.8
0.7
1.7
3.7
0.4
9.9
-79
34
60
15
14
45
Cell disruption pressure of 275 psi was utilized.
Beckman L5-50 centrifuge at 4 for 90 min at 100,000 g. The gradients are

eluted in 1-ml fractions. Densities are determined by refractometry (Abbe
Refractometer, Fisher Scientific Co., Cat. No. 13-964), the fractions
dialyzed against phosphate-buffered saline (PBS), and the volumes adjusted to 2 ml with PBS. The protein content and relative activity of a
membrane marker should be determined in each fraction.
Density gradients have not proven advantageous in enhancing plasma
membrane purification from murine lymphoblastoid cell lines. As shown
in Fig. 3, Thy-1.1 activity is associated with the predominant protein band
in a potassium tartrate gradient at p = 1.14-1.16 g/cm3. Although complete recovery of the total Thy-1.1 activity applied to the gradient could
be achieved, no significant additional purification over the microsomal
pellet was obtained. By this criterion the microsomal pellet represented a
rather homogeneous preparation.
II. Acquisition of Components from the Culture Medium
In addition to providing quantities of subcellular components, suspension cultures of mammalian cells offer a rich source of products normally
released by cells into growth medium that can be purified by appropriate
procedures. One major advantage of isolating cellular products from the
spent medium is that acquisition prior to cell damage simplifies purification and minimizes enzymic degradation of cellular products associated
with cell-disruptive procedures. Suspension culture medium of certain
murine lymphoblastoid cell lines can be a ready source of endogenous
murine leukemia virus (MuLV). In addition, mycoplasmas, either intentionally or unintentionally present in lymphoblastoid cell lines, also reside
226
BASE M E T H O D S
[18]
1.24
,~
,oo I
I '~
,.,,
o~
,~-
U
tw
IM
,.14
z ~ -
9< ~
82
WI.12
a
5W
,.,o
t o oa.~ 34 -~
>-
1.08
2U.I
i oC
LOS, ~
0
~
5
~-..9-~,-~-.~-~,.
10
15
20
25
FRACTION NUMBER
FIG. 3. Isopycnic centrifugation o f purified BW5147 m e m b r a n e s on a p o t a s s i u m tartrate
gradient, showing coincidence of Thy-l.l antigenic activity with protein in membranecontainingfractions.

in the cellular supernatant of infected cultures. The isolation and partial
purification of these two components from BW5147 cells is outlined to
illustrate fundamentals of processing supernatant products.
1. General Considerations of Virus Acquisition

Certain conditions must be met for successful recovery of endogenous
MuLV from murine lymphoblastoid cultures. First, the cells must produce virus in sufficient quantities. Second, culture medium requirements
should be determined that insure good cell viability, high virus production, and lack of interference during purification. The type and concentration of serum is important in meeting these criteria. As an illustration,
BW5147 cells grow well in medium supplemented with 2% fetal calf serum
(FCS), but virus production is much reduced compared to levels obtained
with medium supplemented with 10% FCS. In addition 2% FCS increases
the fragility of cells, which leads to excessive fragmentation during harvesting that physically and enzymically interferes with virus purification.
Horse serum is inappropriate because of its high lipid content that decreases virus yield during purification and results in highly impure virus
preparation. For this cell line, medium containing 5% FCS is an appropriate compromise resulting in high virus production and efficient harvest of
[18]
227
cells. A third general condition for purification of supernatant products is

that cell contamination be closely monitored. Mycoplasmas that are not
cultivatable by ordinary techniques e will grow in suspension cultures of
lymphoblastoid cells. Although we have found no indication that mycoplasma contamination of BW5147 cell cultures decreases MuLV production, the purification of virus is hampered by the presence of mycoplasma
in supernatant liquid from the culture.
2. Isolation Procedures
The low-speed cell supernatant fluid obtained from the initial centrifugation step (see Fig. 1) is further centrifuged in 250 ml polypropylene
bottles (Fisher Scientific Co., Cat. No. 5579) at 6000 g for 30 min at 4 in a
Sorvall RC-2 centrifuge with a GSA rotor to remove residual cells and
cellular debris.
A step gradient is formed in a Beckmart CF-32 continuous-flow zonal
rotor by sequential addition of STE buffer (200 ml), 15% (w/w) sucrose
(120 ml), 43% sucrose (110 ml), and 50% sucrose until STE buffer is eluted
from the rotor. Taking care not to disturb the sediment, the clarified
supernatant liquid is removed and continuously pumped at 2.5 liters/hr
into the rotor spinning at 30,000 rpm. After the sample has been applied,
STE buffer is pumped through the rotor for about an hour to allow partial
equilibration of the sedimenting virus. The gradient is then eluted from the
rotor by pumping 55% sucrose into the bottom of the gradient, and 10-ml
fractions are collected. Virus is found distributed in the gradient at densities between 1.13-1.18 g/cm ~. The fractions are pooled, diluted, and
centrifuged in aellulose nitrate tubes (Beckman No. 302237) at 100,000 g
in a Beckman SW27 bucket rotor at 4 for 60 min.
If cultures are contaminated with mycoplasma, this pellet may contain
large quantities of these organisms. Virus can be further purified by
isopycnic centrifugation on linear gradients of potassium tartrate. Prior to
this step, thorough dispersion of the 100,000 g pellet in STE buffer is
necessary and may be carded out either by vigorous homogenization in a
Dounce homogenizer fitted with a Teflon pestle or by gentle sonication.
The latter technique should be carefully controlled and possibly avoided if
structural integrity of the virus is essential. We have observed that envelope components of MuLV may be disrupted by excessive sonication.
This suspension is applied to a linear gradient of 5--40% (w/w) potassium
tartrate in STE buffer, and isopycnic centrifugation is conducted in celM. F. Baffle, in "Cell Culture and Its Application" (R. T. Acton and J. D. Lynn, eds.), p.
291. Academic Press, New York, 1977.
228
BASIC METHODS
[18]
lulose nitrate tubes (Beckman No. 302237) at 100,000 g in a Beckman

SW27 bucket rotor at 4 for 90 min. Potassium tartrate offers an advantage
over sucrose gradients in the shorter time needed to reach equilibrium,
due to the lower viscosity of this solution. A visible virus band can be
seen at a density between 1.15-1.17 g/cm 3. Mycoplasma from infected
cultures band at p = 1.20-1.24 g/cm3, r The virus bands removed from
gradients by aspiration with a Pasteur pipette or by gradient elution are
diluted 1 : 3 with STE in Beckman cellulose nitrate tubes (No. 331101),
and centrifuged at, 100,000 g for 1 hr in a Beckrnan SW40 rotor at 4 to
remove tartrate and concentrate the virus. Identification of virus is routinely performed by electron microscopy using negative staining with
phosphotungstic acid. For additional virus identification, or more sensitive determination of virus distribution on tartrate gradients when the viral
band is not visible, immunological methods 8 or determination by UV absorption may be employed.
3. Other Methods
A second method for treating large volumes, i.e., 5 liters or more,

requires concentration of the cell supernatant fluid (Fig. 1) prior to clarification. Filtration with a Diaflo DC10 continuous recirculating hollow-fiber
ultrafiltration system (Amicon Corp.) with a 100,000 nominal molecular
weight cut-off provides a rapid procedure for concentrating the macrosolute up to 50-fold. The concentrate can be clarified and, depending on the
final volume obtained, the virus may be purified by zonal centrifugation or
by tartrate gradient isopycnic centrifugation as previously described.
For acquisition of smaller molecular weight supernatant components,
hollow-fiber ultrafiltration can provide a simple method for concentrating
molecules of different sizes by sequential filtration through fibers of varying retention limits.
If small amounts of virus are needed or if the cell line is an excellent
producer of virus, smaller volumes of culture medium may suffice. In that
case, the cell supernatant is clarified at 6000 g as noted. However, virus
from the clarified supernatant liquid can be sedimented by centrifugation
in 71-ml polycarbonate bottles (Beckman No. 339168) at 100,000 g for 45
min at 4 in a Beckman type 45 Ti fixed-angle rotor. The virus pellet is
then dispersed and applied to a tartrate gradient for final purification.
7 K. S. Wise, G. H. Cassell, and R. T. Acton, Proc. Natl. Acad. Sci. U.S.A.
(1978).
a K. S. Wise and R. T. Acton, Protides Fluids, Proc. Colloq. 25, 707 (1977).
75, 4479
[18]
H A R V E S T I N PRODUCTS
G
OF CELL GROWTH
229
III. Comments
The procedures described were designed to allow the effective purification o f cells, subcellular organelles, and components from suspension
culture medium. By varying this general approach one should be able to
harvest products from the growth of various cell types in suspension
culture in volumes of 1-12 liters. When several products are utilized from
a given suspension culture it b e c o m e s a very economical approach and
enhances the possibility of gaining further information on the molecular
biology o f the cell.
Acknowledgments
This work was supported by U.S. Public Health Service Grants CA 15338, CA 18609,
and CA 13148from the NCI, IM33C from the American Cancer Society, GB-43575Xfrom
the Human Cell Biology section of the NSF, and the Diabetes Trust Fund. R. K. Z. and
K. S. W. were supported in part by an institutional National Research Service Award No. T32
GM07561 from the NIGMS. This work by R. T. A. was done during tenure of an Established
Investigatorship of the American Heart Association.
[19]
CELL CYCLE ANALYSIS BY FLOW CYTOMETRY
233
[19] Cell C y c l e A n a l y s i s b y F l o w C y t o m e t r y 1
By
J. W. GRAY
and E
COFFINO
Growth of cell populations can be analyzed by considering individual

cells to progress sequentially through a series of compartments or phases.
The processes of mitotic division and DNA synthesis can be recognized
by light microscopy and by incorporation of [aH]thymidine followed by
autoradiography, respectively. Application of these methods led to a description of the cell cycle composed of four phases: mitosis (M), gap 1 (G1)
preceding DNA synthesis, synthesis of DNA (S), and, finally, gap 2 (G2)
following DNA synthesis and preceding the next mitosis.
Experimental methods are often required to answer the following
questions about a population of tissue culture cells: What is the fraction of
ceils in each phase? At what rate are cells progressing from one phase to
the next? What is the locus of action in the cycle of an agent that perturbs
cell cycle progression?
Techniques that depend on light microscopy and autoradiography
have been used extensively to answer these questions. The method of
autoradiography (see this volume [22]) and the analysis of labeling index
and mitotic index experiments2 have been described in this series. Flow
cytometric (FCM) methods provide an alternative method of analysis.
Cells are stained with fluorescent dyes that bind to DNA. By exciting the
dye and measuring the emitted fluorescence of single cells as they flow in
single file at high rates, the DNA content of each cell can be measured and
its location in the cell cycle thereby inferred. The development of specific
and convenient staining techniques and the commercial availability of
sophisticated instruments have made this method of analysis increasingly
accessible to biologists.
First we shall review flow cytometric principles, and then we shall
describe selected methods for the preparation of cultured cells for flow
cytometry. Three examples of the use of flow cytometry in cell cycle
studies will be given.
i This report was prepared as an account of work sponsored by the United States Government. Neither the United States nor the United States Energy Research & Development
Administration, nor any of their employees, nor any of their contractors, subcontractors,
or their employees, makes any warranty, express or implied, or assumes any legal liability
or responsibility for the accuracy, completeness or usefulness of any information, apparatus, product or process disclosed, or represents that its use would not infringe privately owned rights.
2 j. B. Kurz and D. L. Friedman, this series, Vol. 40, p. 44.
ISBN 0-t2-181958-2
234
[19]
SPECIALIZED TECHNIQUES
TABLE I
COMMERCIALLY AVAILABLE FLOW CYTOMETERS AND FLOW SORTERS
Instrument
type
Company a
Becton-Dickinson FACS
Ortho Instruments
Coulter Electronics, Inc.
Flow
Flow
Flow
Flow
Flow
Address
sorter 4
cytometer
cytomete~ '7
sorter
sorter 6
506 Clyde Ave., Mountain View,

California 94040
410 University Ave., Westwood,
Massachusetts 02090
590 West 20th St., Hialeah, Florida 33010
a Reference to a company or product name does not imply approval or recommendation

of the product by the University of California or the U.S. Department of Energy to the
exclusion of others that may be suitable.
Flow Cytometry Principles

A variety of flow cytometric instruments are available commercially
(Table I). The capabilities of these instruments are varied; procedures that
can be accomplished readily with one instrument might be impossible
with another. Therefore, in selecting an instrument for a particular application, one should understand the principles involved in flow cytometry
and know which capabilities are common to all flow cytometers or sorters
Sample(A)
[~
Fluorescence
detector
Sheathwater ~ . 3._-f~ ~
Deflection
pJat~
/:~
~.~ ~'~
" :
~
"~::-..__ Microscopeobjective
""~G.
~ "-'--"....
..... "--~ ....
Sorted
"'-,DLaserbeam
(~~illection tubes
FIG. 1. Schematic representation of a flow sorter. Fluorescently stained cells in aqueous

suspension are introduced into the sorter chamber (A). Cells traverse the chamber and
emerge one at a time in a liquid jet, where they are illuminated by a laser beam (B). The
resulting fluorescence is collected by a microscope objective and projected through an
optical filter (to remove scattered laser light) onto a photomultiplier. The photomultiplier and
associated amplifier produce a voltage pulse whose height is proportional to the fluorescence
intensity. The height of the voltage pulse is the criterion for cell sorting. Drops containing
cells to be sorted are charged as they separate from the liquid jet. Empty drops or drops
containing unwanted cells are not charged. The charged droplets are separated from the
others as they fall through an electric field generated by charge deflection plates.
[19]
CELL CYCLE ANALYSIS BY F L O W CYTOMETRY
235
and which are machine-specific. Generally speaking, in a flow system

such as the sorter illustrated in Fig. 1, fluorescently stained cells are
transported in a liquid medium, one by one, through an intense light that
excites the dye. The resulting fluorescence that is measured and recorded
is proportional to the amount of the cellular component (in this discussion,
DNA) to which the dye is bound. The rate of cell analysis is usually about
103 cells/sec. In addition, cells can be sorted (separated) on the basis of
their fluorescence.
A brief discussion of flow cytometric concepts, including cell transport, fluorescence excitation and detection, cell sorting, and data aquisition, emphasizing features that are important in DNA-content measurements, follows. Flow cytometric principles have been reviewed in detail
by Van Dilla and Mendelsohn 3 and by Herzenberg et al. 4 (also see
Kamentsky et al. and GiShder).
Cell Transport
All flow cytometers operate on the principle that stained cells in aqueous suspension will flow with the suspending medium. By channeling this
flow properly, the cells can be forced through a region where their optical
properties can be measured. Usually the medium containing the cells is
surrounded by a sheath of similar fluid.
Figure 2 shows a liquid jet emerging from a nozzle into the air; this
occurs in most flow sorters. The black stream was produced by injecting
ink into the cell sorter instead of a cell suspension. In an instrument such
as the one pictured here, cells are measured immediately after the jet is
formed. In other instruments, cells are injected into glass capillary tubes
or water-filled chambers where they are measured. An intense light
source illuminates the cells in the sample stream. If the diameter of the
sample stream is not sufficiently small, some cells may be illuminated less
intensely than others and thus the fluorescence frequency distributions
will be distorted. The sample stream diameter can be decreased by increasing the pressure on the sheath relative to that on the sample. When
the cell concentration is low, the number of cells analyzed per second
should not be increased by increasing the sample stream diameter, because of the danger of improper illumination. Instead, the cell sample
should be concentrated.
3 M. A. Van Dilla and M. L. Mendelsohn, in "Flow Cytometry and Sorting." Wiley, New
York (in press).
4 L. A. Herzenberg, R. G. Sweet, and L. A. Herzenber~, Sci. Am. 234, 108-117 (1976).
5 L. A. Kamentsky, Adv. Biophys. Med. Phys. 14, 83 (1973).
A. Brunsting, in "Flow Cytometry and Sorting," pp. 79-88. Wiley, New York (in press).
7 This flow cytometer operates according to somewhat different principles than those described in this review. For more information, see W. G6hde, in "Fluorescence Techniques
in Cell Biology," pp. 79-88. Springer-Verlag, Berlin and New York, 1973.
236
[19]
FIG. 2. A liquid jet ejected into air by a cell sorter. The sorter nozzle was formed by a
drawn glass capillary with a 50-/.Lminside diameter. The black stream was formed by injecting ink in place of the cell suspension; the confinement of the ink to the center of the jet is
clearly visible. In practice, the sample ~tream should be even smaller than the ink stream
shown in the figure.
F l u o r e s c e n c e Excitation a n d Collection
This is a c o m p l e x subject to discuss in specific t e r m s b e c a u s e of the

different requirements o f the m a n y D N A stains currently in use. We will
describe only the requirements for the d y e s propidium iodide (PI) and
c h r o m o m y c i n A3 (CA3). Detailed staining protocols for these reagents
are given below.
T h e stain content in a cell is m e a s u r e d as the cell flows at constant
velocity through an intense light. T h e dye is stimulated b y the light, and
the resulting fluorescence is detected and is assumed to be proportional to
the a m o u n t of d y e in the cell; this is not always true for a s y m m e t r i c cells. 8
T h e wavelength o f the exciting light should be m a t c h e d as closely as
possible to the excitation m a x i m u m o f the dye (see Fig. 3 for corrected
excitation and emission spectra of the d y e s CA3 and PI b o u n d to D N A ) . a
T h e excitation m a x i m u m of CA3 is a b o u t 430 nm, while that o f PI is about
535 rim.
The light source of m o s t flow c y t o m e t e r s is an argon-ion laser; an
exception is the Ortho flow c y t o m e t e r 7 which uses an arc lamp. The laser
a B. L. Gledhill, S. Lake, L. L. Steinmetz et al., J. Cell. Phys. 87, 367-375 (1976).
0 These data were kindly supplied by Dr. R. Langlois, Lawrence Livermore Laboratory,
Livermore, California.
[19]
/,/
>~ 1.0
f'~
;,\
/
/,
i/11
4OO
\1
ChrornomycinA3-
50O
237
6OO
\
700
Wavelength, nrn
FIG. 3. Excitation and emission spectra ~ for the dyes propidium iodide (A) and
chromomycin A3 (B). Dashed lines represent excitation spectra and solid lines emission
spectra.
should be adjusted to emit light at 457 nm to excite CA3 and at 514 nm to

excite PI. A system with a smaller laser might operate only at 488 nm. PI
can be excited at this wavelength, but CA3 cannot.
The light collected from each cell as it traverses the exciting light beam
is due partly to cellular fluorescence and partly to scattered laser light. To
measure only fluorescence, the scattered light is removed by an optical
filter that transmits light at wavelengths longer than that illuminating the
cell. It is important to select a filter that transmits at longer wavelengths
than the laser light because the scattered light may be several orders of
magnitude more intense than the fluorescence. At the same time, a filter
should be chosen that transmits as much fluorescence as possible. After
the fluorescence from a cell passes through the optical filter, it strikes a
photomultiplier tube and produces an electrical pulse, the amplitude of
which is proportional to the fluorescence intensity. Each pulse is
amplified and its amplitude digitized and added to the memory of a multichannel analyzer. The analysis of a large number of DNA-stained cells
results in a frequency distribution of cellular DNA content. The DNA
distributions of $49 cells stained by the CA3 and PI procedures are shown
in Fig. 4. The flow cytometer was adjusted by changing the gain of the
pulse amplifier so that the mode (highest point) of each DNA distribution
was 50 on theX axis. This adjustment can be made in a variety of ways on
different instruments: by changing the pulse amplifier gain, the photomultiplier high voltage, or the laser intensity. If a choice is available, an
adjustment of the amplifier gain is preferable; every effort should be made
238
50O0
I
Chromomycin A3
[19]
I
Propidium Iodide
"6
j~ 2S00
=E
Z
....... J
,,.....~/
50
100
50
Relative fluorescence
],~ ..........
100
FIG. 4. Fluorescence distributions of $49 mouse I.ymphoma cells stained with (A)
chromomycin A3 and (B) propidium iodide. The data points are solid circles. The solid lines
are the result of least squares best fits to the data; each solid line is the sum of the dotted
component curves.
to avoid changing the laser power. The G1 peak mode should be adjusted
to the same location at the start of each experiment to standardize data.
Most flow cytometers can measure and record more than one parameter for each cell. For example, both the dye fluorescence and the scattered laser light might be recorded to form a two-parameter distribution,
the dye fluorescence proportional to DNA content and the scattered light
proportional to cell size. The exact relationship between scattered light
and cell size is difficult to quantitate; it depends on the light collection,
angle, and aperture, index of refraction of the suspending fluid, etc. Usually, however, the scattered light increases monotonically with cell volume.
Cell Sorting
The selection of cells on the basis of their DNA content can be coupled
to tracer technology. For example, the cell cycle location of a radioactively labeled cohort of cells can be determined by sorting cells according
to their varying DNA contents (and hence position in the cell cycle) and
assaying the fractions for radioactivity by autoradiography or liquid scintillation counting.
Commercially available cell sorters use the electrostatic deflection
principle illustrated in Fig. 1 to fractionate cells. The cells are ejected into
air and their stain content measured immediately. The cells continue
down the liquid jet until the jet breaks into droplets. When a cell to be
[19]
CELL CYCLE ANALYSIS BY F L O W CYTOMETRY
239
sorted reaches this point, a brief voltage pulse is applied to the jet, and the
droplet containing the cell is charged when it separates from the liquid
column. The distance from the measurement point to the droplet breakoff
must be constant during operation of the sorter. This distance determines
the time delay between cell measurement and the application of the charging pulse and is stabilized by vibrating the flow chamber at high frequency, e.g., 30 kHz. The distance can be disturbed by a change in the
pressure producing the liquid jet, by the occurrence of a bubble within the
flow nozzle, or by a piece of debris lodged in the nozzle. Should one of
these events occur, the wrong droplets will be charged. Since such malfunctions are not uncommon, the droplet breakoff point should be
checked frequently.
Methods
Cell Dispersal
All flow cytometric techniques require analysis of individual cells.
Therefore, prior to analysis, the population must be dispersed into a suspension of single cells.
Cells grown in suspension culture usually require no dispersal, but
those grown in monolayer cultures must be dissociated. To accomplish
this, the medium is decanted and the monolayer gently rinsed with 5 ml
cold PBS (solution 1). 1 The PBS is decanted, 2 ml of trypsin (solution 2)
are added, and the cells incubated for about 5 min at 37. The incubation
should be restricted to the minimum time sufficient to free the cells from
the flask. The cells will leave the monolayer in sheets. This process can be
hastened by striking the culture flask sharply with the heel of the hand.
Dispersal of clumped cells will be aided by gently passing the suspension
in and out of a pipette. After dispersal, 8 ml of culture medium containing
serum are added to inhibit trypsin and the mixture is transferred to a
plastic centrifuge tube. The cells from the monolayer culture now should
be a suspension of single cells. Subsequent steps are identical for cells
grown in monolayer or in suspension. The cells are centrifuged (250 g for 5
min in this and subsequent steps), the supernatant liquid aspirated, and
the cells resuspended vigorously in 10 ml cold PBS with a vortex mixer.
Centrifugation and aspiration are repeated and the cell pellet is vigorously
resuspended in the residual fluid and fixed by adding 5 ml of cold 70%
ethanol dropwise during continuous agitation with a Vortex mixer. After
fixation and cell suspension can be stored almost indefinitely at 4 .
~0 All working solutions are described in the Apperidix.
240
[19]
Staining
In our experience, the simplest and most reliable DNA stain is the dye
chromomycin A3 (Calbiochem). The staining procedure is adapted from
that of Crissman and Tobey who used the dye mithramycin, n Mithramycin and chromomycin A3 have similar properties, and chromomycin A3
was chosen because of its availability and high purity.
Cells should be fixed in 70% ethanol for at least 30 rain before staining.
A suspension containing approximately 5 106 fixed cells is centrifuged
and the supernatant liquid removed by aspiration. The cells are resuspended by a Vortex mixer in the CA3 staining solution (solution 3). After
30 rain of incubation in the dark at room temperature, the cells are ready
for flow cytometric analysis. This procedure is elegant in its simplicity;
unfortunately CA3 fluorescence cannot be excited by some flow cytometers (see discussion of flow cytometry above).
When CA3 cannot be used, propidium iodide (PI) can be substituted
with a slight modification of the staining procedure. TM Cells must be
treated initially with RNase because the dye binds to all nucleic acids. To
accomplish this, approximately 5 x 106 cells from the fixative are re~
moved by centrifugation and resuspended with a Vortex mixer in
RNase-buffer solution (solution 4) and incubated for 15 rain at 37. The
cells are centrifuged and the supernatant solution aspirated. Then the cells
are resuspended by vigorous agitation by a vortex mixer in the PI staining
solution (solution 5). The stained cells should remain in the dye solution
for 30 min at room temperature before analysis.
Sorting for Radioactivity Assay

Cells for autoradiographic analysis can be collected directly on microscope slides. At least 1000 cells should be sorted onto each slide. When
the cells are sorted, the slides are air dried. Prior to autoradiography, the
cells are fixed to the slides in a series of three successive 5-min ethanol
baths (95%, 95%, 100%) and then washed for 1 min in distilled water to
remove salt crystals (the sheath fluid in sorters is commonly isotonic
saline).
Alternatively, cells may be analyzed by liquid scintillation counting
for incorporated radioactivity. They are sorted directly onto glass-fiber
filters to minimize the cell loss that occurs if the sample is collected in one
vessel and transferred to filters. When the sorter is adjusted properly, the
collection efficiency is greater than 95%, so that an accurate measurement
of the radioactivity per sorted cell can be made.
n H. A. Crissman and R. A. Tobey, Science 184, 1297-1298 (1974).
1, W. Dittrich and W. Grhde, Z. Naturforsch., Teil B 24, 360-361 (1969).
[19]
241
Filters containing the sorted cells are washed in a Millipore filtration

apparatus with sequential 10-ml portions of 10% cold trichloroacetic acid
(TCA), 5% cold TCA, and 70% ethanol and then loaded into scintillation
vials. To each vial is added 125/zl water and 1 ml tissue solubilizer (NCS,
Amersham); after 1 hr, 10 ml of scintillation fluid (Liquifluor, New EngEland Nuclear) are added. The vials are then ready for counting.
Applications
Estimates of cell cycle parameters such as the fractions of cells in the
GI, S, and G2 + M phases, average durations and variabilities of the three
phases, and rates of DNA synthesis can be obtained from flow cytometric
data. However, proper estimation methods might require relatively
sophisticated computer techniques. We make no attempt to discuss data
analysis techniques in detail; instead, we refer the interested reader to a
review article by Gray et al. 18 for further information and references.
Analysis of Phase Fractions
The most common application of flow cytometric techniques related to
the cell cycle is the determination of the fraction of cells in the G1, S, and
G2 + M phases. This information is obtained from DNA distributions
such as those shown in Fig. 4, measured for asynchronous, exponentially
growing $49 mouse lymphoma cells from the same culture; one aliquot
was stained with CA3 and the other with PI. In each distribution, the peak
at 1 DNA content (relative fluorescence, 50) is produced by diploid, G1
phase cells. The peak at 2 DNA content (relative fluorescence, 100) is
produced by G2 + M phase cells, and the intermediate continuum is produced by S phase cells in which varying amounts of DNA have replicated.
The areas under each of these regions of D N A distribution are proportional to the fractions of cells in the corresponding cell cycle phase.
The G1, G2 + M, and S phase areas are determined by simultaneously
fitting the DNA distribution with the sum of two normal distributions and
a broadened second-order polynomial. 14The normal distributions approximate the GI and Gz + M phases, and the polynomial approximates the S
phase. The estimated fractions in each phase are shown in Table II.
Phase fraction analysis can be used in the assessment of cellular
growth conditions. For example, FCM measurements and cell counts
were made in parallel to assess the effect of dilution of suspension cultures
13 j. W. Gray, P. N. Dean, and M. L. Mendelsohn, in "Flow Cytometry and Sorting."
Wiley, New York (in press).
14 p. N. Dean and J. Jett, J. Cell Biol. 60, 523-527 (1974).
242
[19]
TABLE II
ESTIMATED FRACTIONS OF CELL CYCLE PHASES IN A CULTURE OF
ASYNCHRONOUS 849 MOUSE LYMPHOMA CELLS
Propidium iodide
Chromomycin
0.20
0.68
0.12
0.27
0.63
0.10
G1
S
GeM
of Chinese hamster ovary (CHO) cells on the growth rate and the distribution of cells in the cycle. Figure 5A presents the growth curves for two
exponentially growing cultures. One culture was maintained as a control;
the second was established by a 4-fold dilution of one-fourth of the control
culture at time zero. The control culture was started 48 hr, i.e., almost
four population-doubting times, prior to dilution to ensure the decay of
any synchrony that might have been induced initially. The growth curves
of the two cultures, based on frequent sampling, indicated that both grew
exponentially for at least 24 hr after the dilution step. At later times the
cultures entered stationary phase with saturation densities greater than
1 106 cells/ml.
Because of the apparent continued exponential growth after dilution,
the noticeable perturbation in the DNA distribution of the diluted culture
(Fig. 5B) was unexpected. The control culture contained a relatively constant proportion of cells in the G1 phase, both before and after the time
,oo1,,,
,_
_../B'
b0L
'
'
'
~8o
1
-40
-20
20
40
60
201
I
,
I
80
-10 0
20
Time after dilution (hours)
I
40
l
60
i
80
FIG. 5. The response of Chinese hamster ovary cells to dilution of the culture. At - 4 4 hr
a spinner flask was established with 1100 ml culture medium and allowed to reach asynchronous, exponential growth (O--O). At time zero, 275 ml of this control culture was
poured into a second spinner flask containing 825 ml of fresh 37 medium (D--D), giving a
1 : 4 dilution. Cell counts (A) and estimates of the fraction in the G1 phase (B) were made
periodically for both diluted and undiluted cultures.
[19]
243
zero point and up to about 26 hr. This uniformity indicated asynchrony

and was in agreement with the growth curve. In contrast, in the diluted
culture the fraction of cells in the G~ phase increased and reached a peak
at about 8 hr after dilution. The Ga fraction then quickly declined, returning to the control value at about 17 hr, more than one generation time after
the dilution. By 30 hr the fraction of cells in the G~ phase in the control
culture began to increase noticeably while the growth curve showed no
obvious change in slope. When growth in both cultures slowed and entered stationary phase, the fraction of cells in G~ phase increased dramatically, as expected, because many conditions of growth restriction normally produce G1 (or Go) phase arrest.
Perturbed Population Analysis

We can take advantage of the time-dependent changes in the DNA
distributions resulting from perturbations to study the effect of the perturbing influence on cell cycle progressions. The effect on other cellular
properties such as size can be assayed as well by measuring another
parameter for each cell simultaneously. Figure 6 illustrates the results of
measuring DNA content (CA3 fluorescence) and size (90 fight scatter) of
$49 mouse lymphoma cells at several time points after addition of dibutyryl cyclic AMP (dbcAMP) to an asynchronous, exponentially growing culture. The qualitative effect of the treatment was a G1 phase block.
Most of the cells have the DNA content of the G~ phase after incubation
for 25 hr; their size is about the same as that of early G1 cells in the
unperturbed population. Before entering the G1 block, cell growth appears unaffected; the cells cycle and increase in size at the normal rate.
These observations can be quantitated by computer modeling. On the
right-hand side of Figure 6 is a series of distributions of DNA content vs.
size predicted for the time-dependent response of $49 cells to dbcAMP.
The model was generated using the parameters listed in Table III (the
reader is referred to Gray TM for a detailed description of a similar oneparameter computer model). These parameters were adjusted so that the
computer simulations matched the data; the parameters obtained from
this process are assumed to be valid describers of the population.
The second parameter in such dual-parameter analyses is limited only
by one's imagination. For example, the recent development of a variety of
fluorescent enzyme substrates by Dolbeare and Smith TM points to the practicality of simultaneous assessment of DNA content and enzyme content.
is j. W. Gray, Cell Tissue Kinet. 9, 499-516 (1976).
is F. A. Dolbeare and R. E. Smith, in "Flow Cytometry and Sorting." Wiley, New York (in
press).
244
Experimental
[19]
Simulation
10 Hours
i Pl-'l/
FIG. 6. Light-scatter and D N A - c o n t e n t distributions o f S49 m o u s e l y m p h o m a cells responding to the administration of d b c A M P at zero time. Experimental distributions are
s h o w n on the left and distributions simulated according to the p a r a m e t e r s in Table III are
s h o w n on t h e right.
Combined D N A Distribution-Radioactive Tracer Analysis
Another sophisticated and powerful application of flow cytometry to

cell cycle studies involves the measurement of radioactive tracer uptake
as a function of cell cycle position. This has been accomplished in the past
by measuring tracer uptake in synchronized cell populations (see this
[19]
245
TABLE III
KINETIC PARAMETERS ASSOCIATED WITH THE RESPONSE OF $49
MOUSE LYMPHOMACELLS TO dbcAMP a
Average
phase duration
(hours)
Coefficient
of variation
2.1
12.1
2.5
0.25
0.25
0.25
G1
S
G= + M
a The intensity of 90 light scatter was assumed to increase linearly with cell volume; the
light scatter from mitotic cells is twice that from G1 cells. The effect of the dbc AMP
in the model was assumed to be an immediate and complete block in early G1 phase.
The rate of DNA synthesis as a function of DNA content was described by a normal
distribution with a coefficient of variation of 0.4 and the mean centered in mid-S phase.
(This choice in the model reflects the data in Fig. 7.)
1500
~o
o
100
2O0
Relative
DNA
content
FIG. 7. Incorporation of [aH]TdR into cellular D N A as a function of D N A content. The

[3H]TdR was administered to asynchronous, exponientially growing CHO cells as a 30-rain
pulse. The [2H]TdR uptake was assayed by liquid scintillation counting. The bars span
replicate measurements on cells from the same culture. The solid lines connect the means of
the measurements at each DNA-content region.
246
[19]
volume [20]). Such studies were limited by the synchrony techniques and
could not be applied in vivo. The use of flow sorting to select prelabeled
cells according to size and/or DNA content eliminates the need for
synchronization.
An example of such tracer techniques is illustrated by the technique
used in the estimation of the rate of DNA synthesis as a function of DNA
content for CHO cells. 1~ Prior to sorting, cells were pulse labeled for 30
min with 0.5/xCi/ml tritiated thymidine ([ZH]TdR) (specific activity - 2 0
Ci/mM). The cells were then fixed, stained with CA3, and processed
through a cell sorter where 104 cells were sorted from each of ten DNAcontent regions into which the S phase was divided. The sorted cells from
each region were collected on glass filters and the radioactivity measured
by liquid scintillation counting. Figure 7 shows the results of these measurements. The rate of [ZH]TdR uptake is maximal in mid-S phase, decreasing to minimal values during early and late S phase.
Application of the methods described here and related techniques provides a rapid and convenient means for answering questions related to cell
cycle kinetics that would be otherwise difficult or impossible to study. The
rapidity of the fixation and staining methods makes it possible to monitor
results while experiments are in progress. The recent development of
fluorescent DNA-staining techniques that do not affect cell viabilityTM contributes further to the utility of these methods. Somewhat balancing these
advantages, however, are the instrumental complexity and the considerable hardware cost of this technology, which may limit its accessibility for
the occasional user.
Appendix: Working Solutions for Cell Dispersal and Staining

Solution 1---PBS
Component
NaC1
KC1
Na~PO4
KHsPO4
H20
Amount
800 mg
200 mg
1150 mg
200 mg
1000 ml
17 j. W. Gray, Y. S. George, K. Brown, P. Dean, in preparation.

is D. J. Arndt-Jovin and T. M. Jovin, J. Histochem. Cytochem. 25, 585-589 (1977).
[19]
247
Solution 2--Trypsin
Component
NaEDTA
Trypsin (Difco
Bacto-trypsin
#0153-60)
PBS
Amount
4 ml (stock solution:
1 g NaEDTA/100 ml H20)
1 vial reconstituted
in 10 ml PBS
186 ml
The final solution should be sterilized by filtration through 0.2 /zm

filters.
Solution 3--CA3 Stain

Component
Chromomycin A3
MgC12.4 H20
H20 (4)
Amount
10 mg
1.5 g
500 ml
The chromomycin A3 will not dissolve in warm water.
Solution 4---RNase
Component
RNase (Sigma;
71 Kunitz
units/rag)
Na2HPO4.7 H20
Na~HPO4
H20
Amount
10 mg
55.6 mg
168.9 mg
10 ml
Solution 5---PI
Component
Propidium iodide
H20
Amount
10 mg
1000 ml
248
[20]
Acknowledgments
The authors gratefully acknowledge helpful technical comments from Y. S. George. This
work was performed under the auspices of the Department of Energy, Contract #W-7405ENG-48 with support from USPHS grant 5R0114533, NIH grant GM16496, and NSF grant
PCM 75-06764. P. C. is the recipient of a Research Career Development Award from the
NIH, Institute of General Medical Sciences.
[20] C e l l S y n c h r o n i z a t i o n
By TSUKASA ASHIHARA and RENATO BASERGA

The growth of cells in cultures can be divided into three stages: (1) a
lag period, immediately after the inoculum; (2) a growth phase, during
which the cell number increases rapidly and usually in an exponential
fashion; and (3) a plateau or stationary phase, during which the cell number remains constant.
In exponentially growing populations the cells are distributed asynchronously throughout the cell cycle in its four phases--G1, S, G2, and M
(mitosis). Under conditions restrictive for growth, some cell lines of rodent (3T3) or human (WI-38) origin will accumulate in aGo--G1stage of the
cell cycle, i.e., the cells have a 2n content of DNA (G1 content), do not
traverse the cell cycle, and, when stimulated to resume growth, will enter
first S and subsequently mitosis.X However, most cell lines, especially
virally transformed cell lines, are asynchronously distributed throughout
the cell cycle, even when growth is markedly slowed down by nutritional
deficiencies, z Since metabolic processes are different in different phases of
the cell cycle, synchronized populations of cells are not only useful but
almost necessary if one wishes to study cell-cycle related events. As an
example we can cite the case of histone mRNA, which is absent from the
cytoplasm of G1 cells, is abundant in the cytoplasm of S phase cells, and
rapidly disappears again from the cytoplasm when the cell proceeds from
S to M.a Clearly, when one wishes to study the regulation of histone
mRNA synthesis and processing one must have synchronized populations
of cells.
Theoretically, at least, cells can be synchronized in any one phase of
the cell cycle, including Go. However, the most common sites of arrest are
G~, Go, and M, while G2 arrest is difficult to achieve and S phase arrest is
F. Wiel~el and R. Baserga, J. Cell. Physiol. 74, 191 (1969).
2 j. C. Bartholomew, H. Yokota, and P. Ross, J. Cell. Physiol. 88, 277 (1976).
a M. Melli, G. Spinelli, and E. Arnold, Cell 12, 167 (1977).

ISBN 0-12-181958-2
[20]
CELL SYNCHRONIZATION
249
often lethal. Ideally synchronized populations of cells in culture should

meet these criteria: (1) they should be perfectly synchronized at a specific
point of the cell cycle; (2) the procedure used for synchronizing cells
should have little or no effect on the metabolic processes of the cell; and
(3), especially for the biochemist, the method should allow the harvesting
of synchronized cells in large quantities. One can quickly say that no
synchronizing procedure fully achieves these three goals, but there are a
few methods to be related below, that come close to the proposed goals.
We shall discuss the degree of synchrony that can be obtained with
various synchronizing procedures, the effect that these procedures have
on the metabolic processes of cells in culture, and the amount of cells that
can be obtained by various synchronizing procedures. As a general rule,
we act on the proposition that 108 cells are sufficient for most biochemical
determinations. In fact, a great number of biochemical determinations can
be carried out accurately with 107 cells, but 106 cells constitute the lower
limit for biochemical determinations even if methods are scaled down
considerably. This number of cells, i.e., a million, roughly corresponds to
7-15/xg of DNA, depending on the ploidy of the cells under consideration.
However, simpler determinations, as for instance incorporation of
[3H]uridine into total RNA or of [3H]thymidine into DNA, can be carried
out with as little as 104 or 105 cells.
Evaluation of the Degree of Synchrony
There are several methods for determining the degree of synchrony of
populations of cells in culture.
1. Autoradiography with [3H]thymidine

Unfortunately, light microscopy combined with [3H]thymidine autoradiography is still the best method for an analysis of the cell cycle and
an evaluation of the degree of synchrony. We say unfortunately because
autoradiography is a tedious and time-consuming procedure and also because most biochemists have an ancestral fear of the microscope. Yet
autoradiography gives more information than any other method that is
currently used for an analysis of the cell cycle. By looking at a coverslip
of cells labeled with [3H]thymidine one can obtain information not only
about the number of cells in DNA synthesis, but also about the number of
cells in mitosis. At the same time, one can ascertain whether the population is homogenous or heterogeneous, whether there are dying or dead
cells, and whether the cultures are contaminated or not with mycoplasma.
The technique for autoradiography is not complex. It has been de-
250
[20]
100
x
X/x
/
35o
_?
"E
X/"
,t
o
ib
~o
Jo
Time offer stimulotioe (in hours)
FIG. 1. Cumulative percentage of cells in DNA synthesis, after quiescent K12 cells were
stimulated to proliferate by serum. K12 cells, a ts mutant that arrests in G1 at the nonpermissive temperature, were continuously exposed to [aH]thymidine. The ordinate gives the
cumulative percentage of cells labeled by [aH]thymidine, and the abscissa gives the time
after stimulation, at the permissive (X---X) or at the nonpermissive (O--O) temperatures.
scribed in detail in a monograph by Baserga and Malamud 4 and is presented in this volume [22]. Briefly, to determine the degree of synchrony
the cells can be continuously exposed to [3H]thymidine or, alternatively,
pulse labeled with [3H]thymidine for a brief period of time, 30 rain or so.
In the former case the concentration of [3H]thymidine (6.7 Ci/mM) in the
medium should not exceed 0.1 /~Ci/ml. We use, satisfactorily, a concentration of 0.05 t~Ci/ml. In the latter case a concentration of 0.5/~Ci/ml is
permissible. The cells are fixed at the desired intervals after the synchronization procedure, usually in Carnoy, and then autoradiographed
with Eastman Kodak NTB nuclear emulsion (at the concentrations mentioned above, an exposure time of 7-10 days is usually sufficient), developed, fixed stained with hematoxylin and eosin, and scored at the light
microscope.
A typical result with this procedure is illustrated in Fig. 1, which
shows the cumulative labeling index of K12 cells synchronized by serum
deprivation. Serum-deprived cells (48 hr in 0.5% serum in coverslip cultures) were stimulated by addition of fresh medium plus 10% serum and
immediately exposed to [3H]thymidine, 0.1 /zCi/ml. Duplicate coverslips
were then fixed at the intervals indicated on the abscissa, autoradiographed, and analyzed. As one can see from Fig. 1, the entry of serumstimulated K12 cells into S is not sharply synchronized. At the time of
serum stimulation most of the cells are in G0--G1, and about 20 hr after
stimulation, 80% or more are in S phase. However, they are not synchronized at the same point. In other words, if one wishes to obtain cells
synchronized in the samephase of the cell cycle (G1 orS), then the method
4 R. Baserga and D. Malamud, "Autoradiography. Techniques and Application." Harper
(Hoeber), New York, 1969.
[20]
251
described in Fig. 1 has achieved a reasonable degree of synchronization.

But if one wishes to obtain cells synchronized at aprecise point o f the cell
cycle, then this procedure is not sufficient. This is a problem that affects
most of the synchronizing procedures used in most laboratories, with but
a few exceptions that will be described below. The reader is cautioned to
be exceedingly careful in interpreting the synchronization curves that are
actually published. In many instances the authors, consciously or subconsciously, increase the appearance o f synchronization by simply shortening the abscissa and lengthening the ordinate. This, of course, results in
very steep curves of entry into S. The reader always should check carefully the numbers on the abscissa, which usually refer to the time, and he
will find that some of the steepest curves published still have slopes of
about 8-10 hr between the first and last cells entering S.
For this reason, we insist that autoradiography data be plotted on
arithmetic scales as illustrated in Fig. 1, although it has b e c o m e fashionable lately to plot the exit o f cells from G , on a semilogarithmic scale. 5,6 As
e v e r y biochemist knows, 7 logarithmic or semilogarithmic plots tend to
hide the scattering of data and complex kinetics and, especially in our
case, can give a false impression of the degree of synchronization.
2. Incorporation o f [3H]thymidine into D N A
A quick method for evaluating the degree of synchrony, much quicker

than autoradiography, consists in determining by liquid scintillation
counting the amount of [3H]thymidine (or [14C]thymidine) incorporated
into acid-insoluble material, s Although quicker, this method is not as informative as autoradiography. The procedure is the same except that
instead of fixing the cells and processing them for autoradiography the
cells are harvested after extensive washing with a balanced salt solution.
The D N A from the harvested cells is extracted by the method o f Scott et
al., 9 and the radioactivity is assayed by liquid scintillation counting using
a Triton X-100 Toluene liquifluor. 1 Alternatively, the cells are simply
extracted with perchloric acid to remove the acid-soluble fraction, and
then they are dissolved in a liquid scintillation vial and counted. We find
this p r o c e d u r e useful as a screen for determining whether a method is
worthwhile pursuing or is simply hopeless. H o w e v e r , after screening a
5 R. F. Brooks, Cell 12, 311 (1977).
6 L. J. De Asua, M. K. O'Farrell, D. Clingan, and P. S. Rudland, Proc. Natl. Acad. Sci.
U.S.A. 74, 3845 (1977).
N. R. Cozzarelli, Annu. Rev. Biochem. 46, 641 (1977).
s E. Stubblefield, R. Klevecz, and L. Deaven, J. Cell. Physiol. 69, 345 (1967).
a j. E Scott, A. P. Fraccastoro, and E. B. Taft, J. Histochem. Cytochem. 4, 1 (1956).
~0M. S. Patterson and R. C. Greene, Anal. Chem. 37, 854 (1965).
252
[20]
given p r o c e d u r e b y this method we always prefer to determine the extent

of s y n c h r o n y in a m o r e accurate w a y by autoradiography. The advantage
is that, while the autoradiographs are being e x p o s e d and prepared, one
can p r o c e e d with other experiments.
3. F l o w Microfluorimetry
This is a m u c h faster method in which the distribution of cells o v e r the

cell cycle can be obtained with as little as 5 x 104 cells. T h e method is
essentially b a s e d on the determination of the a m o u n t of D N A per cell 11 b y
flow microfluorimetry.12 It divides a population of cells into cells with a G 1
content of D N A , cells with a G2 or M content of D N A , and cells in
b e t w e e n , with variable amounts of D N A depending on the position they
have in the S phase. By c o m p u t e r processing o f data, the percentage of
cells in e a c h of these three phases can be calculated with a high degree of
accuracy. The disadvantage is that it gives s o m e w h a t less information
than a u t o r a d i o g r a p h y and requires expensive instrumentation. H o w e v e r ,
the instrument also can be used for m o r e sophisticated endeavors, e.g.,
the simultaneous determination o f the a m o u n t of R N A and D N A per cell,
as described b y D a r z y n k i e w i c z et al. 13 (see also [19]).
Methods for Synchronizing Cells in Culture

1. P h y s i c a l M e t h o d s
A. SYNCHRONIZATION BY MITOTIC DETACHMENT

The p r o c e d u r e described is a slight modification of that of Terasima
and Tolmach. 14
Procedure. As an illustration, we present the p r o c e d u r e used for mitotic synchronization o f K12 cells, a ts m u t a n t o f the Chinese h a m s t e r line
WG, originally isolated b y R o s c o e et al. 15 and characterized b y Smith and
Wigglesworth. 16
K12 cells are grown in D u l b e c c o ' s Modified Eagle's Medium containing 10% (v/v) calf serum, at the permissive t e m p e r a t u r e of 34 . Exponen11T. T. Trujillo and M. A. Van DiUa, Acta Cytol. 16, 26 (1972).
12D. M. Holm and L. S. Cram, Exp. Cell Res. 80, 105 (1973).
13 Z. Darzynkiewicz, F. Traganos, T. Sharpless, and M. R. Melamed, Proc. Natl. Acad. Sci.
U.S.A. 73, 2881 (1976).
14T. Terasima and L. J. Tolmach, Exp. Cell Res. 30, 344 (1963).
15D. H. Roscoe, H. Robinson, and A. W. Carbonell, J. Cell. Physiol. 82, 333 (1973).
16B. J. Smith and N. M. Wigglesworth, J. Cell. Physiol. 84, 127 (1974).
[20]
253
10o
50-
FIG. 2. Cumulative percentage ofceUs in DNA synthesis in K12 cells collected by mitotic
detachment and replated at either the permissive (--) or the nonpermissive (X--X) temperatures. Cells were exposed to [SH]thymidine from the time of replating.
tially growing populations are detached from the plate with 0.25% trypsin
in Hanks's solution for 2 min. They are plated in 100-mm dishes (surface
of 78.5 cm2), at a concentration of 1.5 10~ cells per dish. By microscopic observation of the dishes with an inverted microscope, the time at
which mitotic figures are most frequent can be determined. For K12 cells,
this time interval is about 18 hr after plating. Mitotic cells are collected by
shaking the plates gently. The yield is about 5-8% of the total number of
cells, and the percentage of mitotic cells in the collected fraction is 90% or
more. The mitotic cells are then plated in 35-mm dishes, at a density of
2 x 104 cells/cm 2. The degree of synchronization can be established by
continuous labeling with [3H]thymidine at a concentration of 0.1 /zCi/ml,
added at the time of plating, and by counting the number of labeled cells at
various intervals after plating (see above). The results of such an experiment are shown in Fig. 2.
Degree of Synchrony. The degree of synchrony (Fig. 2) is satisfactory
and such cell populations can be used for a study of G1 events and events
in the S phase. The labeling index at the beginning is less than 5%. At the
permissive temperature, 80% of the cells at 34 go from G1 into S in a
period of only 2 hr. It is possible to follow such cell populations and to
determine how this degree of synchrony is maintained by observing the
wave of mitoses that follows the period of DNA synthesis. With separate
cultures it also can be shown that the number of cells doubles in a rather
brief interval, corresponding to the wave of mitoses. A note of caution is
in order: When mitotic cells are plated, the number of cells plated is
calculated from the number of mitotic cells that one actually seeds in
different dishes. Within 15 min after plating most of the mitotic cells have
completed mitosis, so that the number is already doubled. This should be
kept in mind if one wishes to determine the increase in cell number at a
later period.
254
[20]
lOft
8
|
Hours after mitotic detachment
FIG. 3. Cumulative percentage of cells in DNA synthesis in ts AF8 cells collected by

mitotic detachment and replated at either the permissive (O--C)) or the nonpermissive
(X--X) temperatures. The cells are synchronized in aphase of the cell cycle (at the nonpermissive temperature they are this side of the ts block) but not at a precise point.
In general, the degree of synchrony decreases as cells leave the S

phase. This has been discussed in detail in a review by Nias and Fox, lr to
which the reader is referred.
The degree of synchrony varies from one cell line to another, as shown
in Fig. 3, which refers to a mutant of BHK cells called AF8 cells, originally described by Burstin et al. 18 These cells were treated in essentially
the same manner as K 12 cells, except that the peak of mitoses occurred at
30 hr after plating instead of 18 hr. The collected mitotic cells were plated
and their entry into S determined as before. Notice the gradual slope of
the curve of entry into S. K12 cells have a shorter cell cycle time than AF8
cells, and this may explain the differences in the degree of synchronization. However, the reasons why some cell lines synchronize quite well
after mitosis (see, e.g., HeLa cells in the original paper by Terasima and
Tolmach 14arid in the review by Nias and Foxlr), while others synchronize
poorly even by mitotic selection, are not really understood.
Variations. The first important variation to keep in mind is that the time
of harvesting varies among cell lines. With the procedure described
above, since we use a very high density, one can collect a reasonable
number of cells, i.e., as much as 105 cells/100-mm dish. However, the
peak of mitoses after plating varies among cell lines, and there is no
simple rule that allows the investigator to select the ideal timing without
prior experimentation. If one wishes a rule, one could say that the peak of
mitoses generally occurs in most cell lines between 18 and 30 hr after
plating, but it is.imperative that the investigator should carefully monitor
1~ A. H. W. Nias and M. Fox, Cell Tissue Kinet. 4, 375 (197!).
is S. J. Burstin, H. K. Meiss, and C. Basilico, J. Cell. Physiol. 84, 397 (1974).
[20]
255
the plated dishes for the appearance of mitoses every time that he deals
with a different cell line.
The yield of mitotic cells can be increased in a variety of ways. Increase in the yield is often desirable because, as mentioned above, a dish
will yield only about l0 n cells. This is sufficient for experiments in cell
biology, but biochemical studies demand larger numbers of cells; one can
easily calculate that 100 100-mm dishes would be necessary to obtain 107
mitotic cells. To increase the yield of mitotic cells several techniques are
available. The most popular one is the use of Colcemid. This technique
was first devised by Stubblefield and Klevecz TM who found that Chinese
hamster cells could be treated for 2 hr with Colcemid, 0.06/~g/ml; the
effect was to arrest cells in metaphase in a reversible manner. These investigators also used a brief period of trypsinization, 45 sec at 4, resulting in
an increased yield of about 8-12% of the total population. In general, the
use of Colcemid for brief periods (2-3 hr) has no remarkable effect on the
biochemical events in mitotic cells or in synchronized G1 and S phase
cells. However, we occasionally have observed that cells treated with
Colcemid had different kinetics of entry into S phase from cells that had
not been so treated. Again, a simple rule cannot be given, and the investigator will have to determine for each cell line whether the increased
yield of mitotic cells that can be obtained with Colcemid is not counterbalanced by possible toxic effects to the mitotic cells. Nias and Fox, 17 for
instance, found that Colcemid treatment beyond 2 hr leads to an increased
number of aberrant mitoses; we have similar experience with several cell
lines.
Another way of increasing the yield of mitotic cells is tO use repeated
harvesting, a procedure first introduced by Petersen e t a l . 2 With this
procedure the monolayers were shaken at 10-min intervals, and each harvest was rapidly cooled to 4. Again, previous experiments had shown
that cells stored at 0 for a period of 4 hr are still capable of completing
mitosis in a manner identical to that of control cultures when plated in
warm medium. Repeated shaking at 10- or 30-min intervals can therefore
yield a larger number of mitotic cells. These may be stored at 4 without
loss of viability for a period of up to 8 hr, and the collected mitotic cells
can then be plated. Undoubtedly a combination of Colcemid and repeated
harvesting leads to an increased yield of mitotic cells. Repeated shaking,
however, may decrease the purity of the preparation, i.e., the percentage
of mitotic cells in the mitotically detached population. Provided one moni~a E. Stubblefield and R. Klevecz, Exp. Cell Res. 40, 660 (1965).
20 D. F. Petersen, R. A. Tobey, and E. C. Anderson, Fed. Proc., Fed. Am. Soc. Exp. Biol.
28, 1771 (1969).
256
[20]
tors the degree of synchrony, these techniques can be used effectively to

increase the yield of mitotic cells.
In our experience, as well as in the experience of other investigators,
Colcemid is less toxic than colchicine. Other mitotic inhibitors are not
suitable. For instance, vinblastine is quite effective in arresting cells in
mitotis but the arrest, unfortunately, is irreversible. 21
Another method of collecting cells in mitosis 22 consists of gentle shaking in medium free of calcium. It has been known that calcium-free
medium decreases the cell substrate binding and therefore favors the detachment of cells. The problem with the low calcium method is that nonmitotic cells also can be easily detached.
Advantages and Disadvantages. The advantage of mitotic detachment
is that this is probably the technique that gives the highest degree of
synchrony without the use of toxic drugs. The synchrony is not only
phase-specific, but is almost point-specific since the duration of
metaphase-anaphase is usually of the order of 30 min. If the degree of
synchrony is not satisfactory, as in the case of the AF8 cells shown in Fig.
3, one should remember that this is inherent to certain cells in culture.
Terasima and Tolmach 14 had originally observed that the asynchrony that
develops during the course of a single cell cycle arises mainly during the
G1 period. Thus with HeLa cells the fraction of cells in DNA synthesis
began to increase 7 hr after mitosis and continued to increase until 14 hr
after replating. The curve of entry into S was paralleled by an increase in
cell number commencing some l0 hr later. Synchronization in S can be
improved by adding hydroxyurea during the G~ period (see below). The
technique is simple and if care is taken to use warmed media, the effect on
cells is minimal.
Among the disadvantages of the mitotic synchronization method, the
only two that we can think of are the low yield of mitotic cells and the fact
that quite obviously this technique is applicable only to cells that can be
grown in monolayers. The low yield can be remedied by using the proper
number of culture dishes. For cells that only grow in suspension other
techniques are necessary for their synchronization.
B. OTHER PHYSICAL METHODS
Other physical methods for thesynchronization of cells in culture have
not been very successful. In general they are based on gradient techniques
that hopefully separate, on the basis of volume selection, cells in different
phases of the cell cycle. Terasima and Tolmach 14have indeed shown that,
during the mitotic cycle, cells double in volume so that it should be possi2~N. Bruchovsky,A. A. Owen, A. J. Becket, and J. E. Till, Cancer Res. 25, 1232(1965).
22E. Robbins and P. C. Marcus, Science 144, 1152 (1964).
[20]
257
ble to select cells from different points of the cycle on the basis of volume,
and thereby produce a synchronized population. In practice, pure populations of G 1, S, or G2 cells have not been obtained, although some of the
physical methods used have produced enriched populations. Linear sucrose gradients, Ficoll gradients, and gradients in fetal calf serum in
phosphate-buffered saline have been used. The several methods have
been discussed at length in the review by Nias and Fox 17 who concluded
that with all the physical methods " a high degree (more than 90%) of
synchronization has been demonstrated only in cells harvested from
monolayer cultures by the mitotic selection technique." Two recently
introduced methods, one by Everson et al. 23 and the centrifugal elutriation
method of Mitchell and Tupper, 24 are described in more detail.
C. FICOLL GRADIENT ACCORDING TO EVERSON et al. 23
This has been used to separate human lymphoblastoid cells growing in
culture. Since these cells are of interest because they are the only human
diploid cells that grow indefinitely, the technique for separation on a
FicoU gradient is given below, as described by Everson et al. 23
A 5-20% (wt/w) Ficoll linear continuous gradient is generated in a density gradient generator, resulting in a total volume of 80 cm 3 contained in
cylindrical polycarbonate tubes 3 cm in diameter and 10.5 cm in length.
On top of this gradient 10 cm 3 of 5% (w/w) Ficoll are layered to modify the
initial slope of the gradient. A suspension of cells in Hanks's balanced salt
solution is then carefully layered on top of this 5% Ficoll buffer zone. The
cell load can reach approximately 5 107 cells, suspended in a final concentration of 1 107/cm3.
Centrifugation is carded out at 4 in a PR.2 centrifuge using a
swing-out rotor at 80 g for 20-25 rain. Fractions are collected by placing a
stainless-steel tube through the center of the gradient to the bottom and
sampling via a polyethylene tubing with a polystatic pump.
The Ficoll is prepared to a starting concentration of 40% (w/w) with
water. It is then subsequently adjusted to a final concentration of 20%,
with an equal volume of double-strength Hanks's balanced salt solution.
This 20% Ficoll solution is then diluted 1 : 4 with isosmolar Hanks's balanced salt solution to obtain a 5% (w/w) FicoU solution. The two concentrations are used to generate the necessary gradient.
D e g r e e o f S y n c h r o n y . The degree of synchrony is modest, as can be
seen from the original figures in the paper by Everson et al. 2~ However,
reasonably good populations of G 1 and S can be obtained.
z3L. K. Everson, D. N. Buell, and G. N. Rogentine, Jr., J. Exp. Med. 137, 343 (1973).
Z4 B. F. Mitchell and J. T. Tupper, Exp. Cell Res. 106, 351 (1977).
258
[20]
Advantages and Disadvantages. The disadvantage, as mentioned

above, is the modest degree of synchrony; the major advantage is, as with
the isoleucine method (see below), the possibility of obtaining a large bulk
of cells for biochemical studies.
D. CENTRIFUGAL ELUTRIATION
This method uses the Beckman JE6 elutriator rotor and the elutriation
system described in the literature of the manufacturer, Spinco publication
DS-125B. With centrifugal elutriation, the sedimentation rate of cells reflects their size and consequently their position in the cell cycle since,
generally speaking, cells in G ~are smaller than cells in other phases of the
cycle. With this method, Mitchell and Tupper 24 have isolated different
fractions of 3T3 and SV40-3T3 cells. The results are not spectacular because of considerable overlapping, but some synchronization is obtained.
In general, the centrifugal elutriation method enriches certain fractions in
G~ cells, but there is considerable overlapping and the other phases are
totally indistinguishable.
2. Chemical Methods
A. SYNCHRONIZATION BY ISOLEUCINE DEPRIVATION
Procedure. The procedure described below is the original one by Ley

and Tobey. z5 Chinese hamster cells (CHO) are cultured in suspension as a
monodispersed population in spinner flasks. The cells are usually propagated in F10 medium, supplemented with 10% calf serum and antibiotics.
According to Ley and Tobey 25 the original F10 formula has been modified
in their laboratory by omitting stock solutions D and E (which contain
iron, copper, zinc, and calcium) and also sodium pyruvate. The growing
cells are synchronized by transfer to F10 medium prepared without
isoleucine and glutamine; 10% calf and 5% fetal calf sera are included
after they are dialyzed against 10 volumes of Earl's balanced salt solution
for 6 days at 3 with changes of salt solution every other day. In this
isoleucine-deficient F10 medium, CHO cells rapidly reach a stationary
phase; the stationary phase consists of cells arrested in the G0--G1 phase of
the cell cycle. Arrested cells remain viable in the G0--G~ phase for 60 or
more. Cell cycle traverse is promptly resumed when isoleucine and
glutamine are added to the deficient F10 medium. The concentrations of
glutamine and isoleucine that are added to the arrested cultures are 40/~M
and 4 /zM, respectively. Upon addition of these amino acids, the cells
divide in a fairly synchronous fashion as early as 30 hr later.
25 K. D. Ley and R. A. Tobcy, J. Cell Biol. 47, 453 (1970).
[20]
259
Degree of Synchrony. The degree of synchrony is fair. The cells resume

cell cycle traverse and enter DNA synthesis with a slope that varies from
1-12 hr after resuspension in medium containing isoleucine and
glutamine, i.e., the first cells start entering DNA synthesis I or 2 hr after
resuspension and the last cells enter DNA synthesis 12 hr later. This
degree of synchrony is acceptable if one wishes to study a large number of
cells in G 1or in the S phase with a small percentage of out-of-phase cells.
Advantages and Disadvantages. The obvious advantage of this procedure is that it allows the synchronization of cells in suspension and therefore makes available large quantities of cells at a reasonable cost. Since
cells in the stationary phase can reach a concentration, at least for CHO,
of 1.5 10~ cells/ml it can easily be seen that 500 ml of such a suspension
will yield as many as 108 cells. Among the disadvantages is that the
technique is not applicable to all cell lines, not even in suspension culture.
A number of investigators have found that the level of synchrony for a
variety of cell lines is not satisfactory when the isoleucine deprivation
method is used.
B. SYNCHRONIZATION BY SERUM DEPRIVATION AND HYDROXYUREA
Procedure. The detailed technique for synchronization by serum restriction and hydroxyurea is as follows. Cells are plated at a density of
5 105 cells per 100-mm dish for 3 days. The medium is removed, the
plates are washed thoroughly, and the cells are exposed to their usual
medium containing 1% (or 0.5%) serum for 48 hr. After 48 hr, the serumdeficient medium is replaced by a medium containing 10% donor calf
serum. Six hours later, concentrated hydroxyurea stock solution is added
to each dish to reach a final concentration of 1.5 mM hydroxyurea. After
the cells are blocked with hydroxyurea for 14 hr the medium is removed,
the plates are washed, and new medium without hydroxyurea is added.
During the entire process, prewarmed washing solutions and medium are
always used. These precautions are taken to avoid the effect of
temperature-change shock on the synchronization of cells.
Degree of Synchrony. With BHK cells and derivatives of BHK cells,
the above-mentioned procedure gives a good synchrony and little toxicity
(Fig. 4). ~6 Upon removal of hydroxyurea, the cells quickly enter S phase
in a synchronized fashion. The procedure is therefore satisfactory if cells
synchronized in S are desired. The amount of DNA per dish increases by
about 80%, which is about as much as one can get with hydroxyurea using
concentrations that will synchronize the cells at the G1/S boundary.
Lower concentrations will let cells go through the S phase, although
slowly; higher concentrations are more toxic.
26 H. Chang and R. Baserga, J. Cell. Physiol. 92, 333 (1977).
260
[20]
12-
100
IG
lo
90
!o
I-3
FIG. 4. Synchronization of ALl06 by a combination of serum restriction and hydroxyurea (HU) treatment. ALl06 cells were plated on 100-mmpetri dishes for 3 days. Cells
were then cultured in medium containing 1% serum for 2 days. Fresh medium containing
10% serum was then given and HU (1.5 mM) was added 6 hr later. HU was removed at zero
time on the abscissa. Cells were pulsed with [aH]dT for 3 hr at the times indicated for
determination of specific activity. Cells were continuously labeled for autoradiography. Bar,
specific activity of DNA; OpO, /.,g DNA/dish; X--X, % labeled cells. (Reprinted, with
permission, from Chang and Baserga.~e)
Variations. The time periods given do not apply to all cells, as one can
easily understand. Some cells may require more than 48 hr of serum
restriction to achieve quiescence. Other cells may have a longer prereplicative phase and, therefore, h y d r o x y u r e a must be added at a later time.
Generally speaking, one should try out the optimal conditions for each cell
line. Our suggestion is that one should not use more than 14 hr of exposure
to h y d r o x y u r e a ; otherwise toxicity can be very high. Similarly, it should
be emphasized that the final h y d r o x y u r e a concentration of 1.5 m M can
apply to certain cell lines and not to others. For instance, AF8 do better at
1.25 m M whereas a hybrid between AF8 and L N S V , called A L l 0 6 , does
better at 1.5 mM. Again, the investigator is advised to test the optimal
concentration of h y d r o x y u r e a which will induce a G ~/S b o u n d a r y block
with a minimal amount of toxicity.
Mironescu and Ellem ~r have suggested that pretreatment o f serumrestricted cells with either h y d r o x y u r e a or cytosine arabinoside, before
they are stimulated to proliferate, increases the degree of synchrony of
entry into S phase. The effect has no relationship to the role of hyd r o x y u r e a or cytosine arabinoside as specific inhibitors o f D N A synthesis,
2~S. Mironescu and K. A. O. Ellem, J. Cell. Physiol. 90, 281 (1976).
[20]
261
since the serum-restricted cultures were mitotically quiescent; and a similar enhanced response can be induced in density-inhibited cultures by
only 2 hr of prestimulation exposure to cycloheximide, an inhibitor of
protein synthesis. Since we do not have any direct experience of the
ability of this treatment to increase the synchrony of density-inhibited
cells, the reader is referred to the original paper 27 for details.
Advantages and Disadvantages. This is an excellent method for obtaining cells synchronized at the G1/S boundary and the subsequent S phase.
With an appropriate choice of manipulations (length of serum deprivation,
percentage of serum during the restriction period, amount of and length of
exposure to hydroxyurea), toxicity can be low.
C. SYNCHRONIZATION BY DOUBLE-THYMIDINE BLOCK
The double-thymidine block has been found to be unreliable as the

only synchronizing agent. The concentration of thymidine, which in randomly growing cultures inhibits the rate of cell division by more than
90%, allows a considerable degree of DNA synthesis. ~8 Therefore, the
concentrations of thymidine commonly employed to produce cell synchrony do not arrest the cells at the GI/S boundary, but allow slow progress into S. However, the double-thymidine block can be used in conjunction with mitotic selection to obtain populations of synchronized cells.
For instance, in the case of HeLa cells the following procedure is
applicable.
HeLa cell cultures in suspension in the exponential growth phase are
treated for 16 hr with thymidine at a concentration of 2 mM. They are then
grown for 8 hr in regular medium and then again for 16 hr in medium
containing 2 mM thymidine. The cells are then plated in plastic dishes and
after 4 hr Colcemid is added. Beginning 2 hr after the addition of Colcemid, cells in mitosis are collected by shaking at half-hour intervals and
pooled as described above. This method allows the use of thymidine for a
partial synchronization and of mitotic selection for sharp synchronization.
At the same time, the damage done to cells by thymidine is repaired
during the plating period between the removal of the second thymidine
block and the addition of Colcemid. Again, note that it is not necessary to
use Colcemid, although several experiments have shown that Colcemid at
proper concentrations does not interfere with macromolecular synthesis
and other processes of HeLa cells in culture.
D. SYNCHRONIZATION BY CALCIUM DEPRIVATION
A decrease in the concentration of calcium may have a synchronizing

effect, especially on normal untransformed cells such as WI-38. Boynton
2~ G. P. Studzinski and W. C. Lambert, J. Cell. Physiol, 73, 109 (1969).
262
[20]
et al. 2a have shown that, when the concentration of calcium in the ex-
tracellular medium is reduced to 10/~M, WI-38 go into a state of quiescence and do not enter DNA synthesis. It may take more than 48 hr to
bring most of the WI-38 cells out of the cycle. Increasing the calcium
concentration to 1.35 mM at this point immediately produces a burst of
DNA synthesis which is highly synchronous. The rapidity with which this
burst occurs suggests that a decreased amount of calcium may synchronize the cells very closely to the onset of DNA synthesis. It should be
noted that as usual transformed cells cannot be synchronized by this
method but continue to grow even in very low concentrations of calcium. ~9
E. OTHER CHEMICAL METHODS
A number of other chemical methods for synchronizing cells in culture

have been proposed at one time or another and have already been discussed in the review by Nias and Fox ~r and elsewhere. Most of them have
only historical interest, while others may be useful in very specific situations but are not generally applicable. For instance, partial synchronization can be achieved by suicide with high concentrations of high specific
activity [aH]thymidine. Cells in S, which have incorporated huge amounts
of [aH]thymidine, are selectively killed. Clearly, such methods have only
a marginal interest for the biochemist.
Recommendations
These can be summarized as follows:
1. Synchronization by mitotic detachment is the best single method. It
is excellent for studying cells synchronized in mitosis, G~, and S. Its
limitations are the number of cells that can be obtained and the fact that it
is not applicable to cells in suspension.
2. To select cells in S or G2, a combination of serum deprivation
followed by serum stimulation in the presence of hydroxyurea is recommended. It gives larger yields than mitotic detachment, but again it is
limited to monolayer cultures.
3. To obtain large quantities of cells in G1 or S, the isoleucinedeprivation method is very good for some cell lines in suspension cultures. For other cell lines, one may try a double-thymidine block followed
by mitotic detachment (if the cells can be plated).
4. Other methods, at present, give only partially satisfactory results.
However, some of these methods are amenable to improvements that
should considerably enhance their value.
29 A. L. Boynton, J. F. Whitfield, R. J. Isaacs, and R. Tremlay, J. Cell. Physiol. 92, 241
(1977).
[21]
RECONSTITUTED
BASEMENT MEMBRANE RAFTS
263
[21] N e w T e c h n i q u e s f o r C u l t u r i n g D i f f e r e n t i a t e d Cells:
Reconstituted Basement Membrane Rafts
By LOLA M. REID and MARCOS ROJKIND
This article describes techniques, currently under investigation, that

should eventually provide routine methods for culturing differentiated
cells, whether normal or malignant. Indeed at present the techniques are
sufficiently developed that they can greatly lengthen the time that differentiated cells retain their tissue-specific functions in vitro.
The need for model systems of differentiated mammalian cell types to
be used in biological and clinical research has not been sufficiently
satisfied because of the inherent difficulties in establishing differentiated
cells, especially normal cells, in vitro. Epithelial cells have proven particularly difficult to establish in cell culture, a fact that has severely hampered
certain fields of research including some important cancer studies. Since
85-90% of all malignancies are carcinomas, malignancies of the
epithelium, the need for normal and malignant culture models of epithelial
cells is acute.
Over many years standard methods have evolved for keeping
explanted cells alive and functioning. Most methods used currently, as for
example those described in this volume, are for organ culture and for cell
culture. Analysis of these techniques and the response of cells to them
reveals methodologic limitations and certain requirements for cellular differentiation and growth. To facilitate the discussion we shall define several of the terms used in culture studies:
Organ culture: maintenance of organs dissociated from their central
vascular supply but with the organ, as an entity, left intact.
Tissue culture: culture of tissues or fragmented organs. The "sociocellular" relationships of the tissue architecture are preserved.
Cell culture: culture of an individual cell type divorced from other cell
types.
Subcell culture: maintenance of one or more subcellular componems of
homogenized cells.
A marked difference in response of explanted cells occurs when they
are cultured by cell rather than organ or tissue culture techniques. Methods which retain tissue architecture permit retention of tissue-specific
functions including hormone and pharmacological responses. However,
the tissue normally degenerates within a few weeks due primarily to difficulties in vascularizing all the cells within the tissue. Cell culture procedures overcome this limitation since they use explanted tissues which are

ISBN 0-12-181958-2
264
[21]
disaggregated into single cells. The cells are adapted to grow as a

monolayer on treated plastic, or as a cell suspension, and can be maintained in culture indefinitely. Nutrients are supplied by a liquid medium of
defined basal composition supplemented with one of various sera. Ideally,
one isolates a clonal cell line, i.e., a single cell whose progeny are maintained in continuous cell culture. There is genetic uniformity, easier
maintenance of the cells, and reduction of variables associated with a
multicell culture system. Nevertheless, cell culture procedures usually
result in distortion of cellular phenotype and karyotype; normal cells
rarely adapt as permanent cell lines without developing abnormal
karyotypes 1-3 or losing tissue-specific functions; 4-s malignant cells adapt
more easily than do benign tumor cells or normal cells; 4,s and fibroblasts
or stromal components are selected preferentially over epithelial cells. 7
The difficulties of establishing differentiated cells in cell culture have been
attributed to many causes. These include an inadequately defined basal
medium 9-12 (see also this volume [5]); inadequately defined hormone requirements; ~3 the static conditions of cell culture in which nutritional and
oxygen gradients develop and limit the growth and functioning of cells
with strict nutritional and oxygen requirements; 6,14"~5and loss of or damage of cell-cell junctions, perhaps essential in growth and/or differentiation or both by the cell culture procedures of mechanical and enzymic
J. L. Biedler, in "Human Tumor Cells in Vitro" (J. Fogh, ed.), p. 359. Plenum, N e w York,
1975.
2 H. L. Goldblatt, L. Friedman, and R. L. Cencher, Biochem. Med. 7, 241 (1973).
3 R. Parshad and K. K. Sanford, J. Natl. Cancer Inst. 47, 1033 (1971).
4 R. M. Cailleau, in "Human Tumor Cells in Vitro" (J. Fogh, ed.), p. 70. Plenum, N e w
York, 1975.
5 D. J. Giard, S. A. Aaronson, G. J. Todaro, P. Amstein, J. H. Kersey, H. Dosik, and W. P.
Parks, J. Natl. Cancer Inst. 51, 1417 (1973).
6 R. Knazek, P. Gullioo, P. Kohler, and R. Dedrick, Science 178, 65 (1972).
7 G. H. Sato, L. Zaroff, and S. Mills, Proc. Natl. Acad. Sci. U.S.A. 46, 963 (1960).
s y. Yasumura, A. H. Tashjian, Jr., and G. Sato, Science 154, 1186 (1966).
9 R. G. Ham, Proc. Natl. Acad. Sci. U.S.A. 53, 288 (1965).
10 A. Leibovitz, in "Human Tumor Cells In Vitro" (J. Fogh, ed.), p. 23. Plenum, N e w York,
1975.
tx W. McKeehan, W. G. Hamilton, and R. G. Ham, Proc. Natl. Acad. Sci. U.S.A. 73, 2023
(1976).
~2 C. Waymouth, in "Growth, Nutrition and Metabolism of Cells in Culture" (G. H.
Rothblat and V. J. Cristofalo, eds.), Vol. 1, p. 11. Academic Press, N e w York, 1972.
13 G. Sato and L. Reid, in "Biochemical and Mode of Action of Hormones" Series II. (H. V.
Rickenberg, ed.), Ser. II, Chapter 7. Univ. Park Press, Baltimore, Maryland, 1978.
14 j. Leighton and R. Tchao, in " H u m a n Tumor Cells In Vitro" (J. Fogh, ed.), p. 23. Plenum,
N e w York, 1975.
15 W. F. McLimans, D. TI Mount, S. Bogtich, E. J. Crouse, G. Harris, and G. E. Moore,
Ann. N.Y. Acad. Sci. 139, 190 (1966).
[21]
RECONSTITUTED BASEMENT MEMBRANE RAFTS
265
dissociation into single cell suspensions. 16-19 Undoubtedly all of these

have contributed to the impasse in maintaining differentiated cells in cell
culture. Yet despite some progress on these various fronts, the goal of
routinely culturing normal differentiated cells remains elusive. It is likely
that still other considerations must be made to achieve this goal.
We believe that substantial progress in culturing differentiated cells,
including normal cells, will be made possible by study of cell-cell interactions within the tissue matrix. The relevance of tissue architecture to
proliferation and functioning of cells has not yet been fully ascertained,
but the studies discussed below suggest that some of the interactions
within the tissue matrix are critical for normal cellular physiology. An
almost ubiquitous and, therefore, probably primary relationship within
tissues is that between the epithelium and mesenchyme. In vivo, normal
epithelial cells capable of proliferation or long-term survival are attached
to a basement membrane which, in turn, is associated with mesenchymally derived cells, most commonly fibroblasts. The basement membrane,
a layer of secretion located between and produced by the epithelium and
the mesenchymal cells, is assumed to contain the extracellular subset of
factors affecting the epithelial-mesenchymal interactions.
The epithelial-mesenchymal relationship has been studied extensively
in developmental systems since the 1950s. 2-26 Many laboratories have
shown that differentiation of and the direction of differentiation of the
epithelium are dependent on factors derived from the mesenchyme. More
recently, Green and his colleagues have found that differentiation of skin
epithelium depends on factors derived from the mesenchyme. 27 Martin
found that an embryonal carcinoma cell line proliferates as a stem cell as
long as it is cultured on a feeder layer and differentiates when it is sepa-
16 I. Fentiman, J. Taylor Papadimitriou, and M. Stoker, Nature (London) 264, 760 (1976).
lr W. R. Loewenstein, Proc. Can. Cancer Res. Conf. 8, 162 (1969).
~8 N. S. McNutt, R. A. Hershberg, and R. S. Weinstein, J. Cell Biol. 51, 805 (1971).
~a J. D. Pitts and R. R. Burk, Nature (London) 264, 762 (1976).
20 G. R. Cunha, Int. Rev. Cytol. 47, 137 (1976).
21 p. Davies, A. C. Allison, and C. J. Cardella, Philos. Trans. R. Soc. London 271, 363
(1975).
22 R. Fleischmajer and R. Billingham, eds., "Epithelial-Mesenchymal Interactions." Williams & Wilkins, Baltimore, Maryland, 1968.
23 C. Grobstein, Science 11, 52 (1953).
24 R. L. Pictet, and W. L. Rutter, in "Cell Interactions in Differentiation" (M. KarkinenJaafkelainen, L. Fafen, and L. Weiss, eds.), p. 339. Academic Press, New York, 1977.
25 j. W. Saunders, Jr., M. T. Gasseling, and M. D. Gfeller, J. Exp. Zool. 317, 39 (1958).
2~ F. L. Vaughan and I. A. Bernstein, Mol. Cell. Biochem. 12, 171-179 (1976).
2r H. Green, J. G. Rheinwald, and T. Sun, in" Shape and Surface Architecture," p.493. Allen
R. Liss, Inc., New York, 1977.
266
[21]
rated from the feeder cells. 2ra The type of stromal cell associated with the
epithelium is also important as shown by Sakakura and her associates who
demonstrated that the differentiation of mouse mammary gland depends
on association with mammary stroma but not other types of stroma. 2s
Conclusive evidence for epithelial-mesenchymal interactions in adult
tissues is not yet available. However, such a relationship is suggested by
the histopathological observation that the anatomical relationship between the epithelium and the mesenchyme exists in normal and benign
tumor tissue but is missing in malignant tissues. Malignancy is, in fact,
defined in solid tissues when the basement membrane is disrupted 29.3and
when either epithelial cells alone (carcinoma) or mesenchymal cells alone
(sarcoma) proliferate with concomitant loss of the epithelialmesenchymal association. This observation may explain, in part, why
malignant ceils establish in cell culture more easily than do normal cells;
normal cells may be interdependent with other cells in the tissue matrix,
whereas malignant cells may be qualitatively or quantitatively independent of these interactions.
Some of the secretions of the epithelium and the mesenchyme found in
the basement membranes are known to be important for the maintenance
of cultured epithelial cells. Best studied of the basement membrane components are the collagenous proteins, adhesion proteins, and
glycosaminoglycans.
Collagen has been used frequently for establishing primary cultures
since most normal epithelial cells will attach more efficiently to collagen
than to other cell culture substrates. 4 This is true despite the fact that
basement membrane collagen (Type IV collagen) is genetically distinct
from the collagen typically used in cell culture (Type I). More detail on
collagen types is given in subsequent sections. Floating collagen substrates have been used by Pitot's group for normal liver cells 31 and by
Pitelka's group az for normal mammary epithelium to extend the survival
of tissue-specific functions of those epithelial cells for up to a month. The
advantages of floating collagen substrates over collagen plates (in which
collagen is present at the bottom of the plate) are thought to be increased
oxygen tension at the air-media interface and the ability of epithelial cells
to contract the raft and, thereby, to undergo a change in cell shape apparently essential for differentiated cellular functions, al-aa
27a G. Martin, personal communication.
T. Sakakura, Y. Nishizuka, and C. J. Dawe, Science 194, 1439 (1976).
29 L. Liotta, L Leinerman, P. Catanzaro, and D. Ryrnbrandt, J. Natl. Cancer Inst. 58, 1427
(1977).
a0 S. Read, Guy's Hosp. Rep. 123, 53 (1974).
3~ G. Michalopoulos and H. C. Pitot, Exp. Cell Res. 94, 70 (1975).
3~ j. T. Emermanand D. R. Pitelka, In Vitro 13, 346 (1977).
83 T. Allen and C. S. Potten, Nature (London) 264, 545 (1976).
es
[21]
267
Adhesion proteins such as LETS (Large External Transformation

Substance) are present; also called fibronectin and CSP (Cell Surface
Protein), this is a high-molecular-weight protein attached to cell surfaces
and known to bind cells to collagen or to the plastic dishes. 8-~4-41 Digestion of plasma membrane components by nonspecific proteases reduces
the amount of adhesion proteins such as LETS 42,4athat have been implicated in adhesion phenomena. 35 Lack of these proteins on the cell surface
and in the basement membranes is correlated with changes in cell morphology, 44 alterations in the organization of the cytoskeleton, 44,45 ceil
motility, 46 and alterations in the growth pattern in culture. 47 Hepatocytes
isolated after perfusion of the liver with bacterial collagenase show profound alterations in morphology 45 that appear to return to normal after
several hours in culture. Bacterial collagenase has been used because of
its high specificity in digesting the collagen in the stroma of tissues. 48
However, most commercial preparations of collagenase are contaminated
with nonspecific proteases that could be attacking the adhesion proteins
on cells. Purified bacterial collagenase is ineffective in releasing hepatocytes from rat liver (unpublished observations), suggesting that the release of hepatocytes from the collagen stroma requires enzymic digestion
by enzymes other than, or in addition to, collagenase. LETS protein is
decreased in many malignantly transformed cells 35 and disappears from
the cell surface of fibroblasts dissociated by trypsin; 4z'43 it reappears approximately 10 hr after the cells are plated in a serum-supplemented medium.43
Glycosaminoglycans, also called mucopolysaccharides, are polymers

s4 E. Engvall and E. Russlahti, Int. J. Cancer 20, 1 (1977).
ss R. O. Haynes, Biochem. Biophys. Acta 458, 73 (1976).
3s M. Hi56k, K. Rubin, A. Oldberg, B. Obrink, and A. Veri, Biochem. Biophys. Res. Commun. 70, 726 (1977).
37 R. J. Klebe, Nature (London) 250, 248 (1974).
38 K. M. Yamada, S. S. Yamada, and I. Pastan, Proc. Natl. Acad. Sci. U.S.A. 72, 3158
(1975).
3a K. M. Yamada, S. S. Yamada, and I. Pastan, Proc. Natl. Acad. Sci. U.S.A. 73, 1217
(1976).
40 K. M. Yamada, D. Schlesinger, D. Kennedy, and I. Pastan, Biochemistry 16, 5552 (1977).
41 M. Zimmerman, T. Devlin, and M. Pruss, Nature (London) 185, 315 (1960).
42 p. Bornstein and J. F. Ash, Proc. Natl. Acad. Sci. U.S.A. 74, 2480 (1977).
43 j. Wartiovaara, E. Linder, E. Rouslahti, and A. J. Vaheri, J. Exp. Med. 140, 1522 (1974).
D. Goldman, C. Chang, and J. F. Williams, Cold Spring Harbor Symp. Quant. Biol. 39,
601 (1974).
45 j. C. Wanson, P. Drochmans, R. Mosselmans, and M. F. Ronveaur, J. Cell Biol. 74, 858
(1977).
4e M. H. Gall and C. W. Boone, Exp. Cell Res, 70, 33 (1972).
4T E. Martz, H. M. Phillips, and M. S. Steinberg, J. Cell Sci. 16, 401 (1974).
4s S. Seifter, and E. Harper, in " T h e Enzymes" (P. D. Boyer, ed.), 3rd ed., Vol. 3, p. 649.
268
[21]
of sulfated or acetylated disaccharides that are often associated with proteins forming complexes referred to as proteoglycans. The glycosaminoglycans and proteoglycans are present in all connective tissues where they
form a gelatinous charged matrix. The best known and most studied of the
glycosaminoglycans and proteolgycans include hyaluronate (skin, aorta),
dermatan sulfate (skin, tendon, aorta), chondroitin sulfate (cartilage,
bone, cornea, notochord, skin), heparan sulfate (aorta, lung, liver), and
heparin (lung, liver, skin). Their chemical structures, biosynthesis,
specific tissue localization, variations among species, and evolution are
reviewed elsewhere. 49
In dovelopmental systems it has been shown that glycosaminoglycans
and proteoglycans together with collagenous proteins can have a significant influence on cellular differentiation, 5'51 proliferation, ~2aggregation, 53
and migration. 54 Hyaluronate stimulates aggregation of chick embryo
ceUs. 54 In the development of the cornea and the notochord, hyaluronate is
thought to play a role in collagen fibril formation that in turn induces
cellular migration. Further development of these tissues involves replacement of hyaluronate with sulfated glycosaminoglycans that are the
typical polyanionic glycans of the adult tissues. Development of the
mouse salivary gland requires the presence of chondroitin sulfate at the
cell surface. Specific enzymes that eliminate the chondroitin 6-sulfate
from the developing tissues arrest their development; morphogenesis can
be reactivated by renewed biosynthesis of the proteoglycan. 51'55-57
In adult tissues the functions of the proteolgycans and glycosaminoglycans in cellular differentiation and proliferation are uncertain.
Kraemer's studies 5s-6 indicate that mammalian cells in cell culture are
coated with a variety of glycoproteins and polyanionic glycans among
which the most common is heparan sulfate. Kraemer hypothesizes that
49 M. B. Mathews, "Connective Tissue: Marcromolecular Structure and Evolution."
Springer-Verlag, Berlin and New York, 1975.
50 E. A. Balazs and B. Jacobson, in "The Amino Sugars" (E. A. Balazs and R. W. Jeanloz,
eds.), Vol. 2A, pp. 281-309. Academic Press, New York, 1966.
51 M. R. Bernfield, S. D. Banerjee, and R. H. Cohen, J. Cell Biol. 52, 674 (1972).
52 M. Lippman, in "Epithelial-Mesenchymal Interactions" (R. Fleischmajer and R. E. Billingham, eds.), p. 208. Williams & Wilkins, Baltimore, Maryland, 1968.
B. Pessac and V. Defendi, Science 175, 898 (1972).
54 B. P. Toole and R. L. Treistad, Dev. Biol. 26, 28 (1971).
55 M. R. Bernfield and S. D. Banerjee, J. Cell Biol. 52, 664 (1972).
58 R. H. Cohn, S. D. Banerjee, and M. R. Bernfield, J. Cell Biol. 73,464 (1977).
57 R. H. Cohn, J.-J. Cassiman, and M. H. Bernfield, J. Cell Biol. 71,280 (1976).
58 p. M. Kraemer, in "Biomembranes" (L. A. Manson, ed.), Vol. 1, p. 67. Plenum, New
York, 1971.
59 p. M. Kraemer, Biochemistry 10, 1437 (1971).
a0 p. M. Kraemer, Biochemistry 10, 1445 (1971).
[21]
269
heparan sulfate is crucial to some cellular process(es). 6 Chiarugi and

Vannucchi 61 also focused on heparan sulfate as perhaps a significant factor
in the differentiation of cells. However, no conclusive evidence exists
confirming these suggestions, and the possible roles of the glycosaminoglycans await further experimentation.
On the basis of these previous studies it seems reasonable that basement membrane components might facilitate cultures of differentiated cell
types. We propose that new culture procedures should be developed for
differentiated cells that involve utilization of these basement membrane
components and co-culturing of multiple, interacting cell types representing a subset of the relationships existing in the tissue matrix. Generic
characterization of the procedures might be called "sociocell" culture.
One such technique which we are developing simulates epithelialmesenchymal relationships and utilizes reconstituted basement membrane rafts on which are floated epithelial cells over primary cultures of
mesenchymal cells normally in association with the epithelial cells. Others
have already shown that use of collagen rafts for primary cultures of
normal tissues can significantly enhance the retention of tissue-specific
functions of normal cells in cell culture. 31'32 We have modified the raft's
composition to more nearly resemble the in vivo substrate for epithelial
cells, the basement membrane. Then we float the rafts over primary cultures of mesenchymal cells in order to provide other epithelialmesenchymal factors that may be labile, may be taken up by the cells,
and/or are the result of ongoing interactions between the two cell types.
Our efforts to define the raft's composition are focusing on three of the
major constituents of the basement membrane: collagen types, adhesion
proteins, and glycosaminoglycans. Since normal epithelial cells may be in
contact with other collagen types, we are assaying the efficacy of
collagen rafts made with collagen types other than Type I. Adhesion
proteins, and glycosaminoglycans. Since normal epithelial cells may be
in contact with other collagen types, we are assaying the efficacy of
cell types that have lost their adhesion proteins due to neoplastic transformation. A possible complication with some cell types is the cellular
secretion of proteases; addition of adhesion proteins to such cells may or
may not facilitate attachment, depending on the amount of proteolytic
activity present. Addition of glycosaminoglycans, e.g., dermatan sulfate,
keratan sulfate, chondroitin sulfate, and heparan sulfate, is being pursued
to define empirically their participation in the proliferation and/or functioning of epithelial cells.
Since normal differentiated epithelial cells, e.g., endocrine cells, rarely
undergo mitosis in vivo in an adult organism, it is expected that success at
61 V. P. Chiarugi and S. Vannucchi, J. Theor. Biol. 61,459 (1976).
270
[21]
defining the reconstituted basement membrane rafts will result in the

cells' survival and functioning in "sociocell" culture but not in growth.
The reverse might be expected for stem cells such as the skin epithelium
or the stem cells from the bone marrow. For stem cells, an effective
basement membrane raft and feeder layer may result in proliferation of
the cells without differentiation. Differentiation might occur when the
stem cells are detached from the raft.
The techniques presented here are: (1) preparation of the collagen raft,
a modified procedure of the original one developed by Michalopoulos and
Pitot; ~1 (2) purification procedures of the several collagen types and other
basement membrane components; and (3) the procedures for preparing
epithelial cell suspensions and feeder layers of mesenchymal cells. Although the efficacy of some of these procedures has been tested with only
a few cell types, we believe they will form the basis for future procedures
for culturing many differentiated normal and malignant cell types.
Culture Techniques Simulating the Epithelial- Mesenchymal Relationship

The techniques involve purification of collagenous proteins, adhesion
proteins, and glycosaminoglycans; mixing and gelling of these components into a substrate; preparation of pure epithelial cell suspensions;
addition and attachment of these cells to the substrates; detachment of
substrates to form rafts; preparation of feeder layers of stromal cells;and
transfer of rafts plus epithelial cells to dishes where they float over the
primary cultures of mesenchymal cells. For short-term studies (3-4
weeks), one can dispense with the feeder layers and use only the rafts of
reconstituted basement membrane as substrates for maintaining differentiated epithelial cells.
A. Purification of Collagenous Proteins and Other Basement Membrane

Components
1.
PREPARATION OF COLLAGEN EXTRACTS
The collagen class of proteins is a heterogeneous group of proteins that

nevertheless is characterized generally by a unique amino acid composition (about 30% glycine, 20% proline, 20% hydroxyproline, and a variable
content of hydroxylysine). To date, three distinct collagens have been
isolated from interstitial tissues and characterized on the basis of the
uniqueness of their individual polypeptide chains. These are designated as
Type I, Type II, and Type III collagens. Type I collagen has a chain
composition of [od(I)]2a2; it is the only collagenous component of bone,
tendon, and tooth. This type of collagen is also present with other colla-
[21]
271
gens in skin, liver, heart, and kidney, but is absent from normal hyaline
and elastic cartilages. Type II collagen has a chain composition of
[al(II)]a; it is the only collagenous component of hyaline and elastic cartilages. Type III collagen has a chain composition of [al(III)]a; it is present
in most tissues containing Type I collagen except for bone, tendon, and
tooth. 62-~ Collagen also has been isolated from structures identified morphologically as basement membranes. Despite suggestions that these collagens are homogeneous with a 'chain composition of [al(IV)]a, the evidence is more convincing that basement membranes constitute another
class of proteins containing collagenous sequences, ranging in molecular
weights from 25,000--100,000. 65-67 Collagens isolated from basement
membranes differ in amino acid composition from interstitial collagens in
that they contain 3-hydroxyproline, relatively more hydroxylysine, and
relatively less alanine. Recently tissues containing Type I and Type III
collagens have been found also to contain collagenous components similar
in amino acid composition to those found in basement membrane collagens. 68"6a Since these collagens are extracted from tissues and not from
just the basement membranes alone, they are referred to as basement
membrane-like collagens. The basement membrane-like collagens as well
as Type III collagen are solubilized mainly by limited proteolysis with
pepsin.
Collagen used as a cell culture substrate 21'7.71 or as a raft 31,32 is obtained from skin or tendon by extraction with neutral salt solutions or
dilute acid. These extracts contain primarily Type I collagen that has been
shown to substantially lengthen the survival in culture of liver cells al and
mammary epithelium, az Since other types of collagen are part of the in
vivo substrate for epithelial cells (e.g., in normal liver, Type III collagen 7z'ra and basement membrane-like collagens TM rather than Type I collagen are in close association with the hepatocytes), these collagens also
62 p. M. Gallop and M. A. Pax, Physiol. Rev. 55, 418 (1975).
K. I. Kivirikko and L. Risteli, Med. Biol. 54, 159 (1976).
64 E. J. Miller, Mol. Cell. Biochem. 13, 165 (1976).
6.~ B. G. Hudson and R. G. Spiro, J. Biol. Chem. 247, 4229 (1972).
66 B. G. Hudson and R. G. Spiro, J. Biol. Chem. 247, 4239 (1972).
67 T. Sato and R. G. Spiro, J. Biol. Chem. 251, 4062 (1976). See also, this volume [6].
68 E. Chung, K. Rhodes and E. J. Miller, Biochem. Biophys. Res. Commun. 71, 1167 (1976).
69 R. L. Trelstad and K. R. Lawley, Biochem. Biophys. Res. Commun. 76, 376 (1977).
70 M. B. Bornstin, Lab. Invest. 7, 134 (1958).
71 W. D. Hillis and F. B. Bang, Exp. Cell Res. 17, 557 (1959).
72 L. Biempica, R. Morecki, C. H. Wu, M. A. Giambrone, and M. Rojkind, Gastroenterology 73, 1213 (1977).
73 S. Gay, P. P. Fietzek, K. Remberger, M. Eder, and K. Kuhn, Klin. Wochenshr. 53, 205
(1975).
272
[2 l ]
should be used as substrates in attempts to improve the survival and/or

functioning of specific epithelial cell types.
a. Tail Tendon Collagen (Type I Collagen). Collagen is extracted from
tendon fibers dissected from the tail of the rat. ~4 Approximately 1 g of
tendon fibers is suspended in 100 ml of dilute acetic acid (0.01-0.25 M).a2
After incubation with constant stirring at 4 for 48 hr, solubilized collagen
is collected in the supernatant obtained by centrifugation at 27,000 g for 30
min. The collagen thus obtained can be used without further purification
for preparation of substrates or can be stored at - 2 0 until needed. The
residual tissue can be reextracted with acetic acid to improve the yields of
collagen. The collagen so extracted may contain small amounts of noncollagenous material that may or may not gel together with collagen. If
purified collagen is to be used, the acid-solubilized collagen can be precipitated a few times by dialysis against 0.02 M NA2HPO4 and redissolution in
acetic acid. 74
The concentration of solubilized collagen is estimated from the amount
of hydroxyproline after acid hydrolysis; 75it is assumed that each polypeptide chain of approximately 100,000 daltons contains 100 residues of hydroxyproline. 62 Although collagen can be estimated by the Lowry technique, 76a collagen standard is needed; collagen produces low color yields
due to its low tyrosine content.
b. Skin Collagen (Type I Collagen). Although skin from any animal may
be used, rat, mouse, and guinea pig skins are most commonly used. The
animals are killed, shaved, washed with soap and water, and thoroughly
rinsed. The animals are skinned, and the skins are weighed. From l0 g of
skin one obtains approximately 60-70 mg of collagen (about 1-2 mg of
collagen are needed per 35 mm cell culture plate). The skins are ground
and washed for 24 hr in 0.45 M NaC1 at a ratio of I g of tissue to 100 ml of
salt solution. Saline solutions extract small amounts of collagen, but the
yields are much smaller than those obtained with dilute acid extraction.
Thus, the saline Wash can be used to eliminate saline-soluble contaminants, an advantage since there are more contaminants in skin than in tail.
tendon, i.e., skin is tissue, whereas the tail tendonis primarily collagenous
protein. The washed mince is extracted with dilute acetic acid (0.1-0.25
M) and purified as described for tail tendon collagen. All procedures
should be carried out at 4 .
c. Other Collagen Types. Type III collagen and the group of basement
membrane-like collagens can be solubilized by limited proteolysis with
74 D. A. Hall, " T h e Methodology of Connective Tissue Research." Johnson-Brouvers, Ltd.,
Oxford, 1976.
75 M. Rojkind and E. Gonzalez, Anal. Biochem. 57, 1 (1974).
78 O. H. Lowry, N. J., Rosenbrough, A. L. Farr, and R. J. Randall, J. Biol. Chem. 193, 265
0951).
[21]
273
pepsin at 40. 68,77 The concentration of pepsin to be used is 0.1-1 mg/ml,

and the ratio of enzyme to substrate is 1 : 100. Collagen rendered soluble
after enzymic digestion for 6 hr at 4 is collected in the supernatant after
centrifugation at 27,000 g for 30 min and neutralized immediately to pH
7.0 in order to inactivate pepsin. The neutralized solution is then dialyzed
for 24 hr against 0.05 M Tris-chloride (pH 7.4) containing 0.45 M NaC1.
The collagen that precipitates is removed after centrifugation. The clear
supernatant is dialyzed for 24 hr against Tris-chloride (pH 7.4) containing
1.7 M NaCI. Type III collagen will precipitate and can be collected by
centrifugation as described. The clear supernatant liquid obtained after
removal of Type III collagen is dialyzed for 24 hr against the Tris buffer.
Type I collagen will precipitate and can be collected by centrifugation.
The remaining clear supernatant fluid containing a mixture of the basement membrane-like collagens is then dialyzed against 0.02 M NaH2PO4 in
order to precipitate these collagens. The collagens so obtained are soluble
in dilute acid or in neutral salt solutions.
The pepsin-solubilized collagens lack short segments of the nontriple
helical portions of the individual collagen chains. Since such segments
may participate in gel formation, the extent of digestion should be carefully controlled. In fact these collagens may not form firm gels by themselves so that mixtures of Type I collagen with varying proportions of
other collagens should be used to create rafts with which to test the
efficacy of other collagen types on cellular proliferation and functioning.
2. ADHESION PROTEINS (LETS)
LETS protein has an apparent molecular weight of 220,000 and is
present in relatively large amounts on the cell surface of normal fibroblasts. 35 The protein is organized in a reticular fashion 42,43,78 and mediates
the attachment of fibroblasts to collagen in culture.37"79It binds strongly to
collagen and has been purified recently by affinity chromatography on
collagen bound to Sepharose 4B. 34
To prepare LETS protein, monolayers of chick embryo fibroblasts are
plated in disposable 690 cm 2 roller bottles and fed daily. After reaching
confluency, monolayers are washed at 37 with 50 ml of Hanks's balanced
salt solution 4 times, and then they are rinsed 60 min with 25 ml of
serum-free medium (Dulbecco's Modified Eagle's Medium or Diploid
Growth Medium) containing 2 mM phenylmethylsulfonylfluoride (PMSF)
as a protease inhibitor. After another rinse with salt solutions, the
monolayers are extracted for 2 hr with 25 ml of serum-free medium conr~ E. Chung and E. J. Miller, Science 183, 1200 (1974).
7a A. Vaheri, E. Ruoslahti, B. Westermark, and J. Pont6n, J. Exp. Med. 143, 64 (1976).
r9 E. Pearlstein, Nature (London) 262, 497 (1976).
274
[21]
taining PMSF plus 1.0 M urea (ultrapure grade). After centrifugation of

the mixture at 25,000 g for 15 min, LETS protein is precipitated by adding
solid ammonium sulfate to 70% saturation and adjusting the pH to 7.4 with
NH4OH. After 30 min at 4 , the solution is centrifuged at 25,000 g for 15
min. The pellet is solubilized with 1/20 volumes of 10 mM cyclohexylaminopropane sulfonate (CAPS), p H I 1.0, containing 0.15 M NaC1
and 1 mM CaCI2 at a concentration of about 1 mg/ml. The solution is
adjusted to pH 11.0 with 5 N NAOH, dialyzed overnight against two
changes of 400 volumes of CAPS buffer, and stored at - 7 0 . The above
conditions provide a concentrated, nonaggregated preparation of LETS
protein. If LETS protein is to be added to collagen solutions or to substrates, the pH of the solution should be adjusted to pH 7.4 with HCI.
This method was described by Yamada et al. zs-40 and yields approximately 50% of the LETS protein present in cultures of chick fibroblasts.
3. GLYCOSAMINOGLYCANS/PROTEOGLYCANS
Purification procedures for the various glycosaminoglycans and proteoglycans are presented by Hall 74 and in a review by Rod6n et al.; a
accordingly they will not be presented here.
B. Preparation o f Collagen or Reconstituted Basement Membrane

(RBM) Substrates
About 20 years ago, undenatured collagen in neutral salt solutions at

37 was shown to "precipitate as a mass of typical cross-striated fibrils in
an opalescent gel."sl This is not to be confused with a gelatin (denatured)
gel. All the methods used for the preparation of collagen gels require the
removal of any acetic acid and/or equilibration of the collagen solution
with a buffer at a neutral pH and containing NaC1. 82 The ideal concentration for making rigid gels is approximately 1 mg/ml. If an extract is too
dilute, it should be concentrated by dialysis overnight at 4 against 0.02 M
Na2HPO4, and the resulting collagen precipitate should be centrifuged at
27,000 g for 30 min. The pellet is suspended in a volume of dilute acetic
acid solution appropriate to yield a 1 rag collagen per milliliter of solution.
The acid extract should be dialyzed extensively against a neutral buffer
using any buffer appropriate for subsequent experiments. Dialysis and all
handling of the collagen solutions should be at 4 to prevent collagen from
coming out of solution before addition to plates. For culture experiments,
L. Rod6n,J. R., Baker,J. A. Cifonelli,and M. B, Mathews,in "Methodsin Enzymology"
(V. Ginsburg, ed.), Vol. 28, Part B, p. 73. AcademicPress, New York, 1972.
sl j. Gross and J. Kirk,J. Biol. Chem. 233, 355 (1958).
a2j. (]ross and C. M. Lapiere,Proc. Natl. Acad. Sci. U.S.A. 48, 1014.
so
[21]
R E C O N S T I T U T E D BASEMENT MEMBRANE RAFTS
275
acetic acid extracts can be dialyzed against phosphate-buffered saline for

5-6 hr with several changes of saline in order to eliminate most of the
acetic acid. Completion of neutralization can then be done with serumfree medium containing a pH indicator to permit following neutralization
visually. The advantages of this form of neutralization are that the subsequent gels are impregnated with the medium to be used for the cells and
that one avoids overshooting neutralization to an alkaline pH where collagen molecules may be cleaved. The neutralized extract is viscous and
should be plated quickly onto petri or tissue culture dishes (1-2 ml p e r
35-mm dish and 3-4 ml per 60-mm dish). Gelled substrates should form
within an hour at 37. Collagen substrates should be sterilized by irradiating the plates with UV overnight or by irradiating with 50,000 rad of
cobalt 60 or cesium 135. The large amount of irradiation is necessary to
eliminate radiation-resistant bacterial spores. To minimize contamination
prior to sterilization one can add antibiotics (penicillin at 0.12 mg/ml and
streptomycin at 0.27 mg/ml) to the collagen extract and to the serum-free
medium during the final dialysis.
To make reconstituted basement membranes (RBMs) one may add
LETS, and/or any glycosaminoglycan or proteglycan, to the collagen
extract during the final dialysis and copolymerize them onto the culture
plates. One can also try coating the surface of the collagen gel with any of
these components. The exact concentrations necessary for a given cell
type are undefined at present, but approximate concentrations derived
from previous culture experiments suggest 50 fig/ml for LETS 39 and 50
/~g/ml for glycosaminoglycans.
C. Preparation of Pure Epithelial Cell Suspensions

1. CELL LINES OR CELL STRAINS
Since cell lines and cell strains are cloned populations of cells, they are
a source of pure epithelial cells that is the easiest to use. Although cell
lines and cell strains consist of cells that are adapted to cell culture conditions and therefore do not require for cell culture a simulated epithelialmesenchymal relationship, there may be experiments in which feeder
layers are necessary for the cells. For these circumstances one can readily
use collagen rafts to facilitate separation of the epithelial cells from the
feeder layer. We do not know at present whether established, functional
epithelial cell lines will regain any tissue-specific functions when cocultured with other cell types.
To prepare cell lines or cell strains for"sociocell" culture, the medium
is removed from culture plates of the cells which are then washed with
phosphate-buffered saline and trypsinized with a 0.1% trypsin solution at
37 for several minutes. The cells should be observed closely during tryp-
276
[21]
sinization; as soon as the cells begin to detach from the plates, medium
containing serum is squirted onto the plates. All serum (chicken serum is
an exception) contains a trypsin inhibitor. Thus, squirting the cells with
the serum-supplemented medium will complete the detachment from the
plate and inactivate trypsin. The cell suspension is centrifuged in a desktop centrifuge at 900 rpm at 4, the supernatant liquid is removed, and the
pellet of cells resuspended in cold (4) medium containing serum. The cell
suspension should be kept on ice until ready for addition to the collagen or
RBM substrates.
2. TRANSPLANTABLE TUMORS
Transplantable carcinomas can be cleaned of nonepithelial cells by

transplanting the tumors into immunologically suppressed, xenogenic
hosts for one to two passages (see this volume [31]). Carcinomas transplanted into a xenogeneic host undergo replacement of their stroma and
vascular cells by host stroma and vascular cells. Treatment of primary
cultures or cell suspensions of transplanted tumors with antiserum against host ceils will provide suspensions or plates of cells containing only carcinoma cells. For example, nonmouse carcinomas can be
transplanted into athymic nude mice, whose T-cell deficiency permits
heterograft transplantation, s3,84 Tumors successfully transplanted will
contain tumorigenic epithelial cells associated with mouse stroma and
vascular components. 8~ The tumors may be excised, minced finely to
release cells, and the cell suspension treated with antimouse antiserum.
Procedures for making antimouse antiserum (or antiserum against any
host used for the tumor transplantations) are given below.
a. Preparation of Rabbit Antimouse Antiserum (or Antiserum for Other
Hosts). Mouse cells (host cells), ideally a cell line derived from mouse
cells, are grown to confluence in culture, subcultured, and 2 108 cells
injected subcutaneously into each of several rabbits. Ten days later, the
rabbits receive booster shots containing 2 108 of the same cells. Seven
days later the rabbits are bled and the serum assayed for its titer against
mouse (host) cells. The rabbits are given booster shots every 2 weeks and
test bled for antibody titer until the titers are high (the minimum dilution
to achieve 99% kill should be 1 : 100). The antiserum is tested against the
host cells used as the antigen and against the tumor cells used as a control.
Rabbit blood is collected in a sterile Falcon conical tube, allowed to clot at
room temperature for 2 hr, and then refrigerated overnight. The clot is
83 C. O. Povlsen and J. Rygaard, Acta Pathol. Microbiol. Scand., Sect. A 79, 159 (1971).
84 j. Rygaard, "Thymus and Self: Immunobiology of the Mouse Mutant Nude." Wiley, New
York, 1973.
~5 L. Reid and S. Shin, in "The Nude Mouse in Experimental and Clinical Research," (J.
Fogh and B. Giovanella,.eds.), pp. 313-351. Academic Press, New York, 1978.
[21]
277
centrifuged at 1000 rpm for 10 min, and the serum is decanted into a fresh
sterile Falcon conical tube. Varying dilutions from 1 : 4 to 1 : 500 are made
o f antiserum in serum-free media.
b. Test for Antibody Titer. The antiserum at the varying dilutions is
added as a 1% solution in serum-free medium along with 1% unimmunized
rabbit serum as a source of complement; it is tested on both the host cells
and the tumor cells. A dilution is selected that gives a high titer of activity
against host cells and a minimal one against tumor cells. To test the titer, a
cell suspension or plate of cells attached to a substrate is rinsed with
phosphate-buffered saline, treated with the appropriate dilution of antiserum + complement, and the cells incubated at 37 for 1 hr after which
they are rinsed with phosphate-buffered saline and incubated in fresh
medium containing serum. Within an hour, the host cells will lyse and
only t u m o r cells remain.
3. PRIMARY TUMORS AND NORMAL TISSUE
Enriching primary tumors or normal tissu.es for epithelial cells is more

difficult than using either clonal cell lines or strains or transplantable
tumors. If one is not concerned with having pure epithelial cells, primary
cultures o f normal tissues or primary tumors can be plated onto the substrates. The t u m o r or normal tissue is dissected from the host under sterile
conditions, minced finely, and suspended in medium supplemented with
serum. The suspension is added to the substrates. Attachment of the cells
occurs within hours or at most overnight. Following attachment of the
cells, the substrates are detached and allowed to float. Such primary
cultures on floating rafts remain functional for 3-4 weeks.
For pure epithelial cell suspensions one can resort to a variety of
techniques that, although not entirely satisfactory, do give enriched populations of epithelial cells. One of the more straightforward procedures is
to layer dissociated cells (from mechanical dissociation, mincing or sieving, or from enzymic dissociation) onto a density gradient containing
sterile Ficoll or sucrose. The cells are centrifuged at 4 at 1000 rpm or
sedimented by gravity permitting the cells to separate according to size
and weight. Bands are collected that are enriched in epithelial cells and
centrifuged at 4 at 1000 rpm. The supernatant liquid is removed, the
pellet of cells suspended in medium supplemented with serum, and the
cells added to the substrates.
D. Attachment of the Cells to the Substrates and Release of the Substrates

to Form Rafts
Epithelial cell suspensions a r e added to sterilized collagen or RBM
substrates and incubated overnight at 37 in an incubator flushed with 5%
278
[21]
CO~ in air. Once the cells attach to substrates, the substrates can be
released to form rafts by rimming the gelatinous layer with a sterile
spatula. The substrates will float and can be transferred from one plate to
another by gentle pipetting with a large-mouth pipette. In culture, the rafts
with attached epithelial cells will contract to as little as one-half their
original size.
E. Preparation of Feeder Layers: Primary Cultures of Stromal Cells

Ideally one prepares primary cultures of stromal cells that are normally in association in vivo with the epithelial cells to be cultured. Thus,
for pancreatic epithelial cells, one would prepare pancreatic fibroblast
cultures; for breast epithelial cells, breast fibroblast cultures; and for skin
epithelium, skin fibroblast cultures. Tissue is dissected under sterile conditions from the animal, minced finely, and the mince suspended in medium containing serum. The suspension is added to regular cell culture
plates (plastic). The cultures gradually select for fibroblast cells. 7 When
plates are confluent, they can be irradiated with 5000 rad of cobalt 60 or
cesium 135 to eliminate all mitotic activity. Irradiation of the feeder layers
may prove to be unnecessary since the epithelial cells are separated from
the stromal cells by being on the rafts, and the irradiation may even impair
some epithelial-mesenchymal interactions. This possibility is currently
under investigation. The floating substrates of collagen or RBM with attached epithelial cells are transferred to and floated over these primary
cultures of fibroblasts.
[22]
AUTORADIOGRAPHY
279
[22] Autoradiography
By
GRETCHEN
H.
STEIN
and
ROSALIND
YANISHEVSKY
Autoradiography is a technique for detecting radioactivity through the

formation of silver grains in a photographic emulsion. An emulsion consists of crystals of silver halide, usually silver bromide, suspended in
gelatin. To increase their sensitivity, nuclear emulsions designed for autoradiography have a higher ratio of silver halide to gelatin than do emulsions designed for light photography. They have a very high efficiency for
low-energy 13 panicles such as those emitted by tritium (18 keVmax),
carbon-14 (155 keVmax), phosphorous-33 (250 keVmax), sulfur-35 (167
keVma~), and iodine-125 (35 keVmax). The efficiency is somewhat less for
high-energy /3 panicles, such as those emitted by phosphorus-32 (1710
keVmax).
Autoradiography is based on the same principle as photography except that the energy for conversion of silver bromide to metallic silver is
derived from ionizing radiation rather than from photons of light. As /3
panicles pass through an emulsion layer, they set electrons free in the
silver bromide crystals. These crystals contain sensitivity specks, which
are irregularities in the crystal lattices. The free electrons migrate to
sensitivity specks and attract silver ions, which combine with the electrons to form atoms of silver. A latent image of the autoradiogram is
formed when enough silver atoms accumulate at a sensitivity speck to
form a nucleus of metallic silver that catalyzes the conversion of the entire
crystal into a silver grain during development. Developers are reducing
agents that furnish electrons to reduce silver ions to silver atoms. Because
crystals that have latent images are reduced to metallic silver more
quickly than other crystals, development can be stopped when only the
latent images have developed into silver grains. Unreduced crystals are
preferentially dissolved in the fixer, leaving behind a pattern of silver
grains that denotes the presence of radioactive material.1
Materials
Nuclear Emulsions. Eastman Kodak (Rochester, N e w York) manufactures four liquid emulsions. NTB, NTB-2, and NTB-3 produce large
grains suitable for light microscope autoradiography. NTB grains are the
largest (approximately 2700/~ diameter) and NTB-3 grains are the smal-
l w. D. Gude, "Autoradiographic Techniques: Localization of Radioisotopesin Biological

Material." Prentice-Hall, Englewood Cliffs, New Jersey, 1968.

All fights of reproduction in any form reserved.
ISBN 0-12-181958-2
280
[22]
lest (approximately 2300 /~ diameter) of the three, z NTE, which is designed for electron microscope autoradiography, produces very small
grains (approximately 500/~ diameter). 3 NTB-3 is approximately twice as
sensitive as NTB-2, which is twice as sensitive as NTB or NTE. 4,s Because increased sensitivity means that less energy is needed to make an
activated crystal develop into a silver grain, the background increases
with the sensitivity.
Ilford Ltd. (Ilford, Essex, Great Britain) also manufactures nuclear
emulsions that cover a range of sensitivities and grain sizes. Ilford L4 is
frequently used for electron microscope autoradiography because it has a
small grain size (1200 A) and is much more sensitive than NTE. Recently,
Kodak has produced an improved emulsion for electron microscope autoradiography, which is also more sensitive than NTE. It is called Kodak
Special Product Type 129-01.
Dipping Vessels. We use Lab-Tek No. 4310 Cyto-Mailers as dipping
vessels (see Fig. 1). These vessels are thrown away when the emulsion is
discarded.
Slide Boxes. Small black Bakelite slide boxes (Scientific Products),
which hold 25 slides, are used for storing the slides during the exposure
period. Some investigators use this type of slide box as a light-tight box,
provided that it is sealed with tape. However, we prefer to store the taped
slide boxes inside a second light-tight box.
Desiccant. We include small packets of Drierite in the slide boxes
during the exposure period because nuclear emulsion is most sensitive
when it is dry and because moisture can cause latent image fading. The
packets of Drierite, including some blue indicator Drierite, are wrapped in
several layers of cheesecloth and secured with tape.
Autoradiography Darkroom
An autoradiography darkroom should exclude all light. This demand
exceeds the normal requirements of a photographic darkroom. (A model
darkroom for autoradiography is described by Kopriwa.6) The darkroom
should be inspected for light leaks when the eyes have adapted to darkness (a minimum of 15 min). Sources of light entry may be window and
door frames, door steps, ventilation shafts, or where pipes pass through
floors. Light may also come from electrical contacts that spark on switch2 L. G. Caro, Methods Cell Physiol. I, 327-363 (1964).
a M. M. Salpeter, Methods Cell Physiol. 2, 229-253 (1966).
4 A. Ron and D. M. Prescott, Methods Cell Biol. 4, 231-240 (1970).
5 A. W. Rogers, "Techniques of Autoradiography." Am. Elsevier, New York, 1973.
6 B. M. Kopriwa, J. Histochem. Cytochem. 11,553 (1963).
[2 ]2]
AUTORADIOGRAPHY
281
FIc. 1. A Lab-Tek Cyto-Mailer, which is used as a dipping vessel, is shown on the left.
The position of a slide rack used for drying emulsion-coated slides is shown on the right.
ing, equipment containing vacuum tubes that emit light from the rear,
clocks and watches that have luminous dials, and indicator lights such as
those on waterbaths. These objects should be removed or taped over to
shut out the light. Fluorescent lights continue to glow for an appreciable
time after being turned off and, if possible, should not be used in the
darkroom. The darkroom should have an entrance corridor with double
doors so that one may enter or leave without letting light into the darkroom.
A safelight may be used during autoradiography according to the
emulsion manufacturer's recommendations. For example, a dark red
Wratten No. 2 filter can be used for Kodak NTB, NTB-2, and NTB-3
emulsions. The power of the light bulb should be no greater than 15 W.
The safelight housing must be light tight so that light escapes only through
the filter. Safelights and filters may be purchased from Kodak. The
safelight should be mounted about 1 m above the working space. Turn off
the safelight when it is not needed, especially when the slides are drying,
because dry emulsion is the most sensitive.
Basic Procedure for Autoradiography of Tissue Culture Cells
1. Cell Culture
Cells are grown on 22 x 22 mm glass coverslips, thickness number 1,

in 35-mm plastic tissue culture petri dishes or on 22 50 mm coverslips in
four-well multiplates (Lux No. 5278, available from Flow Laboratories).
282
[22]
The coverslips are cleaned before use by: (1) 5 min in 95% ethanol to
remove grease; (2) 5 min in 1 N HCI; (3) 5 min in deionized water, followed by four short rinses in deionized water; (4) two rinses in 95%
ethanol; and (5) two rinses in 100% ethanol. They are sterilized in a 250
oven for 7 hr.
Cells seeded in small petri dishes and in four-well multiplates tend to
settle more heavily in the middle of these vessels. This creates different
growth conditions at the center and periphery. In addition, the cells may
be too crowded in the center for single cell analysis by autoradiography.
We have found that filling the vessels very full when the cells are seeded
reduces this problem. After the cells have attached, the volume of medium may be reduced. We use 5 ml medium to seed cells in 35-mm petri
dishes and 12 ml medium per well in four-well multiplates.
Tissue sections fixed to slides, or cell suspensions smeared on slides,
may be processed for autoradiography in the same way as cells grown on
coverslips. 1The use of a cytocentrifuge to flatten cells in suspension onto
slides is discussed in the section on self-absorption.
2. Fixation
We fix the cells by adding to the growth medium an equal volume of
3 : 1 methanol : acetic acid fixative that has been freshly prepared. After 10
min, the half-strength fixative is removed by aspiration and replaced with
an equal volume of undiluted methanol: acetic acid fixative. After 10 rain,
the coverslips are removed and allowed to air dry. Methanol : acetic acid
fixative can cause some bursting of cell plasma membranes; however, it
flattens the cells well. By adding the fixative to the medium first, we have
eliminated this problem for our experiments with human cells.
3. Removal of Soluble Pools of Radioactively Labeled Precursors

For many experiments using radioactive nucleotides, fixation with 3 : 1
methanol acetic acid adequately extracts the soluble pools of these precursors. For other experiments, such as when the cells are relatively
crowded (Fig. 2) or when they are labeled with radioactive amino acid~,
the cells are: (1) extracted with 5% trichloracetic acid (TCA) at 4 for 5
min; (2) washed 3 times with 70% ethanol, for 5 min each; (3) rinsed 2
times with 95% ethanol; (4) rinsed 2 times with 100% ethanol; and (5) air
dried, r TCA extraction is preferable to perchloracetic acid (PCA) extraction because PCA is difficult to wash out of cells and will inactivate the
emulsion if not completely removed.
r D. M. Prescott, Methods Cell Physiol. 1, 365-370 (1964).
[22]
AUTORADIOGRAPHY
283
FIG. 2. T98G human glioblastoma multiforma cells were labeled with 4 /~Ci/ml
[aH]thymidine(specific activity 50 Ci/mM) for 10 min. The cells were fixed in methanol : acetic acid (3 : 1) and extracted with 5% TCA. The slides were processed according to our basic
procedure, using an exposure time of 3 days. The cells were stained with Giemsa stain.
Silver grains are concentrated over the nuclei that synthesized DNA during the labeling
period.
4. Coverslip Mounting
The coverslips are mounted cell side up on clean microscope slides
using a drop or two o f P e r m o u n t (Fisher) or Euparol (Carolina Biological).
Generally we let the mounting medium dry for several hours at 60 , or
overnight at r o o m temperature, before the slides are dipped.
5. Preparation of Emulsion
We routinely use K o d a k NTB-2 for analysis of [aH]thymidine incorporation into tissue culture cells because it has low background, produces
grains large enough to count at x400 magnification, and needs only 3-4
days exposure to develop 25-50 grains/labeled nucleus under our experimental conditions (0.01 /~Ci/ml [aH]thymidine for 24 hr).
The emulsion is a gel at room temperature and is melted in a waterbath
at 39-43 . Although NTB-2 can be handled under a K o d a k Wratten No. 2
284
[22]
safelight filter, we suggest keeping the light off whenever possible, e.g.,
when the emulsion is melting. Gently stir the emulsion with a clean glass
rod to check its fluidity after approximately 30 min, being careful not to
create air bubbles.
It is useful to test the background of a new bottle of emulsion by
dipping a clean slide into the bottle, letting the slide dry for 2-3 hr in a
light-tight box, and then developing it. One bottle of emulsion (118 ml) is
enough to coat 1000-3000 slides. 7
Because repeated heating of the emulsion to melt it causes an increase
in the background, we dispense new emulsion into Lab-Tek Cyto-Mailers
and store them at 4. For subsequent experiments, a single Cyto-Mailer of
emulsion is melted for 15 min and used as a dipping vessel. In general, 10
ml of NTB-2 and 10 ml of distilled water are added to each vessel and
mixed gently. For relatively flat samples, diluted emulsion forms an adequate layer of emulsion. The major advantage of using diluted emulsion is
that subsequent staining is easier. For preparations that have an uneven
surface, undiluted emulsion covers better.
6. Dipping and Drying

A Cyto-Mailer of melted emulsion is held upright on the workbench in
a small beaker of 39-43 water or in the waterbath. We dip one slide at a
time, remove it with a slow even motion, and touch the bottom end of the
slide to a paper towel to drain the excess emulsion. The slide is allowed to
dry horizontally with the cell side up by supporting it in a test-tube rack
that has been turned on its side and tilted at a 45 angle (Fig. 1). In this
way, the wet slide is supported by its edges. The safelight is turned off
while the slides are drying or the test-tube rack with slides is placed in a
large light-tight box containing Drierite until the slides are dry, i.e., about
1 hr. They are placed in slide boxes containing packets of Drierite and
sealed with black electrical tape to exclude moisture and light. The sealed
slide boxes are placed inside a second light-tight box and exposed at 4.
There are many variations of this dipping procedure. Two slides can
be placed back to back to be dipped simultaneously and then pulled apart
to dry separately. 7 Slides can be dried vertically in a test-tube rack. However, vertical drying can cause a small gradient of increasing emulsion
thickness from top to bottom, and it may cause artifacts as discussed
below under "Autoradiographic Background." The emulsion can be
wiped off the back of the slide immediately after dipping rather than after
development. A drop of emulsion can be placed on the cells and spread
over the surface using the side of a 9" Pasteur pipette tip as a roller, s This
8L. E. Allred, personal communication.
[22]
AUTORADIOGRAPHY,
285
is an easy way to coat small samples, e.g., coverslip fragments. It can be

used without waiting for the coverslip mounting medium to dry. For large
coverslips, however, it is difficult to spread the emulsion evenly and without causing mechanically induced background.
7. Exposure
We expose autoradiograms at 4 to minimize latent image fading. 5
However, for short-term exposures room temperature is also satisfactory.
The duration of the exposure depends on the experimental procedure
used and the grain dengity desired. The way to determine an exposure
time is to include several test slides in the experiment. These can be
developed at various times and used to estimate the correct exposure time
for the experiment.
8. Development
We use Kodak D19 developer and Kodak Fixer 197-1746, a generalpurpose hardening fixer. These solutions are stored in brown bottles and
are discarded after 2 months. If the D19 turns yellow before then, it is
discarded. To develop slides, unused developer and fixer are poured into
staining dishes and are used at 18-20 maximum. The slides are allowed to
warm to room temperature and are placed in staining racks. Slides coated
with either NTB, NTB-2, or NTB-3 are developed for 2 min in D19, rinsed
10-30 sec in 1% acetic acid stop bath, and fixed for 5 min in Fixer. After
fixing, the slides are rinsed gently in running tap water for 20 min, given a
final rinse in distilled water, and air dried. Ideally, the running tap water
should be at the same temperature as the other solutions used to avoid
stressing the gelatin by temperature changes. A thermostatic regulator,
which may be purchased from a photographic supply store, can be used to
maintain the appropriate temperature. However, a regulator is a convenience rather than a necessity.
9. Staining
We use Giemsa Blood Staining Stock Solution (J. T. Baker Chemical
Co.) diluted 1:25 in 0.01 M sodium phosphate at pH 7.1. Developed
autoradiographs are stained for 50 min and destained for 5 min in the same
buffer. Thereafter, the film of emulsion on the back of the slide is wiped
off with a damp Kimwipe. This procedure gives good definition of the
nucleus and cytoplasm without staining the nucleus so darkly that grains
over it are difficult to count. Much shorter staining and destaining periods
286
[22]
may also be used successfully. The choice of a staining procedure depends on the nature of the experiment. Other stains are discussed below.
Other Techniques
Prestaining
Prestaining a specimen before applying nuclear emulsion generally
yields a clearer image than poststaining the developed autoradiogram.
This is so because, in prestaining, the gelatin of the emulsion layer is not
stained and the specimen stains more vividly and precisely. However,
only a few stains may be used for prestaining because most stains cause
increased background (positive chemography) even if they are removed
before the emulsion is applied. Prestains are useful for observing morphological details that may be obscured by overlying silver grains after
autoradiography. For example, chromosomes can be more easily identified before autoradiography. Stains used to detect enzymic reactions are
often used as prestains because the autoradiographic process alters or
destroys the enzymic reaction of interest)
Basic fuchsin, used in the Feulgen procedure, and aceto-orcein are the
most commonly used prestains. Even these acceptable prestains can
cause a high background if they are of low purity or are old. Loss of
radioactive material during staining is also a potential problem in using
prestains. The hydrolysis step of the Feulgen procedure may remove
some radioactivity from the specimen; however, for most qualitative studies, the loss of radioactivity is insufficient to alter the interpretation of the
results. ~,~0Aceto-orcein, used as a prestain, fades during the photographic
processing. Furthermore, aceto-orcein cannot be used as a poststain because it removes developed silver grains. The Feulgen procedure also
cannot be used for poststaining because grains are removed during the
hydrolysis step.
Poststaining
An effective poststain must overcome the presence of a layer of emulsion over the specimen and the chemical changes brought about by the
photographic processing. If the gelatin takes up too much stain, there will
be little contrast between the specimen and the stained gelatin. If the
gelatin affects the staining reagents, or provides an environment of inapR. Baserga and K. Nemeroff, Stain Technol. 37, 21 (1962).
10 W. Lang and W. Maurer, Exp. Cell Res. 39, 1 (1965).
[22]
AUTORADIOGRAPHY
287
propriate p H , or if the reactive groups on the specimen are altered, the

stains m a y be rendered totally ineffective.
M a n y o f the methylene blue--eosin type stains (Giemsa, Wright) are
effective poststains. The p r o c e d u r e for G i e m s a staining used in our laboratory is presented a b o v e in " B a s i c P r o c e d u r e . " H e m a t o x y l i n and eosin
m a y be preferable for greater cytological detail although this p r o c e d u r e is
m o r e time consuming. T h e r e are several lists o f stains that h a v e been used
with autoradiography, 11-14 and recipes for t h e m can be found in basic
histology t e x t b o o k s . 15-1r
R e m o v a l o f Emulsion
Occasionally, it m a y be useful to r e m o v e the emulsion f r o m a completed autoradiograph, e.g., if the b a c k g r o u n d is too high or if visualization of the underlying specimen is to be facilitated. Techniques for rem o v a l of emulsion h a v e been described by Bianchi et al. 18 and Rogers. 5
Stripping Film
K o d a k AR- I0 stripping film is a p r e p a r e d l a y e r o f gelled emulsion that
is 5 / z m thick, has a p p r o x i m a t e l y the same sensitivity as N T B , and produces grains suitable for light m i c r o s c o p e autoradiography. Its principal
a d v a n t a g e is that it gives a uniform layer of emulsion. Its disadvantages
are that it is m o r e difficult to apply than liquid emulsion, the stripping
process creates some static discharges that can cause background, and it
is m o r e difficult to stain the specimen through 5 /zm of emulsion. A detailed description of the technique for using AR-10 has been presented. ~
Double Isotope Autoradiography
T w o isotopes that h a v e different energies can be analyzed in a single
specimen. T h e y are distinguished b y the distance that their 0-particles
H c. P. Leblond, B. M. Kopriwa, and B. Messier, in "Histochemistry and Cytochemistry"
(R. Wegmann, ed.). Pergamon, Oxford, 1963.
12j. M. Thurston and D. L. Joftes, Stain Technol. 38, 231 (1963).
13R. Baserga, Methods Cancer Res. 1, 45-116 (1967).
~4L. K. Schneider and S. L. Pierce, Stain TechnoL48, 69 (1973).
15S. N. Thompson, "Selected Histochemical and Histopathological Methods." Thomas,
Springfield, Illinois, 1966.
le A. G. Pearse, "Histochemistry: Theoretical and Applied," 3rd ed. Longmans, Green,
New York, 1969.
~r G. L. Humason, "Animal Tissue Techniques," 3rd ed. Freeman, San Francisco, California, 1972.
~8N. Bianchi, A. Lima-de-Faria, and H. Joworska, Hereditas 51,207 (1964).
288
[22]
penetrate the emulsion. The most frequently used combination of isotopes

is tritium and carbon-14. The 13particles from tritium range in energy from
zero to 18 keV and those from carbon-14 range from zero to 155 keV.
Consequently, the discrimination between the two isotopes cannot be
complete because approximately 15% of the/3 particles from carbon-14
have the same energy as tritium's 13 particles.
In the double-exposure method, 19 slides are dipped in emulsion, exposed, and developed after a period of time appropriate for the amount of
tritium in the specimen. The ratio of tritium to carbon- 14 should be very
high, so that only a small fraction of the grains produced during this short
exposure are from carbon-14. The slides are then dipped in celloidin embedding solution (no. M-4700, Randolph Products Co., Carlstadt, New
Jersey) that has been diluted 1 : 2 in alcohol-ether. Finally, the slides are
dipped a second time for a period appropriate for the amount of carbon-14
in the specimen. The 13particles from tritium are not sufficiently energetic
to penetrate to the second layer of emulsion. Therefore, grains in the
second layer are produced only by carbon-14. The focal plane of the
grains in the two layers will be different and can be distinguished. Two
emulsions that produce different-size grains may be used for each layer to
allow easier discrimination between them. Alternatively, the silver grains
in one layer can be colored by dye-coupling. 2
A new procedure uses only a single layer of emulsion, which is expanded in glycerol after the silver grains have been developed. 21 Silver
grains at different levels above the cells can be counted separately as in
the double-exposure method.
High-Speed Scintillation Autoradiography

High-speed scintillation autoradiography is a technique for amplifying
the number of silver grains produced per radioactive disintegration. Several procedures have been published for the application of this method to
cells fixed on slides, zz-26 The basic procedure involves immersion of the
emulsion-coated specimen in a solution of scintillator, e.g., PPO (2,5diphenyloxazole) plus dimethyl POPOP [14-bis-2-(4-methyl-5-phenyl19 R. Baserga, J. Cell Biol. 12, 633 (1962).
20 E. O. Field, K. B. Dawson, and J. E. Gibbs, Stain Technol. 40, 295 (1965).
2~ S. W. Perdue, R. F. Kimball, and A. W. Hsie, Exp. Cell Res. 107, 47 (1977).
22 R. J. Przyblyski, J. Cell Biol. 43, 108a (1969).
23 G. S. Panayi and W. A. Neill, J. Immunol. Methods 2, 115 (1972).
24 B. G. M. Durie and S. E. Salmon, Science 190, 1093 (1975); B. G. M. Durie,ibid. 195, 208
(1977). Includes a correction to preceding paper.
25 A. S. Mukherjee and R. N. Chatterjee, Histochemistry 52, 73 (1977).
26 W. Sawicki, K. Ostrowski, and E. Platkowska, Histochemistry 52, 341 (1977).
[22]
AUTORADIOGRAPHY
289
oxazolyl)benzene] dissolved in dioxane. The emulsion becomes impregnated with scintillator molecules. Photons are released as /3 particles
pass through the scintillator. These photons activate more silver crystals
in the emulsion than would have been activated by the/3 particles alone.
Theoretically, the efficiency of the high-speed method of tritiumlabeled molecules could be 30-fold that of conventional autoradiography. 26 However, different investigators have had variable success with
this technique; in some cases only a small increase in efficiency was
produced and was accompanied by an increase in the background.
Autoradiography of Diffusible Substances

There are several techniques for preparing autoradiographs of cells
containing radioactivity in diffusible substances. Briefly, cells can be
freeze-dried and placed under a film of dried emulsion. 27 Cells can be
frozen and kept at low temperature until completion of the autoradiogram,
thus preventing solute redistribution during the process, z8 Cells can be
applied directly to dry emulsion-coated slides and air-dried although controis to monitor positive or negative chemography are necessary because
the wet specimen can interact with the emulsion, z9
Potential Problems
Self-Absorption
Increasing the thickness of a radioactive specimen beyond a certain
point does not increase the number of/3 particles entering the emulsion.
This phenomenon is due to self-absorption of the/3 particles by the specimen. The distance that/3 particles can penetrate depends on their initial
energies and on the density of the specimen. Because most of tritium' s /3
particles penetrate less than 3 /zm through cellular material, 3 selfabsorption is an important factor in autoradiography of tritium-labeled
whole cells. Nevertheless, tritium is useful for autoradiography because it
has a short pathlength and, therefore, can be localized better than more
energetic isotopes.
Tissue culture cells are sufficiently large that they must be in a flat
configuration for tritium autoradiography in order to minimize self~70. L. Miller, Jr., G. E. Stone, and D. M. Prescott, Methods Cell Physiol. 1, 371-379
(1964).
zs S. B. Horowitz, Methods Cell Biol. 8, 249-275 (1974).
2a W. E. Stumpf, Methods Cell Biol. 13, 171-193 (1976).
30 W. Maurer and E. Primbsch, Exp. Cell Res. 33, 8 (1964).
290
[22]
absorption. Many types of cells flatten out on the solid substrate on which
they are grown. Other types of cells are round, e.g., cells that are
grown in suspension. In addition, flat cells become round when they are in
mitosis and when they are removed from the substrate, e.g., with trypsin.
Spreading may take from 2-12 hr, depending on the cell type, the substrate, and the culture conditions. Generally, cells spread more rapidly on
tissue culture plasticware than on glassware. Coating glass with a layer of
evaporated carbon enhances spreading, al When round cells are to be used
for autoradiography, they can be flattened onto microscope slides with a
cytocentrifuge (Shandon Southern Instruments, Sewickley, Pennsylvania). The chambers of a cytocentrifuge are designed so that the centrifugal force drives the cells against a slide at the back of the chamber.
The cells are flattened on the slide and are then fixed as usual.
The choice of fixative affects the degree of flattening of the cells. Even
cells that are well spread in culture are several micrometers thick. For
example, Chinese hamster ovary cells in monolayer culture are about 7
/~m thick and can benefit from additional flattening. We have found that
3 : 1 methanol : acetic acid flattens cells much more effectively than neutral formalin fixative. The ratio of methanol : acetic acid may have to be
modulated to obtain an intact, well-flattened specimen; increasing the
concentration of acetic acid will flatten cells but may disrupt cellular
integrity.
Uniform self-absorption in flattened cells simply reduces the efficiency
at which/3 particles are detected by the emulsion, whereas nonuniform
self-absorption can create artifacts. For example, cells increase in size
during the cell cycle, and large G2 cells may not be flattened to the same
thickness as small G1 cells. Consequently, the efficiency could be less in
the G2 cells. Nonuniform self-absorption is also inherent within individual
cells because the density of various organelles differs: the nucleus is less
dense than the cytoplasm, which is less dense than the nucleolus, z
Autoradiographic Background
The most common causes of background are listed and discussed
briefly. More detailed discussions are given by Boyd, n2 Baserga, a3 and
Rogers. 5
1. Photographic Processing. Excessive background will occur if the
temperature of the developing solutions is high, or if the duration of
development is long. This problem can usually be recognized because
31G. A. Meek, "Practical Electron Microscopyfor Biologists." Wiley,New York, 1970.
32G. A. Boyd, "Autoradiographyin Biologyand Medicine." Academic Press, New York,
1955.
[22]
AUTORADIOGRAPHY
291
background grains will be scattered randomly throughout the emulsion

and will be smaller than the grains that arose from the radioactive material
in the specimen.
2. Chemography. Chemical reactants, particularly reducing agents,
present in the biological specimen or in solutions that come in contact
with the undeveloped autoradiograph, can produce a latent image in the
emulsion. This is termed positive chemography. The opposite, negative
chemography, occurs when chemical groups cause rapid latent image
fading, i.e., loss of the nucleus of metallic silver by recombination of the
silver atoms with bromine atoms to form silver bromide. Chemography is
uncommon in cell culture experiments unless caused by prestaining with
an inappropriate stain.
Chemography due to the specimen itself may be looked for by exposing unlabeled specimens to normal emulsion and to emulsion that has been
fogged by a brief exposure to light. Positive chemography will produce
grain densities higher than background in the normal emulsion. These
grains are often large and irregular and sometimes clumped. Negative
chemography will produce faded areas having grain densities lower than
the uniformly black background in the fogged emulsion.
3. Light Exposure. A high background may occur if the emulsion has
inadvertently been exposed to light, e.g., if the storage facilities for slides
are inadequate or if the darkroom has light leaks. A proper safelight is
especially important because the emulsion will be exposed under it for a
number of minutes.
4. Static Discharges. Electrostatic discharges may be emitted when
tape is stripped from its roll and used to seal a slide box or when it is
stripped from the slide box itself. Similarly, there may be electrostatic
discharges when stripping film is stripped from its support. Sometimes the
discharges may be accompanied by flashes of light. The discharges and
the light can cause background, which often appears as streaks of silver
grains running parallel to one another across the developed emulsion. This
problem can be controlled by cutting strips of tape before the emulsion is
removed from its light-tight bo.x and by untaping slide boxes slowly.
Increasing humidity and decreasing temperature in the darkroom also
help reduce these effects.
5. Pressure Effects. Emulsion is sensitive to pressure, e.g., from
scratches or fingerprints. Contraction of the emulsion as it dries also can
produce a high background if it occurs too quickly or goes too far. When
the emulsion layer is very thin, it may dry too fast. Drying slides coated
with dilute emulsion in the horizontal position rather than in the vertical
position is useful in avoiding the creation of a very thin layer of emulsion
at the top of the slide.
292
[23]
6. Environmental Radiation. Radiation from the environment, e.g.,

cosmic rays, is usually a relatively minor source of background and is
acceptable for most work. However, if desired, the background may be
reduced by shielding the emulsion and coated slides.
7. Other. Occasionally a latent image speck will form spontaneously.
This will occur more often with the more sensitive emulsions. Hence, the
choice of emulsion should be based on the least sensitive one that is
adequate for the experiment. Although the rate of formation of spontaneous background increases with age of the emulsion, we find a reasonable
background for 2-3 months after the expiration date suggested by the
manufacturer. However, one should always dip a test slide in the emulsion and check it before use in experiments in which a low background is
important.
Finally, everything from the coverslips used for cell culture to the
darkroom itself should be meticulously clean. Dust, spilled chemicals
including developer and fixer, and materials that come in contact with the
emulsion all may cause background. Glass, plastic, or high-grade
stainless-steel utensils are generally safe to use in contact with the emulsion but other metals, particularly copper, are to be avoided.
Acknowledgments
We wish to thank Dr. David M. Prescott, Dr. John Heumann, and Dr. DorothyPumo for
their help in preparing this manuscript. It was writtenwhile we were supported by American
Cancer Societygrants NP-156B (G.H.S.) and PF 1292(R.Y.).
[23] I n d u c t i o n a n d P r o d u c t i o n o f I n t e r f e r o n
By ROBERT M. FRIEDMAN
Many cells in culture can be stimulated to produce at least some
interferon. This is a convenient property because it permits the study of a
useful cell product with a specific biological activity, inhibition of virus
growth. Stimulation of interferon by cells in culture appears to be due to
induction of a protein that is not ordinarily made. Recent studies have
suggested that interferon production is controlled by a repressor ~ and is
determined by specific loci on chromosomes that have in some cases been
identified. 2
I R. L. Cavalieri, E. A. Havell, J. Vil~ek, and S. Pestka, Proc. Natl. Acad. Sci. U.S.A. 74,
4415 (1977).
2 p. p. Creagan, Y. H. Tan, S. Chen, and F. H. Ruddle, Fed. Proc., Fed. Am. Soc. Exp. Biol.
34, 2222 (1975).

ISBN 0-12-181958-2
[23]
INDUCTION
A N D P R O D U C T I O N OF I N T E R F E R O N
293
In general, it is best to use confluent monolayers of cells to produce

interferon. Cells that have been aged for a week after confluence in
monolayer cultures are often better producers of interferon than are fresh,
recently trypsinized cells. 3 In some diploid cultures the production of
interferon varies with the passage level of the cells; with excessive
passages the ability of a diploid cell line to produce interferon declines
significantly.4 In this case it is, of course, important to know the history of
the cells being used. In many systems, however, the passage history of
the cells does not affect the yield of interferon. This is certainly true of
heteroploid lines such as mouse L cells, which are usually excellent interferon producers at any passage level.
Interferon production in tissue cultures may be stimulated by viral and
nonviral inducers. Some viruses are usually excellent interferon inducers,
the best in general being among the group A arboviruses (also called
togaviruses) and the paramyxoviruse~. 4 With these viral inducers interferon production is usually maximal by 18-24 hr after induction. The group
A arboviruses are good inducers whether active or heat inactivated, and a
multiplicity of infection of 20-40 virus particles per cell usually suffices to
give optimal interferon production; too high a multiplicity of virus may be
inhibitory. The best generally available inducer in this group is
Chikugunya virus, which may be obtained from the American Type Culture Collection (ATCC, 12301 Parklawn Drive, Rockville, Maryland).
Chikungunya virus is usually propagated by intracerebral inoculation of
suckling mice with about 104 plaque-forming units (pfu) of virus. After 40
hr the mice are killed and their brain tissue homogenized to yield a 10%
suspension in tissue culture medium. The virus can be assayed by plaque
formation in Vero cells. Mouse brain suspensions produced in this manner
usually have titers well in excess of 109 pfu/ml.
To produce interferon with Chikungunya virus, human RSTC-2 cells
are infected at a virus : cell multiplicity of 30 for 1 hr at 37. Cultures are
washed 5 times and incubated for 22 hr in fresh medium with 2% serum.
In order to inactivate residual virus that would interfere with the assay for
interferon, the culture fluids are harvested and treated with drops of N
HC1 until the pH is 2. The medium is immediately brought back to neutrality with N NaOH. In cultures induced by this means, yields of interferon are of the order of 1500 mouse international reference units/ml if the
RSTC-2 cells are treated with actinomycin D (0.1 /zg/ml) for 1 hr before
infection with virus (see below on superinduction). Much lower interferon
yields are obtained from RSTC-2 cells if actinomycin D pretreatment is
omitted, but in many cell systems high interferon titers can be stimulated
E. A. Havell and J. Vil~ek, Antimicrob, Agents & Chemother. 2, 476 (1972).
4 M. W. Myers and R. M. Friedman, J. Natl. Cancer. Inst. 47, 757 (1971).
294
[23]
by infection with Chikungunya virus in the absence of antimetabolite

pretreatment. 4
Among the paramyxoviruses, Newcastle disease virus (NDV) is most
frequently employed as an interferon inducer. NDV is an excellent inducer in most systems other than chick cells. This is probably due to
replication of NDV with resultant severe cytopathic effects in chick cells.
In other systems NDV does not multiply significantly but does induce
high titers of interferon. NDV also may be obtained from the American
Type Culture Collection and, for this purpose, it is probably best to use
the B 1 vaccine strain which has little toxicity even for chickens. The virus
can be propagated in 9-10 day old embryonated chick eggs. The infected
embryos are harvested after 5 days when the yields are about 1000
hemagglutinin units (HAU)/mt. One HAU is equivalent to about 106 virus
particles. NDV can be assayed easily by its capacity to hemagglutinate
fowl red blood cells.
Interferon induction by NDV in human or mouse cells can be effected
with either active or inactivated virus. NDV is rapidly inactivated by
ultraviolet (UV) light (5000 erg/sec, 10-20 sec at 18 cm). In human cells,
20 HAU/106 cells have been used to induce interferon production; in other
systems, UV-inactivated virus is often employed. A multiplicity of 300
plaque-forming units of UV-inactivated virus per cell (virus titer as measured before inactivation) is required to give optimal yields in mouse L
cells. Cells are infected and handled in the manner described for production with Chikungunya virus. Since NDV is far more pH-stable than
Chikungunya virus inactivation must be carded out with great care. To be
on the safe side, it is best to leave NDV preparations at pH 2 for 4 days.
With mouse L cells, titers of interferon as high as 10,000 international
mouse reference units/ml have been obtained by this procedure. 5
For most purposes the only nonviral inducer of interferon that should
be considered is polyriboinosinic : polyribocytidylic acid (poly IC). Poly IC
is an inducer in many tissue culture systems, and excellent preparations of
poly IC are available from several commercial sources. It is important
that the instructions of the manufacturer be followed carefully in attempting to dilute poly IC preparations. Otherwise, the viscosity of the polymer
will result in an unevenly mixed solution. On the other hand, excessive
shaking may shear the molecule and this can alter its properties. A range
of 0.1-100/~g/ml should be tried to determine the optimal concentration of
poly IC necessary. With poly IC induction, interferon yields are usually
maximal 8 hr after induction. In order to obviate removal of poly IC from
the interferon preparation obtained, it is best to treat cells with the inA. J. Hay, J. J. Skehel, and D. C. Burke, J. Gen. Virol. 3, 175 (1968).
[23]
295
INDUCTION AND PRODUCTION OF INTERFERON

TABLE I
SUPERINDUCTION OF INTERFERON IN HUMAN F1BROBLAST CELL CULTURES
Cell line
(reference)
RSTC-2 (4)
RSTC-2
FS-4 (3)
FS-4
Treatment
Time
(hours)
Interferon
yield
poly IC (50 p,g/ml)

+ DEAE-dextran (25 t~,/ml)
poly IC + DEAE-dextran
+ puromycin (20 g,g/ml)
+ actinomycin D (1 /~tg/ml)
0-1
0-1
0-4
1-4
500 a
10,000 a
poly IC (100/~g/ml)
poly IC
+ cycloheximide (50 tag/ml)
+ actinomycin D (1 ttg/ml)
0-2
0-- 2
2-6
5.5-6
100b
10,000 b
a Yield in reference units per milliliter at 8 hr after inducer addition.

b Yield at 24 hr after inducer addition.
ducer for only an hour and then to wash the cells 3-5 times with saline or
medium. In addition, DEAE-dextran (10-100 ~g/ml) may be used to increase poly IC adsorption. The yield of interferon in human FS-4 cells
induced with poly IC alone (20/~g/ml) is usually less than 1000 units/ml. 3,4
Several steps may be taken with poly IC induction to increase interferon yields, but the single most effective way to do this is a superinduction
procedure that employs actinomycin D and cycloheximide or puromycin. 3,4 The timing of the addition of antimetabolites is critical. Monolayers
of cells confluent for at least 6 days are usually used, especially if interferon is being induced in diploid cells. After the cultures are washed with
buffered saline, Eagle's Medium (MEM, serum-free) containing poly IC
(5-20/zg/ml) and cycloheximide is added. After 5 hr, actinomycin D (final
concentration 1-5/zg/ml) is added to the medium for 1 hr and, 6 hr after
induction, the medium is removed and discarded. The cells are washed 4
times with saline and MEM with 0.2% human plasma protein (U.S.P.) or
2% fetal calf serum is added. The medium containing interferon is collected 30 hr after the onset of the production run. The yields of interferon
in FS-4 or RSTC-2 human diploid cells with this procedure are at least
10,000-20,000 reference units/ml so that a very significant potentiation of
interferon production may be obtained. However, it is important to note
that superinduction with antimetabolites does not work in all systems.
Mouse L cells, for instance, are not superinductable for interferon production. Table I reviews the results obtained with superinduction of interferon in two systems.
296
[24]
Priming is another practical procedure that has yielded high titers of

interferon with either viral or nonviral inducers.3.6 In this context priming
means exposure to interferon several hours before an inducer is added.
Primed cells are often much better interferon producers than nonprimed
cells. In the case of human diploid FS-3 cells, preincubation with 100
units/ml of interferon for 16 hr followed by induction with 100/xg/ml of
poly IC results in a yield of 512 units/ml of interferon; unprimed FS-3
yielded only 8 units/ml in the same study. However, superinduced FS-3
cells produced over 10,000 units/ml. Unfortunately a combination of
superinduction and priming did not give rise to significantly higher titers
of interferon than did superinduction alone. Some cell lines cannot be
primed, and in a few instances interferon treatment actually inhibits interferon production.
Once interferon is obtained it is necessary to assay it in order to
determine how many biological units are present. International standards
for mouse, rabbit, and human interferons can be obtained in small quantities from the Resources Research Branch, NIAID, Westwood Building,
NIH, Bethesda, Maryland. These are necessary to standardize individual
laboratory interferon preparations that in turn should be employed in each
individual assay as standards.
Two relatively fast biological assays have been described for interferon that employ rapidly growing viruses (Sindbis virus or EMC virus)
with a wide spectrum of infectivity; a simple hemagglutinin assay is also
available. 7,s These are the generally employed procedures that are easiest
to arrarfge and can be applied to assay a variety of interferons.
6 R. M. Friedman, J. lmmunol. 96, 872 (1966).
7 H. K. Oie, C. Buckler, C. Uhlendorf, D. A. Hill, and S. Baron, Proc. Soc. Exp. Biol.
Med. 140, (1972).
s p. Jameson, M. A., Dixon, and S. E. Grossberg, Proc. Soc. Exp. Biol. Med. 155, 173
(1977).
[24] E v a l u a t i o n o f C h e m i c a l C a r c i n o g e n i c i t y b y in Vitro
Neoplastic Transformation
By NOi~L BOUCK and GIAMPIERODI MAYORCA

The malignant transformation of cultured mammalian cells in response
to chemical carcinogens, first observed I and documented z many years
ago, has now been demonstrated in a number of different cells with a wide
variety of in vitro assays for transformation. Of the many combinations of
' W. R. Earle and A. Nettleship, J. Natl. Cancer Inst. 4, 213 (1943).
2 y. Berwald and L. Sachs, Nature (London) 200, 1182 (1963).
METHODS
IN ENZYMOLOGY,VOL. LVIII

All rights of reproduction in any formreserved.
ISBN 0-12-1819:58-2
[24]
CHEMICAL CARCINOGENITY
297
cell type and assay procedure that have been recorded, one of the more
rapid and reliable is described in this article; a it uses a fibroblast line of
baby hamster kidney cells, BHK 21/cl 13, 4 and assays for its transformation after a brief treatment by suspected carcinogen by testing the ability
of the treated cells to form colonies in soft agar. 5
This established BHK cell line offers advantages over cells newly
placed in culture in that cells grow rapidly (12-hr doubling time at 37),
transform with high frequency in response to a variety of carcinogens, 3,6-8,9'1'11 clone readily to give homogeneous populations, and are
amenable to culture for periods of time sufficient to carry untreated controls to the end of an experiment. The technique is sometimes criticized
because the cells may have already undergone one or more steps in the
process leading to malignant transformation, being "immortal" whereas
normal diploid cells are not. However, in defense of the method it should
be noted that recently cloned BHK cells are not tumorigenic in vivo at
moderate doses, TM whereas their virally and chemicallys transformed derivatives are. Thus the transformation induced in vitro clearly represents a
very crucial step in any series of changes that may be required for
tumorigenesis.
The soft agar assay for in vitro transformation, although slightly more
cumbersome than some others, has the advantage of selecting in a single
step for cells capable of anchorage-independent growth. This particular
characteristic of the transformed phenotype of a fibroblast cell is the only
selective one that is consistently correlated with in vivo tumorigenicity. 13
Cells and Culture Conditions
A culture of BHK 21/cl 13 should be obtained, preferably from someone currently working with it, at as low a passage level as possible. It
a N. Bouck and G. di Mayorca, Nature (London) 264, 722 (1976).
4 M. Stoker and I. Macpherson, Nature (London) 203, 1355 (1964).
5 I. Macpherson and L. Montagnier, Virology 23, 291 (1964).
6 y. Ishii, J. A. Elliott, N. K. Mishra, and M. W. Lieberman, Cancer Res. 37, 2023 (1977).
7 R. F. Newbold, C. B. Wigley, M. H. Thompson, and P. Brookes, Murat. Res. 43, 101
(1977).
s G. di Mayorca, M. Greenblatt, T. Trauthen, A. Soller, and R. Giordano, Proc. Natl.
Acad. Sci. U.S.A. 70, 46 (1973).
9 I. F. H. Purchase, E. Longstaff, J. Ashby, J. A. Styles, D. Anderson, P. A. Lefevre, and
F. R. Westwood, Br. J. Cancer 37, 873 (1978).
10 j. A. Styles, Br. J. Cancer 36, 558 (1977).
1i j. Ashby, J. A. Styles and D. Anderson, Br. J. Cancer 36, 564 (1977).
12 O. Jarrett and I. Macpherson, Int. J. Cancer 3, 654 (1968).
13 S. Shin, V. H. Freedman, R. Risser, and R. Pollack, Proc. Natl. Acad. Sci. U.S.A. 72,
4435 (1975).
298
[24]
grows well when cultured in Dulbecco's Modified Eagle's medium 14

(DME; see GIBCO catalogue) containing 10% autoclaved Difco Bacto
tryptose phosphate broth (TPB) and 10% calf serum in a humidified 10%
COz incubator at 37. (Hereafter this combination is referred to as complete medium.) When the cell density is very low, the serum concentration
is increased to 20%.
Selection of a Low Background Subclone of BHK
In order to obtain maximum sensitivity from a transformation assay
with this cell line, it is necessary to clone it so as to remove the background of spontaneous transformants and to select a clone in which they
will not arise again with too high a frequency.
1. Trypsinize a culture of BHK cells, dilute, and plate them at a very
low density; prepare, for example, 20-30 100-ram dishes of 200 cells per
dish, using complete medium with 20% serum. Incubate the dishes at 37
for 8-11 days until very heavy single colonies can be seen.
2. Carefully select about l0 of the colonies which, when compared to
their neighbors, appear to be unusually light and flat and have especially
feathery edges. Pick the colonies, trypsinizing them within a cloning cylinder secured with autoclaved silicone grease, into small, 30-mm dishes
containing complete medium with 20% serum. Allow them to grow and
passage them whenever they become 60-70% confluent.
3. As soon as there are sufficient cells, about passage 3, freeze several
vials of each clone and also test an aliquot of each by assaying the cloning
efficiency in liquid and agar. Plate 200 cells/60-mm dish containing complete medium with 20% serum and plate l06 cells/60-mm dish in soft agar
as described below under "Soft Agar Assay." Choose for subsequent
experiments a clone that gives a reasonable plating efficiency in liquid
(about 20%) and the minimal plating efficiency in agar (less than one
colony per l0 e cells).
When growing the cloned cells for use in carcinogenesis experiments,
culture cells from the frozen vials, using them prior to passage 7. Maintain
the cells at densities of less than 70% confiuency so as to avoid providing
any selective advantage to spontaneously arising transformants that can
grow more effectively than normal cells in crowded cultures.
Carcinogen Treatment
1. Prepare a fresh carcinogen stock solution, 1-10 mg/ml, in water or
in dimethylsulfoxide, depending on solubility. Sterilize by filtration. Di=
14j. D. Smith, G. Freeman, M. Vogt, and R. Dulbecco,
Virology 12, 185 (1960).
[24]
299
lute the stock solution in TD buffer 14 (0.8% NaC1, 0.038% KCI, 0.01%
Na~HPO4, 0.3% Trizma base, and HCI to pH 7.0 to give a range of concentrations that are double the final desired doses. The final concentration
of dimethylsulfoxide in contact with the cells should be less than 2%.
2. Harvest by trypsinization cells that have been grown from the selected clone, wash them by centrifugation in TD buffer containing 2% calf
serum, and resuspend them at 2 106 cells/ml in TD buffer without
serum.
3. Mix equal volumes of cell suspension and diluted carcinogen, usually 1-3 ml each, in sterile tubes containing small magnetic stirring bars
and stir very gently for 1 hr at 37. Include a zero-dose control tube
containing solvent and cells but no carcinogen.
4. At the end of the hour, dilute the cell-carcinogen mixture with 5-10
volumes of complete medium containing 20% serum, recover the cells by
centrifugation, and wash and resuspend them in fresh medium with 20%
serum.
5. To determine the amount of cell killing due to the carcinogen treatment, dilute an aliquot of the treated cells and plate in quandruplicate in
complete medium with 20% calf serum at densities of 200, 2000, or 20,000
cells/60-mm dish. The number of cells added per dish depends on the
expected killing. It is safest to plate at several cell densities from a given
dose. Count cplonies arising on these plates after incubation for 7 days at
37 .
6. Plate another aliquot of the treated and washed cells at a concentration of about 2 105 cells/100-mm dish in complete medium with 20%
serum and grow at 37 for 4 days. If, during this time, any dishes become
more than 70% confluent, the cells should be subcultured. At the end of
this expression period, cultured cells from each dose should be assayed in
soft agar as described below.
The Soft Agar Assay for Transformation

I. For each culture to be assayed, prepare 10 60-ram dishes, each
containing 5 ml of complete agar medium. Complete agar medium consists
of 40% double-strength D M E , 10% Difco tryptose phosphate broth, 10%
calf serum, and 40% of an agar solution. The agar solution is prepared
by mixing 1.275g of Difco Bacto agar per I00 ml of glass-distilledwater.
The solution is sterilizedby autoclaving and cooled to 45 in a water bath.
The other ingredients are warmed to 45 and mixed with the agar. The
complete agar medium can then be distributed with a 5-ml automatic
Cornwall syringe from the vessel in the water bath to culture dishes.
300
[24]
These agar base layers should be allowed to solidify at room temperature

and, if possible, used within the hour. The ingredients for a second flask of
complete agar medium of sufficient volume to provide more than 11 ml per
culture to be assayed should be ready for mixing just prior to the actual
plating of the cells. For this second flask, filter-sterilize the calf serum to
remove any clumps that could provide a growing surface for the normal
cells. Keep all agar medium at 45 until used.
2. Harvest by trypsinization the cells that have been cultured from
each carcinogen dose, and either strongly pipette them several times
through a drawn out Pasteur pipette or force them through a 20-gauge
syringe needle in order to remove clumps of normal cells that can form
spurious colonies in agar. Suspend the cells at a concentration of 106
cells/ml in complete medium made with filter-sterilized calf serum.
3. Assay an aliquot of the suspended cells for liquid cloning ability by
plating 200 cells/per dish in quadruplicate in complete medium with 20%
serum. Count visible colonies which arise after 7 days of incubation at 37.
4. Assay a second aliquot of the suspended cells for ability to clone in
soft agar by mixing 5.5 ml of the suspended cells with 11 ml of complete
agar medium and pipetting the resulting slurry on top of the prepared agar
plates, using 1.5 ml slurry per dish. This results in 10 dishes of 5 x l05
cells per dish with the cells suspended in a final agar concentration of
0.34%. The cell concentration, number of dishes, etc. may be varied, but
the final agar concentration must be exact. Incubate the plates at 37 for
2~-3 weeks in a well-humidified incubator until visible colonies arise. The
transformed cells will form colonies that grow progressively to greater
than 0.2 mm diameter and can be counted by eye or with a bacterial
colony counter without staining.
5. In order to be assured that the clones seen in agar represent stable
transformants, a small sample of them may be picked from the agar plates
with a sterile Pasteur pipette, suspended in DME, pipetted strongly to free
each colony from the surrounding agar, and transferred to dishes with
one-tenth volume of TPB and one-fifth volume of calf serum. The resulting cultures should be cloned again to eliminate any normal cells inadvertently picked up with the transformed colony. After 4-5 passages, the
clones may be retested by assaying them in soft agar, using 10a cells per
dish. Stable transformed lines should clone in agar with an efficiency of
1-40%.
If desired, it is possible to test the in vivo tumorigenicity of these
clones in Syrian hamsters. 1z,15
15M. Stoker, Virology 18, 649 (1962).
[24]
301
Presentation and Interpretation of Data

1. Survival Curve. From the liquid platings done on the day of carcinogen treatment, calculate the surviving fraction and graph its log versus
carcinogen dose. A shoulder in this curve may indicate repair of carcinogen damage; a sharp break in slope suggests that cell killing is going on by
more than one process, sometimes due to toxic carcinogen breakdown
products.
2. Transformant Induction. Calculate from agar plate counts the number of transformants per cell plated, divide by the liquid plating efficiency
(determined at the time the cells were put into agar), and correct by
subtracting the average number of spontaneous transformants estimated
from the zero-dose platings, to give an estimate of the transformation
frequency per survivor. A graph of the log of this value versus dose
should rise rapidly and will usually plateau when the rates of induction of
new transformants and cell killing come into balance.
It is also expected that a graph of the log of the surviving fraction
against the transformants per survivor would, at least at low doses, be
linear.16 Estimate the number of transformants per cell initially treated.
When the log of this value is graphed against dose, the resulting curve
should rise significantly from its starting point, indicating an absolute
increase in the number of transforrnants with increasing dose. The curve
should then fall in parallel with the survival curve, indicating that the
transformants are as sensitive to killing by the carcinogen as the remainder of the population. These calculations assume that during the growth
period between carcinogen treatment and agar assay the transformants do
not enjoy any growth advantage over the normal cells in the population.
This assumption is reasonable if the cultures are not allowed to become
crowded since chemically transformed BHK derivatives do not multiply
faster than their normal parents. 8 The potencies of different carcinogens
can be compared by comparing the transformants per survivor at 37%
survival.
Variations
If the action of a particular carcinogen requires that the cells be actively growing during exposure, they can be treated on plates instead of in
suspension. The problem with treating growing cells is that it provides an
opportunity for the carcinogen to act as a selective agent and enrich the
proportion of preexisting transformants.
~6R. J. Munsonand D. T. Goodhead,Mutat. Res. 42, 145 (1977).
302
[25]
B H K can activate effectively a variety of carcinogens, and its hydrolytic enzymes are present at a good constitutive level. However, if there is
any doubt about the capacity of the cells to activate a given carcinogen, a
liver extract 17 may be added before and/or during the treatment with the
carcinogen. 10
A period of growth is usually required between carcinogen treatment
and challenge with soft agar to attain the maximum expression of transformation by newly induced transformants. Recently, however, a B H K
subclone has been reported 8 that apparently is able to carry out the divisions necessary for maximum expression while suspended in soft agar,
thereby eliminating the growth period in liquid culture. Substituting
Noble agar 6,1 or agarosC a for Difco Bacto agar may also allow one to
skip the growth period in liquid. This is useful for large screening programs, although the number of background colonies is high and maximum
transformation frequencies may not be achieved.
17 B. N. Ames, W. E. Durston, E. Yamasaki, and F. D. Lee, Proc. Natl. Acad. Sci. U.S.A.
70, 2281 (1973).
is I. Macpherson, in "Tissue Culture Methods and Applications" (P. F. Kruse and M. K,
Patterson, eds.) p. 277. Academic Press, New York (1973).
[25] I n d e p e n d e n t
Control of the Local Environment of Somas

and Neurites
By
ROBERT
B.
CAMPENOT
This article describes a newly devised, three-chamber culture system

that has been used for studying the development of sympathetic neurons
dissociated from superior cervical ganglia of newborn rats. The neuronal
somas are placed in a central chamber and send their neurites across
tightly sealed barriers into separate side chambers. The medium in the
three chambers can be independently altered at any time during the
growth of the neurons, so that the action of agents such as nerve growth
factor (NGF) can be tested separately on the somas and proximal portions
of the neurites (central chamber) or on the distal portions of the neurites
(side chambers). Moreover, the neurons readily can be stimulated chronically by passing current across the barriers.
Description of Three-Chamber Culture Dishes
The three chambers are formed by a Teflon divider (Fig. I)joined with
silicone grease to the collagen-coated coverslip floor of a modified 3 5 - m m
METHODS IN ENZYMOL(X}Y,VOL. LVIII

All rights of reproduction in any form reserved,
ISBN 0-12-181958-2
[25]
CONTROL OF THE LOCAL ENVIRONMENT OF SOMAS
303
FIG. 1. A petri dish (35 mm, Falcon, viewed from above) divided into three chambers by
a Teflon divider (7.5 mm high) sealed to the floor of the dish (a collagen-coated coverslip)
with silicone grease. The floor of the narrow (1 x 5 mm) central chamber (a), in which the
neurons are plated, is transected by 20 parallel scratches that serve to guide the growing
neurites into the left (b) and fight (c) side chambers.
petri dish. The grease is applied to the entire face of the divider that is to
mate with the coverslip, and the two are pressed together. Before assembling the system, 20 parallel scratches are made in the collagen-coated
coverslip (Fig. 1); since the growing neurites tend not to cross these
scratches and are thus confined to the collagen-coated channels between
them, the scratches guide the growing neurites into the side chambers (see
below). Just before assemblying the system, a drop of L-15 COs medium
containing Methocel (composition given in this volume [53]) is placed on
the scratched region of the coverslip; this prevents the silicone grease on
the divider from adhering to the coverslip in this region. After applying
the drop of medium, the Teflon divider and coverslip are gently pressed
together so that the scratched, medium-coated region spans the narrow
central chamber and extends into the side chambers to the left and right
(Fig. 1). Although the silicone grease does not adhere to the coverslip in
this region, it prevents significant bulk flow of medium between the central and side chambers; fluid-level differences of about 5 mm between the
304
[25]
central and side chambers are supported for 4 days without appreciable
equalization.
The somas of principal neurons dissociated from superior cervical
g a n g l i a o f n e w b o r n r a t s ( p r o c e d u r e s a r e gPcen in this v o l u m e [53]) a r e t h e n
p l a t e d i n t o t h e n a r r o w c e n t r a l c h a m b e r . T h e n e u r i t e s g r o w p a r a l l e l to t h e
s c r a t c h e s a n d t r a v e r s e t h e b a r r i e r s , p r e s u m a b l y in a t h i n film o f m e d i u m
b e t w e e n t h e g r e a s e a n d t h e c o v e r s l i p . T h e n e u r i t e s e m e r g e i n t o t h e side
c h a m b e r s a f t e r a b o u t 3 o r 4 d a y s in c u l t u r e ( s e e Fig. 2).
Construction a n d U s e of t h e System
Modified Petri Dishes

P l a s t i c p e t r i d i s h e s (35 m m , F a l c o n ) a r e p r o v i d e d w i t h p o l y s t y r e n e
c o v e r s l i p floors in a m a n n e r s i m i l a r t o t h a t p r e v i o u s l y d e s c r i b e d ( s e e this
v o l u m e [53]). B r i e f l y , a h o l e 22 m m in d i a m e t e r is c u t in t h e b o t t o m o f t h e
FIG. 2. Photomicrographs of 25-day-old neurons grown in a three-chamber culture dish.

The upper montage (A) spans the central chamber and shows a cluster of somas (arrow) and
bundles of neurites on one of the collagen channels in the central chamber. The scratches
that border the 200-/~m-wide channel are at the upper and lower edges, and the silicone
grease seals are at the left and right edges of the montage. The lower montage (B) shows the
neurite bundle on the same collagen channel emerging into the right chamber. The grease is at
the left edge of this montage.
[25]
305
petri dish, and the coverslip is bonded over the hole on the outside suface
of the dish with Sylgard 184 encapsulating resin (Dow Coming). A day or
two before plating the neurons, the coverslip is coated with a dried collagen film (see this volume [53]). Twenty parallel scratches about 200/zm
apart are made in the collagen-coated coverslip with a rake made by
cementing together 20 insect pins; a rubber mat under the coverslip helps
to make the depth of the scratches uniform. The dishes are then sterilized
with ultraviolet light.
Teflon Dividers
Teflon dividers are machined from I-in. rod stock. They are 7.5 mm in
height; the central chamber is about 1 mm wide and 5 mm long. The side
chambers hold about 0.5 ml of medium; the central chamber communicates with the rest of the dish that holds 1-2 ml. The slotted shape of the
central chamber protects the neuronal somas from disturbance when the
medium sloshes during handling of the dish.
The first time the dividers are autoclaved, they are distorted and their
faces must be sanded flat with fine sandpaper; subsequent autoclaving
does not distort them further. Before each use, the dividers are wiped
clean of silicone grease and sequentially washed briefly in Nochromix
(Godax Laboratories Incorporated), in 95% ethanol for 24 hr, and in
flowing, deionized water for several hours; they are then autoclaved in
glass-distilled water, rinsed several times in glass-distilled water, and finally placed in a glass petri dish and sterilized by autoclaving. They must
be completely dry before use.
Silicone Grease Prepdration

Silicone grease (Dow Coming high-vacuum grease) is loaded into a
1-ml glass syringe provided with a 22-gauge hypodermic needle cut flat at
the end. The grease and syringe are sterilized by autoclaving. If the grease
is applied to the divider before sterilization and then autoclaved, a skin
forms on the grease and the results are poor.
Assemblying the System

The component parts of the system are assembled under sterile conditions in a hood, with the aid of a dissecting microscope. The Teflon divider
is held by a hemostat with right-angle jaws under the microscope. Silicone
grease is applied to the entire face of the divider. A petri dish, with a drop
of medium on the scratched region of the coverslip, is inverted and placed
306
[25]
gently on top of the greased divider so that the scratched, medium-coated

region crosses the central chamber (see Fig. 1). Final seating of the divider is accomplished by gently pressing the coverslip with a pair of fine
forceps until the silicone grease is seen to adhere to the coverslip
everywhere b u t in the medium-coated region. In the medium-coated region, the grease is pressed firmly against the coverslip but does not adhere.
Then the hemostat is turned over, and the assembled system is released
from its jaws. The side chambers, but not the central chamber (see below), are filled with medium, and the lid is placed on the petri dish.
Plating the Neuron

Dissociated sympathetic neurons from newborn rats are suspended in
L-15 C.O2 medium and loaded into a sterile syringe. The Methocel in the
medium prevents neurons from settling during the plating procedure. Approximately 50 p.1 of cell suspension are injected into the central chamber.
To help confine the suspension to the central chamber, a small d a b of
silicone grease is usually placed on the coverslip at the entrance during
the assembly procedure. The cultures are incubated overnight to allow
the neurons to settle on the coverslip. Medium, 1 - 2 ml, is added to the
main body of the petri dish, ensuring that confluence is established with the
medium in the central chamber. Within 2 days after plating, neurites on
the channels between the scratches in the central chamber have reached
the grease seals to the left and right. After 3-4 days the neurites emerge
from the"seals into the side chambers (see Fig. 2).
Effectiveness of the Seal

Silicone grease seals prepared in the above manner will support a
5-mm difference in fluid levels between the central and side chambers for
at least 4 days without appreciable equalization. However, they have low
electrical resistance, and thus are permeable to ions. Evidently, a thin film
of medium remains between the grease and the coverslips. Doubtless,
molecules diffuse through this film from one chamber to another, and
unequal fluid levels between the chambers must produce at least a slight
bulk flow of medium. Tracers have not yet been used to assess the effectiveness of the seals, but experiments on the local effects of NGF on the
neurons suggest strongly that biologically significant amounts of this protein do not cross the barriers, a (a) It was found that neurites will not enter
a chamber to which no NGF has been added even though the chamber
containing the somas is supplied with a high concentration of NGF. (b)
R. B. Campenot, Proc. Natl. Acad. Sci. U.S.A. 74, 4516-4519 (1977).
[25]
307
NGF withdrawal from a chamber into which the neurites have grown
stops the growth of the neurites; it also appears to result in slow degeneration of the neurites, even though the somas and proximal portions of the
neurites continue to have access to NGF. (c) When NGF is withdrawn
from the chamber containing the somas, the neurons appear to survive
only if their neurites have crossed into chambers where NGF is continuously available. In these experiments, the fluid level in the NGF-free
chambers was always kept at a higher level than that in adjacent chambers
that contained NGF, thereby eliminating transfer of NGF by bulk flow.
Long-Term Electrical Stimulation of the Neurons

Although the barriers do not have high electrical resistance, they restrict current flow enough so that current passed between the chambers
can stimulate the crossing axons to produce action potentials. This was
shown by recording the membrane potential of the somas in the central
chamber with intracellular microelectrodes while passing current between
the chambers with platinum wires immersed in the medium. In virtually
every neuron, action potentials were produced by pulses of 1 msec duration and amplitudes of 1-5 V; in most cases, the threshold was slightly less
than 1 V. The stimulation was apparently effective on a long-term basis
without any damage to the neurons. One culture was stimulated for 3 days
at 1/s (pulse amplitude of 4 V). After the 3 days, the threshold voltage for
action potentials was about 1 V. In another culture, stimulation at 4 V and
1/s was begun the first day after the neurons were plated, i.e., before the
neurites had crossed ttie barriers; the neurites crossed the barriers while
the stimulation was in progress, indic~ttitig that the stimulation was not
harmful. Using this method of stimulation, it has been shown that electrical activity during development can determine whether sympathetic
neurons develop cholinergic or adrenergic transmitter function. 2
Acknowledgments
This work was performed in the laboratoryof Edwin J. Furshpan and David D. Potter.
Support was provided by National Institutesof Health Research Grants NS-02253, NS03273, NS-11576, and Training Grants NS-07009 and NS-07112.
2 p. A. Walicke, R. B. Campenot, and P. H. Patterson, Proc. Natl. Acad. Sci. U.S.A. 74,
5767-5771 (1977).
308
[26]
[26] M u t a n t I s o l a t i o n
B y LARRY H. THOMPSON
During the last decade somatic cell genetics has developed into a
recognized discipline--largely because of the development of techniques
for somatic cell hybridization and methods for the isolation of a variety of
mutant phenotypes. Already the spectrum of mutant cell lines available is
sufficient to have a substantial effect on approaches to studying metabolic
regulation and growth control in animal cells. In this article the term
" m u t a n t " will be used to refer to heritable phenotypic changes that appear to arise from changes in DNA structure, as evidenced by stability,
response to mutagens, the presence of altered gene product, and other
criteria. Despite some earlier concerns, an epigenetic basis for variant
phenotypes has received little documentation and does not appear to be a
major complicating factor in the isolation and use of mutations. Most
mutant lines that have been isolated behave in a reproducible way and are
suitably stable in phenotype, provided a minimal number of technical
precautions are taken.
The diploid nature of somatic cells undoubtedly remains a major barrier to isolating mutants. Since the great majority of mutations are recessive when tested in somatic cell hybrids, such mutations will only be
expressed in autosomal genes when both alleles are altered. There is some
evidence suggesting that certain established cell lines, particularly the
well-studied Chinese hamster ovary (CHO) lines, may contain significant
hetero- or hemizygosity. 1'2 The precise extent of this functional haploidy,
however, is still unclear since other evidence indicates that the cells behave essentially as diploid. 3 In general, when planning a mutant isolation,
one should assume that the gene coding for the protein (or RNA) molecule
of interest is probably present in two copies. This constraint does not
imply that mutants for such loci cannot be isolated, but it does suggest
that a m u c h greater effort will be required than in the case of an X-linked
function such as the commonly studied hypoxanthine phosphoribosyltransferase (HPRT).
Because many mutants are found at very low frequency, even after
potent mutagenesis, the probability of a successful isolation may be directly proportional to the magnitude of the experiment, i.e., to the number
1L. Siminovitch,Cell 7, 1 (1976).
L. Siminovitchand L. H. Thompson,J. Cell. Physiol. 95, 361 (1978).
3 M. J. Siciliano, J. Siciliano, and R. M. Humphrey,Proc. Natl. Acad. Sci. U.S.A. 75,
1919 (1978).
ISBN 0-12-181958-2
[26]
MUTANT ISOLATION
309
of cells screened. Herein lies a major difficulty. At best, one can handle
about l0 s cells per liter of medium under selective conditions; in
monolayer culture this number of cells would typically be distributed
among 50 or more 100 mm-petri dishes. These plating requirements set a
practical lower limit on detectable mutant frequencies at = 10-a, i.e., one
mutant from 500-1000 dishes (5-10 liters of medium). Thus, the isolation
of mutations, especially of the temperature-sensitive conditional class, at
loci that require simultaneous alteration in both alleles will often be very
difficult. However, for certain "recessive" mutations it is possible to
obtain the homozygous ( - / - ) state by a two-step process because the
heterozygous ( + / - ) state is selectable on the basis of a slight difference in
phenotype (such as degree of drug resistance, discussed subsequently for
the aprt locus) from the wild type. Only a few mutations having dominant
(or incompletely dominant expression) have been identified; these include
resistance to a-amanitin, ouabain, Methotrexate, ricin, and colchicine.
Choice of Cell Line and Culture Requirements
Chinese Hamster Ovary Cells as a Model System
The biochemist who wants to select mutants and is not committed to a

particular cell system should choose a line that combines the most favorable genetic properties with ease of handling. Chinese hamster cells are
commonly used because of their favorable karyotypic properties. The
Chinese hamster ovary (CHO) line has been one of the most popular,
partly because of its early introduction as a system for isolating auxotrophic mutations. 4 To date, more than 50 different recessive mutations
have been isolated in CHO cells, including purine and pyrimidine auxotrophy, 5,e lectin resistance, 7 and temperature-sensitive conditional mutations for aminoacyl-tRNA synthetases. 8 This variety of mutations has
established the genetic utility of CHO strains and provides substantial
background information for attempts to isolate new mutant classes.
Although extensive chromosomal ,rearrangement is evident in the
near-diploid CHO cell line, the karyotype appears to be reasonably stable
from cell to cell within a clone, and upon repeated subcloning, a This
stability should minimize the opportunity for phenotypic variation poten4 F.-T. Kao and T. T. Puck, Proc. Natl. Acad. Sci. U.S.A. 60, 1275 (1968).
5 D. Patterson, Somat. Cell Genet. 2, 189 (1976).
D. Patterson and D. V. Carnright, Somat. Cell Genet. 3, 483 (1977).
r p. Stanley and L. Siminovitch, Somat. Cell Genet. 3, 391 (1977).
8 G. M. Adair, L. H. Thompson, and P. A. Lindl, Sornat. Cell Genet. 4, 27 (1978).
9 R. G. Worton, C. C. Ho, and C. Duff, Somat. Cell Genet. 3, 27 (1977).
310
[26]
tially associated with chromosomal change. Mutant lines that have been
carefully examined were found to resemble closely the wild type. 9
Another attractive feature of CHO cells is that they are not unreasonably fastidious in terms of culture requirements, and they can be
grown under a variety of conditions. A high plating efficiency, an important property for quantitative mutant isolation, is routinely achievable.
Unlike most other hamster lines, CHO cells can be grown in liquid suspension culture, in soft agar, on top of soft agar, 1 as well as in monolayer.
In ot-MEM medium lacking nucleosides (K. C. Biological, Inc., Cat. No.
DM-325) but containing 10% fetal bovine serum CHO cells have a doubling time of about 12 hr both in monolayer and suspension culture.
Not surprisingly, a number of different wild-type CHO strains exist,
having different karyotypes, growth rates, morphology, ease of growth in
suspension, and other properties. This variety should be kept in mind
when choosing a strain for mutant isolation. There is as yet no evidence
that any particular strain is generally more mutable than another, except
in some specific instances where certain clones behave as though they are
hemizygous at a specific autosomal locus that was not intentionally sub:
jected to selection pressure. 2
Suspension vs. Monolayer Culture
Since suspension culture methodology is less common than monolayer

culture, it is worth emphasizing here some of the major advantages of
liquid suspension as it relates to mutant isolation: (1) In suspension, cells
such as CHO are grown either in roller tubes that revolve on a drum, or in
spinner flasks in which the cells are kept suspended by stirring bars. Since
the cells exist as single cells, culture growth rate can be monitored easily
by sampling aliquots. This means that cultures for experimentation can be
readily standardized with reference to cell concentration. (2) The production of bulk quantities of cells for mutant characterization is more readily
achievable because large spinner cultures are commercially available. (3)
Precise temperature control is much easier in water baths than in forcedair incubators; the selection and characterization of temperature-sensitive
mutants depend on good control, especially for restrictive temperatures.
Potential Problems with Media and Sera
Since the difficulty of quality control of serum-supplemented culture

medium is probably the greatest limitation in doing quantitative cell cull0 M. W. Konrad, B. Stori, D. ,~. Glaser, and L. H. Thompson,Cell 10, 305 (1977).
[26]
MUTANT ISOLATION
311
ture, a few c o m m e n t s are in order. N o t only should the s e r u m supplement

be pretested, but it should be tested after dialysis if the use of dialyzed
s e r u m is anticipated. Dialysis often impairs the growth-supporting activity of serum. It is not u n c o m m o n for mutant lines (auxotrophic,
temperature-sensitive, etc.) to grow m o r e poorly than wild type in certain
s e r u m lots. Thus, it is important to select lots of sera that provide vigorous growth of wild-type cells.
Strategy and Timing for Mutant Isolation

In t e r m s of typical biochemical experiments, cell culture and mutant
isolation p r o c e d u r e s m a y s e e m slow and tedious. Because of the length of
time required to obtain results, the i m p o r t a n c e of good incubators, rigorous sterile technique, and high-quality culture m e d i u m should not be underestimated. The essential steps for isolation of mutants are given in
Table I along with an indication of the a p p r o x i m a t e m i n i m u m time required for each step. It is advisable to freeze cultures at intermediate
stages, such as before recloning, to guard against loss f r o m contamination
or other p r o b l e m s . Although cells can be stored temporarily at - 8 0 ,
liquid nitrogen t e m p e r a t u r e is required for maintaining viability for a period o f years.
TABLE I
GENERALIZEDOUTLINEFOR MUTANTISOLATION
Steps in the isolation of a mutant

1. Treatment of culture with mutagen (mutagenesis)
2. Growth of the cells for fixation and expression of mutations
3. Treatment of culture with selective agent and outgrowth of presumptive mutant colonies
4. Growth of isolated colonies to mass cultures (2 x 107 cells)
5. Subculture and growth of clones for biochemical assay and other
tests for drug sensitivity, mycoplasma, etc.
6. Recloning of good mutants to ensure genetic homogeneity
7. Freezing of multiple aliquots of freshly cloned mutants, preferably in liquid nitrogen
Total time
a A s s u m i n g an average generation time o f 18 hr.
Approximate
number of cell
generations
1
6
12
12
2
24
57 generations
(43 days a)
312
[26]
Induction of Mutants
Should a Mutagen Be Used?

A mutagen is not recommended unless it is necessary in order to
obtain the desired phenotype from a reasonable number of cells. Potential
complications with the mutagen may occur from DNA damage at many
other sites than the target gene; such secondary mutations may affect the
growth rate or other properties of the cells, thereby interfering with the
characterization of the mutant phenotype. When a mutagen is not used, it
is important to remember that the frequency of a particular mutation will
be higher, on the average, in cultures that have not been recently cloned
or diluted heavily. 11However, in most instances a mutagen will be necessary to elevate the mutant frequency to a detectable level.
Choice of Mutagen and Conditions of Exposure

Commonly used mutagens for mammalian cells include MNNG (Nmethyl-N'-nitro-N-nitrosoguanidine), MMS (methyl methanesulfonate),
EMS (ethyl methanesulfonate), ENU (ethyl nitrosourea), and ultraviolet
(UV) light. EMS has been the most popular because it is easy to use and
has proved reliable for enhancing the mutation frequency at a large number of loci, producing deficiency mutations in which enzyme activity is
lost (such as 6-thioguanine resistance or numerous auxotrophies) as well
as phenotypes that are presumed to involve primarily missense mutations
(temperature sensitivity, or reduced affinity of a protein for an inhibitor
such as a-amanitin or Methotrexate).
All chemical mutagens should be handled carefully since most are
known to be carcinogenic. The following standard procedure for EMS
mutagenesis of CHO cells is recommended as an example:
1. An exponentially growing suspension culture is collected by centrifugation and resuspended in fresh medium at a concentration of
2-3 x 105 cells/ml.
2. EMS (Sigma Chemical Co.) (density = 1.2 g/ml), diluted to a concentration of 1% in water or serum-free medium, is added to the culture to
give a final concentration in the range of 200-300 ~g/ml. The exact concentration depends on the particular clone of CHO and the composition of
the medium, such as the presence of certain ribonucleosides in the medium, which may enhance the sensitivity to mutagen. For an optimal frequency of mutant induction without excessive damage to the cells, a
surviving fraction after exposure of 0.1 or higher should be used.
11 L. H. Thompson and R. M. Baker, Methods Cell Biol. 7, 209 (1973).
[26]
MUTANT ISOLATION
313
3. After incubation for 16 hr the mutagen is removed. This is accomplished by centrifugation and resuspension in buffered saline to rinse the
cells, and then suspension in fresh medium at a reduced cell concentration
to allow for growth of the culture.
Requirement for Mutation Expression

A period of recovery from the mutagen is necessary to allow fixation
of the mutations and outgrowth of the surviving cells. Full expression of a
mutant phenotype generally requires dilution of preexisting gene product
by successive cell divisions. The number of divisions necessary for maximal expression of new mutations varies considerably with the phenotype
being selected. However, in most cases a 50- to 100-fold increase in cell
number (5-7 doublings) should be adequate. An excessively long interval
will result in the selective loss of mutants if their growth rate is less than
that of wild type.
During this mutation expression interval it is very important to maintain a culture of adequate size by low ratio dilution during subculture so
that mutants are not lost. Thus, when the survival is low the size of the
culture should be expanded at the first subculture so that mutants have an
opportunity to multiply. For the actual selection, some fraction of the
total number of cells can then be screened. If the selection efficiency for
mutants is expected to be low, e.g., 10% as in [3H]amino acid suicide
discussed in a subsequent section, it is desirable to screen a proportionately larger number of progeny cells than the number of survivors
after mutagenesis treatment since a large fraction of the mutants will be
killed by the selection.
Independent Cultures
It is often important to know that individual mutants represent independent mutational events and not siblings. In order to achieve this condition, the experiment should be designed so that separate cultures are
mutagenized or that a large culture is subdivided immediately following
mutagen treatment before there has been opportunity for mutations to
become replicated as siblings within the population.
Direct Selection of Drug-Resistance Mutants
As a general class of mutants, drug-resistance mutations are among the
easiest to isolate because the selective situation favors the direct outgrowth of the mutant clones. This approach can be used in any situation in
314
[26]
which a toxic compound can be directed at a specific target molecule,

including cases in which the mechanism of toxicity is unknown. Examples
of these mutants that have been isolated in CHO cells include resistance
to lectins, Methotrexate, colchicine, t~-amanitin, histidinol, emetine,
8-azaadenine (or 2,6-diaminopurine), ouabaln, hydroxyurea, cycloheximide, 5-bromodeoxyuridine, 6-thioguanine (or 8-azaguanine), azetidinecarboxylic acid (proline analog), diphtheria toxin, and excess
adenosine.
In all of these cases mutants were isolated as clones growing in
monolayer under plating conditions. The isolation of resistant cells by
continuous exposure and subculturing in the presence of the drug is not
recommended because of the difficulty of interpreting the number of genetic changes associated with a phenotype. Since many laboratories will
be interested in isolating new classes of mutants, the present discussion
will focus on the principles and problems involved in drug selections
rather than the use of particular drugs.
Choice of Drug Concentration

Because a discussion of this point has been given previously, 11 only
the most important aspects will be reiterated here. As indicated above,
drug selections are accomplished in monolayer in order to ascertain the
individual mutant clones by visual examination and microscopy. A drug
concentration is chosen at which wild-type cells are satisfactorily killed
while the mutant cells of interest survive with high efficiency. The use of
too high a concentration may well result in a failure to obtain any mutant
colonies. This can be illustrated nicely with the results that are obtained in
selections carried out for 8-azaadenine resistance, which is normally associated with loss of activity of the autosomal enzyme, adenine phosphoribosyl transferase. A mutant cell in which one of the two alleles is
altered to produce defective enzyme is usually expected to have ->50% of
wild-type transferase activity; such a heterozygous mutant remains drugsensitive. However, the mutant will be slightly more resistant than wildtype cells--probably, at most, 2-fold in terms of drug concentration. On
this basis the heterozygous mutant can be discriminated from the wild
type if the drug dose is chosen carefully. At a surviving fraction of = 10-5
in the drug (=8/xg/ml in dialyzed fetal calf serum), a sizeable percentage
of the large colonies will actually possess increased resistance, and will
thus be presumptive heterozygotes with respect to the enzyme.lla Under
these conditions, the choice of drug dose is influenced by the number of
alleles one expects to be associated with the gene locus under considera11a G. E. Jones and P. A. Sargent Cell 2, 43 (1974).
[26]
MUTANT ISOLATION
315
tion. In this example o f selecting a presumptive heterozygote, it is essential that the drug concentration be carefully adjusted for the desired
survival value. It is also necessary to demonstrate explicitly an increased
drug resistance and reduced e n z y m e activity for the presumptive mutant
colonies after they have been isolated since some colonies may not be
mutant for statistical reasons, 1~ and resistance may involve other loci.
Often, more than one gene locus is involved in the mutation to resistant phenotypes, as in the case of Methotrexate, TM diptheria toxin, 13 or
colchicine; TM with each agent, isolation of mutations in the intracellular
target protein is complicated by the presence of permeability variants.
Since different genetic classes of resistance may occur at different drug
concentrations, the drug dose must be chosen accordingly.
Details o f Plate Inoculation and Incubation
Standard procedure involves inoculating lO0-mm plastic petri dishes

with cells in drug medium and incubating until macroscopic clones are
isolatable. The maximum n u m b e r of cells per dish that can be plated must
be arrived at empirically since it depends on such factors as the cell type,
rate of cell lysis, and extent of metabolic interaction between mutant cells
in contact with wild type. The phenomenon of contact feeding (metabolic
cooperation) can greatly suppress the r e c o v e r y of mutant colonies when
the cells are too dense; this has been a c o m m o n complication with purine
analogue resistance selections. 15 In general, the cell inoculum for a
100-mm dish with 20 ml of culture medium will be in the range of 1 105
to 1 l06 cells. After the dishes are filled with drug medium, the cells are
introduced in a small volume, e.g., 0.1-1 ml of a concentrated suspension.
It is important to distribute the cells uniformly on the dish. This is accomplished by gently sliding the dishes back and forth in one direction, then
applying the same motion in the orthogonal direction. This manipulation
should be done very soon after pipetting the cells onto the dish because
they can attach within minutes, even at room temperature. Caution must
be exercised since medium spilled to the outside of the dish is an invitation for contamination.
Especially with CHO cells, it is important not to move the dishes
during the colony formation interval because of the severe problem of
satellite colony formation. If a dish containing colonies is moved and then
returned to the incubator, one typically gets many migrating cells that
12w. F. Flintoff, S. V. Davidson, and L. Siminovitch, Somat. Cell Genet. 2, 245 (1976).
13T. J. Moehring and J. M. Moehring, Cell 11, 447 (1977).
14j. E. Aubin, A. Chase, F. Sarangi, and V. Ling,J. CellBiol. 72, No. 2, Part 2,298a (1977).
15C. M. Corsaro and B. R. Migeon, Exp. Cell Res. 95, 39 (1975).
316
[26]
were loosely attached or floating over the colony but then reattached
elsewhere on the dish. The dishes should be moved only if it is absolutely
necessary to replenish the medium because of drug instability or other
problem. The presence of large numbers of dead cells on the dish is not
esthetically pleasing but generally does not interfere with the selective
conditions and can be ignored.
Mutant colonies that are to be isolated for characterization should be
taken while fairly small, approximately 1 mm in diameter, to minimize the
risk of cross-contamination associated with large colonies on the dish. At
a normal growth rate, CHO cells produce this size colony in about 6-7
days, but mutants under drug selection may require incubation of 2-3
weeks. In any mutant selection, control dishes with cells in drug-free
medium should be set up with 200-300 cells per dish in order to determine
the plating efficiency. This control is important for calculating the mutant
frequency, and it gives an indication that the incubation conditions were
valid. Control dishes are usually stained sooner than selection dishes and
should, therefore, be placed on separate incubator trays to avoid the
problem of colony dispersal mentioned above. Procedures for isolating
the colonies are discussed in the final section.
Multistep Selections
When either more than one gene or more than one allele can be altered
to effect resistance, it may be desirable to produce highly resistant cells
by performing additional selections at increasing drug concentration on
clonal isolates. For example, it might be necessary to obtain a permeability mutant initially, before selecting for an alteration in an intracellular
molecule. Mutagenesis can be carded out as needed at each round of
selection.
Indirect Selection of Conditionally Lethal or Auxotrophic Mutations

Auxotrophic mutants and conditionally lethal (mainly temperaturesensitive, ts) mutants must be isolated either by nonselective procedures
such as replica plating, or by schemes that preferentially kill the wild-type
cells during temporary exposure of the population to selective conditions
that allow for reversible arrest of the mutants. Most of the selective
agents that have been used are ones that act during the S phase of the
cycle; included are cytosine arabinoside, 5-fluorodeoxyuridine, tritiated
thymidine (high specific activity), and 5-bromodeoxyuridine (BrdUrd) followed by exposure to visible light. The last method, developed by Puck
[26]
MUTANT ISOLATION
317
and c o - w o r k e r s , 16 has been very successful for the isolation of many

different auxotrophic mutants defective for steps in the synthesis of
purines and pyrimidines, cholesterol, unsaturated fatty acids, amino
acids, and carbohydrate metabolism. The BrdUrd/light selection methodology has been described in detail 17 and will not be reviewed here. However, it should be noted that a recent modification of the procedure greatly
reduces the time required for light exposure and may also improve the
selection efficiency.TM The DNA-binding dye, 33258 Hrechst, added to
cultures 2-3 hr before light exposure, greatly sensitizes the BrdUrdsubstituted DNA to photolysis. As a general class of mutants, auxotrophic
mutants have been particularly useful because they are often readily
identifiable.
Nearly all of the methods for selecting mutants that are temperature
sensitive (ts) for growth are similar in principle tO the auxotrophic selections, but the mutants obtained usually prove to be extremely difficult to
characterize in terms of the molecular defect. Generally speaking, mutants that cannot be identified biochemically are of somewhat limited usefulness. For this reason, procedures for the isolation ofnonspecific ts mutants will not be detailed here. The reader can refer to previous discussions
that describe the use of tritiated thymidine 11and 5-fluorodeoxyuridineTM as
selecting agents for such mutants.
Selection of Protein Synthesis Mutants with [3H]Amino Acids

The selection procedures for mutants with defects in protein synthesis
are the only methods that have been developed that recover specifically ts
mutants for a particular component of metabolism. Two kinds of procedures have been used: (1) killing of wild-type cells with tritium suicide
from [3H]amino acid incorporation, z and (2) killing by the incorporation
of cytotoxic amino acid analogs. 21 The first approach will be described in
detail here since it has been used extensively on CHO cells to generate a
large number of mutants, s Although the procedures were designed with
the intention of selecting a broad spectrum of protein synthesis mutations,
essentially all of the mutants obtained to date from such selections l~ave
been identified as aminoacyl-tRNA synthetase mutants, z2 Despite the var1~T. T. Puck and F.-T. Kao, Proc. Natl. Acad. Sci. U.S.A. 58, 1227 (1967).
17 F.-T. Kao and T. T. Puck, Methods Cell Biol. 8, 23 (1974).
18 G. Stetten, S. A. Latt, and R. L. Davidson, Somat. Cell Genet. 2, 285 (1976).
19 C. Basilico and H. K. Meiss, Methods Cell Biol. 8, 1 (1974).
20 L. H. Thompson, C. P. Stanners, and L. Siminovitch, Somat. Cell Genet. 1, 187 (1975).
21 j. j. Wasmuth and C. T. Caskey, Cell 9, 655 (1976).
22 L. H. Thompson, D. J. Lofgren, and G. M. Adair, Cell 11, 157 (1977).
318
[26]
ious degrees of temperature sensitivity displayed by these mutants, nearly

all have an altered dependency for the corresponding amino acid. Thus,
the recovery of particular classes of synthetase mutants can be favored or
discouraged by changing the levels of the appropriate amino acids during
the selection. In the procedure described below, temperature sensitivity is
treated as the primary phenotypic property, and emphasis is placed on
testing for altered amino acid responses to identify the mutants.
1. CHO suspension cultures at 34 in exponential growth at a concentration of 3 105 cells/ml, mutagenized and allowed to express as described in a previous section, are used to initiate the selection. A minimum of 1 108 cells should be used, but several times this number may
be required to obtain several different classes of mutants.
2. The cells are harvested by centrifugation, rinsed thoroughly with a
large volume of phosphate-buffered saline, and resuspended at a concentration of 3 x 106 cells/ml in selection medium. This medium must be
lacking in the amino acid that is to be added as the selecting agent (see
below); therefore, dialyzed fetal bovine serum is used, at the regular concentration of 10%. The other 19 amino acids are normally reduced to 10%
of their standard concentrations ( 0.1) since mutant frequencies appear
to be lower when normal amino acid levels are used. When a particularly
common class of synthetase mutant, e.g., asparagyl-tRNA synthetase, is
to be excluded, its selection efficiency is specifically reduced by raising
the concentration of the cognate amino acid to 1 in the selection
medium. (See Adair et al. 8 for an example of this effect.)
3. The culture in selection medium is placed at the restrictive temperature of 39.5 (___0.2 control or better) for 4 hr.
4. The [all]amino acid used for selecting ([aH]valine, [3H]lysine, or
others) at a specific activity of ~18 Ci/mmol (available from AmershamSearle in sterile aqueous solution at 1 mCi/ml concentration) is added
directly to the culture at a final concentration of 20/zCi/ml and incubation
at 39.5 continued.
5. After 4 hr incubation with the [SH]amino acid, the culture is centrifuged and resuspended at 1 108 ceUs/ml in regular medium and incubated at the permissive temperature of 34 for 1 hr to promote recovery of
protein synthesis in mutant cells and to remove the isotope from the
intracellular pool.
6. The culture is placed on a roller wheel .(Model TC7, New
Brunswick Scientific) at 2-4 for 4 days to allow the accumulation of
radiation damage sufficient to kill the wild type to a surviving fraction of
- 1 10-r. The nonspecific killing from cold storage for mutant or wild
type is minimal, i.e., less than 20%. 8
7. The cells are plated into 100-mm petri dishes containing 30 ml of
[26]
MUTANT ISOLATION
319
medium by pipetting directly 1.0 ml of cell suspension into each dish and
dispersing it.
8. The dishes are incubated at 34 for 16-18 days before isolating the
colonies. (See section below on clone isolation.) Other versions of the
above procedure can presumably be designed to obtain a specific synthetase mutation by conducting the experiment at a single temperature and
altering a particular amino acid concentration to a low level in the selective medium, and to a high level during mutagenesis expression and colony formation. Such a procedure should select for specific amino acid
hyperauxotrophs--mutants that require an abnormally high concentration
of an amino acid for growth. These will likely prove to be temperature
sensitive also in many cases.
Screening for Mutant Phenotypes

The testing of the colonies is generally done in two stages. The first test
identifies those colonies that are ts mutants, and it may also indicate
whether there is an altered amino acid dependency. Each colony is picked
and transferred into a tube containing 6 ml of regular ot-MEM medium,
pipetted several times for dispersal, and inoculated as 2-ml aliquots into
three replicate wells (CoStar 24-well cluster dish 2534) in separate trays.
All replicates are first incubated at 34 for 4-5 days to allow for cell
multiplication, but to a density substantially subconfluent. One tray remains as the control at 34 in standard medium while the other two are
placed at 39.5 . One of the latter is kept in regular medium and the other
changed to medium in which all 20 amino acids are reduced to x 0.1; some
mutants die only in medium with reduced amino acids. The responses in
x 1 and in x 0.1 amino acids at 39.5 are closely compared for an indication
as to whether a mutant is ts, hyperauxotrophic, or both. Cultures under
test should be monitored daily for the first 3 days by microscopic examination for cessation of cell division and lysis. Complete cell lysis by 24 hr
is characteristic of strong protein synthesis mutants. Wells at 39.5 that
exhibit growth inhibition or lysis are compared to the control well at 34
before terminating the test at about day 7. It is important to monitor the
rate of lysis carefully in the two media at 39.5 since an early difference
may reflect an amino acid dependency that will not be seen subsequently if
lysis has occurred in both media. Typically, about 20% of the colonies are
mutants under these conditions.
More extensive amino acid tests on clones identified as mutants usually suggest which aminoacyl-tRNA synthetase is defective. This identification is based on either (1) a protective effect from an excess of a particular amino acid at 39.5 (or growth stimulation at 34), or (2) growth
320
[26]
inhibition/killing resulting from reduced amino acid at either 34 or 39.5 ,

depending on the mutant. Some mutants die only under the doubly restrictive conditions of 39.5 and reduced amino acid. These tests are done by
raising or lowering individual amino acids by a factor of 10 compared to
the standard a-MEM concentrations. The tests are also performed in the
24-well cluster trays by inoculating with 20 different media containing
elevated (or reduced) amino acids. The remaining four wells are used for
control medium with normal or 10-fold uniformly reduced amino acids. To
each well, 5 103 cells are added in a small volume of medium or saline.
Thus, for a given mutant as many as four trays are set up including both
10 and 0.1 amino acids at each temperature. The wells are stained
after about 7 days incubation at 39.5 and 8 days at 34. An abnormal
response to a particular amino acid is evident from the staining intensity.
The preparation of media with reduced amino acids is greatly facilitated by reconstituting an amino acid-free basal medium with concentrated, sterile stock solutions of the 20 amino acids. Eighteen of the amino
acids can be prepared as 400 stock solutions; these are stored at room
temperature or 2, except for asparagine and cysteine which are frozen.
Glutamine is stored as a frozen stock solution at 100 and, for convenience, tyrosine is treated as a " 400" liquid stock although it is insoluble
and forms a fine suspension. The 400 stocks can be combined further,
three or four at a time, to 100 stocks.
Replica Plating Procedures
The lack of simple, rapid methods for replica plating mammalian cell
colonies has been a major limitation in isolating auxotrophic,
temperature-sensitive, DNA repair, and other classes of mutations that
may be difficult to obtain by selective methods that enrich for nongrowing
cells. For cell lines that grow in suspension, some attempts have been
made to replicate colonies o n soft agar using velveteen cloth in a manner
similar to that used so successfully with microorganisms on agar. However, the only method 23 introduced to date that appears to have potential
is one that involves making a replicate of colonies attached to a petri dish
by using a fine-mesh nylon cloth that is laid down on the colonies for a
period of 3--4 hr. Cells from each colony attach to the cloth, which is then
transferred to a second dish for further incubation. It appears that few
laboratories have seriously evaluated this method; probably conditions
will have to be optimized for each cell line used. Our laboratory has just
begun to utilize this approach; it appears that for CHO cells a polyester
cloth (Pecap HD7-10 Super, Tetko Inc.) may work better than nylon
23 T. D. Stamato and L. K. Hohman, Cytogenet. Cell Genet. 15, 372 (1975).
[26]
MUTANT ISOLATION
321
because it is more rigid and less water absorbing. With this procedure it
seems possible to screen the colonies from 104-105 mutagenized cells
when as many as 500-1000 colonies per dish are replicated. While it is not
clear that more than one useful replicate can be made, even one would
facilitate the isolation of mutants such as the UV-sensitive clone of
CHO-K1 cells that has been reported. 24
Mutant Colony Isolation and Cloning
Isolating Colonies from Petris Dishes
As indicated earlier, when colonies are being taken from petri dishes
as presumptive or prospective mutants, it is preferable to do this when
they are about 1 mm in diameter; larger colonies are more prone to shed
loose cells. Our standard procedure involves removing the colonies with a
Pipetman autopipette, using the 0.2 ml plastic disposable tips that are
autoclavable. The colonies should be rinsed twice with serum-free medium to r e m o v e loose cells as soon as they are taken from the incubator,
and then circled with a marking pen. Each colony is removed simply by
scraping the circled area with the pipette tip while applying simultaneous
suction by slow release of the plunger. This effectively removes the cells
with little damage. A dish should be rinsed once or twice again before
each additional colony is taken to minimize the chance of crosscontamination. The purity of colonies isolated in this manner is sufficiently high for effective mutant screening when done with the precautions indicated. The method is very simple and amenable to the isolation
of large numbers of colonies using standard laboratory equipment. Another, perhaps more rigorous, method involves isolating each colony with
a stainless-steel cylinder by using autoclaved vacuum grease, and then
trypsinizing the cells.17,25
Cloning in Multiwell Plastic Trays
To ensure genetic homogeneity, it is desirable to clone rigorously cultures that have been identified as mutant. This is done conveniently with
96-well cluster trays having good optical quality, e.g., the CoStar ( # 3 5 % )
or N u n c (#N-1480) trays. Wells are inoculated with 0.2 ml of a cell suspension in medium such that about one well in every 3 to 4 receives a
viable cell. The low concentration minimizes the probability of getting
two cells in one well. When the colonies are recognizable with an inverted
24T. D. Stamato and C. A. Waldren, Somat. Cell Genet. 3, 431 (1977).
~5T. T. Puck, P. I. Marcus, and S. L. Cieciura, J. Exp. Med. 103, 653 (1956).
322
[?7]
m i c r o s c o p e , they should be e x a m i n e d and m a r k e d so that double colonies

that might b e c o m e overlapping are excluded. When the colonies h a v e
r e a c h e d m a c r o s c o p i c size, they are r e m o v e d b y trypsinization and transferred to a larger culture vessel.
Acknowledgment
This work was performed under the auspices of the U.S. Department of Energy, Contract No. W-7405-ENG-48.
Notice
This report was prepared as an account of work sponsored by the United States Government. Neither the United States nor the United States Energy Research & Development
Administration, nor any of their employees, nor any of their contractors, subcontractors, or
their employees, makes any warranty, express or implied, or assumes any legal liability or
responsibility for the accuracy, completeness or usefulness of any information, apparatus,
product or process disclosed, or represents that its use would not infringe privately-owned
rights.
[27] Karyotyping
By
RONALD G. WORTON and CATHERINE DUFF
Introduction
K a r y o t y p i n g should not be an e n d e a v o r restricted to the cytogeneticist. Rather, it is an e x t r e m e l y valuable r e s e a r c h tool for anyone who
works with cell cultures, regardless o f the p r i m a r y field of interest.
M o s t biologists k n o w that the g e n o m e o f higher plants and animals is
organized into discrete c h r o m o s o m e s , that the c h r o m o s o m e s can be set
out in groups to f o r m a k a r y o t y p e , and that the k a r y o t y p e is characteristic
o f the species. M a n y will be less familiar with the fact that in p e r m a n e n t l y
established cell lines there can be extensive k a r y o t y p i c variation f r o m line
to line derived f r o m the s a m e species, or f r o m cell to cell within a line.
The biochemist, therefore, should b e aware of this genetic heterogeneity
in order to properly interpret cell culture data, and m a y on occasion use
this heterogeneity to a d v a n t a g e in designing n e w experiments.
When tissues are explanted and put in culture, the cells that grow
generally display the k a r y o t y p e of the individual as, for example, in direct
bone m a r r o w cultures, short-term l y m p h o c y t e cultures, or p r i m a r y fibroblast cultures. H o w e v e r , w h e n cell cultures b e c o m e established the rigid
control o v e r the k a r y o t y p e is generally lost, and c h r o m o s o m a l variation

ISBN 0-12-181958-2
[9. 7]
KARYOTYPING
323
becomes commonplace. Because of the variation in chromosome number

one usually characterizes an established line by its modal number rather
than its absolute number of chromosomes. In general, when the modal
number is close to the normal diploid number the variation about the
mode is less than when the modal number is greatly elevated. The former
are often termed pseudodiploid or quasidiploid cell lines, whereas the
latter are often referred to as heteroploid cell lines.
In addition to numerical variation, many cell lines undergo extensive
chromosomal rearrangement (deletion, translocation, inversion, etc.) resuiting in new " m a r k e r " chromosomes. While the details of such structural variation in the khryotype will be of greatest interest to the somatic
cell geneticist, the concept of karyotypic evolution ought to be of some
concern to all who work with cell cultures as model systems. As seen
below marker chromosomes can be very useful in research with cultured
cells.
Uses of Karyotyping
1. Gene Dosage. In many types of biochemical research, including, for
example, studies on regulation of enzyme synthesis, it may be desirable to
know whether the cell line is diploid, pseudodiploid, or heteroploid.
Clearly, such studies will be more valid when carried out with diploid
cells rather than pseudodiploid or heteroploid cells. Furthermore, information is rapidly accumulating on the human chromosomal map location
for many enzymes so that in the future the biochemist may want to know
the details of which chromosomes or chromosomal parts are duplicated or
deleted in order to estimate the role that gene dosage may play in the
regulatory phenomenon under investigation.
2. Identification of Cell Lines. Marker chromosomes generated spontaneously by structural rearrangement in cell cultures are usually characteristic of the line, and three or four such unique markers can provide an
unequivocal identification test for a cell line. That such an identification
test is important was realized a few years ago when it was found that a
number of presumably different cell lines, thought to have been derived
from a variety of different human tissues, were not different lines at all,
but were in fact sublines of the heteroploid HeLa cell line derived many
years ago from a human cervical carcinoma. Much of the proof for this
startling discovery consisted of the finding of several marker chromosomes, characteristic of HeLa cells, in the other cell lines. 1Contamination
of one cell line by another is a potentially enormous problem for the
investigator who studies a tissue-specific enzyme or tissue-specific celluW. A. Nelson-Rees and R. R. Flandermeyer, Science 191, 96 (1976).
324
[27]
lar function. Since most established lines do carry one or more characteristic marker chromosomes, karyotyping provides a handy method for
verifying that one is working with the correct cell line. Indeed, in many
instances the karyotype may be the only means available to check the
identity of a cell line.
3. Monitoring of Cell Lines. In many experiments, the cell line being
studied comes in deliberate contact with other cell lines, with subcellular
preparations derived from other cell lines, or with viruses grown in other
cell lines. Examples of direct cell contact would include enzyme crossfeeding experiments or the passage of tumor cells through animals. Experiments involving contact with subcellular preparations such as chromosome transfer experiments carry the danger of cross-contamination by
live cells from the donor line. In all these cases karyotyping provides a
method for monitoring cell lines to confirm that cross-contamination has
not taken place in the course of an experiment.
4. Verification of Hybrid Formation. Cell hybrids (described in this volume [28]) are extensively used for biochemical studies. In inter-species
hybrids the chromosomes provide a definitive set of markers to confirm
that the putative hybrid is in fact a hybrid of the desired type. In intraspecies hybrids, the use of parental cell lines with different characteristic
marker chromosomes can often provide a means of verifying that the
hybrid is the desired one.
5. Cytogenetic Research. In the four examples cited above karyotyping
is used as an aid in a research project that is not cytogenetic in nature.
Other work is concerned with gene mapping or karyotype evolution in
which karyotyping plays a major role. These projects often involve the
routine procedures described in detail below. They may also make use of
more highly specialized procedures outlined briefly below but the details
for which are beyond the scope of this article.
Chromosome Banding and Other Specialized Procedures

Prior to 1970, chromosomes only could be stained uniformly over their
length (solid staining) (Fig. 1A). Since then, however, several procedures
have been developed for differentially staining certain regions of the
chromosomes to give them a banded appearance (Fig. 1B). Standard
banding techniques in common use include: (1) Q-banding, or quinacrine
banding, in which chromosomes are stained with quinacrine, a fluorescent
compound, resulting in patterns of alternating bright and dull fluorescence
along the chromosome; (2) G-banding or Giemsa banding, a technique in
which chromosomes are treated with trypsin followed by staining with
Giemsa (a complex mixture of dyes) producing a series of dark staining
[:27]
KARYOTYPING
325
II
FIG. 1. A human metaphase spread stained first by solid staining (A), then destained by
placing the slide in 95% ethanol for 10 min, and G-banded (B).
bands (Fig. 1B) that correspond to the bright quinacrine bands; (3)
R-banding or reverse banding, a technique involving heat treatment of the
slides and Giemsa staining that results in banding patterns in which the
alternating light and dark regions are the reverse of those seen with
G-banding.
In this article we will deal with solid staining and G-banding in considerable detail, and the reader is referred to the original literature or to
review articles z-5 for details of Q- and R-banding.
In addition to the three major banding techniques there are several
other specialized procedures that will not be detailed in this article although they are mentioned briefly here to provide an idea of the available
technology. These include: (1) C-banding, a procedure for specifically
staining the centromere regions of chromosomes as well as regions of
constitutive heterochromatin; 6 (2) a technique in which a fluorescent dye
is used to stain centromeres of mouse but not human chromosomes, useful for identifying the species of origin of chromosomes in human-mouse
2 T. C. Hsu, Annu. Rev. Genet. 7, 153 (1973).
3 p. Pearson, J. Med. Genet. 9, 264 (1972).
4 0 . J. Miller, D. A. Miller, and D. Warburton, Prog. Med. Genet. 9, 1 (1973).
5 B. Dutrillaux and J. LeJeune, Adv. Hum. Genet. 5, 119 (1975).
6 F. E. Arrighi and T. C. Hsu, Cytogenetics 10, 81 (1971).
326
[27]
hybrids; 7 (3) a technique employing incorporation of bromodeoxyuridine

(BUdR) into DNA (in place of thymidine) allowing one to differentially
stain and therefore to distinguish chromatin with one BUdR substituted
strand from that with 0 or 2 substituted strands; the method has been used
to study replication kinetics of the individual chromosome bands 8 and to
differentially stain sister chromatids of the same chromosome allowing
quantitation of sister chromatid exchanges.9 Other specialized procedures
include (4) nucleolus organizing region (NOR) staining in which ammoniacal silver is used to specifically stain the nucleolus organizing regions of chromosomes containing the sites of the 18 S and 28 S ribosomal
RNA genes;l (5) cytological hybridization, a procedure in which radioactively labeled RNA or DNA is hybridized to denatured chromosomes
followed by autoradiography to identify the sites of complementary
chromosomal DNA; 11and finally (6) a series of immunological procedures
employing fluorescently labeled antinucleoside antibodies to identify, for
example, chromosomal sites of high methylcytosine concentration. 12
Detailed Methods
In this section we provide details for preparing microscope slides of
air-dried metaphase chromosome spreads, and the detailed techniques for
solid staining and for G-banding. The solid staining requires less expertise
than G-banding and is adequate for counting chromosomes or for identifying uniquely shaped marker chromosomes. The G-banding procedure is
more demanding and requires higher-quality metaphase spreads, but it
gives considerably more karyotypic information than does solid staining
(see Fig. 1A and B)
Preparation of high-quality chromosome spreads suitable for banding
contains an eleliaent o f " a r t , " and because of this the preparations usually
become better with practice. As is the case for the typical enzyme purification procedure, quality varies slightly from day to day, and the optimal
procedures vary somewhat from one cell type to another.
Slide Preparation
There are basically five steps in the preparation of high-quality slides

of air-dried metaphase cells. These are (1) optimization of cell growth to
7I. Hilwigand A. Gropp, Exp. Cell Res. 75, 122 (1972).
s S. A. Latt, Proc. Natl. Acad. Sci. U.S.A. 70, 3395 (1973).
9 S. A. Latt, Science 185,74 (1974).
10S. E. Bloomand C. Goodpasture, Hum. Genet. 34, 199 (1976).
H j. G. Gall and M. L. Pardue, this series, Vol. 21, p. 470.
~2B. W. Lubit, T. D. Pham, O. J. Miller, and B. F. Erlanger, Cell 9, 503 (1976).
[2 7]
KARYOTYPING
327
achieve a high mitotic index, (2) treatment with a mitotic inhibitor to allow
cells to accumulate at mitosis, (3) swelling of the cells in a hypotonic
environment, (4) fixation of the cells, and (5) spreading and drying of the
mitotic cells on a slide. Some general considerations with regard to each
step are described below.
1. Optimization of Growth. For cultured cells, a high mitotic index is
usually easy to achieve. For some cell lines, changing the medium the day
before making chromosome preparations may be helpful. Rapidly growing
lines are usually subcultured 1-2 days before and slower growing lines 3-4
days before "harvesting" the cultures for chromosome studies.
2. Metaphase Arrest. Mitotic spindle inhibitors such as colchicine or
deacetylmethyl colchicine (Colcemid) are normally used for blocking cells
in metaphase. Exposure time depends on the length of the cell cycle
(longer treatments for slower growing cells), but 0.5-2 hr is usually sufficient. The concentration of Colcemid is not critical: we have found 0.11.0/zg/ml adequate for many cell lines. Higher concentrations can cause
excessive contraction of the chromosomes reducing the clarity of banding
patterns.
3. Hypotonic Swelling. Swelling of the cells is a critical step and is
essential to achieve well-spread chromosomes. A variety of hypotonic
solutions are in common use, but 0.075 M potassium chloride is the most
popular and is a good choice for the initial attempt. Treatment time in
" h y p o " varies from one cell line to another, but 15-45 min is standard
and, if periods greater than 45 min are required, the hypotonic solution
might be diluted to increase its effectiveness.
4. Fixation. This is another critical step, and it always follows the
hypotonic treatment. The near universal fixative is 3 parts methanol to 1
part glacial acetic acid. It is usually made fresh each day and kept cold
until use. It is always added slowly to prevent cells from clumping. Some
protocols suggest removal of the hypotonic solution from the cells and
subsequent addition of fixative, while others call for slow addition of the
fixative to the hypotonic solution containing the cells. Both methods
work, but a given cell line may respond better to one or the other method.
To insure complete fixation the fixative is changed 3-4 times before drying
the cells onto a slide.
5. Air Drying. Spreading of the cells on a slide is also a crucial step. If
not correctly performed either the mitotic cells will fail to spread and the
chromosomes will remain in a heap in the cell or they will spread too
much with rupture of the cell membrane and release of the chromosomes
from the cell. Cells are spread by drying them onto a slide. Optimum
spreading of the cells seems to be related to the drying of the fixative,
which in turn depends on the temperature and relative humidity in the
328
[27]
laboratory. Drying time can be manipulated by blowing on the slide,

fanning the slide in the air, or placing the slide on a warm surface or hot
plate. We prefer a combination of blowing and moderate heat. The old
standard procedure of flame drying yields chromosome spreads that are
difficult or impossible to band.
Below we provide details of two techniques that are used routinely in
our laboratory. One is designed for handling cell suspensions and is suited
to cultured lymphocytes, long-term lymphoblast cultures, or any cell line
that grows in suspension. Any monolayer culture also can be trypsinized
and then handled as a suspension culture. The second technique is an in
situ method whereby monolayer cultures are processed right on their
growth surface in specially designed culture chambers that consist of a
microscope slide base with removable plastic chamber walls.
While the in situ method is not as generally applicable as the suspension method, we present it because we prefer it in certain circumstances
and use it routinely for some cell types. For example, it generally requires
fewer cells and is less wasteful of cells than the suspension method, an
advantage for a slow-growing line where mitotic cells are rare or for
human amniotic fluid cells where the results of karyotyping are required
as quickly as possible. However, it has one disadvantage in that the
quality of the metaphase spreads is quite sensitive to overcrowding of the
cells on the slide making it a little more difficult for the beginner to obtain
good spreads.
The suspension method described below applies specifically to longterm hutnan lymphoblast cultures and the in situ method to human fibroblast cultures. Suggested modifications in these basic protocols for a number of other cell lines follow.
A. THE SUSPENSION METHOD (As APPLIED TO HUMAN
LYMPHOBLASTS)
Equipment
Pasteur pipettes and rubber bulbs

Suction apparatus with collecting flask
Centrifuge, low-speed, refrigerated if available
Centrifuge tubes--10-15 ml siliconized glass, or plastic
Microscope slides (1 3 inch slides with ground-glass surface at
one end), precleaned. Before use place several slides in a beaker of
ice water.
Hot plate (e.g., Coming model PC 351) covered with paper towel and
set to maintain 100 ml of water in a 100-ml beaker at 60-70
[2 7]
KARYOTYPING
329
Solutions
Colcemid (Grand Island Biological Co., #521L), 10/zg/ml, stored at
4
Hypotonic: 75 mM KC1 warmed to 37 (5 ml per culture)
Fixative: 3 parts methanol to 1 part glacial acetic acid, made fresh on
day of harvest and kept on ice in a tightly stoppered vessel (20 ml
per culture)
Steps
1. OPTIMIZATION OF GROWTH. TWO days before harvesting the culture for chromosome spreads, subculture so that cells will be in mid to late
exponential growth phase on the day of harvest.
2. METAPHASE ARREST. TO 3 ml of lymphoblast culture add 2 ml of
fresh medium and 0.1 ml Colcemid. Incubate at 37 for 30 min.
3. HYPOTONIC SWELLING. Centrifuge cells at approx. 1000 rpm (200
g) for 8 min. Remove all but 0.2 ml of supernatant with suction apparatus.
Resuspend cells in this small volume, add 5 ml of warm hypotonic solution, and invert to mix. (Do not pipette cells because swollen cells are
fragile.) Incubate at 37 for 15 min.
4. FIXATION. Centrifuge cells at 1000 rpm for 8 min. Remove all but
about 0.2 ml of supernatant, and resuspend cell pellet in this small volume
by gentle agitation (no pipetting). Add first 1 ml of cold fixative slowly,
drop by drop with constant agitation. Add another 4 ml of fixative and
leave on ice for 15 min. Repeat fixation twice as above, but the fixative
can be added more quickly on the 2nd and 3rd fixation. Allow 15 min
between successive fixations. At this point cells can be stored overnight at
4 or slides can be made immediately.
5. AIR DRYING. Centrifuge cells at 1000 rpm for 8 min. Remove
supernatant and suspend cells in about 0.5 ml of cold fresh fixative. (Make
fresh fixative if cells have been stored overnight). Shake excess water and
ice from a microscope slide and, using a Pasteur pipette, drop 3-4 drops
of cell suspension onto the slide. Blow gently over the surface of the slide
to spread the cell suspension thinly, wipe excess liquid from the underside
of slide, and place on hot plate. Leave 5 min to dry thoroughly. Store
slides at room temperature in slotted trays or boxes so that surfaces do
not touch. (Note: It is our practice to prepare one slide and check it in the
microscope under phase contrast so that adjustments can be made on
subsequent slides if cells are too crowded or if spreading is poor--see
Table I.)
330
[27]
TABLE I
COMMON PROBLEMS IN SLIDE PREPARATION AND SOME SUGGESTED SOLUTIONS
Problem
Suggested solutions
Very few metaphase cells on slide despite

presence of many nonmitotic cells
For in s i t u method mitotic cells may be

washed off so reduce time in hypotonic
o r handle more gently o r blow less vigorously on slide. For either method, the
problem could be poor growth so increase
time in Colcemid o r change medium the
day before harvest to stimulate growth o r
adjust time between subculture and
harvest to take advantage of any partial
synchronization induced by the
subculture.
Very few metaphase cells and nuclei without

cytoplasm (extensive cell lysis)
Reduce time in hypotonic o r if your cells are

especially sensitive to hypotonic swelfing
increase salt concentration in the hypotonic o r handle cells more gently avoiding
pipetting.
Many broken cells with scattered

chromosomes
Blow less vigorously on slide.
Underspread cells with overlapped

chro/nosomes or with chromosomes
embedded in a thick layer of cytoplasm
(Fig. 3A)
Blow more vigorously on slide o r increase

time in hypotonic up to 40 min o r reduce
salt concentration of hypotonic o r if your
cells are especially resistant to hypotonic
swelling add trypsin to the hypotonic at
the concentration used for subculture.
Overcondensed chromosomes or widely

spaced chromatids
Reduce Colcemid concentration to O. 1 gg/ml

o r reduce time in Colcemid.
B. T H E
In
Situ
M E T H O D ( A s APPLIED TO H U M A N FIBROBLASTS)
Equipment
Lab-Tek chambers (single chamber, #4801, Lab Tek Products, Miles
Laboratories)
Pasteur pipettes and 22-gauge needle
Suction apparatus with collecting flask
Hot plate (e.g., Corning, model PC 351) covered with paper towel
and set to maintain 100 ml of water in a 100 ml beaker at 60-70 )
[27]
KARYOTYPING
331
Solutions
Colcemid (Grand Island Biological Co., #521L) 10/~g/ml, stored at 4
Hypotonic: 0.075 M KC1, warmed to 37 (2 ml per culture)
Fixative: 3 parts methanol to 1 part glacial acetic acid, made fresh on
day of harvest and kept on ice in a tightly stoppered vessel (10 ml
per culture)
Steps
1. OPTIMIZATION OF GROWTH. T w o days before harvesting, subcul-
ture cells from the flasks or plates in which they normally grow into two or
more Lab-Tek chambers. The cells should cover no more than 5-10% of
the growth surface after they are attached, and the final volume of medium is 4 ml. In order to prevent cells from settling primarily around the
edges, first add medium to the chamber and then pipette the concentrated
ceils into the center of the chamber and rock gently to distribute outwards. Incubate for about 2 days at 37. (Slow-growing strains may require 3 or more days.)
2. METAPHASE ARREST. On the day of harvest, cultures should be
nonconfluent (about 10-20% of surface covered by cells---see Fig. 2A)
and actively growing (several rounded-up mitotic cells or doublets should
be seen). Add 0.2 ml Colcemid and incubate for 90 min.
3. HVPOTONICSWELLING. With as little agitation as possible so as not
to detach loosely adhering mitotic cells, carry chamber slides from incubator. Very slowly, aspirate all of the culture medium by placing a
22-gauge needle connected to a suction flask into one corner of the
chamber. Add 2.0 ml of warm hypotonic solution drop by drop over the
surface of the chamber. Incubate for 30 min at 37.
4. FIXATION. Remove chamber slides very carefully from incubator
and, without removing any of the hypotonic solution, add 2.0 ml of cold
fixative drop by drop over the surface of the chamber. Let sit on the bench
for 20 min. Repeat fixation 3 times more at 15-min intervals, each time
aspirating all the fixative and replacing it with 2-3 ml fresh fixative. The
15-min interval is a minimum time, and longer times can be used if more
convenient. Slides can be dried immediately or stored at 4 overnight.
5. AIR DRYING. Aspirate all the fixative, remove chamber walls from
slide, blow gently over the surface of the slide to spread the cells, wipe
excess fixative from the underside of the slide, and place slide on hot plate
to dry. Leave at least 10 rain to dry thoroughly. Remove any rubber
cement remaining after removal of chamber walls and store slides at room
temperature in slotted trays or boxes so that slide surfaces do not touch.
332
[27]
FIG. 2. Monolayers of cells viewed in an inverted microscope at low ( 10 objective)

power. (A) human fibroblasts in a Lab-Tek chamber at a cell density suitable to begin harvest
by the in situ procedure. (B) CHO cells in a Lab-Tek chamber at a cell density suitable to
begin harvest by the in situ procedure. (C) CHO cells in a flask at a cell density suitable to be
trypsinized and harvested by the suspension protocol (note many rounded-up mitotic cells
and doublets indicating extensive mitotic activity). The cell density is slightly less in (B) than
in (C) since cells need room to spread by the in situ procedure.
[27]
KARYOTYPING
333
C. SUGGESTED MODIFICATIONS FOR OTHER CELL LINES
Since different cell lines can differ considerably in their nuclear size,
amount of cytoplasm, number and size of chromosomes, and response to
a hypotonic environment, any single protocol will not work well for all cell
lines. We have had experience with a number of cell lines of human,
hamster, and mouse origin and summarize below some general guidelines
for them.
Most primary cultures of fibroblast-like cells, regardless of the species
or tissue of origin, can be harvested with the in situ procedure described
above for human skin fibroblasts.
Human amniotic fluid cells are primarily of two types, fibroblast-like
and epithelial-like, and these cultures respond well to the same in situ
procedure. In fact, for amniotic fluid cultures set up for prenatal diagnosis
it is common to initiate the cultures in Lab-Tek chamber slides and to
examine the chromosomes of individual colonies of cells in these primary
cultures. This has two advantages. It allows diagnosis to be made in the
shortest possible time (no subculturing is required), and it provides information regarding the colony of origin of any cell with an unusual
karyotype.
The CHO cell line, a permanent line of Chinese hamster ovary origin,
grows in suspension culture and can be harvested with the suspension
method described above but with a longer period in Colcemid (60-90
min).
CHO cells also grow in monolayer but are less well attached than
primary fibroblasts; at mitosis the cells round up and are easily dislodged
from the surface (Fig. 2B and C). Thus, while CHO can be harvested by
the in situ method (at the cell density shown in Fig. 2B), great care has to
be taken to avoid washing away all the mitotic cells during the hypotonic
and fixation steps. An alternative procedure, starting from a monolayer
(at the cell density shown in Fig. 2C), is to treat for 90 min with Colcemid,
then tap or shake the flasks (or plates) to detach many of the mitotic cells,
removing the medium with detached cells to a centrifuge tube. Any remaining mitotic cells can be recovered by lightly trypsinizing: the trypsin
solution used for subculture is added for 30-60 sec. After pooling the
trypsinized cells with the mitotic cells, we centrifuge, resuspend in
hypotonic solution, incubate at 37 for 30 min, and add an equal volume of
fixative to the hypotonic solution before the next centrifugation. The cells
are fixed twice more, and slides are prepared according to the suspension
protocol.
A number of other commonly used cell lines of Chinese hamster origin
(V-79, GM-7, M3-1, 18-1, DON), mouse origin (3T3, L, LMTK), or
334
[27]
human origin (HeLa, HT 1080) have characteristics similar to CHO, or

intermediate between CHO and primary fibroblasts. We have harvested
all of these successfully using either the in situ method, or starting from a
monolayer (one 75 cm 2 flask, 1 day before becoming confluent), trypsinizing, and following the suspension method. Colcemid treatment is usually
60-90 min and hypotonic treatment about 30 rain.
Hybrid cell lines, whether of intra-species or inter-species origin, usually have characteristics intermediate between the two parental cell types
and can often be harvested with the procedure used for the parental line
that is most fibroblast-like.
Staining Techniques
A. SOLID STAINING
While many different stains have been used to give uniform staining of
the chromosomes, Giemsa stain seems to be the most common and is the
one we use routinely. We use commercially prepared concentrated
Giemsa (Giemsa stain, Original Azure Blend Type, Harleco #620). Like
many cytogenetic labs we dilute into Gurr's buffer at pH 6.8 (Gurr's buffer
tablets, Searle Diagnostic, England, 1 tablet per liter of distilled water) but
10 mM potassium phosphate at pH 6.8 works equally well.
For solid staining, set up 3 coplin jars (glass jars with flat sides and
ridges that hold 5 slides) or 50-ml beakers as follows:
1. 50 ml of Gurr's buffer (or 10 mM phosphate at pH 6.8) containing
1.5 ml of concentrated Giemsa;
2. 50 ml of water;
3. 50 ml of water.
Place slides into the stain for 3 min and rinse by dipping 8-10 times in
each beaker of water. Shake off excess water and stand on end on a paper
towel to drain and dry.
If chromosomes are not sufficiently dark the slide can be returned to
the stain for a few minutes. The diluted Giemsa can be used to stain up to
about 50 slides and is discarded at the end of the day.
I . GIEMSA (G) BANDING
Giemsa banding or G-banding was so named because it was the first
banding procedure described to utilize this stain. However, it is
clearly a misnomer because the Giemsa stain is not the critical ingredient
in the procedure and also because Giemsa can be used for solid staining
(above) as well as in other banding procedures. In the original G-banding
technique the critical step was incubation of the slide at high temperature.
[27]
KARYOTYPING
335
This has largely been replaced by incubation of the slide in a crude trypsin
solution, and that has been our standard banding method for several
years.
The trypsin is a crude preparation (Bacto trypsin, #0153, Difco Laboratories, Detroit, Michigan) dissolved in 10 ml of distilled water and
stored frozen in 1-ml portions. The Giemsa stain is the same as that used
for solid staining (Original Azure Blend Type, Hareleco #620). The
Giemsa has a shelf life of only about 3 months at room temperature;
thereafter it may be used for solid staining but not for G-banding. As for
solid staining, the Gurr's buffer in which the stain is diluted can be replaced with 10 mM phosphate at pH 6.8.
A prerequisite for good G-banding is a high-quality slide. Chromosomes should be straight and slender with sister chromatids lying close
together. Cells should be well spread with few overlapped chromosomes,
but the cells should not be broken. Even on a good slide, only one
metaphase spread out of 10 or 20 will fit these criteria. Therefore, many
mitotic cells should be provided on the slide. These properties can be
checked by phase-contrast microscopy before proceeding with banding.
Our technique, modified only slightly from the original trypsinGiemsa technique of Seabright, 13is done at room temperature by passing
the slide through a series of 7 coplin jars or 50-ml beakers as follows:
1. 50 ml of 0.15 M NaCl--brief rinse
2. 50 ml of 0.15 M NaCI containing 1 ml of trypsin stock solution-30-120 sec
3. 50 ml of 0.15 M NaCl--brief rinse (dip slide 8-10 times)
4. 50 ml of 0.15 M NaCl--brief rinse
5. 50 ml of Gurr's buffer (or 0.01 M phosphate buffer, pH 6.8) containing 1.5 ml of concentrated Giemsa--l-2 min
6. 60 ml of distilled water brief rinse (dip slide 8-10 times)
7. 50 ml of distilled water--brief rinse
Slides are placed on end on a paper towel to drain and dry.
Since the time required in trypsin depends to some extent on how the
slides were prepared, especially the temperature of the hot plate, a few
trial slides may be necessary.
Following either solid staining or G-banding, slides are ready for microscopic study. We do not bother to mount a coverslip on the slides since
the presence of a coverslip does not seem to significantly affect the quality
of the microscopic images. When using oil-immersion lenses we place the
immersion oil on the surface of the slide and then remove it immediately
after use by dipping the slide into fresh xylene (xylol) several times. The
,3 M. Seabright, Lancet 2, 971 (1971).
336
[27]
new imersion oils, if left on the slide, will extract the stain resulting in
fading of the image.
If a coverslip is to be mounted, add a drop of Permount (Fisher Scientific Co., #SO-P-15), or a 50:50 mixture of Permount and xylene. Coverslips can be removed if necessary by soaking the slide overnight in
xylene.
Suggestions for Improving Suboptimal Slides

For a simple chromosome count with solid staining you should encounter little difficulty. Even with suboptimal slides, some of the metaphase spreads should have chromosomes that are sufficiently distinct for
counting. The only possible problems might be too few cells if the starting
culture is small or a low proportion of metaphase cells if the culture was
not actively growing. Of course, these problems can only be corrected by
altering the culture conditions.
For G-banding or any other banding technique the requirement for
high-quality slides is greater, and the novice may therefore experience
difficulty in the beginning. Since all cell lines are not alike, even an experi,
enced cytogeneticist will sometimes fail to achieve good banding with a
new cell line. In these cases, examination of the rejected slides may give
important clues regarding appropriate adjustments in technique. Before
rejecting a set of slides, however, keep in mind that some slides can be
better than others and that on any given slide only a fraction of the
metaphase cells will be well spread and optimally banded. It is therefore
wise to scan a few representative areas of several slides before rejecting
the whole preparation.
In Tables I and II some of the common problems encountered in slide
making and in banding are listed together with a few suggestions for their
correction. Figure 3 illustrates a few of the problems.
Microscopy and Photography

Because of the great variety available, it is not possible to provide
details of microscopic and photographic procedures. We therefore outline
briefly some general principles and leave the details to the information
booklets describing individual equipment. The following comments apply
to solid-stained or G-banded preparations, but not to slides stained with
fluorescent compounds.
The first requirement for a chromosome study is a good binocular
laboratory microscope. Two objective lenses should be sufficient, one low
power ( 10 or 16) for scanning the slide to look for metaphase spreads,
[27]
KARYOTYPING
337
TABLE II
COMMON PROBLEMSIN BANDINGAND SOME SUGGESTEDSOLUTIONS
Problem
Suggested solution
Darkly stained but unbanded chromosomes
Increase time in trypsin o r decrease time in

stain.
Individual chromatids partially banded, but

separated (undertrypsinized--Fig. 3D)
Increase time in trypsin (chromosomes swell

causing the two chromatids to come
together and knobs on chromatids to
coalesce into bands) o r if chromatids are
too far apart or too condensed the problem is one of slide preparation--see
Table I.
Lightly stained swollen chromosomes with

ragged edges (over trypsinized--Fig. 3C)
Reduce time in trypsin; o r this can happen

even after 10 sec in trypsin if slides are not
dried with heat.
Lightly stained chromosomes with indistinct bands--sometimes chromosomes

appear to be embedded in cytoplasm
(Fig. 3A)
Decreasing trypsin time may help o r usually

problem is due to underspreading during
slide preparation--see Table I.
Halo effect--pale chromosomes with dark

outlines
Reduce time in trypsin; o r reduce time on

hot plate during slide making; o r reduce
number of cells on slides (seed Lab-Tek
chamber with fewer cells in case o f in s i t u
technique).
Failure to band despite long treatment times

in trypsin and well-spread slides
Try a new batch of Giemsa stain especially

if old batch is over 3 months old. Some
batches of Giemsa give poor banding even
when new so test different lot numbers.
a n d o n e high p o w e r ( 100) (oil i m m e r s i o n ) for c o u n t i n g , e x a m i n i n g , a n d

p h o t o g r a p h i n g c h r o m o s o m e s . It is u s e f u l if the l o w - p o w e r o b j e c t i v e is a
p h a s e - c o n t r a s t lens (with m a t c h i n g p h a s e rings in the c o n d e n s e r ) so t h a t
u n s t a i n e d slides c a n b e s c a n n e d b y this m e a n s . By s i m p l y r e m o v i n g the
p h a s e ring f r o m the c o n d e n s e r t h e s a m e o b j e c t i v e lens c a n b e u s e d with
bright field i l l u m i n a t i o n to s c a n s t a i n e d slides.
I f p h o t o g r a p h i c r e c o r d s o f the c h r o m o s o m e s are r e q u i r e d t h e n the
m i c r o s c o p e m u s t b e e q u i p p e d with a c a m e r a . A n a u t o m a t e d p h o t o m i c r o s c o p e is n o t n e c e s s a r y . I f y o u a l r e a d y o w n or h a v e access to a good
l a b o r a t o r y m i c r o s c o p e , t h e n c h a n c e s are a 3 5 - m m c a m e r a c a n b e m o u n t e d
338
[27]
FIG. 3. (A) A portion of a poorly spread cell, trypsin banded. The thick layer of cytoplasm over the chromosomes causes background staining (blue through the microscope, grey
on the print) and prevents the chromosomes from becoming clearly banded. It also refracts
the light making it impossible to focus clearly on the chromosomes. (B) A portion of a
well-spread, clearly banded cell for comparison with (A), (C), and (D). (C) A portion of an
over-trypsinized cell. (D) A portion of an under-trypsinized cell.
on the top at a reasonable cost. Automatic film a d v a n c e is an expensive
luxury and not necessary in m o s t cases. A u t o m a t i c e x p o s u r e control is
also u n n e c e s s a r y since the a m o u n t o f light reaching the film is relatively
constant f r o m one m e t a p h a s e spread to another, and f r o m slide to slide.
Thus, once e x p o s u r e time has been determined for a particular combination of m i c r o s c o p e , c a m e r a , and film, it need not be altered. A green filter
is c o m m o n l y used and gives sharper images on black and white film.
The choice o f film is not critical and any m e d i u m - c o n t r a s t black and
white film should provide adequate pictures. I f y o u develop y o u r o w n
films a local photographic supplier should be able to provide advice on the
choice of film and developer. One combination that we have used successfully is Tri-X (Kodak) film d e v e l o p e d in Microdol X (Kodak) d e v e l o p e r
according to the m a n u f a c t u r e r ' s instructions. Fixation in Rapid Fix
(Kodak) and w a s h p r o c e d u r e s are standard.
[27]
KARYOTYPING
339
For printing of pictures (approximately 5 7 inches) any photographic

enlarger that handles 35-mm film can be used. We print on Kodak singleweight Ektamatic SC + N + A paper (contrast grade 2 without filters)
and develop automatically in an Ektamatic Processor (Kodak). In the
absence of automatic processing equipment prints can be made on any
medium-contrast (contrast grade 2 or 3) paper and developed in trays
according to instructions with the developer.
Whether processing one's own films or sending them out for processing, one or more test films must be obtained. Choose a typical metaphase
spread and photograph it several times using two or three different light
levels aiming for an exposure time of 1-3 sec. Once the best exposure
time for given equipment is determined, note all the settings, e.g., condenser diaphragm opening, aperture diaphragm opening, current to light
source, choice of filters, and auxilliary magnification lenses, and use these
standard conditions for all future photographic work.
Arranging a Karyotype
With only a little experience solid-stained chromosomes can easily be
counted in the microscope with high-power magnification (total magnification x 1000-1400). However, for any further analysis, such as identification of specific marker chromosomes, or determination of the number of
copies of individual chromosomes, it is not only necessary to use banding
techniques but it is usually necessary to photograph and print the chromosome spreads and then to cut out the individual chromosomes in order to
arrange them in a karyotype.
Figure 1 shows two photographs of the same human metaphase cell,
first solid stained and then G-banded. It is clear that counting chromosomes is much easier with solid staining than with G-banding, and that
matching up and identifying homologous chromosomes in the banded cell
would be difficult without actually cutting out the individual chromosomes
and arranging them in a karyotype.
Figures 4, 5, and 6 are normal diploid (male) karyotypes of human,
Chinese hamster, and mouse, respectively. These three are presented
because many of the animal cell cultures in common use are derived from
these three species. The arrangement of the human karyotype was standardized in 1971 at the Paris Conference on Standardization in Human
Cytogenetics. Not only was each chromosome assigned a specific number, but all the bands were also numbered and a shorthand system was
described for noting chromosomal abnormalities. TM Many human genes
14D. Bergsma, ed., "Paris Conference: Standardizationin Human Cytogenetics,"Birth
Defects, Vol. 8, No. 7. Natl. Found., New York, 1972.
340
[27]
II If
tm
Ill
FIG. 4. ,The human male karyotype, G-banding. Top row: Groups A (chromosomes 1-3)
and B (4 and 5). Second row; Group C (chromosomes 6-12). Third row: Groups D (chromosomes 13-15) and E (16-18). Fourth row: Groups F (chromosomes 19 and 20) and G (21 and
22) and the sex chromosomes X and Y.
FIG. 5. The Chinese hamster (Cricetulus griseus) male karyotype, G-banding. Top row:
Chromosomes 1-5. Bottom row: Chromosomes 6--10 plus X and Y.
[27]
KARYOTYPING
341
ii
FIG. 6. T h e m o u s e (Mus musculus) male karyotype, G-banding. Top row: C h r o m o s o m e s

1-5. Second row: C h r o m o s o m e s 6-10. Third row: C h r o m o s o m e s 11-15. B o t t o m row:
C h r o m o s o m e s 16-19 plus X and Y. Figure prepared by Dr. P. W. Aliderdice.
also have been assigned to specific chromosomes or regions of

chromosomes. 15
The Chinese hamster karyotype in Fig. 5 follows the proposed standard karyotype of Ray and Mohandas. 16While this will likely become the
accepted standard, recent publications have appeared using an alternate
format in which the #2 chromosomes are placed the other way around
(inverted) and the X and Y are referred to as the #3 pair. Thus, pairs 3-10
of Fig. 5 are numbered 4-11 in the alternate scheme.
The mouse karyotype has also been standardized, 17 and the genetic
linkage groups that were established several years ago have all been assigned to specific chromosomes.18
The karyotypes of a great many other species have been published,
but it is beyond the scope of this article to reference them all. There are
hundreds of different cell lines in existence, many of which have not been
15 V. A. M c K u s i c k and F. H. Ruddle, Science 196, 390 (1977).
16 M. Ray and T. M o h a n d a s , Cytogenet. Cell Genet. 16, 83 (1976).
tr " S t a n d a r d K a r y o t y p e o f the M o u s e , Mus musculus," J. Hered. 63, 69 (1972).
18 O. J. Miller and D. A. Miller, Annu. Rev. Genet. 9, 285 (1975).
342
[27]
T A B L E III
KARYOTYPE CHARACTERISTICS OF A FEW CELL STRAINS AND ESTABLISHED CELL LINES
Species
Human
(2n = 46)
Chinese
hamster
(2n = 22)
Syrian
hamster
(2n = 44)
Mouse
(2n = 40)
Rat
(2n = 42)
Modal no. o f
chromosomes
No. o f
marker
chromosomes
Reference to
karyotype
46
63-78
0
16--25
~.b
b-~
(HeLa
contaminants)
46
b-e
f
diploid
CHO
CHW
22
20-21
22
0
10-13
4
a.t
J
diploid
BHK
44
44
0
0
k
t
diploid
CAK
RAG
A9
LMTK-
40
41
approx. 60
approx. 60
52
0
1-3
approx 15
several
approx. 90
m
n
P
q
r
diploid
42
Name of
line
WI-38
(diploid
flbroblast)
HeLa
KB,Hep-2,
D98,HEK,
etc.
HT-1080
Drosophila
melanogaster
(2n = 10)
GM2 and GM3
Variable
with time
a D. B e r g s m a , ed., " P a r i s C o n f e r e n c e , " Birth Defects, Vol. 8, No. 7. Na. Found.,

N e w York, 1972.
b O. J. Miller, D. A. Miller, P. W. Alderdice, V. G. Dev, a n d M. S. Grewal, Cytogenetics
10, 338 (1971).
c C. C. Lin and S. Goldstein, J. Natl. Cancer Inst. 53, 298 (1974).
aW. K. H e n e e n , Hereditas 82, 217 (1976).
e W. A. N e l s o n - R e e s and R. R. F l a n d e r m e y e r , Science 191, 96 (1976).
[27]
KARYOTYPING
343
karyotyped. Although it is impractical to list all of the lines that have been
karyotyped, a few of the more commonly used lines are listed in Table III
with references to the detailed karyotypes.
In attempting to karyotype one's own cell line for the first time, there
are a few facts to keep in mind. First, chromosomes are continually contracting as the cell proceeds into mitosis and in so doing several fine bands
present on a chromosome of a prophase cell or early metaphase cell will
coalesce into a smaller number of thicker bands by mid-metaphase. Thus,
cells with long thin chromosomes are generally in an early stage of mitosis
and have more total bands (and hence different patterns) than cells with
shorter thicker chromosomes. With long Colcemid treatment, all cells
tend to be at the same stage, that represented in Figs. 4, 5, and 6, thereby
considerably reducing this source of variation.
Second, there is subtle variation in chromosomes between the individual members of a species. In humans, for example, this variation can be in
the width of certain bands that stain darkly by C-banding, or in the
fluorescent intensity of certain bands after staining with quinacrine. However, such inherited chromosomal variants or polymorphisms are not
readily apparent by solid staining or G-banding and should not present
major difficulties in karyotyping by these techniques,
Third, in comparing the karyotype of one's cell line with published
karyotypes, keep in mind that karyotypes evolve with continued culture.
rS. Rasheed, W. A. Nelson-Rees, E. M. Toth, P. Arnstein, and M. B. Gardner, Cancer
33, 1027 (1974).
~M. Ray and T. Mohandas, Cytogenet. Cell Genet. 16, 83 (1976).
hL. L. Deaven and D. F. Petersen, Chromosoma 41, 129 (1973).
~R. G. Worton, C. C. Ho, and C. Duff, Somatic Cell Genet 3, 27 (1977).
J P. A. Gee, M. Ray, T. Mohandas, G. R. Douglas, H. R. Palser, B. J. Richardson, and
J. L. Hamerton, Cytogenet. Cell Genet. 13, 437 (1974).
kN. C. Popescu and J. A. Di Paoio, Cytogenetics 11,500 (1972).
i R. Marshall, Chromosoma 37, 395 (1972).
m "Standard Karyotype of the Mouse," J. Hered. 63, 69 (1972).
hR. A. Farber and R. M. Liskay, Cytogenet. Cell Genet. 13, 384 (1974).
o The number of marker chromosomes is enhanced in mouse cell lines because the
normally telocentric (centromere at one end) mouse chromosomes tend tp fuse
together at the centromere creating biarmed markers that are not as unique as those
created by random breakage and reunion.
P E S. Hashmi, E W. Allderdice, G. Klein, and O. J. Miller, Cancer Res. 34, 79 (1974).
qE W. Allderdice, O. J. Miller, D. A. Miller, D. Warburton, P. L. Pearson, and
H. Harris, J. Cell Sci. 12, 263 (1973).
rA. R. Rushton, Can. J. Genet. Cytol. 15, 791 (1973).
' "Standard Karyotype of the Norway Rat," Cytogenet. Cell Genet. 12, 199 (1973).
t S. F. Dolfini, Chromosoma 58, 73 (1976).
344
[27]
For example, a CHO line may have several normal Chinese hamster
chromosomes, .several marker chromosomes characteristic of CHO lines
already described, and additional new marker chromosomes not previously described that have evolved by structural rearrangements during
continuous cell culture. Because of this evolution, it is not valid to assume
that one's own CHO line will have the same karyotype as CHO lines in
other laboratories. If it is important for the interpretation of biochemical
or other data, then the karyotype of the specific subline under study must
be determined.
Comments
Karyotyping can be an extremely valuable research tool for anyone
working with cultured cells. In some instances, a quick chromosome
count may be all that is required. For this, solid staining would be sufficient, and photographs of chromosomes would not likely be required. The
solid-staining technique is easily learned, and for many cell lines one can
expect adequate chromosome spreads on the first attempt (also see this
volume [27]). Thus, it should take no more than a few hours to obtain a
chromosome count on 20-25 cells.
If more detailed information is necessary, then chromosome banding
may be required. The G-banding technique presented above will be adequate for most applications, although some situations may require one of
the special techniques that have been noted. Since the requirement for
well-spread chromosomes is greater for banding than for solid staining, a
few days may have to be invested in optimizing the techniques for one's
own cell line. Banded chromosomes are also difficult to analyze in the
microscope without considerable experience, thus making it necessary to
photograph, print, cut out, and arrange the chromosomes in a karyotype.
However, once the technique is developed, even a detailed banded analysis can be performed in a day or two.
While several days or even a few weeks invested in developing
karyotyping expertise may seem excessive, it is completely inconsequential compared to the many years that already have been spent on HeLa cell
research by scientists who thought they were studying something else. It
is similarly inconsequential when compared to the time that may be involved in studying gene regulation in heteroploid cell lines where abnormal gene dosage relationships may preclude the operation of normal regulatory mechanisms.
Put in these terms we suggest that karyotyping expertise is not only a
valuable tool, but often an essential one for the biochemist who works
with cultured cells.
[28]
CELL FUSION
345
[28] Cell F u s i o n
By ROGER H. KENNETT
Cell fusion is a biological phenomenon that has been studied and developed from two distinct points of view. Biochemists and cell biologists
have been interested in the role that the fusion of biological membranes
may play in normal cellular processes such as plasma membrane formation and in the movement of vesicles in the processes of endocytosis and
exocytosis.l On the other hand much progress in the areas of eukaryotic
genetics, 2 virology, 3 and the analysis of malignancy4 has been made possible by application of cell fusion techniques to the production of heterokaryons and hybrid cells. Heterokaryons result from the fusion of plasma
membranes of two different cell types so that one or more nuclei from
each type are included within a common cytoplasm. A hybrid cell is
formed when growth and division processes take place in a heterokaryon
and the genetic materials from the two types of nuclei are included in the
single nucleus of a proliferating cell. Such hybrid cells can then be propagated in culture with the growth characteristics usually depending on the
cell types fused. Generally, the hybrids can be maintained in culture
under the same conditions and for as long as the most actively growing of
the two parental cells. Several selective systems have been devised that
allow preferential growth of hybrid cells and eventual elimination of the
two parental types. These methods will be described below.
There has been interaction among those approaching cell fusion from
these two points of view at the level of the various agents used to promote
the fusion of cellular membranes. These promoters of membrane fusion
fall into two general categories: enveloped viruses and lipophilic or lipolytic reagents such as lysolecithin and polyethyleneglycols.5
I will concentrate here on those methods for cell fusion that allow
production of a high proportion of viable heterokaryons or hybrids and
will discuss the numerous ways in which such fused cells are useful to
investigators involved in analysis of biochemical and genetic properties of
cultured cells. For those who would like a more detailed account of the
i G. Poste, Int. Rev. Cytol. 32, 157 (1972).
2 V. A. McKusick and F. H. Ruddle, Science 196, 390 (1977).
3 H. Koprowski, Fed. Proc., Fed. Am. Soc. Exp. Biol. 30, 914 (1971).
4 G. Barski, in " T i s s u e Culture: Methods and Applications" (P. F. Kruse, Jr. and M. K.
Patterson, eds.), p. 469. Academic Press, N e w York, 1973.
5 G. Poste and A. C. Allison, Biochim. Biophys. Acta 300, 421 (1973).
METHODS IN ENZYMOLOGy,VOL. LVIII

ISBN 0-12-181958-2
346
[28]
historic development of cell fusion techniques, the two divergent views of

Ephrussi 6 and Harris 7 provide interesting reading.
Heterokaryons and Hybrids: Their Use in Analysis of Eukaryotic Gene
Expression
The ability to include genetic material from two cell types within the
same cytoplasm makes it possible to pose questions concerning the interactions of two sets of genes. The two types of questions that have been
asked are:
1. What positive or negative regulatory effects can be observed when
genetic material from two cells in different states of differentiation are
included in the same cytoplasm? Do elements from one genome turn on or
off genes in the other genome?
2. Is genetic complementation observed when nuclei from cells, each
deficient in some function, are included in the same cytoplasm?
In general, it can be concluded that when two cells expressing different
states of differentiation are fused, there is usually suppression of differentiated functions. 8 For example, fusion of melanomas with fibroblasts
produces hybrids that do not produce melanin as do the melanoma parental cells.a Similarly, fusion of a Friend erythroleukemia mouse cell line,
which can be induced to synthesize hemoglobin, with a human fibroblast
suppresses globin message formation when the mouse globin loci are present as shown by nucleic acid hybridization studies. TM
A notable exception to this rule is that fusion of a rat liver tumor cell
line synthesizing several liver-specific proteins with other types of cells
sometimes l'esults in hybrids that synthesize some of these proteins. 1~
There.are even reports that, when a rat liver tumor cell is hybridized with
some types of cells, e.g., human fibroblast, not only are the rat liver
proteins synthesized but, in some clones, liver-specific proteins are synthesized from the genes of the other species included in the hybrid.12 In
hybrids in which these liver-specific functions are suppressed, one often
finds reexpression of one or more of the liver-specific functions after the
6 B. Ephrussi, "Hybridization of Somatic Cells," Princeton Univ. Press, Princeton, New
Jersey, 1972.
7 H. Harris, "Cell Fusion," Harvard Univ. Press, Cambridge, Massachusetts, 1970.
s R. C. Davidson, Syrup. Soc. Study Dev. Biol. 31, 295 (1973).
9 R. L. Davidson, B. Ephrussi, and K. Yamamoto, J. Cell. Physiol. 72, 115 (1968).
10 A. Deisseroth, R. Velez, R. D. Burk, J. Minna, W. French Anderson, and A. Nienhuis,
Somatic Cell Genet. 2, 373 (1976).
~1 G. J. Darlington and F. H. Ruddle, Mod. Trends Hum. Genet. 4, 111 (1976).
12 G. J. Darlington, H. P. Bernhard, and F. H. Ruddle, Science 185, 859 (1974).
[28]
CELL FUSION
347
cells are grown in culture for several passages.13 This may result from loss
of chromosomes and a resulting change in relative gene dosage favoring
reexpression of one or more of the liver-specific proteins. These observations are consistent with the finding that when a near tetraploid derivative
of the liver cell line is fused with a diploid cell, the liver-specific functions
are expressed in most of the hybrids.14
The suppression and reexpression of differentiated functions in hybrids, and the possibility that this results from chromosome loss, point out
a significant difference between hybrids and heterokaryons. Hybrids start
out after the fusion event with at least one full set of chromosomes from
each parental cell. As the hybrids are continued in culture, variants arise
that have lost chromosomes derived from one or both of the parental cell
types. 15 This presumably is a consequence of unequal distribution of
chromosomes during mitosis and cell division and the growth advantage
that some of the resulting variants have over the original hybrid clone.
How rapidly a variant with a reduced chromosome number comes to
predominate in the hybrid culture depends on the types of cells fused and
the species from which they each originate. An extreme case of chromosome loss exists in the formation of most rodent cell human cell hybrids. In most cases, the human chromosomes are rapidly lost leaving only
a few human chromosomes in each hybrid clone.16 Since each clone has a
different set of a few human chromosomes, mouse human hybrids have
in fact been very useful tools for human gene mapping as indicated below. lr
One confirmed exception to the rule of extinction in heterokaryons
occurs when fibroblasts fused with chick erythroblast nuclei are reactivated and begin to synthesize globin. This also happens when the erythroblasts are fused with cells from which the nuclei have been removed. TM
In regard to the second question above, genetic complementation has
been demonstrated in several systems either by assaying the individual
heterokaryons or the mass culture after the fusion for the particular factor(s) missing from the unfused cells. 19,2 Another method is to detect
13 M. C. Weiss, R. S. Sparkes, and R. Bertolotti, Somatic Cell Genet. 1, 27 (1975).
14 j. A. Peterson and M. C. Weiss, Proc. Natl. Acad. Sci. U.S.A. 69, 571 (1972).
15 E. Engel, J. Empson, and H. Harris, Exp. Res. 68, 231 (1971).
~e M. Weiss and H. Green, Proc. Natl. Acad. Sci. U.S.A. 58, 1104 (1967).
17 D. Bergsma, ed., " H u m a n Gene Mapping III," Birth Defects: Orig. Art. Ser. XII(7).
Karger, Basel, 1976.
is T. Ege, J. Zerthen, and N. R. Ringertz, Somatic Cell Genet. 1, 65 (1975).
19 Reviewed in N. R. Ringertz and R. E. Savage, "Cell Hybrids." Academic Press, New
York, 1976.
20 E. Weed-Kastelein, W. Keijzer, and D. Bootsma, Nature (London), New Biol. 238, 80
(1972).
348
[28]
complementation by using selective conditions under which neither of the

unfused cell types will grow but in which the fused cells can grow if
complementation takes place. By this procedure, Puck, for instance, has
shown that several glycine auxotrophs can be classified into four complementation groups. 21
T h e Use of H y b r i d Cells for Gene Mapping
Because most mouse x human hybrids lose human chromosomes,
most clones in a collection of such hybrids contain different combinations
of human ~chromosomes. By relating the expression of a human gene
product to the presence of a specific human c h r o m o s o m e one can identify
the c h r o m o s o m e containing the gene for that particular gene product. This
has made possible the chromosomal localization of several hundred
human genes. Mouse human hybrids, made with certain combinations
of cell types, result in retention of human chromosomes with an apparently random loss of mouse chromosomes. This, o f course, allows the
analysis of the mouse genome by similar methods. 2"16'17'22
Hybrids and Malignancy
If a malignant cell is fused with another cell type, one can ask if the
hybrid retains the malignant phenotype. One may thus inquire as to
whether malignancy is a dominant or recessive characteristic and, at this
point, it appears that the answer depends on the specific malignant cell
that is chosen. In most cases, the malignancy is defined by the ability to
grow as a tumor in immunologically deficient " n u d e " mice or to grow in
semisolid medium without attachment to a substrate 4e2a~ (see also this
volume [30]).
Detection of Viruses in Nonpermissive Cells and Study
of Viral Integration ~ites
Cell fusion has been used as an effective tool for detecting the presence of viruses in nonpermissive cells. By fusing a cell that permits the
replication of a virus with a cell line that contains the virus genome but is
21F. T. Kao, L. Chasin, and T. T. Puck, Proc. Natl. Acad. Sci. U.S.A. 64, 1284(1969).
22C. Croce, A. Talavera, C. Basilico, and O. J. Miller,Proc. Natl. Acad. Sci. U.S.A. 74,694
(1977).
2~a S. Shin, V. H. Freedman, R. Risser, and R. Pollack, Proc. Natl. Acad. Sci. U.S.A. 72,
4435 (1975). See also this volume [31].
23C. O. Poulson and J. Rugaard, Acta Pathol. Microbiol. Scand. 79, 397 (1975).
[28]
CELL FUSION
349
nonpermissive, virus replication is initiated and the virus is t h e r e b y detected. 3

Croce et al. h a v e used cell hybrids to detect the c h r o m o s o m a l location
o f viral (SV40) integration in h u m a n fibroblasts t r a n s f o r m e d with SV40
virus. The t r a n s f o r m e d cells were fused with m o u s e peritoneal macrophages that do not normally divide in culture. Hybrids isolated from a
given fusion or series of fusions with the same SV40 t r a n s f o r m e d line all
contained the s a m e h u m a n c h r o m o s o m e ; some had only that single h u m a n
c h r o m o s o m e . All the hybrids contained the SV40 genome, expressed
SV40 antigens, and were of a malignant p h e n o t y p e when tested by the
criteria stated above. 24
Hybrids as a Source of Cell Products for Immunological

and Biochemical Analysis
It has recently become clear that somatic cell hybrids will become an
important source of specific cellular products that cannot bc obtained
from short-term primary cultures. The best c:~amplc of this is the system
developed by Milstcin and others for the production of hybrid myclomas
making monoclonal antibody against an antigen of choice.2,26If an antigen
is injected into mice the spleen cells can bc removed from the mouse and
put into culture. These cells will divide and produce antibody but will last
for only a short period in culture and then stop dividing.27 Cell hybridization makes it possible to combine the spleen cells making antibody with
mouse plasmacytoma lines. The mouse lines arc plasma cell tumors that
can bc cultured continually or passaged from mouse to mouse as a tumor.
They secrete their o w n mycloma protein, a monoclonal antibody molecule, the binding specificity of which is usually unknown. By hybridizing
the continuous cell line with mouse spleen cells and screening the resulting hybrids for production of antibody against the injected antigen, one
effectively inserts the genes for the desired antibody into a continuous cell
line that has the necessary machinery to express these genes. A continuous supply of the desired genc product is thereby provided which, in this
case, is an antibody that binds the antigen with which the mouse was
immunized. The procedure can bc refined so that a variant of the plasmacytoma line that no longer produces its own myclorna protein is used
for hybridization and the resulting hybrid secretes only the antibody m0124c. Croce, D. Aden, and H. Koprowski, Proc. Natl. Aead. Sci. U.S.A. 72, 1397 (1975).
2~G. Kohler and C. Milstein, Nature (London) 256, 495 (1975).
2~G. Galfre, S. C. Howe, C. Milstein, G. W. Butcher, and J. C. Howard, Nature (London)
266, 550 (1977).
z7 N. Klinman and J. L. Press, Transplant. Rev. 24, 41 (1975).
350
[28]
ecules produced by the antibody genes introduced into the hybrid by the
spleen cell nucleus, zs The culture medium from the hybrids contains microgram per milliliter amounts of the specific antibody. If the "hybrid
tumor" is injected into mice that have been pristane primed 2a to encourage ascites fluid formation, the tumor will grow and produce the antibody
in milligram per milliliter amounts in the ascites fluid.
Such hybrids will provide a constant supply of monoclonal antibody
against antigens of choice. These antibodies can be used as reagents for
any procedures for which antibodies were previously used, but with the
added advantages of higher levels of discriminatiori, lower background,
and a continuously available supply. The other obvious application of the
procedure is that large amounts of homogeneous antibody against specific
antigens will be available for sequence analysis; using current techniques,
the nucleic acid for these specific antibody genes can be isolated from the
hybrid cell lines and characterized, s It will soon be possible to compare
the genes and gene products of several cell lines that produce antibodies
against the same antigenic determinant. One could even choose lines
making such antibodies on the basis of those types of antibodies that
appear early in mouse neonatal development and those that appear later.31
The analysis of antibody and gene structure could have significant implications in regard to the genetic mechanism involved in the generation of
antibody diversity.
The above system may be but a prototype for others in which a chosen
cell line is hybridized with a specific cell type that produces a desired
product; the "immortalized" primary cell type thereby provides a continuous hybrid line producing enough of the chosen product to allow its
use as a reagent or to make it available for biochemical analysis. Systems
of this sort may be applicable to the production of cell lines expressing
specific immunological T cell receptors, a2 hormone receptors, or such
other cell surface molecules as differentiation antigens or tumor antigens.
It may also provide the means of producing significant quantities of hormones and other cellular products.
Some general principles that may be applicable to other possible systems can be derived from the work with hybrid plasmacytomas. It is
important that the cell line used has the capacity to produce the desired
28 D. Margulies, N. M. Keuhl, and M. D. Scharff, Cell 8, 405 (1976).
29 M. Potter, Physiol Rev. 52, 631 (1972).
a0 S. Tonegawa, C. Brack, N. Hozumi, and R. Schuller, Proc. Natl. Acad. Sci. U.S.A. 74,
3518 (1977).
31 N. Klinman and J. L. Press, Fed. Proc., Fed. Am. Soc. Exp. Biol. 34, 47 (1975).
az R. Goldsby, B. Osborne, E. Simpson, and L. A. Herzenberg, Nature (London) 267, 709
(1977).
[28]
CELL FUSION
351
product. As indicated previously, hybridization of an antibody-producing

cell with another cell type such as a fibroblast usually results in the suppression or drastic reduction in the amount of differentiated product produced. Also, certain cell lines of the same type may hybridize better than
others. We have been working with two mouse plasmacytoma lines for
producing hybrids P3X63Ag83a and 45.6TGI.7. s4 Which of these cell lines
produces the most hybrids in a fusion depends on the source of primary
mouse cells used in the fusion. P3X63 Ag8 produces more hybrids when
fused with adult immunized spleen cells whereas 45.6TG1.7 produces
more hybrids when fused with stimulated neonatal spleen cells or cells
from in vitro stimulated monoclonal spleen fragments, a5 Some other plasmacytoma lines do not produce sufficient hybrids to be of practical use in
these experiments.
Another factor to consider is the metabolic state as well as the source
of primary cells used for t h e fusion. In our experience, neonatal spleen
cells give very few hybrids when fused with the above cell lines. However, when the spleen cells are stimulated in vitro with lipopolysaccharide
(LPS), a mouse B cell mitogen, for only a few hours the number of hybrids produced is increased dramatically. 35 This is consistent with a report that mitotic cells are fused preferentially when polyethyleneglycol
(PEG) is used as a fusing agent, a6 A similar phenomenon may explain
why a large proportion of hybrids produced by fusion of spleen cells
with a plasmacytoma line are making antibody against a recently
injected antigen. 25~a Thus the production of the desired hybrids will
depend upon selection of the proper combination of cell line and primary
cell.
Another factor that must be considered is that there must be a method
for selecting the hybrids and eliminating the parental nonhybrid population after cell fusion.
Hybrid
Selection Procedures
Littlefield 37 reported the first use of selective medium allowing the

growth of hybrid cells but killing the two parental cell types. This selection is based on the use of HAT medium, which was devised by Szybal~ki
33 G. Kohler, S. C. Howe, and C. Milstein, Fur. J. Immunol. 6, 292 (1976).

34 D. Margulies, N. Geplinski, B. Dhermgrongartama, M. L. Gefter, S. L. Morrison, T.
Kelley, and M. D. Scharff, Cold Spring Harbor Syrup. Quant. Biol. 41,781 (1977).
35 K. Denis, personal communication.
3~ D. Hansen and J. Startler, Somatic Cell Genet. 3, 471 (1977).
37 j. Littlefield, Science 145, 709 (1964).
352
[28]
e t al. 38 to kill cells that lack hypoxanthine-guanine phosphoribosyl trans-
ferase (HGPRT, EC 2.4.2.8.). The medium used by Littlefield contains

hypoxanthine (0.1 mM), thymidine (16/xM), and aminopterin (0.4/zM)
Aminopterin inhibits dihydrofolate reductase. Blocking the reductase inhibits d e n o v o synthesis of purines and pyrimidines. The medium also
must contain glycine (3/~M) because, in the presence of aminopterin, the
conversion of serine to glycine is blocked (the medium is sometimes designated as GHAT medium).
With d e n o v o synthesis of the nucleotides blocked, cells depend on the
transferase and thymidine kinase to utilize the hypoxanthine and thymine
in the medium. Only cells with both of these enzymes will grow in HAT
medium. Littlefield used two cell lines: one lacked thymidine kinase and
was selected by growth in the presence of the toxic thymidine analog,
5'-bromodeoxyuridine;39 the other lacked the transferase and thus was
resistant to the toxic analogs, 8-azaquanine or 6-thioquanine. 38 Both of
these parental cell types are killed in HAT medium whereas hybrids, i.e.,
cells containing both enzymes by complementation, will grow in HAT.
A modification of this selective system, the so-called half-selective
system, can be used: a cell line that does not grow in HAT is fused with a
cell type that grows slowly, e.g., a senescent human fibroblast, or does
not grow at all in culture, e.g., peripheral blood cells. Only the hybrids
form colonies in the HAT medium while the two parental cells either are
selected against or do not form colonies. A further modification allows the
fusion of HAT-sensitive cells with cells that normally grow in suspension
such as lymphoblastoid lines; hybrids are retained as attached colonies
and the parental cells that grow in suspension can be washed off from the
culture by frequent medium changes.
A similar medium (AA) contains alanosine, an antibiotic blocking conversion of inosine-5'-phosphate to adenosine-5'-phosphate and adenosine.
This medium will inhibit cells lacking adenine phosphoribosyltransferase. 4 The same system selects against cells lacking adenosine
kinase. 41 Puck e t al. have derived a auxotrophic mutant requiring an
amino acid or nucleotide that has been used in selective medium without
the required component. Such media allows growth of hybrids in which
genetic complementation takes place but inhibits the auxotrophic mutants. 42
as W. Szybalski, E. Szybalski, and G. Ragni, Natl. Cancer Inst., Monogr. 7, 75 (1962).
W. J. Rutler, R. L. Pictet, and P. W. Morris,Anna. Rev. Biochem. 42, 607 (1973).
40T. Kusano, C. Long, and H. Green, Proc. Natl. Acad. Sci. U.S.A. 68, 82 (1971).
41T. Chan, R. Creagen, and M. P. Reardon, Somatic Cell Genet. 4, 1 0978).
42F. T. Kao, R. Johnson, and T. T. Puck, Science 164, 312 (1970).
39
[28]
CELL FUSION
353
Another system of selection uses the drug ouabain 43as a specific inhibitor of the N a - K activated ATPase of the plasma membrane, the enzyme responsible for the active transport of K into the cell and the
extrusion of Na . Fusion of a cell that can be selected by HAT medium or
by its growth characteristics but that is not sensitive to a chosen concentration of ouabain, with another cell type that is sensitive to the same
concentration, allows selection of hybrid cells. Since human cells are
killed by low concentrations (10 -7 M) and mouse cell by higher concentrations (10 -a M) of ouabain this agent has become useful for selecting
mouse human hybrids.
One recently described selective medium, glucose-flee medium, results in selective growth of hepatoma cell lines producing gluconeogenic
enzymes.44 Variations of the medium select for hybrid clones reexpressing
one or more of these enzymes from hybrid lines in which liver-specific
functions were extinguished in the initial hybridization.
It should be clear that many selective systems are potentially available. One is limited only by the ability to derive cell lines that are differentially sensitive to metabolic inhibitors or differing in metabolic substrate
requirements.
Selection procedures are usually applied to hybrid cells 24-48 hr after
fusion to assure expression of the functions necessary for survival although it is not evident from our experience that this delay is really necessary when HAT medium is used. It is important in selective media such as
HAT, media in which cells must grow to be selected against, that the
fused cells should have enough surface area to divide 2-3 times after the
selective medium is added. Plating of the cells at too high a density will
slow the growth and increase the time that the cultures must be handled
before hybrid clones are isolated.
Fusion Protocols
Fusion with Sendai virus or polyethylene glycol can be done with cells
in mixed monolayers or with cells in suspension. In both cases it is essential that healthy cells with.high viability be used for fusion. When cells are
removed from a plastic or glass surface in order to fuse in suspension the
high viability should be established by phase microscopy or dye exclusion
after the cells are in suspension. During the washing and fusion procedures, care should be taken to avoid extremes in pH (maintain 7.2-7.6).
This is particularly important with Sendai virus fusions that involve long
43L. H. Thompsonand R. M. Baker, Methods Cell Biol. 6, 209 (1973).
44R. Bertolotti,Somatic Cell Genet. 3, 579 (1977).
354
[9-8]
incubation periods w i t h large n u m b e r s of metabolizing cells in a small

volume of medium.
After cells are fused they should be treated as any cells with which one
desires to obtain m a x i m u m cloning efficiency. Actually, the hybrids are
being cloned at a v e r y low cell density b e c a u s e selective conditions will
often quickly kill the parental cell types. T h e r e f o r e one should use a rich
medium with 20% fetal calf s e r u m and avoid fluctuations in the p H of
cultures. It is helpful to use organic buffers such as H E P E S to maintain
the p H and thereby avoid the p H fluctuations of bicarbonate-buffered
media. 45,46
Fusions with Sendai Virus (Parainfluenza I)
The efficiency with which preparations o f Sendai virus p r o m o t e cell
fusion is variable. A given b a t c h should be tested b y fusing two cell types
that are k n o w n to fuse well. Virus can be obtained f r o m Microbiological
Associates, B e t h e s d a , Maryland. W h e n a good virus preparation is identiffed it should b e dispensed as concentrated allantoic fluid and stored in
liquid nitrogen or at least at - 7 0 . It can be thawed as needed and inactivated with ultraviolet light 4r or/3 propiolactone. 4s In our experience inactivated virus can be refrozen at least once without significantly reducing
its fusing efficiency. It is clear that Sendai virus will p r o m o t e the fusion o f
some cell t y p e s and not others. F o r instance, Scndai virus has not increased the a m o u n t o f fusion a b o v e that of the spontaneous fusion frequency w h e n m o u s e limploid cell lines are used. 49
Because polyethylene glycol can p r o m o t e fusion o f a wider variety of
cell types and is m u c h easier to obtain than an effective preparation of
Sendai virus, it will soon replace Sendal virus as a fusing agent. Therefore,
fusion methods with P E G one described in greater detail whereas Sendai
virus fusion techniques are presented in outline.
Sendai Virus Fusion
K l e b e et al. 50 and Giles and Ruddle 51 describe the p r o c e d u r e s for virus
preparation and mixed m o n o l a y e r fusions and for fusion of cells in suspension. I describe here the p r o c e d u r e used in m y l a b o r a t o r y for fusion of
4s C. Ceccarino and H. Eagle, Proc. Natl. Acad. Sci. U.S.A. 68, 229 (1978).
4e C. Croce, H. Koprowski, and H. Eagle, Proc. Natl. Acad. Sci. U.S.A. 69, 1953 (1972).
~r G. Yergarian and M. B. Nell, Proc. Natl. Acad. Sci. U.S.A. 55, 1066 (1966).
4s j. M. Neff and J. F. Enders, Proc. Soc. Exp. Biol. Med. 127, 260 (19685.
49D. H. Margulies, N. M. Kuehl, and M. D. Scharff, Cell 8, 405 (1976).
50R. J. Klebe, T. Chen, and F. H. Ruddle, J. Cell Biol. 45, 74 (1970).
51R. E. Giles and F. H. Ruddle, in "Tissue Culture: Methods and Applications" (P. F.
Kruse, Jr. and M. K. Patterson, eds.), p. ,*75. Academic Press, New York, 1973.
[28]
CELL FUSION
355
cells in suspension, a method that is more generally applicable than

monolayer fusion because it can be used with cells that grow in suspension as well as those in monolayers. When the precautions of maintaining
proper pH control, using healthy cells, and plating so as to allow rapid
selection mentioned above are taken, this method yields more hybrid
clones than does a monolayer fusion with the same cell types. 5z We use
modified Eagle's medium (MEM) with nonessential amino acids, Dulbecco's MEM with high glucose (4.5 g/liter), or RPMI-1640 medium, depending on which of these media are routinely used to grow the cells to be
fused.
1. Cells are grown for at least 1 day in medium containing 20% serum
that is buffered with 15 mM HEPES at pH 7.6. 45
2. The fusion is done in Hanks's balanced salt solution without glucose, adjus~ted to pH 7.6 with Tris HC1 (10 raM) (Modified Hanks,
MH).53
3. Harvest cells and wash twice in MH by centrifugation. Centrifugations are done in IEC MS centrifuge (radius: 15 cm) at 1000 rpm for 5 min
at room temperature. The cells may be mixed in appropriate ratio and
washed together. Use 1 : 1 ratio unless it is evident that one cell type fuses
better than the other. The 1 : 1 ratio may be used for initial experiments,
and if this does not provide sufficient hybrid clones, other ratios should be
tried.
4. Mix 3 x 106 of each cell type in a round-bottomed tube (Falcon
2001) in 1 ml of MH.
5. Chill on ice. 54
6. Add 1000 hemagglutination units (HAU) of inactivated virus in 0.5
ml MH.
7. Maintain on ice at pH 7.6 for 5 min or until clumped.
8. Place tube into 37 water bath for 5 min.
9. Add 0.3 ml of dialyzed fetal calf serum. The pH of this should be
adjusted so that addition of 0.3-1.5 ml of MH does not change pH from
7.6.
10. Incubate tube at 37 by rocking gently for 30 min.
11. Add medium with 20% serum and gently dispense into 20 35-mm
petri dishes or other appropriate culture vessels. One should consider that
separation into several separate plates will assure that colonies isolated
from each plate arise from separate fusion events.
12. Change medium to HAT medium on the next day. Littlefield's
original concentration37 of HAT constituents has worked satisfactorily in
R. Kennett, unpublished observations.
53 y. Okada and J. Tadokoro, Exp. Cell Res. 32, 117 (1963).
54 There is an indication that although this is a routine procedure it is really unnecessary [L.
Donner and L. Turek, Somatic Cell Genet. 2, 77 (1976)].
52
356
[9-8]
all H A T selections we h a v e done. Change to flesh m e d i u m 2-3 times per

week.
13. Colonies usually a p p e a r macroscopically b e t w e e n 10 and 20 days.
14. Pick individual colonies into 35-mm petri dishes and grow large
enough quantities so that stocks of several ampules, containing at least 106
cells each, can be frozen in 95% s e r u m - 5 % dimethyl sulfoxide
(DMSO).53"55
Fusion with P o l y e t h y l e n e Glycol (PEG)
Polyethylene glycol, which had been used to p r o m o t e fusion o f plant
protoplast,s, was applied to fusion of m a m m a l i a n cells by Pontecorvo. 5s
Davidson and Gerald modified conditions to include a shorter exposure to
P E G and fusion of cells in suspension. 5r'Ss O ' M a l l e y and Davidson found
that the frequency of hybridization of cells that are not attached to
monolayers could be increased if the cells to be fused were pelleted onto a
coverslip before e x p o s u r e to PEG.59 Other investigators have described
the effectiveness of fusion of P E G s of different molecular weights, concentrations, t e m p e r a t u r e s , and fusion conditions, s-62 N o r w o o d e t al. have
shown that e x p o s u r e of cells to D M S O increases P E G - m e d i a t e d fusion, s3
Care must be taken not to e x p o s e cells to D M S O in medium containing
H E P E S since D M S O will allow the buffer to gain entrance into the cells
with resultant toxicity.
The a m o u n t o f e x p o s u r e that a cell can tolerate depends on the molecular weight: and concentration of PEG. sz,.~8T h e s e factors m a y have to be
adjusted for each type of cell in order to maximize the fusion/toxicity
effects o f P E G treatment. Davidson et al. discuss the effects of these
factors.5r,ss
The method used here has been effective for several types of cells. The
p r o c e d u r e s are similar to those reported b y Gefter e t al. 64 for fusion of
mouse plasmacytomas.
F o r fusion o f certain types of cells it m a y be necessary to increase the
P E G concentration to 50%, and the addition of D M S O m a y increase the
55w. s. Sly, G. Sekhon, R. Kennett, W. F. Bodmer, and J. Bodmer, Tissue Antigens 7, 165
(1976).
56G. Pontecorvo, Somatic Cell Genet. 1,397 (1975).
s7 R. L. Davidson and P. S. Gerlad, Somatic Cell Genet. 2, 165 (1976).
58R. L. Davidson, K. O'Malley, and T. B. Wheeler, Somatic Cell Genet. 2, 271 (1976).
59K. O'Malley and R. L. Davidson, Somatic Cell Genet. 3, 441 (1976).
e0 A. Hales, Somatic Cell Genet. 3, 227 (1977).
6~V. L. Vaughn, D. Hensen, and J. Stadler, Somatic Cell Genet. 2, 537 (1976).
e2 Z. Steplewski, H. Koprowski, and A. Leibovitz, Somatic Cell Genet. 2, 559 (1976).
63T. H. Norwood, C. J. Zeigler, and G. M. Martin, Somatic Cell Genet. 2, 263, (1976).
64 M. L. Gefter, D. Margulies, and M. D. Scharff, Somatic Cell Genet. 3, 231 (1977).
[28]
CELL FUSION
357
fusion efficiency.6a With increased PEG concentration, the time of exposure will probably have to be decreased to avoid toxicity. It is our experience that the centrifugation step does increase the frequency although we
have fused several combinations of cells simply by exposing the loosened
pellet of cells to 1 ml of 50% PEG 1000 or PEG 6000, mixing the pellet
with PEG solution by gently rotating the tube, and, after 1-2 min of
exposure, adding 10 ml of medium without serum taking care that the
PEG is mixed and diluted with the medium.
The ratio of cells fused may vary from 1 : 1 for two cell lines to 5-10 : 1
when spleen cells or other primary cells that do not divide in culture are
fused with a cell line. In the cases in which 5-10 : 1 spleen cells : cell line
are used, it is clear that the unfused spleen cells act as a feeder layer and
increase the cloning efficiency of the hybrid cells. When we fused human
peripheral blood cells with mouse plasmacytoma cell lines the number of
hybrid clones derived was increased significantly by adding 1000 irradiated (4500 R) human diploid fibroblasts per microwell (Linbro FB
96TC) as a feeder layer.
The procedure below is used for fusion of plasmacytoma lines, cells
that are killed by HAT medium, with spleen cells from immunized
mice. 2~'65 We have also used it for making mouse x human hybrids with
several combinations of cell types. Spleen cells for the fusion are removed
3-4 days after the mouse is immunized. The spleen is placed in a 60-mm
petri dish in tissue culture medium with serum and is perfused with medium by injecting with a 26-gauge needle at several sites thereby forcing
medium into the spleen. This is continued until most of the cells are
removed; the number of spleen lymphocytes recovered is usually 5-10
107 per organ. Care must be taken not to repeatedly draw the suspended
cells up into the syringe and force them through the needle. Remove the
cells to a centrifuge tube, pellet them (1000 rpm, IEC MS), remove the
supernatant liquid, and suspend them in 5 ml of cold 0.17M NIL C1. Keep
the tube on ice for 10 min to lyse erythrocytes and then add 10 ml of cold
medium with 20% serum. Pellet the cells, count and check viability, and
mix with 107 of the plasmacytoma cell line.
1. Wash the cells in medium free of serum by centrifugation in a
round-bottomed tube (Falcon 2001).
2. Remove all the supernatant liquid by suction and loosen the pellet
by tapping the tube.
3. Add 0.2 ml of 30% PEG in medium without serum. The 30% PEG is
made by adding 3 ml of PEG 1000 at 41 to 7 ml of medium without serum
at 41 . PEG tends to lower the pH of the medium which should be ad65 R. H. Kennett, K. A. Denis, A. S. Tung, and N. R. Klinman, Current Topics
in Microbiol. and Immunol. 81, 77 (1978).
358
[28]
justed to approximately 7.6 with NaOH. After mixing the PEG solution it
is kept at 37 until used.
4. The cells are maintained in the 30% PEG for 8 min. During the
8-rain period, cells are pelleted in the 30% PEG (1000 rpm, IEC MS,
radius 35 cm). The centrifugation time should be 3-6 min. It is our impression that the longer the centrifugation, the more hybrids. The limitation,
as indicated above, is the toxic effect on the particular cells used. This
may be monitored by phase microscopy or trypan blue exclusion. At the
end of 8 min the PEG is diluted with 5 ml of medium without serum and
then 5 ml of medium with 20% serum is added.
5. The cells, in diluted PEG, are pelleted and resuspended in 30 ml of
culture medium with serum. We have found the following medium optimal
for cloning and growth of hybrid plasmacytoma lines:
Dulbecco's MEM with high glucose (4.5 g/liter) to which are added
10% NCTC 109 medium,2S 20% fetal calf serum, 0.150 mg/ml oxaloacetate, 0.050 mg/ml pyruvate, 0.200 units/ml bovine insulin (Sigma),
and antibiotic of choice.
In addition to the purine and pyrimidine bases present in the NCTC 109
medium, thymidine (16 ~M) and hypoxanthine (0.1 mM) are added to the
medium in which the fused cells are plated.
6. The 30 ml of cells are evenly suspended and gently distributed into
6 microplates (Linbro FB96TC), 1 drop (about 50/zl) per well. The next
day, an additional drop of the above medium with aminopterin (0.8/zM,
twice Littlefield's concentration) is added to make HAT selective
medium.
7. The wells are fed two additional drops of medium 6-7 days later;
clones appear macroscopically within 2 weeks.
8. If necessary, the clones may be fed weekly by removing most of the
medium in the well and replacing it with fresh medium.
9. In most fusions, clones arise in nearly all the wells of the six microplates. The valuable wells are identified by screening the supernatant
liquid for production of the desired antibody, and these clones are transferred to larger wells (Linbro FB16-24-TC) with not more than 0.5 ml of
medium.
10. At early stages, the clones must be watched carefully so that they
are not allowed to grow to too high a density nor diluted too sparsely.
11. When the cells have been transferred into four of the larger
wells, they may be transferred to flasks, grown, and stocks prepared and
frozen.
12. The frequency of hybrids usually requires that the cells be cloned
in agarose over a feeder layer of human fibroblasts to be sure that the cells
[29]
FUSION OF HIGHER PLANT PROTOPLASTS
359
are truly clonally derived. We have often cloned the cells from the original
well into two cloning plates successfully. -66
13. If the reversion rate of the parental lines is not known, control
cells should be put under selective conditions. For those cells selected
against a HAT medium it is wise to grow the parental cells in the appropriate selective drug, 5-bromodeoxyuridine or 8-azaguanine, before the fusion. The hybrid nature of the clones can of course be assessed by testing
for expression of parental enzymes, antigens, or chromosomes? 1
In addition to being used to fuse cells to form heterokaryons and
hybrids, the fusion methods with Sendai virus and PEG also have been
used recently to incorporate nuclei into enucleated cells, TM to transfer a
few chromosomes that are enclosed in microceUs into other cells, s7,ss and
to transfer strbstances enclosed in lipid vesicles e9 or erythrocyte ghosts TM
into cells. These techniques are likely to become important and useful
applications of membrane fusion techniques in the future.
Acknowledgments
The author is supported by NIH Grant CA 18930, NSF Grant PCM 76-82997, and the
Cystic Fibrosis Foundation. I thank Kathleen Denis for her discussions a n d work on hybridizations of mouse plasmacytomas with mouse neonatal cells and spleen fragments.
K. Sato, R. Slesinski, and J. Littlefield, Proc. Natl. Acad. Sci. U.S.A. 69, 1244 (1972).
67 R. E. K. Fournier and F. H. Ruddle, Proc. Natl. Acad. Sci. U.S.A. 74, 3937 (1977).
6s R. E. K. Fournier and F. H. Ruddle, Proc. Natl. Acad. Sci. U.S.A. 74, 319 (1977).
69 G. Poste, D. Papahadjopoulos, and W. J. Vail, Methods Cell Biol. 14, 34 (1976).
70 M. Furusawa, M. Yamaizumi, T. Nishimura, T. Uchida, and Y. Okoda, Methods Cell
Biol. 14, 73 (1976).
[29] F u s i o n o f H i g h e r P l a n t P r o t o p l a s t s
B y A L B E R T W . RUESINK
For many years the higher plant cell wall prevented any significant
research utilizing the fusion of higher plant cells, but the development of
techniques for removing the cell walls to leave behind li~,ing protoplasts 1
has recently provided exciting opportunities for research on genetics, development, and crop modification. In contrast to animal cells, plant cells
are often totipotent, and whole plants can be regenerated from individual
cells, from isolated protoplasts, or even from the product of the fusion of
two protoplasts, 2 which is called a somatic hybrid. Though the route to
1 A. W. Ruesink, this series, in press, (1979).
2 p. S. Carlson, H. H. Smith, and R. D. Dearing, Proc. Natl. Acad. Sci. U.S.A. 69, 2292
(1972).

ISBN 0-12-181958-2
360
[29]
useful genetic engineering by means of protoplast fusion will surely be

long and difficult, a tremendous potential exists. Both the problems and
possibilities have been reviewed elsewhere. 3-5 Because of the difficulty of
controlling the nature of the fusion product when one fuses whole cells,
another promising approach to genetic manipulation involves various
methods being developed to transfer only a selected part of one cell's
genome into another. 6 However, isolated DNA is very often degraded in
transit in contrast to the minimal genetic modification that occurs during
the process of protoplast fusion. In addition to providing a novel means
for studying nuclear genes, the fusion of protoplasts also permits a new
approach to the study of cytoplasmic genetic factors. 7
An appropriate fusion process will yield up to 35% of the cells in a
preparation as fusion products. Many different species have been successfully fused. Though the fusion itself usually is readily achieved, the subsequent selection and regeneration of useful progeny are not. This article
will emphasize the fusion process and suggest strategies for dealing with
the later processes needed for genetic engineering. Two short general
articles on fusion already exist, one in a handbook on tissue culture methodology 8 and the other in a symposium volume, a
Protoplast Release and Preparation for Fusion

Protoplasts are readily released from many different plants by an array
of commercially available enzymes. 1 Leaf mesophyll cells (the green
photosynthesizing cells within the leaf) and suspension-cultured cells release protoplasts most readily, but almost any cell type that lacks thick
secondary cell walls has at one time or another been used to prepare
protoplasts. Although the osmotic strength of the solution surrounding
intact plant cells is usually of small concern, once the cell wall is
weakened or removed the osmotic strength of the medium is critical.
Usually a solution of mannitol or a similar compound at a concentration
3 p. S. Carlson and J. C. Potacco, Science 188, 622 (1975).
4 p. R. Day, Science 197, 1334 (1977).
5 H. H. Smith, BioScience 24, 269 (1974).
6 D. Hess, in "Microbial and Plant Protoplasts" (J. F. Peberdy et al., eds.), p. 125. Academic Press, New York, 1976.
r G. Belliard, G. Pelletier, and M. Ferault, C. R. H e b d . Seances A c a d . Sci., Ser. D. 284,
749 (1977).
s K. N. Kao, in "Plant Tissue Culture Methods" (O. L. Gamborg and L. R. Wetter, eds.),
p. 22. Nat. Res. Counc. Can., Ottawa, 1975.
a T. Eriksson, H. Bonnett, K. Glimelius, and A. Wallin, in "Tissue Culture and Plant
Science 1974" (H. E. Street, ed.), p. 213. Academic Press, New York, 1974.
[29]
361
of about 0.5 M effectively stabilizes protoplasts. Sometimes protoplast

viability is enhanced by the additional presence of a complete nutrient
medium.
A clean preparation of healthy protoplasts is required for fusion studies. The health of protoplasts can be monitored by measuring the number
showing normal cyclosis, 1 by using vital dyes, 11 or by determining the
fraction of the protoplasts that survive subsequent culture. To remove
protoplasts from undigested tissue, from the digesting enzymes, and from
various debris, the preparation first can be poured gently through a nylon
net or stainless-steel screen with a pore size of 75-300/zm. Several subsequent washings with fresh osmoticum will remove both the enzymes used
to release the protoplasts and most cytoplasmic debris from any cells that
have broken, although the fusion capability may decrease quickly following removal of the enzymes as will be described. To separate good protoplasts from cell wall fragments and other particulate debris, several kinds
of density gradients may be used I although such a separation is usually
unnecessary for fusion studies. Being decidely fragile, protoplasts must
be handled gently throughout the isolation procedures and must never be
centrifuged in excess of 500 g; usually about 100 g is better.
Protoplasts are not suitable for fusion if they have already formed a
new wall. The best way to prevent wall regeneration is simply to work
quickly enough so that a wall has no time to form. In one case, the
available working time was as short as 10 min after removing the enzyme,
at which time the fusion rate of pea mesophyll and Vicia suspension culture protoplasts had started to decrease, a decrease that correlated with
the formation of new wall.12 Though workers apparently do not use them
for fusion studies, two methods do exist for reducing the rate of wall
regeneration--the use of an ionic osmoticum 13and the use of high osmotic
strength, a
The Fusion Process
Three main methods for promoting protoplast fusion have been reported, but the most recent one is so superior to the earlier ones that the
first two will be mentioned only briefly.
The first successful method used NaNO3 as the fusing agent. 14 ProtolO A. W. Ruesink and K. V. Thimann, Proc. Natl. Acad. Sci. U.S.A. 54, 56 (1965).
11 j. M. Widholm, Stain Technol. 47, 189 (1972).
1~ G. Weber, F. Constabel, F. Williamson, L. Fowke, and O. L. Gamborg, Z. Pflanzenphysiol. 79, 459 (1976).
1~ R. K. Horine and A. W. Ruesink, Plant Physiol. 50, 438 (1972).
14 j. B. Power, S. E. Cummins, and E. C. Cocking, Nature (London) 225, 1016 (1970).
362
[29]
plasts were prepared and washed in a sucrose osmoticum. As soon as they

were placed into 0.25 M NaNO3 the fusion process began. Centrifugation
of the protoplasts in the NaNOz promoted fusion, but the number of
protoplasts fusing remained low. 2 Moreover, protoplasts treated with this
salt are markedly altered in their uptake capabilities. 15
The second method utilized 0.05 M CaC12 in 0.4 M mannitol with the
pH at 10.5 (0.05 M glycine-NaOH buffer) and the temperature at 37o.16
When protoplasts of tobacco leaves were centrifuged in this medium, they
aggregated at once and started to fuse in 10 min. By 40 min, 20-50% of the
protoplasts were involved in fusion events. Many were still viable and
could readily be cultured, although very large fusion products became
unstable and disintegrated. Magnesium did not substitute for the calcium
requirement.
Essentially all present work on plant protoplast fusion is carried out
with polyethylene glycol (PEG) as the fusing agent, an agent that will also
induce fusions of animal cells even though the other agents that have been
used successfully to fuse animal cells seem not to work with plant protoplasfs. Two laboratories did the early work with the PEG method.If'Is The
following method is one that gives good results with many kinds of
protoplasts.
Release protoplasts from two kinds of tissues in separate containers,
planning the wall digestions so that the releases are completed simultaneously. Obviously the two tissues should have some characteristic that
permits the detection of hybrid protoplasts either visually or with appropriate selection media. The nature of the stabilizing osmoticum is not too
critical as far as fusion itself is concerned, but is important to the stability
of the protoplasts. Mix the protoplasts together and pour them through an
80-/zm screen. Centrifuge for 3 min at 100 g to sediment the protoplasts.
Remove the supernatant. Wash the protoplasts with 10 ml of a solution
such as one of the following:
1. 0.5 M glucose, 3 mM CaCI2, 0.7 mM KH2PO4 (pH 5.9)
2. nutrient salts with 0.1 M sucrose, 0.25 M sorbitol
Centrifuge as above, remove the supernatant liquid, and resuspend in
the same medium to give a 6% (v/v) suspension of cells. Proceed immediately to the following fusion process:
1. Place 0.2 ml of suspension onto a coverslip in the bottom of a small
petri dish and cover the dish.
2. Let the protoplasts settle for 5 min.
is A. W. Ruesink, Physiol. Plant. 44, 48 (1978).
le W. A. Keller and G. Melchers, Z. Naturforsch., Teil C 28, 737 (1973).
17 K. N. Kao and M. R. Michayluk, Plants 115, 355 (1974).
is A. Wallin, K. Glimelius, and T. Eriksson, Z. Pflanzenphysiol. 74, 64 (1974).
[29]
363
3. Add 0.6 ml of PEG solution drop by drop. (Make the PEG solution
by dissolving 1 g of polyethylene glycol (mol wt 1500) in 2 ml of 0.1 M
glucose, 10 mM CaCI2, and 0.7 mM KH~PO,.) PEG sources: PEG 1540,
Polysciences, Inc., Warrington, Pennsylvania; MODO-PEG 1500: Modokemi, Stenungssund, Sweden.
4. Incubate at room temperature for 40 min. Occasional rocking of the
petri dish should prove helpful if the protoplasts are not touching each
other.
5. Gently add 0.5 ml of protoplast culture medium that should contain
about 5 mM Ca. A pH of 5.7 is usually used, but a higher pH, up to 10,
will enhance the number of protoplasts fusing.
6. After 10 min, add another 1.0 ml of medium.
7. Wash the preparation several times with 2-ml aliquots of culture
medium, using a Pasteur pipette to remove the old medium each time. Be
sure to leave a thin layer of medium over the protoplasts at all times.
8. Culture the protoplasts or study them microscopically while they
are still on the coverslip. Alternatively, gently wash them free from the
glass and embed them in soft agar (0.6%, procool/ed before adding protoplasts). TM An optimal arrangement is to embed the protoplasts in a 1-mm
thick layer on top of a previously solidified 5-mm layer.
Occasionally, protoplasts will fuse spontaneously as they are being
released from tissue, apparently by expansion of the plasmodesmata linking cells together in the tissue. 2 Such spontaneous fusions are rare and
are of little concern to workers interested in interspecific fusions. The
method described above will produce up to 35% interspecific hybrids in a
preparation. The rest are unfused cells and intraspecific fusion products.
Although fusion products resulting from the fusion of just two protoplasts
predominate, the products of triple, quadruple, and multiple fusions are
also present. Very large cells resulting from the fusion of many protoplasts
tend to be unstable and hence are typically present only in small numbers.
Both the molecular weight and the concentration of PEG are critical to
inducing successful fusions. Polyethylene glycols of less than 1000 mol wt
are not able to produce tight adhesion, and those of up to 6000 are more
and more active per mole in inducing fusions but produce too viscous a
solution for easy handling. The choice of 1500 is a good compromise. The
concentration used above (about 0.33M) provides maximal fusions, with
significantly less fusion at both lower and higher concentrationsY 1
A distinction must be made between aggregation, where two or more
protoplasts just stick to each other, and true fusion, where mixing of
lg I. Takebe, G. Labib, and G. Melchers, Naturwissenschaften 58, 318 (1971).
20 L. A. Withers and E. C. Cocking, J. Cell Sci. 11, 59 (1972).
zl K. N. Kao, F. Constable, M. R. Michayluk, and O. L. Gamborg,Planta 120, 215 (1974).
364
[29]
cytoplasm occurs. Treatment with g ~ l a t i n , 22 antisera, 2s or Concanavalin

A 24 causes only aggregation and does not promote fusion. Only by careful
microscopic work can one ascertain whether there are aggregations or
fusions.
The nature of the change at the membrane surface that leads to fusions
is not totally clear. With sodium nitrate, some electrophoretic evidence
suggests that the elimination of negative charges on the protoplast surface
is a key aspect. 25'26 The effect of PEG is even less clear. The PEG may
provide a bridge by which Ca 2+ can link membrane surfaces together, or it
may somehow lead to a disturbance in the surface charge during PEG
washout. 21 On the other hand, certain concentrations of PEG are known
to cause~a phase separation, 27 and such a separation at the plasma membrane surface could conceivably alter its potential for fusion.
Brief mention must be made of other methods of inducing protoplast
fusions. When protoplasts are formed from cultured meiotic cells, they
fuse very readily. Merely tapping a depression slide containing them is
enough to cause considerable fusion. 28 Treatment with sea water or
lysozyme somewhat promotes fusions in petunia. 29
Hybrid Fusions
M u c h of the interest in plant cellfusion resultsfrom its abilityto bring
together new combinations of genetic information. The following tabulation liststhe cases where hybrid fusions have occurred and indicates that
usually P E G was the fusion-inducing agent. Often, some cell divisions
subsequently occurred. In only two cases, however, has it been possible
to regenerate a hybrid plant.
Considerable attention has been devoted to means by which hybrid
fusions can be detected. One important way is the use of selectivegrowth
media as will be described below, but microscopic methods are also used.
Workers have looked for: nucleus size differences,3 nucleus staining
22 T. Kameya, Planta 115, 77 (1973).
2a j. Burgess and E. N. Fleming, Planta 118, 183 (1974).
24 K. Glimelius, A. Wallin, and T. Eriksson, Physiol. Plant 31, 225 (1974).
25 A. W. Ruesink, Plant Physiol. 47, 192 (1971).
26 B. W. W. Grout and R. H. A. Coutts, Plant Sci. Lett. 2, 397 (1974).
2r R. Kanai and G. E~ Edwards, Plant Physiol. 52, 484 (1973).
2s M. Ito, Plant Cell Physiol. 14, 865 (1973).
29 H. Binding, Z. Pflanzenphysiol. 72, ,122 (1974).
3o C. W. Jones, I. A. Mastrangelo, H. H. Smith, H. Z. Liu, and R. A. Meek, Science 193, 401
(1976).
INTERSPECIFIC FUSION OF HIGHER PLANT PROTOPLASTS

Species
Fusion method
Maize oat roots

Torenia baillonii T. fournieri petals
Nicotiana glauca N. langsdor~i
mesophyll
Vicia culture x pea mesophyll
NaNO3
NaNO3
Unreported
Unreported
NaNOa
PEG
& Ca
PEG
& Ca
PEG
& Ca
PEG
& Ca
PEG
& Ca
PEG
& Ca
PEG
& Ca
PEG
& Ca
PEG
& Ca
PEG
& Ca
PEG
PEG
PEG
Hybrid plant
Some cell divisions
Soybean culture x barley mesophyll

Soybean culture Vicia mesophyll
Soybean culture corn mesophyll
Soybean culture x rapeseed
mesophyll
Soybean culture pea mesophyll
(2 vars.)
Soybean culture tobacco mesophyll
(4 vars.)
Soybean culture x Colchium
mesophyll
Tobacco mesophyll HeLa culture
Petunia hybrida mesophyll

P. parodii mesophyll
Carrot petunia
Petunia Atropa
Tobacco mesophyll x chicken red cell
Fate
Some cell divisions
Ref.
a
b
Some cell divisions

Some cell divisions
Colonies up to 10cells
Some cell divisions
Some cell divisions
Some cell divisions
Nuclei cohabit cells
up to 6 days
Hybrid plant
Small colonies
Small colonies
Mixed cytoplasm;
intermediate cell
surface structure
J
k
t
J. B. Power, S. E. Cummins, and E. C, Cocking, Nature (London) 225, 1016 (1970).

hi. Potrykus, Nature (London), New Biol. 231, 57 (1971).
cp. S. Carlson, H. H. Smith, and R. D. Dearing, Proc. Natl. Acad. Sci. U.S.A. 69,
2292 (1972).
dK. N. Kao and M. R. Michayluk, Planta 115, 355 (1974).
eK. N. Kao, F. Constabel, M. R. Michayluk, and O. L. Gamborg, Planta 120, 215
(1974).
SK. K. Kartha, O. L. Gamborg, F. Constabel, and K. N. Kao, Can. J. But. 52, 2435
(1974).
g F. Constabel, G. Weber, J. W. Kirkpatrick, and K. Pahl, Z. Pflanzenphysiol. 79, 1
(1976).
h C. W. Jones, I. A. Mastrangelo, H. H. Smith, H. Z. Liu, and R. A. Meck, Science 193,
401 (1976).
~J. B. Power, S. F. Berry, E. M. Frearson, and E. C. Cocking, Plant Sci. Lett. 10, 1
(1977).
J J. Reinert and G. Gosch, Naturwissenschaften 63, 534 (1976).
k G. Gosch and J. Reinert, Naturwissenschaften 63, 534 (1976).
ZG. E. Willis, J. X. Hartmann, and E. D. deLamater, Protoplasma 91, i (1977).
366
[29]
properties, 31pigment differences,32 plastid differences,33 chromosome size

distinctions,~ isozyme production, 35 and the tritiation of one type of
nuclei.3
Once the cells fuse, nuclei can fuse in either or both of two ways: (1)
by a fusion of the nuclear membranes and a direct mixing of the interphase chromatin, and (2) by synchronous division of two nuclei in such a
manner that chromosomes become intermingled and the two daughter
nuclei each receive a full set of chromosomes from each species. The
relative frequency of these two fusion methods is not yet clear. 31 It is
possible that better development of hybrid cells may occur where the two
parent lines have similar generation times. However, in one report in
which two fused species had division times of 1 day in one case and over 5
days in the other, the hybrid cells divided rather uniformly at day 2 or 3. 36
Interspecific hybrid cells have been shown to ~etain their hybrid character
for more than 4 weeks in culture,--for five or six generations--but
throughout that time the chromosomes of the two species entered
metaphase with a slight asynchrony and assembled in groups rather than
as a true cell plate. 3r
Development Following Protoplast Fusion
For most research needs, the limiting factor is not the fusion itself, but
rather what is done with that initial fusion product. The demonstration
that a somatic hybrid produced by fusing protoplasts o f N i c o t i a n a glauca
and N i c o t i a n a langsdorfii could be both selected from other protoplasts
and regenerated into an intact plant 2 was convincing proof that significant
work with fusion products is possible. In that case, ~he selection involved
plating all protoplasts on a nutrient medium whose hormone content was
such that none of the parent protoplasts would multiply, a method that
unfortunately does not have general applicability.
Once protoplasts from two species have been fused, there are four
subsequent problems to be solved when using protoplasts for development studies or genetic engineering. (1) The desired fusion product must
be selected from among cells that are not the desired hybrid. (2) The
fusion product must be induced to divide. (3) The desired portion of the
3~F. Constabel, D. Dudits, O. L. Gamborg, and K. N. Kao, Can. J. Bot. 53, 2092 (1975).
32I. Potrykus,Nature (London), New Biol. 231, 57 (1971).
aa K. K. Kartha,O. L. Gamborg,F. Constable,and K. N. Kao,Can. J. Bot. 52, 2435(1974).
34G. Gosch and J. Reinert, Naturwissenschaften 63, 534 (1976).
35L. R. Wetterand K. N. Kao, Z. Pflanzenphysiol. 80, 455 (1976).
as F. Constabel,G. Weber,J..W. Kirkpatrick,and K. Pahl,Z. Pflanzenphysiol. 79, 1(1976).
3r F. Constabel,G. Weber, and J. W. Kirkpatrick,C. R. Hebd. Seances Acad. Sci., Ser. D
285, 319 (1977).
[29]
367
genome from each parent plant must be selected and maintained in the
culture. (4) An intact plant must be regenerated. Significant progress has
been made on all problems but number 3.
To select a desired fusion product, the genetic complementation of
auxotrophic mutants has been used with such systems as Neurospora 3s
and liverworts. 39 So far, it has not been successful with higher plant cell
fusions, apparently because of cross-feeding between cells that allows
parent cells to grow up. 2 Differential drug sensitivity of parent species has
recently shown promise as a selection device in a petunia system, 4 and
related work indicates that complementation, bringing green coloration
into albino tissue culture cells, may prove useful. 4~An alternative method
has been to use semilethal, recessive chlorophyll-deficient mutants that
complement in diploids. 42
To procure some cell divisions from protoplast fusion products is not a
major problem as can be seen from the table. However, the appropriate
culture conditions for maintainance of continued proliferation of the cells
have not been worked out. An equally difficult problem has been the
regeneration of whole plants from protoplasts. Whole plants can be regenerated from tobacco protoplasts more readily than from those of most
species, although there is no intrinsic reason against regeneration of plants
from the protoplasts of many other species. Only six plants have yielded
to efforts to obtain whole plants from protoplasts: tobacco, TM c a r r o t , 4a
petunia,44 asparagus,45Atropa, 46and rapeseed. 47 A number of laboratories
that want to perform genetic engineering work are now focusing their
primary attentions on the elucidation of basic information needed to convert tissue-cultured cells, particularly of crop plants, into differentiated
plants; these researchers realize that a much better understanding of that
preliminary process is essential before protoplast fusions can be used in
genetic engineering.
3a L. Ferenczy, F. Kevei, and M. Szegedi, Experientia 31, 50 (1975).
39 O. Schieder, Z. Pflanzenphysiol. 74, 357 (1974).
40 j. B. Power, S. F. Berry, E. M. Frearson, and E. C. Cocking,Plant Sci. Lett. 10, 1 (1977).
4~ E. C. Cocking, D. George, M. J. Price-Jones, and J. B. Power, Plant Sci. Lett. 10, 7
(1977).
42 G. Melchers and G. Labib, Mol. Gen. Genet. 135, 277 (1974).
43 H. J. Grambow, K. N. Kao, R. A. Miller, and O. L. Gamborg, Planta 103, 348 (1972).
44 E. M. Frearson, J. B. Power, and E. C. Cocking, Dev. Biol. 33, 130 (1973).
45 D. Bui-Dang-Ha and I. A. Mackenzie, Protoplasma 78, 215 (1973).
46 G. Gosch, Y. P. S. Bajaj, and J. Reinert, Protoplasma 86, 405 (1975).
4r K. K. Kartha, M. R. Michayluk, K. N. Kao, O. L. Gamborg, and F. Constabel,PlantSci.
Lett. 3, 265 (1974).
368
[30]
[30] Cell T r a n s f o r m a t i o n
By IRA PASTAN
Cultured fibroblastic cells are frequently used in studies of "malignant
transformation"; the term "transformation" is also often used by itself.
"Malignant transformation" was coined to distinguish it from the "transformation" process by which DNA is introduced into bacteria. A "transformed" cell is one that has undergone a stable heriditable change that
enables it to grow into a tumor in an appropriate recipient animal. Since
the animal from which a cell is derived may not be available for testing for
tumorgenicity in vivo, other tests have been developed. Nude mice have
an immunological deficiency that interferes with their ability to reject
foreign cells, thereby allowing their use for tumorgenicity testing (see this
volume [31]).
Since it is cumbersome to test whether a cell has become transformed
by injecting it into an animal, other simpler tests are widely used. One that
strongly correlates with results of animal tests is the ability of transformed
cells to grow in the absence of substratum (agar; methylcellulose; suspension culture). 1 Another prominent metabolic feature of transformed cells
that has long been recognized is their rapid metabolism of glucose; 2 cells
develop this feature shortly after transformation)
Frequently, transformed cells have a different morphologic appearance than normal cells in culture. Normal fibroblasts are elongated and
flattened. Transformed cells often are more rounded. This change in
shape is accompanied by changes in the cell surface. When viewed by
scanning electron microscopy, cells are usually covered with many microvilli and lamelopodia; at times the surfaces are covered with blebs. 4
Another property of transformed fibroblasts is their increased ability to be
agglutinated by plant lectins) ,6 Several theories have been advanced to
explain their enhanced agglutinability. The simplest of these is that
agglutinability is enhanced because the microvilli and other irregularities
i V. H. Freedman and S. Shin, in "The Nude Mouse in Experimental and Clinical Research" (J. Fogh and B. Giovanella, eds.). Academic Press, New York, 1978 (in press).
2 B. Warburg, " T h e Metabolism of Tumors." R. N. Smith, New York, 1931.
a S. G. Martin, S. Venuta, M. Weber, and H. Rubin, Proc. Natl. Acad. Sci. U.S.A. 68,
2739-2741 (1971).
4 K. Porter, D. Prescott, and J. Frye, J. Cell Biol. 57, 815-836 (1973).
5 M. M. Burger, Biochemistry 62, 994 (1969).
6 B. Inbar, and L. Sachs, Proc. Natl. Acad. Sci. U.S.A. 63, 1418 (1969).

ISBN 0-12-181958-2
[30]
CELL TRArqSVORMATIOrq
369
on the transformed cells' surface provide large surfaces that are available
for the lectins to "glue" the cells together. 7
A very important factor contributing to the abnormal shape of transformed cells is their decreased adhesion to substratum.S Mutant cells have
been obtained that have the morphologic phenotype of transformed cells:
rounded shape, decreased adhesion to substratum, increase in surface
microvilli, and increased agglutinability by lectins. In such cells the primary defect is a decrease in cell adhesion2 Apparently decreased adhesion to substratum causes the cell to assume a rounded shape with microvilli or blebs on the cell surface. Although such adhesion-defective
mutants have the morphologic phenotype of transformed cells, they are
not, in fact, transformed since they neither form tumors in animals nor
grow in agar2 Thus, in the absence of other information, it is not safe to
use morphologic or adhesive features for establishing that the phenomenon of transformation has occurred.
Selection of Transformed Cell Lines for Study
Many transformed cell lines are available for study and can be conveniently obtained from the American Type Culture Collection or other
sources (see this volume [37]). Some of these lines have been in culture
for many generations and may have developed characteristics that are not
related to the original transformation event. Therefore, it is generally best
to use recently transformed cells in which the normal parent is available
for comparative purposes.
High and Low Transformation Systems
It is possible to transform almost all the cells in a culture into cancer
cells within a few days with RNA tumor viruses (high-frequency transformation). A widely used system is that of chick embryo fibroblasts
(CEF) transformed by various strains of Rous sarcoma virus (RSV).
Methods for preparing pure clones of RSV and obtaining RSVtransformed chick embryo cells are described in this Volume [32] and [33].
Similarly, murine sarcoma virus (MSV) when combined with a helper
virus murine leukemia virus (MuLV) will rapidly transform various mouse
and rat cell lines. 1 An advantage of the Rous sarcoma system is the
r M. Willingham, and I. Pastan, Proc. Natl. Acad. Sci. U.S.A. 72, 1263-1267 (1975).
8 K. K. Sanford, B. E. Barker, M. W. Woods, R. Parshad, and L. W. Law, J. Natl. Cancer
Inst. 39, 705-718 (1967).
9 j. Pouyssegur, M. Willing,ham, and I. Pastan, Proc. Natl. Acad. Sci. U.S.A. 74, 243-247
(1977).
10 j. W. Hartley and W. P. Rowe, Proc. Natl. Acad. Sci. U.S.A. SS, 780-786 (1966).
370
SPECIALIZED
TECHNIQUES
[31]
availability of viruses with temperature-sensitive mutations in the gene s r c

responsible for the transformation process (reviewed in Vogt e t al.11).
These viruses should be maintained in the same manner as for wild-type
virus, but only at the permissive temperature in order to prevent selection
of revertants. They then can be checked at a nonpermissive temperature
prior to use. Mutants with deletions in the s r c gene are also available) ~
In general, the DNA tumor viruses transform only a small fraction of
cells they infect (low-frequency transformation). This applies to SV40, TM
polyoma, TM and adenovirus (see this volume [35]). Therefore, a number of
transformed clones should be studied to determine if the event of interest
is "transformation-specific." With SV40-infected cells, temperaturesensitive mutants in the gene responsible for transformation are
available. 13
It is also possible to transform cells with chemical carcinogens. A
detailed method for obtaining such cells is included in this volume [23].
tl p. K. Vogt, R. A. Weiss, and H. Hanafusa, J. Virol. 13, 555-554 (1974).
12 j. Tooze, in "The Molecular Biology of Tumor Viruses" (J. Tooze, ed.), Chapter 6, pp.
350-402. Cold Spring Harbor Lab., Cold Spring Harbor, New York, 1973.
13 T. J. Kelley and D. Nathans, Adv. Virus Res. 21, 86 (1977).
[31 ] U s e o f N u d e M i c e f o r T u m o r i g e n i c i t y T e s t i n g a n d M a s s
Propagation
By
SEUNG-IL
SHIN
The hairless nude phenotype in the mouse results from an inherited

developmental defect, which is determined by an autosomal recessive
mutation that segregates according to the simple Mendelian laws. In addition to being virtually hairless, homozygous mutant nude mice also suffer
from a congenital failure to develop a normal thymus gland. As a result,
nude mice are completely deficient in the thymus-dependent (T-cell) immunological functions.X
Since immunological rejection of heterologous tissue grafts in an animal depends primarily on the T-cell-mediated immunity of the host, the
nude mouse is unable to reject implanted cells or tissues from a genetically nonidentical donor, z This discovery led to the suggestion that the
nude mouse should be useful as a test system for determination of cellular
J. Rygaard, "Thymus and Self: Immunobiology of the Mouse Mutant Nude." F.A.D.L.,
Copenhagen, 1973.
2 j. Rygaard and C. O. Povlsen, Acta Pathol. Microbiol. Scand. 77, 758 (1969).
Copyright 1979 by Academic Press, lnc.

ISBN 0-12-181958-2
[31]
NUDE MICE FOR TUMORIGENICITY TESTING
371
tumorigenicity and as a biological "incubator" for mass propagation of

animal cells that can form tumors in this host. a,4 Subsequent developments confirmed that nude mice are indeed a useful adjunct to the
biochemist who needs to grow a large quantity of cells or tumor tissues in
a relatively simple and reliable manner. Using a number of cell mutants
carrying specific biochemical and genetic markets, Freedman and coworkers demonstrated that mass production of mammalian cells by inoculation of small numbers of cells in nude mice does not cause changes in
these markers. ~
Heteroploid mammalian cell lines, most of which exhibit variable degrees of transformed phenotypes in vitro and are tumorigenic in nude
mice, can be grown as large tumors in nude mice by injecting 108 or fewer
cells. The maximal size of tumors that can be obtained depends upon the
cell types used, but large single tumors weighing as much as 20 g have
been observed. The usefulness of the nude mouse as an experimental
vehicle for production of animal cells is particularly obvious for the cell
types that cannot be grown in tissue culture and for the tumors that cannot
be transplanted in the usual experimental animals. For example, a number
of different human primary tumors removed at surgery have been successfully propagated in nude mice as serial transplants. 6,7
For genetic and biochemical studies of cellular transformation, the
availability of a generally applicable assay for neoplastic growth potential
in vivo is sine qua non. Cells transformed in vitro by oncogenic viruses,
chemical carcinogens, or other agents usually express tumor-specific cell
surface antigens. Established heteroploid cell lines not deliberately exposed to known transforming agents may also express new surface antigens through spontaneous mutations. Therefore, even cells originally derived from inbred strains of experimental animals may elicit immunological rejection when they are tested for tumorigenicity in the strain of origin.
For cells derived from noninbred animals, notably human, an immunologically neutral test system is clearly even more essential.
Several studies established that the nude mouse provides a rare and
unique advantage in this regard, since cell surface antigens that normally
induce graft rejection in normal animals fail to do so in the thymusless
3 Vo H. Freedman and S. Shin, Cell 3, 355 (1974).

4 S. Shin, V. H. Freedman, R. Risser, and R. Pollack, Proc. Natl. Acad. Sci. U.S.A. 72,
4435 (1975).
5 V. H. Freedman, A. L. Brown, H. P. Klinger, and S. Shin, Exp. Cell Res. 98, 143 (1976).
6 M. Schmidt and R. A. Good, J. Natl. Cancer Inst. 55, 81 (1975).
7 y. Shimosato, T. Kameya, K. Nagai, S. Hirohashi, T. Koide, H. Hayashi, and T. Nomura, J. Natl. Cancer Inst. 56, 1251 (1976).
372
[31]
nude mice (for a review, see Freedman and ShinS). Some recent experiments suggest that nude mice contain elevated levels of certain nonTcell-mediated immunity, and that they are in fact capable of inhibiting the
growth of some tumorigenic cells. 9 It may thus become necessary to
employ additional immunosuppressive techniques on nude mice to assess
the tumorigenicity of selected cell types. However, for the general purpose of cellular tumorigenicity assay of animal cells in culture, the nude
mouse offers the best system presently known. The association between
cellular tumorigenicity in nude mice and various in vitro parameters of
cellular transformation has been reviewed recently. 8 The parameter best
correlated with tumor formation in vivo is the "anchorage-independent"
phenotype,-as measured by colony-forming efficiency in soft agar or
methylcellulose.
Source, Maintenance, and Breeding of Nude Mice
Due to their severe immune deficiency, homozygous mutant (nu/nu)
mice are highly susceptible to infectious diseases. During the early days
after the original discovery of this mutation, when the nature of the genetic defects was unidentified and therefore conventional animal husbandry techniques were used, nude mice rarely survived beyond the first few
weeks. However, if relatively simple precautions are taken it is now possible to maintain nude mice f o r a t least several months in most research
laboratory environments.
For intermittent short-term uses of nude mice, the most practical arrangement would be to obtain young adult nude mice from commercial
breeders. Once received, the animals should be strictly segregated from
other experimental animals, especially rodents; this is a most important
factor in maintaining good viability. Nude mice in different inbred backgrounds can now be purchased from commercial suppliers in the United
States and several other countries. Preferably, the mouse cages should be
placed inside a laminar-flow air filtration system. Conventional laminarflow hoods designed for cell culture can be used for this purpose. For
short-term experiments, the use of clean filterpaper covers that fit over
the cage tops has also proven to be adequate. Cages, bedding, cage tops,
and water bottles should be autoclaved before use if possible. Drinking
water should be boiled beforehand, but normal animal feed can be used
without further modification. Access to the animals should be limited to
8 V. H. Freedman and S. Shin, in " T h e Nude Mouse in Experimental and Clinical Research" (J. Fogh and B. Giovanella, eds.), p. 353. Academic Press, New York, 1978.
L. M. Reid, J. Holland, and C. Jones, submitted for publication.
[31]
373
the investigator or the research assistant as much as possible. The use of

disposable surgical gowns, masks, and gloves is recommended for experiments that require longer duration.
With a greater awareness of the nature of immune deficiency in the
nude animals, breeding of nude mouse colonies has become increasingly
practical for many research laboratories not otherwise engaged in animal
breeding programs. In most cases, heterozygous (+/nu) females, which
are phenotypically normal, are mated with homozygous mutant (nu/nu)
males. However, breeding and maintaining an independent nude mouse
colony would still require resources not usually available in a biochemical
laboratory. Detailed information and practical guides for breeding of nude
mice have been published. 1'11 Information on various aspects of experimentation on nude mice in different countries can be obtained from
the Nude Mouse Secretariat, c/o Pathological Anatomical Institute,
Kommunenhospitalet, DK 1399 Copenhagen K, Denmark. In the United
States, research workers may also receive practical information from the
Veterinary Resources Branch, National Institutes of Health, Bethesda,
Maryland.
Procedures
Test of Cellular Tumorigenicity

Harvest cells from the culture vessels in the usual manner, by trypsinization of monolayer cells or by centrifugation of cells grown in suspension. Determine the number of viable cells by the trypan blue exclusion
test. (Suspend the cells in phosphate-buffered saline, containing 0.2%
trypan blue, for 5 min and count the unstained cells on a hemacytometer.)
Take up an appropriate volume of the cell suspension to contain enough
viable cells for 2 106 cells per injection point. Wash the cells once with
serum-free culture medium, and resuspend the cells in the same medium
to a final cell concentration of 107 cells/ml.
Inject 0.2 ml of the suspension for each assay point, using sterile
1-ml-capacity tuberculin syringes fitted with a 21-gauge needle. Injecting
the cells into a mouse can be made much easier if a second person holds
the mouse, by firmly grasping at the base of the tail and at the back of the
1o "Guide for the Care and Use of the Nude (Thymus-Deficient) Mouse in Biomedical
Research," a Report of the Committee on Care and Use of the " N u d e " Mouse. I L A R
N e w s 19, No. 2 (1976).
11 R. Ediger and B. C. Giovanella, in "The Nude Mouse in Experimental and Clinical
Research" (J. Fogh and B. Giovanella, eds.), p. 16. Academic Press, New York, 1978.
374
[31]
head behind the ears, and keeps the mouse stretched out. The person who
holds the syringe can then lift the fold of skin on the flank of the mouse
midway between the front and the hind legs, and use the other hand for
injection. It is neither necessary nor desirable to anesthetize the mouse
for this purpose. The cells should be deposited subcutaneously at a single
point, as far away as possible from the point of penetration in order to
prevent the cells from oozing out when the needle is withdrawn. Immediately after removal of the needle, hold the needle opening briefly between
two fingers. The fact that the cells are injected subcutaneously, and not
into the abdominal cavity or the rib cage, can be seen easily during the
injection process because the needle is visible throughout due to the translucency of the skin of the hairless mouse.
In order to increase the assay points per nude mouse, injections can be
made bilaterally. For the determination of.the threshold number of cells
required to initiate the tumor growth, etc., the number of cells injected
accordingly can be decreased or increased. Up to 108 cells have b e e n
injected at a single site.
Check the injected mice at least every 2 days thereafter, by both visual
inspection and manual palpation. The latency period between the injection of cells and the first appearance of a palpable nodule at the site of
injection varies greatly, depending upon the malignancy of the cells and
the inoculum used. 8 A typical tumorigenic cell line, such as H e L a or
mouse A9 cells, generates palpable nodules 1-2 weeks following injection
of 2 x l0 s viable cells. The minimum threshold number of cells required
to initiate tumor growth also varies widely. With the most tumorigenic
cells, such as the mouse melanoma cell line PG19, less than 10 cells can
initiate tumor growth, while near-diploid Chinese hamster cell lines such
FIG. 1. A phenotypically normal, heterozygous (+/nu) BALB/c mouse (left), and its hairless, thymus-deficient homozygous mutant nude littermate (right).
[31]
375
as CHO and WOR-6 form growing tumors only if 106 cells or more are
injected. Nontumorigenic cells, such as human diploid cell strains, will fail
to form tumors even with l0 s cells. A tumorigenic cell line should result in
the formation of a progressively growing tumor that, if left alone, would
eventually kill the host. Positive results are rarely ambiguous (see Fig. 1).
Test of Mycoplasma Infection

Infection of t e s t cells by mycoplasma can drastically alter their
tumorigenicity. 12It is therefore imperative that the absence of contaminating mycoplasma in the cell lines be established by appropriate tests for
mycoplasmas before the cell lines are injected into nude mice. This subject has been reviewed extensively TM (see this volume [2]).
Mass Production of Cells in Nude Mice

Procedures identical to those described above are used for the generation of large tumors in nude mice, except that greater numbers of cells
should be injected in order to shorten the latency period. It is practical to
inject up to 4 107 cells in 0.2 ml volume each at both sides of a mouse.
In general, tumors induced in nude mice by heterologous malignant cells
grow as well-encapsulated masses and do not as a rule metastasize to
distant sites. The tumor mass can therefore be removed from the host
relatively free of the host tissue. Variable amounts of host blood elements,
however, are present in the tumor masses. 5 The final weight of the tumor
that is produced in nude mice depends primarily on the cell type used.
Many of the commonly used established cell lines such as HeLa, mouse L
cells, and Syrian hamster B H K 21 can produce enormous tumors that
weigh more than the host mouse itself. 5 Certain other cell lines, such as
SV40-transformed 3T3, rarely result in large fast-growing tumors. The
factors that determine the final tumor size in nude mice are not well
understood, even though the residual tumor immunity of the nude mouse
against specific cellular antigens may play a role.
In Tables I, 3,14--17 II,3--5,1S,18 and I I I , 3,4.1e'ls-21 representative cel~ lines
that have been shown to be tumorigenic in nude mice are presented. It can
be seen that a very wide variety of cell types can be grown in nude mice as
tumors. In Table IV, commonly used established cell lines that have been
shown to be non tumorigenic are listed. Tumorigenicity o f virus-transformed cells in nude mice has been discussed elsewhere. 22
12 O. P. van Diggelen, S. Shin, and D. M. Phillips, Cancer Res. 37, 2680 (1977).
ta G. J. McGarrity, D. G. Murphy, and W. W. Nichols, eds., "Mycoplasma Infection of Cell
Cultures," Vol. 3. Plenum, New York 1978.
376
[31]
TABLE I
REPRESENTATIVE
Cell line
TUMORIGENIC
CELL
LINES
DERIVED
FROM HUMAN
Tissue of origin
TISSUES
References
HeLa
D98/AH2
HS 0643
RPMI 2650
KE- I
SH- 1
PA- 1
NB20
Adult cervical carcinoma

8-azaguanine-resistant mutant of HeLa
Lung carcinoma
Nasopharyngeal carcinoma
Fibrosarcoma
Melanoma
Ovarian teratoma
Lymphoblastoid cell line, established in vitro
3
3
3
14
1~
15
~5
S. Shin,
unpublished
WI-L2
A5/FG/HEK
LN-SV
Lymphoblastoid cell line, established in vitro

Embryonal cells, transformed in vitro by adenovirus 5
Lesch-Nyhan fibroblasts, transformed in vitro by simian
virus (SV) 40
14
16
17
TABLE II
REPRESENTATIVE TUMORIGENIC CELL LINES DERIVED FROM MOUSE TISSUES
Cell line
Tissue or cell line of origin
L929
A9
CI-I-D
Ehdich/Lettr~
RAG
Normal connective tissue, transformed in vitro

8-azaguanine-resistant mutant of L929
BrdUrd-resistant mutant of L929
Ascites tumor, culture-adapted
Renal adenocarcinoma; 8-azaguanine-resistant
mutant
Spontaneous melanoma
Carcinoma of adrenal cortex
Lewis lung carcinoma
Teratocarcinoma from strain 129;
8-azaguanine-resistant mutant
Plasmacytoma, a derivative of MPC 11
Myeloma
B-16
Y-1
LLC
PCC4.azal
MPC 45-6
MOPC 315
SV101
Py3T3
KA31
EL4
3T3 (Swiss), transformed in vitro by SV40

3T3 (Swiss), tranformed in vitro by polyoma
virus
3T3 (Balb), transformed in vitro by Kirsten
murine sarcoma virus
Lymphoma, induced by Moloney virus
References
3
3
5
3
3
15
18
15
S. Shin, unpublished
S. Shin and H. Eisen,
unpublished
3
18
[31]
377
TABLE III
TUMORIGENICCELL LINES FROM OTHER SPECIES
Cell line
Species
GH3
Rat
LG
Rat
C6
F17
Rat
Rat
SVRE-9
Rat
(M-MuSV)NRK
Rat
(KSV)NRK
Rat
CHO
Chinese
hamster
Chinese
hamster
WOR-6
PSN4
BHK 21
BHK-T6a
DMN6A
Syrian
hamster
Syrian
hamster
Syrian
hamster
Syrian
hamster
References
Pituitary tumor secreting growth

hormone
Clonal line of embryonal myoblast
19
Chemically induced glioma

Embryonal cells, transformed in
vitro by adenovirus 2
vitro by SV40
Rat kidney cells (NRK), transformed in vitro by Moloney
sarcoma virus
NRK, transformed in vitro by
Kirsten sarcoma virus
Normal ovary, spontaneously
transformed
Normal lung fibroblast, mutant
resistant to 8-azaguanine and
ouabain
Embryonal cells, chemically
transformed
Baby kidney, spontaneously
transformed
BHK 21, 8-azaguanine-resistant
mutant
BHK 21, chemically transformed
in vitro
PySplA
Syrian
hamster
BHK 21, transformed in vitro by

polyoma virus
B(a)P104C1
Guinea pig
DENA HM2C1
Guinea pig
RKT-TG1
Rabbit
RSV-CE-Pr
Chick

vitro by benz(a)pyrene
vitro by diethylnitrosamine
Normal adult kidney, transformed
in vitro by SV40
Embryo fibroblast, transformed
in vitro by Rous sarcoma virus
S. Shin and
B. Nadal-Ginard,
unpublished
18
16
20
H. Klinger,
unpublished
3
P. Canary
and N. Bouck,
unpublished
P.. Canary
and N. Bouck,
unpublished
21
S. Shin and
R. Junghans,
unpublished
378
[31]
TABLE IV
REPRESENTATIVE ESTABLISHED CELL LINES THAT ARE
NONTUMORIGENIC IN NUDE MICE
Cell line
Species
References
CV-1
MDCK
Balb/3T3
3T3 (Swiss)
SVR95
Monkey
Dog
Mouse
Mouse
Mouse
4
is
3
3
4
F1-SV
M22
Mouse
Mouse
NRK C13
BRL
F2408
Rat
Rat
Rat
Normal adult kidney

Normal kidney
Embryonic fibroblasts
Embryonic fibroblasts
3T3 (Swiss), partially transformed in vitro
with SV40
Revertant of SV40-transformed 3T3 (Swiss)
Revertant of KA31, Balb/3T3 transformed in
vitro by Kirsten murine sarcoma virus
Normal kidney
Normal liver
Normal embryonic fibroblasts
SVRE 12
Rat
Embryonic fibroblast, partially transformed

in vitro by SV40
4
4
20
18
S. Shin,
unpublished
4
Reestablishment of Cell Cultures from Tumors

Tumors produced in nude mice by established culture lines can easily
be reintroduced into culture for further characterization. All steps should
be carried out under strict aseptic conditions to prevent contamination of
the cultures:
Sacrifice the tumor-bearing mouse, either by placing it in a chamber
with a loose-fitting lid that contains Dry Ice or by cervical dislocation.
Pin the animal to a cork board and move it into a cell culture hood for the
14 C. D. Stiles, W. Desmond, G. Sato, and M. H. Saier, Jr., Proc. Natl. Acad. Sci. U.S.A.
72, 4971 (1975).
15 B. C. Giovanella, J. S. Stehlin, and L. J. Williams, J. Natl. Cancerlnst. 52, 921 (1974).
16 L. B. Chen, P. H. Gallimore, and J. D. McDougall, Proc. Natl. Acad. Sci. U.S.A. 73, 3570
(1976).
17 C. M. Croce, D. Aden, and H. Koprowski, Proc. Natl. Acad. Sci. U.S.A. 72, 1397 (1975).
Is C. D. Stiles, W. Desmond, L. M. Chuman, G. Sato, and M. H. Saier, Jr., Cancer Res. 36,
1353 (1976).
lg S. Shin, A. L. Brown, and F. C. Bancroft, Endocrinology 103, 223 (1978).
20 j. A. Bilello, V. H. Freedman, and S. Shin, J. Natl. Cancer Inst. 58, 1691 (1977).
21 C. H. Evans and J. A. DiPaolo, Cancer Res. 36, 128 (1976).
32 C. D. Stiles and A. A. Kawahara, in " T h e Nude Mouse in Experimental and Clinical
Research" (J. Fogh and B. Giovanella, eds.), p. 385. Academic Press, New York, 1978.
[32]
AVIAN SARCOMA VIRUSES
379
subsequent operations. After spraying the area around the tumor with
95% ethanol, open the tumor with scissors and remove a section of tumor
about 5 5 x 5 mm from an internal area free of necrotic tissue. Place
the tumor tissue in a sterile petri dish and remove the mouse from the
hood. Rinse the tumor tissue 3 times with at least 5 ml of phosphatebuffered saline or culture medium, placing the tissue into a new petri dish
each time. Mince the tissue into very small fragments of 1 mm s or less and
remove all obvious fatty material. Many tumors can be reduced to mostly
single cells and cell clumps in this manner. Tumor tissues that are not
easily dissociated into single cells can at this point be passed through a
sterilized stainless-steel wire mesh to further disaggregate the cells by
rubbing the fragments against the mesh with blunt-ended scissors. Alternatively, stepwise, gentle trypsinization can be used to dissociate single
cells from tumor fragments. In any case, it is not necessary to obtain pure
single-cell suspensions in order to start a new culture from the tumor.
Suspend the cell or minced tissue preparation in 10-25 ml culture medium
containing antibiotics, and distribute the contents into a set of 60-mm
culture dishes. Change the medium after overnight incubation; most of
the unattached lymphocytes and cell debris can be washed off at this time.
Red cells, loosely attached to the dish, may last for many days, but will
not prevent the growth of the cells. Tumors originally produced by established cell lines almost always give rise to vigorous cultures in a few days.
[32] B i o l o g i c a l T e c h n i q u e s f o r A v i a n S a r c o m a V i r u s e s
By ERIC HUNTER
Rous sarcoma virus (RSV), an avian retrovirus, is capable of inducing
rapidly growing solid tumors of the connective tissue in fowl, and morphological transformation of cells in culture. This virus group is therefore
of particular interest as a model system for studies on viral oncogenesis.
Moreover, it has become clear that the avian sarcoma viruses provide an
excellent system for studies of the control of eukaryotic gene expression
and of the mechanism of integration of viral nucleic acids into a host cell
genome.
Experiments of this nature require methods for obtaining a biologically
and biochemicaUy homogenous virus population and quantitating the biological activity of such a population. This article describes a quantitative
and reproducible assay of RSV, the focus assay, l'z and methods for obR. A. Manaker and V. Group6 Virology 2, 838 (1956).
z H. M. Temin and H. Rubin, Virology 6, 669 (1958).
M~I'HODS IN E N Z Y M O ~ Y , VOL. LVIII

ISBN 0-12-181958-2
380
[32]
taining cloned preparations of RSV, through both the focus and softagar-colony assays. Large-scale growth of RSV is considered in this volume [33].
Infection of chick embryo fibroblasts by RSV in vitro results in a
gradual change in the fibroblastic morphology of the infected cells to that
of a rounded, refractile transformed cell. Division of the initially transformed cell, in addition to infection and transformation of adjacent cells,
results in the formation of an island or "focus" of morphologically transformed cells within the uninfected monolayer of fibroblasts. A single particle of RSV is sufficient to produce a focus since the relationship between
virus qoncentration and focus count is strictly linear. The shape of the
transformed cell as well a s the morphology of the focus are largely controlled by the viral genome, 3 so that a focus induced b y the SchmidtRuppin strain of RSV is normally diffuse with interdispersed cells of
normal morphology while those induced by the Bryan high-titer strains of
RSV are generally compact and contain only transformed cells. To a
lesser extent focus characteristics are also influenced by the physiological
state of the cells in culture. For instance, cells that have been transferred
several times may tend to respond with the production of transformed
cells with a fusiform morphology.
Most strains of RSV are nondefective for replication; that is, infection
of a cell with a single virion results in both infectious-virus replication and
cell transformation. Viruses of the Schmidt-Ruppin strain of RSV, the
Prague strain of RSV, and the B77 strain of avian sarcoma virus (ASV) are
of this type. The most notable exception is the Bryan high-titer (BH)
strain of RSV, a replication-defective virus that requires a coinfecting
helper virus in order to produce infectious progeny. BH-RSV is a deletion
mutant, lacking most of the envelope glycoprotein gene. 4 It can be complemented by an accompanying helper virus that provides the missing
envelope glycoproteins that are necessary for infectivity. ~ BH-RSV
strains exist primarily, therefore, as "pseudotypes" in conjunction with a
nontransforming helper virus. Such a mixture may be unsuitable for some
biochemical experiments where a single homogenous population of virus
is required.
Viruses of the Schmidt-Ruppin and Prague strains of RSV and the B77
strain of ASV are most frequently used for biochemical studies, since they
can be readily cloned and yield good titers of virus. The initial isolates of
the Schmidt-Ruppin and Prague strains of RSV were mixtures of viruses
3 HI M. Temin, Virology 10, 182 (1960).
4 p. H. Duesberg, S. Kawai, L.-H. Wang, P. K. Vogt, H. M. Murphy, and H. Hanafusa,
Proc. Natl. Acad. Sci. U.S.A. 72, 1569 (1975).
5 H. Hanafusa, Natl. Cancer Inst., Monogr. 17, 543 (1964).
[32]
381
with different host range characteristics. The Schmidt-Ruppin strain

yielded virus of subgroups A, B, and D while the Prague strain was found
to consist of viruses of subgroups A, B, and C. 6 Subgroup D virus of the
Schmidt-Ruppin strain of RSV (SR-D) and the B77 strain of ASV (subgroup C) are worth noting since they infect mammalian cells with a relatively high efficiency. Indeed, there is evidence to. suggest that B77 virus
can infect rat cells as efficiently as it can chicken cells; with mammalian
cells, however, infection may not necessarily be accompanied by expression of viral genetic information, r even though the cells contain an integrated provirus that can be rescued after fusion with permissive chicken
cells.
Preparation and Culture of Chicken Embryo Fibroblasts

Materials
1. Avian Embryos
Fertile chicken eggs, most frequently of the white leghorn breed, may
be obtained from commercial flocks, many of which have only a low (less
than 5%) incidence of congenital leukosis virus infection. Among those
breeders in the United States whose flocks have proven satisfactory are:
Hy-Line International, Dallas Center, Iowa; H & N Inc., Redmont, Washington; and Spafas Inc., Norwich, Connecticut. Of these H & N and
Spafas also offer eggs from leukosis-free flocks. For crucial experiments,
however, tests for leukosis virus in the embryo are worthwhile.
All chicken cells contain, integrated within their chromosomal DNA,
the genome of an endogenous leukosis virus, RAV-0. The degree to which
the endogenous virus genes are expressed varies with different chicken
flocks, and ranges from undetectable expression to assembly and release
of infectious RAV-0 virions. In most commercial flocks release of complete virions is rare, but the synthesis of the internal structural or gs
proteins and envelope glycoproteins of the virus is common. Since the
latter cells can complement the BH-RSV strain by providing the missing
envelope glycoproteins, they are generally termed chick helper factor
(chf)-positive cells.
Detectable transcription of the endogenous viral genome may be undesirable, and for such studies, embryos from chicken flocks negative for
6 p. K. Vogt, personal communication.
7 D. Boettiger, Cell 3, 71 (1974).
382
[3 9-]
both chf and gs expression can be obtained from Spafas Inc. and H & N
Inc. For crucial experiments, quantification of chf expression, as described later in this article, is recommended.
Fertile eggs of other avian species generally can be obtained from local
vendors. Cells from the Japanese Quail and Peking Duck are particularly
useful in some experiments since they appear to lack any endogenous
viral sequences related to RSV.
2. Sera
a. Calf Serum. Not all serum lots offered commercially will support
focus formation by RSV in tissue culture, and samples from several different lots should be tested before larger quantities are purchased. Satisfactory sera have been obtained from Biocell Inc., Venice, California;
Flow Laboratories, Bethesda, Maryland; and Grand Island Biological
Company, Grand Island, New York. Newborn calf serum is usually not
required for cultivation of chick cells. In fact, bovine cadet serum, collected from animals up to 6 months of age, has proved highly satisfactory.
Calf sera are not inactivated. They may be stored for as much as 1 year at
- 2 0 .
b. Chicken Serum. Chicken serum also may be purchased from the
suppliers mentioned above. It should not contain antibody to RSV. Since
these sera are not usually obtained from leukosis-free flocks, they must be
inactivated at 56 for 2 hr prior to use in order to destroy any infectiousleukosis viruses that may be present. Storage up to I year at - 2 0 is
permissible.
3. Media and Reagents
a. Cell Culture Media. Several media are satisfactory both for the
growth of chick embryo cells and assay of RSV. Nutrient mixture F108 is
commonly used, but Minimal Essential Medium, 9 and Medium 1991may
be substituted. Single- and double-strength concentrations are required.
Commercially available solutions and powders are acceptable. The latter
are particularly suitable where large volumes of media are used, since
each batch can be tested in advance for its ability to support cell growth
and focus formation. Sterilization of tissue culture media is achieved by
pressure filtration through 0.22-/zm (average pore size) filters. Penicillin
s R. G. Ham, Exp. Cell Res. 29, 515 (1963).
g H. Eagle, Science 130, 432 (1959).
10 j. F. Morgan, H. J. Morton, and R. C. Parker, Proc. Soc. Exp. Biol. Med. 73, 1 0950).
[32]
383
(1 105 IU/liter) and streptomycin (100 mg/liter) are added prior to

sterilization.
b. Tris-Buffered Saline. This contains 8 g/liter NaC1, 0.37 g/liter KC1,
0.1 g/liter Na HPO4, 1 g/liter glucose, 3 g/liter Tris, 105 IU/liter penicillin,
100 mg/liter streptomycin and is adjusted to pH 7.2 with 10 N HC1. Trisbuffered saline is used for washing cells and as a general diluent for
trypsin and other reagents. It is sterilized by pressure filtration.
c. Trypsin. This is prepared as a 0.25% solution in Tris-buffered saline
from Trypsin 1:300 (Grand Island Biological Company). The powdered
trypsin is stirred overnight in & the final desired volume, clarified, and
filtered under pressure through a 0.22-/zm filter. Prefiltration through filters of 0.8 /zm and 0.45 /zm average pore size can help prevent rapid
clogging of the final 0.22-/zm filter. The sterile 0.25% solution may be
stored at - 2 0 , and distribution in volumes of approximately 100 ml
makes it feasible to use freshly thawed trypsin for preparation of primary
chick embryo cultures.
d. Tryptose Phosphate Broth. Bacto tryptose phosphate broth (Difco
Laboratories) is prepared according to the manufacturer's directions and
sterilized by autoclaving.
e. Agar. Purified agar (Difco Laboratories) is prepared at a concentration of 1.8% in double-distilled water and sterilized by autoclaving.
Purified agar is preferable to Bacto agar (Difco Laboratories) for the
nutrient overlay of primary cultures, but is unsuitable for use in the RSV
focus assay where Bacto agar is used.
f. Sodium Bicarbonate. A 2.8% solution of sodium bicarbonate is prepared in distilled water and sterilized by autoclaving.
g. Folic Acid. A 100 stock solution of folic acid is prepared by dissolving 80 mg folic acid in 100 ml of 1 N sodium bicarbonate. The solution
is sterilized by filtration and stored in small aliquots at - 2 0 .
h. Vitamins. A x 100 stock solution of the following nonessential vitamins may be purchased commercially or prepared in the laboratory and
sterilized by filtration: biotin, calcium pantothenate, choline chloride, inositol, niacinamide, pyridoxine HC1, riboflavin, thiamine HCI.
i. Dimethyl Sulfoxide. Analytical grade dimethyl sulfoxide (DMSO) is
used as an additive to media and for freezing cells.
4. Complex Media Formulations
The media solutions and reagents described above are combined to
form complex media, such as liquid growth medium for primary cultures
(PGM); semisolid "overlay" medium for primary cultures (POL); growth
medium for secondary cultures (GM); seeding medium (SM); and agar
384
SPECIALIZED
[32]
TECHNIQUES
TABLE I
COMPLEX MEDIA FOR PRIMARY AND SECONDARY CULTURES OF CHICK EMaRYO CELLS
Media
Nutrient mixture FI0, a 1 concentration
Nutrient mixture F10, a 2 concentration
Calf serum
Chicken serum
Tryptose phosphate broth
Sodium bicarbonate (2.8%)
Purified agar c
Bacto agar
Dimethyl sulfoxide
PGM
POL
SM
GM
OL
80 ~
-8
2
10
2.0
--
40
20
8
2
10
2.0
20
80
-1
-10
2.0
--
80
-5
-10
2.0
--
-40
5
-10
2.0
--
.
--
.
--
.
--
1.0 d
40
1.0
a Other culture media may be substituted for nutrient mixture F10; see text.
b E x p r e s s e d in p a r t s o f t o t a l v o l u m e .
c P u r i f i e d a n d B a c t o a g a r ( 1 . 8 % ) a r e m e l t e d in a b o i l i n g - w a t e r b a t h o r m i c r o w a v e
o v e n . B o t h a g a r a n d m e d i a c o m p o n e n t s a r e t h e n e q u i l i b r a t e d in a 45 . w a t e r b a t h
prior to mixing.
d G M is s u p p l e m e n t e d w i t h 1 % D M S O w h e n v i r u s is t o b e h a r v e s t e d f r o m t r a n s f o r m e d cells; s e e t e x t .
overlay medium for the RSV focus assay (OL) (Table I). The formulations
for the bottom agar (BA) and top agar (TA) media of the soft-agar-colony
assay and for cloning medium (CM) used in conjunction with this assay
are presented in Table II.
T A B L E II
COMPLEX MEDIA FOR SOFT-AGAR CLONING OF R S V
Media
BA
TA
CM
Nutrient mixture FI0, 1 concentration

Nutrient mixture F10, 2 concentration
Calf serum
Chicken serum
Tryptose phosphate
x 100 v i t a m i n s
x 100 f o l i c a c i d
Quail conditioned medium b
Bacto agar
Dimethyl sulfoxide
-35 ~
I0
3
15
1.0
1.0
-20
6
2
10
1.0
1.0
40
20
--
70
-10
3
15
1.0
1.0
--0.5
--
35
--
a E x p r e s s e d in p a r t s o f t o t a l v o l u m e .
b P G M h a r v e s t e d f r o m q u a i l p r i m a r y c u l t u r e p l a t e s 72 h r a f t e r s e e d i n g .
[32]
385
Cell Culture Methods
1. Preparation of Primary Chick Embryo Cell Cultures

Fertile chicken eggs, incubated for 9-11 days, are candled and the
position of the air sac marked. The shell above the air sac is swabbed with
70% alcohol and cracked with an egg punch (Tri R Instrument Co.,
Jamaica, New Jersey), which allows this piece of the shell to be removed
essentially in one piece with a pair of sterile forceps. From this point, it is
important to use separate instruments for each embryo in order to avoid
possible cross-contamination. Using sterile curved forceps for each operation, the air sac membrane is peeled away carefully and the embryo
removed to a sterile petri dish. The embryo is decapitated with the aid of
another pair of sterile forceps, and then it is transferred to a 50-ml conical
centrifuge tube (Coming or Falcon Plastics). Contaminating red blood
cells are removed by addition and removal of 10 ml of Tris-buffered saline,
and the embryo is minced to 1-2 mm square pieces using the spoonshaped end of a sterile stainless-steel spatula. Ten milliliters of warm,
freshly thawed trypsin (0.25%) are added, and the tissue suspension is
pipetted up and down gently 10-12 times in a 25 ml wide-mouth serological pipette in order to free cells from the tissue matrix. Pieces of tissue and
cell clumps are allowed to settle, and the supernatant liquid is decanted
into 20 ml of ice-cold PGM. Gentle swirling ensures that further digestion
of the cells in suspension is inhibited by the high levels of calf serum in the
PGM. Five milliliters of warm Tris-buffered saline are added to the remaining tissue, and cells that are freed by pipetting gently as before are
again decanted into ice-cold PGM after any tissue pieces have been allowed to settle. This washing procedure with Tris-buffered saline is carried out of a total of 4-5 times and is usually sufficient to free most cells from
the embryo tissue. If significant amounts of tissue remain at the end of this
procedure, a further 5 ml of warm trypsin may be added, and the washing
procedure repeated.
The cell suspension obtained is pelleted by low-speed centrifugation in
an PR2 refrigerated centrifuge I.E.C. or equivalent (600-800 rpm, 10 min),
and gently resuspended in 20 ml of PGM by means of a wide-mouth
pipette or a Vortex mixer. Since some tissue pieces and cell clumps are
unavoidably transferred during decanting, these are allowed to settle and
the supernatant liquid is carefully decanted once more to a sterile 50-ml
centrifuge tube. A small aliquot is removed, diluted 1 : 10 in PGM, and the
number of cells determined by counting the suspension in a hemocytometer. Care should be taken to avoid counting red blood cells (elipseshaped, nucleated cells) and badly damaged embryo cells. The yield from
a 10-day-old embryo should be between 1.0 and 1.5 x 108 viable cells.
386
[32]
Approximately 0.8--1.2 x 107 cells are required for tests of susceptibility to RSV infection and endogenous virus gene expression; the remainder
may be seeded in 100-mm petri dishes in l0 ml PGM (5-7 x l0 ~ ceils/
plate) for later use as secondary cultures. Plates containing PGM are
equilibrated in a 37-38 incubator, gassed with 95% air-5% CO2, for 1-2 hr
prior to seeding, and returned to it immediately after addition of the cells.
The primary cultures are overlayed with POL (12.5 ml/100-mm plate)
3 days after seeding. This stimulates the growth of the cells and maintains
them in a viable, healthy condition for up to 9 days after seeding.
The general procedure described above can be used to establish primary embryo cultures from most avian species, with only minor modifications to the method described here.
2. Assay of Primary Cells for Susceptibility to Virus Infection

In general, chicken embryos obtained from the larger commercial
breeders, particularly Spafas and H & N, will show uniform susceptibility
to specific virus subgroups. For example, H & N offer embryos susceptible to all subgroups (C/O) or susceptible to only A and C subgroups
(C/BDE). However, it is worthwhile to test the cells from each embryo in
order to confirm the susceptibility of the cells to the particular virus under
study and, in the case of less well defined flocks, to detect possible congenital infection with avian leukosis virus.
A portion of the cells from each embryo is seeded into 35-ram petri
dishes containing 2 ml PGM (1.0 x 106 cells/plate), and within 4 hr of
seeding these cultures are infected with both a high and a low concentration o f the RSV under study. In general, approximately 1 104 and
1 x 10z focus-forming units are added to the respective plates. Viruses of
subgroups A, B, and C are routinely added in this fashion to each embryo
to ascertain its susceptibility to RSV. The inoeulum is left on the cells
overnight, and the following day the liquid medium is replaced with 3 ml
of OL which is allowed to harden at room temperature for 10-20 rain
before the plates are returned to the incubator.
A qualitative estimate of susceptibility to the various virus subgroups
can generally be obtained on the third or fourth day after seeding by
observing the morphology of the cells on those plates receiving the most
virus. Since the large uninfected plates are generally ready to be used for
secondary cultures at this time, cells with the desired susceptibility can be
assured.
The number of foci developing on those plates receiving the lower
quantity of virus can be determined 7 days after infection and used as a
quantitative measure of the susceptibility of the cells.
[32]
387
Cellular resistance of primary chicken cultures to a particular RSV can

result from an inheritable characteristic in the cells or from congenital
infection of the embryo with an avian leukosis virus. The latter type of
resistance can be transmitted with cell-free supernatant fluids to susceptible cultures and is due to viral interference, whereas the former cannot.
Without carrying out this type of test it is not possible to distinguish
between the two types of resistance. However, as mentioned previously,
the flocks of the large commercial breeders have well-defined susceptibility to virus infection. Furthermore, Vogt 11 has pointed out that in commercial white leghorn flocks most resistance against subgroup B is of
genetic origin, whereas practically all resistance to subgroup A is caused
by congenital infection with a subgroup A leukosis virus.
3. Assay of Primary Cellsfor Chick Helper Factor (chJ) Expression

For some experiments it is necessary to know the extent to which the
endogenous virus genes are being expressed. The most convenient
method for assaying such expression is to determine the ability of the cells
to complement the defective Bryan high-titer (BH) strain of RSV in the
chick helper factor test. 12"1a For this test, cells are inoculated with
RSV(RAV-7), a virus mixture obtained after dual infection of cells with
BH-RSV and the leukosis virus, RAV-7. The BH-RSV in this mixture is
referred to as a pseudotype since, having incorporated the envelope
glycoproteins of RAV-7, it possesses the host range and antigenic properties of the leukosis virus. Infection of embryo cells that are not expressing
chf will result in the replication of more RSV(RAV-7) with the host range
of RAV-7, whereas infection of cells expressing chf results in the synthesis of both RSV(RAV-7) and RSV(RAV-0); the latter is a BH-RSV
pseudotype that has incorporated the envelope glycoproteins of the endogenous leukosis virus, RAV-0. The virus produced from "chfpositive" cells thus has the host-range of both RAV-7 (subgroup C) and
RAV-0 (subgroup E) and can be distinguished from RSV(RAV-7) by its
ability to form foci on quail embryo cells. RSV(RAV-7) is a particularly
useful pseudotype virus for the chf test since, having the subgroup C
host-range, it can infect most commercially available chicken embryo
cells; it differs from most subgroup C viruses, however, in that it is comn p . K. Vogt, in "Fundamental Techniques in Virology" (K. Habel and N. P. Salzman,
eds.), p. 198. Academic Press, New York, 1969.
n T. Hanafusa, H. Hanafusa, and T. Miyamoto, Proc. Natl. Acad. Sci. U.S.A. 67, 1797
(1970).
13 R. A. Weiss, W. S. Mason, and P. K. Vogt, Virology 52, 535 (1973).
388
[32]
pletely excluded from quail embryo cells, and thus provides a very sensitive system for the detection of RSV(RAV-0) pseudotypes.
For the test, a portion of the embryo cells (2 x 106/plate) are seeded
into a 35-mm petri dish containing 2 ml of PGM + polybrene (2/xg/ml-see below) and inoculated with approximately 105 focus-forming units of
RSV(RAV-7) shortly after seeding. On the second day after seeding, 1 ml
of the culture medium is removed and assayed on secondary cultures of
quail embryo cells for infectious virus as described in the following section. Control plates are inoculated with RSV(RAV-7) and appropriate
dilutions of a stock of RSV(RAV-0). Culture medium from embryos that
are chf~positive will generally induce 102 to l0 s foci in this quail focus
assay.
Quantitation of Transforming Virus: The Focus Assay
1. Preparation of Secondary Chicken Embryo Cultures

Quantitative assays of RSV are generally carded out on secondary
chick embryo cultures since more cells are available and the population is
more uniform. Two methods for making secondary cultures have been
employed in this laboratory and are described below.
A. METHOD 1
The soft-agar overlay is aspirated or poured gently from the primary
cultures, and the cell monolayer is washed with warm Tris-buffered
saline. Warm trypsin (0.05% in Tris-buffered saline, 5 ml/100-mm dish) is
added, and the plate is rocked gently for 1 min after which the trypsin is
aspirated from the dish. When the monolayer takes on a granular appearance, the cells are removed from the surface of the dish in 5 ml of GM by
means of a 5-ml hand pipette and rubber bulb of corresponding size. The
cell suspension, which should consist predominantly of single cells, is
transferred to a sterile tube; the plate is rinsed with another 5 ml of GM to
remove any remaining cells. This washing is added to the tube, and after
gentle mixing, the total number of cells obtained is determined with a
hemocytometer or an electronic cell counter; between 10-20 106 cells/
dish can be expected from one 100-mm primary plate, depending on the
number of days the cells have been in culture. Cells are seeded in 60-mm
tissue culture dishes containing 4 ml of GM (0.%1.2 x 106 cells/dish;
other sizes of dishes receive cells in proportion to their surface area, e.g.,
4-5 105 cells/35-mm dish). The density at which cells are seeded is
particularly important for the RSV focus assay since overseeding can lead
to suppression of focus formation.
[32]
389
B. METHOD 2
The medium is removed from the primary cultures and the monolayers
washed with Tris-buffered saline as described above. Only 2.5 ml of
0.05% trypsin are added to a 100-mm plate, and this is left on the cells
until the monolayer takes on a granular appearance. The cells are suspended with a hand pipette and rubber bulb and are added to 3 ml of
transfer buffer (TB: 10% calf serum in Tris-buffered saline). After determining the cell concentration, the suspension is seeded into 60-mm tissue
culture dishes containing 3 ml SM, at 1.2 106 cells/plate. The cells
settle and spread rapidly in this medium, which is changed, 0.5-2 hr after
seeding, to 5 ml GM.
Each method has its advantages. The first generally avoids any possibility of overtrypsinizing the cell monolayer since excess trypsin is removed from the plates. It also avoids the obligatory medium change of
method 2 that can be excessively time consuming in large assays. In
general, assays of RSV set up by method 2 have somewhat cleaner background monolayers since unhealthy or dying cells are removed with the
SM.
2. Inoculation and Agar Overlay

Secondary chick embryo cultures may be inoculated immediately after
seeding if prepared according to method 1 or after the medium change in
method 2. In general, virus is added to the cells within 2-4 hr after seeding. The cells remain susceptible to virus infection for about 1 day after
seeding, but inoculation with virus beyond this time reduces the efficiency
of the assay since one round of cell division appears necessary for the
expression of viral genes. 14,1s Suitable dilutions (usually 10-1, 10-2, 10-3)
of the virus are inoculated directly into the GM and left on the cells
overnight. The next day, the GM is replaced by OL (7 ml/60-mm plate),
which is allowed to harden for 10 min at room temperature before the
plates are returned to the incubator. The extended period of incubation of
virus and cells prior to addition of OL is acceptable since under conditions
of low multiplicity infection progeny virus is not detected until 16-18 hr
after infection. Secondary infection by progeny virus is thereby avoided.
The efficiency with which viruses of subgroups B, C, D, and E infect
cells is increased significantly (up to a 40-fold enhancement for focus
formation) if polycations are added to the infecting medium.16 Of these,
14 E. H. Humphries and H. M. Temin, J. Virol. 10, 82 (1972).
15 E. H. Humphries and H. M. Temin, J. Virol. 14, 531 (1974).
16 K. Toyoshima and P. K. Vogt, Virology 38, 414 (1969).
390
[32]
polybrene (Aldrich Chemical Co.) is the most efficient and least toxic to
cells. It may be added to the GM (2/~g/ml) on a routine basis unless
viruses of subgroup A are to be assayed, since infection by these viruses
is inhibited to a limited extent by polycations.
Focus assays on secondary cells of avian species other than chicken
are carded out in essentially the same manner. However, the OL added in
focus assays on quail embryo cells should always contain 1% chick serum
since this increases the efficiency of the assay severalfold.
3. Quantitation of Foci
The focus assay is usually counted 6--8 days after infection, although
foci may be visible from the fifth day. In order to quantitate the number of
foci a glass plate with a 2-mm2 grid (Technical Instrument Co., San Francisco, California) is attached to the slide holder on the mechanical stage of
an inverted microscope. The optimum total magnification for counting
foci is between 30 and 40 since this allows a 4 mm wide section of the
plate to be counted at one time. A 10 objective may be useful for
determining fine details of focus morphology. Cell clumps may resemble
foci, but generally can be distinguished from them by the presence of cells
radiating from the clump and by the absence of the rounded, refractile cell
type that characterizes RSV foci. Counting foci is usually facilitated if 1-2
ml of GM are added on top of the agar 2 hr prior to counting since this
appears to increase the refractile nature of the transformed cell and distinguishes it further from the background monolayer.
Cloning of R S V
The use of a homogenous virus population is crucial in many biochemical experiments. This is particularlytrue for R S V since in addition to the
natural variation that might be expected in a virus population, the growth
of nondefective sarcoma viruses in cellsis accompanied by the generation
of transformation-defective (td) mutants of RSV. 17,1sIn fact, continued
passage of a virus stock can result in a population that is predominantly
transformatiot~-defective virus. Therefore, it is important to produce
cloned virus stocks with which to perform experiments with RSV. T w o
methods are currently used to clone RSV: in the first,the virus is cloned in
the focus assay described above; in the second, the virus is cloned by
means of the soft-agar-colony assay.
17 p. K. Vogt, Virology 46, 939 (1971).
is S. Kawai and H. Hanafusa, Virology 49, 37 (1972).
[32]
391
1. Focus Assay Cloned RSV

Appropriate dilutions of the virus to be cloned are inoculated onto
secondary chick cells so that between 10 and 20 foci will develop on each
plate. The focus assay is carried out in the normal manner, except that an
intermediate reoverlay with 2 ml OL can be useful on the fifth day. A
well-isolated individual focus is chosen, and the transformed cells together with a small plug of overlying agar are withdrawn with a finely
drawn-out sterile Pasteur pipette. This material is transferred to a tube
containing 1 ml of TB, sonicated gently to destroy live cells, and used as
a fresh inoculum in a second focus assay. Several dilutions are again
inoculated onto secondary chick cells in order to obtain widely separated
foci. A total of three such clonings are generally .carried out prior to
growing up stocks of the virus. Usually, a focus from the third serial focus
assay is transferred directly onto a monolayer of secondary chick embryo
cells (1.2 106/60-mm plate). The transfer of these cells, after 4-5 days,
directly to a 100-mm plate results in complete transformation of the cell
population.
Virus can be harvested at this stage for future experiments. Significantly higher titers of cloned virus are obtained if the GM is supplemented
with 1% DMSO, and supernatant fluids are harvested on a daily basis.
Variations on this method are used by others (see this volume [33]).
Preparation of cloned virus stocks in this manner is sufficient for most
experiments. However, in this method the virus in the final cloned stock is
the result of several cycles of infection and will, therefore, even at this
stage, contain a small td virus component. In those experiments in which
any td virus is unacceptable the following method should be used.
2. Soft-Agar-Colony Assay Cloned RSV

The generation of td virus appears to be associated with a stage in the
replication of the nondefective sarcoma virus and may in fact occur during
reverse transcription.19 Cloned sarcoma virus lacking a td component can
thus be obtained if a single RSV transformed cell is propagated into a
virus-producing clone of cells. Since cells transformed by RSV acquire
the ability to grow as colonies when suspended in a soft-agar medium, ~ it
is possible to obtain clones of cells that have been transformed by a single
nondefective sarcoma virus.
The method for the agar-colony assay 21 employs a bottom layer of
~9 j. Hillova, D. Dantchev, R. Mariage, M. P. Plichon, and M. Hill, Virology 62, 197 (1974).
2o H. Rubin, Exp. Cell Res. 41, 149 (1966).
~1 j. A. Wyke and M. Linial, Virology 53, 152 (1973).
392
[32]
0.6% agar medium (BA) containing Japanese quail feeder cells and a top
layer of 0.36% agar medium (TA) into which infected cells are suspended.
Bacteriological dishes (60 mm) are used in preference to tissue culture
dishes in order to prevent adherence and overgrowth of the quail feeder
cells. A volume of 3.0 ml of bottom agar containing 5 x 105 quail cells is
added to each dish and allowed to harden at room temperature.
Chicken cells are infected in suspension by mixing 1-2 106 cells in 1
ml GM with virus at a multiplicity of infection of between 0.1 and 0.01
focus-forming units per cell. The mixture is incubated at room temperature for 1-2 hr, after which aliquots of dilutions of the infected cells are
added to 3 ml of TA in tubes at 45 . The agar suspension of cells is then
poured immediately onto the bottom agar medium. After the top agar has
hardened (20-30 min), the dishes are moved carefully to incubate at 37.
At day 4 and day 7, dishes are overlaid with 2 ml of OL medium. Colonies
of transformed cells are generally.visible to the naked eye by 7-12 days.
In some instances clumping of cells can occur during the 1-2 hr incubation of cells and virus. To ensure that only sifigle cells are suspended in
the top agar layer, the cell-virus mixture may be seeded into a 35-mm
plate and retrypsinized from the plate 4--6 hr later.
In order to establish transformed cell clones, colonies (about 1 mm in
diameter) are aspirated into a finely drawn-out Pasteur pipette and placed
into 16-mm wells of a 24-well cluster plate (Costar, Cambridge, Massachusetts) containing 1 ml of cloning medium. In general, no more than
12 colonies should be grown up in each 24-well plate in order to reduce the
possibility of cross-contamination. Plates are incubated at 37 and
checked every second day to determine the stage of growth. When the
cells are confluent they can be transferred to a 35-mm culture dish and
from there, via a 60-mm dish, to a 100-mm plate, from which cloned virus
can be harvested. It is important when transferring transformed cell
clones not to underseed the tissue culture dish; in general, seeding cells
onto twice the original area results in rapidly growing healthy cells. Once
established, transformed cell clones can be grown in GM containing 1%
DMSO.
Frozen Storage of Chick Embryo Fibroblasts
Chick embryo cells can be frozen and stored in a medium containing
high concentrations of serum and D M S O . The cells are trypsinizcd and
suspended in G M as described previously. After centrifugation(600-800
rpm, 5 min), the cells are resuspcnded in G M at a concentration of about
20 106 cells/ml. A n equal volume of freezing medium (11 parts G M , 5
parts calfserum, 4 parts dimethyl sulphoxide) is added drop by drop, and
[33]
GROWTH
OF
ROUS
SARCOMA
VIRUS
393
the cell suspension is added in 0.5-ml aliquots to 2-ml freezing vials (Vangard International, Neptune, New Jersey). The suspension is allowed to
freeze slowly (l/min is optimal) to - 7 0 before being transferred to a
liquid nitrogen freezer. The cells can be stored at - 7 0 , with reduced
survival.
Acknowledgments
Many of the techniques described in this article were developedin the laboratory of Dr.
P. K. Vogt, Universityof Southern California. The work of the author is supportedby Grant
VC-215 from the American Cancer Society.
[33] L a r g e - S c a l e G r o w t h o f R o u s S a r c o m a V i r u s
By RALPH E. SMITH
Biochemical investigations frequently require large quantities of cells
and virus. For example, the investigator may wish to isolate a specific
subcellular component that comprises a small fraction of the total material
present in the cell. In addition, biochemical investigations of viral structural components frequently require milligram quantities of purified virus.
Finally, the investigator may need to prepare large quantities of highly
infectious virus to take advantage of the unique ability of nondefective
Rous sarcoma virus (RSV) to rapidly and uniformly transform all cells in a
susceptible population. This article describes techniques for the growth of
transformed cells. It also outlines procedures for the preparation of
purified RSV. Certain of the biological techniques that are useful for working with the avian sarcoma viruses are presented separately in this volume
[32].
Media Reagents, Viruses, and Cells
Media and Reagents

Growth Medium (GM). Ham's F10 nutrient mixture (Flow Laboratories) is supplemented with 10% (v/v) tryptose phosphate broth (Difco), 5%
calf serum (Gibco), streptomycin (50/zg/ml), penicillin (50 U/ml), Fungizone (2 ~g/ml), and dimethylsulfoxide (1%). Vitamin B12 (Gibco; crystalline) is added to a concentration of 0.1% (w/v).
Maintenance Medium (MM). Ham's F10 is supplemented as above for
growth medium, except that the calf serum concentration is reduced to
1%.
METHODS IN ENZYMOL(~Y, VOL. LVIII

ISBN 0-12-181958-2
394
[33]
Primary Growth Medium (PGM). Ham's F10 is supplemented as above

for growth medium, except that the serum concentration is 8% calf serum
and 2% chicken serum (heat inactivated at 56 for 1 hr).
Overlay Medium. Part A: 2% Bacto agar (Difco) is prepared in water,
autoclaved, and stored at room temperature. Just before use, the agar is
melted in a boiling-water bath, cooled to 45 , and mixed 40 : 60 with part B
at 45 . Part B: Double-strength Ham's F10 (prepared by reconstituting
powdered medium with half the usual amount of water), 10% calf serum,
20% tryptose phosphate broth, 100/zg/ml streptomycin, 100 U/ml penicillin, and 4/zg/ml Fungizone (Squibb No. 43760, E. R. Squibb and Sons,
Inc., Princeton, New Jersey).
Phosphate-Free Maintenance Medium. Maintenance medium is made
from powdered Ham's F10 that has no phosphate (available as a special
order from most media suppliers), made without tryptose phosphate
broth, and buffered with 5 mM Tricene at pH 7.2.
Polybrene. Available as desiccated powder from Aldrich Chemical Co.
Prepare as a 2 mg/ml stock in water, autoclave, and store at 4 . Dilute in
Tris-buffered saline 1 : 10 to make a working stock of 0.2 mg/ml; add 0.05
ml (one drop from a Pastuer pipette is sufficiently accurate) to 5 ml medium to make a 2 tzg/ml solution.
Dilution Buffer. Ten percent calf serum in Tris-buffered saline.
Tris-Buffered Saline. O.14 M NaCI, 25 mM Trizma base, 5 mM glucose,
5 mM KC1, 0.7 mM Na~HPO4, 100 Ug penicillin/ml, 100 /zg
streptomycin/ml. The pH is adjusted to 7.4 with 1 N HC1 (final concentration of HC1 is approximately 0.018 N), and the material is sterilized by
filtration.
TryFsin in Tris-Buffered Saline. Trypsin (Gibco catalog no. 70730), at a
final concentration of 0.25%, is dissolved by constant stirring for 3-4 hr in
Tris-buffered saline and sterilized by filtration.
TE. 5 mM "Iris HC1 (pH 8.6) and 1 mM sodium ethylenediaminetetraacetate.
STE. 0.1 M NaC1, 10 mM Tris. HC1 (pH 7.5), and 0.1 mM sodium
ethylenediaminetetraacetate.
Viruses
Any of the sarcoma virus strains can be used for large-scale culture.
The most commonly used strains are the Schmidt-Ruppin strain of RSV,
the Prague strain of RSV, and the B77 strain of avian sarcoma virus.
Although most are used interchangeably, differences exist which merit
discussion. Both the Schmidt-Ruppin and Prague strains were originally
mixtures of viruses that have been cloned to reveal three viruses with
different antigenic and host-range properties. The Schmidt-Ruppin RSV
[33]
GROWTH OF ROUS SARCOMA VIRUS
395
strain was cloned into members of the A, B, and D subgroups, while the
Prague RSV strain was purified into members of the A, B, and C
subgroups. 1
Members of the A subgroup are difficult to work with because extracellular virus does not accumulate reproducibly to a high level. The
reason for this is not clear, but seems to be related to the tendency of
viruses of subgroup A to accumulate at the cell surface in large aggre-.
gates, which are difficult to remove and purify. An occasional preparation
will show a high titer of virus, but this result is not reproducible. Viruses
of subgroups B and D are slightly cytopathogenic, and prolonged passage
of infected cells is usually difficult. In contrast, viruses of subgroup C (the
B77 avian sarcoma virus and the C subgroup of Prague RSV) are attractive, since they are not cytopathogenic and extracellular virus accumulates readily.
Cells
Chicken cells are used for most biochemical procedures. Methods for
the preparation of chicken embryo fibroblasts (CEF) will not be given,
since detailed descriptions are available 2 (see also this volume [32]). However, a brief discussion of the appropriate cell for study is useful. All
chicken cells possess an endogenous virus (RAV-0), which may be either
completely latent or expressed as infectious particles. For certain biochemical applications, it may be important to know whether the endogenous virus is being expressed in infected cells. Procedures have been published for determining the level of RAV-0 production. 3 Breeding programs
have been initiated to provide fertile eggs from flocks that are free of
RAV-0 expression. Such fertile eggs are available from commercial
sources (SPAFAS, Inc., Storrs, Connecticut; Heisdorf and Nelson, Redmon, Washington).
Fibroblasts that are free of the endogenous virus of the chicken may be
obtained from other avian species, but these cells may have endogenous
viruses of their own. Further, other avian species often have unusual
susceptibility patterns, and only certain subgroups of virus may be used
efficiently. The most common nonchicken avian species are the Japanese
quail and the Pekin duck. Both are readily available from commercial
breeders (Truslow Farms, Inc., Chestertown, Maryland) and can be purchased throughout the year. Other more exotic avian species, such as the
i j. Tooze, " T h e Molecular Biology of Tumor Viruses." Cold Spring Harbor Lab., Cold
Spring Harbor, New York, 1973.
z p. K. Vogt, in "Fundamental Techniques in Virology" (K. Habel and N. P. Salzman,
eds.), p. 198. Academic Press, New York, 1969.
3 See this volume [32].
396
[33]
pheasant, are usually more difficult to obtain, and fertile eggs may be
available only once per year.
Mammalian cells can be transformed by avian sarcoma viruses, but
replication of the virus is blocked. Mammalian cells can therefore be used
to grow virus-transformed cells in the absence of virus replication. Infection of mammalian cells is an inefficient process, and large quantities of
highly infectious virus are usually required to obtain a few transformed
cells. Furthermore, one should keep in mind that the mammalian cells
should be frequently recloned, and cells selected for the particular attribute desired.
Preparation of Virus Stocks
The preparation of high-titer, well-characterized stocks of virus is a
key ingredient for the successful use of RSV in biochemical investigations. Stock virus preparation is usually carried out in two stages: cloning
of the virus stock and preparation of an intermediate stock.
Cloning. RSV cloning is started by performing a focus assay, in which
a monolayer (1.2 106 CEF in 5 ml GM/60-mm tissue culture dish) is
infected with 10-fold dilutions of RSV (dilutions are prepared in dilution
buffer). Infection with subgroups B, C, and D RSV are performed in the
presence of 2 /xg/ml polybrene. After 18-24 hr, the growth medium is
replaced with overlay medium, and the cells are incubated for an additional 7 days (an intermediate overlay of 3 ml of overlay medium at day 4
is usually helpful). Individual well-isolated loci are identified by visual
inspection and removed by drawing up a plug of agar directly over the
focus with a drawn-out sterile capillary pipette. The small amount of
material in the capillary pipette is transferred to 1 ml of dilution buffer.
After brief sonication in an ultrasonic bath, this material may be used to
start a second round of cloning, since sufficient virus is usually present to
obtain loci at a 10-1, 10-2, or 10-z dilution. Using this procedure, a virus
can be consecutively cloned 3 times in as many weeks. 4
Intermediate Stock Preparation. The relatively low virus titer of the
clone necessitates the preparation of a consistent, high-titer intermediate
virus stock for routine use. After investment of a large amount of time and
effort in the cloning of the seed virus, it is desirable to preserve the clone
in a pristine condition. This is accomplished by leaving the clone in its
milliliter of dilution buffer and entering the tube infrequently. To prepare
the intermediate stock, 0.1 ml of fluid from the isolated clone is put onto a
monolayer (1.2 106 CEF/60-mm dish) in the presence of 2 /zg/ml of
4 T. Graf, H. Bauer, H. Gelderblom, and D. Bolognesi,
Virology 43, 427 (1971).
[33]
G R O W T H OF ROUS SARCOMA VIRUS
397
polybrene, and cells are allowed to grow without overlay until all the cells
are transformed (one to two transfers). If other viruses are being grown at
the same time, it is a good practice to set aside a bottle of medium
specifically for the feeding of this stock in order to minimize the possibility of cross-virus contamination. The size of the-intermediate stock is
contingent upon the anticipated use and available storage space. One
should attempt to harvest sufficient virus to last for 1-2 years; otherwise,
the frequent removal of fluid from the original clone will necessitate recloning at an early date. Storage should always be at - 7 0 or lower.
Aliquots of the intermediate stock should be made in sufficiently small
volume that repeated freezing and thawing are not necessary, since the
titer of the virus will then drop below a useful level. An effective method
is to prepare several large containers of the virus stock (150 ml in a
screw-cap 200-ml polyallomer Sorvall centrifuge bottle works well), and
then thaw and dispense the virus into smaller samples (10 ml or less) as
the need arises.
Large-Scale Culture
Large amounts of virus are usually prepared by culture of virusproducing cells in roller culture bottles. Roller culture bottles are generally maintained at 37 , and since the bottles are sealed with screw caps,
humidification and carbon dioxide addition are unnecessary. Roller culture bottles are rotated at the rate of 6-10 revolutions per hour.
Preparation of Cells
There are two common methods for the preparation of virus-producing
cells in roller culture bottles. In method A, a small number of cells are
infected with virus in plastic tissue culture dishes, and these cells are
transferred until sufficient numbers are available to seed roller culture
bottles. The advantage of this method is that little intermediate stock
virus is used, and few cells are required for the initial seeding. Further,
the cells transform during passage in plastic dishes, and transformed cells
are easier to grow than nontransformed cells.
Specifically, method A involves the following procedure. Secondary
CEF are seeded into 150-mm plastic tissue culture dishes at 20 106
cells/dish and infected with 0.2 ml virus (undiluted) per dish. Cells are
subsequently trypsinized and replated into plastic tissue culture dishes at
20 x 106 cells/dish every 2-3 days until sufficient number of cells are
available for a roller culture bottle. In order to estimate the number of
cells required for a roller culture bottle, approximately 4 107 cells can
398
[33]
be obtained from a single 150-mm dish of confluent CEF. A single roller

culture bottle (1330 cm2; Bellco) requires inoculum from about five tissue
culture dishes. Cells are seeded into the roller culture bottle (2 108 cells
in 200 ml growth medium per bottle), and the medium is replaced the next
day. After an intervening day without feeding, the growth medium in the
roller culture bottle is replaced daily. After about 4-7 days, the cells can
be further subdivided by trypsinization and transfer to roller culture
bottles. The following procedure is used to tryp.sinize confluent CEF in
roller culture bottles. The medium is poured off, and 50 ml of warm 0.25%
trypsin in Tris-buffered saline are added to the bottle. The bottle is tilted
to distribute the trypsin solution over the entire surface of the bottle, and
then the trypsin solution is poured off. When visual inspection indicates
that the cell sheet has started to become granular (a few seconds is generally sufficient), 100 ml of growth medium are added, and the bottle is
vigorously rotated by hand to strip the cells from the surface of the bottle.
The cell suspension is transferred to a glass bottle and set aside momentarily. Since the roller culture bottle that has been trypsinized almost always
has cells remaining on the glass surface, it is fed with 200 ml of growth
medium and returned to the roller rack. The trypsinized cells are counted
and 2 108 cells are seeded into each new roller culture bottle in 200 ml
of growth medium. This procedure can be repeated as frequently as necessary to obtain the desired number of roller culture bottles. Four to five
days after seeding, sufficient numbers of cells can be recovered to seed at
least one new bottle (which amounts to a 2 : 1 split, since the trypsinized
bottle is retained).
One disadvantage of method A is that 2 weeks may be required to
obtain sufficient cells for seeding a single roller culture bottle. Although
this time interval makes it difficult to respond to fluctuations in experimental needs, one can compensate by having several large-scale stocks in
various stages of preparation. For example, it may be advisable to start
new large-scale preparations weekly.
In method B, cells are obtained for seeding roller culture flasks by
trypsinizing chick embryo fibroblasts, and the cells are seeded directly
into roller culture bottles without an intervening period of growth in plastic tissue culture dishes. One advantage of method B is that the investigator can rapidly respond to experimental needs by preparing large
numbers of roller culture bottles in a short time. Another advantage is that
no time is spent in plastic tissue culture dishes, and the risk of contamination is consequently diminished because less handling of the cells is
involved.
Specifically, method B involves the following procedures. Eleven- to
twelve-day-old embryos are pooled, so that two embryos are used to
[33]
399
prepare cells for one roller culture bottle. The embryos are decapitated,
minced, .and cells are dissociated with 0.25% trypsin in Tris-buffered
saline. A fluted trypsinization flask (Bellco) is helpful for this step. After
two consecutive trypsinizations (using 50 ml trypsin per embryo), the
cells are centrifuged at 1500 rpm for 15 min (800 g) in an International
refrigerated centrifuge, and the trypsin is decanted from the cell pellet.
The cell pellet is resuspended at the rate of 100 ml PGM per embryo, and
25 ml of a high-titer virus stock are added per embryo. Roller culture
bottles are seeded in 250 ml PGM per bottle. It is important to note that
polybrene cannot be used in this procedure, since it appears to change the
surface charge of the glass and prevents adherence of the cells. The primary cells will form a monolayer within 4-5 days, at.which time the roller
culture bottle is trypsinized (see method A) and cells are transferred to a
new roller culture bottle. Transformation is obtained within 4-5 days.
A disadvantage of method B is the requirement for large quantities of
intermediate virus stocks. It is impractical to infect a roller culture bottle
with less than 50 ml of virus-containing supernatant fluid, because transformation and high-titer production are considerably delayed when small
amounts of virus are used. The reason for the delayed appearance of
transformation probably stems from the large number of chicken cell
fragments and nonviable cells present in primary embryo fibroblast preparations that may absorb virus and prevent infection of the minority of
cells that eventually emerge to form the monolayer. Another disadvantage
of method B is that the cell monolayer is usually less. uniform than
monolayers prepared as outlined in method A. However, the rough appearance of the cell sheet is rapidly eliminated once trypsinization and
transfer of the cells in the roller culture bottles are initiated. Cells in roller
culture bottles are harvested, trypsinized, and transferred in the same
manner, whether prepared by method A or B.
Harvest of Virus and Cells

Once cells are confluent, virus can be harvested from roller culture
bottles for an extended period of time. 5 The amount of medium used, and
the frequency and duration of harvest, are influenced by the intended use
of the virus.
When the investigator is interested in harvesting virus for examination
of structural polypeptides, virus can be collected daily or every other day,
using 200 ml of growth medium per bottle for the first 7-10 days of culture, then maintenance medium for harvest periods of up to several
R. E. Smith and E. H. Bernstein, Appl. Microbiol. 25, 346 (1973).
400
[33]
months. Prolonged harvest periods are only possible with RSVtransformed cells, since they will grow back into areas of the roller culture
bottle that have been exposed due to overgrowth of the cell monolayer
and loss of the cell sheet. The investigator determines when to stop harvesting by monitoring the virus yield. A good rule of thumb is that a liter
of supernatant fluid should yield 1 mg of purified virus. When production
falls below this level, it is hard to justify the time and expense required for
continued harvest.
When the investigator is interested in harvesting virus for the isolation
of 70 S RNA, daily harvesting is adequate. The duration of harvest is
usually not as long, and 1 month of production is generally all that should
be attempted. When the investigator is interested in harvesting virus for
the isolation of 35 S RNA, the roller culture bottles should be harvested at
intervals of 2 hr, but can usually be harvested for only 3 weeks. An
automated system is available for this type of harvest, and this subject
will be covered in the next section.
Virus-producing cells can be harvested at any time during the preparation or harvest of large-scale cultures, as long as the cells have not begun
to decline due to prolonged harvest. Cells in roller culture bottles are
washed with 50 ml of Tris-buffered saline and detached from the surface
of the roller culture bottle with the aid of a rubber-bladed scraper (Bellco).
The cells are rinsed into a centrifuge bottle with the aid of 200 ml of
Tris-buffered saline and centrifuged at 800 g for 15 min in a refrigerated
centrifuge. The cell pellet is then processed further for experimental purposes, or stored frozen at - 7 0 .
Mechanization of Virus Production

The requirement for frequent harvest intervals for the production of 35
S RNA leads to considerable expenditure of time and effort when large
quantities of RNA are needed. For example, it is difficult to manually
harvest several roller culture bottles more than 3 times in a working day,
especially when one is also required to process the supernatant fluid. The
automated collection of roller culture bottle supernatant fluids 6 is therefore helpful. The frequent collection of supernatant fluid also has the
additional benefit of providing virus that has a high specific infectivity. 7
Roller culture bottles are prepared for automated harvesting in much
the same way as outlined previously. Cells are seeded into roller culture
bottles and grown until nearly confluent; then the roller culture bottles are
fitted with special caps and placed onto a harvesting device (Bellco Glass,
6 R. E. Smith and F. Kozoman, Appl. Microbiol. 25, 1008 (1973).
7 R. E. Smith, Virology 60, 543 (1974).
[33]
G R O W T H OF ROUS SARCOMA VIRUS
401
Inc.). It is helpful to start with subconfluent roller culture bottles, since

cell growth is usually vigorous as a consequence of frequent feeding.
Roller culture bottles are usually fed at the rate of 200 ml of maintenance medium per bottle per day, or about 17 ml per bottle every 2 hr.
This small amount of medium must evenly cover the bottom of the bottle,
so an accurate leveling of the harvesting device is required.
Bottles are harvested for 3 weeks, at which time virus production
declines. 8 When necessary, the monolayers in the roller culture bottles
may be inspected by placing an inverted microscope on a laboratory cart
and manipulating the bottles to view the monolayers without removing the
attached tubing.
We have experimented recently with the addition of 5% COz to the
roller culture bottles while they are being harvested. Two types of CO2
addition are satisfactory. In the first method, bottles are continuously
flushed with 5% CO2 in air at the rate of 30 ml/min for five bottles. The
second method consists of adding a 70-see pulse of 5% CO2 at the same
flow rate (30 ml/min for five bottles) immediately after each harvest. The
first method gives faster cell growth, and thus high virus yields are
achieved sooner. However, cells overgrow rapidly, and harvest is terminated sooner. When the second method is employed, roller culture bottles
require more time to achieve maximal virus yield, but high production is
maintained for a longer period of time, and less CO2 is consumed.
A special application of automated harvesting is the collection of zzp_
labeled virus for RNA. a Cells are grown in roller culture bottles until
confluent; then phosphate-free maintenance medium is added (50 ml/
bottle) for 4 - 6 hr. Ten to twelve milliCuries of [aZP]orthophosphate
(carrier-free; New England Nuclear) are added in 50 ml of fresh
phosphate-free maintenance medium, and the cells are labeled overnight
(12-18 hr). The supernatant fluid is removed, the roller culture bottle is
fitted with a special cap, and the bottle is harvested every 2 hr for up to 8
days after labeling.
Purification of Virus
The purification of RSV is divided into two steps, concentration and
purification.
Concentration
The large quantities of fluid generated by roller culture bottles require

the investigator to reduce the volume of the sample prior to processing the
a R. E. Smith, S. Nebes, and J. Leis, Anal. Biochem. 77, 226 (1977).
9 R. E. Smith and K. Quade, Anal. Biochem. 70, 354 (1976).
402
[33]
virus. The most convenient method is to pellet the virus with ultracentrifugational forces. This is done in three stages. The first stage is to
remove cells and large debris by centrifugation at 800 g for 15 min. The
second stage is to centrifuge the supernatant fluid at 8000 g for 15 min to
remove subcellular debris and improve the subsequent purification steps.
The third stage is to pellet the virus. Three fluid-volume capacities are
used. For up to 210 ml, a Beckman SW27 rotor is used, and centrifuged 30
min at 24,000 rpm (80,000 g). For up to 420 ml, a Beckman 35 rotor is
used, and centrifuged 45 min at 24,000 rpm (46,000 g). For up to 1200 ml, a
Beckman 19 rotor is used, and centrifuged 105 min at 18,000 rpm (21,000
g). For even larger fluid-volume capacities, zonal rotors are available that
process several liters per hour.
The virus pellet is resuspended (at a volume ~ to z-~ of the original) in
a buffer appropriate for the use of the virus. For most applications, TE (1
mM EDTA, 5 mM Tris HC1, pH 8.6) is a suitable buffer. Two special
cases deserve mention. The first is that the virus pellet can usually be
extracted for 70 S RNA, in which case it is appropriate to resuspend the
virus in STE (0.1 M NaC1, 1 mM EDTA, 10 mM Tris, pH 7.4). The second
is that virus pellets may be used to make complementary DNA, in which
case the investigator resuspends the virus pellet in a buffer appropriate for
DNA synthesis.
Purification
Highly purified virus is obtained in a two-step process: equilibrium
(isopycnic) density gradient sedimentation, followed by velocity (rate
zonal) density gradient sedimentation. The concentrated virus pellet is
resuspended thoroughly (a brief sonication in an ultrasonic cleaning bath
is helpful) and layered onto a 15-60% (w/v) sucrose (in TE) gradient. The
Beckman SW27 rotor is used most frequently in our laboratory for this
application. The virus suspension is layered onto the sucrose gradients,
and the rotor is centrifuged at 24,000 rpm (80,000 g) for 3-6 hr, at which
time the virus is visualized near the center of the gradient. The virus band
is removed with a Pasteur pipette, diluted 4-fold with TE, and the virus is
pelleted at 24,000 rpm (80,000 g) for 35 min in a Beckman SW27 rotor. The
virus pellet is resuspended in TE, layered onto a 20-40% sucrose gradient, and centrifuged 90 min at 24,000 rpm (80,000 g) in a Beckman SW27
rotor, or 40 min at 36,000 rpm (175,000 g) in a Beckman SW41 rotor. The
virus band is diluted with TE, and the virus is pelleted as before. This
virus can be stored at - 2 0 indefinitely. Virus concentration can be mea-
[33]
403
sured by the method of Lowry e t al. lo or estimated by an optical density

measuremen01 in which 1 optical density unit equals 158/zg protein.
Large-Scale Growth of Nontransforming Viruses
Growth of nontransforrning viruses in large-scale cultures is performed in essentially the same fashion as outlined previously for RSV.
Virus cloning is more difficult,since the viruses do not form loci in chick
embryo fibroblasts. However, viruses of subgroups B, D, and F and
certain members of subgroup A form plaques on CEF, 1z'13and virus can
bc recovered from a plaque by aspiration into a Pasteur pipette. The
preparation of rollerculture bottles of ceils infected with nontransforming
viruses is performed in the same manner as outlined for RSV, except that
growth medium is used for a longer period of time. Nontransformcd cells
produce less organic acids than transformed cells, and therefore they
sccm less susceptible to overgrowth or toxic effects from feeding on an
every-other-day schedule. The production of virus is followed by the size
of the virus pellet obtained upon concentration of the virus by ultraccntrifugation. Subgroups B, D, and F viruses arc somewhat cytopathogenic,
and healthy rollerculture bottles can decline duc to the replication of the
virus. This effectcan bc overcome by periodically adding I-2 10s uninfected cells to the roller culture bottle. Once the cells arc confluent, cell
transfer by trypsinization is performed in the manner outlined for RSV.
Cell yields arc lower than with transformed cells, so a longer period of
time is required to obtain a large number of virus-producing cells. Virus
yields are also lower, a consequence of a smaller number of cellsper roller
culture bottle.
Acknowledgments
The author wishes to express his thanks to Jann Bodie, Christine Chastain, Kay Izard,
and Sue Nebes, each of whom have substantially contributed to the development of the
procedures outlined in this review. Work performed in the author's laboratory was supported by National Institutes of Health Grant ROl-CA12323.
10 O. H. Lowry, N. J. Rosebrough, A. L. Farr, and R. J. Randall, J. Biol. Chem. 193, 265

(1951).
11 R. E. Smith and E. H. Bernstein, Appl. Microbiol. 25, 346 (1973).
12 T. Graf, Virology 50, 567 (1972).
13 C. Moscovici, D. Chi, L. Gazzolo, and M. G. Moscovici, Virology 73, 181 (1976).
404
[34]
[34] P r e p a r a t i o n o f S i m i a n V i r u s 4 0 a n d Its D N A
by GEORGE KHOURY a n d C H I N G - J U H L A I
Simian virus 40 (SV40), a small (45-nm diameter) protein-encapsidated

DNA virus, is a member of the papovavirus B genus (polyomavirus) of the
papovaviruses. 1 Its nucleoprotein core consists of a double-stranded
superhelical closed circular DNA genome with a molecular weight of
3.6 106, complexed with host-cell-derived histones. The genome contains approximately 5200 base pairs, which is enough genetic information
to code for only five or six proteins. SV40 was discovered in 1960 by
Sweet and Hilleman 2 as an inapparent infectious agent in rhesus monkeys.
It was capable of producing a lytic infection in African green monkey
kidney (AGMK) cells characterized by cytoplasmic vacuolization. Initially, SV40 was of medical interest because it was present as a contaminant of certain batches of poliovirus vaccine that had been administered to
humans. In addition, SV40 could produce tumors in suitable laboratory
animals and was able to transform human cells in tissue culture. More
recently, investigators have focused on the ability of SV40 to undergo
different interactions with various host cells. While SV40 replication in
permissive cells such as AGMK leads to amplification of the virus and cell
death, quite a different response occurs in nonpermissive cells. The inoculation of mouse cells with SV40, for example, results in an "abortive
infection" with only partial expression of the viral genome. SV40 DNA
does not replicate in mouse cells, and progeny virions are not produced. A
subpopulation of these abortively infected cells eventually become stably
transformed by the virus. The transformed state results from stable integration of the viral DNA into the host chromosome, continued expression
of the viral genes, and an altered cellular morphology and growth pattern
reflecting the presence of the viral function.
The simplicity of the SV40 genome and its ability to undergo transforming as well as lytic interactions with various cells has made it a model
system for virologists, geneticists, and molecular biologists who are int e r e s t e d in the study of gene regulation in eukaryotic cells. SV40 is probably the best understood of all DNA animal viruses and has provided
insight into the mechanisms of DNA replication, transcription, and posttranscriptional controls, translational regulation, and DNA-protein interactions. The use of purified viral DNA has been crucial to most of these
1 j. L. Melnick, A. C. Allison, J. S. Butel, W. Eckhart, B. E. Eddy, S. Kit, A. J. Levine,
J. A. R. Miles, J. S. Pagano, L. Sachs, and V. Vonka, Intervirology 3, 106 (1974).
z B. H. Sweet and M. R. Hillernan, Proc. Soc. Exp. Biol. Med. 105, 420 (1960).

ISBN 0-12-181958-2
[34]
SIMIAN VIRUS 4 0 AND ITS D N A
405
studies. In addition, SV40 DNA has been used to study the physical and
chemical properties of supercoiled molecules, to assay restriction enzymes, and to evaluate DNA-protein interactions. The recent determination of the complete nucleotide sequence for SV40 a,4 will certainly enhance its value for numerous future biological and biochemical investigations. Simple methods for the production of SV40 and SV40 DNA are
described below.
A. Growth of SV40
Cell Cultures
SV40 produces a lytic infection in monolayer cultures of African green
monkey kidney cells. Several continuous cell lines are available including
BSC-1,5 CV-1,6 and Vero 7 as well as derivative clones of these lines. Titers
between 10s and 10a plaque-forming units (PFU) per milliliter of cell lysate
usually can be obtained. Primary or secondary AGMK cells apparently
yield 5-10 times this amount of virus. However, primary cells are expensive and tedious to prepare, which makes them a less satisfactory choice
for most laboratories.
Stock Virus
A number of strains of SV40 are commonly used including 776 (SVS),
777, and Va 4554. Many strains differ slightly in their DNA sequence and,
in some cases, this difference is reflected in variations in the restriction
enzyme cleavage patterns or in the morphology of the plaques that the
virus induces.8 Stocks of virus are generally prepared by infecting cells (in
Eagle's minimal essential medium, for example, supplemented with antibiotics, 0.03% glutamine, and 2% fetal bovine serum; abbreviated,
MEM-2) with less than 0.01 PFU/cell of a seed virus preparation. The low
multiplicity avoids the generation of defective virions that have been
a V. B. Reddy, B. Thimmappaya, R. Dhar, K. N. Subramanian, B. S. Zain, J. Pan, C. L.
Celma, and S. M. Weissman, Science (in press).
4 W. Fiers, R. Contreras, G. Haegeman, R. Rogiers, A. Van de Voorde, H. Van
Heuverswyn, J. Van Herreweghe, G. Volckaert, and M. Ysekaert, Nature (London) (in
press).
5 H. E. Hopps, B. C. Bernheim, A. Nisalak, J. H. Tjio, and J. E. Smadel, J. lmmunol. 91,
416 (1963).
6 F. C. Jensen, A. J. Girardi, R. V. Gilden, and H. Koprowski, Proc. Natl. Acad. Sci.
U.S.A. 52, 53 (1964).
r E. Earley, P. H. Peralta, and K. M. Johnson, Proc. Soc. Exp. Biol. Med. 125, 741 (1967).
8 T. J. Kelly and D. Nathans, Adv. Virus Res. 21, 85 (1977).
406
[34]
shown to arise after high multiplicity infections or during undiluted serial

passage of virions. 8
For inoculation, a small volume of virus, sufficient to cover the cell
sheet, is added to confluent monolayer cultures. After adsorption at 37
for 1-2 hr, cultures are fed with a suitable maintenance medium, such as
MEM-2. Cultures infected by wild-type SV40 are incubated at 37,
whereas cultures infected by temperature-sensitive mutants of SV40 are
usually grown at 32. A suitable stock of virus is prepared by simply
freeze-thawing monolayer cultures of SV40-infecjted cells 3 times in their
own maintenance medium. This is generally done at a time when the cell
monolayer shows extensive cytopathic effects (rounding or detachment of
greater than 80% of the cell sheet), an event that occurs, for example,
10-14 days after incubation at 37 of cells infected at a multiplicity of 0.01
PFU per cell. The stock virus can be titered, by the plaque assay of serial
dilutions on monolayers of primary or continuous African green monkey
kidney cultures, a SV40 is quite stable and retains its titer for long periods
of time when frozen in portions at - 2 0 to - 7 0 . Since a significant amount
of virus remains associated with cell membranes, clarification of viral
stocks by pelleting and discarding cellular debris will result in the loss of
large fractions of the virions. If it is important to clarify or concentrate
stocks of SV40, this can be achieved without significant losses by purifying virions. A convenient method for virus purification, which employs
the equilibrium banding of virus in CsC1, is described below in the second
procedure for preparation of viral DNA.
B. Preparation of SV40 DNA

Since defective virions do not generally arise in significant quantities
before several serial passages at high multiplicities of infection, it is convcnient to infect confluent monolayers of kidney cells at a multiplicity of
approximately l0 PFU/cell for the preparation of SV40 DNA. There are
two basic methods of viral DNA preparation; the choice of technique
depends on the intended use for the SV40 DNA as is explained below.
Preparation of SV40 DNA by the Selective Extraction Method of Hirt

In 1967, Bernard Hirt developed a convenient method for selective
extraction of polyoma virus DNA (and other low-molecular-weight DNAs
including SV40 DNA) from infected cells. 10 This method, which is based
g R. Dulbecco and M. Vogt, J. Exp. Med. 99, 167 (1954).
10 B. Hirt, J. Mol. Biol. 26, 365 (1967).
[34]
SIMIAN VIRUS 40 AND ITS D N A
407
on the removal of high-molecular-weight cellular DNA by precipitation in

the presence of high salt concentrations and sodium dodecyl sulfate
(SDS), has a number of advantages. It is both rapid and efficient, producing as much as 2-6 times the yield of SV40 DNA per equivalent of infected
cells as the alternative method mentioned below(e.g., 10-20/~g per 150
cm 2 flask of cells). Since the Hirt extraction procedure is based on the
isolation of intracellular DNA, rather than encapsidated viral DNA, cells
are harvested earlier than if DNA were to be prepared from purified
virions. A disadvantage of the Hirt method is the presence of small
amounts, less than 1%, of contaminating cellular DNA in most preparations. While this level of contamination is acceptable for most purposes,
there are certain circumstances under which essentially pure SV40 DNA
is required and for which the second method is preferable, e.g., in vitro
labeling of viral DNA (see below). A slight modification of the Hirt extraction procedure in which the supercoiled SV40 DNA is further purified by
dye-bouyant-density centrifugation n is described below.
Reagents and Materials
Hirt lysing solution (0.6% sodium dodecyl sulfate, 10 mM EDTA at
pH 7.4)
Sodium chloride, 5M
RNase (Worthington Biochemicals, Freehold, New Jersey; preheat a
1 mg/ml solution to 80 for 10 min to inactivate traces of DNase)
Potassium acetate solution (20%) at pH 6.0
Ethidium bromide (aqueous solution, 2 mg/ml)
Standard saline citrate (SSC solution = 0.15 M sodium chloride, 15
mM sodium citrate)
Isopropyl alcohol saturated with CsC1 by adding a few crystals to
about 10 ml isopropyl alcohol
Phenol, saturated with a solution of 10 mM Tris HCI at pH 8.0, and
10 mM NaC1
Sucrose solutions: 5% and 30% (w/v) in 0.1 M NaC1, 10 mM
Tris HC1 at pH 7.5, and 2.5 mM EDTA
Cell monolayers are infected with SV40 at 1-10 PFU/ceI1. Between 3
and 5 days after infection when cells show early cytopathic effects (some
cells are rounded or vacuolated, but all cells remain attached to surface),
cultures are harvested as follows. The cell sheet is gently washed twice
with cold PBS. After draining, about 1-2 ml of Hirt lysing solution is
added per 150-cm ~ flask and allowed to remain on the surface for 20 min at
room temperature. The cell lysate is then scraped into a Sorvall centrifuge
H R. Radloff, W. Bauer, and J. Vinograd, Proc. Natl. Acad. Sci. U.S.A. 57, 1514 (1967).
408
[34]
tube, the volume measured, and 5 M NaC1 added to a final concentration

of 1 M. The tube is covered, inverted gently (to reduce potential shearing
forces) l0 times, and stored at 4 overnight. After centrifugation for 40 min
at 25,000 g and at 0, the supernatant fraction is gently removed into
another glass centrifuge tube. RNase is added to a concentration of 10
/~g/ml followed by incubation at 37 for 30 min. After the addition of an
equal volume of saturated phenol, the mixture is shaken for 10-15 min at
room temperature, and the phases are separated by centrifugation at
3000-5000 rpm for 10 min at room temperature. The aqueous (upper)
phase is removed into a centrifuge tube and to it is added 0.1 volume of
20% potassium acetate and 2-2.5 volumes of cold ethanol. After mixing,
this material is stored at - 1 0 to - 2 0 for at least 8 hr. The DNA that
precipitates in ethanol is pelleted by centrifugation at 10,000 rpm for 20
min at 0 in a Sorvall SS-34 rotor. After the supernatant liquid is carefully
decanted, the DNA precipitate is dissolved in 1-2 ml of SSC solution at
pH 7. For subsequent dye-density equilibrium centrifugation, the DNA
solution should contain ethidium bromide at approximately 200/.~g/ml and
CsC1 with a density of 1.56 g/ml. This is conveniently accomplished by
first weighing an empty nitrocellulose or polyallomer (see below) centrifuge tube, and then adding the DNA solution that has been increased to
6.5 ml with SSC solution. After the addition of 0.75 ml of ethidium
bromide stock solution (2 mg/ml), the tube is again weighed and the content weight increased to 7.5 g by the further addition of SSC. After dissolving 6.5 g of CsCl into each tube, the density should be approximately 1.56.
This may be checked by weighing 1 ml of the solution or by measuring its
refractive index n (n = 1.3880). Tubes are filled with mineral oil, capped,
and centrifuged to equilibrium, e.g., 40,000 rpm for 48 hr at 10 in a
Beckman 50.1 rotor. Other rotors, e.g., SW41 and type 40, also can be
used by appropriately adjusting the speed and time of centrifugation.
Nitrocellulose centrifuge tubes are preferable to polyallomer tubes
since they are more transparent, i.e., bands can be more easily visualized,
and are easier to puncture. However, DNA adheres more readily to nitrocellulose. Thus, for the preparation of very small amounts of DNA,
some investigators prefer to use polyallomer ultracentrifuge tubes and
siliconized glass centrifuge tubes.
After centrifugation to equilibrium, two bands of DNA should be detectable near the center of the tube separated by approximately 1 cm. The
lower of these bands contains the supercoiled DNA (higher density) and
can be removed by tapping from the bottom of the tube or by pipetting
from the top. Small amounts of DNA are more easily visualized with an
ultraviolet (UV) lamp in the dark. However, exposure to UV should be
[34]
SIMIAN VIRUS 40 AND ITS DNA
409
limited to avoid thymidine dimer formation or nicking of DNA. Ethidium

bromide is frequently removed by passing the DNA through a column of
cationic resin, e.g., Dowex-50. Perhaps the easiest way to remove the dye
is by extracting several times with equal volumes of isopropyl alcohol
saturated with CsC1. The pink, intercalating dye is fractionated into the
organic (upper) phase and removed. The CsC1 can then be removed from
the aqueous phase by dialysis against an appropriate buffer (e.g., 10 mM
NaC1 and 10 mM Tris-chloride at pH 8.0). Since supercoiled DNA molecules other than SV40 may still be present, i.e., mitochondrial DNA,
SV40 DNA can be further purified by sedimentation in 5-30% sucrose
gradients at 30,000 rpm for 9 hr at 4 with an SW41 rotor. These conditions
will place supercoiled SV40 DNA near the center of the gradient.
Preparation of SV40 DNA from Purified Virions

As an alternative to the Hirt procedure, SV40 DNA can be isolated
from purified SV40 virions. The principle advantage of this method is the
high degree of purity of the DNA released from SV40 capsids. This purity
is esPecially important when viral DNA is subsequently labeled to very
high specific activities by in vitro methods (see "In vitro labeling" below)
in order to detect small amounts of SV40 DNA in transformed cells. As
mentioned, the disadvantages of this method are that it requires more
time, both for the growth of virions and for the additional steps in the
purification procedure, and that the yield of DNA is generally lower.
Reagents and Materials

CsC1 solution, density = 1.4 (n = 1.3723)
CsC1 solution, density = 1.34 (n = 1.3670)
Tris-NaCl buffer (10 mM Tris HC1, pH 8.0, 10 mM NaC1)
Sarkosyl: 10% solution of sodium sarcosinate
Cell monolayers are infected at about 10 PFU/ceI1. At 6--10 days, when
monolayers show extensive cytopathic effects (see above), the remaining
cells in infected flasks are scraped from the surface. Media and cells are
poured into large centrifuge tubes, and the cells are pelleted at about
5000 g for 10 min at 4. The supernatant fluid is decanted and saved; the
cell pellet from 5-10 flasks is resuspended in 16 ml of the supernatant fluid. After the addition of 2 ml of 0.25% trypsin (0.025% final
concentration) and 2 ml of 13% desoxycholate (1.3% final concentration),
the resuspended cells are incubated at 37 for 20 min. Cellular debris is
removed by pelleting at 10,000 g for 10 min at 4, and the supernatant fluid
is gently layered onto a 10-ml cushion of CsCl (p = 1.4 g/ml) in an SW27
410
[34]'
nitrocellulose tube. This tube and additional tubes are filled to capacity
with the medium saved from the infected cells; centrifugation is carried
out at 24,000 rpm for 3 hr (or 12,000 rpm overnight) at 4 in an SW27 rotor.
The virions, which under these conditions enter about 1 cm into the CsC1
cushion, can be visualized best with a high-intensity lamp in a dark room
as a blue translucent band. The band is removed in 1-2 ml of CsCI,
preferably by dripping through the bottom of a punctured tube. The density of this virus-containing CsC1 should be 1.33-1.34 g/ml (n = 1.3660),
and this material can be diluted directly into 7-8 ml of CsC1 of the same
density (1.34 g/ml) in a nitrocellulose tube for either an SW40 or a 50Ti
rotor. The virus is centrifuged to equilibrium, e.g., at 35,000 rpm for 16 hr
at 4 in a 50Ti rotor. After visualization and removal of the SV40 band
(which should be near the center of the tubes; 19 = 1.34), the virus is
dialyzed for 24 hr against the Tris-NaC1 buffer at 4 with 3-4 changes of
the dialysate. The protein capsids can be removed from virions by incubation at 50 for 30 min in 1% Sarkosyl. After two extractions with phenol
that is saturated with Tris-NaC1 buffer, the viral DNA, which remains in
the aqueous phase, can be either dialyzed at room temperature or precipitated in 2.5 volumes of cold ethanol prior to use. A further purification of
supercoiled SV40 molecules from relaxed molecules can be obtained as
described in the previous purification method by dye-bouyant-density
centrifugation and/or sucrose gradient sedimentation. However, when
SV40 DNA is extracted from virions these steps are frequently
unnecessary.
C. Radiolabeling and Quantitation of SV40 DNA
Radiolabeling of SV40 DNA

The intended use for SV40 DNA frequently requires that it be
radiolabeled. Addition of [3H]thymidine (10-100/zCi/ml) or [~4C]thymidine
(0.01-0.1 /zCi/ml) to the medium of infected cells generally provides
labeled DNA with a specific activity of 105 cpm//zg and 10~-10 a cpm//zg,
respectively. Radiolabeling with carrier-free [a2P]orthophosphate can
produce SV40 DNA with specific activities of 1-2 106 cprn//zg if the
cellular phosphate pools are lowered. This can be achieved by the following protocol.
Cells are washed with phosphate-free medium several times prior to
infection and fed with phosphate-free medium supplemented with 2%
dialyzed fetal calf serum, antibiotics, and glutamine. Twenty-four hours
after infection, the monolayers are again washed three times with
phosphate-free medium. Cells are fed agaitl with phosphate-free medium
[34]
SIMIAN V l n U S 4 0 AND ITS D N A
411
that contains approximately 100 /zCi/ml of carder-free [32p]orthophosphate. The time for harvesting radiolabeled viral DNA depends on
the method to be used for purification (see above):
In vitro Labeling of SV40 DNA
A number of laboratories now are routinely labeling DNA preparations in vitro using a method referred to as nick-translation. There are at
least two recently published procedures that describe in detail the methodology involved. ~2'13 Both procedures require DNase I and DNA
polymerase I and involve the nicking and removal of nucleotides from the
unlabeled input DNA followed by repair synthesis with radiolabeled nucleoside triphosphates. The advantages of in vitro labeling compared with
in vivo labeling include the ability to obtain radiolabeled DNA within
hours by a relatively simple procedure and the production of DNA with
extremely high specific activities, e.g., l0 s cpm//.~g using deoxynucleoside
triphosphates, two of which are o32p-labeled. Lower specific activities
also can be achieved by limiting the DNase concentration, the specific
activity of the radiolabeled precursors, or the time of incubation.
In vitro labeled SV40 DNA recently has been used to obtain qualitative
and quantitative information relating to the integrated viral genomes in
transformed cells. These high-specific-activity probes have been successfully used for reassociation kinetic analyses la and for detection of viral
DNA transferred to nitrocellulose filters t4 by the Southern blotting procedure.15
Quantitation and Assay of SV40 D N A
The yield of SV40 DNA is most easily determined spectrophotometrically. The concentration of a sample in milligrams per milliliter is approximately equal to the optical density at 260 nm divided by 20. Small
amounts of viral D N A are quantitated more accurately by reassociation
kinetics 16 or by gel electrophoresis. All three forms of SV40 DNA, i.e.,
form I (supercoiled molecules), form II (relaxed circles), and form III
(linear duplexes), are easily separated by agarose gel electrophoresis. In
order to quantitate small amounts of SV40 DNA (approximately 0.05-1.0
/zg), a DNA sample of unknown concentration is electrophoresed in parallel with serial dilutions of a DNA preparation of known concentration.
12 T. Maniatis, S. G. Kee, A. Efstratiadis, and F. C. Kafatos, Cell 8, 163 (1976).
~3 p. W. J. Rigby, M. Dieckmann, C. Rhodes, and P. Berg, J. Mol. Biol. 113, 237 (1977).
14 M. Botchan, W. Topp, and J. Sambrook, Cell 9, 269 (1976).
15 E. Southern, J. Mol. Biol. 98, 503 (1975).
~6 L. D. Gelb, D. E. Kohn, and M. A. Martin, J. Mol. Biol. 57, 129 (1971).
412
[35]
The gel is stained with ethidium bromide solution (1 /~g/ml in the electrophoresis buffer) for 15-30 min and visualized under UV light. By comparing the intensity of staining of the unknown sample with a similar
standard, one can rather accurately assess the DNA concentration.
Perhaps the best criterion of purity for a viral DNA preparation is its
restriction enzyme cleavage pattern. Restriction cleavage maps of the
SV40 genome are known for numerous restriction enzymes 17 (also see
Kelly and Nathansa). In a typical assay, a small quantity of DNA is
cleaved with a restriction endonuclease, and the product is subjected to
agarose or polyacrylamide gel electrophoresis. If the DNA is 32p-labeled,
the ge! can be examined by autoradiography; otherwise, the restriction
pattern of the cleaved DNA can be analyzed by ethidium bromide staining
as described above.
A number of experiments have recently been carried out to directly
test the biological activity of either native SV40 DNA or segments of the
viral genome. In the first approach, permissive cells can be effectively
inoculated with the viral DNA in the presence of DEAE-dextran.18 This
technique has made it possible to assay the relative infectivity of various
viral DNA preparations. In addition, the technique has enabled investigators to propagate altered forms of viral genomes, either naturally occurring molecules or those constructed in vitro. An alternative procedure
for introducing viral DNA into nonpermissive cells involves coprecipitation of the DNA with calcium phosphate~a; this method had proven more
reliable, for obtaining morphologically transformed cells, and it has been
employed successfully to determine the limits of the viral DNA regions
required for establishment and maintenance of cell transformations. Finally, direct microinjection of viral DNA into cells has been valuable for
studying several biological interactions between viruses and cells. 2 While
this technique is limited by the number of cells that can be microinjected,
the ability to inoculate efficiently large numbers of molecules into single
cells has led to a number of important observations.
~7 K. J. D a n n a and D. N a t h a n s , Proc. Natl. Acad. Sci. U.S.A. 69, 3097 (1969).
18 j. M c C u t c h e o n and J. M. Pagano, J. Natl. Cancer Inst. 41, 351 (1968).
19 F. L. G r a h a m and A. J. v a n der Eb, Virology 52, 456 (1973).
2o A. G r a e s s m a n , Exp. Cell Res. 60, 273 (1970).
[35] P u r i f i c a t i o n a n d A s s a y o f M u r i n e L e u k e m i a V i r u s e s
By CHARLES J. SHERR and GEORGE J. TODARO
Murine type C viruses have been classified into two major groups
based on their potential for producing neoplastic disease. The mouse

ISBN 0-12-181958-2
[35]
MURINE LEUKEMIA VIRUSES
413
leukemia viruses (MuLVs) can replicate in cultures of fibroblastic and

epithelioid cells and generally do not produce cytopathic effects or morphological transformation. By contrast, the mouse sarcoma viruses
(MSVs) are defective in their ability to replicate, require "helper"
MuLVs for their continued propagation in cell culture, and produce morphological transformation of susceptible cells in vitro. This article describes techniques for growing murine leukemia viruses in tissue culture
and discusses methods for purifying virus in sufficient quantities for subsequent biochemical studies. Since the success of these techniques involves determinations of viral titer, several of the commonly used, quantitative in vitro assays.for viral replication are also described.
Growth and Purification of Murine Leukemia Virus
Sources of Cells and Viruses
MuLVs have been classified into several different host-range types.

These include ecotropic viruses that replicate predominantly or exclusively in cells from the same or closely related species; xenotropic viruses, restricted from replicating in mouse cells, but able to grow in cells
from heterologous species; and amphotropic viruses that grow in both
mouse and nonmouse cells. Ecotropic viruses from laboratory mice (Mus
musculus) can be further distinguished by their ability to grow preferentially in cells derived from certain prototype mouse strains. The N-tropic
viruses replicate Preferentially in cells derived from NIH Swiss mice (Ntype cells) while B-tropic viruses grow best in cells derived from BALB/c
mice (B-type cells); NB-tropic viruses grow readily in either N- or B-type
cells. 1 The restriction of N-tropic viruses in B-type cells and vice versa is
controlled by the Fv-1 locus. 2,3
Successful propagation of MuLVs in culture depends on the host range
of the virus chosen for study as well as the cell line chosen to support its
replication. For routine propagation of ecotropic virus stocks, the feral
mouse SC- 1 cell line is widely used since it lacks Fv- 1 restriction and can
be infected with N-, B-, or NB-tropic viruses. 4 Alternatively, cell lines
such as NIH/3T3 5 or BALB/3T3 n established from the prototypic N- and
B-mouse strains, respectively, can be employed. Xenotropic viruses can
J. W. Hartley, W. P. Rowe, and R. J. Huebner, J. Virol. 5, 221 (1970).
2 T. Pincus, J. W. Hartley, and W. P. Rowe, J. Exp. Med. 133, 1219 (1971).
3 T. Pincus, W. P. Rowe, and F. Lilly, J. Exp. Med. 133, 1234 (1971).
4 j. W. Hartley and W. P. Rowe, J. Virol. 19, 19 (1976).
5 j. L. Jainchill, S. A. Aaronson, and G. J. Todaro, J. Virol. 4, 549 (1969).
6 S. A. Aaronson and G; J. Todaro, J. Cell. Physiol. 72, 141 (1968).
414
[35]
be propagated to high titer in the SIRC rabbit corneal cell line (CCL 60,
American Type Culture Collection [ATCC], Rockville, Maryland), in
MvlLu mink lung cells (ATCC line CCL 64), or in Cf2Th canine thymus
cells (Naval Biomedical Research Laboratories, Oakland, California). Viably frozen virus seed stocks can be obtained from the ATCC, from Pfizer
Laboratories (Maywood, New Jersey), from ElectroNucleonics, Inc.
(Bethesda, Maryland), or from the Office of Resources and Logistics of
the National Cancer Institute (Bethesda, Maryland). Seed cultures of
cells infected with virus also can be obtained frona numerous investigators
working in the field.
Infection of Cultured Cells

Cells used for virus production are grown at 37 on the surfaces of
75-cm~ (250-ml) plastic tissue culture flasks (Falcon Labware, Cockeysville, Maryland) or in 490-cm2 polystyrene roller bottles (Coming Glass
Works, Coming, New York) at 1.0-1.5 rpm using Dulbecco's modified
Eagle's medium supplemented with 4 mM L-glutamine, 4.5 mg/ml glucose, 10% heat-inactivated fetal calf serum, 100 units/ml penicillin, and
100/zg/ml streptomycin (Grand Island Biological Co., Grand Island, New
York). All cell lines should be periodically monitored for opportunistic
infections by fungi, adventitious viruses, mycoplasma, and resistant
bacteria.
Prior to infection, cells are seeded at 1 10~ cells per 75 cm2 tissue
culture flask in 12 ml of complete medium containing 2 tzg/ml polybrene
(Aldrich Chemical Company, Milwaukee, Wisconsin). After 24 hr, the
medium is removed and the cells are exposed at 37 to 1.5 ml of undiluted,
filtered (0.45 ~m) medium from a virus-producing culture, or with a 10fold dilution of concentrated virus seed stock obtained as described
above. After 1 hr, 12 ml of complete medium are added to the culture, and
the medium is changed after 24 hr and subsequently at 3-day intervals.
Cells are subcultured when they reach confluence using 0.1% trypsin in
phosphate-buffered saline to dissociate the cells from the plastic substrate
and from each other. Supernatants from the cultures are tested at weekly
intervals for the production of viral RNA-dependent DNA polymerase
(see "Supernatant Reverse Transcriptase Assay" below). In general, cultures become highly positive for viral replication from 1-2 weeks after
infection.
Virus Purification
Infected cells are grown and seeded into plastic 490-cmz roller
bottles (0.5 rpm) each containing 50--70 ml of medium. After cells attach,
[35]
415
the roller belt speed is increased to 1.0-1.5 rpm. Cells should be maintained at 60--100% confluency to insure high levels of virus production,
but should not be allowed to remain at confluency for long periods as virus
production will diminish, and cells will tend to slough from the surfaces of
the bottles. Medium containing virus is sterilely decanted at 24- to 48-hr
intervals and is rapidly cooled and kept at 4 . If maximum infectivity is
desired, roller bottles should contain only 20--40 ml of medium and should
be "rapidly harvested" at 3- to 6-hr intervals. In general, 10TM viral particles (approximately 1-2 mg of protein) can be obtained from approximately 5 liters of medium.
All further steps are performed at 4 . Virus-containing fluids are
clarified of cells and debris by centrifugation at 12,000 g for 10 min at 4.
The supernatant fluid is then centrifuged over a cushion of 20% glycerol
containing 0.05 M "Iris HC1, pH 7.8, and 0.1M KC1 (105,000 g for 90 min)
to pellet the virus. If the supernatant fluid is large, virus can be concentrated by ultrafiltration prior to pelleting (30 psi, XM 300 membrane;
Amicon Corp., Lexington, Massachusetts). Virus pellets are resuspended
in 0.05M Tris HC1, pH 7.8, containing 10 mM KC1 and 0.1 mM EDTA
using gentle passage through a #18 gauge syringe needle. The suspension
is then loaded over 20-60% (w/w) linear sucrose (ribonuclease-flee) gradients and banded isopycnically at a density of 1.16 g/cc. Visible gradient
bands can be harvested from the top of the tube using a Pasteur pipette, or
gradients can be collected by needle puncture from the bottom, and the
density of the band determined by refractometry. If the latter technique is
employed, aliquots of each tube can be assayed for viral reverse transcriptase activity (as described below) to confirm the position of the viral
band. Pooled, isopycnically banded virus is then diluted in the same buffer
used for banding and is repelleted over a glycerol cushion as described
above. For studies of viral structural proteins, viral pellets can be directly
frozen at - 7 0 or can be suspended in 0.05 M Tris HC1, pH 7.8, containing 0.1 mM EDTA prior to freezing. For maximum viral infectivity, preparation of viral RNA, or for synthesis of complementary DNA transcripts,
suspended virus particles should not be frozen but should be stored ,at 4
and used as soon as possible.
With very large volumes of virus-containing medium, continuous-flow
centrifugation is the most efficient method for purifying viral particles.
Virus fluids are clarified of cells and debris by a single pass through a
Beckman J-CF continuous-flow rotor (15 liters/hr at 12,000 rpm; calculated to retain particles with a sedimentation coefficient of -> 10,000 S).
Clarified virus fluids (20 liters maximum volume) are then passed through
a CF-32 rotor at 5 liters/hr, 32,000 rpm, and sedimented into a 20-60%
(w/w) linear sucrose gradient. Particulate and protein bands are allowed
to equilibrate for an additional 0.5 hr after cessation of virus fluid flow.
416
[35]
The rotor contents are unloaded at 2000 rpm using a dense " c h a s e "
solution (62% sucrose) and are fractionated on the basis of optical density
(254 nm) and buoyant density (1.120-1.180 g/cc). Virus-rich fractions
(concentrated 500-1000 times at p -1.16 g/cc) are normalized to ---15%
sucrose by dilution with 0.05 M Tris HCI, pH 7.8, and centrifuged in a
Beckman type 35 rotor at 35,000 rpm for 1.5 hr (143,000 g). The virus
pellets are then suspended and stored as described above.
Assays for Murine Leukemia Virus

I. Enzymatic and Immunological Techniques
A. SUPERNATANTREVERSE TRANSCRIPTASE ASSAY

This test depends upon the packaging of RNA-dependent DNA
polymerase activity in extracellular type C viral particles, r The extremely
sensitive assay is useful for recognizing cultures infected by any type C
virus, including all classes of MuLVs. a'a In general, a linear correlation
exists between the amount of reverse transcriptase activity in a viral stock
and the number of infectious viral particles. 1 Supernatant fluids can be
serially collected and assayed from a single cell culture allowing the study
of the time course of viral infection in a single flask.
Procedure
1. Five to fifteen milliliters of medium from an infected culture are

clarified by centrifugation (12,000 g for 10 rain at 4).
2. Virus in the supernatant is pelleted by centrifugation through a
cushion of 20% glycerol containing 0.05 M "Iris HC1, pH 7.8, and 0.1 M
KC1 (105,000 g for 90 rain at 4).
3. Pellets are resuspended in 0.1 ml of0.05M Tris HC1, pH 7.8, containing 0.1 M KC1, 10-~ M dithiothreitol (DTT), and 0.1% Triton X-100.
4. An aliquot of the suspension (0.01 ml) is added to a 0.060 ml total
reaction mixture containing 0.05 M Tris. HCI~ pH 7.8, 0.06 M KCI,
2 10-a M DTT, 6 10-4 M manganese acetate, 0.02 A2m units poly rA
(Miles Laboratories, Kankakee, Illinois), 0.02 A2e0 units oligo dT,2-1s
(Collaborative Research, Waltham, Massachusetts), and 3 10-6 M
[all] Thymidine 5'-triphosphate, tetrasodium salt ([aH]TTP, 60,000 dpm/
pmol; New England Nuclear Corporation, Boston, Massachusetts).
r H. M. Teminand D. Baltimore,Adv. Virus. Res. 17, 129 (1972).
8j. Ross, E. M. Scolnick,G. J. Todaro, and S. A. Aaronson,Nature (London), New Biol.
231, 163 (1971).
9 M. M. Lieber, C. J. Sherr, and G. J. Todaro,Int. J. Cancer 13, 587 (1974).
10j. R. Stephenson, R. K. Reynolds,and S. A. Aaronson, Virology 48, 749 (1972).
[35]
417
5. The reaction mixture is incubated for 60 min at 37 and terminated

by the addition of 3.0 ml 10% trichloroacetic acid (TCA).
6. Acid-precipitated mixtures are collected on Millipore filters (type
HA, white plain), washed 5 times with 5% TCA, dried, and counted in a
toluene-based scintillation cocktail.
Remarks. This test can be used for assaying not only type C viruses but
also the other classes of reverse transcriptase-containing viruses. 11 In
particular, certain mouse cell lines or cultured tissues produce type B
viruses like the mouse mammary tumor virus (MTV) as well as another
newly described group of "B-like" retroviridae (M43212). The latter two
groups of viruses contain reverse transcriptases that prefer magnesium
(10-z M) to manganese as the divalent cation in reactions performed with
synthetic templates? T M
Some cells also release nonviral coded polymerases into the culture
medium. Some of these enzymes can incorporate [3H]TrP into acidprecipitable poly(dT) product. The cellular polymerases have different
template-primer requirements, T M and control assays using different
template primers [e.g., poly (rC)-oligo (dG), poly (rA)-poly (dT)] can be
performed in conjunction with the reverse transcriptase test. Inhibition of
reverse transcriptase activity with specific antibodies also can be used to
confirm the nature of the polymerase being measured? 5
In the above assay, confluent uninfected host cells in one 75-cm2
(250-ml) flask generally produce less than 3 103 cpm of acid-precipitable
poly (dT) product. Polymerase activities greater than 1 104 cpm/flask
are scored as positive for viral replication, although activities in positive
cultures are routinely much higher than this minimum level. Although the
assay can be performed in the absence of synthetic templates, thus allowing only the endogenous viral RNA to be transcribed, the endogenous
reaction is considerably less sensitive.
B . RADIOIMMUNOASSAYS FOR V I R A L STRUCTURAL PROTEINS
The genomes of murine type C viruses code only for a limited number
of viral proteins. These include the reverse transcriptase, the major envelope glycoprotein (gp70), and at least four additional low-molecularweight polypeptides (p30, p15, ppl2, and pl0). By convention, the virion
11 A. J. Dalton, J. L. Melnick, H. Bauer, G. Beaudreau, P. Bentvelzen, D. Bolognesi, R.
Gallo, A. Graft], F. Haguenau, W. Heston, R. Huebner, G. J. Todaro, and U. I. Heine,
lntervirology 4, 201 (1974).
12 R. Callahan, C. J. Sherr, and G. J. Todaro, Virology 80, 401 (1977).
13 E. M. Scolnick, E. Rand, S. A. Aaronson, and G. J. Todaro, Proc. Natl. Acad. Sci.
U.S.A. 67, 1789 (1970).
14 M. Green and G. F. Gerhard, Prog. Nucleic Acid Res. 14, 187 (1974).
15 W. P. Parks, E. M. Scolnick, J. Ross, G. J. Todaro, and S. A. Aaronson, J. Virol. 9, 110
(1972).
418
[35]
structural proteins are designated " p " for protein, "gp" for glycoprotein,
and " p p " for phosphoprotein, followed by their apparent molecular
weight in thousands of daltons. 16Although standard immunological techniques such as immunofluorescence, immunodiffusion, and complement
fixation can be employed to assay for type C viral structural proteins, the
ability to purify viral proteins to homogeneity, label them to high specific
activity with 1251, and generate antisera that react specifically with them
has led to the development of extremely sensitive radioimmunoassays
(RIAs) for each of the above virion components.
Viral antigenic determinants can be assigned operationally to one of
three major classes of immunological reactivities, although in actuality
there exists a broad spectrum of different antigenic reactivities for each of
the viral proteins. Antigenic determinants that are shared in common by
type C viruses isolated from different mammalian species are designated
"interspecies-specific," while antigenic determinants shared in common
among a group of isolates from a single species (e.g., all MuLVs from M.
musculus) are classified as "species-specific" or "group-specific" determinants. In addition, "type-specific" antigenic determinants are those
that distinguish different isolates within a group (e.g., xenotropic and
ecotropic MuLVs).16
Four virion proteins (reverse transcriptase, p30, pl0, and gp70) possess predominantly group-specific and interspecies-specific reactivities
while p15 and ppl2 appear to exhibit predominantly type-specific reactivities. For general assays of MuLVs isolated from laboratory mice, p30
and gp70 are the most useful proteins for radioimmunoassay studies since
they are relatively easy to purify, are present in higher concentrations
than other proteins in virions, and contain readily detectable group- and
interspecies-specific antigenic determinants. Techniques for the purification of antigenically reactive MuLV p30,17"1sgp70,19 p15, p12, 2 and pl0 ~1
structural proteins as well as MuLV reverse transcriptase 22 have been
described in detail. Each of these proteins can be labeled with 125Ito high
specific activity (5/xCi//xg) using the chloramine T method~ and can be
used as "tracer antigens" in competitive RIAs that can detect subnanogram quantities of competing viral protein.
le j. T. August, D. P. Bolognesi, E. Fleissner, R. V. Gilden, and R. C. Nowinski, Virology
60, 595 (1974).
17 S. Oroszlan, C. L. Fisher, T. B. Stanley, and R. V. Gilden, J. Gen. Virol. 8, 1 (1970).
is E. M. Scolnick, W. P. Parks, and D. M. Livingston, J. Immunol. 109, 570 (1972).
is M. Strand and J. T. August, J. Biol. Chem. 248, 5627 (1973).
s o R. C. Nowinski, E. Fleissner, N. H. Sarkar, and T. Aoki, J. Virol. 9, 359 (1972).
21 M. Barbacid, J. R. Stephenson, and S. A. Aaronson, J. Biol. Chem. 251, 4859 (1976).
22 j. M. Krakower, M. Barbacid, and S. A. Aaronson, J. Virol. 22, 331 (1977).
23 F. C. Greenwood, W. M. Hunter, and J. S. Glover, Biochem. J. 39, 114 (1963).
[35]
419
A representative RIA protocol for MuLV p30 proteins is described

below. This assay utilizes an antiserum prepared to the feline leukemia
virus (FeLV) and detects those antigenic determinants shared by the p30
proteins of MuLVs and FeL. For screening infected cells in tissue culture, assays for interspecies determinants are generally most useful, since
detectable immunological differences between various MuLV isolates are
minimized. Competition assays are initiated using conditions in which
50% of the l~5I-labeled MuLV tracer p30 protein is bound in soluble immune complexes formed with antibodies to F e L p30 protein. Following
incubation with various dilutions of competing protein, immune complexes are precipitated with a second antiserum to 7 S immunoglobulins,
and residual radioactivity in the precipitates is determined. The dilution of
antiviral serum required to bind 50% of the labeled tracer antigen must be
determined by titration as described below. Although the outlined procedure utilizes a rabbit antiserum to viral p30 protein and a goat antiserum
to rabbit IgG, antiviral sera produced in other species are equally suitable
as long as appropriate "second antibodies" are employed. The procedure
can be adapted for assaying the different viral proteins and depends only
on the availability of tracer antigens and antisera.
Reagents
1. RIA buffer: 0.01 M potassium phosphate, pH 7.5, containing 0.1 M
NaC1, 0.01 M ethylenediaminetetraacetate (EDTA), and 0.5% normal
rabbit serum.
2. 125I-labeled p30 protein from MuLV (Rauscher strain or some hightiter equivalent). Specific activity should be approximately 5/~Ci//zg.
3. Rabbit antiserum to p30 protein of FeLV.
4. Goat antiserum to rabbit 7 S globulin ("second antibody").
I. Procedure for Titration of Antiviral Antibodies

1. Serial dilutions of rabbit antiserum to viral p30 protein are prepared
in RIA buffer. The dilutions required will depend upon the titer of antibodies in a particular serum but, for most sera, 2-fold dilutions from
10-2-10 -6 will suffice.
2. Titration reaction mixtures (0.5 ml) include: 0.3 ml RIA buffer, 0.1
ml of 125Iantigen (diluted in RIA buffer to give 104 cpm/0.1 ml), and 0.1 ml
of rabbit antiserum at various dilutions. Two "background" control tubes
without antiserum should be included and should contain 0.4 ml of RIA
buffer and 0.1 ml of 1251antigen.
3. Reactions are incubated for 2 hr at 37 and for 18 hr at 4.
4. A titered excess of second antibody is added (i.e., enough to precipitate all rabbit IgG in a 0.5 ml reaction) to each tube.
420
[35]
5. Incubation is continued for 1 hr at 37 and for 3 hr at 4.

6. Immune precipitates are collected by centrifugation (5000 g for 10
min at 4), washed with phosphate-buffered saline, and radioactivity in the
pellets is determined in a gamma counter.
7. The dilution of rabbit 'antiserum that binds 50% (5000 cpm) of 125Ilabeled tracer antigen is chosen as the antiserum titer for competition
assays.
2. Procedure for Competitive RIA
1. Competing antigen extracts of infected, cultured cells are prepared

by scraping the cells from the culture flask with a rubber policeman,
washin~ the cells in serum-flee medium, and suspending the washed cells
in 0.5 ml phosphate-buffered saline. The cells are lysed by two cycles of
rapid freezing and thawing in acetone-Dry Ice, followed by the addition
of 10-15 volumes of ether. The ether is evaporated under a nitrogen
stream, debris is removed by centrifugation (5000 g for 10 min), and the
supernatant is used as the competing antigen. Protein is determined by the
Lowry method 24 and should ideally be at least 2 mg/ml.
2. Two-fold serial dilutions of competing antigen are prepared in RIA
buffer.
3. Two background control tubes contain 0.4 ml RIA buffer without antiserum or competing antigen. Two "binding" control tubes contain
0.3 ml RIA buffer and 0.1 ml titered rabbit antiserum at appropriate dilution (see procedure 1 above).
4. Competition reaction mixtures contain 0.25 ml RIA buffer, 0.1 ml
titered rabbit antiserum at appropriate dilution, and 0.05 ml diluted competing antigen extract.
5. Reaction mixtures are incubated for 2 hr at 37.
6. 12q-labeled tracer p30 protein (diluted in RIA buffer to give 104
cpm/0.1 ml) is added to all tubes.
7. Reaction mixtures are incubated for 1 hr at 37 and for 18 hr at 4.
8. Immune complexes are precipitated using the second antibody as
described for steps 4-6 in procedure 1 above.
9. Calculations: The "working range" (R) of the assay is determined
by subtracting [average cpm in binding control tubes (Bi)] - [average
cpm in background control tubes (Ba)]. The degree of competition (% C)
for each competition reaction mixture is determined as:
%C = [cpm in competition reaction mixture] - Ba 100
R
2~ O. H. Lowry, N. J. Rosebrough, A. L. Farr, and R. J. Randall, J. Biol. Chem. 193, 265
(1951).
[35]
421
Remarks. All assays should include a standard competition curve generated using known concentrations of purified, unlabeled viral p30 protein
as the competing antigen. This permits quantitative estimation of the
amount of p30 antigen in any heterogeneous competing protein mixture, if
the total quantity of competing protein at each dilution is known. 25 Technical problems in RIAs for type C viral proteins are similar to those
encountered using general RIA techniques and arereviewed elsewhere. 26
Antisera to viral proteins can be prepared by immunization with
detergent-disrupted, isopycnically banded virions. 25 While such sera contain antibodies to many different viral structural proteins, the specificity
of the RIA depends only on the purity of the 125I-labeled tracer antigen,
Second antibodies from several species are commercially available, but
can be prepared at minimal cost using purified IgG as the immunizing
antigen.
Assays for group-specific determinants of p30 proteins can be designed using antiserum to the homologous p30 protein (i.e., anti-MuLV
p30) instead of antiserum to FeLV. TM In such tests, minor type-specific
differences between different MuLV p30 proteins will be reflected in the
slopes of the competition curves as well as in the final percent competition
at "saturating" levels of competing protein. 25'26
While RIAs can be used as a general screen for MuLVs from laboratory mice (Mus musculus), they also can be used for detecting MuLVs
from several other Mus species (e.g., M. caroli 27 and M. cervicolor28). The
immunological properties of some of the latter isolates are, however,
distinctly different from those of laboratory mouse MuLVs.
H. Virological Assays
A. MIXED CULTURE CYTOPATHOGENICITY(THE "XC TEST")
The XC cell line was established from a Wistar rat tumor induced with
the Prague strain of Rous (avian) sarcoma virus. 29 These cells contain the
Rous sarcoma virus genome but do not release infectious viral particles.
When infected with certain MuLVs, XC cells undergo syncytium formaz5 C. J. Sherr and G. J. Todaro, Virology 61, 168 (1974).
26 W. H. Hunter, in " T h e Handbook of Experimental Immunology" (D. M. Weir, ed.), p.
608. Davis, Philadelphia, Pennsylvania, 1967.
27 M. M. Lieber, C. J. Sherr, G. J. Todaro, R. E. Benveniste, R. Callahan, and H. G. Coon,
2~ R. E. Benveniste, R. Callahan, C. J. Sherr, V. Chapman, and G. J. Todaro, J. Virol. 21,
849 (1977).
29 V. Klement, W. P. Rowe, J. W. Hartley, and W. E. Pugh, Proc. Natl. Acad. Sci. U.S.A.
63, 753 (1969).
422
[35]
tion resulting in the appearance of plaques. The number of plaques produced is a direct function of the number of infectious viral particles, a
Procedure
1. Mouse cells (see below) are seeded sparsely into 60-mm plastic
petri dishes ( - 1 105 cells/plate) and are grown at 37 in Dulbecco's
modified Eagle's medium containing 10% fetal calf serum and antibiotics.
2. One day later, the cells are exposed to medium containing 25/zg/ml
DEAE dextran (Sigma Chemicals, St. Louis, Missouri) or 2 #g/ml Polybrene (Aldrich Chemical Company) for 1 hr at 37.
3. The medium is removed, and cultures are infected with dilutions of
virus-containing tissue culture medium ( - 1 . 0 ml) for 2 hr on a rocking
platform. The medium is changed after infection, and cultures are kept at
37 for 5 additional days until the cells reach confluence.
4. The medium is removed, and the cell layers are exposed for 10-30
sec to two GE germicidal bulbs at - 1 0 cm (60 erg/mm2/sec).
5. XC cells in fresh medium (10 6 cells/plate) are added to the irradiated monolayer, and cultures are incubated at 37 until plaque formation occurs (generally about 4 days after the addition of XC cells). Medium is changed 2 days after the addition of XC cells.
6. Cultures are washed with phosphate-buffered saline, fixed with
methanol, and stained with Giemsa (Grand Island Biological Corp.,
Grand Island, New York). The plaques appear as holes in the XC cell
sheet which contain some multinucleated giant cells.
Remarks. The XC test is useful for detecting most ecotropic MuLVs
that replicate efficiently in culture, but it cannot be used for either
xenotropic or amphotropic viruses, or for certain genetically transmitted
ectropic viruses that replicate poorly. Thus the test detects only a subset
of MuLVs, and a negative XC test does not mean that virus is not present.
For routine assays, the mouse SC-1 cell line is ideal since it lacks Fv-1
restriction and can be infected with N-, B-, or NB-tropic viruses. 4 Mouse
secondary embryo cultures a may be used to type N- and B-tropic isolates, the NII-I/3T35 and BALB/3T36 cell lines can also be used as prototype N- and B-cells, respectively.
Ultraviolet (UV) irradiation of the mouse cells prior to the addition of
XC cells kills the mouse cells at a faster rate than it inactivates their
capacity to produce virus. The dose of irradiation is approximate and
should be determined empirically for each mouse cell line. In addition, the
time after infection when UV irradiation is performed must be long
enough to allow plaque development but short enough to prevent the
a0 W. E Rowe, W. E. Pugh, and J. W. Hartley, Virology 42, 1136 (1970).
[35]
423
development of secondary, "satellite" plaques. 30 This will depend on the

relative "infectivity" of the viral stock and the susceptibility of the mouse
cells employed.
B . SECONDARY FOCUS-FORMATION
("S+L - T E S T S " )
The Moloney sarcoma-positive, leukemia-negative (S+L -) strain of

mouse sarcoma virus can transform susceptible cells from a variety of
mammalian species. Nonproducer transformed clones containing the
S + L - genome do not produce infectious viral particles but do express
certain viral coded antigens in addition to the function(s) required for
transformation, al Several nonproducer cell lines undergo further morphologicai alterations when infected by MuLVs. These changes (focus or
plaque formation) form the basis for several quantitati.ve assays for infectious MuLVs.
Nonproducer dedvants used in these assays include clones derived
from mouse 3T3 cells (line "Cl16"3z), from feline CCC cells (line
"SC"as), or from mink MvlLu cells (line MiCllO4). Mouse C116 cells support the replication of ecotropic MuLVs while 8C and MiCI~ cells are
employed in assays for xenotropic MuLVs. In general, the procedures
employed are similar for each cell line. The number of loci or plaques
produced is directly proportional to infectious virus concentration. 3z-a4
Procedure
1. Cl16 cells are grown in Dulbecco's modified Eagle's medium with

10% calf serum, MiCI~ cells in RPMI-1640 medium with 1.0% fetal calf
serum, and 8C cells in McCoy's 5A medium with 15% fetal calf serum. All
cultures are supplemented with antibiotics.
2. Cells are seeded in 60-mm plastic petri dishes at - 1 105 cells/
plate.
3. On the next day, the medium is withdrawn and cells are treated
with medium containing 25 /zg/ml DEAE dextran (8C cells) or with 1
/zg/ml Polybrene (MiCI~ or Cl16 cells) for 1 hr.
4. Serial dilutions of virus-containing medium (0.5-1.0 ml/plate) are
adsorbed onto the cells for 1 hr at 37 with occasional rocking of the
plates.
5. Four milliliters of medium are then added, and plates are incubated
at 37 in a 5% COE atmosphere.
at R. H. Bassin, L. A. Phillips, M. J. Kramer, D. K. Haapala, P. T. Peebles, S. Nomura, and
P. J. Fischinger, Proc. Natl. Acad. Sci. U.S.A. 68, 1520 (1971).
az R. H. Bassin, N. Turtle, and P. J. Fischinger, Nature (London) 229, 564 (1971).
az p. j. Fischinger, C. S. Blevins, and S. Nomura, J. Virol. 14, 177 (1974).
a4 p. T. Peebles, Virology 67, 288 (1975).
424
[35]
6. The medium is changed 24 hr after infection, and again at 3-day

intervals until foci can be scored (generally 5-12 days after infection).
Remarks. The foci induced in Cl16 or 8C cells appear as lytic areas
containing a few grape-like clusters of round, refractile cells. This occurs
because cells undergoing morphological change tend to round up and
detach from the plate. By contrast, foci produced in MiCll cells appear as
clustered, refractile cells that overgrow the monolayer and tend to remain
attached to the plate. As a result MiCI1 foci are most readily scored by
direct microscopy while C116 or 8C "plaques" can be enumerated most
readily after fixation and staining of the cells. For quantitative assessment
of MuLV titers, foci should be read as early as possible after infection to
avoid the possibility of satellite focus-formation resulting from further
rounds of infection, a~-a4 Use of early-passage MiCll cells is necessary
since, at late passages, these cells spontaneously revert to multilayer
growth and tend to round up and float in the culture medium.
Concluding Remarks
Investigators working with murine type C viruses should be aware that
genetic information coding for these viruses is vertically transmitted in the
germ line of all mouse species, and that cells in tissue culture can begin to
produce "endogenous," genetically transmitted viruses. Moreover,
mouse cells have the capacity to produce at least four distinct classes of
retroviridae with different host-range, antigenic, and genomic properties. z8 Activation of endogenous viruses may occur after infection of
mouse cells with standard MuLVs; therefore, viruses purified for biochemical studies should be tested for the acquisition of new properties not
detected in the original infecting stock. The mouse cell lines recommended here for virus propagation (SC-1, BALB/3T3 and NIH/3T3) all
have endogenous viral information that can be activated by the infectious
leukemia virus and that may be able to recombine with the infecting virus.
For example, stocks of NB-tropic Rauscher MuLV propagated in cells
derived from Balb/c mice are frequently contaminated by endogenous
xenotropic viruses that were activated in the host cells. These viral stocks
contain mixtures of virions with different antigens and genomes, as well as
phenotypically mixed particles and potential recombinant viruses. Such
events would not be detected by assaying viral particles for reverse transcriptase activity but could be scored by appropriate immunological or
virological tests. The prevalence of many different retroviral classes
among different species of Mus also suggests that currently employed
assays for viral replication may be insufficient in screening for new isolates. Thus, the failure to detect routine type C viruses in certain assays
for viral replicative functions may not mean that viruses are absent.
[36]
HUMAN ADENOVIRUSES
425
[36] H u m a n A d e n o v i r u s e s : G r o w t h , P u r i f i c a t i o n , a n d
Transfection Assay
By MAURICE GREEN and WILLIAM S. M. WOLD

Human adenoviruses (Ads) provide some of the most powerful
eukaryotic systems for the experimental analysis of DNA tumor virus
replication, cell transformation, and the molecular biology of human cells
(reviewed in Green et al. 1-3). Thirty-one well-defined human adenovirus
serotypes (Adl-31) have been identified. Most types fall into five distinct
groups based upon several properties including DNA-DNA homology,
immunological cross-reactivity o f " e a r l y " viral proteins (tumor antigens),
and tumorigenicity in newborn hamsters. The groupings are as follows:
group A (Adl2, 18, 31), group B (Ad3, 7, 11, 14, 16, 21), group C
(Adl, 2, 5, 6), group D (Ad8-10, 13, 15, 17, 19, 20, 22-30), and group E
(Ad4).
Human Ads have a duplex, linear, noncircularly permuted DNA
genome of 20-25 106 daltons. 4 The DNA molecules have inverted terminal repetitions and a protein linked, probably covalently, to both 5'termini. 3 The Ad virion (virus particle) is assembled from about 13 viral
polypeptides and the viral DNA genome into an icosahedron structure,
consisting of an external capsid and an inner core containing viral DNA
and at least two basic proteins.
The productive infection of human KB cells by Ad2 has been the most
widely used model for Ad replication. 5 Recently, Ad5 (very closely related to Ad2) and HeLa cells have been used as well. The genomes of
these viruses have been dissected with a variety of restriction endonucleases, and detailed physical and genetic maps constructed.6 -9 Productive
1M. Green, Annu. Rev. Biochem. 39, 701 (1970).
2 M. Green, J. T. Parson, M. Pina, K. Fujinaga, H. Cattier, and I. Landgraf-Leurs, Cold
Spring Harbor Symp. Quant. Biol. 35, 803 (1970).
W. S. M. Wold, M. Green, and W. Biittner, in "Adenoviruses" (D. P. Nayak, ed.).
Dekker, New York, 673-768 (1978).
4 M. Green, M. Pina, R. Kimes, P. C. Wensink, L. A. MacHattie, and C. A. Thomas, Jr.,
5M. Green and G. E. Daesch, Virology 13, 169 (1961).
6 j. Sambrook, M. Botchan, P. Gallimore, B. Ozanne, U. Pettersson, and J. Williams, Cold
Spring Harbor Syrup. Quant. Biol. 39, 615 (1974).
rp. A. Sharp, P. H. Gallimore, and S. J. Flint, Cold Spring Harbor Symp. Quant. Biol. 39,
457 (1974).
8 j. F. Williams, C. S. H. Young, and P. E. Austin, Cold Spring Harbor Syrup. Quant. Biol.
39, 427 (1974).
9 H. S. Ginsberg, M. J. Ensinger, R. S. Kauffman, A. J. Mayer, and U. Lundholm, Cold
Spring Harbor Syrup. Quant. Biol. 39, 419 (1974).

All rights of reproduction in any form reservea.
ISBN 0-12-181958-2
426
[36]
infection of "permissive" human cells results in the production of approximately 200,000 virus particles per cell, and the eventual death of
the infected cell. 1 The viral genome is transcribed and replicated in the
cell nucleus. There are at least two stages of expression of the Ad genome,
"early," before initiation of viral DNA replication at 6-7 hr postinfection,
and "late. ''2 Early genes, which are transcribed by host cell enzymes, are
arranged in four noncontiguous gene blocks, two on the r-strand and two
on the 1-strand, and represent about 30% of the single-stranded genome.
Proteins encoded by early genes (as many as 12 or 14 early polypeptides
have been identified) probably function in viral DNA replication, regulation of transcription, and cell transformation. Late genes, which code
virion structural proteins, are mainly on the r-strand, and represent most
of the remainder of the genome. Most exciting, late mRNA consists of
"spliced" mRNA molecules (reviewed in Wold et al. 3), wherein the main
coding body of the RNA is covalently linked to short "leader" RNA
sequences coded by DNA regions far upstream from each gene.
Ad infection of nonpermissive or semipermissive rodent cells may
result in the stable transformation of a small fraction (10-5--10-~) of the
cells. The transforming information [presumably gene(s)] has been
localized on the left-hand end of the r-strand within map position 1 to 7.5
(the entire genome consists of 100 map units), in that cells can be transformed by transfection with DNA restriction fragments containing this
region, H and all virion-transformed cells retain this region6 and express it
as RNA 7 and protein. The transforming "genes" of Adl2 (group A) and
Ad7 (group B) are in a similar region. The transforming region lies within
an early gone block.
In this article, we describe in detail methods used in our laboratory for
growth of cells for the propagation and plaque assay of all human Ad
serotypes. We also describe the isolation of viral DNA, and the transfection assay to determine Ad DNA infectivity and to localize Adtransforming genes.
Growth of Cultured KB Cells Used for the Propagation of Adenoviruses
The 31 human Ads (prototype strains are available from the American
Type Culture Collection) are grown in suspension cultures of human KB
cells for large-scale virus propagation, and in monolayer culture (petri
dish) for plaque assay of Ad infectivity. HeLa cells are used in some
laboratories with comparable results. There is some evidence that HeLa
lM. Green, Cold Spring Harbor Syrup. Quant. Biol. 27, 219 (1962).
~1F. L. Graham, P. J. Abrahams, C. Mulder, H. L. Heijneker, S. O. Warnaar, F. A. J. de
Vries, W. Ficrs, and A. J. van der Eb, Cold Spring Harbor Syrup. Quant. Biol. 39, 601
(1974).
[36]
HUMAN ADENOVIRUSES
427
and KB cell lines are derived from the same H e L a cell parental line. Both
lines are available from the American Type Culture Collection. Below are
described the preparation of growth media and the cultivation of KB cells
for large-scale production of virus and for infectivity assay.
Preparation of Cell Culture Media
Triple-distilled water is used to prepare all media. Monolayer cultures
are grown in serum-supplemented Eagle's minimal essential medium
(MEM), TM a synthetic medium consisting of a balanced salt solution and
amino acids and vitamins essential for growth of cultured human cells; it
is often supplemented with nonessential amino acids. Joklik-modified
MEM which lacks calcium (to prevent cell clumping) and contains a 10fold higher phosphate concentration (increased buffering capacity) is used
for suspension cultures. There are several ways to prepare media: by
assembly of laboratory-prepared or commercially available concentrates
( 10 to 200) of balanced salts, amino acids, and vitamins; by purchase
of complete 1 or x 10 media; and by hydration of powdered media
(complete except for sodium bicarbonate). Most commercial powdered
media appear to be as satisfactory as laboratory-prepared media for the
growth of cells, and they cost about the same.
As a precaution against contamination and errors in media components, we routinely maintain cells on two independent media (A and B),
using sera, media, phosphate-buffer saline (PBS), and trypsin-EDTA,
from two different sources or lots.
Preparation of 20 Liters of MEM from Powdered Medium. Two packages of powdered media [for 10 liters each, Grand Island Biological Co.
(GIBCO) or KC Biologicals]are dissolved in 18 liters of water in a
stainless-steel pressure vessel (Millipore Corp., 5-gallon capacity).
Twenty milliliters of a stock x 1000 solution of antibiotics are added
(either penicillin and streptomycin, or x 1000 Gentamycin), followed by 44
g of NaHCOa for monolayer MEM or 40 g of NaHCOs for suspension
MEM. The pH is adjusted to 7.2 with 1 N N a O H or 1N HC1, made up to
20 liters with water, and sterilized with a pressure pump by filtration
through a 127-mm disc prefilter (Millipore AP2512750) placed on top of a
142-mm, 0.20-#m metricel filter (Gelman, from Fisher Scientific Co. ) or a
0.22-p.m GS filter (Minipore). Media are sterility-tested for at least a week
prior to use, using both thioglycolate broth (Difco) and blood agar plates
[tryptic soy agar containing 5% sheep blood (Gibco Diagnostics)], and
stored at 4. Before use, 5% or 10% sera (donor calf or donor horse
obtained from KC Biologicals, Flow Laboratories, Inc., or Gibco) are
added.
12H. Eagle, Science 130, 432 (1959).
428
[36]
Preparation of Minimal Essential Medium from Individual Components.

Assembling medium from individual components provides control of quality and the opportunity to alter the concentration of specific substances
for experiments, e.g., reduction in phosphate concentration in order to
label nucleic acids with zzPO4.
Minimum Essential Medium for Suspension Culture. The following materials are assembled and sterilized by filtration as described above: 15
liters of water, 2000 ml of 10 suspension salt A (the preparation of
concentrates is described below), 200 ml of 100 suspension salt B, 1000
ml of 20 MEM essential amino acids, 200 ml of 100 MEM nonessential
amino acids, 200 ml of x 100 MEM vitamins, 1380 ml of water containing
12.5 g of L-glutamine (Sigma Chemical CoO and 100 mg of phenol red
(sodium salt, Fisher Scientific Co.), and 20 ml of either 1000 penicillin
and streptomycin or x 1000 Gentamycin.
Minimal Essential Medium for Monolayer Culture. The procedure described above for suspension medium is used except that monolayer salt
A and B concentrates (see below) are substituted for monolayer salt
concentrates.
Balanced Salt Concentrates (stored at 4, nonsterile):
SUSPENSION SALTS. Salt A (10) is 68 g NaCI, 4 g KC1, 15 g
NaH2PO4" 1-120, 10 g dextrose, and 20 g NaHCOa (added slowly) per
liter. Salt B ( 100) is 20 g MgCI2 6 HzO per liter.
MONOLAYER SALTS. Salt A (10) is 68 g NaC1, 4 g KC1, 1.4 g
NaH2PD4 H20, 10 g dextrose, and 20 g N a H C O per liter. Salt B (x 100) is
20 g MgCI2 6 H20 and 25 g CaC12 2 H20 per liter.
Essential Amino Acids (20). The following amino acids (Sigma) are
dissolved in 4 liters of water and 80 ml of concentrated HCI: 25.3 g
L-arginine-HC1, 4.8 g L-cystine, 8.2 g L-histindine .HC1, 10.4 g
L-isoleucine, 10.4 g L-leucine, 14.6 g L-lysine, 3 . 0 g L-methionine, 6.4 g
L-phenylalanine, 9.6 g L-threonine, 2.0 g L-tryptophan, 7.2 g L-tyrosine,
and 9.2 g L-valine. Then, 5920 ml of water are added and the solution is
sterilized by pressure filtration and stored at 4 .
Nonessential Amino Acids (100). The following amino acids (Sigma)
are dissolved in 2 liters of water and stored frozen in 100-ml portions: 1.78
g L-alanine, 3.00 g L-asparagine, 2.66 g L-aspartic acid, 2.94 g t-glutamic
acid, 1.5 g glycine, 2.1 g L-serine, and 2.3 g L-proline.
Vitamins (100). The following vitamins (Sigma) are dissolved in 1500
ml of water: 24 mg niacinamide, 41 mg pyridoxal. HC1, 64 mg
thiamine HC1, 8 mg riboflavin, 48 mg D-pantothenic acid (hemi-calcium
salt), 280 mg choline chloride, and 108 mg myo-inositol. Then 5.4 ml of 0.1
N N a O H containing 5 mg d-biotin and 88 mg folic acid are added, and the
solution is adjusted to 2 liters. Aliquots (200 ml) are stored at - 2 0 .
[36]
HUMAN ADENOVIRUSES
429
Penicillin and Streptomycin (1000). Twenty million units of penicillin

G (Sigma) and 20 g of streptomycin sulfate (Sigma) are dissolved in 200 ml
of water. The solution is sterilized by suction filtration through a 47-mm,
0.22-/zm filter (Millipore) and stored at 4 .
Gentamycin (1000). A sterile solution containing 50 mg/ml of Gentamycin is purchased from the Schering Corporation and stored at 4 .
Phosphate-Buffered Saline (PBS). For washing cell sheets prior to
plaque assay, PBS containing magnesium and calcium is used; it is usually
supplemented with 0.1% bovine serum albumin (BSA) (Sigma, crystalline). To wash cell sheets before trypsinization and for the preparation of
trypsin-EDTA, PBS lacking magnesium and calcium is used. PBS is made
by dissolving with stifling 80 g NaC1, 2 g KC1, 1 g CaC12 2 H20, and 1 g
MgC12 6 H20 in 8 liters of water. Then, 11.5 g Na2HPO4 and 2 g KH2PO4
are dissolved in 2 liters of water and added to give 10 liters of PBS. The
solution is sterilized by pressure filtration and stored at 4 in 100-ml portions. PBS lacking magnesium and calcium is prepared as described
above except that CaC12 2 HzO and MgC12 6 1-120 are deleted. BSA is
prepared as a 5% stock solution, sterilized by suction filtration, and stored
at 4 . To prepare PBS-0.1% BSA, 2 ml of stock BSA are added to 100 ml
of PBS.
Trypsin-EDTA. Four grams of trypsin (Difco Laboratories, 1 : 250), 2 g
of disodium EDTA, 4 g of dextrose, 20 mg of phenol red (sodium salt,
Fisher), and 4 ml of 1000 stock solution of penicillin and streptomycin
are dissolved in PBS without magnesium and calcium and adjusted to 4
liters. The solution is filtered and stored at 4.
Tris-Saline-Glycerol. This is used to maintain the infectivity of purified
Ad stocks at - 3 5 . A 5 stock of Tris-saline is 40 g of NaC1, 1.9 g of KC1,
0.5 g of Na,zHPO4, 5 g of dextrose, and 150 ml of 1 M Tris HC1 buffer (pH
8.1 at 25), made up to 1 liter, and sterilized by filtration. Tris-saline-30%
glycerol is made by adding 100 ml of 5 Tris-saline to 150 ml of sterile
glycerol (autoclaved at 120 for 15 min) and adjusting the volume to 500 ml
with sterile water.
Agar Overlay Medium for Plaque Assay. The following sterile solutions
are combined to prepare 100 ml of x2 overlay medium: 18.0 ml of 10
Earle's balanced salt solution, 2.0 ml of 100 MEM essential amino acids,
2.0 ml of 100 MEM vitamins, 2.0 ml of I00 L-glutamine, 0.7 ml of 2.1%
L-arginine, 0.4 ml of 1.0N NaOH, 12 ml of horse serum, 12 ml of chicken
serum (KC Biological), ( recently, we have used 12 ml of fetal calf serum
plus 12 ml of water instead of horse and chicken serum ) 2.0 ml of antibiotic mixture (see below), and 43 ml of sterile water. The solution is stored
at - 2 0 . (1) Earle's salt solution (10). Solution A is 136 g of NaC1, 8 g of
KC1, 2.88 g of KH2PO4, and 2 g of MgSO4 in 800 ml of water. Solution B is
430
[36]
4.0 g of CaCI~ 2 H~O in 200 ml of water. Solution C is 40 g of dextrose

and 100 ml of 0.2% phenol red in 200 ml of water. A, B, and C are
autoclaved separately and cooled. A and B are mixed, and then C is
added. The solution is adjusted to 2 liters with sterile water and stored at
4 . (2) Glutamine (200 mM, xlO0). L-glutamine, 29.2 g, is dissolved in 1
liter of water, sterilized by filtration, and 100 ml amounts stored at - 2 0 .
(3) Phenol red (0.2%). Phenol red (2.0 g) is dissolved in 50 ml of 95%
ethanol and 950 ml of water and stored at 4. (4)Arginine. L-arginine
(Sigma), 2.1 g, is dissolved in 100 ml of water, sterilized by autoclaving,
and stored at 4. (5) Antibiotic mixture. During plaque assay, petri dish
cultures are incubated at 37 for up to 20 days in a humidified COs incubator. Fungal and bacterial contamination are often a problem, and
therefore an antibiotic mixture containing neomycin and Fungizone in
addition to penicillin and streptomycin is used. Six million units of penicillin, 2 g of streptomycin, 0.1 g of amphotericin B (Fungizone, Squibb
Pharmaceuticals), and 1 g of neomycin sulfate (Upjohn) are dissolved
aseptically in 200 ml of sterile water and stored in 10-ml amounts at - 2 0 .
Growth of Monolayer and Suspension Cultures. Cells are incubated at
36--37. Monolayers are cultured in 32-ounce Saniglass prescription bottles or plastic cell culture flasks with growth surface areas of 25-500 cm 2
(Nunc, from Vanguard International; Falcon Plastics, from Fisher Scientific; or Coming Glassworks, from Fisher Scientific). Cells are grown in
suspension in screw-cap spinner flasks, ranging from 50 ml to 8 liters, with
adjustable hanging magnetic bars (Bellco Glass, Inc.). The suspension is
stirred with a magnetic stirrer. Cells are subcultured and media changed
using aseptic techniques in an isolation cubical or, preferably, in a
laminar-flow hood. Glassware is cleaned with nontoxic detergents designed for cell culture, and sterilized by autoclaving, or by heating in an
oven at 180 for 4-8 hr.
A KB cell clonal line, grown in our laboratory for the past 19 years, is
readily converted from monolayer to suspension by scraping monolayers
with a rubber policeman and culturing in suspension. KB cells grow in
suspension indefinitely and readily form monolayers. Suspension cultures
are propagated in MEM supplemented with 5% horse serum for the routine growth of Ads. Cells are counted each day with a hemocytometer
(American Optical Corp., Bright-Line) under a microscope (x 100 magnification); the number of cells deposited on 10 large squares (five on each
side of the hemocytometer) x 1000 is the cell density (cells/ml). The cell
density is adjusted to 2 x 10s cells/ml by the addition of fresh medium
each day. The cell doubling time ranges from 18-24 hr.
Monolayers maintained in Eagle's MEM with 10% calf serum are used
mainly to seed petri dishes for plaque assay. Cultures are observed every
other day with an inverted microscope for cell culture (x40 to x l00
magnification). Confluent monolayers are subcultured about twice a week
[36]
HUMAN ADENOVIRUSES
431
as follows. After removal of the medium by aspiration or pouring, the cells

are scraped into 8 ml of fresh medium with a rubber policeman, dispersed
gently with a 10-ml pipette (fitted with Propipet), and seeded into four or
five new flasks (about 2 x l0 s cells per 32-ounce flask) containing 40 ml of
medium.
Large-Scale Growth and Purification of Adenoviruses
Infection of KB Cell Suspension Cultures. 13-15 Exponentially growing

KB cells (3-6 105 cells/ml, 2-20 liters total) are collected by centrifugation at 180 g for 15 min at room temperature in the International PR-J
centrifuge (No. 276 rotor) in sterile I-liter polypropylene screw-capped
bottles (International No. 2939). Cells are suspended in 1/20th volume of
MEM (no serum), infected by the addition of 20--100 plaque-forming units
(PFU) per cell of a sterile Ad stock, and stirred for 1 hr at 37. The
suspension is diluted with warm MEM (Joklik-modified) containing 5%
horse serum to give 3-4 105 cells/ml. The infected cells are incubated
with stirring at 37 until 30-40 hr after infection (48 hr for Adl2 strain
Huie), chilled in ice, and harvested by centrifugation at 230 g for 20 min at
4 . The cell pellet (over 90% of virus remain cell-associated) is resuspended in 10 ml of 10 mM Tris HC1, pH 8.1, per 3 liters of infected cells,
and frozen until purified (usually within 3 days after harvesting).
Purification of Adenoviruses. 13.14 Infected cell-pellet suspensions are
thawed briefly in a 37 water bath. All subsequent operations are at 0-4 .
Cell pellets are suspended in an additional 40 ml of 10 mM Tris HCI, pH
8.1, and the cells disrupted with a Raytheon DF-101 sonic oscillator at full
power for 5 min. The suspension is homogenized with 40 ml of trichlorotrifluoroethane (Taylor Chemical) in a 200-ml stainless-steel cup of a Sorvail Omnimixer for 1 min at about 10,000 rpm, and then centrifuged in a
250-ml polycarbonate screw-capped bottle (Nalgene No. 3123), or in a
disposable sterile 250-ml screw-capped centrifuge tube (Coming No.
25350), using the PR-J centrifuge (No. 259 rotor) at 1000 g for 10 min. The
upper aqueous layer is removed with a pipette, and the lower layer is
homogenized with 20 ml of 10 mM Tris HC1, pH 8.1. The aqueous layers
are combined and carefully layered over 10 ml of CsCI at 1.43 g/cm 3 (e.g.,
Schwarz-Mann, biological grade, 43 g of CsC1 in 60 ml of 10 mM
Tris HC1, pH 8.1) in a cellulose nitrate tube (1 3.5 inch). After centrifugation in a SW-27 rotor for 1 hr at 20,000 rpm, the supernatant liquid
above and most of the CsCI solution below the opalescent virus band are
removed. The virus band (3-5 ml) is adjusted to 1.34 g/cm 3 with powdered
CsCI, placed in a 0.5 2.5 inch (6.5 ml) cellulose nitrate tube (which fits
13M. Green and M. Pina, Virology 20, 199 (1963).
14M. Green and M. Pina, Proc. Natl. Acad. Sci. U.S.A. 51, 1251 (1964).
15M. Pina and M. Green, Virology 38, 573 (1969).
432
[36]
the Ti 50 rotor with an adaptor), filled with CsC1 solution (density -- 1.34
g/cm a, 56 g of CsC1 in 116 ml of 10 mM Tris HCI, pH 8.1), and centrifuged for 16-24 hr at 30,000 rpm. The visible band in the middle of the
gradient, at a buoyant density of 1.34 g/cm a, is collected from the side by
tube puncture, adjusted to 6.5 ml by the addition of CsC1 solution at 1.34
g/cm a, and recentrifuged as described above. The twice-banded virus is
collected and stored at 4 .
Preparation of PurifiedAdenovirus Stocks and PlaqueAssay of Infectivity.

It is essential for biochemical studies that the amount of infecting virtis
be known, because all cells must be infected, and because the course of
the viral infection and the quantity of viral macromolecules synthesized
varies somewhat with the multiplicity of infection. Virus is purified as
described above, with the use of aseptic procedures, for the preparation
of Ad stock viruses for infection. All materials are either autoclaved or
soaked for at least 30 min in 70% ethanol (e.g., cellulose nitrate tubes).
Virus, banded once, is diluted in 10 ml of sterile Tris-saline-30% glycerol
per liter of initial infected cells, and stored in 5-ml aliquots in plastic
snap-cap tubes at - 3 5 . The infectivity of Ad stocks remains constant for
years. Stocks range in titer (assayed on KB cells) from 2-5 x 1011PFU/ml
for Ad2, 6 x 101 to 2 1011 PFU/ml for Ad7, and 2 l0 s to 1 109
PFU/ml for Adl2 (Huie).
A plaque assay TM developed for Ad2 has been used for all human
Ads. 17 A plaque is a cluster of infected, nonstained, nonviable cells,
produced by infection of a single cell by one virus particle with secondary
infection of contiguous cells by progeny virus, and is surrounded by uninfected, viable cells stained by neutral red. The plaquing efficiency of
purified virus ranged from virus particle:PFU ratios of 11 : 1 for Ad3 to
2300 : 1 for Ad25; the size and time of appearance of plaques varied with
the Ad serotype. 17 Assays are performed in 60-ram plastic petri dishes
seeded usually the day before with 1-1.5 106 KB cells. Cells are collected from suspension cultures by centrifugation at 180 g for 7 min, and
resuspended in MEM with 10% calf serum to give a density of 2-3 x 105
cells/ml; 5-ml amounts are seeded in petri dishes. Alternatively, KB cell
monolayers are harvested with trypsin-EDTA, resuspended in MEM with
10% calf serum, and petri dishes seeded as described above. Dishes are
incubated in a COz incubator (flushed with 5% CO2 and maintained at
90-100% humidity, e.g., Forma Hydro-Jac, Forma Scientific) until cell
sheets are 70-80% confluent. The medium is removed by aspiration with a
Pasteur pipette, and cells are washed by gentle addition and aspiration of
5 ml of PBS-0.1% BSA at 37. Serial 10-fold dilutions of virus samples are
16H. C. Rouse, V. H. Bonifas, and R. W. Schlesinger, Virology 20, 357 (1963).
~7M. Green, M. Pina, and R. C. Kimes, Virology 31, 562 (1967).
[36]
HUMAN ADENOVIRUSES
433
prepared in cold PBS-0.1% BSA (or alternatively in MEM with 2% calf

serum), and 0.5-ml amounts (incubated briefly in a 37 water bath) are
added to triplicate dishes. Dilutions are chosen to contain about 20-200
PFU/0.5 ml. Ad2 plaques are about 5 mm in diameter, and therefore over
50-100 plaques per plate will produce confluent, noncountable areas,
whereas as many as 500 Adl2 plaques (0.5 ram) per plate can be scored. 17
Dishes are rocked gently and placed in a CO2 incubator for 90 min to
permit virus adsorption. Five milliliters of agar overlay medium are
slowly added at the edge of each dish (to avoid dislodging cells) with a
10-ml pipette. Overlay medium is prepared at 46 just before use by mixing 94 ml of 2 overlay medium (containing 6% horse serum and 6%
chicken serum or 6% fetal calf serum) with 100 ml of melted 1.8% Bactoagar and 6 ml of sterile 7.5% sodium bicarbonate. After the agar overlay
has hardened at room temperature (15-30 min), dishes are incubated in a
humidified CO2 incubator. Five milliliters of agar overlay medium are
added again after 5 days. An additional 5 ml of agar overlay medium,
containing 0.001 or 0.002% neutral red (Fisher), are added after 10 days.
Plaques are visible after 12-14 days with most Ads (13-15 days with Adl2)
and are counted macroscopically against a lighted background. Small
plaques of Ad2, Adl, and Ad5 can be seen as early as 8 days after infection if the neutral red overlay is added on the 7th day. ~r
Isolation and Purification of Adenovirus
D N A 14'18
DNA is isolated from virus by digestion with a protease, treatment

with SDS, and phenol extraction. Protease treatment is necessary to remove a tightly bound protein is that is covalently linked to the termini of
Ad DNA molecules. 19 Twice-banded virus (in CsC1 solution) is dialyzed
against two changes of 200 volumes of 10 mM Tris HC1, pH 8.1, 1 mM
EDTA at 4 for 2 hr or longer. To each milliliter of virus are added 150 tzl
of 50 m M EDTA (pH 6.9), 150 ~1 of 1 M sodium phosphate (pH 6.0), 15 #d
of 1 M cysteine-HC1, and 30/zl of papain (Sigma, twice crystallized, 25
mg/ml in 50 mM sodium acetate, pH 4.5). After 1 hr at 37, 150 fd of 5%
SDS are added per milliliter of virus, the solution is incubated at room
temperature for 30 rain, and it is extracted with an equal volume of phenol
(saturated with 10 mM Tris HC1, pH 8.1, 1 mM EDTA) by gentle rotation (to avoid shear degradation of Ad DNA molecules) in a multipurpose
rotator (Scientific Industries, Model 150 V) for 15 rain at 4. After centrifugation at 7800 g at 4 for 10 rain, the upper aqueous phase is removed
and extracted twice more with phenol. The aqueous phase is extracted 3
18M. Green and M. Pina, Proc. Natl. Acad. Sci. U.S.A. 50, 44 (1963).
I~D. M. K. Rekosh, W. C. Russell, A. J. D. Bellet, and A. J. Robinson, Cell 11, 283 (1977).
434
[36]
times with water-saturated ether, and then dialyzed against 200 volumes
of 10 mM Tris. HC1, pH 8.1, containing 1 mM EDTA for 24 hr. The
absorbance at 250, 260, and 280 nm is determined. The amount of DNA is
calculated based on an A2~0 unit equivalent to 50/~g DNA/ml. The 260
nm/280 nm and 260 nm/250 nm ratios are about 1.9 and 1.1, respectively.
Yields of 60-80% homogeneous, intact duplex DNA molecules, sedimenting at 30-31 S, are obtained. 4 Approximately 0.5 mg of Ad2 DNA is
isolated from virus produced in 1 liter of cells.
Transfection-InfectivityAssay for Viral D N A and D N A Restriction
Fragments
Transfcction assays with purified viral D N A (rather than virions) arc
useful for several kinds of studies. For example, recent recombinant D N A
procedures, and related procedures, arc beginning to bc exploited for the
construction of A d mutants in vitro.20In addition, transfectionprocedures
using A d D N A and D N A restrictionendonucleasc fragments have been
used to map the Ad-transforrning genes. The infectivityof Ad D N A and
the transforming activity of Ad D N A and restrictionD N A fragments are
assayed with the calcium phosphate procedure, 11"21modified to include
treatment of ceils with dimethylsulfoxide (DMSO) 2~2 to enhance the
transfection activity. Ceils used for transfection-infectivity assay arc
permissive human K B cells in petri dishes as described in preceding sections. Cells (nonpermissive) used for transfection-transformation assay
are prepared from 16-day rat (Wistar) whole embryos. Cells prepared
from baby rat (6- to 7-day-old) kidneys are used by Graham et al. i~ A
permanent rat cell line for transfection-transformation experiments has
also been described. 2a
Cell procedures are performed under aseptic conditions. Several embryos are minced with scissors, washed 2 or 3 times with PBS lacking
calcium and magnesium, and digested with warmed (37 ) trypsin-EDTA
for 10-20 min at room temperature. Cells are centrifuged for 10 min at
230 g, chilled, suspended in MEM with 10% calf serum, and filtered
through two layers of gauze. Cells are collected from the filtrate by centrifugation, a 0.5% suspension is prepared in MEM containing 10% calf
serum, and 60-ml amounts are seeded in 32-ounce culture flasks; usually
one embryo provides sufficient cells for one flask. When monolayers are
nearly confluent (3-4 days), cells are subcultured by trypsin-EDTA treat20 G. Chinnadurai, S. Chinnadurai, and M. Green, J. Virol. 26, 195 (1978).
2~F. L. Graham, J. Smiley, W. C. Russell, and R. Nairn, J. Gen. Virol. 36, 59 (1977).
22N. D. Stow and N. M. Wilkie, J. Gen. Virol, 33, 447 (1976).
2a G. Kimura, A. Itagaki, and J. Summers, Int. J. Cancer 15, 694 (1975).
[36]
HUMAN ADENOVIRUSES
435
ment into 60-mm dishes in MEM wlth 10% fetal calf serum for transfection (5 105 cells/dish). When the cell sheets are 80-90% confluent, 0.5 ml
of solution, containing either Ad DNA or DNA restriction fragments, is
added to each dish without removing the medium. Ad DNA is prepared as
described above. Preparative isolation of Ad DNA restriction endonuclease fragments is described in detail in Wold et al. 24 The viral DNA
solution is prepared as follows.
(1) Ad DNA (0.1/xg/ml to 0.5/zg/ml for DNA fragments smaller than
2-3 x 106 daltons, or 1.0/zg/ml to 10/xg/ml for larger fragments) is diluted
into HEPES-buffered saline (a 2 solution is prepared from 8 g of NaC1,
370 mg of KC1, 125 mg of Na2HPO4 2 I-I20, 1 g of glucose, and 5 g of
N-2-hydroxyethyl-piperazine-N'-2-ethanesulfonic acid, made up to 500
ml and pH 7.05, and sterilized by filtration) containing 10/~g/ml of salmon
sperm DNA.
(2) CaC12 is added (from a 2.5 M stock solution sterilized by autoclaving) to a final concentration of 125 mM. After 15-20 min at room temperature, the mixture becomes slightly turbid and is added to petri dishes
containing rat embryo cells. After 4 hr of incubation at 37 in a CO2
incubator, the medium is removed, the cultures washed once with medium, and 1 ml of 25% DMSO in HEPES-butfered saline is added to each
plate. After 2-4 min, the DMSO is removed, the cultures washed with
medium, and 5 ml of fresh MEM with 10% fetal calf serum are added.
Three or four days later the medium is replaced with calcium-free MEM
with 5% fetal calf serum. Incubation is continued (the medium is changed
twice weekly) until transformed cell foci appear (positive controls are
always included in assays). Generally, about 2--4 weeks are required for
foci to develop. Foci appear macroscopically as raised colonies, and microscopically as multiple layers of densely packed, criss-crossed cells.
Transformed colonies can be quantitated under a dissecting microscope at
a magnification of x30 to 40. Alternatively foci can be counted after the
cultures have been fixed by a treatment with absolute methanol for 5 min,
air dried, and stained with Giemsa. The efficiency of transfection of an
Ad2 DNA HindlII digest is approximately 10 foci per microgram of DNA.
As final proof that foci represent cells transformed by viral DNA, clonal
lines must be isolated and shown to contain integrated viral DNA and to
synthesize viral RNA. For this purpose, individual foci are taken up into a
Pasteur pipette by gentle aspiration with a rubber bulb and transferred
into a new petri dish containing calcium-free MEM with 10% fetal calf
serum. Clonal lines are then developed by selection and subculture and
used for subsequent biochemical analyses.
24 W. S. M. Wold, M. Green, and J. K. Mackey, Methods CancerRes. 15, 69--161"(1978).
[37]
SOURCES
OF
STABLE
CELL
LINES
439
[37] S o u r c e s o f S t a b l e C e l l L i n e s
B y GERARD J. MCGARRITY
I. Introduction
T. C. Hsu summarized the advantages of cell culture systems: 1
1. They provide a continuous supply of homogenous cellular material
for biochemical experiment~ as well as for practical use in medical and
public health work.
2. The cells in vitro can be manipulated advantageously in many ways.
This cannot be done with cells in vivo.
3. They can be stored in a deep-frozen state without changing their
growth rate and genetic composition, and they c a n b e revived at will.
4. Using cell cultures is more economical than rearing animals and
performing experiments with intact animals.
5. They save lives of animals.
To obtain full benefit of these advantages, cell cultures used in research and diagnosis should be standardized as fully as possible and subjected to extensive quality-control procedures. (For detailed methods on
this topic, see this volume [2] and [13].) Quality control should include
selection of the cell culture to be used and the source of the culture. Many
cell cultures contain undetected adventitious agents such as bacteria,
yeast, viruses, and mycoplasmas. In this laboratory the annual mean
infection rate of cell cultures with mycoplasmas has ranged from 3.414.9%. This is based on testing of more than 6000 cell cultures.
Mycoplasma-infected cell cultures are useless in controlled standardized
procedures. Introduction of a mycoplasma-infected culture into a laboratory generally results in spread of the infection to all cultures in the
laboratory. 2
A significant number of cell cultures are contaminated with other cells,
especially HeLa. Gartler concluded that many of the permanent human
cell lines may in fact be HeLa, based on electrophoretic variant forms of
isoenzymes as genetic markers; a this observation has been reinforced. 4,5
More recently, Nelson-Rees and Flandermeyer have reported that 41 of
1T. C. Hsu, in "The Future of Animals,Cells, Models,and Systemsin Research Development, Educationand Testing," p. 180. Natl. Acad. Sci., Washington,D.C., 1977.
2 G. J. McGarrity,In Vitro 12, 643 (1976).
a S. M. Gartler, Natl. Cancer Inst., Monogr. 26, 167 (1967).
40. J. Miller,D. A. Miller,P. W. Allderdice,V. G. Dev ,and M.S. Grewal,Cytogenetics 10,
338 (1971).
5 W . Nelson-Rees, R. R. Flandermeyer,and P. K. Hawthorne,Science 184, 1093 (1974).
ME~FHODS IN ENZYMOLOGY, VOL. LVIII

All rights of reprodt~ction in any form reserved.
ISBN 0-12-181958-2
440
SPECIFIC CELL LINES
[37]
253 cell cultures examined (16%) were not as purported; HeLa cell contamination and wrong species accounted for 36 of these 41 errors. 6
To protect against mycoplasma infection, interspecies and intraspecies
contamination, and other problems, cell cultures should be acquired only
from sources that have performed reliable quality-control checks. The
common practice of acquiring cultures from colleagues is risky and is
discouraged; several episodes of mycoplasma infection have been traced
to this practice. Even reassurances of negative mycoplasma assays made
months, even weeks, before is no guarantee of a culture's present status.
The purpose of this article is to list the major sources of cell cultures
and information pertinent to each source.
II. Cell Repositorirs
American Type Culture Collection (ATCC)

The ATCC was founded in 1925 for the collection, preservation, and
distribution of authentic cultures of living microorganisms and animal
cells. Its 1975 catalogue lists 153 characterized reference animal cell lines
derived from 40 different species. An additional 170 human skin fibroblast
lines are derived from apparently normal individuals and from patients
with various diseases, including genetic disorders. Animal species represented include: amphibian, aves, bat, buffalo, cattle, dog, dolphin, gerbil, goat, hamster, horse, human, insect, minipig, mink, monkey, mouse,
muntjac, pig, pisces, potoroo, rabbit, raccoon, rat, and reptile. Widely
used human cultures include Hela and several HeLa derivatives; EB-3
and RAJI from Burkitt's lymphoma; and WI-38 and MRC-5, both human
diploid lung. IMR-90, a human diploid lung, although not listed in the 1975
catalogue, is available from ATCC.
A copy of the latest catalogue can be purchased (approx. cost: $10,
includes postage). The catalogue lists a history of each cell line and a
description of the repository reference seed stock that includes freeze
medium, viability, growth medium, growth characteristics, plating efficiency, age of culture since origin, morphology, karyology, sterility tests,
species confirmation, virus susceptibility, and isoenzyme patterns, if
available. All ATCC stocks of questionable human origin (possible HeLa
contaminants) are indicated in the catalogue.
Frozen ampules of cell cultures can be purchased from ATCC. Current
prices are $28.00 per ampule to nonprofit institutions and $43.00 per ampule to commercial firms. Information regarding purchase of cultures or
6 W. Nelson-Rees and R. R. Flandermeyer, Science 195, 1343 (1977).
[37]
SOURCES OF STABLE CELL LINES
441
catalogues can be obtained directly from the ATCC, 12301 Parklawn

Drive, Rockville, Maryland 20852.
Additional information regarding handling of frozen ampules and
guidelines for selection of cells and deposition of new cell lines are also
listed in the catalogue.
Special arrangements can be made with ATCC for safe-deposit of
individual cell cultures. These cultures are not accessioned into the general stock. The depositor retains all proprietary rights. Other special services are also available on a for-fee basis.
Human Genetic Mutant Cell Repository
This repository was established at the Institute for Medical Research,

Camden, New Jersey in 1972 by the National Institute of General Medical
Sciences. The objective of this repository is to store cultures in low passage with single and multiple gene defects, both defined and undefined at
the molecular level; chromosome abnormalities, including translocations,
inversions, and deletions; polymorphisms (antigens, isoenzymes) as well
as carder, sibling, and control cultures. 7 Fibroblast, lymphocyte, and
amnion cultures are included. Most are human cells.
The purposes of this repository are:
1. The study of cells from a number of persons with the same inherited
disorder which may reveal subtle variations in the underlying defect that
would be important to its diagnosis and treatment.
2. Use of identical cell lines for study in different laboratories should
facilitate the comparison of data and interpretation of research results.
3. Cell lines from some rare diseases which otherwise would not be
readily available to all interested investigators.
4. Investigators in different laboratories can develop and maintain
numerous cell lines.
5. The availability of the skills for characterization and identification of
genetic mutant cells will assure uniformity and reliability of cells from the
repository.
Cell cultures in the repository are tested for sterility, including mycoplasma, species of origin, karyotype, viability, and expression of the biochemical or chromosomal defect.
The Fourth Edition of the Catalogue of the Human Genetic Mutant
Cell Repository was published in October 1978 and is available free of
charge. The catalogue lists over 1700 cultures, including approximately
200 biochemical mutants and 250 chromosome aberrations. Where
7 L. L. Coriell, Science 180, 427 (1973).
442
SPECIFIC CELL LINES
[37]
known, McKusick numbers and HL-A antigens are listed after appropriate cultures.
Starter cultures are supplied. These are obtained by inoculating a
frozen ampule into a T25 flask and growing for several days prior to
shipping. The cost per T25 flask is $20 plus shipping. The catalogue can be
obtained by writing to the Human Mutant Cell Repository, Institute for
Medical Research, Copewood Street, Camden, New Jersey 08103.
Aging Cultured Cell Repository

The repository was established at the Institue for Medical Research by
the National Institute on Aging. The purpose of this repository is to collect, store, and distribute cultures of interest to aging research, a The
repository contains fibroblast cultures from apparently normal individuals
including large quantities of well-characterized early passage cultures
from fetal lung; cultures from clinical conditions associated with cellular
aging and growth disorders; cultures from tumor patients; and transformed cell lines. The human diploid fibroblast cell line IMR-90 was established from lung tissue of a female fetus by this repository and is
available) A limited supply of early passage IMR-90 is also available to
qualified investigators by special action of the repository's advisory
committee.
Another human diploid fibroblast, IMR-91, has been established from
lung tissue of a male fetus. IMR-91 will be available from this repository
when characterization studies have been completed. Two human diploid
fibroblast lines will be established from adult tissue and will be available
when characterization studies have been completed.
The list of available cultures is contained in the Catalogue of the
Human Genetic Mutant Cell Repository described above. Currently, 157
cell lines are stored in the Aging Cultured Cell Repository. Cultures are
tested for sterility, including mycoplasma, species of origin, karyotype,
viability, and, where applicable, expression of the specific characteristic
of interest.
Shipping and prices are the same as listed for the Human Genetic
Mutant Cell Repository. Cultures are free to laboratories engaged in aging
research. Further information can be obtained from the Aging Cultured
Cell Repository, Institute for Medical Research, Copewood Street, Camden, New Jersey 08103.
a D. G. Murphy and W. W. Nichols, Cytogenet. Cell Genet. 15, 30 (1975).
W. W. Nichols, D. G. Murphy, V. J. Cristofalo, L. H. Toji, A. E. Greene, and S. A.
Dwight, Science 196, 60 (1977).
[37]
SOURCES OF STABLE CELL LINES
443
III. National Science Foundation Cell Culture Centers

The NSF has established Cell Culture Centers a't Massachusetts Institute
of Technology and the University of Alabama at Birmingham. These are
intended to serve as a facility and research resource for scientists who
wish to obtain large-scale cell and virus production for highly meritorious
research projects.
The MIT Center is designed for large-scale monolayer and suspension
cell cultures and viral production. The University of Alabama, Birmingham, has been participating in large-scale production of lymphocyte cell
cultures. The objective of both Centers is to produce cells and viruses on
a scale sufficient to permit investigators to perform experiments that could
not otherwise be performed.
Proposals to use these facilities should be made to the individual Centers. Each proposal is evaluated by a review committee. Proposals should
list the amount of culture required and a description of the objectives and
significance of the research and supporting materials. Further information
is available from:
N S F Cell Culture Center
E17-21
Massachusetts Institute of
Technology
77 MassachuSetts Avenue
Cambridge, MA 02139
N S F Cell Culture Center

University of Alabama Hospitals
& Clinics
1808 7th Avenue, South
Birmingham, AL 35294
IV. Other Sources

Certain commercial firms sell various types of cell cultures. Some of
these cultures are established by the firms and others are obtained from
other sources, including the American Type Culture Collection. Further
information can be obtained by consulting the catalogues of suppliers of
cell culture reagents for types of cultures available and types of qualitycontrol tests performed on these cultures.
V. Comments
Selection of the proper cell culture system is an integral component of

any in vitro experimental procedure. The source of the cell culture can be
just as critical. Where possible, cultures should be obtained from sources
that have performed adequate quality-control tests. If such tests have not
444
SPECIFIC CELL LINES
[38]
been performed, recipient laboratories should quarantine the culture until

tests (especially for mycoplasma) are complete. Procedures for sterility
testing of cell cultures and media are described elsewhere in this text.
Quality control should be performed periodically to detect any significant alteration in the cell culture that could influence results of experimental procedures.
[38] I s o l a t i o n o f F i b r o b l a s t s f r o m P a t i e n t s
By WILLIAM S. SLY and JEFFREY GRUBB

Over the past 10 years, fibroblasts cultured from biopsies of human
patients have provided unique opportunities for exciting biochemical
studies. Studies of Lesch-Nyhan syndrome fibroblasts made important
contributions to knowledge of purine metabolism and gout, to X-chromosome inactivation, and to somatic cell hybridization. 1 Studies of
fibroblasts from patients with mucopolysaccharide storage diseases defined the enzymes involved in mucopolysaccharide degradation and led to
discovery of secretion and receptor-mediated uptake of lysosomal enzymes by fibroblasts. 2 Studies of fibroblasts from patients with familial
hypercholesterolemia have revolutionized our thinking about the uptake
and metabolism of cholesterol and of diseases involving defective cellsurface receptors. 3 These three examples provide dramatic evidence that
fibroblasts, despite their reputation as cells in which many differentiated
cell functions are not expressed, can contribute enormously to solution of
certain biochemical problems.
In this article, we will describe how one obtains a skin biopsy and
propagates cultured fibroblasts for biochemical studies from this source.
Before doing so, it is worth emphasizing that cell lines from patients with
many human genetic disorders are available on request from the Human
Genetic Mutant Cell Repository, Institute for Medical Research, Camden,
New Jersey. This is a federally financed collection of human mutant cell
lines developed under contract for the National Institute of General Medical Sciences. Its purpose is to provide material from patients with human
genetic diseases to investigators who wish to study these diseases, but
who may have difficulty gaining access to cells from affected patients.
Mutant strains are established from skin biopsies solicited from around
the world, grown up, tested for contamination, and stored frozen in liquid
1 j . E. Seegmiller, A d v . H u m . Genet. 6, 75 (1976).
2 E. F. Neufeld, T. W. L i m , a n d L. J. Shapiro, A n n u . Rev. Biochem. 44, 357 (1975).
3 M. S. B r o w n and J. L. Goldstein, Science, 191, 150 (1976).

All fights of reproductionin any form reserved.
ISBN 0-12-181958-2
[38]
ISOLATION OF FIBROBLASTS
445
nitrogen. For a small charge, samples are thawed, started in culture, and
shipped to investigators on request. Investigators contemplating initiating
a cell culture laboratory for studies on normal or mutant human fibroblasts would be well advised to obtain the catalogue of existing strains.
1. Obtaining a Skin Biopsy
Materials
Biopsy tray with 2 4-mm punch biopsy instruments, 2 curved scissors, and 2 forceps. A second set of instruments is provided in
case one gets contaminated accidentally while attempting the
biopsy.
Sterile gauze
Isopropyl alcohol, 70%, 30 ml
Lidocaine without epinephrine, 1%
1 tuberculin syringe
1 screw-capped tube containing 10 ml of medium
Note: All glassware items--dishes, tubes, coverslips--are acid
washed with concentrated HNO3 and rinsed extensively with
deionized water before sterilizing by autoclave.
The site of removal of tissue depends on the purpose. Studies of testosterone binding and/or metabolism should be done on fibroblasts from
genital skin. 4 For nearly all other general purposes, we take a biopsy from
the subdeltoid area of the upper arm (the lateral aspect of the arm ~ the
distance from the shoulder to the elbow). Others use the anterior aspect of
the forearm, 2 inches below the elbow crease. 5 The area is first scrubbed
twice thoroughly with a sterile gauze fiat moistened with 70% isopropyl
alcohol. Then an area about 1-1.5 cm 2 is anesthetized by injection of 0.2
ml of 1% Lidocaine without epinephrine intradermally with a tuberculin
syringe. After 2-3 min, a 4-mm sterile punch biopsy instrument is rotated
with pressure to cut a circular flap through the epidermis from the center
of the anesthetized area. The edge of the flap is raised with a sterile
forceps, and the flap is removed by cutting under it with a small curved
scissors. The removed flap is dropped into a sterile 125-mm screw-capped
tube containing 10 ml of tissue culture medium (MEM-Earle's containing
15% heat-inactivated fetal calf serum, penicillin, and streptomycin, see
below).
Biopsies are usually "planted" immediately, but they may be planted
up to 4 days after the biopsy, if stored at room temperature in cell culture
4 j. E. Griffin, K. Punyashthiti, and J. D. Wilson, J. Clin. Invest. 57, 1342 (1976).
5 j. T. Cooper and S. Goldstein, Lancet 2, 673 (1973).
446
SPECIFIC CELL LINES
[38]
medium. If the biopsy is to be mailed, it may be transferred to a sterile

5-ml polypropylene cryotube (Vangard International), which should be
filled to the top, sealed, and mailed in a well-insulated shipping carton.
2. "Planting" the Skin Biopsy and Initial Propagation of Fibroblasts
Materials
1 100-mm sterile glass petri plate
2 Bard-Parker disposable scalpels with No. 15 blades (may be autoclaved and reused)
2 60 x 15 cm tissue culture dishes
4 25-mm round coverslips, Corning 2915 "circles," thickness No. 2;
these have been acid washed, and distributed between layers of
filter paper in 100-mm glass petri dishes and autoclaved
Dow Coming silicone stopcock grease, placed in 13 x 75 screw-cap
vial and autoclaved
Flat-ended forceps for handling coverslips (sterilized by immersion of
operational end in 95% ethanol, and flame-dried prior to use)
Growth medium: Eagles Modified Minimal Essential Medium containing penicillin, 100 U/ml; streptomycin, 100/z g/ml; and sodium
pyruvate 0.10 g/liter (all components are commercially available in
solution or may be made up from commercially available powdered
medium)
Heat-inactivated fetal calf serum (75 ml) is added to each 500-mt
bottle of medium (fetal calf serum is heat-inactivated prior to use
by placing in a 60 water bath for 30 min)
Tissue culture trypsin 0.25% (obtained commercially as 2.5% trypsin
and diluted to 0.25% in 0.01 M, Tris- HC1, pH 7.0, and 0.015M
NaCI)
This operation is carried out in a laminar-flow hood to avoid contamination of the sample. The aim is to cut the sample into small pieces,
distribute it into two dishes that will be fed from separate media sources
to reduce chances of loss by contamination, and to hold the tiny pieces in
place by placing them under a coverslip which in turn is held in place by
silicone grease (see Fig. 1). 5,6
Pour the biopsy and medium from the tube to a sterile glass petri dish.
Use two round-bladed scalpels held opposing each other like scissors
blades. Bisect the skin biopsy by pressing the two blades down against the
glass and rolling the edges as if closing them in opposite directions (as a
6 R. J. Warren and C. De La Cruz, Exp. Cell Res. 71, 238 (1972).
[38]
.SK,
447
REASE
60 mm DISH
FIG. l. View of tissue culture dish in which explants (fragments from a skin biopsy) have
been "planted" beneath 25-mm coverslips that are held in place with silicone grease.
closing scissors). Then each piece is further divided in two and subdivided
until one has 12-16 pieces of 1 mm 3 or less.
Transfer the pieces to two dry 60-mm tissue culture dishes. The samples can easily be transferred with the tip of the blade of one of the
scalpels and will adhere to the dry dish. Arrange samples to fit under a
round 25-mm coverslip as shown. Use a sterile wooden applicator or a
disposable needle to place a spot of sterile silicone grease for one quadrant of the coverslip. Lower a coverslip into place with a forceps so that it
covers the skin pieces and the silicone grease. If skin pieces lie beyond it,
tuck them under the coverslip with a scalpel blade before pressing the
coverslip down tightly enough to secure them. Add about 0.2 ml medium
with a small pipette (1 ml) at the margin of the coverslip and allow it to
move across by capillarity to displace the air beneath the coverslip. Once
e coverslips have medium beneath them (and no large air bubbles), 5 ml
medium are added to the dishes directly to the top of the coverslips.
e dishes are placed in trays, incubated in a humidified CO2 incubator
with 5% CO2 in air, and not touched for 10 days. At 10 days, the dishes are
examined with an inverted microscope. By this time an epithelial outgrowth may be evident at the margin of the skin pieces. From this point,
dishes are " f e d " 3 times a week. Medium is aspirated from the edge of the
tilted dishes and replaced with 5 ml of fresh medium. By 2 weeks, fibroblasts are evident at the margins of the epithelial layer and moving out
from the explant. Four to six weeks from the time the biopsy was taken,
the cells will have grown well beyond the coverslip and be ready for
transfer (splitting).
At this point, the medium is removed, and the dishes are washed with
2 ml of 0.25% trypsin, following which 1.0 ml of 0.25% trypsin is added
and the dishes placed at 37 for 5-7 min. The cell layer will be seen
detaching from the dish. Then the coverslips are dislodged to be certain
448
SPECIFIC CELL LINES
[38]
the trypsin has had access to cells beneath them, and the cell layer is
dispersed by adding 4 ml of serum-containing medium and pipetting up
and down 3-4 times. The cells are sedimented in a sterile tube by centrifuging at 1000 rpm, resuspended in 5 ml of serum-containing medium,
and added to a 60-mm dish. Within 2-4 days, this dish should be confluent
and ready to split further. The trypsinization procedure outlined in this
paragraph is repeated. Resuspended cells from each 60-mm dish are distributed into 2 100-mm dishes, i.e., a 1:5 split, which are fed 3 times
weekly with 10 ml of medium. Within 7 days, the dishes should be confluent and may be propagated further by splitting 1:5 to 1:8 (see Section 4)
or prepared for storage by freezing (Section 3).
3. Freezing Cells for Storage and Recovering Them from Frozen Storage
Materials
Dimethylsulfoxide (DMSO), not sterilized

Fetal calf serum: heat inactivated as in Section 2
Freeze medium is prepared by adding 5 ml of DMSO to 95 ml of
heat-inactivated fetal calf serum.
Pro Vials (Cooke), also called sterile serum freeze-vials (Fisher), 2-ml
size (2-ml sterile cryotubes, Vangard International are equivalent)
- 70 freezer
It is generally advisable to freeze cells in early passage to guarantee a
source of material for future investigation. We have found 5% dimethylsulfoxide (DMSO): 95% heat-inactivated (60 for 30 min) fetal calf serum a
good freeze medium for every type of fibroblast examined. However, the
exposure to DMSO prior to freezing should be as brief as possible, and the
DMSO should be diluted out promptly on thawing. For many cell lines
(but not mucolipidosis II cell lines), serum-containing medium containing
10% glycerol is also a satisfactory freeze medium. (See this volume [3].)
The following explains the method to prepare five freeze vials from
two 100-mm dishes. This can be done as soon as one has two subconfluent
to confluent 100-mm dishes from the primary explant. The freezing may
be delayed one or two passages if one wants to prepare a considerably
larger number of vials of a line that he expects to use for a long time. Cells
are detached from two 100-mm dishes at a time by trypsinization. Aspirate medium, wash with 2.0 ml of 0.25% trypsin, and then allow 1.0 ml of
trypsin to bathe the cells for 5 min at 37. Disaggregate the detached cells
by pipetting cells in serum-containing medium (5 ml/dish) and sediment.
Resuspend in 5.2 ml of freeze medium (which should be at 4) and
transfer 1 ml into each of five freeze vials. Cap and move immediately to a
[38]
449
- 7 0 freezer to allow them to freeze. The vials of frozen cells are transferred to a liquid nitrogen storage tank the next working day. We routinely
thaw one vial and initiate the culture to be certain that the cells are
recoverable from frozen storage and not contaminated. An alternate
method to freezing in a -70 freezer is use of a BF-5 cap attachment for a
Linde Liquid nitrogen tank which allows slow freezing of nine vials at a
time in the vapor phase. They are held here for 90 min before transfer to a
cannister for storage.
To recover cells from frozen storage, one warms the bottom of the vial
to 37 in a water bath immediately on removal from the freezer. As soon
as the sample has melted (which is obvious as the sample turns from ice to
liquid) the vial is wiped with 95% ethanol to sterilize the outside, air dried,
and the cells removed from the opened vial with a sterile 3-cc syringe with
a 19-gauge needle. The cells are diluted with 10 ml of serum-containing
medium to dilute out the DMSO; cells are sedimented, resuspended in 5
ml of serum-containing medium, and placed in a 60-mm dish. Within 1-2
days at 37 , this sample is usually ready to propagate and is "split" 1:5
into two 100-mm dishes.
From that point, most cell lines can be split every 7-8 days from 1:5 to
1:8 to produce large amounts of cells.
4. Propagation of Cultured Fibroblasts

Fibroblasts can be propagated by passaging, or splitting, every 7-8
days. Splitting 1:5 means detaching cells by trypsinization and distributing
them into vessels with 5 times the original surface area. Note that the
actual surface area available for growth is smaller than one would estimate from the outside diameter of plastic dishes (see the table). Plating
human fibroblasts at low density (such as produced by 1:15-1:20 splits) is
followed by a lag in initial growth rate with a disproportionately long time
required to achieve confluence. To generate large amounts of cells, one
can easily split 1:5-1:8 every week for most early passage cell lines. After
20-30 passages, growth of cell lines will slow progressively. Unless one is
Commonly used plastic

disposable dishes
Diameter of
growth surface (mm)
Area of
growth surface (cm)
Media per
feeding (ml)
100 mm x 20 mm
60 mm x 15 mm
35 mm x 10 mm
81
51
32
51.5
20.4
8.5
10
5
2
450
SPECIFIC
CELL
LINES
[39]
interested in studying the phenomenon of senescence in culture, it is best

at this point to thaw early passage cells to produce more material.
Growth of cells in roller bottles has considerable advantage when large
amounts of cells are desired. Single bottles can provide the surface area of
10-30 100-mm dishes. Once growth is established in roller bottles, feeding
each bottle requires far less labor and somewhat less material than feeding
10-30 plates with the corresponding surface area.
The Coming disposable tissue culture bottles (490 cm 2) provide a convenient amount of material for many diagnostic studies, such as surveys
of lysosomal enzyme levels. HEPES buffer, 15 mM, is added to the tissue
culture medium. Bottles are started from two 100-mm dishes (a 1:5 split)
and fed twice weekly with 100 ml of media for 2 weeks. Cells may be
harvested by trypsinization, or without trypsin by washing the cell layer
with phosphate-buffered saline, detaching the cells by scraping the surface with a special rubber policeman (Bellco), and sedimenting in the
phosphate-buffered saline, after which the cell pellet may be frozen,
lyophilized, or disrupted to produce a cell extract.
[39] Cell L i n e s f r o m I n v e r t e b r a t e s
By W. FRED HINK
Cell lines from about 70 different species of invertebrates have been
reported. Most are from insects, but several lines have been established
from ticks and snails. The lines have developed from primary cultures of
embryos, hemocytes, ovaries, imaginal discs, fat bodies, and macerated
larvae, pupae, or adults. This article presents general information that is
applicable to the culturing of most of these lines. It does not deal with
primary cultures or details about each specific cell l i n e .
Maintenance of Cell Lines
Incubation Temperature and Atmosphere

Invertebrate cell lines grow most rapidly within the range of 25-30 .
The Trichoplusia ni (TN-368) moth line is a typical example and has maximum growth rates at 26-30 . At 37 the ceils are in poor morphological
condition and at 20 they grow slower but remain viable for longer periods
of time than when grown at 26 or 30. An incubator, set at 28, is an
adequate environment for culturing all invertebrate cell lines.
MIL~HODS IN ENZYMOLOGY, VOL. LVIII

ISBN 0-12-181958-2
[39]
INVERTEBRATES
451
Many vertebrate cell lines are grown in an atmosphere of 5% C02 in

air to reduce the loss of C02 from the bicarbonate buffers. Since invertebrate cell culture media are not primarily buffered with bicarbonate, the
cells are incubated under a normal atmosphere.
Subculturing
While almost any cell culture vessel is adequate (TC petri dish, glass
T-flask, polystyrene disposable TC flask, etc.), most lines are maintained
in TC flasks with 5-ml volumes of media and 25-cm 2 growth surfaces, and
this discussion will center around the use of this size flask. Subculturing is
done by transferring an aliquot of cells from a parent culture to a new flask
containing fresh medium. New cultures are initiated with 1-3 x 105cells/ml.
The ratio of the volume transferred from the parent culture to the volume
of the new culture is termed the split ratio. A 0.5-ml aliquot of suspended
cells from a 5.0-ml parent culture transferred to 4.5 ml fresh medium is a
1:10 split ratio. Split ratios for different lines vary between 1:2 and 1:25,
and subculture intervals range from 2 days to several weeks. When one
first receives a specific line and attempts initial maintenance, it is advisable to follow the split ratio and subculture interval as used by the parent
laboratory.
If difficulty is encountered in keeping cells in good condition, the
subculture techniques may be modified. Cells should be subcultured while
in the exponential growth phase or just as they are entering the stationary
phase. To determine when to subculture, growth curves should be done
soon after receiving cells. The single cell suspensions are counted with a
hemocytometer. Growth curves from representative insect cell lines show
that after initiation of new cultures, there are lag periods of several hours
to 3-4 days. Exponential growth may last for 2-7 days after which the cell
population numbers level off or decline. Population doubling times during
exponential growth range from 16 hr for T. ni to 48 hr for Antheraea
eucalypti, and maximum cell densities are from 1 x 106 to 1 x 107 cells/ml.
Many insect lines grow loosely attached or in suspension, and these
are gently agitated to obtain a homogenous suspension prior to
subculturing.
Attached cell lines must be released from the culture flask and dispersed as single cells before subculturing. For most attached lines, the
cells are removed by scraping with a rubber policeman and resultant
clumps dispersed by gentle pipetting, xa Other techniques employ a small
i p. E. Eide, J. M. Caldwell, and E. P. Marks, In Vitro 11,395 (1975).
J. L. Vaughn, R. H. Goodwin, G. J. Tompkins, and P. McCawley,In Vitro 13, 213 (1977).
452
SPECIFIC CELL LINES
[39]
TABLE I
R I N A L D I N I ' S SOLUTION a
Compound
mg/100 ml solution
NaC1
KCI
Na.zPO4 H20
D-glucose
NaHCOz
NaaC6H507 "2 H20
Demineralized water
800
20
5
100
100
67.6
to 100 ml
a Rinaldini.3
magnetic spin bar followed by pipetting or simply pipetting medium over

the cell layer to flush them off. Insect cells are fragile, and vigorous
agitation or shearing must be avoided.
If cells are damaged by physical techniques, they are released from the
culture flask with enzymes. Since Ca z and Mgz ions apparently function
in stabilizing cellular adhesiveness, enzymes are dissolved in Ca 2- and
MgZ+-free saline solutions. Rinaldini's solution (Table I), 3 i.e., Tyrode's
solution4 with Ca and Mg salts replaced with sodium citrate, is one of the
most often used.
Trypsin is usually used at 0.1 or 0.2%. 5.6 To prepare, add 100 mg or
200 mg to about 1.0 ml Rinaldini's solution and stir to make a paste. With
continuous stirring, add the additional 99 ml of the salt solution. If necessary, the pH should be adjusted to 7.8 and the solution sterilized by
filtration. SinGe enzymic activity is lost over a period of several .weeks
when stored at 4 , the preparation should be divided into convenient volumes and stored frozen at - 2 0 . Prepared trypsin solutions (2.5%) in
salines without Mg and Ca may be purchased from commercial sources.
To subculture attached cells, all medium is removed from the flask and
2 ml of trypsin solution are added. The trypsin and cells are incubated for
2-5 min (depending on cell line), and the solution is gently pipetted to flush
cells from the flask. Cells are transferred to sterile centrifuge tubes containing 1 ml of heat-inactivated fetal bovine serum (FBS) to stop enzyme
activity. After centrifugation at 380 g for 10 min, the serum is removed by
pipette and fresh medium added. Aliquots are transferred to new culture
flasks to obtain the appropriate split ratio.
3 L. M. Rinaldini, J. Physiol. (London) 123, 20P (1954).
a M. V. Tyrode, Arch. Int. Pharmacodyn. Ther. 20, 205 (1910).
5 I. Schneider, J. Cell Biol. 42, 603 (1969).
6 j. Mitsuhashi, Annot. Zool. Jpn. 48, 139 (1975).
[39]
INVERTEBRATES
453
Storage
If cells are not transferred to fresh medium when they reach maximum
density, they begin to die and, depending on the cell line, all will be dead
in 10-20 days. The longevity of cell viability can be increased by transferring newly subcultured cells to 5. In my laboratory, the following lines
were stored at 5 in fresh media and samples taken every week to determine if they would recover: T. ni (TN-368) can be stored 14 days; A.
eucalypti for 21 days; Grace'sAedes aegypti for 40 days; and Lasperesia
pomonella (CP-1268 and CP-169) for 90 days. In all cases, cell growth was
slower after storage, requiring about five subcultures for the growth rate
to return to normal.
As with vertebrate cells, insect cells also may be frozen in liquid
nitrogen or ultra-low temperature freezers at - 7 0 to - 9 0 . The cell culture
media are supplemented with either sterile 10% glycerol or 8% dimethysulfoxide. Cells from a culture in late stage of exponential growth
are centrifuged at 380 g for 10 rain and resuspended in glycerolsupplemented medium at one-half the original volume. The cell suspension
is dispensed in 1.0-ml volumes into glass ampules and immediately sealed
by use of a propane torch and glass rod. Animal cells should be slowly
frozen at rates of 1-3 per minute to about -30; programmable freezers
may be used for this purpose. We obtain slow freezing by wrapping the
ampules in insulation before placing them in a freezer and, after 2 hr,
transferring them to liquid nitrogen. Cells are thawed quickly by removing
them from the nitrogen and plunging the ampules in a 30 water bath
where they are agitated to accelerate melting. The content of one ampule
is added to 4 ml of fresh medium in a TC flask. For most lines, the cells
will attach within 2-4 hr at which time the medium with the glycerol
supplement is removed by pipetting and 5 ml of fresh medium added. If
cells do not attach, the cryoprotective agent may be removed by centrifuging freshly thawed cells, discarding the supernatant fluid, and resuspending cells in fresh medium.
Quality Control
Occasionally, a batch of basal medium, serum, hydrolysate, or other
ingredient will be toxic or will not support normally expected growth
rates. For this reason, the lot number of all ingredients in the complete
medium should be recorded. When an ingredient with a new lot number is
to be employed and before supply of the previously used batch is depleted,
it should be incorporated in a small batch of complete medium that contains no other new ingredients. This should be used to culture cells for at
454
SPECIFIC CELL LINES
[39]
least 10 subcultures; the new ingredient is judged satisfactory if cells and

growth are normal.
Suspension Culture
Large numbers of single cells may be obtained by growing cells in
vessels where they are kept in suspension. Spin flasks, with magnetically
driven impellers designed to avoid shear damage, flasks on gyratory shakers, or fermentors are employed. These cells may be used in molecular
and cellular biology studies involving such techniques as extraction of
macromolecules or isolation of cellular organelles.
Mosquito (Aedes albopictus) cells, 7 established and normally grown in
Mitsuhashi and Maramorosch medium (Table VI) have been adapted to
grow in vertebrate culture medium consisting of Eagle's medium
(Joklik-modified) with 1% seven nonessential amino acids, 10% FBS, and
1% lactalbumin hydrolysate, s The salt concentrations and osmotic
pressures of these two media are similar. Upon transfer of cells to the new
medium, there was a brief lag in growth before multiplication commenced. Cells were cultured in stationary flasks for several months before
attempting to adapt them to suspension culture. Thereafter, ceUs were
alternately passed in stationary and suspension cultures until a rapidly
growing suspended subline was obtained. Cells in suspension clump with
several hundred cells per clump. To quantify cells accurately, they were
treated with the detergent, NP40, and nuclei were counted with a
hemocytometer. The population doubling time was 21 hr at 25. After
adaptation to suspension culture, the lactalbumin hydrolysate supplement
was removed and cells could then be labeled with radioactive leucine.
The same A. albopictus cell line was grown in 150-ml spin flasks in
Mitsuhashi and Maramorosch medium with no medium modifications. 9
Singh's A. aegypti and Hsu's Culex quinquefasciatus10 and Culex
tritaeniorhynchus 11 were also cultured under the same conditions. 9
Fruit fly (Drosophila melanogaster, line 2) cells, 12 established and
normally grown in Schneider medium (Table VIII), were also adapted to a
vertebrate cell culture medium. TM The medium was Dulbecco's modified
Eagle's supplemented with 10% FBS, 0.5% lactalbumin hydrolysate, and
7 K. R. P. Singh, Curr. Sci. 36, 506 (1967).
s A. Spradling, R. H. Singer, J. Lengyel, and S. Penman,Methods CellBiol. 10, 185 (1975).
9 T. K. Yang, E. McMeans, L. E. Anderson, and H. M. Jenkin, Lipids 9, 1009 (1974).
10 S. H. Hsu, W. H. Mao, and J. H. Cross, J. Med. Entomol. 7, 703 (1970).
11 S. H. Hsu, S. Y. Li, and J. H. Cross, J. Med. Entomol. 9, 86 (1972).
12 I. Schneider, J. Embryol. Exp. Morphol. 27, 353 (1972).
13 j. Lengyel, A. Spradling, and S. Penman, Methods Cell Biol. 10, 195 (1975).
[39]
INVERTEBRATES
4.55
Gibco MEM nonessential amino acids. The pH was adjusted to 6.9 and cultures gassed with 5% CO2 in air. A series of manipulations consisting of
addition of various supplements and culturing in roller bottles were required to obtain suspension cultures. Cells in suspension reach densities
of 1-4 106 cells/ml with a generation time of 30 hr. Another D.
m e l a n o g a s t e r line, (GMz), 14 was grown in insect cell culture media in
Erlenmeyer flasks on a gyratory shaker at 180 rpm with a 1-inch radial
stroke.15
The moth cell line (T. n i , TN-368) 16 grew in suspension in 100-ml spin
flasks and 4-liter fermentors. 17 The culture medium (TNM-FH, Table IV)
was modified by addition of 0.1% methylcellulose (50 cps) to prevent cell
clumping. Aeration of cultures produced higher growth rates and final cell
densities. Periodic adjustment of medium pH also resulted in higher maximum cell densities.
Media
According to my most recent tabulation, there are 34 different formulations of media for culturing invertebrate cell lines. It is impractical to
consider all of these here, and I will therefore discuss only the most
common media. Before dealing with specific media, some general comments will be made.
Insect cell culture media differ from media designed for vertebrate
cells by having different ion balances, elevated amino acid concentrations,
generally phosphates as buffers rather than carbonates, usually lower pH,
and higher osmotic pressures. There is some question as to the significance of these differences since a few insect cell lines grow in vertebrate
media. 8,1a,18,19 Established insect cell lines appear to be rather insensitive
to Na/K ratios altered from 0.67-1.38 and calcium and magnesium levels
reduced by one-half and three-fourths, respectively. 2 However, as with
vertebrate cells, excess potassium is toxic. 21 The increased amounts of
amino acids are present in some formulas because insect blood has high
concentrations of these compounds. Requirements for specific amino
acids vary with different lines, and patterns of utilization are different
14G. Mosna and S. Dolfini,Chromosoma 38, 1 (1972).
15T. Miyake, K. Saigo, T. Marunouchi, and T. Shiba, In Vitro 13, 245 (1977).
18W. F. Hink, Nature (London) 226, (1970).
17W. F. Hink and E. Strauss, in "Invertebrate Tissue Culture" (E. Kurstak and K.
Maramorosch, eds.), p. 297. Academic Press, New York, 1976.
18M. Pudney, personal communication.
~9A. H. Mclntosh, K. Maramorosch, and C. Rechtoris,In Vitro 8, 375 (1973).
20j. L. Vaughn,In Vitro 9, 122 (1973).
21T. J. Kurtti, S. P. S. Chaudhary, and M. A. Brooks,In Vitro 11,274 (1975).
456
SPECIFIC CELL LINES
[39]
even between cell lines from the same insect species, z2 In fact, some
amino acids essential for growth of living intact insects are not essential
for cells in vitro .23
Osmotic pressure and pH are important parameters in media formulation. Studies with moth cell lines indicate that growth is reduced when
osmotic pressure varies more than about 40 msM/kg on either side of the
osmotic pressure (316 msM/kg) of the "normal" medium. 24 Our policy is
to have less than _ 10 msM/kg variation in batches of media. The pH of
most media range from 6.2-7.0 with media for som.e cockroach lines being
as high as 7.4. The optimum initial pH for moth (Heliothis zea) cells is
between 6.5-7.025 and for two leafhopper cell lines it is 6.3-6.4. 26 When
the pH of the medium was held at preset levels of 5.8, 6.0, 6.3, 6.5, and 6.7
for the life of suspension cultures ofT. ni cells, the optimum was between
pH 6.0-6.5, whereas pH 6.7 was clearly detrimental.
Supplements and Additives to Basal Media
For purposes of discussion, the basal medium is that portion of the
complete medium that contains synthetic chemically defined compounds.
There are no chemically defined media for invertebrate cell lines, and all
basal media are supplemented with undefined natural ingredients. Fetal
bovine serum, at concentrations of 5-20%, is used in nearly all media, and
it is usually heat inactivated at 56 for 30 rain. Other complex ingredients
such as lactalbumin hydrolysate furnish amino acids, and yeast hydrolysate provides both amino acids and vitamins.
Antibiotics are often added, but their use should be discouraged because they can mask microbial contamination. If contamination is present, it is better to know it immediately instead of carrying undetected
low-level contaminants and having them affect experimental results. It
appears that dependence on the presence of antibiotics leads to the development of rather careless nonsterile techniques. The two most common
antibiotics and concentrations are penicillin G at 100 U/ml and streptomycin sulfate at 100/xg/ml. A 100 stock solution is prepared by dissolving
10,000 U penicillin and 10 mg streptomycin/ml PBS, filter-sterilized; 1.0 ml
is added to 99 ml of complete medium just before use.
22 W. F. Hink, B. L. Richardson, D. K. Schenk, and B. J. Ellis,Proc. Int. Colloq. lnvertebr.
Tissue Cult., 3rd, 1971 p. 195 (1973).
23 j. Mitsuhashi, J. Insect Physiol. 22, 397 (1976).
z4 T. J. Kurtti, S. P. S. Chaudhary, and M. A. Brooks, In Vitro 10, 149 (1974).
25 T. J. Kurtti and M. A. Brooks, in "Insect and Mite Nutrition" (J. G. Rodriguez, ed.), p.
387. North-Holland Publ., Amsterdam, 1972.
26 G. Martinez-Lopez and L. M. Black, In Vitro 13, 777 (1977).
[39]
INVERTEBRATES
457
M e d i a Sterilization
Media are usually sterilized by pressure filtration through membranes

with 0.22-0.20 /zm pore sizes. A c o m m o n procedure is to filter basal
media and then add sterile sera, as purchased from commercial suppliers,
aseptically to the basal media. Since sera may contain microbial contaminants, we make up complete media and then filter. It is advisable to
c h e c k media for sterility before using them to culture cells. One may take
a small aliquot from storage bottles o f media and incubate at 28 for at
least a week or leave all the media at room temperature for 7 days. We use
the latter a p p r o a c h as it assures us that the entire contents o f the storage
bottles are free of viable microorganisms.
F o r m u l a s a n d Preparation o f M e d i a
GRACE MEDIUM
This medium (Table 11)27 is designed to resemble the chemical composition of silkworm, B o m b y x m o r i , h e m o l y m p h and is a modification o f
W y a t t ' s medium, z8 The basal medium is supplemented with various combinations and amounts of FBS, chicken egg ultrafiltrate, yeast extract,
lactalbumin hydrolysate, and bovine serum albumin prior to use. It is
used to culture cell lines from the Australian emperor gum moth, A.
eucalypti; zr silkworm, B. Mori; 29 spruce b u d w o r m , Choristoneura
f u m i f e r a n a ;3o cotton bollworm, H . zea ;31 codling moth, L. p o m o n e l l a ;32
g y p s y moth, L y m a n t r i a dispar; 33 forest tent caterpillar, M a l a c o s o m a disstria; 34 t o b a c c o hornworm, M a n d u c a s e x t a ; 1 cynthia moth, S a m i a
cynthia ;35 cabbage looper, T. ni ;16 mosquito, A . aegypti ;36 mosquito, A e d e s
v e x a n s ;37 and mosquito, Culiseta inornata .37
PROCEDURE FOR MAKING l0 LITERS OF GRACE MEDIUM 3s
1. Weigh out amino acids (B) and combine them in a 400-ml beaker.
2. Weigh out sugars (C) and combine them in a 400-ml beaker.
27T. D. C. Grace, Nature (London) 195, 788 (1962).
2s S. S. Wyatt, J. Gen. Physiol. 39, 841 (1956).
30S. S. Sohi, Proc. Int. Colloq. Invertebr. Tissue Cult., 3rd, 1971 p. 75 (1973).
31W. F. Hink and C. M. Ignoffo,Exp. Cell Res. 60, 307 (1970).
32 W. F. Hink and B. J. Ellis, Curr. Top. Microbiol. Immunol. 55, 19 (1971).
33j. M. Quiot, personal communication.
34 S. S. Sohi, Proc. Int. Colloq. lnvertebr. Tissue Cult., 3rd, 1971 p. 27 (1973).
35j. Chao and G. H. Ball, Curr. Top. Microbiol. Imrnunol. 55, 28 (1971).
37 B. H. Sweet and J. S. McHale, Exp. Cell. Res. 61, 51 (1970).
38W. F. Hink, unpublished data.
458
SPECIFIC CELL LINES
[39]
TABLE II
GRACE INSECT TISSUE CULTURE MEDIUM (g/lO liters)a
Salts
NaH2PO4 H~O
NaHCO3
(A) KCI
MgCI2 6 1-120
MgSO4 7 H~O
(I) CaCI~.2 H20 (sep.)
I 0.08
3.50
22.40
22.80
27.80
13.25
Amino acids
L-arginine HCI
L-aspartic acid
L-asparagine
L-alanine
B-alanine
L-glutamic acid
L-glutamine
L-glycine
L-histidine
L-isoleucine
L-leucine
L-lysine HC1
L-methionine
L-proline
L-phenylalanine
DL-serine
L-tryptophan
L-threonine
L-valine
L-cystine HCI (sep.)
L-tyrosine (sep.)
7.0
3.5
3.5
2.25
2.0
6.0
6.0
6.5
25.0
0.5
0.75
6.25
0.5
3.5
1.5
11.0
1.0
1.75
1.0
0.25
0.5
(B)
(G)
(H)
Sugars
Sucrose
Fructose
Glucose
266.8 /
4.0
7.0
(C)
Organic acids
Malic
Alpha-ketoglutaric
Succinic
Fumaric
6.7 1
3.7
0.6
0.55
(D)
Vitamins
Thiamine HC1
Riboflavin
Ca pantothenate
Pyridoxine HC1
p-Aminobenzoic acid
Folic acid
Niacine
Isoinositol
Biotin
Choline chloride
0.0002
0.0002
0.0002
0.0002
0.0002
0.0002
0.0002
0.0002
0.0001
0.002
(E)
Antibiotics
Penicillin G, Na salt
Streptomycin sulphate
0.3 "~
1.0 /
(F)
a Grace.2r
3. Weigh out organic acids (D) and combine them in a 100-ml beaker.
4. Weigh out L-cystine (G) and L-tyrosine (H) and combine them in a
50-ml beaker.
5. Weigh out calcium chloride (I) and place in a 100-ml beaker.
6. A x 1000 stock solution of vitamins (E) is made up by adding 2 mg
thiamine, 2 mg riboflavin, 2 mg Ca pantothenate, 2 mg pyridoxine, 2 mg
p-aminobenzoic acid, 2 mg folic acid, 2 mg niacin, 2 mg isoinositol, 1 mg
biotin, and 20 nag choline chloride to a 100-ml volumetric flask and bringing up to volume with water. Filter-sterilize, dispense in 10-ml portions,
and store at - 2 0 .
7. If antibiotics (F) are to be used, they are made up in a x 100 stock
[39]
INVERTEBRATES
459
solution in H a n k s ' s PBS, filter-sterilized, and added to complete medium

just before use.
8. Put (C) into a 15-liter beaker. Add 4 liters water using some o f this to
rinse the beaker that held the dry sugars. Stir with a magnetic stirrer.
9. Weigh out salts (A) starting with NaH2PO4" H20, transfer to 15liter beaker, and proceed in order down through the list in Table II. To
prevent precipitation, each salt must be added separately and dissolved
before adding the next salt.
10. Put (B) into a separate 4-liter beaker, add 2 liters water, and stir
until dissolved.
11. Put (D) into a separate 4-liter beaker, add 2 liters water, and stir
until dissolved.
12. Make up 250 ml 1 N K O H by placing 13.03 g K O H in a 250-ml
volumetric flask and bring up to volume with water.
13. Add the 1 N K O H to (D) with continuous stirring to bring p H to
5.9. Note volume of K O H used (usually 170-200 ml).
14. Make up 1 N HCI by adding 6 ml concentrated HCI (37%) to 54.0
ml with water.
15. Add 15 ml of the 1 N HCI to (GH) and dissolve these amino acids.
16. Add 50 ml water to (I) and dissolve.
17. Add (B) and pH-adjusted (D) to 15-liter beaker.
18. Add (GH) to 15-liter beaker.
19. Add (I) to 15-liter beaker.
20. Adjust p H of the contents of 15-liter beaker to 6.2 with 1 N HC1.
N o t e volume of HCI added.
21. Add 10 ml of 1000 stock vitamins (E).
22. While combining ingredients, the volumes are recorded in order to
know how much additional water is needed to bring final volume to 10
liters. H e r e is an example of our calculations for a typical batch:
Fraction
A
B
C
D
KOH
GH
I
HC!
E
Totals:
Volume occupied by
chemicals when
dissolved (ml)
37.0
53.0
170.0
5.0
2000.0
2000.0
2OOO.0
2000.0
--
0.5
3.0
--268.5
Volume of
liquid (ml)
175.0
15.0
50.0
27.0
10.0
8277.0 =
8545.5 m l
460
SPECIFIC CELL LINES
[39]
23. Add water to bring final volume to 10 liters.

24. Filter through 0.22-~m membrane and store at 5.
25. As part of our quality-control program, we take the osmotic
pressure of all batches, and variation from normal indicates that errors
were made during formulation. We also make up 100 ml complete medium
from a new batch of Grace basal medium. This is evaluated through 10
subcultures before concluding that it is satisfactory.
Tables l i P 8a and IV give two examples of complete media that use
Grace basal medium.
GOODWIN IPL-52 MEDIUM39
This medium (Table V) supports growth of cell lines from moths,
S p o d o p t e r a f r u g i p e r d a , H . z e a , and T. ni. a9 It also has been used to obtain
quickly developing monolayers from primary cultures of pupal moth

ovaries.
PREPARATION OF GOODWIN I P L - 5 2 M E D I U M 4
1. Prepare stock solution (A) by dissolving the following in 100 ml

water:
(NH4)MoTO24 4 H20
COC12 6 H~O
CuCI~ 2 H20
MnC12 4 H20
ZnC12
4.0
5.0
19.5
2.0
4.0
mg
mg
mg
mg
mg
2. Prepare stock solution (B) by dissolving the following in 100 ml

water:
FeSO4" 7 H20
Aspartic acid
82.3 mg
53.2 mg
3. Dissolve all ingredients in Table V from, and including, L-arginine

to, and including, folic acid in order listed in 900 ml H20.
4. While stirring, add 1 ml stock solution (A) and 0.67 ml stock solution (B).
5. Adjust pH to 6.3 with 10% NaOH (approximately 5 ml).
6. Dissolve CaCI2 separately in 50 ml H20 and add slowly while stirring.
7. Bring total volume to 1000 ml with H20.
8. Sterilize by filtration.
3sa C. E. Yunker, J. L. Vaughn, and J. Cory, Science 155, 1565 (1967).
aa R. H. Goodwin, In Vitro 11, 369 (1975).
40 R. H. Goodwin, personal communication (1977).
[39]
INVERTEBRATES
461
TABLE III
YUNKER, VAUGHN, AND CORY MEDIUMa
90.0 ml Grace medium
10.0 ml FBS
10.0 ml Egg ultrafiltrate
1.0 g Bovine plasma albumin
a Yunker
e t a l . asa
TABLE IV
HINK (TNM-FH) MEDIUMa
90.0
8.0
0.3
0.3
ml Grace medium
ml FBS
g Lactalbumin hydrolysate
g TC Yeastolate
a Hink and Strauss. ~r
TABLE V
GOODWIN IPL-52 MEDIUM (mg/liter)a'b
I(NH4)MorO24 4 H20
[ C o C l 2 6 H20
AJCuClz 2 H20
[MnC12 4 H20
[ Zn C12
B J'FeSO4_ 7 HzO
LAspartic acid
L-Arginine hydrochloride
L-Aspartic acid
L-Asparagine
L-Cystine (dissolve in
10% NaOH)
L-Glutamic acid
L-Glutamine
L-Glycine
L-Histidine
Hydroxy-L-proline
L-Isoleucine
L-Leucine
L-Lysine hydrochloride
L-Methionine
L-Proline
L-Phenylalanine
a Goodwin?9
b Goodwin.40
0.040
0.050
0.195
0.020
0.040
0.0551
0.0356
800
1000
1300
I00
1300
1000
400
200
800
500
400
700
1000
600
1000
DL-Serine
L-Threonine
L-Tryptophan
L-Tyrosine (dissolve in
10% NaOH)
L-Valine
Dextrose
Maltose
MgSO4 7 H20
KCl
NaHC03
NaHzPO4 " H20
TC Yeastolate
Acetyl B-methyl choline
chloride
Cyanocobalamin (B~2)
Isoinositol
Folic acid
CaCI2
Turkey serum
Chicken serum
Calf serum
600
200
100
250
500
5000
1000
1880
2600
350
1160
5000
250
1
10
1.2
50O
3%
3%
3%
462
SPECIFIC
CELL LINES
[39]
TABLE VI
MITSUHASHI AND MARAMOROSCH MEDIUM (g/liter) a
NaX-X2VO4 1-120
MgC12 6 1-120
KCI
CaC12 2 H20
NaC1
NaHCO3
0.20
0.10
0.20
0.20
7.00
0.12
D-glucose
Lactalbumin hydrolysate
Yeastolate
FBS
Penicillin
Streptomycin
4.00
6.50
5.0
20%
100 U/ml
100/,~g/ml
a Mitsuhashi and Maramorosch. 41
9. Before use, the above basic medium is supplemented with heattreated (60 for 30 min) sera as given in Table V.
MITSUHASHI AND MARAMOROSCH M E D I U M 41
Many mosquito cell lines are grown in this medium (Table VI) which
contains relatively few ingredients. It supports growth of cell lines from
mosquitoes A. aegypti, r A. albopictus, 7 Aedes novalbopictus, 42 A.
taeniorhynchus, 43Aedes w-albus, 44 and Armigeries subalbatus. 43 This medium also is used for cell lines from the leafhopper, Agallia constricta, 45
and the potato tuber moth, Gnorimoschema operculeUa. 4e
PREPARATION OF MITSUHASHI AND MARAMOROSCH M E D I U M 43'47
1. The lactalbumin hydrolysate is dissolved separately in 300 ml HzO

by stirring with a magnetic stirring bar.
2. The remainder of the components (except FBS and antibiotics) are
dissolved one at a time in 400 ml water.
3. The two solutions are combined, and the pH is adjusted to 6.5 with
0.1 N KOH. The pH of 6.5 is used for leafhopper cells whereas a pH of
6.8--7.0 is used for mosquito cells.
4. Bring volume to 800 ml and filter-sterilize.
5. FBS (200 ml) is added aseptically before use.
6. If antibiotics are to be incorporated, add 20 ml of sterile stock
antibiotics consisting of 5000 U penicillin and 5000/zg streptomycin/ml.
41 j. Mitsuhashi and K. Maramorosch, Contrib. Boyce Thompson Inst. 22, 435 (1964).
42 U. K. M. Bhat and P. Y. Guru, Expl. Parasitol. 33, 105 (1973).
4~ I. Schneider, personal communication (1977).
44 K. R. P. Singh and U. K. M. Bhat, Experientia 27, 142 (1971).
4~ R. Chiu and L. M. Black, Nature (London) 215, 1076 (1967).
46 U. Pant, A. F. Mascarenhas, and V. Jagannathan, Indian J. Expl. Biol. 15, 244 (1977).
47 D. E. Lynn, personal communication (1977).
[39]
INVERTEBRATES
463
TABLE VII
LEIBOVITZ L-15 MEDIUM (mg/liter) a
Amino acids
DL-alpha alanine
L-Arginine (free base)
L-Asparagine
L-Cysteine (free base)
Glycine
L-Histidine (free base)
DL-Isoleucine
(A). L-Leucine
L-Lysine
DL-Methionine
DL-Phenylalanine
L-Serine
DL-Threonine
L-Tryptophane
DL-Valine
(B) L-Glutamine
(C) L-Tyrosine
Salts
[ NaCI
(D) ~! KCI
| MgCI2 6 H20
[ MgSO4 " 7 H20
/
450.0
500.0
250.0
120.0
200.0
250.0
250.0
125.0
75.0
150.0
250.0
200.0
600.0
20.0
200.0
300.0
300.0
8000.0
400.0
200.0
200.0
CaC12
KH2PO4
Na~HPO4
Phenol red
Sodium pyruvate
140.0 (E)
60.0 ]
50.0 / (~
10.01
550.0 (G)
Vitamins
DL-Ca pantothenate
Choline chloride
Folic acid
Inositol
Nicotinamide
Pyridoxine HCI
Riboflavin-5' phosphate
Thiamine monophosphate
Na2HPO4
1.0
1.0
2.0
1.0 ' (H)
1.0
0.1
1.0
140.0
Sugar
Galactose
900.0
FBS
1.0"
10%
a Leibovitz.4S
L E I B O W l T Z L - 1 5 M E D I U M 48
This medium (Table VII), which was originally developed for vertebrate cells, supports cell lines from the hard tick, Rhipicephalus appendiculatus, 49 three other tick species, 5 and the blood-sucking bug,
Triatoma infestans. ~s The medium is usually supplemented with 10% tryptose phosphate broth and 10% FBS. The final pH of L-15 is 7.6 and must
be adjusted to pH 7.0 before use for tick cell lines.
PREPARATION OF L - 1 5 M E D I U M 48
1. Amino acids (A) are prepared as 10 solutions, filtered, divided

into 100-ml aliquots, and stored at 4 .
48 A. Leibovitz, Am. J. Hyg. 78, 173 (1963).
49 M. G. R. Varma, M. Pudney, and C. J. Leake, J. Med. Entomol. 1 1 , 6 9 8 (1975).
~o p. y. Guru, V. Dhanda, and N. P. Gupta, Indian J. Med. Res. 64, 1041 (1976).
464
SPECIFIC CELL LINES
[39]
2. Vitamins (H), L-glutamine (B), sodium pyruvate (G), and galactose

(I) are prepared as 100 solutions, filtered, divided into 10-ml aliquots,
and stored at - 2 0 .
3. L-tyrosine (C) is prepared by dissolving 400 rag/liter of water with
heat and continuous stirring. Then 750 ml are added to a l-liter flask,
autoclaved, and stored at room temperature.
4. Salts (D), (E), and (F) are made up so that when combined they are
a x 10 solution. (D) is dissolved in 600 ml water, (E) in 100 ml water, and
(F) in 275 ml water plus 25 ml of 0.4% phenol ted (see step 5). Each
solution is autoclaved separately and cooled; solution (E) is added to (D),
and (F) is added to combined (E) and (D). This sequence must be followed
to prevent formation of calcium phosphates.
5. Make up 0.4% phenol red (pH indicator) by adding 1 g of the dye
into a 500-ml flask and slowly adding 0.05 N NaOH until the phenol red is
almost in solution (about 60.0 ml). Then add NaOH, drop by drop, until
solution is attained; the solution should be deep red. Add water to 250 ml.
6. The solutions are aseptically combined to make 1 liter:
Tyrosine (x 1)
Vitamins (x 100)
Pyruvate (x 100)
Galactose (x 100)
Salts (D, E, and F) (x 10)
Amino acids (x 10)
Antibiotics (x 100)
750.0
10.0
10.0
10.0
100.0
100.0
10.0
ml
ml
ml
ml
ml
ml
ml
7. The combined ingredients are divided in 99-ml portions to which

1.0 ml of glutamine and 10% FBS are added immediately before use.
SCHNEIDER M E D I U M 43'~1
This medium (Table VIII) is for culturing cell lines from fruit flies D.
melanogaster, 12 Drosophila immigrans, ~ and Drosophila virilis. 5z There
are several other media, not discussed here, that are used for other D.
melanogaster lines.
PREPARATION OF SCHNEIDER M E D I U M 43
1. Salts and organic acids (A) are dissolved in 30 ml water. The pH

will be about 3.5 and should be adjusted to 6.0 with KOH.
2. Sugars (B) are dissolved in 5 ml water.
3. Yeastolate (C) is dissolved in 5 ml water.
~1 I. Schneider, J. Embryol. Expl. Morphol. 15, 271 (1966).
52 I. Schneider a n d A. B. Boumenthal, in "Genetics and Biology of Drosophila,"
(M. Ashburner and T. R. F. Wright, eds.) Vol. 2a. Academic Press, New York, 1978.
[39]
INVERTEBRATES
465
TABLE VIII
SCHNEIDER MEDIUM (mg/100 ml)a'~
Salts and
organic acids
~NaCI
Na2HPO4
KH, PO4
KC1
(A), MgSO4.7 H20
a-Ketoglutaric acid
Succinic acid
Fumaric acid
Malic acid
2 l0
70
45
160
370
20
10
l0
10
Sugars
_. fGlucose
B)'~Trehalose
(C) TC Yeastolate
200
200
200
Amino acids
/fl-Alanine
~ L-Arginine
(D) | L-Aspartic acid
[ L-Cysteine
50
40
40
6
L-Cystine
L-Giutamic acid
L-Glutamine
Glycine
L-Histidine
L-Isoleucine
L-Leucine
L-Lysine HCl
L-Methionine
L-Phenylalanlne
L-Praline
L-Serine
L-Threonine
L-Tryptophan
L-Tyrosine
L-Valine
CaCl2
NaHCOa
l0
80
180
25
40
15
15
165
~(D)
80
15
170
25
35
l0
50
3(
60 (E)
40 (F)
a Schneider.43
b Schneider.51
4. All a m i n o acids (D) e x c e p t c y s t i n e and t y r o s i n e are dissolved in 40
ml water. Dissolve c y s t i n e in 5.0 ml hot acidic w a t e r and t y r o s i n e in 5.0 ml
alkaline water. A f t e r t h e s e t w o amino acids are in solution, t h e y are a d d e d
slowly to the rest o f the a m i n o acid solution.
5. CaClz (E) is dissolved in 5.0 ml water.
6. N a H C O s (F) is dissolved in 4.0 ml water.
7. T h e c o m p o n e n t s in (B), (C), (E), and (F) dissolve quickly. This is
not so with (A) a n d (D) w h i c h require stirring for a b o u t 30 min with a
m a g n e t i c stirrer.
TABLE IX
HANSEN S-301 MEDIUMa
Schneider's medium diluted to 22%
Galactose
1.3 g/liter
Lactalbumin hydrolysate
4.5 g/liter
Fetal bovine serum
13%
a Hansen.~
466
SPECIFIC CELL LINES
[40]
8. All solutions, except (F), are added to (A) in alphabetical order.

9. Adjust pH to 6.45 and add NaHCOa (F). This brings the final pH to
6.68 without further adjustment.
10. Filter-sterilize.
H A N S E N M E D I U M ~a
This medium (Table IX) is for culturing a cell line (Bge) from the snail,
Biomphalaria glabrata. This is the first, established, rapidly growing cell
line from an invertebrate other than the arthropods. The pH is 7.1-7.3,
and osmotic pressure is much lower than insect cell culture media.
E. H a n s e n , in " I n v e r t e b r a t e Tissue Culture: R e s e a r c h Applications" (K. M a r a m o r o s c h ,
ed.), p. 75. A c a d e m i c Press, N e w York, 1976.
[40] C o l d - B l o o d e d V e r t e b r a t e Cell a n d T i s s u e C u l t u r e
By KEN WOLF
Introduction
Methods of cell and tissue culture I range from the very simple to the
highly complex and demanding. In general, many of the methods can be
routine, but cultures are biological systems that can and at times do evade
absolute control. The biochemist planning to use cell cultures will find
that cultures at time show greater variability than the biochemical determinations that are performed.
Most of today's animal cell and tissue culture involves materials of
mammalian and secondarily, of avian origin. Reptiles, amphibians, and
fishes--the poikilotherm vertebrates--are used the least. Without doubt
the present status of homeotherm cell and tissue culture reflects healthrelated research on man and his domestic livestock. Lower vertebrate cell
and tissue culture has much to offer the researcher; it extends the
phylogeny of vertebrates, allows work to be done through a wide range of
temperatures, and is less demanding than culture of homeotherm
materials.
Fortunately for all concerned, the physiology and functional systems
of the foregoing five classes of vertebrates have many similarities. Within
the limits to be described, the methods can be considered realistically as
simply being vertebrate cell and tissue culture.
1 This article e m p l o y s t h e standard terminology o f the Tissue Culture Association as set
forth by S. Federoff, T C A M a n u a l 1, 53 (1975).
Copyright 19"/9by AcademicPress, Inc.

ISBN 0-12-181958-2
[40]
COLD-BLOODED VERTEBRATE
467
Although commonly referred to asfishes, the elasmobranchs (sharks,

skates, rays, and allied forms) and the cyclostomes (lampreys and
hagfishes) differ considerably from teleosts or bony fishes. Culture of cells
and tissues from the last two classes of vertebrates is best described as
being in the exploratory or developmental stage.
One reason for the wide use of cell and tissue culture today is that
quality equipment, media, and supplies are readily available from scientific and biological supply houses. Commonly used biologicals were originally formulated for application with homeotherm cells and tissues; with
minor modification or, in most cases, without alteration these products
also can be used for reptilian, amphibian, and teleostean cells. In other
words, the biochemist can buy ready-made tissue culture supplies; virtually everything he will need is readily available.
The intent of this article is to enable the reader to culture lower
vertebrate cells or tissues successfully. However, the reader first should
be familiar with aseptic technique and should have had at least some
tissue culture experience. Obviously there are space limitations here, and
the unitiated will probably benefit from Part I of this volume and from cell
culture textbooks. In addition, there are review type references on culture of reptile materials by Clark, z on that of amphibians by Freed and
Mezger-Freed a and Rounds, 4 and on fishes by McKenzie and Stephenson, 5 Sigel and Beasley, 6 Sigel et al.,7 and Wolf. a The most comprehensive
work on fish cell and tissue culture is that of Wolf and Quimby. 9
Some differences between homeotherm and poikilotherm vertebrates
are worth noting, namely their environments and health histories. Microbial contamination is a hazard in initiating almost any primary culture.
However, a great many of the mammals and avian embryos used as tissue
donors are essentially healthy and usually laboratory-reared. Therefore,
2 H. F. Clark, in "Tissue Culture: Methods and Applications" (P. F. Kruse, Jr. and M. K.
3 j. j. Freed and L. Mezger-Freed, in "Tissue Culture: Methods and Applications" (P. F.
4 D. E. Rounds, in "Tissue Culture: Methods and Applications" (P. E Kruse, Jr. and M. K.
L. S. McKenzie and N. G. Stephenson, in "Tissue Culture: Methods and Applications"
(P. F. Kruse, Jr. and M. K. Patterson, Jr., eds.), p. 143. Academic Press, New York, 1973.
6 M. M. Sigel and A. R. Beasley, in "Tissue Culture: Methods and Applications" (P. F.
7 M. M. Sigel, E. C. McKinney, and J. C. Lee, in "Tissue Culture: Methods and Applications" (P. F. Kruse, Jr. and M. K. Patterson, Jr., eds.), p. 135. Academic Press, New
York, 1973.
s K. Wolf, in "Tissue Culture: Methods and Applications" (P. F. Kruse, Jr. and M. K.
9 K. Wolf and M. C. Quimby, Fish Physiol. 3, 253 (1969).
468
SPECIFIC CELL LINES
[40]
the risk of contamination in such material is low. In contrast and for all
practical purposes, the reptiles, amphibians, and all but aquarium or
hatchery-reared fishes are wild animals, and their health history is unknown or uncertain. Such feral animals can harbor systemic parasites,
microorganisms, and viruses that can contaminate primary cell cultures.
On the other hand, there are differences between homeotherms and
poikilotherms that are to the investigator's advantage. For example, cell
cultures from lower vertebrates usually require less frequent handling and
attention than homeotherm cultures, and they are far more tolerant of
neglect. Intrinsically, the coldblooded vertebrates are adapted to and
have a range of body temperatures that cannot be matched by
homeotherms. Some cold-water fish cells metabolize at temperatures
from near 0--25 whereas many warm-water fishes, reptiles, and amphibians span the range of 10-37 . If rigid temperature control need not be
maintained, many kinds of cells are grown on the bench at room temperature, i.e., 20-25 .
Physiological Salines
The usual composition of physiological salines or balanced salt solutions (BSS) for vertebrates is an isotonic salt solution containing the essential physiologic ions, a system for maintaining pH in the physiological
range, an energy source (usually glucose), and, unless contraindicated,
phenol red as a pH indicator. With appropriate adjustment of osmolarity
the common BSS are appropriate for cell cultures of mammals down
through bony fishes (Tables I).
Virtually all routine coldblooded vertebrate cell culture needs will be
met with Earle's or Hanks' BSS and Dulbecco and Vogt's phosphatebuffered saline (PBS). All are available commercially. Without their
sodium bicarbonate, Earle's and Hanks' BSS are also available as
nonsterile powders that can be dissolved in high-purity water and decontaminated by membrane filtration (0.22/zm mean pore diameter). Alternatively, these solutions are easily prepared and decontaminated or
sterilized in one's own laboratory.
Physiological salt solutions find many applications in cell culture.
Earle's BSS with 2.2 g/liter of sodium bicarbonate is intended to equilibrate in the physiological range in an atmosphere containing about 5% COs.
In normal atmosphere, the pH of Earle's BSS rises comparatively
rapidly--the quickest of the three. Hanks' BSS has a phosphate buffer
and much less sodium bicarbonate; therefore its pH changes less rapidly
in normal atmosphere. For the easiest maintenance of pH, PBS is recommended. Physiological salines may be buffered with about 15 mM Tris,
[40]
C O L D - B L O O D E D VERTEBRATE
469
TABLE I
OSMOLARITY ADJUSTMENTS SUGGESTED FOR PHYSIOLOGICAL SALINES AND MEDIA
USED FOR CULTURING LOWER VERTEBRATE CELLS AND TISSUES
Vertebrate
class
Reptile
Amphibian
Teleost
Elasmobranch
Cyclostome
Habitat
Terrestrial or
freshwater
Marine
Terrestrial or
freshwater
Freshwater
Marine
Freshwater
Marine
Freshwater
Marine
Adjustment
None
Add 0.06 M NaC1 a
Dilute 20% with water
or NaCl-free BSS
None
Add 0.07 M NaCI ~
Add urea to isotonic level
Add 2 M urea
None
Not determined
a Add 17.6 ml of 3.4 M stock NaCI solution per liter.

b Add 20.6 ml of 3.4 M stock NaCI solution per liter.
HEPES, TES, TRICINE, or BES buffer providing that about 10 mM

NaHCOa, an essential constituent of media for vertebrate cell cultures, is
also present.
Culture Media
Based as they are upon one or another of the aforementioned balanced
salt solutions, culture media for poikilotherm vertebrate cells are off-theshelf media that were originally designed for mammalian cell and tissue
culture. At this time lower vertebrate cells have been only maintained and
not truly grown, in chemically defined medium. If the reader requires a
chemically defined medium, it should be kept in mind that, while the cells
may not grow, their viability can be retained and they will metabolize for
at least several days in such medium.
In most situations, animal cell cultures are expected to divide and the
biological mass to increase. Literature on teleost fishes, amphibia, and
reptilia shows that investigators use one or another of four common media
for growing cells and that the media are usually supplemented with 10%
serum; in a few instances 20% serum is used. Without hesitation, fetal
bovine serum is recommended as the growth stimulant of choice. Fetal
bovine serum is rather expensive but, if cost is a factor and time permits,
comparative testing of agamma calf, horse, bovine, or porcine serum is
470
SPECIFIC CELL LINES
[40]
recommended since they m a y do as well. Probably because of innate

resistance to change and the generally excellent results with fetal bovine
serum, it is the single supplement that is recommended for culture of all
vertebrate cells.
While it is likely that most of the common commercially available
media will be satisfactory for poikilotherm vertebrate cell or tissue cultures, the most widely used are Eagle's minimal essential medium
(MEM), Eagle' s basal medium (BME), Medium 199, and Leibovitz L- 15.
The latter has been specifically designed to maintain a pH in the physiological range under conditions of normal atmosphere.
Contrary to what might appear to be logical, serum from the donor species more frequently than noLproves to be unsatisfactory for the usual cell
cultures. In addition, a sustained supply is usually difficult to obtain, and
quality control is less easily achieved than in commercial sources.
For most work with poikilotherm vertebrate cultures a 10% level of
serum is standard, but when initiating primary cultures or when vigor
appears to be low, an additional 5% serum or 5% whole egg ultrafiltrate
can sometimes be used to advantage. For reasons of economy, some
workers substitute part of the serum with 0.5% lactalbumin hydrolysate,
but the advantage is only marginal.
Leukocyte culture differs somewhat from other cell culture in that a
20% level of serum is usually employed, and some have found it advantageous to use homologous serum. This appears to be particularly true for
elasmobranch leukocyte culture; sharks, skates, and rays have little or no
serum albumin, and isotonicity with the marine environment is maintained
by retention of very high levels of urea.
pH Control
As a general guideline and in the absence of specific information to the
contrary, an initial pH of 7.3--7.4 is suggested for starting lower vertebrate
cell cultures. At this range, phenol red indicator is reddish orange. Until
one is familiar with the color changes of phenol red, a set of pH standards
will be useful. Some cells tolerate a pH well above or below the optimal
physiological level, but values much above or below will result in an
extension of the lag phase of growth.
Temperature of Incubation
On the premise that near-optimal growth is desired, the choice of
incubation temperature can be approached with the following rule of
thumb: use a temperature slightly above that preferred by the intact ani-
[40]
471
TABLE II
PROXIMATE GUIDELINES FOR INCUBATION TEMPERATURES FOR LOWER VERTEBRATE
CELL AND TISSUE CULTURES
Temperature range (C)

Vertebrate
class
Reptile
Amphibian
Teleost
Elasmobranch
Cyclostome
Nearoptimal
Environment
Low
High
Temperate
Tropical
Temperate
Tropical"
Temperate
Cold water
Warm water
Tropical
Temperate
Cold water
Warm water
Temperate
20
not known
15
not known
23-25
30
25-26
28
37
37
37
37
4
13
20
20
25
25
26
30-37
37
4
not known
4
10-15
not known
15
not known
not known
20
mal. Once the cultures are growing, experimentation can determine an

optimal or near-optimal temperature. The advantage to be found with
poikilotherms is that their cells can be chosen for growth and metabolism
in the range of near freezing through 37 (Table II).
Cold-water fishes such as salmon and trout prefer temperatures of
about 8-12 , but the optimum for salmonid cell cultures is about 20 .
Salmonid cells will grow, however, at 4, and their thermal death point is
about 26 or 27 . Interestingly, the growth rates near their maximum temperature tolerance are less than at 20 . Warm-water fish cells seldom grow
at 5-10; their optimum is generally in the range of 22-25 , and their upper
limit is about 30 or higher. With increments of about 1 at each subculture, warm-water fish cells may be adapted to grow at 37 .
Amphibian cells can be handled much like those of warm-water fishes
and grown at about 22-25 . They too can be kept at 10-15 to slow metabolic rates. Accordingly, they can be stored at 15 for weeks and in some
cases for months without attention. Since most plastic cell culture vessels
allow diffusion of CO2, long-term storage of active cultures is better done
in glassware.
Reptiles are designated as being terrestrial, aquatic (fresh water), or
marine; they are further classified as temperate climate or tropical animals. Reptilian cell cultures therefore should be grown in medium of
proper tonicity and also incubated at an appropriate temperature (Table
472
SPECIFIC CELL LINES
[40]
II). Cells from donors of a temperate climate are incubated at room temperature and those from tropical latitudes at about 30.
Leukocyte Culture
Poikilotherm vertebrate leukocytes lend themselves to the usual in
vitro applications: karyology, migration inhibition studies, investigations
of antibody production, and other immunologic research. Viable leukocytes are readily obtained from sterile whole blood drawn with either
plastic or siliconized glass syringes containing cold heparin solution in
physiological saline at a final concentration of 10 IU per milliliter of blood.
If one is unfamiliar with the anatomy of a particular animal, it will be
advantageous to dissect a specimen and locate the heart and major blood
vessels and to orient them with external anatomy. Direct cardiac puncture
with a needle of appropriate gauge works well with normally scaled reptiles. Easy access to the heart of terrapins and turtles, of the size conveniently used in laboratories, is gained by first drilling a small hole in the
plastron immediately beneath the heart. Amphibians can be bled from the
heart, and the frogs and toads also can be bled from the femoral artery.
For best results in culturing amphibian leukocytes, the donors should be
kept at normal temperature prior to bleeding; in that way the lymphocyte
population is maintained.
Fishes also can be bled from the heart, but if the animal is to remain
alive cardiac puncture does run the risk of tamponade formation. Fishes
may be bled by opening the mouth and using a shallow approach to enter
the dorsal aorta immediately behind the last gill arch. Fishes are perhaps
bled most safely, most easily, and repeatedly by lateral or ventral approach in the region behind the vent, the caudal peduncle. By the last
method, either the dorsal aorta or the caudal vein will be entered.
When drawn, blood is mixed with the heparin solution and centrifuged
in the cold at 200-500 g or until the "buffy coat" of leukocytes is clearly
visible above the packed erythrocytes. It is convenient to centrifuge the
blood in an inverted but locked syringe. Locks are easily made from split
plastic tubing that fits around the plunger; during centrifugation the tubing
prevents the barrel from moving down and expressing blood. After centrifugation, a needle is refitted, bent to a 90 angle, and the plasma or
"buffy coat", or both, are gently expressed into culture medium.
Obtaining Tissues
Either external or internal tissues can be cultured, but the latter have
the lower risk of contamination. Almost universally, ovarian tissues do
[40]
473
well in culture. Ovaries may be used at almost any stage of development

and typically provide a sizable mass of usable tissue. Ovaries are highly
recommended for beginning exercises in lower vertebrate cell and tissue
culture. Embryonic tissues also do well because they are typically clean
and usually grow luxuriantly; their only limitations are volume and the
fact that they are typically available only during a limited portion of the
year. Some viviparous laboratory fishes are an exception, and they produce young throughout the year. Other internal tissues that usually lend
themselves well to culture are heart, trachea, gas bladder, mesenteries,
spleen, kidneys, and lungs. However, the last two frequently harbor
metazoan parasites that are esthetically undersirable but which probably
will not grow in vertebrate culture systems. Liver from juvenile animals is
an additional candidate tissue but, because of the great mass of differentiated cells, liver can yield active cultures somewhat more slowly than
other tissues. Portions of the gastrointestinal tract also may be cultured,
but with them the risk of contamination exceeds that of external tissues;
thorough washing and decontamination for an hour or more in a mixture
of the following bactericidal antibiotics is suggested: 500 IU polymyxin B,
500/zg neomycin, and 40 IU of bacitracin per milliliter of sterile water.
Although available in only limited quantity, corneal tissue is excellent
for culture as are fish fins and amphibian skin. Because of its cornified
outer layers the skin of reptiles presents difficulties because of its inert
component and greater-than-normal risk of contamination.
Whether internal or external tissues are to be used, the donor animals
should be deprived of food for several days prior to use. This precaution
will minimize risks of gross contamination by food or feces during
handling.
Immobilization during surgical opening is desirable, if not essential.
Animals can be pithed or killed with a blow to the head; fishes and amphibians can be euthanized in 1 : 5000 solution of tricaine methanesulfonate
and amphibians by immersion in a plastic bag of the anesthetic. Reptiles
and elongate amphibians can have muscular contraction for considerable
time after death. To reduce problems of vigorous bodily contractions
during operations, the carcass can be refrigerated for an hour or more
prior to use. One need not fear tissue or cell death; sterile internal tissues
may be refrigerated in PBS for at least a day and in some instances as long
as 3 days and still yield cultivable cells. If mobility is a problem, the head,
limbs, and tail from the vent rearward also can be restrained with hooks or
stout pins. The plastron of terrapins and turtles must be removed to
expose internal organs.
Decontamination of the integument should be carded out if sterile
internal tissues are to be obtained from any lower vertebrate. Soaking for
474
SPECIFIC CELL LINES
[40]
5 min in a 1 : 10 dilution of household bleach or any hypochlorite solution

containing at least 500 ppm available chlorine is adequate. For reptiles,
thorough scrubbing with disinfectant is further recommended since they
often have external fauna and flora. Strong hypochlorite solutions denature protein and kill living cells; therefore these solutions should not be
used if external tissues of amphibia and fishes are to be cultured. Fins,
skin, tongue, and gills should be washed well with cold, preferably chlorinated, tap water and further decontaminated for an hour or more at 5-10
in a solution of the three bactericidal antibiotics listed earlier.
Strict aseptic technique should be used to remove internal tissues. A
tray of wax or soft wood is used to pin or otherwise immobilize the
animal. Additional pins hold back the abdominal flaps and the gastrointestinal tract so that there is unimpaired access to the visceral cavity and
organs. Fish, of course, are positioned laterally and also opened laterally.
Amphibia and reptiles are best fixed in the supine position and opened
midventrally.
As organs or tissues are removed, they should be transferred to a
sterile petri dish or other covered container maintained on ice. Materials
from lower vertebrates of temperate or tropical origin need not be chilled,
but holding in chilled vessels is not inhibitory. If the amount of tissue
removed is greater than can be used immediately, it may be kept safely at
4 . Protected from dehydration, the tissue can be expected to yield viable
cells for several days.
There is a great deal of evidence that the emphasis of lower vertebrate
cell and tissue culture has been on proliferation of undifferentiated or
dedifferentiated cells. However, if obtained aseptically as indicated in the
foregoing, differentiated tissues or cells, including hepatocytes and endrocrine tissues, among others, can be maintained for days or weeks and
remain capable of at least some of their functions.
Primary Cultures
Primary cultures can be of the explant type, or they can be monolayers
that are derived either from tissues that are simply minced and planted or
minced and enzymically dispersed before planting. Procedures for initiating primary cultures of poikilotherm cells are essentially the same as those
for homeotherms. For details, see this volume [9] and [10].
Monolayer Cultures from Minced Tissues

Monolayer cell cultures can be started simply and quickly from
minced tissues that are washed, drained, and transferred to culture vessels. Bent-tip tissue culture type Pasteur pipettes having about a 2-mm
[40]
475
orifice or spatulas are convenient for planting the tissues. Fragments are
uniformly distributed o v e r the growth surface at a rate of two to four
pieces per cm 2. The culture vessels are then closed and placed on edge for
an hour or so at a favorable incubation temperature. After fluids drain
from the tissue and natural adhesion occurs, the liquid is aspirated, the
vessels inverted, medium is added, and the cultures are m o v e d to an
incubator. The vessels are then oriented so that the medium covers the
tissue. Most fragments adhere and many will send out cellular processes
or extend areas o f epithelial-like cell sheet within a day or two. This initial
growth helps the fragments adhere more tightly. After the first growth
occurs the cultures can be examined safely microscopically. Confluent
sheets o f cells may be formed within a few days, but it may take as long as
2 weeks; it depends on donor age, the organ used, pH, temperature, and
other variables. When confluency is attained and growth is obviously
heavy, the cultures may be divided. Requiring only 2 or 3 hr, this method
is undoubtedly the simplest and least time-consuming means o f setting up
primary monolayer c u l t u r e s ) The cultures thus developed are typically
less homogeneous than monolayers that have been initiated from trypsinized tissues.
Monolayer Culturesfrom Trypsinized Tissues

The most commonly used procedure for starting primary monolayer
cultures o f vertebrate cells is that of " t r y p s i n i z a t i o n . " Details of the
p r o c e d u r e are described in this volume [10], and the general methods are
applicable to lower vertebrate as well as to h o m e o t h e r m tissues. The
following are important comments and exceptions as they apply to reptile,
amphibian, and fish tissue trypsinization. 11
No lower vertebrate tissues should be trypsinized at 37 because the
cells will be killed. Instead, there are two alternative lower temperatures
that can be used. Tissues from poikilotherm vertebrates of warm environments can be trypsinized at 20-25 with harvest being made at intervals o f 15-30 min. Tissues from cold-water fishes may be trypsinized at
15 or lower. Regardless of the environmental temperature preferred by
the animal itself, tissues from all of the lower vertebrates can be trypsinized successfully at 5 . Duration of enzymic treatment will depend
upon the tissue; embryonic tissues may disperse in several hours but
digestion o f from 12-16 hr is more common. The choice of room tempera10For additional reading and a listing by source and catalogue numbers of specific materials
needed to carry out the foregoing methods, see K. Wolfand M. C. Quimby, TCA Manual
2, 445 (1976).
11 Additional readings and a listing of specific materials by source and catalogue number
are to be found in K. Wolf and M. C. Quimby, TCA Manual 2, 453 (1976).
476
SPECIFIC CELL LINES
[40]
ture or cold overnight digestion may be dictated by the work situation, or

one can compare the two temperatures for dispersing tissues. If that is the
case, ciliated cells that occur in various tissues may be used to monitor
the digestion and subsequent culture conditions for at least several days.
Under favorable trypsinization and culture regimes, cilia will continue to
beat actively and provide instant assessment of viability.
Penicillin and streptomycin, a bacteriostatic combination, are almost
traditional in animal cell and tissue culture, and some workers include
them in the digestion mixture. Gentamicin is a bacteriocidal antibiotic and
much more effective in controlling the gram-negative organisms that are
common in lower vertebrates. Levels of 10-100/~g/ml are suggested.
Trypsinization of homeotherm tissues commonly is allowed to proceed
until a monodispersed suspension of cells results. That degree of dispersion can be excessive for some lower vertebrate tissues; the resulting
suspension may contain too many dead cells. Viability is usually excellent
if digestion is allowed to proceed only to the point at which there is a
mixture of individual cells and tissue fragments consisting of 10-100 or so
cells. That kind of determination may be made with 0.5% aqueous trypan
blue; live cells exclude the dye.
Residual trypsin can be inhibitory to cells that are obtained by digestion. Accordingly, the harvest should be washed in PBS or in medium
before planting.
Because of the presence of small fragments of nondispersed tissue in
the suspension of lower vertebrates after the optimal digestion period, cell
counting is pointless. Instead, the harvest is resuspended in growth medium on the basis of its packed volume. For embryonic or soft tissues, the
ratio of cells to medium can be 1 : 2000 to 1 : 4000 or more. For initial
trials, especially when adult tissues are used, a ratio of 1 : 1000 or less is
suggested; prompt success is much preferred to slow results or to failure
that can follow if too few cells are seeded.
While vortex mixers are common in most laboratories, they should
not be used to suspend living animal cells; the vigorous mixing will likely
kill many cells. Cells should be resuspended by repeated gentle pipetting.
Subculturing or Transferring Cell Lines

Coldblooded and homeotherm vertebrate cell cultures are transferred
or divided in essentially the same way. There are advantages, however, in
working with lower vertebrate cells since it is seldom, if, ever, necessary
to feed cultures between transfers, i.e., to provide them with fresh medium. When not used, cell cultures from the poikilotherms retain viability
for months without attention; they typically have a high tolerance for
[40]
477
neglect. Virtually all vertebrate cell cultures can be divided or transferred

at room temperature, but some workers prefer to handle mammalian and
avian cells at 37, a temperature probably excessive for many
poikilotherm cells.
The most widely used method of dispersing lower vertebrate cells for
subculturing is with a solution of EDTA (200/xg/ml) and trypsin (1-2.5
mg/ml). A listing of equipment, solutions, and procedures is available. 12
Cultures first should be examined microscopically for quality, cell morphology, and possible contamination. The medium is then removed and
the EDTA-trypsin solution added at about 5 ml/75 cm 2 of growth area or 2
ml/25 cm 2. After several minutes the cell sheet develops marked opacity.
The EDTA-trypsin solution is removed and replaced with about half as
much fresh dispersant. Cell lines vary considerably in their response.
Many will have been freed after several minutes, but some may require
gentle scraping. Total treatment time should not exceed 10--12 min, after
which fresh medium should be added to stop the action of trypsin. The
cells are uniformly dispersed by pipetting, diluted as necessary in fresh
medium, and planted. Although most lower vertebrate cell lines can be
routinely subcultured with this procedure, some are usually sensitive to
trypsin. If cells lyse or otherwise fail to prosper, they should be centrifuged out of the EDTA-trypsin solution, washed in fresh medium or
BSS, recentrifuged, and resuspended in fresh medium for planting. Cells
grown in the absence of serum or other protein are particularly vulnerable
to trypsin and will probably require greatly reduced trypsin or even alternative methods that omit trypsin. Inexperienced investigators are
cautioned to use low split ratios for their early trials of subculturing.
At times, cell growth may be more rapid than anticipated or transfers
may be unavoidably delayed, so that heavy or dense cell sheets result. In
such cases, the usual 10-12 min treatment time may not be sutficient to
disperse the cells. The incompletely dispersed population can be resuspended, planted, and allowed to grow for several days. During that time,
cells will migrate out of the clumps, and a second EDTA-trypsin treatment can be used to achieve complete dispersion.
In the absence of specific information on density of cells to be cultured, i.e., number per milliliter, it is suggested that the initial density
should be at least several hundred thousand cells or more per milliliter. It
is usually safe to seed at a high population density although cultures will
become confluent quickly and require frequent handling. If cells are
seeded too thinly, the lag time will be lengthy or the population too sparse
to establish itself, so that the culture may decline and die. Adverse affects
are particularly prevalent in media of marginal quality.
12 K. Wolf and M. C. Quimby, Tissue Cult. Assoc. Man. 2, 471 (1976).
478
SPECIFIC CELL LINES
[41]
[41] P l a n t C e l l L i n e s 1
By JOHN F. REYNOLDS and TOSHIO MURASHIGE

As experimental materials, aseptically cultured cells or tissues have
certain advantages over organs and plants that are allowed to develop in
the more natural state. Correlative influences can be minimized, and artifacts attributable to bacteria, fungi, and other small organisms can be
avoided. It also is possible to regulate more precisely the provisions with
respect to nutrients, light, and temperature. Sometimes cultured cells can
be treated and interpreted experimentally like simple microorganisms.
Plant cell cultures may have economic significance as well. When
totipotentiality is manipulatable the cultures can serve as intermediary in
rapid clonal multiplication of plants 1 and in elimination of viruses and
related pathogens. 2 Some are potential sources of medicinals and other
important plant constituents. 3
In the simplest and perhaps most widely practiced method of maintaining plant cell lines the callus, or the unorganized tissue that results on
wounding, is cultured. For biochemical studies, however, liquid suspension cultures of free-living cells are probably preferred, even though they
demand more labor. Indefinitely subculturable callus or cell suspensions
are now attainable with virtually any plant, bryophytes through angiosperms, and from nearly every plant organ, including stem, root, leaf, fruit,
flower, and seed. Callus is easily initiated by simply placing freshly cut
sections of disinfested organs on the surface of an agar-gelled medium.
The nutrient medium of plant callus and cell cultures should contain a
balanced salt mixture, sucrose (3-5%) as carbon source, 4 thiamine HC1
(0.1-10 mg/liter), and usually the growth regulators, auxin and cytokinin.
These substances should be tested for their effectiveness as auxin in the
concentration range 0.1-10 mg/liter: 3-indoleacetic acid (IAA),
1-naphthaleneacetic acid (NAA), and 2,4-dichlorophenoxyacetic acid
(2,4-D). Kinetin, Ne-benzyladenine (BA), and Ne-isopentenyladenine (2iP)
are the more readily available cytokinins; they should be examined in the
range 0.03-3 mg/liter. Myo-inositol in a concentration of 100 mg/liter and
other supplements, e.g., citric acid (2 g/liter) 5 and casein hydrolysate (1-3
g/liter), also may be helpful.
1 T. Murashige, Annu. Rev. Plant Physiol. 25, 135 (1974).
2 T. Murashige, A. W. Dimock Lect., Cornell Univ. No. 2 (1974).
3 W. Barz, E. Reinhard, and M. H. Zenk, eds., "Plant Tissue Culture and Its Biotechnological Application." Springer-Verlag, Berlin and N e w York, 1977.
4 Glucose is used occasionally instead of sucrose.
5 y. Erner, O. Reuveni, and E. Goldschmidt, Plant Physiol. 56, 279 (1975).

All rights of reproductionin any formreserved.
ISBN 0-12-181958-2
[41]
PLANT CELL LINES
479
Most callus cultures grow best in darkness. However, free-living cells

in liquid suspension may benefit from low levels of illumination, perhaps
1000 Ix, provided daily by cool-white fluorescent lamps. The standard
practice with callus and liquid suspension cultures has been to incubate at
a constant temperature of about 27.
Establishing the Callus Culture

Cell and tissue cultures of the tobacco, Nicotiana tabacum L. "Wisconsin 38," have been chosen to illustrate all principles because details of
their requirements are established, the pattern of their development in
vitro is readily manipulated, and totipotentiality can be demonstrated.
Plants should be grown under greenhouse conditions to ensure a high
percentage of clean and responsive explants and used before flowering
begins. The better explants are obtained from the upper or younger regions of the stem. Cell cultures originating from older tissues often manifest retarded growth and organogenic rates as well as high frequencies of
polyploid cells.
Sever the plant at its base and remove all leaves. Trim stem of all leaf
bases and axillary buds, using a surgeon's scalpel fitted with No. 10 blade.
Swab stem with cheesecloth, dampened with 95% ethanol, to remove
some of the stickiness and to achieve preliminary disinfestation. Obtain
1-cm-long stem segments from the region 10-30 cm from the shoot apex
region. Plance segments in a small Erlenmyer flask and add disinfestant
solution. 6 Cover flask loosely with beaker and place under gentle vacuum
for 10 min. Decant disinfestant and rinse stem segments 3 times with
distilled water that has been previously autoclaved and cooled. Working
within a laminar air-flow hood or other protective device, transfer stem
segments to sterile Petri dish. Remove the bark from each of the segments. This is accomplished by grasping the segment at its cut ends with a
pair of forceps, making a shallow incision longitudinally with a surgeon's
scalpel fitted with No. 10 blade, and paring away the bark with the dull
edge of the scalpel. Stand the peeled stem sections on end and divide into
three or four tangential slabs. Transfer slabs to nutrient agar and plant
with the cambium side exposed. If desired, the pith portion that remains
also can be cut into explants. It should be prepared as cubes about 2 mm
thick.
Presterilized, disposable Petri dishes are convenient. In their absence,
the Pyrex type can be used, after autoclaving 30 min at 121 or dry-heating
Laundry bleach diluted to contain 0.5% sodium hypochlorite is used most widely; a few
drops of detergent might be added to enhance wetting.
480
SPECIFIC CELL LINES
[41]
in an oven for 4 hr at 160. Several dishes can be wrapped together in

aluminum foil or placed in a prescribed cannister for sterilization.
Satisfactory callus from the above explants can be expected by employing a medium of the following composition, in mg/l: Murashige and
Skoog inorganic salts; r sucrose, 30,000; myo-inositol, 100; thiamine HC1,
0.4; kinetin, 0.1; 2,4-D, 0.3; and Phytagar, 8 6000. Adjust pH of the solution to 5.7 with I N HC1 or NaOH prior to addition of agar. Wash agar
with distilled water 3 times before use and dissolve by heating. If a
hotplate or burner is employed, stir medium constantly to avoid charring
of agar. A simpler way. to dissolve agar is by autoclaving the medium at
121 for 3-7 min, depending on the quantity of medium. Mix medium well
and dispense 25-ml portions into 25 150 mm glass culture tubes. Cap
tubes with polypropylene closures, a Autoclave 15 rain at 121 to sterilize
medium, then cool tubes at 45 slants. Inoculate each tube with one
tobacco stem explant, and incubate in darkness at 27. Within 2 weeks,
callus will appear as a white growth on the exposed surface of the explant.
After 4 weeks it should have developed sufficiently for subculturing and
establishing of callus stock.
In subculturing the callus, prepare nutrient tubes as above. Select
cultures that show large growths of white callus. Examine tubes carefully
and discard those that show bacteria, fungi, or other microbial contaminants. Contaminated cultures should be autoclaved before disposal to
minimize spread of fungi and bacteria in the laboratory. Using a pair of
long forceps, 1 transfer callused section to sterile Petri dish. Pare away a
2-mm-thick slice of callus from the stem, using a surgeon's scalpel fitted
with No. 10 blade. It may be helpful to use a pair of small forceps to hold
the stem section during the cutting process. Divide the callus slice into
3-5 mm squares and transfer to nutrient tubes, at a rate one tissue section
per tube. The callus is maintained as stock by subdividing established
cultures at 4-week intervals and transferring 2 4 5 mm subsections to
freshly prepared medium. Incubation of established callus is also carried
out in constant darkness and at 27 .
z T. Murashige and F. Skoog, Physiol. Plant. 15, 473 (1962). The salts, in mg/1 are:
NH4NO3, 1650; KNOz, 1900; CaClz. 2 H20, 440; MgSO4.7 HzO, 370; KH2PO4, 170;
Na2EDTA, 37.3; FeSO4 7 H20, 27.8; HzBO3, 6.2; MnSO, H20, 16.9; ZnSO4 7 H20,
8.6; KI, 0.83; Na2MoO4 2 H20, 0.25; CuSO4 ' 5 H~O, 0.025; and CoC12 6 HzO, 0.025.
8 This is agar specially prepared for plant tissue culture by Grand Island Biological Co.,
Grand Island, New York.
9 Examples of plastic caps available are Kimble Kimkap, Bellco Kaput, and Bacti Capalls.
10 Use 9- inch dressing forceps, Model MX6-158, obtainable through Miltex Instrument
Co., New York, New York.
[41]
PLANT CELL LINES
481
Cells in Liquid Suspension Culture

Whereas it is possible to obtain free-living cells in liquid media by
starting directly with a freshly excised plant section, the easier method is
to begin with a suitably prepared callus as inoculum. Cell dissociation in
liquid culture can be improved if the callus is first rendered friable. The
friable texture can be attained sometimes by culturing the callus on an
agar medium that is enriched in salts and auxin. Gibberellin A3 (GAs) may
further enhance the friability. Cytokinin, if necessary, should be included
at only very low levels. Some natural complexes, e.g., protein hydrolysates and yeast extract, may also increase friability. The ease of altering
tissue texture may be an inherent quality, and occasionally cell dissociation may not be possible under any condition. These relationships between nutrient addenda and callus friabihty are perhaps applicable to
freshly excised tissues and might be considered when callus is not
available.
To obtain readily dissociating tobacco callus, use a medium whose
composition is the same as that employed in establishing the tobacco
callus cultures, but with the 2,4-D increased to 2 mg/liter and with an
additional supplement of 1 g/liter casein hydrolysate; u the vitamin component might be altered by raising thiamine HC1 to 10 mg/liter and by
adding 5 mg/liter nicotinic acid and 10 mg/liter pyridoxine HCI. Adjust
the medium pH to 5.7, dissolve the agar in the medium by heating, and
dispense medium into 25 150 mm culture tubes with 25 ml per tube.
Use polypropylene closures and autoclave the medium to achieve sterility. Obtain callus that has been grown for 4 weeks in a given subculture
passage of an established stock. Transfer the tissue to a sterile Petri dish
and cut into 2-mm-thick slices; then cut each slice further into 4- or 5-mm
squares. Inoculate nutrient tubes with callus slices and incubate in darkness at constant 27 .
After 4 weeks a translucent, soft, and relatively watery tissue should
result. The callus will have lost its earlier firmness and white appearance.
A substantial proportion of the cells will be easily separable from one
another by gentle teasing; the cells will be substantially larger than those
of the firm, white callus. Transfer the entire new growth to liquid nutrient.
A micro-spatula may be better than forceps for handling the tissue.
The composition of the liquid medium can be the same as that employed in rendering the callus friable, except for the omission of agar. Use
125-ml Delong flasks, each with 25 ml of nutrient solution. Cap flasks with
Morton stainless-steel closures and sterilize by autoclaving at 121 for 15
min.
1~ Enzymic digest of casein is obtainable from ICN Life Sciences Group, Cleveland, Ohio.
482
SPECIFIC CELL LINES
[41]
Place the inoculated flasks on shaker t2 and agitate constantly at 150

rpm. Provide 1000 lx illumination, from cool-white fluorescent lamps, 16
hr daily and at a constant 27 temperature.
After 10-14 days a dense suspension composed of free cells and cell
aggregates should have developed. Separate free-living cells and small
aggregates from larger aggregates and subculture them in liquid medium
to establish a suspension culture stock. To achieve separation, first, drain
the nutrient solution by filtering the suspension through two layers of
sterile Kimwipe tissue; use a 70-mm-diameter powder funnel to support
the Kimwipe tissue. The funnel may be held in a small beaker or flask.
Sterilization of Kimwipe tissue is accomplished easily by folding and
enclosing several sheets in Petri dishes and autoclaving. Complete drainage of the nutrient solution by gathering together the corners of the Kimwipe tissue and gently wringing the contents. Carefully unfurl the Kimwipe tissue and weigh out l-g samples of cells for subculture. Avoid
including the larger cell clusters. Weigh cells in a sterilized Petri dish.
The medium for subcultures is the same in composition as that used in
initiating the suspension culture. A stock of vigorously dividing and growing cells can be attained by repeatedly subculturing in fresh medium 1-g
portions at 10- to 14-day intervals. Growth of plant cells in liquid suspension cultures can be measured as drained weighL dried weight, 13
sedimented volume of cells after gentle centrifugation,14or as cell counts.
Accuracy of cell counting may be improved if the mixture of cells and
aggregates is first dispersed by treating drained samples with 5% chromic
acid or 0.25% pectinase 1~for 24 hr at room temperature. Protein content
might also be used to estimate plant cell culture growth) 6
Plating to Obtain Strains of Singie-Cell Origin
Tissue strains of single-cell origin are obtainable in large numbers by
allowing the free-living cells of liquid suspension cultures to regenerate
callus in agar plates. The procedure enables establishment of pure lines
from an otherwise heterogeneous population of cells. Heterogeneity
among cultured cells may be attributable to natural and induced variations. 17 Since suspension cultures are usually comprised of a mixture of
12 Satisfactory results can be obtained with a Model G-10 gyratory shaker, made by New
Brunswick Scientific Co., Edison, New Jersey.
la Dry in 70 oven until constant weight is attained.
14 2880 g for 5 min.
t~ Sigma pectinase, for example.
t 6 j..p. Jouanneau and C. Peaud-Lenoel, Physiol. Plant. 20, 834 (1967).
1~ T. Murashige and R. Nakano, Am. J. Bot. 54, 963 (1967).
[41]
PLANT CELL LINES
483
free-living cells and cell aggregates of varying dimensions, it is important

to fractionate the cultures to obtain a component of predominantly freeliving cells. In practice, even the best free-cell preparations will be contaminated by a significant fraction of small aggregates. Thus, to ensure a
single-cell origin of developing tissues, each newly prepared plate should
be examined under a microscope and the single-cell units located and
circled with indelible ink. The tissue arising within the circled area is more
likely to be of single-cell derivation. Further assurance of single-cell origin
is possible by plating protoplasts instead of intact cells. Since protoplasts
have no cell wall, separation into individuals is virtually guaranteed.
The nutrient medium for plating of tobacco cells is the same in composition as the agar medium used in obtaining friable callus. The cells to
be plated are mixed with the nutrient agar while the agar is still fluid, but
sufficiently cooled to be noninjurious. In actual practice, prepare the complete nutrient medium, but dilute to only 90% of its ultimately desired
volume. After pH adjustment and agar dissolution, dispense medium in
9-ml aliquots into 25 150 mm culture tubes. Cap tubes with polypropylene closures and autoclave 15 min at 121. Allow the agar to cool to
about 50" and place nutrient tubes in a water bath adjusted to 40. This
prevents premature gelling, but enables maintenance of a favorable temperature. The nutrient tubes should be held in a suitable rack and emersed
so that the water in the bath is level with that of the medium in the tubes.
Only clean distilled water should be used in the bath. Obtain desired cells
by filtering a 10-day liquid suspension culture through a layer of sterile
cheesecloth. Collect cells that pass into the filtrate by draining the filtrate
through sterile Kimwipe tissue. After gentle wringing, weigh out 1-g quantities and resuspend in 9 ml of sterile liquid nutrient, the composition of
which is the same as that used for plating, but without the agar. Transfer
1-ml portions of the suspension to tubes containing the agar nutrient. Mix
cells and nutrient agar gently, but thoroughly, and pour contents quickly
into sterile 100-mm petri dishes. Spread cells uniformly in the dish. As
soon as the medium gels, seal the petri dish with Parafilm. The seal minimizes evaporation of medium and contamination by airborne organisms.
Observe plates with an inverted microscope and circle single-cell units
with a fine-tipped felt pen. Place cultures in darkness and at constant 27.
Within 2-4 weeks, callus growths of varying sizes, sometimes also diversely textured and pigmented, will become apparent. Remove growths
that have attained 2-3 mm in diameter and reculture or subculture, following the procedure described earlier for callus, and establish callus stocks
of single-cell origin.
Protoplasts for plating are best obtained from tobacco cells that are in
the early exponential phase of cell division in liquid suspension culture,
484
SPECIFIC CELL LINES
[41]
i.e., after 4-5 days of a given passage. ~s Collect cells on sterile Kimwipe
tissue. Transfer 0.5-g portions to 50-ml Delong flasks, each containing 5 ml
of protoplast-release solution.19 Place the suspension on a gyratory shaker
and agitate continuously for 2-3 hr at 50-70 rpm. Filter through nylon
cloth of 50-60/~m. Centrifuge the filtrate at 100 g for 2 min. Decant and
resuspend in 0.7 M mannitol and recentrifuge. Repeat the process of
washing 3 times, by suspending in mannitol and centrifuging. Finally,
suspend the rinsed protoplasts in 2 ml of nutrient medium that has been
prepared earlier and held in a 40 water bath. This medium should contain, in mg/l: Murashige and Skoog salts; sucrose, 15,000; mannitol,
110,000; NAA, 0.6; kinetin, 0.1; thiamine. HCI, 10; pyridoxine. HC1,
10; nicotinic acid, 5; myo-inositol, 100; glycine, 2; and prewashed
Phytagar, 8000. Mix protoplasts and medium gently, but thoroughly, and
pour into 40-mm petri dish by following a procedure similar to cell plating.
Incubate plated protoplasts as done with cells. Cell-wall regeneration will
occur within several hours, and cell division will be observable within a
few days. Transferable cultures will be obtained in 2-4 weeks.
Comments
Some serious misconceptions regarding cultured plant cells need correction. Cultured plant cells are not undifferentiated as often claimed.
They are larger, more vacuolated, lower in cytoplasm content, thicker
walled, and smaller of nuclei than those of the embryo or apical meristem,
where the more truly undifferentiated cells can be found. Very often
phloem and xylem elements may be apparent among them. The cells in
culture may lack organization and are therefore identifiable as being unorganized.
Uniformity also is not the usual trait among cells of callus or liquid
suspension. Obvious variations include shape and size, cytoplasm and
organelle contents, wall thickness, and ergastic deposits. More important,
the cells within a culture may be genetically diverse. For example, variations in chromosome number have not been uncommon. 2 Subculturing,
furthermore, has tended to accentuate the incidence of variants. Retention over long periods of a cell line's characteristics will require employment of cryogenic proceduresY 1
is H. Uchimiya and T. Murashige, Plant Physiol. 54, 936 (1974).
19 Solution containing 1% cellulysin, 0.2% macerase, and 0.7 M mannitol. The enzyme
preparations are obtainable through Calbiochem, La Jolla, California. The pH of the
solution is set at 5-7, and the solution should be filter-sterilized.
20 F. D'Amato, Int. Biol. Programme 2, 333 (1975).
21 y. p. S. Bajaj, Physiol. Plant. 37, 263 (1976).
[41]
PLANT CELL LINES
485
It is hazardous to assume that the behavior of cultured cells invariably

reflects that of their progenitor plant. Frequently, the progenitor' s characteristics are dependent on its organized state. For example, tobacco roots
have been observed to carry on high rates of synthesis of the alkaloids,
nicotine and anabasine, but cultured cells do not. 22 Epigenetic modifications also occur frequently and contribute to aberrant behavior of cultured
plant cells and plants. 2~
When inoculating freshly prepared liquid media of suspension cultures
it is important that cell density requirements are satisfied. Cell division
will not occur below critical concentrations. For tobacco cultures, an
initial count of 60,000 cells/ml might be satisfactory. If it is necessary to
use cell concentrations below critical levels, nurse cells might be employed. The nurse cells first must have been rendered incapable of proliferation by irradiation or other means. Excessive cell numbers in the inoculum are also undesirable. They necessitate more frequent subculturing
of liquid suspension cultures and present difficulty in distinguishing individual callus growth that arises in plate cultures.
A variety of nutrient compositions can be used successfully with plant
cell cultures. But maximum advantage is attained only by assessing
specific nutritional needs systematically. A first step might be the comparison of the more popular basal salt formulations. Subsequent tests may
include vitamins, amino acids, carbohydrates, hormonal substances, and
natural complexes. Data regarding salt formulations, vitamin and amino
acid mixtures, hormones, and other nutrient constituents have been compiled recently. 24 Agar is still the best gelling agent, but its quality varies
according to commercial source, grade, and lot. Regardless of the information contained on the product label, it is good practice to wash the agar
before adding it to the nutrient medium. Often, rinsing 3 times with distilled water may be sufficient. This excludes much of the readily dissolved
and undesirable substances.
Agitation of liquid suspension cultures can be accomplished by various devices. Reciprocating shakers and rotators may be used instead of
gyratory shakers, and agitation rates can be varied from less than 1 to
over 400 rpm. Successful liquid suspension cultures have also been
achieved without agitation, by continuously bubbling air through the nutrient solution.
The question of pH of culture media has often been raised. Unfortunately, there have been few systematic investigations directed at the reso2z M. L. Solt, R. F. Dawson, and D. R. Christman, Plant Physiol. 35, 887 (1960).
23 F. Meins and A. Binns, Proc. Natl. Acad. Sci. U.S.A. 74, 2925 (1977).
24 L. C. Huang and T. Murashige, Tissue Cult. Assoc. Man. 3, 539 (1976).
486
SPECIFIC CELL LINES
[42]
lution of this problem. The common practice has been simply to set a
nutrient medium's starting pH at between 5 and 6, assuming that this is
within optimum range, and maintaining this pH range during the course of
culture.
Acknowledgments
This work was supported in part by the Elvenia J. Slosson Fellowship in Ornamental
Horticulture and NSF Grant OIP 75-10390 awarded to T. M. We thank S. Hamman and
S. Kearns-Sharp for typing the manuscript.
[42] L y m p h o c y t e s as R e s t i n g Cells
By SHELBY L. BERGER
The human peripheral blood lymphocyte is an ideal system for the
study of both resting and growing cell populations. As they are isolated
from the blood, lymphocytes are quiescent. DNA synthesis measured in
such cells reflects mainly ongoing repair processes since less than one cell
in 5000 is engaged in mitosis.l In v&o, the cells are capable of surviving in
the nondividing state for months or years. ~ In culture, resting lymphocytes can be maintained in excellent condition for several days after
which they deteriorate and die. It should be emphasized that these are
physiologically nongrowing cells; despite the presence of autologous
plasma and a complete medium such as Eagle's minimal essential medium
(MEM) or RPMI-1640, the cells remain in Go, outside the mitotic cycle.
Since most of our knowledge of cellular processes derives either from
continuously dividing cells adapted for cell culture, from density-inhibited
cells, or from arrested cells deprived of essential nutrients, it is essential
to expand our understanding by including the great bulk of normal cells
which resemble none of these examples. The lymphocyte provides such
an opportunity.
When confronted with a stimulus, the resting lymphocyte undergoes
changes in virtually every aspect of cellular metabolism culminating in
DNA synthesis and mitosis. In vivo, the challenge is an immunological
one, and the responding cells are those having specific mechanisms for
antigenic recognition. In vitro, growth induction of a large number of cells
Reviewed by H. L. Cooper, in "Drugs and the Cell Cycle" (A. M. Zimmerman, G. M.
Padilla, and I. L. Cameron, eds.), pp. 137-194. Academic Press, New York, 1973.
2 N. B. Everett and R. W. Tyler, in "Formation and Destruction of Blood Cells"
(T. Greenwalt and G. Jamieson, eds.), pp. 264-283. Lippincott, Philadelphia, Pennsylvania, 1970.

ISBN 0-12-181958-2
[42]
LYMPHOCYTES AS RESTING CELLS
487
can be elicited by the addition of any one of a number of nonspecific

mitogenic materials such as phytohemagglutinin or concanavalin A to the
culture medium. Although the latter compounds do not depend on any
known prior sensitization of the cells, the mitogenic response in vitro is
believed to resemble that in vivo in all other particulars. 1
Lymphocytes are also activated when cells from more than one donor
are mixed, a'4 This mixed lymphocyte reaction is of interest to immunologists but presents an impediment to cellular biochemists desiring
large numbers of resting cells for metabolic studies. Moreover, metabolic
activity differs in newly isolated cells from different donors so that, in
practice, an experiment and its control must be performed with lymphocytes from the same unit of blood. Therefore, optimizing the cell yield
from a single donor is essential.
Preparation of Lymphocytes
The method of purification presented here is a modification of
Cooper's procedure5 incorporating improvenients and simplifications developed during the past few years. Changes in the catalog numbers of
essential apparatus also have been included for the convenience of those
desiring to reproduce this scheme exactly.
Preliminary Isolation of Leukocytes

The initial collection of the blood and the preliminary separation of
lymphocytes from erythrocytes require equipment available from Fenwall
Laboratories. 6 Items and their catalog numbers are as follows: "BloodPack," 500 ml, containing heparin for the collection of whole blood
(4R0601); "Transfer Pack Unit," 150 ml with coupler (4R2001); "Plasma
Transfer Set" with 2 couplers (4C2243); plasma extractor (4R4414). For
centrifuging large buckets capable of holding the 500-ml Blood-Pack, a
low-speed centrifuge, such as the PR-6000 (International) fitted with a No.
276 or No. 981 rotor, or a RC-3 (Sorvall) with an HG-4L rotor, is also
necessary.
Blood from a fasting donor is collected in the 4R0601 pack and after
thorough mixing is centrifuged at 1500 rpm (approximately 400 gr) for 5
a B. Bain, M. R. Vas, and L. Lowenstein, Blood 23, 108-116 (1964).
4 B. Bain and L. Lowenstein, Science 145, 1315-1316 (1964).
5 H. L. Cooper, this series, Vol. 32, [62].
Fenwall Laboratories, One Baxter Parkway, Deerfield, Illinois.
r The g values are rounded to one significant figure and refer to the force at the center of the
centrifuge tube.
488
SPECIFIC CELL LINES
[42]
rain at room temperature. On removing the blood pack from the bucket, a
dense layer of erythrocytes can be seen with overlying plasma. The lymphocytes are found as a diffuse band, suspended in the lower region of the
plasma layer. Without disturbing the bands, the unit is carefully hung in
an open plasma extractor by the holes provided in the blood pack and
connected to the 4R2001 Transfer Pack by plugging the attached coupler
into the center port of the blood pack. The blood pack is also connected
by the side port to a receiving vessel, usually a sterile 32-ounce prescription bottle with a foil cover, by means of the 4C2243 Transfer Set. The foil
can be crimped around the coupler to hold it in plfice near the rim of the
bottle. Sterile technique is used throughout, and all steps are performed at
room temperature unless noted otherwise. With the tubing of the transfer
set closed, 50 g of plasma are expressed from the unit into the transfer
pack by releasing the plasma extractor. Then, with the transfer pack
tubing squeezed shut, the remaining plasma and the region of the interphase are expressed into the bottle. To insure complete recovery of the
lymphocyte-rich layer, several milliliters of red cells are expressed into
the bottle as well. Afterwards, 90-100 g of the remaining red cell layer in
the blood pack are added to the transfer pack. As a result of this procedure, most of the lymphocytes are confined to the prescription bottle
while the remainder, which were either trapped in the red cell layer or
suspended in the plasma, are in the transfer pack. Both the blood pack
and the transfer pack are mixed well and centrifuged again under the same
conditions to reextract the plasma-erythrocyte interphase. This step
necessitates removing the transfer set coupler from the receiving bottle,
but not from the side port of the blood pack, and capping it with the
original covering to maintain sterility. Other tubing connections are
clamped but left in place, and the tubing itself is tucked between the two
packs in the bucket during sedimentation. After the centrifugation step the
aim is to combine the lymphocytes in a single container. If plasma is
visible in the blood pack when subsequently squeezed by the plasma
extractor, it is collected into the receiving bottle. The plasma layer and a
few milliliters of erythrocytes from the transfer pack are also expressed
into the prescription bottle. As a result, virtually all the lymphocytes, in
partially purified form, are recovered. To insure against clotting, 5000 U
of heparin (1 ml of Liquaemin Sodium "50 "8) are added at this time.
Many investigators preparing lymphocytes have observed the formation of a "buffy coat" in the discarded blood pack after several hours of
standing. When examined microscopically, this band was found to be rich
in granylocytes and almost devoid of lymphocytes. Thus the preliminary
8 Available from Organon Inc., West Orange, New Jersey.
[42]
489
separation described above removes some, but not all, of the contaminating granulocytic leukocytes.
Nylon Column Filtration

Separation of lymphocytes from polymorphonuclear leukocytes is accomplished by adsorbing the granulocytes on nylon fibers. The nylon
fiber, 3 denier, 1.5 inch, type 200 nylon staple, is obtained from Dupont.
The purification is carried out in a waterjacketed column, 60 cm in length
by 2.5 cm ID, maintained at 37. In order to retain sterility of the fractionated blood, it is helpful if the bottom of the column is made to fit the
needle adapter of a No. 4C2240 Plasma Transfer Set with coupler and
needle adapter (Fenwall). The column effluent can then be conducted to a
32-ounce sterile prescription bottle with the sterile tubing; the coupler end
can be crimped in place with foil as before.
Nylon, as obtained from Dupont, is compressed and clumped, and
contains a fabric finish that must be stripped. The fiber mass is first teased
apart by hand into a fluffy puff of single strands. A column is packed with
3.5 g of the nylon so that the height is approximately 13 cm. This is best
accomplished by rinsing the inside of the column with water to wet the
surface and pushing the nylon to the bottom with a stick. The fabric finish
then can be removed by soaking the teased nylon in situ for 1 hr in 1 N HC1
and washing it for at least 3 hr with running distilled water. Excess water
is suctioned out of the nylon, and the column, with its ends covered with
foil, is autoclaved for 20 min. Prolonged drying is not recommended because the nylon is destroyed by dry heat. If the column is to be used
immediately, it is cooled and made ready by equilibration at 37 , connection to the receiving vessel with the aforementioned transfer set, and
application of approximately 100 ml of Eagle' s MEM to wash out condensate. Traces of the wash solution adhering to the nylon fiber are not
detrimental, but soggy nylon should be avoided. It is advantageous to
prepare sterile nylon columns in advance.
The lymphocyte-rich fraction is warmed to 37 , mixed thoroughly, and
poured into a waiting nylon column of the type described above. With t~ae
bottom outlet clamped, the cell suspension is incubated for 5 min before
passing the entire sample through the nylon at a flow rate of 4 ml/min.
With the apparatus suggested, the flow rate is 20-25 drops every 15 sec. In
this laboratory, the rate of flow has been controlled by first drawing a
partial vacuum above the surface of the blood in the column with a Pharmacia P-3 peristaltic pump during the 5-min incubation, reversing the
pump, opening the outlet, and finally adjusting the speed of the pump to
490
SPECIFIC CELL LINES
[42]
effect the desired flow rate. A tight-fitting sterile stopper with a sterile
cotton plug is required to couple the pump to the column aseptically.
When the cell suspension has been thoroughly drained or pumped
from the column, it is held at 37 while the nylon is rinsed. Approximately
100 ml of warmed MEM are passed through the column at the same rate of
flow to release trapped cells. The wash solution is reserved in a separate
vessel.
Removal of Platelets
The partially purified cell suspension is freed of platelets by differential
centrifugation at 1000 rpm (200 g) for 10 min. Sterile, plastic conical
centrifuge tubes with caps, holding 50 ml, such as Coming (No. 25335),
Falcon (No. 2074), Kimble (No. 58331), or the equivalent are satisfactory
and can be centrifuged in an HL-8 (Sorvall) or a No. 269 (International)
rotor. The plasma layer, containing platelets, is carefully decanted or
removed with a sterile pipette without mixing the two layers. The deep
red pellet contains the lymphocytes. At a later time the platelets are
separated from the plasma centrifugally (2000--3000 g) in 50-ml sterile
tubes at 20 for 15-20 rain. Although room temperature should be acceptable for this process, in practice, the plastic tubes become deformed at
high speeds in a warm centrifuge. The supernatant fraction, composed of
cell-free plasma, is filtered through a Nalge, 0.45-/zm filter9 to insure
sterility and retained for use as a supplement in the final culture medium.
Separation of Lymphocytes from Erythrocytes

The cell pellets, in their original tubes, are fractionated immediately by
sedimentation through a dense medium. Toward this end, they are each
diluted to 35 ml with the reserved wash solution and, if necessary, with
fresh MEM. Each tube is then underlaid with 13-14 ml of a mixture of
Ficoll and diatrizoate salts (density 1.077 --- 0.001 g/ml) (LSM solutiona),
which is kept refrigerated until required. This is best accomplished by
tilling a 10-ml pipette nearly to the sterile cotton plug and slowly expelling
the liquid with the pipette tip held near the bottom of the tube. A Pipetaid 11 simplifies this task. The resultant two-phase system is centrifuged at
1500 rpm (500 g) for 20 min at room temperature to effect cell separation.
9 Available from Nalgene Labware Division, Rochester, New York.
10 LSM stands for Lymphocyte Separation Medium and can be purchased from Litton
Bionetics, Inc., 5510 Nicholson Lane, Kensington, Maryland.
u Pipet-aid is an automatic pipetting device sold by Drummond Scientific Co., BroomaU,
Pennsylvania.
[42]
491
The dense red cells penetrate the LSM layer and sediment to the bottom.
Lymphocytes band at the interphase while granulocytes, if present, are
found as a whitish layer immediately above the erythrocytes. Contaminating platelets band with lymphocytes.
The lymphocyte layer is recovered with a pipette and separated from
LSM by centrifugation at 1500 rpm for 15 min. The final purified preparation is resuspended in 30 ml of MEM and combined in a single tube. The
entire procedure takes 4 hr, once the blood is drawn.
YieM
For counting, approximately 0.2 ml are withdrawn to a spot plate, and
an aliquot is diluted 20-fold with 0.02% crystal violet in 1% acetic acid. A
white blood cell diluting pipette is convenient for this purpose, and a
hemocytometer is required for quantitation.
The yield from one unit of blood (1 pint; 500 ml) is approximately
4-12 x 108 small lymphocytes. Immediately after purification polymorphonuclear leukocytes account for 0--2% of the total purified leukocyte
population. However, these contaminating cells do not survive well in
culture and are gradually eliminated during incubation. An occasional
monocyte is evident on microscopic examination; such cells are essential
for mitogen stimulation of lymph node lymphocytes from guinea pigs TM
and are believed to be necessary for activating other types of lymphocytes
as well.
Culture Conditions for Human Lymphocytes

In this laboratory the purified cells are usually cultured at 2 l0 s per
milliliter in MEM modified for suspension cultures; higher concentrations,
e.g., I0 r per milliliter, can be used for short incubations. The MEM is
supplemented with 10% autologous plasma, 100 tzg/ml streptomycin, 100
U/ml penicillin, 4 mM glutamine, 0.01 M HEPES buffer, TM and 0.4 mM
nonessential amino acids. The cells are incubated without stirring at 37 in
sealed recumbent prescription bottles. A 32-ounce bottle with a volume of
approximately 950 ml holds 200-275 ml of cell suspension. It is important
to use a smaller bottle for smaller volumes in order to maintain the depth
of the cell suspension at about 1.5-2 cm. Incubation for 15-18 hr is recommended before commencing biochemical studies of resting cells.
13 D. L. Rosenstreich, in "Mitogens in Immunobiology" (J. J. Oppenheim and D. L.
Rosenstreich, eds.), pp. 385-398. Academic Press, New York, 1975.
13 HEPES (N-2-hydroxyethylpiperazine-N-2-ethanesulfonic acid) buffer is available as a
1 M solution in saline from Microbiological Associates, Walkersville, Maryland.
492
SPECIFIC CELL LINES
[42]
If the cells are to be stimulated, they can be incubated with mitogens

immediately. A dose of 5 ~g/ml phytohemagglutinin or 10 ~g/ml concanavalin A is optimal for human peripheral blood lymphocytes. 14 Since
the cells, purified by the procedure reported here, undergo vigorous
mitogen-induced proliferation, the method obviously does not completely
deplete monocytes and macrophages. A lower cell density of 2-3 105
per milliliter has been recommended for growing lymphocytes. 15 In this
laboratory a starting concentration of 2 x 106 per milliliter has been used
routinely, and the activated lymphocytes have been cultured for as long as
10 days. During this interval they are resuspended in fresh medium every
48 hr or diluted with 50% more fresh medium on the same schedule.
Nevertheless, the cultures tend to become acid and survive in better
condition if maintained with loose caps in an environment equilibrated
with 8.5% CO2.
The response to mitogens is asynchronous. For example, phytohemagglutinin stimulates one, two, or three rounds of cell division in some
cells and is toxic to others. ~ Thus the total number of cells rarely increases. However, direct determinations of cell number are complicated
by a somewhat variable tendency to adhere to glass between 6 and 10 hr
after incubation with phytohemagglutinin, and by cell aggregation later in
the activation process.
Impurities
The major contaminants in lymphocyte preparations are platelets and
erythrocytes. Both are destroyed by the dilute acetic acid in the staining
solution used for counting the cells but can be viewed microscopically in
unstained samples. On the average, one erythrocyte is seen per 20 lymphocytes after incubating the cells overnight.
Platelet contamination is more difficult to assess. In this laboratory,
platelets account for 8-25% of the packed cell volume. However, such
contamination can be reduced by repeating the centrifugation step in
100% plasma. For best results the cell pellets, containing predominantly
lymphocytes and erythrocytes, should be resuspended in a small amount
of medium while the recovered plasma is cleared of platelets by sedimentation. When the plasma is ready, the impure lymphocytes are mixed with
~4 For a more complete list of mitogens and their properties, see B. A. Cunningham, B.-A.
Sela, I. Yahara, and G. M. Edelman, in "Mitogens in Immunobioiogy" (J. J. Oppenheim
and D. L. Rosenstreich, eds.), pp. 13-30. Academic Press, New York, 1975, or see
N. Sharon, ibid. pp. 31-41.
15 j. L. Bernheim and J. Mendelsohn, in "Regulatory Mechanisms in Lymphocyte Activation" (D. O. Lucas, ed.), pp. 479-505. Academic Press, New York, 1977.
[42]
493
the plasma and separated from remaining platelets by differential centrifugation as indicated above.
In the procedure detailed by Cooper, 5 platelets are removed by
agglutination with adenosine diphosphate.16 This method is capricious in
its usefulness; it ranges from highly effective to completely ineffective
depending on the donor. If platelet-free cultures are essential, defibrination with glass beads is recommended. Although some loss of lymphocytes from clotted blood cannot be avoided, the fibrin clots trap and bind
virtually all of the platelets, lr
Comments
The inconvenient and tedious preparation of nylon can be circumvented by purchasing this material from Fenwall. The Leuko-Pak (No.
4C2401) contains sufficient teased nylon for 3-4 units of blood. There are
two disadvantages: processed nylon is considerably more expensive than
the cruder product from Dupont, and one becomes dependent on the
availability of Leuko-Paks, which can be out of stock for months.
The separation of erythrocytes by LSM solution has been adopted as a
convenience. The Ficoll-Hypaque method as described by Perper et al. 18
and used by Coope# is equally satisfactory.
The composition of the lymphocytes produced by nylon-column filtration has been a matter of conjecture. It has been suggested that nylon may
deplete the population of B cells (bone marrow-derived lymphocytes)
leaving predominantly T cells (thymus-derived lymphocytes). When the
distribution of the two types of lymphocytes was measured by rosetting
with sheep erythrocytes treated with S-2-aminoethylisothiouronium
bromide hydrobromide, TM approximately 85% were nominally T cells in
our preparations. Studies of the composition of lymphocytes purified
without recourse to nylon gave similar results. 19 There is therefore no
evidence for depletion of B cells during the brief exposure to nylon which
forms part of this procedure for purifying lymphocytes.
A major disadvantage of the human peripheral blood lymphocyte in
research is the difficulty of obtaining large amounts of material. For most
purposes, pooled cells are unsatisfactory even if mitogen treatment is
anticipated. This problem is solved in part by leukophoresis, an alternate
16 A. Gaarder, J. Johnsen, S. Laland, A. Hellem, and P. A. Owren, Nature (London) 192,
531-532 (1961).
17 S. Niewiarowski, E. Regoeczi, G. J. Stewart, A. Senyi, and J. F. Mustard, J. Clin. Invest.
51, 685-700 (1972).
18 R. J. Perper, T. W. Zee, and M. M. Mickelson, J. Lab. Clin. Med. 72, 842-848 (1968).
19 M. E. Kaplan and C. Clark, J. lmmunol. Methods 5, 131-135 (1974).
494
SPECIFIC CELL LINES
[43]
yielding large numbers of white cells which must then be purified as described. However, autologous plasma sufficient for culturing the entire
yield is not provided so that pooled AB serum must be used as a substitute. The effect of foreign substances in serum on the cellular biochemistry of cultured cells has not been fully investigated.
[43] M a c r o p h a g e s
By DOLPH O. ADAMS
Macrophages are currently studied in vitro in almost all fields of biology. The mononuclear phagocyte system, which includes macrophages, is
a host-wide system of phagocytic cells with similar properties. 1 Mononuclear phagocytes arise in the marrow, circulate briefly in the blood as
monocytes, and immigrate into the tissues and inflammatory foci where
they mature into macrophages. Macrophages can be further stimulated to
develop altered function and metabolic characteristics--a state often
termed activation. Activation in this broad sense can be induced by many
stimuli and is characterized by the presence of certain markers 2 as well as
by increases in size, adherence, secretory capacity, content of lysosomes,
and various functions such as phagocytosis and chemotaxis. 1 Since most
activated states are not identical functionally, activated macrophages
should be identified by both the eliciting stimulant and the altered capacity
tested.
Macrophages as cultivated cells offer the following advantages: They
are easy to obtain, can be cultured in relatively pure form, are primary
cultures, and are available in numbers sufficient for analytical biochemical
manipulations. On the other hand, macrophages do not generally replicate
in culture, are relatively short-lived, and may be difficult to obtain in
numbers sufficient for preparative biochemistry, It is important to note
that macrophages also are very sensitive to small changes in their environment and are thereby modified considerably from their native state in
vivo, even when delicately handled and observed after very short periods
of culture.
1 D. O. Adams, Am. J. Pathol. 84, 163 (1976).
z p. Edelson and Z. A. Cohn, in " I n Vitro Methods in Cell Mediated and Tumor Immunity"
(B. R. Bloom and J. R. David, eds.), pp. 333-340. Academic Press, New York, 1976.
METHODS IN ENZYMOL,OGY, VOL. LVIII

ISBN 0-12-181958-2
[43]
MACROPHAGES
495
Sources of Macrophages
General
Macrophages can be obtained from blood, lung, spleen, liver, and the
peritoneal cavity, t2-s} However, mononuclear phagocytes from most of
these sources must be used in short-term experiments since they do not
survive much more than 24-48 hr in culture. Alveolar macrophages from
all sources and peritoneal macrophages from small rodents other than
mice are particularly difficult to maintain. The spleen does not yield many
macrophages, and those obtained are often contaminated by other adherent cells. Hepatic macrophages are considerably altered by the techniques necessary for their isolation. Consequently, peripheral blood
monocytes and peritoneal macrophages are the commonly employed
sources of mononuclear phagocytes.
Monocytes
Monocytes are generally obtained from humans, but equine monocytes have been studied, z Monocytes are separated from the blood by
differential centrifugation and subsequent adherence to culture vessels.
Approximately 108 monocytes can be obtained from 500 ml of blood. This
represents the most readily accessible source of human mononuclear
phagocytes. However, these cells are extensively manipulated before
plating, are difficult to culture, and may be contaminated with large numbers of platelets. A detailed protocol for the culture for human monocytes
is available. 2
Peritoneal Macrophages
The peritoneal cavity offers a ready source of mononuclear phagocytes. The unstimulated peritoneal cavity of mice contains usable numbers (2-3 108) of resident macrophages. These may be washed out directly and placed into culture. The resident cells should not possess any of
the characteristics of inflammatory or activated macrophages. 2 If they do,
the likelihood that the mice are infected should be strongly considered.
3 Z. A. Cohn, this series, Vol. 32, pp. 758-765.
4 G. D. Wasley and R. John, in "Animal Tissue Cultures" (G. D. Wasley, ed.), pp. 101-137.
Butterworth, London, 1972.
5 A. E. Stuart, J. Habeshaw, and A. E. Davison, in "Handbook of Experimental Immunology" (D. M. Weir, ed.), pp. 24.1-24.36. Blackwell, Oxford, 1973.
496
SPECIFIC CELL LINES
[43]
The yield of peritoneal macrophages can be augmented considerably

by injection of various irritants 3-5 days before harvest, s The disadvantage
of this procedure is that the elicited macrophages are newly immigrated
and differ markedly from those normally resident in the peritoneal cavity. 1,2
Furthermore, elicited macrophages differ from one another depending on
the irritant. A wide variety of irritants has been used, each of which has
its own particular advantages and disadvantages, s In general, the kinetics
of appearance of various leukocytes should be determined over several
days for any stimulant employed. Many irritants are indigestible and remain in the macrophages for the duration of their culture, thus serving as a
continuing source of metabolic perturbation.
Serum (1 ml of fetal calf serum intraperitoneally 3 days before harvest)
will generally elicit 3-4 x 106 minimally altered inflammatory macrophages. Glycerol trioleate or mineral oil (1 ml, 3-5 days before harvest)
yields large numbers (20 x 106) of enlarged macrophages, heavily laden
with ingested lipid. Lectins, e.g., 40/xg of purified phytohemagglutinin or
concanavallin A, given 3 days before harvest, produce moderate numbers
(6-8 x 106) of macrophages that are large, do well in culture, and do not
visibly contain the eliciting stimulant. Thioglycolate broth (1 ml, 3-5 days
before harvest) elicits many (20-30 x 106) large macrophages, heavily
laden with phagocytic vacuoles. Although there are numerous media containing thioglycolic acid or sodium thioglycolate, Brewer's Thioglycolate
medium (Difco Lab., Detroit, Michigan, catalogue no. B236) should be
used to obtain results comparable to those described by other laboratories. 6
Human peritoneal macrophages can be obtained from cadavers or
from peritoneal dialysis fluid, s These are obviously from diseased or injured individuals and, in the latter instance, present a high risk of exposure of laboratory personnel to hepatitis virus.
Alveolar Macrophages
Alveolar macrophages can be obtained from guinea pigs, rabbits, and
humans, z In the former two circumstances, 20 106 macrophages can
normally be lavaged from the lungs. Up to 20 x 108 macrophages are
obtainable if the animals are given Freund's adjuvant or avirulent tubercle
bacilli intravenously 3-4 weeks before harvest. Human alveolar mac-
6 S. Gordon, Z. Werb, and Z. A. Cohn in "In Vitro Methods in Cell Mediated and Tumor
Immunity" (B. R. Bloom and J. R. David, eds.), pp. 341-352. Academic Press, New
York, 1976.
r T. P. Stossel and Z. A. Cohn, Methods lmmunol, lmmunochem. 5, 261-292 (1976).
[43]
MACROPHAGES
497
rophages can be obtained by direct bronchial lavage of both normal volunteers and patients. 8
Separation and Purification of Macrophages
Only techniques for enrichment of macrophages will be described here.
Procedures for depletion of mononuclear phagocytes are available elsewhere. 9 Enriched populations of macrophages generally are obtained by
centrifugation, adherence, or combinations of the two.
Centrifugation
A fraction of mononuclear leukocytes containing both lymphocytes
and monocytes can be separated from other leukocytes and from red cells
by centrifuging in albumin or in a gradient of Ficoll-Hypaque. 2 Contaminating platelets are then partially removed from the mononuclear
fraction by differential centrifugation. This technique, which is necessary
to obtain monocytes, may injure or alter the cells due to the extensive
separation procedure.
Adherence
Most mononuclear phagocytes, particularly mature or elicited ones,
adhere avidly to culture vessels. 1 This property is effectively exploited in
purification by adding populations of leukocytes to culture vessels and
incubating them in medium containing serum for several hours. The
nonadherent cells are removed, and the adherent cells are washed. The
washings must be vigorous to dislarge other adherent leukocytes such as
neutrophils and stimulated T cells. The adherent cell population can then
be trypsinized to remove fibroblasts or tumor cells. 11 The remaining cells
should be macrophages of high purity (95% or greater). The proportion of
the cell population that is macrophages is determined routinely as described below.
Adherence does possess certain disadvantages. Some of the mononuclear phagocytes will not adhere in several hours and may be lost. However, cultures may be continued without washing for 24 hr, which permits
a larger number of the mononuclear phagocytes to adhere. This interval
also results in death and disintegration of the neutrophils, further con8 D.
K.
10 M.
11 R.
W. Golde, J. Reticuloendothel. Soc. 22, 223 (1977).

Shortman, Contemp. Top. Mol. Irnmunol. 3, 161, (1974).
E. Fedorko and J. G. Hirsch, Semin. Hematol. 7, 109 (1970).
Evans, J. Natl. Cancer Inst. 50, 271, (1973).
498
SPECIFIC CELL LINES
[43]
tributing to the purity of the macrophages. The resultant macrophages

should be used directly in the separation vessel if possible, since they are
extremely difficult to remove. Despite these disadvantages, the adherence
technique yields highly purified macrophages that can be used within
hours of removal and that have been subjected to little manipulation.
Most subsequent experimental manipulations can be conducted without
removing the macrophages by appropriate selection of the vessel used for
purification.
Identification of Macrophages
Precise identification of a given population of cells as mononuclear
phagocytes is vital because of the number of other cells that resemble
them morphologically, particularly lymphocytes. Mononuclear phagocytes are defined as adherent cells that have a characteristic morphology,
the capaciW for extensive phagocytosis, and bear receptors for the activated third component of complement (3) and for the activated C-terminal
portion of immunoglobulin.
Morphology and Histochernistry

The morphology of mononuclear phagocytes is well defined. 1 A population of spread adherent cells in the size range of 20-50/xm and having
characteristic nuclei and ruffled cytoplasm can be tentatively identified as
macrophages. Resistance to removal with trypsin further strengthens this
tentative identification.11 Mononuclear phagocytes contain abundant
quantities of a nonspecific esterase, the presence of which distinguishes
these cells from other leukocytes,n It is worth emphasizing that differences in the nonspecific esterase, as with most histochemical procedures,
represent a quantitative rather than a qualitative difference between macrophages and other cells.
Phagocytic Uptake
Extensive phagocytic uptake by a mononuclear cell population is clear
evidence of a lineage in the mononuclear phagocyte system. Procedures
for determining phagocytic uptake are well detailed. 7 A test particle such
as starch, latex beads, or erythrocytes plus an opsonin (fresh serum or
specific antibody) are incubated for 30-60 rain at 37 with the mac1~ L. T. Yam, C. Y. Li, and W. H. Crosby, Am. J. Clin. Pathol. 55, 283 (1971).
[43]
MACROPHAGES
499
rophages. The cells are then washed and the percentage of mononuclear
cells containing five or more particles within the cytoplasm (not adherent to
the cells) is determined. Lab-Tek chambers (Lab-Tek Products, Napierville, Illinois) or coverslips suspended over slides with a depressed well
(Arthur H. Thomas, cat. no. 6688-D20) are particularly convenient for this
purpose.
Markers
Mononuclear phagocytes are distinguished by the presence of surface
receptors for both C3h and Fc. 13,14These may be demonstrated by use of
appropriately coated red cells. Detailed protocols for these procedures
are available) T M
Culture Techniques
The selection of appropriate culture conditions for macrophages depends upon the intent of the particular experiment and usually represents
a balance between multiple factors. First, basal media are less expensive
and will support viability of the macrophages but may not permit full
expression of all functional capabilities. Second, macrophages are easily
stimulated by alterations in their environment such as enriched media,
high concentrations of serum, trace concentrations of endotoxin, or
phagocytosis of other cells. Such endogenous stimuli can obscure differences between experimental and control cultures. Third, the inherent inconsistencies of tissue culture must be considered. In our hands, for example, Dulbecco's minimum essential medium (MEM) supports accumulation of acid hydrolases much better than does Eagle's MEM, whereas
the reverse is true for the expression of nonspecific cytotoxicity.
General Requirements
Macrophages usually are cultured in moist air containing 5% COz.
Precise regulation of CO2 content with a controller may be advantageous
for critical experiments. Acidity is common with cultured macrophages,
and pH should be checked daily. The pH can be adjusted to 7.2 by
addition of sterile sodium bicarbonate (sterilized by filtration).
Plastic vessels coated for tissue culture are generally suitable for
13 j. Michl, D. J. Ohlbaum, and S. C. Silverstein, J. Exp. Med. 144, 1465 (1977).
14 E. M. Shevach, E. S. Jatfe, and I. Green, Transplant. Rev. 16, 3 (1973).
500
SPECIFIC CELL LINES
[43]
growth of macrophages. We have found clusters of four 60-mm wells

useful for large numbers (10 10n/well) of macrophages and plates containing twenty-four 16-mm wells useful for small numbers (5 105/well).
These plates must be maintained in a moist atmosphere to prevent evaporation. T-flasks do not have this disadvantage, but are difficult to wash
vigorously for purification and difficult to scrape completely when removing the macrophages. Small culture chambers placed over glass slides
(Lab-Tek) are extremely useful for assessing morphology and for autoradiography.
Macrophages generally do best in culture at densities of from
2-4 x 105 adherent macrophages per cm 2. The number of suspended
leukocytes to be added to a culture to achieve this density can be roughly
calculated by multiplying the number of macrophages per milliliter of
suspension by the proportion of macrophages expected to adhere, i.e., by
to ~ if the macrophages are activated and by to k if they are not.
Cultures are washed free of nonadherent cells after 3 or 4 hr, except in
the case of human monocytes, which are washed after 1 hr. Leaving the
cultures intact for 24 hr generally improves the number of adherent cells
and the purity of the resulting population.
Media
Medium 199 and Eagle's minimal essential medium are useful basal
media. These are generally supplemented with streptomycin, penicillin,
and fresh glutamine. HEPES buffer, 5-10 mM, aids in stabilizing pH but
may be toxic to some cultures.
NCTC-135, MEM alpha, RPMI-1640, and Ham's FI2 are richer media
that have been successfully employed in culturing macrophages.
Basal media can be selectively enriched by addition of a variety of
substances including sodium pyruvate (0.11 mg/ml), ascorbic acid (50
/zg/ml), other vitamins, and nonessential amino acids.~5
Serum-free medium may be necessary for some experiments. A variety of supplements have been tried; we have had the most success with
lactalbumin hydrolysate (Grand Island Biological Co.). The cell-free medium of Newman-Tytell is particularly useful, especially when supplemented with 10 mM HEPES. We have recently begun using the supplement described by Guilbert and Iscove of selenite, transferrin, lecithin,
and albumin TM and have found that it improves the viability and maturation
of macrophages when added to serum-free cultures.
15 C. W a y m o u t h , Int. Rev. Cytol. 3, 1 (1954).
an L. J. Guilbert and N. N. I s c o v e , Nature (London) 263, 594 (1976).
[43]
MACROPHAGES
501
Serum
Most cultures of macrophages are supported by 10-40% serum. Fetal
calf serum is routinely employed, but equine and newborn calf serum also
have been used. Newborn calf serum may contain antibodies that stimulate maturation of the macrophages. 3 Human macrophages usually are
cultured with human serum, preferably autologous. For murine macrophages, we routinely used 10% heat-inactivated fetal calf serum.
Serum can vary considerably from lot to lot in its ability to support
both survival and function of macrophages. For critical experiments,
serum may have to be screened by testing the ability of various lots to
support the function being tested. Furthermore, certain components of
serum are apparently labile. In critical experiments, we obtain frozen
serum shipped in Dry Ice from a previously screened lot. All is gently
thawed at 37, heat-inactivated with constant swirling for 30 min at 56,
apportioned into small aliquots, and stored at - 2 0 . On the day an experiment begins, one container of serum is gently thawed and used. Any left
over serum is not reused for an experiment, not even the next day, but is
saved for routine tissue culture.
Detachment of Macrophages
Detaching macrophages that have already adhered to a culture vessel
particularly if the macrophages are activated, is extremely difficult.
Neither trypsin, cold, nor chelating agents remove macrophages well, and
it is generally best to conduct experimental manipulations in the culture
vessel used for purification. Gentle scraping of macrophage cultures with
a rubber policeman will remove the cells, approximately 1/z of which may
be viable. Macrophages may be detached partially by use of a local anesthetic, a7 Purified Lidocaine, without preservatives (Astra Pharmaceuticals, Worchester, Massachusetts) is made to 360 mM in phosphatebuffered saline (PBS) and adjusted to pH 6.6 with 1 M NaOH. The stock is
diluted in medium to 12 mM, and macrophages are incubated in the
medium containing Lidocaine for 5 min at 37. The macrophages should
then appear rounded and are easier to remove, although extensive losses
may be incurred.
Observation and Viability of Cultures

Cultures are observed daily with an inverted phase microscope.
Healthy cultures will appear as monolayers of plump or stellate, ~hase17 C. Nathan, J. lrnmunol. 118, 1612 (1977).
502
SPECIFIC CELL LINES
[43]
dense, well-spread, and closely approximated cells. Unhealthy cultures

will be sparsely populated with rounded, unspread, and phse-lucent cells
and will contain many floating cells.
More stringent criteria for determining viability include testing the
ability to phagocytose particles, to secrete lysozyme, and to hydrolyze
fluorescein diacetate, is
Long-Term Cultures
Macrophages in culture are generally nonreplicating cells except under

special circumstances. 19 Murine peritoneal macrophages and human
monocytes under well-controlled conditions can survive for periods of up
to 2 or 3 weeks.
Preparation of Macrophages for Study
Morphologic Studies
For study of macrophages in exudates or suspension cultures, an excellent method is to prepare them in a Cytocentrifuge (Shandon Southern
Instrument Co., Sewickeley, Pennsylvania) and stain the resultant
smears with a commercially available Wright's stain (Diffquik, Arthur
Thomas Co.). This stain is also convenient for macrophages cultivated on
coverslips or in Lab-Tek chambers.
Phase microscopy is extremely useful for more detailed examinations.
Macrophages are particularly advantageous to study by this method,
since they spread after adherence and thus reveal their cytoplasmic contents in the thinned cytoplasm. Cultures of spread macrophages on coverslips are washed in PBS and fixed 5 rain in 4% osmium tetroxide in
S-Collidine buffer while in a f u m e hood. The coverslips are immersed over
PBS in a trough whose shallow walls of silicone grease are applied to a
glass slide with a syringe and needle. The resultant inverted wet mount is
studied under the phase microscope and can be preserved by coating it
with clear nail polish.
Biochemical Preparation
Studying the intracellular contents of macrophages poses problems,

since macrophages are relatively resistant to lysis. For example, mac~8A. C. Allison,in "In Vitro Methodsin Cell-Mediatedand TumorImmunity"(B. R. Bloom
and J. R. David, eds.), pp. 395--404. AcademicPress, New York, 1976.
19M. Virolainenand V. Defendi, Wistar Inst. Syrup. Monogr. 7, 67 (1967).
[43]
MACROPHAGES
503
rophages can remain intact after multiple cycles of freezing and thawing.
We generally employ the detergent Triton X-100. We add ice-cold PBS
containing 0.2% Triton X-100 to washed cultures of macrophages and
incubate the cultures with the detergent for 30 min. The entire procedure
must be conducted on ice to prevent liberated lysosomal enzymes from
destroying the desired cellular constituent. After the incubation, macrophages are completely removed from the culture vessel by scraping
with a rubber policeman. The preparation of organelles from macrophages has been described in detail. 2
The study of products secreted by macrophages is conducted on medium conditioned by growing macrophages in it. 6 Since many secreted
products, e.g., neutral proteases, are inhibited by serum constituents, the
cultures are frequently conducted in serum-free medium. Siliconized
glassware is useful because many of the secreted products are extremely
labile and adhere to glass. Daily, the cultures of macrophages are aspirated and the aspirates placed in siliconized conical centrifuge tubes on ice
and immediately spun 10 min at 10,000 g at 4. If the desired secretory
product is present in low concentration, the conditioned medium can be
lyophilized or dialized against an adsorbent such as Acquacide II (Calbiochem, San Diego, California) to concentrate the constituent. In our
hands, concentration in a stirred cell against an appropriately sized UM
membrane (Amicon Inc., Lexington, Massachusetts) has proven the most
useful and satisfactory method.
Specific activity of a given constituent within macrophages can be
expressed in relation to either protein or cell number. Determination of
cell number is generally advantageous because cultivated macrophages
usually produce large amounts of protein in culture and thus specific
activity may fall in the face of a rising total amount of a particular enzyme. 21 Determining cell number by hemocytometer count is often inaccurate because a large but variable number of macrophages is lysed while
scraping them from the culture vessel. Determination of DNA content
provides an accurate, reliable, and sensitive method for quantifying cell
number. 2z
Continuous Cell Lines
Several lines of continuously replicating cells having the properties of
macrophages are now available (for review, see Defendi2Z). These neon0z. Werb and Z. A. Cohn, this series, Vol. 31, P. A, pp. 339-345.
21 Z. A. Cohn and B. Benson,J. Exp. Med. 121, 153 (1968).
z2 S. Cooksonand D. O. Adams,J. lmmunol. Methods (1978) in press.
2s V. Defendi, in "Immunobiologyof the Macrophage" (D. S. Nelson, ed.), pp. 275-286.
504
SPECIFIC CELL LINES
[43]
plastic murine lines have the morphology of macrophages and are phagocytic. Depending on the particular line, these cells may express various
other functions generally associated with macrophages. The macrophagelike lines are generally easy to grow and are passaged without difficulty,
since they detach readily from the culture vessels. They offer the obvious
advantage of providing large numbers of similar cells with minimal
difficulty.
Appendix: Protocol for Culturing Stimulated Macrophages
Animals
Two C57 B 1/6J mice of either sex (Jackson Laboratories, Bar Harbor,
Maine) weighing 25-30 g each are used.
Supplies
Small, lidded plastic bucket, containing one-haft pound of Dry Ice
Small dissecting board, covered with sterile barrier
70% ethanol
Two small forceps with teeth
One 3 inch dissecting scissors
Sterile, plugged, Pasteur pipettes with rubber bulbs
Two sterile 23-gauge 1 inch needles
Two sterile 10-ml syringes
One sterile 50-ml conical polypropylene centrifuge tube with cap
30 ml of ice-cold washout medium [Eagle's minimal essential medium
with Earle's salts (Grand Island Biological Co., Grand Island, New
York, cat. no. F-11); to this, add 10 units of heparin per milliliter]
Ice bucket filled with ice
3 cluster dishes with four 60-mm wells of tissue culture plastic
(Falcon Plastics, Los Angeles, California)
250 ml of sterile Eagle's MEM with Earle's salts to wash plates
Hemocytometer
Gilson or similar pipettes, dispensing 20 and 200/A
10 ml sterile plugged pipettes
Glass slides
Coverslips
2 ml of sterile, Brewer's thioglycolate broth (Difco Manufacturing
Co., Detroit, Michigan, cat. no. B-236; prepare and store
according to manufacturer's instructions)
[43]
MACROPHAGES
505
150 ml of complete tissue culture medium to culture macrophages:

Eagle's minimal essential medium with Eade's salts (Gibco no.
F-11) plus 10% heat-inactivated fetal calf serum plus 2.5 x l04
units of penicillin per 100 ml plus 12.5 mg of streptomycin/100 ml
plus 29.2 mg of fresh L-glutamine (Sigma Chemical Corp., St.
Louis, Missouri)/100 ml. Basic medium can be made in advance,
but the final assembly of medium plus serum, antibiotics, and
supplements should be prepared on the day of the experiment.
Stimulation of Peritoneal Exudate Cells (PEC)

Three days before the experiment, inject each mouse sterilely with
1 ml of thoglycolate broth intraperitoneally with a 23-gauge needle. Be
sure that the injection does not perforate or enter the intestines.
Collection of PEC
Lay out all equipment. Load two syringes with 10 ml of chilled washout medium and place them and the centrifuge tube on ice. Note that the
entire collection and culture procedure is to be done under sterile
conditions.
To kill the mouse, place it in the plastic bucket containing the Dry Ice.
Suspend it over the ice by its tail placed under the rim of the closed lid.
After 1 min, remove the dead mouse.
Wash the skin of the abdomen of the mouse vigorously with 70%
ethanol. Gently pick up the abdominal skin with a pair of toothed forceps
in each hand and gently manipulate the skin up and down to separate it
from the peritoneal wall. Grasp the skin with both forceps and gently tear
it, making sure not to tear the peritoneal wall. Pull the torn skin caudally
and rostrally until the carcass of the mouse is completely exposed. Do not
tear the peritoneal wall or the thorax.
Rewash the peritoneal wall with 70% alcohol. Insert the needle into
the peritoneal cavity in the midline and inject the 10 ml of washout medium. Massage the flanks of the animal for several seconds. Elevate theI
shaft of the needle to create a small tent just below the xyphoid process
and aspirate the injected fluid. At least 9 ml should be recovered.
Remove the needle from the syringe and gently introduce the collected
fluid into the centrifuge tube. Cover the centrifuge tube. Wash the second
mouse similarly. Add the fluid from the second mouse to that of the first
and mix gently. Remove a few drops of the fluid and prepare smears on
the Cytocentrifuge, keeping the centrifuge tube on ice.
506
SPECIFIC CELL LINES
[44]
Culture of Macrophages
Immediately centrifuge the collected peritoneal fluid for 10 rain at 250g
at 4. Carefully aspirate the supernant from the resultant cell pellet. Add
10 ml of complete tissue culture medium. Gently disperse the cell button
with a cotton-plugged sterile Pasteur pipette and take a small sample.
Return the capped centrifuge tube to the ice bucket.
Determine the cell number in a hemocytometer, using the Gilson
pipettes for dilution and counting at least 400 cells. Examine the stained
cytocentrifuge smear and do a differential count. Typically, each mouse
should yield 20-30 10n PEC, of which approximately 90% should be
large macrophages. Calculate the final dilution of the PEC. To get 3 105
adherent macrophages/cm 2, a concentration of 2.5 x 10n PEC/ml in 6 ml
will be added to each 30 cm 2 well. (2.5 10n PEC/ml 90% macrophages
~ of elicited macrophages to adhere 6 ml + 30 cm z = 3 105 adherent
macrophages/cm2).
Make this dilution and add 6 ml of the suspension to each well. Shake
the dishes laterally to disperse the cells evenly and incubate the dishes for
3 hr in a humidified CO2 incubator at 37. Aspirate the medium and wash
each well vigorously 3 times with tissue culture medium containing no
serum or supplements. The cultures at this time should consist of
monolayers of well-spread phase-dense macrophages having prominent
cytoplasmic ruffles. Add 6 ml of complete culture medium to each well
and return to the incubator.
Acknowledgments
This work was supported in part by U.S.P.H.S. (}rants CA-14236 and CA-16784.
[44] M o u s e E r y t h r o l e u k e m i a C e l l s
By T. V. GOPALAKRISHNANand W. FRENCH ANDERSON
Mouse erythroleukemia (MEL) or Friend cells grow in suspension
culture and have been extensively used as a model system for studying
erythropoiesis in vitro.l,2 When MEL cells are grown for several days in
the presence of dimethylsulfoxide (DMSO) or a variety of other chemical
agents, 3 they undergo changes similar to the normal maturation of red
1 C. Friend, H. D. Preisler, and W. Scher, Top. DeF. Biol. 8, 81 (1974).
2 p. R. Harrison, Int. Rev. Biochem. 15, 227 (1977).
3 R. C. Reuben, R. L. Wife, R. Breslow, R. A. Rifkind, and P. A. Marks,Proc. Natl. Acad.
Sci. U.S.A. 73, 862 (1976).

ISBN 0o12-181958-2
[44]
MOUSE ERYTHROLEUKEMIA
507
blood cells including alterations in morphology and the production of

hemoglobin, heme, carbonic anhydrase, spectrin, and red blood cell surface antigens. 4-s H o w e v e r , these cells are insensitive to the hormone
erythropoietin, the natural mammalian hormone involved in the in v i v o
development of erythroid cells. 9 Hence, Friend cells are considered by
most observers to be erythroid cells arrested at or before the proerythroblast stage. The reason for the inability of these cells to undergo the
normal maturation process in the absence of any chemical inducers remains unknown.
Because of the ease o f altering conditions in v i t r o , M E L cells are a
potentially useful system for studying the molecular control of hemoglobin gene expressions as well as other biochemical and genetic aspects of
erythroid cell maturation, l'z~-lz The unique advantage o f the globin system is that the immediate gene product, globin messenger R N A (mRNA),
can be assayed directly and quantitatively using a specific molecular
probe, globin c o m p l e m e n t a r y D N A (cDNA). Little or no hybridizable
globin m R N A is present in uninduced M E L cells, but globin m R N A can
be detected 2 days after induction with D M S O and reaches a maximum
concentration 4 days after incubation. 1~ Considerable studies are now
u n d e r w a y attempting to understand the molecular events that o c c u r inside the intact M E L cell between the time of exposure to an inducing
agent and the synthesis of globin m R N A .
Growth of M E L Cells
A . C e l l s . M E L cells can be obtained from the Institute for Medical
Research, C o p e w o o d and Davis Streets, Camden, N e w Jersey. Strains
GM-86 and GM-979 are presently available. Other isolates or mutants can
be obtained from individual investigators. In addition, these cells can be
4 c. Friend, W. Scher, J. G. Holland, and T. Sato, Proc. Natl. Acad. Sci. U.S.A. 68, 378
(1971).
5 D. Kabat, C. C. Sherton, L. H. Evans, R. Bigley, and D. Koler, Cell 5, 331 (1975).
6 p. S. Ebert and Y. Ikawa, Proc. Soc. Exp. Biol. Med. 146, 601 (1974).
r H. Eisen, R. Bach, and R. Emery, Proc. Natl. Acad. Sci. U.S.A. 74, 3898 (1977).
8 y. Ikawa, M. Furusawa, and H. Sugano, Bibl. Haematol. (Basel) 39, 955 (1973).
9 A. W. Nienhuis, J. E. Barker, and W. F. Anderson, in "Kidney Hormones" (J. W.
Fischer, ed.), p. 245. Academic Press, New York, 1978.
10A. W. Nienhuis, J. E. Barker, A. Deisseroth, and W. F. Anderson, Ciba Found. Symp. 37,
329 (New Ser.), (1976).
11T. V. Gopalakrishnan, E. B. Thompson, and W. F. Anderson, Proc. Natl. Acad. Sci.
U.S.A. 74, 1642 (1977).
12D. E. Axelrod, T. V. Gopalakrishnan, M. Willing, and W. F. Anderson, Somatic Cell
Genetics, 4, 152 (1978).
~aj. Ross, Y. Ikawa, and P. Ledcr, Proc. Natl. Acad. Sci. U.S.A. 69, 3620 (1975).
508
SPECIFIC CELL LINES
[44]
obtained as primaries by following the procedure of Friend. l Briefly, susceptile mice are infected with Friend virus complex from which they
develop an erythroleukemia syndrome. Cells from the enlarged spleen are
then inoculated subcutaneously into another susceptible mouse and a
tumor resembling a reticulum cell sarcoma develops. Cells from this
tumor can be cultured indefinitely in suspension as MEL cells.
B. Growth Conditions. MEL cells are grown at 37 in a humidified
water-jacketed incubator equilibrated with 10% COz in air. Growth at 37
is not greatly influenced by slight variations in the CO2 concentration.
Since the CO2 content inside the incubator determines the pH of the
medium (which is buffered with NaHCOz), too much or too little COz will
lower or raise the pH and be lethal for the cells. A CO2 concentration that
lies between 5-10% maintains the pH of any one of the media described
below in a range that is conducive to good cell growth.
C. Medium. MEL cells are easily grown as a suspension culture in any
of the established media supplemented with 10% fetal calf serum
(GIBCO). Routinely these cells have been grown in BME (Basal minimal
medium, GIBCO) or in slightly richer media such as Modified Improved
Minimal Essential Medium (Modified IMEM) or F14 medium (available
from GIBCO as F12 modified medium 72004 and then supplemented with
3.7 g/liter of NaHCO3). Depending upon the medium and growth conditions, MEL cells grow with a doubling time of 10-18 hr. It is important to
maintain the cells in logarithmic growth. This is achieved by diluting the
cells 1:50 twice a week; slower growing cells can be passed at 1:100
dilution once a week. These dilutions are useful for the maintenance of
stock cultures, whereas cells to be used regularly for experiments can be
maintained at 1:10 dilution at frequent intervals.
D. Serum. Fetal calf serum is used with medium at a concentration of
10% (v/v). Several different lots of serum should be tested for their ability
to promote optimal growth as well as for hemoglobin induction of the
MEL cells. Markedly different doubling times and percent inducible cells
are obtained with different lots of serum from the same (or different)
companies.
1. Lot Testing for Growth Support

Growth media supplemented with 10% fetal calf serum of different lot
numbers are prepared. The capacity of different lots of serum to support
cell growth can be tested in two different ways: (a) by cloning efficiency or
(b) by doubling time.
[44]
MOUSE ERYTHROLEUKEMIA
509
(a) Cloning efficiency is determined by suspending in separate tubes

MEL cells in medium plus each different lot of serum, then plating in
microtiter dishes. The cell suspensions are diluted to give a concentration
of 1.5 cells/ml; 0.2 ml of each suspension is added to each well of a 96-well
microtiter dish (Linbro or Falcon Plastics); one microtiter dish is used for
each serum lot being tested. Thus, roughly 30% of the wells will have one
cell. The dishes are then kept in the incubator until visible colonies appear
in the wells. One can see a colony with the naked eye in 5-7 days or
under low power with an inverted phase microscope earlier. If cloning
efficiency were 100%, 30% of the wells (29) would have colonies; the
actual cloning efficiency can be calculated from the actual number of
colonies observed. The serum lot that gives the best cloning efficiency is
then determined.
(b) An approximate doubling time can be determined as follows: 104
MEL cells are seeded in 10 ml of growth medium prepared with different
lots of serum in separate 100-mm Petri dishes in duplicate. The dishes are
incubated for a fixed period of time, normally 3-5 days. At the end of this
period, the total number of viable cells in each dish is determined with a
hemocytometer. The number of hours of growth is divided by the number
of cell doublings.
2. Lot Testing for Hemoglobin Inducibility

For testing the inducibility of hemoglobin by DMSO (or another inducing agent) using the medium prepared with different lots of serum, MEL
cells are centrifuged and washed once with serum-free medium. The cells
are suspended in serum-free medium and their concentration determined.
The cell density is then adjusted by suitable dilution with serum-free
medium to 106 cells/ml. Nine milliliters each of medium prepared with
different lots of serum are placed in separate 100-ram Petri dishes and to
each of these 0.2 ml of DMSO is added and mixed; then 1 ml of the MEL
cell suspension is added and mixed. The dishes are incubated for 5 days at
37, at which time the percentage of benzidine-positive cells in each dish is
determined as described below. The serum lot that gives the maximum
percentage of benzidine-positive cells is selected. It has been observed
that the ability of different lots of sera to bring about hemoglobin induction by 2% DMSO varies more widely than growth support. Therefore, a
lot that gives the highest induction is usually selected over one with a
slightly superior growth support but less induction potential. Once a satisfactory lot is found, it is purchased in bulk and stored frozen at -20 or
below until used.
510
SPECIFIC CELL LINES
[44]
Hemoglobin Induction
For induction experiments, cells are plated in regular growth medium
at a density of 1 105 cells/ml with 1.5-2% (v/v) DMSO. ~4 After 4 or 5
days, the cell pellet appears pink to deep red depending upon the extent of
hemoglobin production. For the quantitative estimation of the percentage
of cells producing hemoglobin, the cells are stained with benzidine as
described below.
DMSO should not be autoclaved for sterility since this inhibits induction of hemoglobin in MEL cells. Hence, a stock solution of 15-20%
DMSO in growth medium is prepared and then sterilized by filtration
through ~algene filters with 0.45-/~m pores. The stock solution can be
stored at 4 . This solution is diluted 1:10 with medium so as to bring the
final concentration of DMSO to 1.5-2.0~. In addition to DMSO, a variety
of other chemical agents bring about the induction of hemoglobin in MEL
cells, a Of these inducers, hexamethylenebisacetamide has proved to be
the most potent inducer for hemoglobin.
Benzidine Staining
A. Stock Solution. The stock solution is 0.2% (w/v) benzidine dihydrochloride in 0.5 M acetic acid. The solution is stable for many months
when stored in a b r o w n bottle and kept refrigerated. Note: ~ i s is a potential carcinogen and, therefore, should be handled using proper precautions.
B. Stain Solution. A fresh solution of 0.4% of 30% H~O2 (v/v) in the
benzidine stock solution (i.e., 10 microliters 30% H~O~ in 2.5 ml stock
solution) is prepared just prior to use. One-tenth volume of the stain
solution is added to a cell suspension that is to be tested for hemoglobin
induction. Routinely we use 4 ml of a suspension of M E L cells that have
been treated with 2% DMSO for 5-6 days; 0.4 ml of the stain solution is
added and mixed. The cells that contain hemoglobin turn blue in 5-10
min. The percentage of benzidine-positive cells is determined by brightfield light microscopy; 200-500 cells are counted for an accurate estimation. It is essential to examine cells in different areas of the dish to insure
that the distribution of benzidine-positive cells is uniform. A potential
source of error arises from the tendency of the floating population of cells
to concentrate at the center of the dish. One way of overcoming this
problem is to mix the cell population very well after the addition of the
stain solution. We have also noted that the benzidine-positive cells have a
tendency to float and the negative cells to sink. This should be taken into
account during counting.
14Obtained from Fisher, Fisher Scientific Co., Pittsburgh, Pa. Since different lots of
DMSO also differ in their ability to bring about hemoglobin induction, DMSO should be
lot tested also.
[45]
SKELETAL MYOBLASTS IN C U L T U R E
511
Spectrometric Determination of Hemoglobin

Hemoglobin induction in a population of cells also can be estimated by
lysing the cells and measuring the hemoglobin content spectrophotometrically. 5 This is done as follows: The cell suspension is centrifuged at 100g
for 3 min and the cell pellet washed once in normal saline. The cells are
lysed by suspending the cells in 3 volumes of ly~ing buffer (50 mM
Tris.HC1, pH 7.0; 25mM KC1; 5 mM MgC12; 1 mM 2-mercaptoethanol;
and 0.3% Triton X-100) at 4 and vortexing briefly. The lysed cell suspension is centrifuged in an Eppendoff centrifuge at maximum speed (8000 g)
for 10--15 min using microcentrifuge tubes. The supernatant liquid is collected, and absorbance is measured in a spectrophotometer at 420 nm
using the lysing buffer as blank. The absorbance at 420 nm is multiplied by
0.0945 to give hemoglobin concentration in milligrams per milliliter. The
hemoglobin content in the cell lysate is normalized per 106 cells or per
milligram of soluble protein.
[45] S k e l e t a l M y o b l a s t s in C u l t u r e
By IRWIN R. KONIGSBERG
Cell culture systems have been employed for the past 17 years 1"2 to
study the differentiation of skeletal muscle fibers from myoblasts, the
embryonic stem cells from which this tissue develops in vivo. When appropriately cultured, these progenitor cells reproduce, with remarkable
fidelity, the sequence of events that have been observed during normal
embryonic development. Cell culture techniques provide a number of
advantages unattainable in the intact organism. One can obtain a uniform,
highly purified population of cells of known developmental fate. These
can be maintained under rigidly controlled conditions and subjected to a
wide range of experimental intervention. Finally one can manipulate culture conditions to impose a greater degree of synchrony than is ever
observed in the organism.
The use of skeletal muscle cell culture as an experimental system is
now widely employed. Although the same overall strategy is used, the
details of the specific techniques used vary widely from one laboratory to
another. At times, therefore, it becomes difficult to compare the results
obtained by different investigators. This article will deal largely with the
techniques and procedures currently in use in our laboratory. Where
1 L. M. Rinaldini, Exp. Cell Res. 16, 477 (1959).
2 I. R. K o n i g s b e r g , Exp. Cell Res. 2 1 , 4 1 4 - 4 2 0 (1960); I. R. K o n i g s b e r g et al., J. Biophys.
Biochem. Cytol. 8, 333 (1960).

1SBN 0-12-181958-2
512
SPECIFIC CELL LINES
[45]
alternatives p r o c e d u r e s are e m p l o y e d b y other investigators these will be

described briefly and referenced. To the extent that we k n o w how variations in protocols and p r o c e d u r e s might affect results and observations,
these will be discussed.
Since cell culture is m o r e art than science, m u c h of the technology is
empirical but based on extensive testing that is, unfortunately, largely
unpublished: W h e r e v e r appropriate, the results of such tests will be cited
and in general the rationale o f our p r o c e d u r e s will be indicated.
Cell Source
E m b r y o n i c muscle f r o m a n u m b e r o f species is c o m m o n l y e m p l o y e d
(chick, 1"2 quail, 3 rat, 4 calf, 5 m o u s e , ~'r and humane), but muscle f r o m such
sources as Drosophila, 9 the butterfly, ~ cricket, 11 and lizard 12 has also been
cultured. We routinely use avian e m b r y o s , both chick (Gallus d o m e s t i c u s )
as well as quail (Coturnix coturnix j a p o n i c a ) , finding the latter species
preferable. F r o m the standpoint of the yield o f viable cells we find 11-day
chick e m b r y o s (either pectoral or leg muscle) and 9-day quail e m b r y o s
(pectoral muscle) to be the m o s t suitable stages. The choice of stage is a
c o m p r o m i s e . With increased age a larger v o l u m e of muscle tissue can be
obtained per e m b r y o . H o w e v e r , as d e v e l o p m e n t increases the ratio of
m y o b l a s t s to nonproliferative nascent multinuclear fibers decreases.
S o m e difficulty is also e n c o u n t e r e d in dissociating tissue f r o m older emb r y o s due, p r e s u m a b l y , to increased cellular packing and the p r e s e n c e of
m o r e substantial a m o u n t s of extracellular connective tissue elements.
Dissection and Mincing the Tissue
In establishing p r i m a r y cultures all operations b e t w e e n sacrificing the
e m b r y o to inoculating the petri plates with single cells must be p e r f o r m e d
as quickly as possible. We use no more tissue than can be p r o c e s s e d
3 I. R. Konigsberg, Dev. Biol. 26, 133 (1971).
4 D. Yaffe, Curr. Top. Dev. Biol. 4, 37 (1969).
5 M. E. Buckingham, D. Caput, A. Cohen, R. G. Whalen, and F. Gros, Proc. Natl. Acad.
Sci. U.S.A. 71, 1466 (1974).
e F. Bowden- Essien, Dev. Biol. 27, 351 (1972).
7 S. D. Hauschka, C. H. Clegg, T. A. Linkhart, and R. W. Lim, Cell. Biol. 75, No. 2, Part 2,
383 (1977).
8 S. D. Hauschka, Dev. Biol. 37, 329 and 345 (1974).
9 R. L. Seecoff, Am. Zool. 17, 577 (1977).
lo T. J. Kurtti, and M. A. Brooks, Exp. Cell Res. 61,407 (1970).
11 R. Wittmann, J. G. Moser, B. Heinkin, and B. Wolf, Cytobiologie 8, 468 (1974).
12 p. G. Cox, and S. B. Simpson, Jr., Dev. Biol. 23, 433 (1970).
[45]
SKELETAL MYOBLASTS IN CULTURE
513
within 1 hr. Embryos are removed sterilely, one at a time, and transferred
to a 100-mm glass petri plate. Skin and other overlying tissues are removed and discarded and the muscle cleanly dissected and transferred to
a 50-mm petri dish containing a drop or two of a phosphate-buffered balanced salt solution. The pooled muscle is then thoroughly minced with a
pair of sharp curved scissors. During this procedure the tissue is confined
to one edge of the dish by tilting the dish and using the edge of the scissors
to push the mince together. This should be done quickly to avoid desiccation. One quickly learns to gauge the proper degree of mincing (by counting scissor strokes and observing the degree of homogeneity of the
mince). If the tissue is not adequately minced this will become obvious at
the next step, which requires diluting and pipetting themince to a tube (or
flask).
Cell Dissociation (Mechanical)
The mince can be dissociated into single cells either enzymically or

mechanically. The two enzymes commonly employed are pancreatic
trypsin (Difco 1:300) or bacterial collagenase (Worthington Type I or Type
II); both preparations are relatively unpure. Crude trypsin is preferred
because one or more of the contaminating enzymes (presumably DNase)
digests a viscous material present in the mince; TM with more highly
purified trypsin the viscous material remains, trapping cells and clogging
whatever sieve is used to remove undissociated clumps of cells. Similarly,
but for different reasons, crude collagenase is preferred to the crystalline
enzyme. Our own attempts to use highly purified collagenase were singularly unsuccessful, leading us to assume that the efficacy of the crude
enzyme preparation was due to the contaminating proteases rather than to
collagenase activity p e r s e . Of the two enzymes we have found collagenase to be preferable since it yields a population of more viable cells
than trypsin.
A practical procedure for the mechanical dissociation of embryonic
skeletal muscle was first devised by Arnold Caplan several years ago. He
I
communicated the technique to several laboratories, including our own.
This technique has since been published TM as well as a number of variations of the original procedure. 15.,6
We adopted mechanical dissociation for the preparation of primary
cultures during a period when we had difficulty obtaining standardized
13 M. S. Steinberg, Exp. Cell Res. 30, 257 (1963).
14 A. I. Caplan, J. Embryol. Exp. Morphol. 36, 175 (1976).
,5 K. Tepperman, G. Morris, F. Essien, and S. M. Heywood, J. Cell Physiol. 86,561 (1975).
16 j. C. Bullaro, and D. H. Brookman, In Vitro 12, 564 (1976).
514
SPECIFIC CELL LINES
[45]
Fig. 1. Cell filter. The device centers a cotton-plugged polypropylene test tube (Nalge
#3110-0180), with the bottom machined off, inverted in a 40-ml Pyrex centrifuge tube with a
#5 silicone rubber stopper. The centrifuge tube is vented by the insertion of a 20-gauge
hypodermic needle. Nitex or gauze filters (see text) are tightly tied to the flared lip of the
inverted tube and discarded after use.
batches of collagenase. The procedure proved satisfactory and we use it

routinely, as follows:
1. The mince from a pectoral muscle of four 10-day quail embryos is
suspended in 2 ml of complete growth medium and transferred to a sterile
25-ml polycarbonate graduated centrifuge tube.
2. The tube and its contents are subjected to the action of a Vortex
mixer (Vortex-Genie, Model K 550 G) for 30 sec at the highest speed
setting.
3. Large clumps are allowed to settle by gravity, and the supernatant
liquid is withdrawn with a curved-tip Pasteur pipette (Bellco Glass) with
rubber bulb.
4. The cell suspension is filtered through a Nitex mesh lr (20-/~m opening) screen positioned in a 40-ml graduated conical Pyrex centrifuge tube
(see Fig. 1).
17 Tetko Inc., Elmsford, New York.
[45]
515
5. While the first cell suspension is filtering, an additional 2 ml of

complete growth medium are added to the settled clumps that are resuspended by a 3-sec vortexing at half-maximum speed.
6. The suspension is filtered through the same Nitex filter, with continous swirling to prevent clogging the Nitex, and pooled with the first
filtrate.
7. The cell number in the pooled suspension is counted in a
hemocytometer, and primary cultures are established by inoculating approximately 107 cells into each of several gelatin- or collagen-treated
Falcon 100-ram petri plates (see Culture Surface) containing 5 ml of complete growth medium each. (Primary plates are pre-equilibrated for 1-2 hr
in a humidified incubator at 36.5 in an atmosphere of 4% CO2 in air before
use.)
8. Unattached cells and debris are removed by replacing the medium

with 10 ml of complete medium 2 hr after the primary plates are seeded.
9. After 16-18 hr of incubation these primary plates are used to prepare secondary cell suspensions (as described below), which are then
used to set up experimental and control cultures.
The establishment of satisfactory primary cultures requires not only a
brief (ca. 1 hr) transit time between dissection and plating but also a
judicious adjustment of inoculum size. It is difficult to accurately count
cells in mechanically dissociated suspensions that are not monodisperse
but contain many cell clumps. With practice, however, one can obtain
reproducible primaries. One should always monitor primary plates by
microscopic inspection before use. Extremely sparse primaries yield poor
secondary cultures. However, in overloaded primary plates, a higher proportion of multinucleated cells are found the next morning, presumably
due to the early initiation of fusion. Moderate overloading thus selects
against the cell type, i.e., the myoblast, of interest. Grossly overloaded
primaries, even though they may not contain proportionately higher numbers of multinucleated cells to the eye, invariably yield cultures that grow
poorly and contain abnormal-looking cells. Our suspicion is that in such
densely seeded primaries metabolic by-products accumulate, the buffer
capacity is exceeded, and the drop in pH damages the cells.
Cell Dissociation (Enzymic)
Cell yield, by mechanical dissociation, is lower than one obtains enzymically since many of the cells are clumped and consequently lost
during sieving. Although we prefer to establish experimental cultures
from secondary cell suspensions, the reader's experimental design may
require the use of freshly isolated primary cells. Since accurate cell
516
SPECIFIC CELL LINES
[45]
counts are more readily obtained using enzymically dissociated cells, the
collagenase procedure previously used by us is given here:
1. Muscle tissue is dissected and minced as described above (see
"Dissection and Mincing the Tissue") using two or four embryos.
2. The mince is transferred with two 2.5-ml portions of nominally
0.1% collagenase in Puck's saline G ~8into a 25-ml Erlenmeyer flask using
a serological pipette with an orifice adequate to accept the minced fragments (usually 10 ml).
3. The suspended tissue is maintained at 37 for 5 min and subjected to
gentle pipetting with a 10-ml serological pipette to disperse the cells.
4. Enzymic digestion is stopped by adding 5 ml of complete growth
medium at about 5, and the suspension is filtered through three double
layers of cheesecloth (see Fig. 1) into a 40-ml Pyrex conical centrifuge
tube.
5. The filtered suspension is centrifuged at 800 rpm with a bench-top
centrifuge (I.E.C., Model H) and the supernatant liquid removed by
aspiration.
6. The cell pellet is dispersed in 2-5 ml of complete growth medium by
repeated pipetting with a hypodermic syringe equipped with a 20-gauge
5-inch spinal-tap needle.
7. Following a final filtration through a Nitex (Nylon monofilament)
screen with a mesh opening of 10/.~m, the suspension is counted with a
hemocytometer chamber.
8. At this point, primary cultures are started in the same manner as
with mechanically dissociated cells and are used on the following day to
prepare secondary cell suspensions that have been enriched for
myoblasts.
Other investigators prefer to establish experimental and control cultures from such primary suspensions. Under these circumstances enrichment can be achieved by exposing the suspension briefly to culture surfaces to which fibroblasts attach preferentially.19
The same protocol can be used with trypsin as well although cell
viability is somewhat poorer. During the period in which we routin/ely
employed this enzyme we used 0.05% trypsin (N.B.C. or Difco t":300)
made up in Puck's saline G.
Secondary Cell Suspension
We prefer to use cultures from secondary suspensions harvested from
briefly cultured primary cells for the following reasons: (1) We feel that we
is T. T. Puck, S. J. Cieciura, and A. Robinson,J. Exp. Med. 108, 949 (1958).
19D. Yaff,Proc. Nail. Acad. Sci. U.S.A. 61, 477 (1968).
[45]
517
get a more accurate count of viable cells (after plating such secondary
suspensions we see relatively few "floaters"). (2) W e c a n more readily
enrich for myoblasts during the harvest of primary cultures.
The protocol we adopted to prepare secondary cell suspensions from
mechanically dissociated primary cultures is described below. In this procedure we now use crystalline trypsin 2 since it has the singular advantage
of being rapidly inactivated by crystalline soybean trypsin inhibitor 2 in
stoichiometric amounts.
The minimum effective concentration of enzyme is empirically determined. Primary cultures are briefly exposed to the enzyme, and enzyme activity is stopped rapidly. We have also abandoned the practice of
centrifuging down the cells, washing, and resuspending the pellet (see
preceeding page, steps 5 and 6). This step, we find, lowers the cell yield
(either by damaging cells or by the loss of cells in undissociable cell
clumps).
1. The medium is aspirated from primary cultures incubated overnight
and each petri plate rinsed with 10 ml of saline G to remove residual
medium.
2. An iced solution of trypsin (Worthington, 3 crystallized), 5 ml
containing 13/zg/ml in saline G, is delivered into each petri plate.
3. The petri plates are maintained at room temperature and observed
periodically at a magnification of 100 under phase-contrast microscopy.
4. Cultures are gently swirled intermittently. When approximately
half of the cells have either rounded up or detached from the surface,
about 5 min, the enzyme solution is gently pipetted over the culture
surface.
5. The resultant cell suspension is quickly transferred to an iced 40 ml
Pyrex conical centrifuge tube of complete growth medium (0.5 ml for each
primary plate used) containing 0.5 mg/ml of soybean trypsin inhibitor ( 3
crystalline, Worthington).
6. The suspension is mixed well, and the cells are enumerated in a
hemocytometer chamber. Suspensions prepared as described above tend
to be dilute and are generally counted by the procedure used for white cell
counts.
Procedures for the enrichment of myoblasts in cell suspensions are
based on the observation that fibroblasts stretch out and are more tenaciously bound to the culture surface 21 (see Fig. 2). In the above protocol,
differential release of myoblasts is achieved by a controlled, brief trypsinization that leaves most of the fibroblasts bound to the petri plate.-We
2o Worthington Biochemical Co.
zl I. R. Konigsberg, Science 140, 1273 (1963).
518
SPECIFIC CELL LINES
[45]
Fig. 2. Morphological differences between cells in a clone of chick muscle fibroblasts (A)
and a clone of myoblasts (B) photographed on the fourth day of culture using phase-contrast
optics. Since fibroblasts attach more tenaciously they appear to be larger. The remarkable
similarity of cell shape predictably identifies the clonal type.21a
prefer to sacrifice yield for purity o f cell t y p e and can, with care, obtain
cell suspensions in which a m i n i m u m of 96% of the cells are myogenic. 22
Inoculum Size, Medium Composition, and Synchrony
One o f the chief a d v a n t a g e s of muscle cell culture as an experimental
s y s t e m is the high degree of s y n c h r o n y of differentiation that can be
achieved. S y n c h r o n y can be i m p o s e d b e c a u s e cell culture is a closed
s y s t e m in which the proliferating cells eventually deplete the m e d i u m o f
heterologous, exogenous mitogens and lag in the G~ p h a s e o f the cell
cycle. T h e protraction o f G~ is the cue for the initiation o f differentiation.
By adjusting the inoculum size and the volume and composition of the
growth medium, the timing o f differentiation can be manipulated within
limits. The limits are i m p o s e d b y the establishment, in static cultures of a
gradient o f depletion in the immediate vicinity of the cells. This can be
21a I. R. Konigsberg, Scientific American, 211, 61-66 (1964).
p. A. Buckley, and I. R. Konigsberg, Dee. Biol. 37, 186-193 (1974).
22
[45]
519
overcome to a certain extent by continuous, gentle perturbation of the

cultures a or by scheduled refeeding. Under such conditions the period of
rapid log growth can be extended, but short of a regime of continuous
perfusion, z3 increasing cell density eventually exceeds the practical limits
of any refeeding schedule. The protocols employed to achieve synchrony,
therefore, represent compromises of one sort or another.
In order to cleanly divide the stages of culture development into a pure
rapid proliferative stage and a stage in which myogenic fusion is initiated
and proceeds rapidly, we employ the following protocol:
1. After determining the number of cells per milliliter in a secondary
cell suspension (usually about 8 104; total volume per primary plate,
3 ml), an aliquot is serially diluted to provide a suspension in which 0.1 or
0.2 ml contains the inoculum per petri plate to be delivered (2 104 cells).
2. Cells are inoculated into each of a series of collagen- or gelatintreated petri plates containing 2 ml of high-growth medium. Culture
dishes containing medium are equilibrated in the incubator before use.
3. Following incubation for 16-18 hr, the medium is aspirated, replaced with 3 ml of high-growth medium, and not replaced thereafter.
In cultures established on this schedule, fusion--in the form of short,
stubby multinuclear cells (two to six nuclei per cell)--is first observed on
day 3. 22 Following the same protocol but using a low-growth medium
(10% serum, 1% embryo extract) instead of high-growth (10%'embryo
extract, 15% serum), fusion is initiated 24 hr earlier (Fig. 3) and still
earlier by further depleting the low-growth medium (see "Conditioned
Media").
Using this tactic of adjusting the rate of depletion of an initially
growth-promoting medium, fusion, once initiated, occurs rapidly but certainly not at the maximum rate. However, greater synchrony can be
achieved by employing a "step-down" medium change; replacing a highgrowth medium with a defined, synthetic medium containing no serum or
embryo extract (see "Media"), with the plate rinsed once with the defined
medium before refeeding. 24
The suggested inoculum size yields satisfactory cultures using secondary suspensions of quail myoblasts. In our experience, inocula 2-3 times
larger are required to obtain comparable cultures of chick myoblasts.
However, even a cursory examination of the current literature indicates
that many investigators use inocula many times heavier than suggested
here, ranging to 1-2 106 cells per 5-cm petri dish and more. This degree
22 I. R. Konigsberg, in "Pathogenesis of Human Muscular Dystrophies" (L. P. Rowland,
ed.), pp. 779-798. Excerpta Med. Found., Amsterdam, 1976.
24 C. P. Emerson, Jr., in "Pathogenesis of Human Muscular Dystrophies" (L. P. Rowland,
ed.), pp. 799-809. Excerpta Med. Found., Amsterdam, 1976.
520
SPECIFIC CELL LINES
[45]
of cell crowding in culture can hardly be considered physiological. Indeed, our experience has been that a smaller percentage of such inocula
actually attach. Synchrony is completely lost in such cultures, myogenic
fusion being initiated during the first 24 hr in culture, and, as one might
expect, cell proliferation is diminished by cell crowding. If, as we suspect,
the use of such heavy inocula reflects an inability to obtain cell survival
and proliferation at lower cell density, the investigator should reexamine
the cell dissociation techniques, media preparation, and incubation conditions that are being used.
The criteria for the adequacy of these parameters are the plating efficiency, colony size, and differentiation of cells plated at clonal density
(200-400 cells per 5-cm dish). Hating efficiencies of about 20% (chick) and
40% (quail) are acceptable.
Media (General Considerations)
Media composition, preparation, and sterilization are discussed elsewhere in this volume [1] [2] [5]. This section will deal with just those
media requirements peculiar to freshly isolated cells and muscle cells
particularly. These cells, unlike cell lines, are neither adapted to nor selected for the culture environment. They are, therefore, more sensitive to
variations in quality of media components than are established cell lines.
I00
80
,7 60
r~
N20
DAYS IN CULTURE
Fig. 3. The increase in percentage of nuclei in nascent multinuclear muscle fibers as a

function of time in culture, using media of differing growth-supporting properties. Each
point represents the average of two cultures each inoculated with 104 cells of a secondary
suspension of quail myoblasts. Fusion is initiated earliest in conditioned low-growth medium, CM 1-10 (A); then in freshly prepared low-growth medium, FM 1-10 (O); and finally in
high-growth medium, FM 10-15 (11). (From Wm. M. Sutherland, Doctoral Thesis, University of Virginia 1977.)
[45]
521
1. We have been forced for logistic considerations to use commercially prepared media components, such as MEM, F10, and FI2. Although these products are satisfactory they are not quite up to the standard of these components as prepared by ourselves from reagent-grade
chemicals.
2. All of the sera that we use are pretested by us, using the ability of
each serum when incorporated in our "high-growth" medium to support
growth and differentiation at clonal density (see above). Generally, one of
two (rarely four) samples of horse serum proves satisfactory, and we
order and store at - 6 0 sufficient serum to last the year.
3. Unlike cell lines, freshly isolated myoblasts require, in addition to
horse serum, the incorporation of a saline extract of chick embryos. In the
absence of embryo extract, little if any proliferation occurs. The protocol
(see "Appendix") for preparing this extract is the result of exhaustive
empirical testing of the mincing procedures, extraction time, centrifugation speeds and times, as well as filtration procedures. We have employed
this protocol, with only minor changes, over the past 10 years with good
results. The extract has a useful life of only 2 weeks even when stored at
-60 and therefore is prepared on a regular schedule. Alternative methods
of preparing extract are employed by other investigators. These alternatives range from essentially the same procedures as ours, 2~to substituting
a brief (20-min) blending step (Waring Blendor) z6 for the mincing procedure we use, to forcing the embryos through a stainless-steel screen followed by freeze-thawing and centrifugation. 27
Constituents of the Medium
The media used in most laboratories in which muscle cell culture is

practiced consists of: (1) one of the standard synthetic mixtures of amino
acids, vitamins, and cofactors (such as Eagle's MEM), (2) serum (usually
equine), (3) embryo extract, and (4) a mixture of antibiotics.
Normally a Ca 2+ ion concentration of 1.4 mM is employed. By reducing the calcium ion concentration to 35 ~M, myoblast fusion is reversibly
blocked without affecting the rate of cell proliferation. 28'29 Under these
conditions the mononucleated myoblasts deplete the medium, differentiate, and withdraw from the cell cycle 29 as mononucleated myocytes.
25 N. K. White, P. H. Bonner, D. R. Nelson, and S. D. Hauschka, Dev. Biol. 44, 346 (1975).
26 S. D. Hauschka, personal communication.
2r M. C. O'Neill and F. E. Stockdale, J. Cell Biol. 52, 52 (1972).
zs B. Paterson and R. C. Strohman, Dev. Biol. 29, 113 (1972).
29 C. P. Emerson, Jr. and S. K. Beckner, J. Mol. Biol. 93,431 (1975).
522
SPECIFIC CELL LINES
[45]
Reduction of the calcium ion concentration is accomplished by either of

two procedures:
1. Supplementing low-calcium synthetic media with dialyzed embryo
extract and serum a (or reducing the calcium ion concentration by ionexchange chromatography31).
2. Chelating calcium in the medium by the addition of EGTA. The
precise concentration of EGTA (made up in the synthetic low-Ca 2+ media
stock, adjusting pH to 7.2) must be determined empirically for each batch
of medium since the concentration of calcium in the embryo extract
varies. This is done by adding back Ca 2+ (from a 25 mM stock solution) to
the level that minimizes cell detachment from the plate surface while still
completely inhibiting cell fusion. Levels of EGTA of approximately 1.9
mM are generally adequate.
The Synthetic Nutrient Component
The synthetic nutrient mixtures most frequently employed in muscle

cell culture are: (1) Eagles MEM, z2 (2) Ham's F10, z3 and (3) Puck's NC
(also called NCI). z4 Other synthetic formulations, as well as mixtures of
two or more, have also been used. Although it is assumed that such usage
is based on comparative tests, it is not obvious that any of the alternatives
to the three most frequently used synthetic nutrients offer any unique
advantage in cultures seeded at mass density. Indeed, one should not
expect to detect differences except when comparing clonal with highdensity cultures. Cells at clonal density are far more sensitive to nutritional deficiencies (as well as to inhibitory concentrations of required
nutrients). Such conditions, one would expect, would be rapidly adjusted by metabolic cooperation at high cell density. On the other hand,
media devised to support clonal growth may be rapidly depleted of one
or more essentia~ components present in low concentration. In fact,
chick myoblasts at mass density grow well in NC-based medium but do
very poorly on F10 medium, probably due to the lower Ca 2+ ion concentration in F10. 2 Conversely, growth and development at clonal
density is far superior in medium containing F10 than in medium containing NC.
ao A. Shainberg, et al., Exp. Cell Res. 58, 163 (1970).
al E. Ozawa, Biol. Bull. (Woods Hole, Mass.) 143, 431 (1972).
az H. Eagle, Science 130, 432 (1959).
aa R. G. Ham, Exp. Cell Res. 29, 515 (1963).
a4 p. I. Marcus, S. J. Cieciura, and T. T. Puck, J. Exp. Med. 104, 615 (Sec P. 616) (1956).
[45]
Variations of Media Composition
523
and the Control of Proliferation
The standard medium used in this laboratory was developed to support the growth and differentiation of avian myoblasts at clonal density
applying the criteria of plating efficiency, colony size, and per cent differentiated clones. For that reason it is a "high-growth" medium, supporting
a rapid rate of cell proliferation. In developing this medium a number of
parameters were systematically varied, the most important of which, it
turned out, were the proportions of serum and embryo extract in the
medium. The concentration of equine serum that proved optimal was 15%
(v/v). We found also that for chick myoblasts the maximum concentration
of embryo extract was 5% (v/v) and that higher concentrations were actually inhibitory (see also Stockdale3~). In adapting this medium for quail
myoblast culture 3 we found that Eagle's MEM was as effective a nutrient
base as F10 and that the quail cells, unlike the chick, would tolerate and
do better on 10% embryo extract medium.
The high-growth ( " F M 10-15") medium is currently employed for
cells at both clonal and mass density (see above) and consists of:
Eagle's MEM (with Earle's salts) 36
Equine serum
Embryo extract
Penicillin-streptomycin stock 37
Fungizone stock 3s
740
150
100
10
2.5
ml
ml
ml
ml
ml
By varying the ratio of serum and embryo extract, a low-growth ( " F M

1-10") medium was derived 29that limits proliferation and promotes earlier
differentiation (fusion and/or the initiation of myosin synthesis). This medium, when appropriately "conditioned" to further reduce the concentration of mitogens, promotes still more limited proliferation and correspondingly earlier differentiation.
Finally, a completely defined synthetic medium (Emerson's F12
[PVP] 24) that contains neither serum not embryo extract has been devised.
It is prepared as follows.
33 F. E. Stockdale, in "Regulation of Cell Proliferation and Differentiation" (W. W. Nichols
and D. G. Murphy, eds.), pp. 165-176. Plenum, New York, 1977.
36 This synthetic medium is modified to contain Eagle's nonessential amino acids (see
Eagle 3~) and contains 1.2 g/liter of NaHCO3 rather than the concentration normally employed.
37 p and S stock contains 106 units of penicillin G (potassium) and 500 mg streptomycin
sulfate per 100 ml of P-C chick (see Pannett and Compton373).
37~ C, A. Pannett and A. Compton, Lancet 206, 381 (1924).
as Fungizone stock contains 50 mg amphotericin B (Fungizone ) in 62.5 ml of distilled H~O.
524
SPECIFIC CELL LINES
[45]
1. Ham's F12 is prepared to contain, in addition to all of the normal

components of F12, the following:
Eagle's essential amino acids in the same concentration as in MEM
146 mg glutamine per liter
221 mg CaC12 2 HzO per liter
2. In place of the concentration of sodium bicarbonate normally employed in F12, this solution is made to contain 1.2 g/liter. The complete
medium has the following composition:
Ham's F12 (as modified above)
Polyvinylpyrrolidone 360 (Sigma)
Penicillin-streptomycin stock
Fungizone stock
990
500
10
2.5
ml
mg
ml
ml
This defined synthetic medium does not support cell proliferation but
will support rapid, synchronous fusion (or myosin synthesis in low-Ca 2+,
fusion-blocking F12[PVP]) when cultures are switched from growth to
defined medium. ~4 This indicates that no constituent of either serum or
embryo extract is required for either fusion or cell-type specific synthesis.
It suggests, in fact, that these heterologous components of the medium,
by reason of stimulating proliferation, prevent or delay the initiation of
differentiation. This suggestion is also supported by the observation that
fusion is inhibited by continuous perfusion with high-growth medium but
not with F12(PVP). 23
Conditioned Media
Conditioning freshly prepared media by subjecting it to the metabolic
activities of cultured cells could conceivably alter media composition in a
number of ways, for example, by the depletion of media constituents or
the accumulation of cell products. Such media were employed in developing the art of cell culture (see this volume [5]) and have also proven useful
in the development of the muscle cell culture system. Specifically, conditioned media have been used: (1) to develop adequate cloning procedures, 21 (2) to study and control the time of initiation of differentiation, 3,2z'2a and (3) to identify subpopulations of myoblasts in early
embryonic development. 25,z9 Further analysis of two of these media has
led to some understanding of the nature of their alteration and eventually
to better-defined substitutes for the conditioning procedures. To a large
extent it is no longer necessary to employ conditioned media to clone
myoblasts (from embryos o f advanced stages) since a gelatin- or
39 N. K. White and S. D. Hauschka, Exp. Cell Res. 67, 479 (1971).
[45]
525
collagen-coated culture surface adequately substitutes for the altered medium. 4,41 In addition synchronous differentiation now can be achieved by
a step-down to defined medium (see above) rather than by employing a
nit rlogen-depleted medium.
H o w e v e r , should the use of any of these conditioned media seem
advantageous to the investigator's aims he would be advised to follow the
published protocols in minute detail, at least initially. All of the protocols
cited above were derived empirically. Media conditioned by each of the
three protocols elicit quite a different response from the cultured myo-.
blast. These protocols differ principally with respect to cell type, population size, and duration of conditioning. The different results achieved with
each most probably reflect a different balance of the large number of
metabolically generated changes that are occurring. Very brief exposure
to moderate numbers of cells might simply promote the accumulation in
the medium of low-molecular-weight biosynthetic products that leak into
the medium. 42 At the other extreme of population size and duration the
over riding effect may reflect the depletion of mitogenic medium
components.
Collagen P r e t r e a t m e n t of the Culture Dish
Myoblasts adhere and stretch out more satisfactorily to a collagen or
gelatin substratum than to the bare polystyrene TC dish, as supplied
by the manufacturer. Furthermore, unless provided with such an attachment surface, the percentage of clones that differentiates is low. The few
clones that do differentiate are small and contain relatively few muscle
fibers. 4'41'43 In mass cultures established from cell suspensions that are
not enriched for myoblasts, differentiation is equally normal in collagentreated and untreated petri plates. Although not strictly required, the use
of pretreated plates is advantageous since it delays the detachment of the
cell sheet that eventually occurs in long-term mass culture.
Collagen can be applied by raising the ionic strength of a concentrated
solution of the protein in 0.15 M acetic acid. A small measure volume is
then spread on the petri plate surface, which is subsequently rinsed. 4~
Solutions of gelatin in distilled water also can be spread and simply allowed to dry. 44
40I. R. Konigsberg and S. D. Hauschka, in "Reproduction: Molecular, Subceltular and
Cellular" (M. Locke, ed.), pp. 243-289. Academic Press, New York, 1965.
41 S. D. Hauschka and I. R. Konigsberg, Proc. Natl. Acad. Sci. U.S.A. 55, 119 (1966).
42H. Eagle and K. Piez, J. Exp. Med. 116, 29 (1962).
43I. R. Konigsberg, Proc. Natl. Acad. Sci. U.S.A. 47, 1868 (1961).
44S. D. Hauschka, in "Growth, Nutrition and Metabolism of Cells in Culture" (G. H.
Rothblat and V. J. Cristofalo, eds.), Vol. II, pp. 67-130. Academic Press, New York, 1972.
526
SPECIFIC CELL LINES
[45]
To insure more uniform distribution and to avoid the tedium of spreading we now flood TC petri plates with a larger measured volume, 1.4 ml,
of more dilute collagen or gelatin. 3,45 The protein is allowed to adsorb to
the surface overnight (about 18 hr) at room temperature in an atmosphere
equilibrated with the protein solvent used. The collagen solution applied is
made up by adding 40/zl of 3.76% NaCl to 1.0 ml of a purified collagen
stock containing 40/xg collagen per milliliter. Gelatin solutions contain 100
/zg/ml of a good bacteriological grade of gelatin (Oxoid) in distilled water.
After aspirating the excess, plates are rinsed twice with distilled water,
placed in a desiccator over silica gel, and stored at room temperature until
used.
Our procedure for preparing and purifying collagen has been published. 3 It is more exhaustive than need be, and our earlier procedure 43 is
perfectly adequate. We find it simpler to prepare and handle collagen
sterilely. However, we have sterilized collagen-treated petri plate surfaces under UV light and found them perfectly satisfactory. Since gelatin
solutions can be autoclaved the investigator may prefer this alternative.
We have found no difference between the native and denatured molecule
in routine culturing. When cultures are to be subjected to constant 23 or
intermittent 3 perturbation, however, myoblasts seem to adhere better to
the collagen substratum.
Appendix: Procedure for Preparing Embryo Extract
1. Incubate fertile chick eggs for 12 days. Candle and discard infertile
eggs and eggs containing dead embryos.
2. Cut a circular hole in the blunt end of the egg with a pair of curved
scissors and remove the embryo.
3. Cut off and discard that portion of the head anterior to the posterior
border of the eyes. Make a longitudinal cut through the body wall from
the cloaca to the base of the neck. Collect the embryos in a 150-mm petri
plate.
4. When all of the embryos have been collected, decant the blood and
empty the petri plates onto paper toweling. Transfer the embryos to a
large square of double-thickness cheesecloth. Suspend the cheesecloth
" b a g " in a beaker and wash with several changes of P-C chick 3ra (until
most of the blood is removed).
5. Place the cheesecloth on several thicknesses of paper toweling (to
drain) and transfer the embryos to a tared beaker and weigh.
45 I. R. Konigsberg, in "Chemistry and Molecular Biology of the Intercellular Matrix"
(E. A. Balazs, ed.), Vol. 3, pp. 1779-1810. Academic Press, New York, 1970.
[46]
HORMONE-PRODUCING
PITUITARY
CELLS
527
6. Run embryos through a machined Lucite Latapie mincer with a

volume of P-C chick equal in milliliter to the weight in grams of the embryos. (Hold back a small quantity of this measured amount of P-C chick.
Weigh out 4 mg [or 1080 U] of hyaluronidase for every 100 g of embryos
and add to this P-C chick.)
7. Transfer the minced embryos to a large beaker, add the
hyaluronidase, and stir at cold room temperature (4--5) for 1 hr.
8. Distribute the extract to 90-ml ultracentrifuge tubes and centrifuge
for 3 hr in the 21 rotor in the Spinco Model L ultracentrifuge at 20,500 rpm
(40,000g) at 1-2 .
9. Decant and collect the supernatant after centrifugation.
10. The extract is clarified by pressure filtration through a 142-ram
diam fiberglass prefilter over a 0.45-/xm millipore filter of the same diameter held in a large-volume "pancake" filter holder (Millipore). The supernatant collected in step 9 is transferred to an 800-ml minireservoir (Amicon) and transferred through heavy-duty silicone rubber tubing through
the filtering system.
11. The filtered extract is recentrifuged as in step 8 (above) overnight
(ca. 18 hr), decanted, and stored at - 6 0 . Sterilization by filtration is performed after the complete medium is made up.
[46] C l o n a l S t r a i n s o f H o r m o n e - P r o d u c i n g
P i t u i t a r y Cells
By ARMEN H. TASHJIAN, JR.
Clonal strains of pituitary tumor cells that synthesize and secrete prolactin, growth hormone and adrenocorticotropic hormone (ACTH) have
been established and serially propagated in culture for periods of up to 15
years. The rates of biosynthesis of the specific hormonal peptides by these
clonal strains are high (up to 10% of total protein synthesis in appropriately stimulated cells), and they respond in culture to many of the same
regulatory factors as normal pituitary cells do in situ. However, because
certain of these regulatory factors, such as the hypothalamic peptides,
affect more than one cell type in the intact pituitary gland, it is not possible to perform unambiguous mechanistic studies with hemi-pituitary
fragments or mixed primary cultures of freshly dispersed pituitary cells.
Therefore, strains of homogeneous populations of functional cells serve as
useful model systems for determining the mechanisms of action of factors
that regulate the release and synthesis of prolactin, growth hormone, and
ACTH.

ISBN 0-12-181958-2
528
SPECIFIC CELL LINES
[46]
E s t a b l i s h m e n t o f P i t u i t a r y Cell S t r a i n s
T h i s s u m m a r y is n o t i n t e n d e d as a r e v i e w o f all f u n c t i o n a l p i t u i t a r y cell
s y s t e m s n o w in u s e , b u t r a t h e r a s a d e s c r i p t i o n o f cells n o w a c t i v e l y b e i n g
s t u d i e d in this l a b o r a t o r y . T h e s e i n c l u d e t h e r a t G H c e l l s a n d m o u s e
A t T 2 0 / D 16 cells.
A variety of cultures from a growth hormone- and prolactin-producing
r a t p i t u i t a r y t u m o r , M t T / W 5 , 1 w e r e e s t a b l i s h e d in 1965. 2 B e c a u s e t h e
organ-specific function originally characterized was growth hormone prod u c t i o n , t h e s e cells b e c a m e k n o w n in t h e l i t e r a t u r e as " G H c e l l s . " I t w a s
l e a r n e d s u b s e q u e n t l y t h a t c e r t a i n G H cell s t r a i n s a l s o p r o d u c e p r o l a c t i n . 3
Cells w e r e a d a p t e d t o g r o w t h in v i t r o b y a l t e r n a t e c u l t u r e a n d a n i m a l
p a s s a g e . 2 D i s p e r s e d t u m o r c e l l s w e r e a l l o w e d to a t t a c h a n d g r o w in p l a s tic c u l t u r e d i s h e s f o r p e r i o d s o f s e v e r a l ~lays to a b o u t 1 m o n t h . T h e
s u r v i v i n g a t t a c h e d cells w e r e t h e n h a r v e s t e d a n d i n j e c t e d i n t o r a t s o f t h e
W i s t a r - F u r t h s t r a i n . W h e n a n e w t u m o r d e v e l o p e d , it w a s r e m o v e d a n d a
s e c o n d g e n e r a t i o n o f c u l t u r e - d e r i v e d t u m o r cells w a s e s t a b l i s h e d in vitro.
This process of alternate passage between culture and animal was repeated
s e v e r a l t i m e s in o r d e r to s e l e c t f o r t u m o r cells t h a t w o u l d a t t a c h a n d g r o w
r e a d i l y in t h e c u l t u r e e n v i r o n m e n t . 4
S e v e r a l e p i t h e l i a l cell s t r a i n s h a v e b e e n c l o n e d , z a n d s o m e h a v e b e e n
m a i n t a i n e d in c o n t i n u o u s c u l t u r e f o r a s l o n g a s 10 y e a r s w i t h o u t loss o f
hormone production. Table I gives the characteristics of several clones of
G H cells, z'a'5-~5 T h e GH1 a n d GH3 s t r a i n s a r e a v a i l a b l e f r o m t h e A m e r i c a n
T y p e C u l t u r e C o l l e c t i o n (12301 P a r k l a w n D r i v e , R o c k v i l l e , M a r y l a n d ) ;
t h e C e l l R e p o s i t o r y N u m b e r s f o r t h e s t r a i n s a r e C C L 82 a n d C C L 82.1,
respectively.
1 H. Takemoto, K. Yokoro, J. Furth, and A. I. Cohen, Cancer Res. 22, 917 (1962).
2 A. H. Tashjian, Jr., Y. Yasumura, L. Levine, G. H. Sato, and M. L. Parker, Endrocrinology 82,342 (1968).
3 A. H. Tashjian, Jr., F. C. Bancroft, and L. Levine, J. Cell Biol. 47, 61 (1970).
4 V. Buonassisi, G. Sato, and A. I. Cohen,Proc. Natl. Acad. Sci. U.S.A. 48, 1184 (1962).
5 A. H. Tashjian, Jr., N. J. Barowsky, and D. K. Jensen,Biochem. Biophys. Res. Commun.
43, 516 (1971).
e T. F. J. Martin and A. H. Tashjian, Jr., Biochem. Actions Horrn. 4, 269 (1977).
7 A. Schonbrunn and A. H. Tashjian, Jr., J. Biol. Chem. 253, 6473 (1978).
8 H. H. Samuels, J. S. Tsai, and R. Cintron, Science 181, 1253 (1973).
9 H. H. Samuels, J. S. Tsai, J. Casanova, and F. Stanley, J. Clin. Invest. 54, 853 (1974).
10H. H. Samuels, Z. D. Horwitz, F. Stanley, J. Casanova, and L. E. Shapiro, Nature
(London) 268, 254 (1977).
H D. K. Biswas, J. Lyons, and A. H. Tashjian, Jr., Cell 11,431 (1977).
12p. S. Dannies and A. H. Tashjian, Jr., J. Biol. Chem. 248, 6174 (1973).
13p. M. Hinkle and A. H. Tashjian, Jr., J. Biol. Chem. 248, 6180 (1973).
14L.-Y. Yu, R. J. Tushinski, and F. C. Bancroft, J. Biol. Chem. 252, 3870 (1977).
15 U. I. Richardson, J. Cell. Physiol. 88, 287 (1976).
[46]
529
HORMONE-PRODUCING PITUITARY CELLS
'~,
,~:=
~,~.
:=
%
J
,.J
I:::
a~
0
.<
'~~:
~.~.~
.~.~ ~
~o
"~
._
[-
0
Z
0
t~
0
o
h.
h,
<
.,<
(J
0
.<
Z
0
~J
Z
~0
m
530
SPECIFIC CELL LINES
[46]
H o r m o n e - p r o d u c i n g mouse pituitary t u m o r cells have also been established in culture. The AtT20/D16 cells are a clonal strain isolated in 1962
by Yasumura TM from a radiation-induced pituitary t u m o r in the mouse.17
These cells have continued to produce large amounts o f A C T H during
serial propagation in culture for 15 years. TM Table II gives the functional
characteristics o f this strain, la-z3 The AtT20/D16 strain is available from
the American Type Culture Collection (Cell Repository N u m b e r C C L 89).
Methods of Culture
Type o f Culture. G H cells can be grown as monolayers on plastic or
glass surfaces, 2 in roller bottles, or in suspension culture. 24
M e d i u m . G H cells are not fastidious in their medium requirements.
They are usually grown in monolayer in this laboratory in H a m ' s nutrient
mixture F1025 supplemented w i t h 1 5 % horse serum and 2.5% fetal calf
serum at 37 _ 0.5 in a humidified atmosphere o f 5% CO2 and 95% air. 2
They also can be grown in H a m ' s nutrient mixture F12, 26 Eagle's minimum essential medium, 27 and D u l b e c c o ' s modified Eagle's medium 28
supplemented with horse serum (5-15%) and 2.5% fetal calf serum. In
addition, for experiments of relatively brief duration (1-48 hr) requiring
the absence of serum, G H cells can be maintained in a functional
hormon-producing and hormone-responsive state in either N e u m a n and
Tytell's serumless medium 29 or in H a m ' s nutrient mixture F10 supplemented with lactalbumin hydrolysate (5 g/liter). Hayashi and Sato have
reported growth of GH3 cells in chemically defined medium supplemented
with certain " h o r m o n e s . ''3
For spinner cultures, modified Eagle's medium for suspension 31 supplemented with 15% horse serum and 2.5% fetal calf serum is used.
16y. Yasumura, Am. ZooL 8, 285 0968).

17j. Furth, Recent Prdg. Horm. Res. ll, 221 (1955).
is U. I. Richardson, Endocrinology 102, 910 (1978).
19R. E. Mains and B. A. Eipper, J. Biol. Chem. 251, 4115 (1976).
2oR. E. Mains, B. A. Eipper, and N. Ling, Proc. Natl. Acad. Sci. U.S.A. 74, 3014 (1977).
zl G. Giagnoni, S. L. Sabol, and M. Nirenberg, Proc. Natl. Acad. Sci. U.S.A. 74, 2259
(1977).
22j. L. Roberts and E. Herbert, Proc. Natl. Acad. Sci. U.S.A. 74, 5300 (1977).
23E. Herbert, R. G. Allen, and T. L. Paquette, Endocrinology 102, 218 0978).
24F. C. Bancroft and A. H. Tashjian, Jr., Exp. Cell Res. 64, 125 (1971).
25R. G. Ham, Exp. Cell Res. 29, 515 0963).
26R. G. Ham, Proc. Natl. Acad. Sci. U.S.A. 53, 288 0965).
27H. Eagle, Proc. Soc. Exp. Biol. Med. 89, 362 0955).
2s j. D. Smith, G. Freeman, M. Vogt, and R. Dulbecco, Virology 12, 185 (1960).
29R. E. Neuman and A. A. Tytell, Proc. Soc. Exp. Biol. Med. 104, 252 0960).
s o I. Haysahi and G. H. Sato, Nature (London) 259, 132 (1976). See also [6].
31 H. Eagle, Science 130, 432 (1959).
[48]
.=.
o
%
m
E
J
m
Ii
m
o
o
m
t,
t...
e-~
il ..~
O
E,
O
,-r
o
<
Z
0
;>-.
m
[..
<
H ,:.:,
r=
o
..
531
532
SPECIFIC CELL LINES
[46]
Subculture. G H cells are released from substrate for subculture by

brief (2-5 min) incubation at 37 with Viokase (0.1%) or trypsin (0.1%) in
either phosphate-buffered saline or nutrient mediun lacking serum, z Split
ratios of 1 : l0 to 1:40 are used. Plating efficiencies are 30-50%.
G r o w t h R a t e s . In monolayer culture in H a m ' s F10 medium, the population doubling time determined from counts of cell n u m b e r or total cell
protein is 45-52 hr. G H cells do not form confluent monolayers, but after a
period of logarithmic growth, they reach a plateau density at which the
cells c o v e r 60-80% of the available surface. 2 Cell division continues with
viable cells being shed into medium at about their rate of division. Hormone production occurs during both exponential growth and plateau density.2"32
Values for alkali-soluble protein and intracellular volume for G H cells
grown in monolayer were 2.1 mg cell protein per 107 cells and 17/~1 per 107
cells, respectively. 3'~
Characteristics of G H Cells
Stability. The rates o f growth and hormone production by G H cells
have generally remained unchanged o v e r many years o f continuous propagation. H o w e v e r , some changes have been noted: for example, GH3 cells
that initially produced about equal quantities of growth hormone and
prolactin 3 now synthesize 5-10 times more prolactin than growth hormone, and in the GH4C, strain derived from GH3 cells, 34 the ratio o f
prolactin to growth hormone is even greater. It is probable that such
changes in the ratio of hormone production reflect some unknown current
environmenthl alteration, such as medium composition, because GH3
cells that are thawed from liquid nitrogen change their ratio o f prolactin to
growth hormone synthesis o v e r a period o f several weeks from about 1 : 1
to the same high ratio (about 10 : 1) as serially propagated GH3 cells.
K a r y o t y p e . G H cells are aneuploid with a modal n u m b e r of chromosomes ranging from 69 to 76. 35 The modal c h r o m o s o m e n u m b e r for the
AtT20/D16 cells is 76. '8
H o r m o n e Production. As described in Table I, most o f the G H cell
strains, except GH,2C1 and GC, produce both growth hormone and prolactin although the ratios of hormone synthesis among different clones
varies considerably, with GH3 and GC being the highest producers of
32O. P. F. Ciausen, K. M. Gautvik, and T. Lindmo, Virchows Arch. B 23, 195 (1977).
a3T. F. J. Martin and A. H. Tashjian, Jr., J. Biol. Chem. 253, 106 (1978).
34A. H. Tashjian, Jr., P. M. Hinkl, and P. S. Dannies, in "Endocrinology" (R. O. Scow,
ed.), p. 648. Excerpta Med. Found., Amsterdam, 1973.
35C. Sonnenschein, U. I. Richardson, and A. H. Tashjian, Jr., Exp. Cell. Res. 61,121 (1970).
[46]
533

TABLE III
RESPONSES OF GH CELLSTO HORMONESAND DRUGSa
Signal
TRH
Effect
A c u t e actions (0-30 min)
Binds to membrane receptors

Quenches membrane fluorescence
Increases cyclic AMP
Releases stored PRL
Enhances uridine uptake
Chronic actions (4-48 hr)
Increases PRL synthesis
Decreases GH synthesis
Decreases TRH receptor number
Decreases uridine uptake
Morphologic changes
Cortisol
Estrogens
Thyroid hormone
Somatostatin
Increases GH synthesis
Decreases PRL synthesis
Increases TRH receptor number
17/3-estradiol increases PRL and
decreases GH production
Antiestrogens increase PRL synthesis
Increases cell growth
Increases GH synthesis
Decreases TRH receptor number
Binds to specific receptors
Decreases GH release
Decreases PRL release
Bromocriptine
Decreases PRL synthesis and release
Prostaglandins
Increases PRL synthesis

No change in GH production
Cations
References
Role of calcium in hormone release and

synthesis
13.40--43
44
45--47
6,42
33.48
5,6.12,49.50
5.6,51
6,52,53
33,48
51,54
10,14,36,55,56
12,36
43
51,57.58
57
8
8--10
59
7
7
7
51,60
61
61
62,63
a Abbreviations: TRH = thyrotropin-releasing hormone; PRL = prolactin; GH = growth

hormone; bromocriptine = 2-bromo-ct-ergokryptine (CB-154); cyclic AMP = adenosine
3' : Y-monophosphate.
g r o w t h h o r m o n e a n d GH4C1 p r o d u c i n g the m o s t p r o l a c t i n . N o c l o n e y e t
isolated synthesizes only prolactin.
T h e specific r a t e s o f g r o w t h h o r m o n e a n d p r o l a c t i n s y n t h e s i s i n G H
cells c a n b e m o d u l a t e d b y a v a r i e t y o f s t e r o i d a n d p e p t i d e h o r m o n e s as
well as b y d r u g s a d d e d to the c u l t u r e m e d i u m (see b e l o w ) . A t m a x i m u m
s t i m u l a t i o n w i t h cortisol a n d t h y r o i d h o r m o n e , GH3 or G C cells c a n pro-
534
SPECIFIC CELL LINES
[46]
duce as much as 50-400/zg growth hormone/mg cell protein/24 hr. 14,36,37In

comparable experiments at maximum stimulation with thyrotropinreleasing hormone, GH4C1 can produce 40-150 /.tg prolactin/mg cell
protein/24 hr. T M Under these conditions, the fraction of total protein
synthesis represented by either growth hormone or prolactin synthesis is
2-10%.
At least 30 serial clones of GH3 ceils have been isolated, each derived
from a single cell of the preceding clone, and each clone produces both
growth hormone and prolactin. This finding suggests strongly, but does
not prove, that both hormones are produced by the same cell.
Somatic Cell Hybrids. The GH12C1 clone was hybridized with Clone 1D
mouse fibroblasts. 38 Six hybrids clones were analyzed in detail and none
produced growth hormone. These hybrid lines were subsequently found
to exhibit unexpected resistance to X-irradiation. 39 Subsequently, GH3
and GH4C1 cells have been hybridized with mouse fibroblasts (Thompson,
Dannies, and Tashjian, unpublished data), and a variety of clones have
been isolated; some produce no hormone, some produce only prolactin,
and some produce both growth hormone and prolactin.
Modulation of Hormone Biosynthesis and Release in GH Cells. Table III
summarizes the more prominently studied responses of GH cells to hormones and drugs known to affect the function of mammotrophs and
somatotrophs in the normal pituitary gland in situ. 5-10,12-14,33,36.40-63
Although less extensive studies have been performed with AtT20/D16
cells, they have been used to examine the negative feedback of adrenoglucocorticoid hormones on ACTH release, 23"e4the ACTH release induced
by corticotropin-releasing factor, 23 the ultrashort loop self-regulation of
ACTH release by ACTH itself, is and the mechanism by which adrenoglucocorticoids enter target cells (using the related AtT20/D-1 cell
strain). 65,66
36 F. C. Bancroft, L. Levine, and A. H. Tashjian, Jr., J. Cell Biol. 43, 432 (1969).
37 j. A. Martial, J. D. Baxter, H. M. Goodman, and P. H. Seeberg, Proc. Natl. Acad. Sci.
U.S.A. 74, 1816 (1977).
as C. Sonnenschein, U. I. Richardson, and A. H. Tashjian, Jr., Exp. Cell. Res. 69, 336 (1971).
39 j. B. Little, U. I. Richardson, and A. H. Tashjian, Jr.,Proc. Natl. Acad. Sci. U.S.A. 69,
1363 (1972).
40 p. M. Hinkle, E. L. Woroch, and A. H. Tashjian, Jr., J. Biol. Chem. 249, 3085 (1974).
41 p. M. Hinkle and A. H. Tashjian, Jr., Endocrinology 97, 324 (1975).
42 p. S. Dannies and A. H. Tashjian, Jr., Nature (London) 261, 707 (1976).
43 A. H. Tashjian, Jr., R. Osborne, D. Maina, and A. Knaian, Biochern. Biophys. Res.
Cornrnun. 79, 333 (1977).
44 T. Imae, G. D. Fasman, P. M. Hinkle, and A. H. Tashjian, Jr., Biochem. Biophys. Res.
Comrnun. 62, 923 (1975).
45 p. S. Dannies, K. M. Gautvik, and A. H. Tashjian, Jr., Endocrinology 98, 1147 (1976).
46 K. M. Gautvik, E. Walaas, and O. Walaas, Biochern. J. 162, 379 (1977).
[46]
535
GH C e l l s as a M o d e l S y s t e m
As summarized above, GH cells produce large amounts of growth
hormone and prolactin. The biosynthesis and release of these two protein
hormones are modulated by specific factors that have been found to influence the pituitary gland in the intact animal. In fact, observations initially
made with GH cells were subsequently shown to operate in v i v o . 5 Unlike
the pituitary gland, however, GH cells are clonal populations that can be
studied in a highly controlled environment. They therefore represent a
useful tool for studies on the molecular mechanisms underlying the actions of hypothalamic peptides on growth hormone and prolactin release
and synthesis as well as how peptide hormone receptors are modulated.
Because GH cells are neoplastic, their functions and responses may not
be identical to comparable cells in the normal animal. It is recognized that
tumor cells can lose functions or express functions inappropriately, but it
is not likely that they invent new functions or novel mechanisms. Nevertheless, it is essential to perform, wherever possible, control experiments
with normal pituitary cells or explants to confirm that mechanisms deduced from tumor cells correspond to normal mechanisms.
4~ p. M. Hinkle and A. H. Tashjian, Jr., Endocrinology 100, 934 (1977).

48 T. F. J. Martin, A. M. Cort, and A. H. Tashjian, Jr., J. Biol. Chem. 253, 99 (1978).
a0 p. S. Dannies and A, H. Tashjian, Jr., Bioehem. Biophys. Res. Commun. 70, 1180 (1976).
5o G. A. Evans and M. G. Rosenfeld, J. Biol. Chem. 251, 2842 (1976).
5t A. H. Tashjian, Jr. and R. F. Hoyt, Jr., in "Molecular Genetics and Developmental
Biology" (M. Sussman, ed.), p. 353. Prentice-Hall, Englewood Cliffs, New Jersey,
1972.
52 p. M. Hinkle and A. H. Tashjian, Jr., Biochemistry 14, 3845 (1975).
5~ p. M. Hinkle and A. H. Tashjian, Jr., in "Hormones and Cancer" (K. W. McKerns, ed.),
p. 203. Academic Press, New York, 1974.
54 R. F. Hoyt, Jr. and A. H. Tashjian, Jr.,Anat. Rec. 175, 374 (abstr.) (1973).
55 p. O. Kohler, L. A. Frohman, W. E. Bridson, T. Vanha-Perttula, and J. M. Hammond,
Science 166, 633 (1969).
56 R. J. Tushinski, P. M. Sussman, L.-Y. Yu, and F. C. Bancroft, Proc. Ndtl. Acad. Sci,
U.S.A. 74, 2357 (1977).
57 p. S. Dannies, P. M. Yen, and A. H. Tashjian, Jr., Endocrinology 101, 1151 (1977).
58 E. Haug and K. M. Gautvik, Endocrinology 99, 1482 (1976).
59 M. H. Perrone and P. M. Hinkle, J. Biol. Chem. 253, 5168 (1978).
6o K. M. Gautvik, R. F. Hoyt, Jr., and A. H. Tashjian, Jr.,J. Cell. Physiol. 82, 401 (1973).
61 K. M. Gautvik and M. Kriz, Endocrinology 98, 352 (1976).
62 K. M. Gautvik and A. H. Tashjian, Jr., Endocrinology 92, 573 (1973).
63 K. M. Gautvik and A. H. Tashjian, Jr., Endocrinology 93,793 (1973).
64 H. Wantanabe, W. E. Nicholson, and D. N. Orth, Endocrinology 93, 411 (1973).
~5 R. W. Harrison, S. Fairfield, and D. N. Orth, Biochemistry 14, 1304 (1975).
R. W. Harrison, S. Fairfield, and D. N. Orth, Biochim. Biophys. Acta 466, 357 (1977).
536
S P E C I F I C CELL L I N E S
[47]
[47] L i v e r Cells
By H. L. LEFFERT, K. S. KOCH, T. MORAN, and M. WILLIAMS

Methods for establishing primary monolayer fetal and adult rat
hepatocyte cultures are presented. Techniques are emphasized rather
than rationale or methodological validation. 1-4 Media preparations (Tables I and II), plating procedures (Tables III and IV), and buffer formulations (Table V) are outlined with explanatory details provided in the
text. Table VI summarizes some differentiated properties of these culture
systems.
Animals
Fetal livers are obtained from pregnant rats 19-21 days in gestation;
14--19-day-old fetuses can be used but hepatocyte yields are reduced.
Adult livers are obtained from rats, 150-300 g body weight, and fed standard Purina Chow and water ad libitum. Animals are housed at 21 with
alternating 12-hr cycles of light and darkness.
Biological Reagents
Fetal bovine serum is purchased from standard suppliers and is pretested to ensure its suitability. Defrost, mix well, and heat the serum at
56 for 30 min. Dialyze against 0.15 M NaC1 (1:80, v/v) for 24 hr at 4,
followed by two similar changes, and store aliquots at - 2 0 for up to 12
months. Dialysis tubing (~ inch; Van Waters and Rogers #25225-226) is
boiled 5 min in water containing 0.1% ethylenediaminetetracetic acid
(EDTA), washed 20 min with deionized water, and stored at 4. All serum
used is dialyzed unless otherwise noted.
Collagenase (Sigma Type I, #C0130) also varies in effectiveness and
mUst be pretested for its suitability. The enzyme preparation is stored
desiccated at - 2 0 for up to 6 months.
Amino acids, hydrocortisone-succinate, vitamins, insine, bovine
serum albumin (BSA), and sodium pyruvate are from Sigma; penicillin G
and streptomycin-sulfate are purchased from Squibb and Pfizer, respecLeffert and D. Paul, J. Cell Biol. 52, 559 (1972).
Leffert and D. Paul, J. Cell. Physiol. 81, 113 (1973).
Leffert, T. Moran, R. Boorstein, and K. S. Koch, Nature (London) 267, 58 (1977).
Leffert, K. S. Koch, B. Rubalcava, S, Sell, T. Moran, and R. Boorstein, Natl.
Cancer Inst., Monogr, 48, 87 (1978).
1 H.
2 H.
3 H.
4 H.
L
L.
L.
L.
METHODS IN ENZYMOL(~Y. VOL. LVIII

All rights of reproduction i~ any formreserved.
ISBN 0-12-181958-2
TABLE I
CONSTRUCTION OF 8 LITERS OF TWICE-CONCENTRATED(X2) MEDIUM
Step
Procedure
1. Thaw:
2.
3.
4.
5.
6.
7.
8.
9.
10.
320 ml amino acid stock

160 ml vitamin stock
40 ml antibiotic stock
Dissolve 1.152 g L-tyrosine in 500 ml boiling water. Add 3.2 g MgSO4 7 H20. Cool to
room temperature.
Dissolve 0.748 g L-cystine in 200 ml water (at 60) containing 7 ml 1 N NaOH. Cool to
room temperature.
Dissolve 4.2 g CaC12 2 H20, 6.4 g KCI, and 102.4 g NaCi in 1500 ml water.
Add amino acid stock to #4 using stirring bar. Add:
24 ml 1% (w/v) phenol red
72 g glucose (anhydrous)
2.3 g NaH2PO4
0.112 g Inositol
16 ml 0.01% (w/v) FeCla
40 ml antibiotic stock
16 ml antimycotic
160 ml vitamin stock
Let stir for 5-10 min.
Add 9.34 g L-glutamine and 1.76 g sodium pyruvate to #5. Stir.
Add cooled solutions (#2, #3) to #6. Stir.
Bring to 3200 ml with water; bubble COs for about 1 hr while adding 800 ml water containing 59.7 g NaHCOa.
Filter through 0.22 p.m Millipore GS filter.
Dispense into sterile bottles and store in the dark at 4 for up to 6 weeks.
TABLE II
STOCK SOLUTIONS FOR ARGININE-FREE MEDIUM
Amino acids (4 liters)

L-Glycine
L-Histidine HC1
L-Isoleucine
L-Leucine
L-Lysine
L-Methionine
L-Ornithine
L-Phenylalanine
L-Serine
L-Threonine
L-Tryptophan
L-Valine
6.00
8.40
20.96
20.96
29.24
6.00
16.80
13.20
8.40
19.04
3.20
18.72
g
g
g
g
g
g
g~
g
g
g
g
g
Vitamins (2 liters)
Choline chloride
Folic acid
Nicotinamide
Pantothenic acid
Pyridoxal HC!
Thiamine
Riboflavin
Antibiotics (330 ml)
Penicillin G
Streptomycin sulfate
Antimycotic (1 liter)
n - Butyl-p -hydroxybenzoate
0.80
0.80
0.80
0.80
0.80
0.80
0.08
g
g
g
g
g
g
g
30 106 U
6.0 g
0.1 g
Can be omitted or added from a standard sterile stock solution (332 mg/100 ml water;
add 2.5 ml/100 ml l medium).
538
SPECIFIC CELL LINES
[47]
TABLE III
FETAL RAT LIVEn CELL CULTURE PLATING PROCEDURE
Step
Manipulation
1. Lightly etherize pregnant rat; immobilize the animal on its back and soak entire abdomen with 95% (v/v) ETOH.
2. Make a midline incision with a sterile scissors, cutting first skin and then muscle layers;
do not perforate intestines, which lie under muscle layer.
3. Remove fetuses one at a time from embryonic sacs. Cut umbilical cord and decapitate
immediately. Place about 10-15 into a sterile 9-cm dish.
4. Dissect out intact livers sterilely and put about 15 into a 5-cm dish containing 5 ml
ornithine-minus medium at 37 supplemented with 10% (v/v) serum. Inspect each liver
to ensure absence of nonhepatic tissue.
5. Cut each liver into its four separate "lobes"; remove medium by aspiration and wash
once with 5 ml of fresh medium.
6. Pool washed, cut tissue and incubate in fresh collagenase-supplemented digestion buffer
(1-2 mi/liver), for 8-10 min at 37 in a 125-ml Erlenmeyer flask containing a magnetic
stirring bar. Set rheostat = 35-40.
7. Tilt flask to allow undigested tissue to settle and remove cell suspension, with a sterile
Pasteur pipette, placing it into a "storage" flask containing 10 ml medium (maintain
on ice).
8. Repeat tissue digestion steps 6 and 7, a total of 4-5 times, pooling each subsequent
harvest. Add an equal volume of fresh ice-cold medium to the final cell suspension and
swirl gently. Dispense 20-40 ml (by pouring) into sterile 50-ml plastic tubes.
9. Centrifuge 10 sec at full speed and apply brake immediately thereafter.
10. Remove supernatant liquid by suction and discard; wash pellet once with 10 ml of
serum-free medium (blow fluid rapidly against sidewall to disrupt pellet and then take
up the cell suspension; blow it out gently against sidewall). Repeat step 9.
11. Discard second supernatant liquid. Resuspend pellets as in step 10, pool (store on ice),
and titer a 0.5-ml aliquot.
12. Pipette cell suspensions into appropriate media and plate using a flow syringe.
tively. Eastman Kodak produces n-butyl-p-hydroxybenzoate.

s u l i n is o b t a i n e d f r o m t h e E l i L i l l y R e s e a r c h L a b o r a t o r y .
P o r c i n e in-
Inorganic Reagents
T r i p l e - d i s t i l l e d w a t e r is u s e d u n l e s s o t h e r w i s e n o t e d . T r i s b a s e is f r o m
Sigma.
Tissue Culture Supplies
Equipment. T i s s u e c u i t u r e i n c u b a t o r s s u p p l i e d w i t h CO2 a n d a i r l i n e
f i l t e r s , r e g u l a t o r y v a l v e s ( s e t t i n g s a r e 0.5 a n d 4 . 0 f o r CO~ a n d a i r , r e s p e c -
[47]
LIVER CELLS
539
TABLE IV
ADULT RAT LIVER CELL CULTURE PLATING PROCEDURE
Step
Manipulation
1. Lightly etherize rat. Immobilize the animal on its back and soak entire abdomen with
95% ethanol.
2. Make a midline incision extending into the thoracic cavity. Exsanguinate the animal by
cardiac puncture, removing 5-8 ml blood, and push the intestinal contents aside
everting them to expose portal vessels.
3. Clamp portal vein (to the left of splanchnic and gastric venous inputs) together with the
celiac trunk (small pulsating arterial system dorsal to everted portal system). Clamp
inferior vena cava above the renal venous inputs but below the hepatic vein.
4. Insert a 20-ml plastic syringe equipped with a 25-gange needle (bevel up) and perfuse
20 ml hormone, purine, and collagenase-supplemented "digestion buffer" for 2-3 rain
through the liver. Complete tissue blanching; a light tan color indicates adequate
perfusion.
5. Excise perfused liver, placing it in a 9-cm plastic petri dish and removing extraneous
tissue if present. Wash liver twice with 10 ml of "complete medium." Add 10 ml of
fresh medium and dissect tissue (two livers) into 5 8 mm pieces and wash twice.
6. Incubate cut tissue at 37 with 40 ml of hormone, purine, and collagenase-supplemented
digestion buffer for 8-10 min in a 125-ml Erlenmeyer flask containing a magnetic
stirring bar. Set rheostat = 35--40.
7. Tilt flask to allow undigested tissue to settle, and remove cell suspension with a sterile
Pasteur pipette, placing it into a flask containing 10 ml of medium (maintain on ice).
8. Repeat tissue digestion (steps 6, 7) four times, pooling each subsequent harvest. Add
160 ml ice-cold "complete" medium (a total volume of about 320 ml) and dispense
40 ml into eight 50-ml sterile plastic centrifuge tubes.
9. Centrifuge 13 sec (full speed) and apply brake.
10. Remove supernatant liquid by suction and discard. Wash pellet once with 10 ml of
" b a s a l " medium; blow fluid rapidly against the sidewall to disrupt pellet and then take
up the cell suspension and blow it out gently against sidewall. Repeat step 9.
11. Remove by suction and discard clear supernatant fluid; pellets should be "white."
Resuspend pellets as in step 10 twice but blow out hard. Transfer to ice-cold sterile
bottle, swirl pooled cell suspension (about 80 ml), and resuspend seven times. Titrate a
0.5-ml aliquot.
12. Redistribute in portions of 20 ml or less to centrifuge tubes and repeat step 9.
13. Resuspend pellets in 10 ml (twice, hard) of appropriate plating media. Bring up to
needed volume and plate using a 10-ml plastic pipette.
tively), and external thermometers are from Napco (#3321) or Shel-Lab

(Portland, Oregon). Water trays are used (gas is bubbled through water) to
maintain humidity. Calibrate temperature (use internal thermometers at
different tray levels) and gas tension (use test media and check pH).
Napco also supplies a small air incubator (#310) used to perform tissue
digestion and trypsinizations at 37 . Cell numbers are determined electronically with a Coulter Particle Counter (Model Fn, Hialeah, Florida).
540
SPEOFIC CELL LINES
[47]
TABLE V
BUFFER SOLUTIONS
Ca2Mge-freeconcentrate (2 liters)
NaCI
KC1
Na2HPO4 (anhyd.)
Tris base
1 N HCI
(Adjust pH to 7.4)
64.0
3.04
0.08
24.0
160
Trypsin concentrate (1 liter)

g
g
g
g
ml
Ca2~Mg~+-free Tris buffer (2 liters)

Ca2+MgZ+-free concentrate
Water
(Adjust pH to 7.4)
500
1500
25 g
1.0 liter
Steps:
1. Dissolve 8 hr at 4 with stirring; let settle.
2. Filter supernatant at 21 using Whatman
No. 1 paper.
3. Sterilize with 0.22/.Lm GS MiUipore and
AP2512450 prefilter.
4. Store 60 ml portions at - 2 0 .
Trypsin stock solution for cell counts (0.6
Tris buffer (3 liters)

Ca2+Mg2+-free concentrate
Water
15% (w/v) CaCI~ 2 H~O
15% (w/v) MgCI~
(Adjust pH to 7.4)
ml
ml
Difco 1:250 trypsin

Water
750
2250
2
2
ml
ml
ml
ml
liter)
Trypsin concentrate
60
1% phenol red
2
(Filter through 0.45/zm Nalgene filter)
Sterile Ca2+Mg2+-free Tris buffer 540
Store 10 ml portions at - 2 0 .
ml
ml
ml
Digestion buffer (1 liter)

KC1
NaCI
NaH2PO4" H~O
D-glucose
Phenol red
Insulin a
Hydrocortisone- succinate
Inosine a
(Adjust pH to 7.4)
0.224
7.587
0.138
1.802
0.0012
0.010
0.010
0.010
g
g
g
g
g
g
g
g
a Freshly added for adult liver cell isolation only.
A desk-top centrifuge containing a timer, brake, and rpm indicator is

obtained from International Equipment Co. (#HN-S, Needham Heights,
Massachusetts). A #958 head, adapted for IEC #325 rings and #305
buckets with rubber stoppers, is used.
The electric rheostat-controlled stirrer, from A. H. Thomas (MagneStir, #10), is placed into the Napco air incubator.
Aluminum tissue culture trays (12 30 cm) with grated open sides
(1-cm high) and bottoms were constructed by this Institute's engineering
shop.
Magnetic stirring bars from Van Waters and Rogers (#58948-171;
2 x ~ inch) are replaced monthly.
[47]
LIVER CELLS
541
T A B L E VI
DIFFERENTIATED PROPERTIES OF PRIMARY MONOLAYER RAT
LIVER CELL CULTURES
Enzymes
Gluconeogenic
Glutathione S-transferase B[EC 2.5.1.18]

Glycogenolytic
Microsomal P44s,4~0
Ornithine transcarbamylase [EC 2.1.3.3]

Pyruvate kinase [EC 2.7,1.40]
Fetal (type III)
Adult (type I)
Tyrosine aminotransferase [EC 2.6.1.5]
Urea " c y c l e "
Secretory proteins
Albumin
Alphal-fetoprotein
Haptoglobin
Hemopexin
Very low density lipoprotein
UItrastructure c
Bile " c a n a l i c u l i "
Desmosomes
Glycogen granules
Peroxisomes
Tight j u n c t i o n s
Fetal
Adult
+
U"
+
U
+
+
+t,
+
+
+
+
U
U
+
+~
+b
+
+
+
+
+
+
_c
+
+b
U
U
+b,c
+
+
+
+
+
+
+
+
+
+
a Unknown.
0 G r o w t h state dependent (H. L. Leffert et al., Proc. Natl. Acad. Sci. U.S.A. 75, 1834
(1978).
c U n p u b l i s h e d observations (D. Weinstein, K. D e m p o , S. Sell, and H. L. Leffert).
Plasticware and Glassware. Falcon Plastics supplies sterile 50-ml

(#2070) and polypropylene test tubes (#2059), petri dishes, 35-ram diameter tissue culture dishes (#3001), and 1-ml (#7506) and 10-ml (#7530)
pipettes. N U N C (Roskilde, Denmark) supplies 5-cm (#1400) and 9-cm
(#N-1415) diameter tissue culture dishes. Dish lot numbers should be
noted in case of toxicity problems.
Nalgene 0.20 and 0.45-/~m filters (#120-0020 and #245-0045) are purchased from Sybron (Rochester, New York) and should be checked for
tightness of seal and detergent contamination (removable with warmwater washes).
Glass Erlenmeyer flasks (125 ml; #26500 or #26650) and other routine
items are purchased from Kimax.
Cornwall continuous pipetting outfits (5-cc "flow syringes") are from
542
SPECIFIC CELL LINES
[47]
Becton Dickinson (#3054) and are fitted with autoclavable I-inch ID 4!inch OD silastic tubing from Dow Coming (#601-365).
Surgery
Sterile technique is preferred but not necessary; alcohol-washed areas
and instruments suffice. Animals are placed onto clean absorbent cotton
pads layered over a simple fiat cork board and operated on between
0900-1200 hours.
Media and Digestion Buffer Construction
Twice-concentrated medium is prepared as described in Table I.

Amino acid, vitamin, and antibiotic stock solutions (Table II) are prepared
in the appropriate portions and stored in darkness at - 2 0 until use. The
antimycotic stock solution is stored in darkness at room temperature.
L-Ornithine is usually present but is not necessary for fetal liver cultures 1-4. Arginine is added after plating only under special conditions. 5-9
One hundred milliliters of x 1 medium at pH 7.4 is prepared fresh as
follows: 45 ml sterile 2 medium; L-ornithine, if desired (see Table II);
serum (0-30 ml); and sterile water to make a final volume of 100 ml. If
contamination occurs, add 0.25 ml of antibiotic stock.
For tile adult system (Table IV), "complete medium" contains
L-ornithine, 15% (v/v) serum, and 10 /zg/ml each of insulin, hydrocortisone, and inosine. "Basal medium" is complete medium minus hormone
and purine supplements. Stock solutions of hormone or purine, 5 ml, are
prepared fresh as 1 mg/ml solutions in polypropylene test tubes in 0.15 M
NaC1 containing 0.3% (w/v) BSA. One drop of 6 N HC1 will dissolve
insulin in the 5-ml solution. Supplements, 1 ml/100 ml medium, are added
to the medium followed by sterilization through a 0.20-/~m Nalgene
filter.
Digestion buffer (Table V) is stored at - 2 0 until use. Fresh collagenase, 3 mg/ml, is added to buffer at 37; after dissolving, the solution is
sterilized (as above) and preincubated at 37 for 1-4 hr prior to use. For
adult liver tissue digestion only, prior to sterilization, 100 ml of buffer
5 H.
6 H.
7 K.
s H.
9 H.
L, Leffert and S. Sell, J. Cell Biol. 61, 823 (1974).

L. Leffert, J. Cell Biol. 62, 767 (1974).
Koch and H. L. Leffert, J. Cell Biol. 62, 780 (1974).
L. Leffert, J. Cell Biol. 62, 792 (1974).
L. Leffert and D. B. Weinstein, J. Cell Biol. 70, 20 (1976).
~7]
LIVER CELLS
543
containing enzyme is supplemented with 1 ml each of the insulin, hydrocortisone, and inosine stock solutions.
Cell Isolation and Plating
Detailed procedures appear in Table III (fetal) andTable IV (adult). All
steps are performed with sterile technique. The stirring speed, i.e., rheostat knob setting, for all tissue digestion is 35-40. For fetal cultures, four
animals are used and usually yield 20-40 fetuses giving rise to about
1-3 108 cells. For adult cultures, two livers are used giving rise to about
1-2 108 cells. For both'systems, the time from surgery to plating should
be about 3 hr or less. At plating, greater than 95% of isolated cells are
hepatocytes 3 with 70-95% viability as judged by exclusion of 0.05% (w/v)
trypan blue at 1 min.
Seeding densities, i.e., numbers of cells per milliliter medium, and
volume of media per dish vary with experimental design. 1-3 For the fetal
system, usual parameters for 35-, 50- and 90-mm diameter dishes are
1.25 x 105 cells/ml, 2 ml; 1.5 x 105 cells/ml, 5 ml; and 2 105 cells/ml, 10
ml, respectively. For the adult system, 5 105 cells/ml, 2 ml, are plated
per 35-mm dish. " H e a l t h y " cultures show cellular aggregates upon the
surface of the dish within 24 hr and flattening (nonrefractile nuclear and
cytoplasmic morphology under phase optics) within 48 hr. Medium
changes are performed as needed but are not necessary. In this laboratory, functional fetal cultures plated without L-ornithine have survived for
6-8 weeks without a medium change; functional adult cultures plated with
L-ornithine have survived for 2 weeks without a medium change.
Cell Counts
At plating, titrations are performed with 0.5 ml of cell suspension
added to 9.5 ml of Isoton (Scientific Products, #B3157-11) and counted
electronically (threshold at 10 ~m; attenuation and aperture settings
of 1 and 256, respectively). Hemocytometer counts should be within
___10% of corrected electronic counts using these settings.
Table V lists buffer solutions used to measure attached cell numbers.
Dishes are washed twice with Tris buffer, pH 7.4 (a volume equal to initial
medium volume). Then an equal volume of Ca2+Mg2+-free Tris buffer with
trypsin (10 ml trypsin stock solution per 180 ml Ca2+Mg2+-free Tris
chloride at pH 7.4) is added and the dishes incubated for 30-40 min at 37.
A 2 ml (#122402010) or 5 ml (#122405010) glass handpipette (Bellco)
equipped with a rubber bulb (Scientific Products, #R5000-1) is useful for
544
SPECIFIC
CELL
LINES
[48]
detaching the cells; vigorous pipetting, about 20 times, over the dish
surface usually reduces cell aggregation. The cell suspension or an aliquot
of it is brought up to 10 ml with Isoton and counted.
Functional Properties
Typical differentiated properties of primary monolayer rat liver cell
cultures are summarized in Table VI. Additional sources, including those
cited above, should be consulted, lo-19
10 S. Sell, H. L. Leffert, U. Mueller-Eberhard, S. Kida, and H. Skeily,Ann. N. Y. Acad. Sci.
259, 45 (1975).
11 H. Watabe, H. L. Leffert, and S. Sell, in "Onco-Developmental Gene Expression" (W.
Fishman and S. Sell, eds.), p. 123. Academic Press, New York, 1976.
12 D. M. Bissell, L. E. Hammaker, and U. A. Meyer,'J. Cell Biol. 59, 722 (1973).
13 R. J. Bonney, In Vitro 10, 130 (1974).
14 C. Guguen, C. Gregori, and F. Schapira, Biochimie 57, 1065 (1975).
t5 G. Michalopolous and H. C. Pitot, Exp. Cell Res. 94, 70 (1975).
~e T. H. Claus, S. J. Pilkis, and C. R. Park, Biochirn. Biophys. Acta 404, 110 (1975).
lr G. M. Williams, Am. J. Pathol. 85, 739 (1976).
18 D. Bernaert, J. C. Wanson, P. Drochmans, and A. Popowski, J. Cell Biol. 74, 878 (1977).
10 H. L. Leffert and K. S. Koch, Ciba Found. Syrnp. 55 (New Ser.), 61 (1978).
[48] L i v e r Cells ( H T C )
By E. BRAD THOMPSON
HTC (hepatoma, tissue culture) cells are a line of cells established
from the ascites form of the rat hepatoma, Morris 7288C. 1 The original
hepatoma was dissected from the liver of a male rat of the inbred Buffalo
strain that had been fed the carcinogen, N,N'-2,7-fluorenylenebis-2,2,2trifluoroacetamide. It subsequently was converted to the ascites form and
carried serially as such. The HTC line was cultured directly from the
ascitic fluid of tumor-bearing rats.
HTC cells have proved useful in studying a number of both liverspecific and general cellular processes at the cellular level. The cells have
been used especially for studies of the effect of steroid hormones on
enzyme induction. Glucocorticoids induce in these cells the hepatic
marker enzyme tyrosine aminotransferase (L-tyrosine: 2-oxoglutarate
aminotranserase, EC 2.6.1.5), as well as ornithine decarboxylase and
glutamine synthetase. In HTC cells, such steroids also increase cell adhe1 E. B. Thompson, G. M. Tomkins, and J. F. Curran,Proc. Natl. Acad. Sci. U.S.A. 56, 296
(1966).

ISBN 0-12-181958-2
[48]
HTC CELLS
545
siveness; reduce the activity of cyclic Y,5'-AMP(cAMP) phosphodiesterase and of plasminogen activator; induce an increase in one of the
isoaccepting forms of phenylalanyl tRNA; induce mouse mammary tumor
virus in infected cells; render cells more sensitive to cAMP analogs; and
inhibit the transport of certain amino acids. Besides studies on the action
of steroid hormones, HTC cells have been utilized in studying the control
of fatty acid and cholesterol synthesis, the viral-like A particles, RNA
viruses (C particles), growth control, phenylalanyl hydroxylase, and an
extensive list of other enzymes. The cells also have been used in somatic
cell genetics, through formation of somatic cell hybrids with various other
cell types and through the isolation of stable cell variants cloned from the
original cell line.
Propagation of HTC Cells in Monolayer Culture
HTC cells are a hardy line that grow readily as monolayer cultures on
either glass or plastic substrata. They have been successfully carded in
both commercially made special glass tissue culture flasks and in simple
medicine bottles laid on side. Plastic tissue culture flasks and dishes from
virtually all major U.S. suppliers also have been employed to grow these
cells. HTC cells in monolayer culture, being transformed cells, grow in an
irregular, trabecular pattern, display both round and polygonal cell morphology, and do not exhibit density-dependent inhibition of growth. After
being seeded so as to form a lawn of individual cells, they .grow atop one
another as well as side to side, and frequently reach stationary phase with
areas of the substratum still unoccupied by cells.
Media
HTC cells are not particularly fastidious with regard to medium. They
were originally established and maintained for years in Swim's medium
#77, (modified from 5103, with hydroxyproline omitted 2) supplemented
with 20% bovine plus 5% fetal bovine serum. The accompanying table
lists the ingredients of this medium. Later it was found that they grov~ in a
number of standard media, including Eagle's minimal essential medium, 3
Eagle's basal medium, 4 Ham's F 12, 5 and improved minimal essential medium, zinc option (IMEM-ZO). n For routine culture, these media usually
2 H. E. Swimand R. F. Parker, J. Lab. Clin. Med. 52, 309 (1958).
3 H. Eagle, Science 130, 432 (1959).
4 H. Eagle,Science 122, 501 (1955).
5R. G. Ham, Proc. Natl. Acad. Sci. U.S.A. 53, 288 (1965).
A. Richter, K. K. Sanford, and V. J. Evans,J. Natl. Cancer Inst. 49, 1705(1972).
546
SPECIFIC CELL LINES

INGREDIENTS OF MEDIUM
$77 AS USED
FOR THE CULTURE OF
[48]
HTC CELLS
Concentration
(mM)
Salts
Calcium chloride (2 H20)

Magnesium sulfate (7 H20)
Potassium chloride
Sodium chloride
Sodium bicarbonate
NaH~PO4 H20
Amino Acids
1.8
0.8
5.5
116.2
26.2
1.2
Vitamins, miscellaneous
Glucose
Choline bitartrate
Folic acid
Calcium pantothenate
Mesoinositol
Nicotinamide
Pyridoxal HCI
Riboflavin
Thiamin HCI
Phenol red (sodium salt)
Concentration
(raM)
5.6
0.01
0.005
0.005
0.01
0.005
0.005
0.001
0.005
0.02
L-Alanine
L-Aspartic acid
L-Arginine (HCi)
L-Cystine (HC1)
L-Histidine (HCL - H20)
Glycine
L-Glutamine
L-Isoleucine
L-Leucine
L-Lysine (HC1)
L-Methionine
L-Phenylalanine
L-Proline
L-Serine
L-Threonine
L-Tryptophan
L-Tyrosine
L-Valine
0.3
0.15
0.8
0.05
0.05
0.15
2.0
0.2
0,2
0,2
0.1
0. I
0.15
0.2
0.4
0.05
0.1
0.4
need supplementation with only 5% fetal calf serum. Although it is not

necessary, heat inactivation of the serum (56 for 60 min) to destroy
complement and to reduce the chance o f contamination with m y c o p l a s m a
is r e c o m m e n d e d . In I M E M - Z O , the serum requirement can be reduced
sharply, and the cells grow readily in as little as 0.25% fetal calf serum.
H T C cells also have been adapted to growth in totally defined, serumfree medium without m a c r o m o l e c u l a r supplements. For this purpose, the
medium o f choice is I M E M - Z O . All these media are bicarbonate-buffered
and therefore should be used in an incubator kept at 37 with a humidified
atmosphere of 5% CO2 in air.
R o u t i n e Culture
Routine cultures are fed 3 times a week by removing the medium from
the monolayer and replacing it with fresh medium at 37 . The cells double
about every 24 hr under these conditions, and for best viability they
should be kept in logarithmic growth phase by subculturing appropriately
into fresh culture flasks or dishes. They can be subcultured from nearconfluent flasks at a ratio o f 1 : 10 or 1 : 20 with virtually 100% viability.
After such subculturing, they should show a lag phase o f no more than 24
hr, and less if handled expeditiously. For subculturing, cells can be re-
[48]
HTC CELLS
547
moved from their substratum in any of several ways. The preferred methods is to use dilute trypsin or trypsin-EDTA as follows:
Remove spent growth medium by suction. Gently introduce 5-10 ml of
phosphate-buffered saline (PBS) (138 mM NaC1, 2.7 mM KC1, 8.1 mM
Na2HPO4, 1.47 mM KH2PO4, pH 7.4) at 25, and tip the dish to wash over
cells. Remove by suction. Introduce 1-2 ml of Puck's Saline A (containing, in grams per liter: 20 mg phenol red, 0.35 g NaHCO3, 8.0 g NaC1, 0.4 g
KC1, and 1.0 g dextrose), which contains 0.05% trypsin and 0.02%
ethylenediaminetetracetic acid (EDTA), prewarmed to 37. Tip to wash
over cells. Remove by suction all but approximately 0.5 ml. Place in a 37
incubator and examine after 5 min. Tap the dish against the heel of the
hand to dislodge cells, which should come free in 5-15 min. As soon as
cells float free, add serum-containing medium or soybean trypsin inhibitor
at 37 and reflux-pipette cells gently about 3 times to give a suspension of
single cells. Plate appropriate portions of cell suspension into fresh
dishes.
Cautions: If trypsin-EDTA is interrupted too soon or allowed to continue overlong, cell clumping will result. In addition, with overlong trypsinization, there may be loss of viability. The pH during these procedures
should not be allowed to rise above 8.0; the inclusion of phenol red in the
trypsin-EDTA solution allows visual monitoring of the approximate pH. If
a large number of cultures are to be subcultured, it is recommended that
they be carried through the procedure a few at a time, and that those
waiting to be handled be kept in the air-CO2 atmosphere of the incubator.
Alternate methods of cell removal for subculturing that can be used
successfully are: scraping with an L-shaped glass rod, either naked or
with its operative end covered by a silicone rubber tip; trypsin alone, up
to 0.1 g/100 ml in saline A or PBS; scraping with a Teflon-coated magnetic
bar inside the tissue culture flask and a stronger magnet outside moved to
manipulate the inner magnet. EDTA alone is not as satisfactory as the
other methods, although it will remove some cells.
For cells grown in serum-free medium, best results are obtained with
the trypsin-EDTA technique, using soybean trypsin inhibitor to stop proteolysis. If the introduction of even this small amount of protein is to be
avoided, scraping is the second-best choice.
Cloning
HTC cells can be cloned with high efficiency by any of several standard techniques in standard medium. If good tissue culture technique is
used, cloning efficiency should be 50-100%. Methods that have produced
good results include cloning in plastic tissue culture dishes or flasks, in
548
SPECIFIC CELL LINES
[48]
multiwell dishes, and in soft agar. To clone in a tissue culture dish or flask,
cells are trypsin-EDTA treated as described above (if grown monolayerstyle) or simply taken as grown if in suspension culture (see below). In
any case, for best results logarithmic-phase cells should be cloned; they
are easier to obtain in monocellular suspension and give higher cloning
efficiency. The original cell suspension should be counted carefully by
hemocytometer or electronic counter and in either case examined by
phase-contrast microscopy to be sure the closest possible approximation
to a monocellular suspension is achieved. Proper attention to procedure
should result in 90--95% of all cells existing independently, with the remainder appearing as doubles and only rare groups of three or more. It is
worth performing preliminary experiments to be sure this degree of dispersion is achieved, since most cloning experiments are aimed at obtaining truly clonal populations, and contamination with greater than 5% with
pseudoclones, derived from more than a single cell, may complicate subsequent results unnecessarily.
The dispersed cells are serially diluted with growth medium at 37 ,
taking care not to pipette too vigorously, until a convenient dilution is
reached. If the cells are to be cloned e n m a s s e in large plates or flasks, a
seeding concentration of 1-2 cells cm 2 results in a yield of colonies convenient for counting or picking for subculture. Larger numbers may be
plated, of course, if selective conditions resulting in survival of only a
fraction of the seeded cells are to be used. To cleanly pick colonies, each
colony is first, isolated in a stainless-steel ring and then trypsinized, as
described by Ham and Puck. 7 Because of the possibility that multiple
colonies obtained from a single dish may include either seed colonies from
an initial site or cross-contaminated colonies from cells detached from
other, large clones, it is recommended that attempts to obtain truly clonal
populations can be carded out either in multiwell dishes or in soft agar.
For the former, good results can be obtained with the standard 96-well
dishes at a volume of about 0.1 ml per well. If cells are diluted to average
one cell per five wells at the time of seeding, then about 95% or more of the
subsequent colonies will have risen from a single cell. Seeding at higher
numbers of cells per well results in significantly greater numbers of
nonclonal colonies. Care must be exercised to keep the parent cell suspension well mixed during the seeding. This can be done by frequent
manual mixing or by use of a sterile Teflon-coated magnet in the suspended " m o t h e r " culture, turning slowly under the control of an external
magnetic stirrer. In addition, using no more than a 2-ml pipette to seed the
0.1-ml aliquots will help avoid nonrandom cell distribution. In larger
pipettes or with very slow pipetting, cells may accumulate along the
7 R. G. Ham and T. T. Puck, This series, Vol. 5, [9] p. 107.
[48]
HTC CELLS
549
pipette, walls, resulting in irregularities in the number of cells delivered

per unit volume. When sufficient cells have grown in the wells, they can
be replica-plated if desired. Commercial devices for replica-plating from
such dishes are available.
To clone in agar, either of two methods may be used. High cloning
etticiencies result with either in a final agar concentration of 0.3-0.5%.
Agarose, at 0.35%, works equally well. In the first procedure, a stock agar
solution is made by autoclaving 2 g agar (Difco) in 100 ml of PBS. This is
kept in solution in a 56 water bath. Cells to be cloned are diluted to 1.25
times the final desired concentration in normal growth medium at 37 . The
agar is allowed to cool to near gelling temperature; then it is quickly
mixed with the cell suspension, 1 part agar to 4 parts cells, and plated in
appropriate dishes. The freshly plated dishes are transferred to a refrigerator for a few minutes to assist uniform gelling and then moved to
the 37 incubator. The 20% dilution of medium nutrients incurred in this
procedure does not seem to affect the cells adversely. An alternate procedure, however, avoids this dilution. The agar or agarose is made as an
aqueous solution and autoclaved as above. A special supply of growth
medium is prepared, containing twice the normal concentration of salts,
nutrients, and serum. Cells grown normally are diluted down to 10-100
times the final desired concentration. These are then rapidly diluted to the
final concentration in the concentrated medium, which is at once mixed
1 : 1 with the agar (or agarose) and plated as described before. When
clones appear, they may be removed from the agar bed with a sterile,
cotton-plugged Pasteur pipette and transferred to other vessels for propagation as monolayer cultures.
Serum-Free Growth of HTC Cells

For some purposes, such as defining growth factors and amino acid
and cofactor requirements, developing auxotrophs, and studying sterol
synthesis and certain hormone actions, cells grown in defined medium
offer many advantages. HTC cells can be adapted to growth in totally
defined medium lacking macromolecular supplements. 8 For this, the medium of choice is IMEM-ZO. The cells can be adapted directly to the
serum-free medium, or carried for a few weeks in intermediate levels of
fetal calf serum, for example 1% followed by 0.25%. In the defined medium, the cells undergo a brief adaptation period and then grow with a
doubling time only a few hours longer than that in 5% fetal calf serum.
Cell transfer of the cells grown without serum can be accomplished by
scraping the monolayer with an L-shaped glass rod with or without a
s E. B. Thompson, C. U. Anderson, and M. E. Lippman, J. Cell. Physiol. 86, 403 (1975).
550
SPECIFIC CELL LINES
[48]
silicone rubber sheath, or by trypsinization. If trypsinization is used it

should be noted that the serum-free cells are exquisitely sensitive to the
proteolytic enzyme, and should be monitored constantly during the process. Overlong treatment results in rapid loss of cell viability; consequently the standard procedure for serum-grown cells is modified as follows. Growth medium is removed and a few milliliters of trypsin-EDTA
solution introduced, washed over the cells, and all but 0.1 ml immediately
removed. The cell layer is left with only this slight film of liquid, and as
soon as the cells loosen from the substratum a slight excess of soybean
trypsin inhibitor in PBS is added. The cells can be centrifuged for 3 min at
800 g and resuspended in growth medium or simply diluted into growth
medium for transfer. They should not be subdivided at a ratio below 1 : 10,
and 1:4 is recommended for best viability. The medium containing the
trypsin and inhibitor is removed as soon as the transferred cells have
adhered to the plastic substratum. The advantage of trypsin for cell transfer is that it results in a well-dispersed cell monolayer; the obvious disadvantage is that the process briefly introduces exogenous proteins into the
culture system.
As with many cell lines, HTC cells grown in defined medium are more
sensitive to mishandling than are serum-grown cells. Care to avoid rough
pipetting, pH extremes, and toxic contaminants in medium, glassware,
and plastics must be taken. The cells clone poorly in serum-free medium,
with an efficiency not exceeding 1%.
Propagation of HTC Cells in Suspension Culture
HTC cells have been adapted to grow in suspension (spinner) culture
in a modified form of Swim's $77 medium, in which CaCI2 is omitted,
NaHCOa is reduced to 0.5 g/liter, glucose is increased to 3 g/liter, and 0.05
M Tricene at pH 7.6 is added. 9 (This medium works equally well with only
1 g/liter glucose.) The medium is supplemented with 10% serum, at least
5% of which should be fetal calf serum. In this medium the cells can be
grown in tightly capped bottles without added COz. Suspension culture
also can be carried out in the usual bicarbonate-buffered $77 monolayer
medium from which CaCI2 is omitted. In this case, to maintain proper pH,
the cultures must be grown in 5% CO2 in air. Any of several vessels have
been used for this form of culture: commercial suspension culture flasks,
with Teflon-coated magnetic stir bars on Teflon pivots (Belco); Wheaton
bottles or screw-capped Erlenmeyer flasks in which a Teflon-coated
magnetic bar simply rests on the bottom; or bottles with a stir bar suspended from a rubber stopper by a stainless-steel "bathtub" chain. All
9 R. S. Gardner, J. Cell Biol. 42, 320 (1969).
[48]
HTC CELLS
551
work well, although the cheapest and simplest by far is the Wheaton
bottle-stir bar combination.
Initiating a suspension culture from HTC cells previously grown exclusively as monolayer cultures requires adaptation by the cells. Considerable cell death is often seen during the adaptation period, and once the
number of viable cells falls below a critical level, the entire culture dies.
Consequently, several cell concentrations should be set out initially. Cells
in monolayer culture should be adapted to the spinner medium for several
days while still grown as monolayers. The cells that come free from the
substratum should be collected by centrifugation and returned to the culture, and not discarded during medium changes. After this initial adaptation, the monolayer-grown cells are collected, with those still adhering to
the substratum removed by trypsin-EDTA. Suspend 1-5 105 cells/ml in
fresh, suspension-culture growth medium at 37 to a total volume of at
least 100 ml. Add the cell suspension to a vessel chosen so that it will be
1 2
r-~
full, containing a stir bar whose length equals 1-1.5 times the radius of
the vessel. Place the vessel on a magnetic stirrer set at a speed to produce
gentle mixing and no turbulence. There should be a slight vortex formed at
the surface of the liquid. During the adaptation period the cells may
multiply quite slowly. Medium changes, without diluting the cells, can be
carried out by separating the cells by centrifugation (5 min at 800 g) or by
removing the flask from the magnetic stirrer and allowing the cells to
settle for 30-60 min, and then removing most of the overlying medium by
suction.
Once the cells adapt to suspension culture conditions, they double
approximately every 24 hr. If fed 3 times per week, and kept between 10~
and l0 Gcells/ml, they will remain in logarithmic growth phase indefinitely.
They have been grown successfully in volumes as low as 50 ml and as high
as 50 liters. The suspension-adapted cells can be again grown as
monolayers, although at first they attach less firmly and have a more
spherical morphology.
Cryopreservation of HTC Cells
HTC cells can be stored by freezing in a viable state, using standard
methods. Medium containing at least 10% serum is supplemented with
10% dimethylsulfoxide (DMSO) or 10-15% glycerol. Cells are harvested
and suspended in this medium at about 106 cells/ml. Aliquots of 1-1.5 ml
are placed in small screw-capped ampules and immediately frozen, lowering the temperature at 1-3/min. This can be achieved several ways: by
use of special instruments, by putting the vials in an alcohol-ice bath to
which frozen COz chips are added, or by simply standing the vials on a
552
[49]
bed of frozen COs or, wrapped in paper towels, in a -75 freezer. For
greater security against contamination, sealed glass ampules may be substituted for the screw-capped type. After the cells are frozen, they are
transferred to a liquid N2 freezer; the lower temperature it provides is
essential for viability after long storage. HTC cells have been regrown
after more than 5 years of such preservation.
When frozen cells are to be grown, the vial is promptly brought to
ambient temperature or to 37 in a water bath and cultured without delay.
Glycerol-frozen cells can simply be diluted with fresh medium and grown.
DMSO-frozen cells should be removed from the toxic freezing medium
either by centrifugation as soon as the cells are thawed or by plating the
thawed cells with fresh medium and then replacing the medium again after
a few hours, when the cells have adhered to the substratum. Initial culture
in any case should be as monolayers, not suspension cultures. Cells grown
serum free can be frozen in serum-supplemented medium as above, or
15-20% polyvinylpyrrolidone can be substituted for serum.
[49] A n E s t a b l i s h e d b u t D i f f e r e n t i a t e d K i d n e y E p i t h e l i a l Cell
Line (MDCK)
By MARY TAUB and MILTON H. SAIER, JR.

The Maden Darby canine kidney cell line (MDCK) is one of the best
characterized epithelial cell lines available for study. MDCK was derived
from the kidney of a normal male cocker spaniel in 1958. Initially, the
cells were used primarily for viral production. 1 However, subsequently,
the differentiated properties of MDCK were examined. Although maintained in culture for almost 20 years, the cells still bear close resemblance
to transporting epithelia present in the distal tubule of the kidney. 2 Thus,
the cells may be used to answer the following questions about kidney
epithelia: (1) What are the mechanisms involved in the regulation of
epithelial cell growth? Previous studies concerned with growth regulation
have been concerned primarily with fibroblast cultures, a and the observations made with fibroblasts do not necessarily apply to epithelial cells. (2)
What are the mechanisms responsible for the biogenesis of the polarity in
these cells? Epithelial cells in general have two distinct surfaces, serosal
J C. R. Gaush, W. L. Hard, and T. F. Smith, Proc. Soc. Exp. Biol. Med. 122, 931 (1966).
z M. H. Saier, Jr., M. J. Rindler, M. Taub, L. Shaeffer, and L. M. Chuman, Proc. Natl.
Acad. Sci. U.S.A. (submitted for publication).
3 B. Clarkson and R. Baserga, eds., "Control of Proliferation in Animal Cells." Cold Spring
H a r b o r Lab., Cold Spring Harbor, N e w York, 1976.

ISBN 0-12-181958-2
[49]
KIDNEY EPITHELIAL CELLS
553
and mucosal, which have distinct structures and enzyme markers. The
mechanisms by which these enzyme markers segregate to the two surfaces are not well understood. (3) What are the mechanisms by which
solutes are transported across kidney epithelia? Studies of kidney function using MDCK are not only feasible, but have the advantage over in
vivo studies that the cells in culture form a homogenous population that
can be cultured indefinitely; they can be grown readily to sizable populations for biochemical studies and are amenable to genetic analysis using
the techniques available to somatic cell geneticists. 4
Several studies have been conducted concerning the differentiated
processes in MDCK cells related to salt and water transport. That MDCK
cells transport solutes in a vectorial manner, from the mucosal to the
serosal surface, was first suggested by Leighton5 who observed the formation of multicellular blisters or domes in confluent MDCK cultures. The
blisters are groups of cells in the monolayer slightly raised from the tissue
culture dish surface. Microscopically, the cells in the blisters are out of
the focal plane of the majority of the cells in the monolayer. Blister formation can be accounted for by the transport of salt and water from the
mucosal surface of the cells (facing the culture medium), to the serosal
surface (facing the dish surface). Presumably, the solutes that have accumulated between the cell layer and the plastic surface give rise to
hydrostatic pressure, which~causes portions of the monolayer to be elevated from the dish surface, i.e., blister formation. This explanation is
supported by both morphological and electrophysiological studies.
A prerequisite for vectorial transport is that the cells have the structural polarity characteristic of transporting epithelia of the kidney; indeed, electron microscopic studies indicate that microviUi are localized
exclusively to the mucosal (upper) surface of the MDCK cells. 6 Adjacent
cells are interconnected by tight junctions. The vectorial transport of salt
and water by MDCK cells was first demonstrated by Misfeldt et al. employing a Ussing chamber. 6 They detected a spontaneous potential difference across the MDCK cell monolayer, as well as net water flux from the
mucosal to the serosal surface of the monolayer.6 More recent studies also
indicate that the cells vectorially transport certain amino acids, r
In addition to studies of differentiated functions, both the in vitro hnd
the in vivo growth characteristics of MDCK have been examined. In vitro,
MDCK cells have a saturation density typical of transformed cells, have a
4 L. H. Thompson and R. M. Baker, Methods Cell Biol. 6, 209 (1973).
5 j. L. Leighton, W. Estes, S. Mansukhani, and Z. Brada, Cancer 26, 1022 (1970).
D. S. Misfeldt, S. T. Hamamoto, and D. R. Pitelka, Proc. Natl. Acad. Sci. U,S.A. 73,
1212 (1976).
r M. White and D. S. Misfeldt, personal communication (1977).
554
SPECIFIC CELL LINES
[49]
low serum requirement, and can grow in methocel suspensions.Sin v i v o ,

when injected into baby nude mice M D C K cells form nodules structurally
very similar to the intact kidney, z Cells injected into adult nude mice
remain viable over time, but nodule growth is not observed. 9 These observations suggest that M D C K cells respond to growth regulatory signals
present in baby mice, but that the signals are deficient in adult animals.
M D C K Cell Growth and Maintenance
M D C K cells are maintained as monolayers at 37 in a 5% CO2 in air
atmosphere, in Dulbecco's modified Eagle's medium with 10% fetal caff
serum, penicillin (1.9 102 IU/ml), streptomycin (0.2 mg/ml), and ampicillin (25 /zg/ml) (DME). U n d e r these culture conditions M D C K cells
have a growth rate of one doubling per 22 hr and reach a final saturation
density o f 4 105 cells/cm 2. The cells are cuboidal in shape and form
continuous tight junctions between adjacent cells, similar to other epithelial cells. At confluency the groups of cells have joined to form a continuous epithelial sheet on the dish surface. In such confluent cultures t h e
multicellular blisters or domes first b e c o m e apparent. The regulation of
blister formation may be studied under these conditions by simply adding
reagents o f interest to the culture medium and microscopically determining their effects on the number and size of~such blisters as a function of
time. 10
As an alternative to monolayer cultures, M D C K cells can be grown as
papillary cysts in suspension, a method developed by John Valentich. 11To
initiate cyst culture development, cells are grown to late confluency (1.21.4 107 per 32-ounce bottle) in DME. The medium is changed 24 hr
prior to harvesting the cells for cyst formation. The next day fragments of
the cell sheet are detached by scraping with a Pasteur pipette, and
aliquots of the resulting cell suspension are placed in 25-ml E r l e n m e y e r
flasks (aliquots are 4 ml per 50 ml DME, when using a 32-ounce prescription bottle). The flasks are shaken at 70 rpm at 37 . Twenty-four hours
later cysts appear and increase in size and n u m b e r o v e r a 3-4 day period.
Solute Transport
The uptake of labeled solutes is determined using M D C K cell
monolayers in 35-ram diameter plastic dishes containing DME. For up8 c. D. Stiles, W. Desmond, L. M. Chuman, G. Sato, and M. H. Saier, Jr., CancerRes. 36,
3300 (1976).
s C. D. Stiles, W. Desmond, L. M. Chuman, G. Sato, and M. H. Saier, Jr., Cancer Res. 36,
1353 (1976).
1oj. D. Valentich, R. Tchao, and S. Leighton, J. Cell Biol. 70, 330a (1976).
11R. Tchao, personal communication (1976).
[49]
555
take studies, confluent cultures are washed 3 times with the uptake buffer
[phosphate-buffered saline (PBS) or 1 mM Tris made isotonic to PBS using
sucrose], and the ceils are incubated with labeled solute at room temperature. After the uptake period, the solution is removed by aspiration, and the
monolayers are washed 3 times with 3 ml of incubation buffer. The labeled
material is solubilized by incubation for 1 hr in deionized water (for cation
transport) or in 5% trichloroacetic acid (for sugar or amino acid transport). The solubilized material is added to scintillation vials containing a
Triton-toluene based scintillation fluid, and the radioactivity is determined. Total label accumulated intracellularly is measured in duplicate
and is corrected for zero time uptake, i.e., for unincorporated material not
removed by the washing procedure.
The label present per cell population is standardized by either the
protein content per dish, determined using the Lowry protein assay as
modified for tissue culture by Oyama and Eagle, 12 or by the total cell
number per dish, determined using a Coulter counter. In order to determine the concentration of solute present intracellularly, the volume is
estimated by measuring the average MDCK cell size with a Coulter
counter. Pollen particles are used as standards, and a 75% water content
is assumed.13 The protein content is used to determine the intracellular
fluid volume per dish since a direct correlation has been made between
these two factors as a function of cell density, lz
Growth for Studies in Ussing Chamber
The transepithelial transport of salt and water by MDCK cells has
been studied with a Lucite Ussing chamber (Fig. 1) that was originally
developed for the measurement of the transepithelial electric potentials
generated by frog skin. For the Ussing chamber studies MDCK
monolayers are grown on Millipore filters, so that the monolayers resemble a continuous epithelium. Membrane filters (Millipore, HAMK 02512)
are boiled 5-10 min to remove the wetting agent and are attached to
plastic tissue culture dishes while still wet using droplets of Millipore
cement (Formulation No. 1). MDCK cells are distributed into the dishes
containing the filters and DME. When the cells on the filters have grown
to form a continuous epithelial sheet, a filter may be gently detached from
the dish and placed between Lucite Ussing chambers containing either
culture medium without serum or a balanced salt solution (for example,
Hanks's balanced salt solution). 6
The Ussing chamber (designed originally by Ussing and Zerahn 14) is
12V. I. Oyama and H. Eagle, Proc. Soc. Exp. Biol. Med. 9, 305 (1956).
13D. O. Foster and A. Pardee,J. Biol. Chem. 244, 2675 (1969).
14H. H. Ussing and K. Zerahn, Acta Physiol. Scand. 23, ll0 (1951).
556
SPECIFIC CELL LINES
[49]
FIc. 1. Diagrammatic sketch of Ussing chamber. Abbreviations used in the figure are as
follows: C, Celluloid or Lucite chamber containing salt solution; S, skin or membrane filter
with MDCK cells; a, air inlets; A and A ', Agar-Ringer bridges, connecting outside and inside
solutions with calomel electrodes and lx~tentiometer;B and B', Agar-Ringer bridges used for
applying outside emf; I.), battery; W, potential divider; P, tube potentiometer.
illustrated in Fig. 1. The epithelium is placed in a space b e t w e e n two
adjacent c h a m b e r s , t h e r e b y separating the solutions in these c o m p a r t ments. Saturated K C l - c a l o m e l electrodes are connected to the solutions
in the two c h a m b e r s b y means o f 1 M N a C I - 4 % agar bridges. A potent i o m e t e r is c o h n e c t e d to these half cells to determine the potential difference b e t w e e n the solutions facing either the mucosal or serosal sides of
the cell layer.
In applying this apparatus to the study of M D C K cells, a B e e k m a n
model 76 p H m e t e r is used as a p o t e n t i o m e t e r and the potential is recorded with a B e c k m a n model 1005 recorder. ~ The s y s t e m has to be
carefully balanced, i.e., the c o m p o n e n t s present on either side of the
bridges h a v e to be symmetrical, in order to m e a s u r e spontaneous potentials. Relatively high potentials induced b y external currents m a y be measured if a Simpson multimeter is placed in series with the potentiometer.
The spontaneous electrical potential of the M D C K cells has been determined w h e n identical solutions were present on either side of the Ussing chamber. 6 I m m e d i a t e l y after initiating the readings, relatively high
potentials are observed; a decline follows after 3-10 min. At this lower
steady-state level a potential of 1.42 ___ 0.26 m V with a negative serosal
surface was determined. Due to the wide range o f readings multiple determinations are n e c e s s a r y to obtain meaningful estimations of the sport-
[49]
557
taneous transepithelial electrical potential. MDCK epithelial layers are

leaky and have a much lower spontaneous current than frog skin or toad
bladder.
The electrical potential across the cell layer also may be determined
when an external current is applied. As illustrated in Fig. 1 an external
power supply is connected in series with the Ussing chamber, and the
voltage is adjusted to the desired amplitude using a potential divider (W).
When applying such an external current, Ussing and Zerahn demonstrated a linear relationship between the Na flux and the measured potential difference across frog skin? 4 In frog skin the measured current is
equivalent to the Na influx rate when the external voltage is adjusted so
that the potential is zero. However, similar relationships have not been
established with the MDCK cells; a determination of the Na influx rate
from the spontaneous potential difference may actually underestimate the
total rate of Na influx, due to the simultaneous flow of cations and anions
across the monolayer.
Misfeldt et al. have demonstrated a constant linear relationship between applied current and transepithelial potential when the serosal side is
negative; 6 a similar relationship holds when the mucosal side is negative,
provided that current densities are below 800/.tA. Under these conditions
the transepithelial electrical resistance may be determined.
The chamber apparatus also has been used to determine the rate of
water flux. 6 Capillary tubes are positioned horizontally from the center to
the outside of each chamber. The chambers are filled with appropriate
solutions, evacuated of air, and the vectorial water flux is measured from
the position of the meniscus of water present in the capillary tubes.
Finally, the Ussing chamber can be used to measure the vectorial flux
of labeled solutes. Labeled material is placed in the buffer adjacent to one
side of the monolayer, and the label present in the buffer in the other
chamber may be measured as a function of time. To establish vectorial
flux one must demonstrate a difference in the rate of movement of labeled
solute from the serosal to mucosal side as compared to the mucosal to the
serosal side of the Ussing Chamber.
cAMP Measurements
cAMP production by MDCK cells is determined by measuring intracellular cAMP or the cAMP excreted into the culture medium using the
Gilman receptor protein binding displacement assay. 15 To assa~ cAMP
production confluent monolayers of MDCK (in 30-mm diameter dishes)
are washed 2 times with PBS and are incubated with the compounds to be
~ A. G. Gilman and F. Murad, this series, Vol. 38, p. 49.
558
SPECIFIC CELL LINES
[49]
tested in serum-free DME at 37 in a 5% CO~ in air incubator. To facilitate

studies of the effects of hormones on MDCK, the cells are incubated
concomitantly with the hormone of interest and isobutyl methylxanthine.
The latter agent inhibits cAMP phosphodiesterase action. At the end of
the incubation period the culture medium is removed, and intracellular
cAMP is extracted from the monolayers with boiling water. Extraction
with boiling water is a simpler and more accurate procedure than extraction procedures utilizing trichloroacetic or perchloric acid which require
subsequent extractions to remove the denaturing agent from the cell extracts.15 To standardize cAMP measurements protein is determined by a
modified Lowry procedure.12
The Gilman procedure assays unlabeled cAMP by its ability to compete with labeled cAMP for binding to the regulatory subunit of a cAMP
protein kinase partially purified from bovine skeletal muscle. The binding
is enhanced in this assay by adding a heat-stable protein inhibitor of the
kinase. After the binding has reached equilibrium, free and bound nucleotides are separated by passing the solution through cellulose ester (Millipore) filters. The concentration of labeled nucleotide that the sample
displaces from the binding protein is determined, which permits the concentration of unlabeled cAMP in the sample to be estimated from standard
curves.
Standard curves are constructed in which log (total [3H]cAMP bound)

is plotted versus log (total cAMP present, labeled and unlabeled), for each
concentratibn of [all]cAMP used. Such plots are linear at saturating
cAMP concentrations, cAMP is assayed at saturating concentrations of
the nucleotide to maximize reproducibility of the assays and to facilitate
estimation of cAMP content. The sensitivity of the assay is enhanced
when lower [3H]cAMP concentrations are employed, but accuracy
suffers.
Routinely, the standard, or unknown, compounds are added to the
tubes (10 75 mm) containing 5/zl of 0.1 M sodium acetate at pH 4.0, 5
/~1 [H3]cAMP (usually 1 pmol, 10,000-20,000 cpm), and 5/zl of inhibitor
protein (9-15/zg). Prior to adding binding protein the solution containing
these components has a total volume of 20 txl. While the tubes are on ice,
30/zl of a solution containing binding protein and bovine serum albumin
(10-15/~g) are added to each tube, raising the fluid volume to 50/~1. The
quantity of binding protein used is kept minimal to ensure saturation of
the protein; saturation may be defined as conditions under which less than
30% of the total [3H]cAMP is bound after a 1-hr incubation at 0. The
reaction mixtures are diluted with 1 ml of 20 mM potassium phosphate,
pH 6, at 0-3 and are passed through 25-mm cellulose ester Millipore
[49]
559
filters (previously rinsed with buffer); the filters are washed with buffer to
remove unbound cAMP. To determine [all]cAMP bound, filters placed in
scintillation vials are dissolved with 1 ml of methyl cellusolve, scintillation
fluid is added (the ratio PPO/POPOP : methyl cellusolve is 3 : 1), and the
vials are counted in a scintillation counter.
Comments
MDCK has both the morphological and enzymic properties of epithelial cells from the distal tubule of the kidney.1 In addition to having high
Na+/K ATPase activity," MDCK cells possess the kidney-specific enzymes alkaline phosphatase and leucine aminopeptidase. The peptidase
has been localized primarily to the mucosal membrane by immunofiuorescence. While the cells lack proximal markers (Na*-dependent sugar
transport, p-aminohippurate transport; disaccharidase), they do have a
distinctive distal marker, responsiveness to the antidiuretic hormones.
When present at physiological concentrations, lysine and arginine vasopressin as well as oxytocin stimulate cAMP production by adenylate cyclase. TM Similarly, in the presence of PGE1, PGE2, glucagon, or cholera
toxin cAMP production is enhanced in MDCK. The stimulation of adenylate cyclase by vasopressin is apparently subject to negative cooperativity; however, these studies indicate that the cells have antidiuretic hormone (ADH) receptors specific to distal kidney cells, and that a single
kidney epithelial cell type is responsive to multiple hormonal agents.
When ADH is present in the distal portion of the kidney, increased
water flux from the kidney lumen into the mucosal surface of the cells is
observed.17 If the MDCK cells provide a good model system for the study
of kidney epithelial function one would expect that the cells would respond to treatment with vasopressin (or to dibutyryl cAMP since the
vasopressin response is cAMP mediated) by increased rates of water
transport. That cAMP levels in MDCK cause increased vectorial fluid
transport is suggested by the observation that incubation of MDCK with
dibutyryl-cAMP results in increased blister formation.l These observations indicate that the MDCK cells provide a novel model system for the
study of distal tubular cell function.
The primary function of distal cells is to transport salt and water,
thereby regulating body levels of these compounds. Misfeldt's experiments indicate that in MDCK cells, as in the kidney, salt and water
16M. J. Rindler, L. M. Chuman,and M. H. Saier, Jr., in preparation.
17E. Koushanpour,"Renal Physiology,Principlesand Functions,"p. 336. Saunders,Philadelphia, Pennsylvania,1976.
560
[SO]
transport occur transepithelially, from the mucosal to the serosal surface. ~ The transport of Na + by the Na + channel (presumably located on
the mucosal surface) has been examined, both kinetically and genetically, is 22Na+ uptake studies indicate that a saturable Na + transport system is present in MDCK cells that is energy independent and insensitive
to ouabain inhibition. 18 Na + uptake by the channel is inhibited by a number of monovalent cations including Li +, K +, Rb +, guanidine, and
amiloride. 22Na+ uptake studies indicated that the system is Subject to two
exchange processes: (1) exchange transport with intracellular Na +, and
(2) Na+/proton antiport. Calcium is implicated as an important regulatory
molecule for the channel, inhibiting Na + uptake when present extracellularly, and stimulating Na + uptake when present intracellularly.
Variant clones of the MDCK line have been isolated by their resistance
to killing by 5 10-4 M amiloride. 18 Since the clones are stable over a
6-month growth period under nonselective conditions, and since their
frequency is increased by mutagens, they appear to have arisen by genetic
mutations. The clones exhibit reduced uptake of Na +, amiloride, and Rb +
(ouabain-insensitive uptake). Since both Rb + uptake by the Na+/K
ATPase and a-amino-isobutyrate uptakes are not affected by the mutation,
the decreased ionic uptake rates may be tentatively explained as resulting
from a mutation affecting the Na channel. The reduced uptake of
amiloride by the Na channel in the resistant cells may confer resistance
to killing by amiloride by decreasing the intracellular concentration of the
drug below the cytotoxic level. Reduced salt transport in these cells apparently results in a decreased propensity for vectorial fluid transport, as
indicated by the lack of blister formation in the variant cells.
is M. Taub, M. S. Rindler, and M. H. Saier, Jr., Cell Biol. 75, 367a (1977).
[50] L a r g e - S c a l e P r e p a r a t i o n o f C h o n d r o c y t e s
By MICHAEL KLAGSBRUN
Cartilage is an unusual tissue in several respects. It has neither blood

vessels nor nerves and contains only one type of cell, the chondrocyte.
Thus, it is relatively easy to obtain a homogeneous population of cells
from cartilage without having to resort to cell fractionation techniques.
Chondrocytes are useful for the analysis of proteoglycan synthesis 1 and
i M. Muto, M. Yoshimura, M. Okayama, and A. Kaji, Proc. Natl. Acad. Sci. U.S.A. 74,
4173-4177 (1977).
METHODS IN ENZYMOI.X)GY,VOL. LVIII

All rightsof reproduction in any formreserved.
ISBN 0-12-181958-2
[50]
PREPARATION OF CHONDROCYTES
561
collagen synthesis. 2 In addition, the metabolic effects of polypeptide

growth factors, a,4 hormones, 5 and vitamins 6 on chondrocytes can be studied. Chondrocytes can be readily grown in cell culture in plastic dishes
and, unlike other normal diploid cells, can be maintained and grown in
suspension, v,8 Furthermore, it has been demonstrated that chick embryo
chondrocytes will grow in soft agar. a Growth in suspension and growth in
agar are properties usually ascribed to transformed cells.
Chondrocytes are obtainable in large numbers from bovine and
human sources. The shoulder of a newborn calf (1-14 days old) is an
excellent source of large amounts of cartilage and is readily available from
slaughterhouses that prepare veal. Since sterile conditions cannot be
maintained in the slaughterhouse, care must be taken to avoid further
contamination of the shoulder before its arrival in the laboratory. The
shoulder is placed in a sterile plastic bag containing 0.15 M NaC1, penicillin (200 U/ml), and streptomycin (200/zg/ml) immediately after slaughter
and packed in ice. The shoulder is then transferred to a laminar flow hood
and placed on a sterile cloth. Within the calf shoulder are two major
sources of cartilage. One is the scapula, which has a piece of cartilage
measuring about 3 cm x 8 cm. The other source is the cartilage on the
articular surfaces of the bones that make up the shoulder joint.
The surrounding muscle and connective tissue must be removed in
order to obtain clean cartilage. This is done by dissection with scalpels
holding sterile stainless-steel #10 and #21 surgical blades (Bard-Parker).
The scapular cartilage is enclosed in a fibrous connective tissue, the
perichrondrium, which is removed by scraping the scapular cartilage with
a #21 blade. Once the scapular cartilage is clean, it is cut away from the
scapular bone. However, the region of about 0.5 cm wide that is proximal
to the bone is discarded, because this section of cartilage is slightly
vascular.
The articular cartilage, which does not have a perichondrium, is easily
removed from the articular surfaces of the long bones by cutting off slivers
using the #21 blade.
z K. v o n d e r Mark, V. Gauss, H. v o n d e r Mark, and P. Muller, Nature (London) 267,

531-532 (1977).
z M. Klagsbrun, R. Langer, R. Levenson, S. Smith, and C. Lillehei, Exp. Cell Res. 105,
99-108 (1977).
4 M. T. Corvoi, C. J. Malemud, and L. Sokoloff, Endocrinology 90, 262 (1972).
S. Meier and M. Solursh, Gen. Comp. Endocrinol. 18, 89-97 (1972).
M. Solursh and S. Meier, Calcif. Tissue Res. 13, 131-142 (1973).
K. Deshmukh and B. D. Sawyer, Proc. Natl. Acad. Sci. U.S.A. 74, 3864-3868 (1977).
s Z. Nevo, A. Horwitz, and A. Dorfmann, Dev. Biol. 28, 219-228 (1972).
a A. L, Horwitz and A. Dorfmann, J. Cell Biol. 45, 434-438 (1970).
562
SPECIFIC CELL LINES
[50]
Both scapular and articular cartilage are diced into pieces approximately 0.5 cm 0.5 cm and washed twice with phosphate-buffered saline
(Gibco) containing penicillin (200 U/ml) and streptomycin (200 /zg/ml).
The diced cartilage is incubated at 37 in 25 ml of 2 mg/ml clostridial
collagenase (Worthington CLS II, 140 U/mg) made up in the phosphatebuffered saline--penicillin-streptomycin mixture. The incubation is carried
out in a 60-mm petri dish that is gently rotated on a rotating apparatus
(Arthur Thomas Co.) for 12-24 hr or until the pieces of cartilage are
completely dissociated and the chondrocytes are liberated from the matrix. The suspension of chondrocytes is passed through a 153-/zm nylon
sieve (Tetko) to remove debris and undissolved cartilage fragments. The
cells are subsequently washed twice with the phosphate-buffered salinepenicillin-streptomycin mixture by means of centrifugation. After the
second wash, cells are counted in a Coulter counter (Coulter Electronics).
A typical scapular or articular chondrocyte preparation contains
1-4 108 cells per shoulder.
A source of large numbers of human chondrocytes is the costal cartilage obtained from children undergoing surgical correction of pectus excavatum, a disorder in which rib cartilage is deformed. These chondrocytes are particularly easy to isolate because the perichondrium, muscle,
and connective tissue are removed from the cartilage during surgery.
Otherwise, the isolation procedure is the same as for the bovine
chondrocytes.
For cell culture, cells (3 106 per 75 cm 2 culture dish) are plated
sparsely in the presence of 10 ml of Dulbecco's modified Eagle's medium
containing 4.5 goiter glucose (Gibco), 10% calf serum (Colorado Serum
Co.), penicillin (50 U/ml), and streptomycin (50/zg/ml) and transferred to
a 37 water-jacketed COz incubator (National Appliance Co.). Chondrocytes are round cells in situ. They gradually flatten in the culture dish over
a period of 2-4 days; growth begins after this initial lag period. In confluent cultures, two patterns of growth are observed; some cells in the dish
grow in a monolayer while others pile up and form nodules. The number
of nodules increases with repeated feeding of the cells. If the cells are
stained with 2% aqueous toluidine blue, metachromasia may be observed
in the nodules, indicating that active synthesis of proteoglycans has taken
place.
Chondrocytes cannot be maintained indefinitely in culture flasks once
they become confluent because the synthesis of the sulfated proteoglycans leads to a sharp decrease in the pH of the media. The cells are
harvested and either subcultured in new flasks by plating sparsely or are
frozen. Harvesting of chondrocytes is accomplished by detaching the
cells with 0.1% trypsin (Gibco) made up in phosphate-buffered saline
lacking calcium and magnesium (Gibco). The medium is removed from a
[50]
PREPARATION OF CHONDROCYTES
563
T-75 culture flask, and 10 ml of 0.1% trypsin solution are added. The cells
are incubated with trypsin at room temperature for about 5 min or until
they round up and are loosely attached to the flask. The trypsin is carefully removed, and 10 ml of fresh media are added. The flask is struck
several times with the palm of the hand in order to detach the cells.
Chondrocytes are frozen in a solution of Dulbecco's modified Eagle's
medium containing 10% calf serum and 7.5% dimethyl sulfoxide (Fisher
Scientific Co.). A stock solution of dimethyl sulfoxide (15%, v/v) in Dulbecco's modified Eagle's medium containing 10% calf serum is prepared.
One milliliter of the dimethyl sulfoxide solution is mixed with 1 ml of
chondrocytes (2.5 10~ cells/ml) resuspended in Dulbecco's modified
Eagle's medium containing 10% calf serum. The cells are transferred into
2-ml ampules (Wheaton Scientific) that are subsequently sealed. The ampules are gradually frozen at - 9 0 in a Revco freezer and are stored either
in the Revco freezer or in a liquid nitrogen tank (Cryogenics-East Co.).
Bovine chondrocytes will remain viable in liquid suspension culture
for at least 7 days. Maintenance in liquid suspension culture can be used
as a technique to remove contaminating fibroblasts since these cells will
not survive in suspension for more than 2 or 3 days. 8 One technique for
keeping chondrocytes in liquid suspension is to coat the bottom of a petri
dish with 0.5% agar. The chondrocytes are unable to plate out in the agar
and remain suspended. Agar plates are prepared as follows: A 1.0% solution of agar (Difco Bacto-agar) is made up in water and sterilized by
boiling for 30 min. The 1% agar is cooled to 45 and mixed with an equal
volume of warm 2 basal Eagle's medium with Earle's balanced salts
(Microbiological Associates) containing 20% calf serum, penicillin (100
U/ml), and streptomycin (100 /zg/ml). Two milliliters of the 0.5% agar
solution are poured into a 35-mm petri dish. After the agar hardens, 3 ml
of chondrocytes (2 l0 s cells/ml) are resuspended in Dulbecco's modified Eagle's medium containing 10% calf serum and plated onto the agar.
The agar plates are maintained at 37 in a water-jacketed CO2 incubator.
A major problem in chondrocyte culture is the possibility of contamination by fibroblasts arising from the perichondrium. Nevo e t al. 8 have
suggested that the fibroblasts can be eliminated by maintaining the chondrocyte preparation in suspension culture for 2 or 3 days. Fibroblasts do
not survive in suspension culture.
In general, chondrocyte cultures can be characterized by several criteria. These include a high level of ~5SO4 uptake into chondroitin sulfate, 8
metachromatic staining with toluidine blue, 9 and the ability of chondrocytes to survive in suspension culture, r-a
Cartilage in v i v o contains Type II collagen that is genetically distinct
from the Type I collagen found in skin, bone, and tendon. Thus, chondrocytes should be the only cell capable of producing Type II collagen. How-
564
SPECIFIC CELL LINES
[51]
ever, the type of collagen produced by chondrocytes in culture has been

shown to be dependent on extracellular conditions. 2'7"1 In general,
monolayer cultures produce Type I and suspension cultures produce Type
II. 7 Therefore, care must be taken in using collagen typing as a criterion
for the presence of chondrocytes.
1o p. D. Benya, S. R. Padilla, and M. E. Nimni, Biochemistry 16, 865-872 (1977).
[5 1] M e l a n o m a Cells
B y JOHN PAWELEK
Melanocytes synthesize melanin in a reaction controlled by the enzyme monophenyl, dihydroxyphenylalanine : oxygen oxidoreductase (EC
1.14.18.8), commonly referred to as either tyrosinase or dopa-oxidase.
Melanocytes in the skin of adult mammals divide infrequently, perhaps
once a year, and attempts at adapting these cells to culture have met with
only limited success. On the other hand, malignant melanoma cells divide
as often as once a day, and several lines have been established in culture.
Commonly used lines are the B16,1 Harding-Passey, 2 and Cloudman3
strains (mouse); the Fortner (RPM1) strain 4 (hamster); and many human
strains. 5"6 Some of these lines are available commercially. 7,s
The Cloudman $91 melanoma was established as a transplantable
tumor in DBA 2/J mice in the 1930sa and was adapted to culture in the
1960s. 3 In 1973 it was reported that the Cloudman cells responded in
culture to melanotropin (MSH) with increases in tyrosinase activity,
melanin formation, and generation time, as well as changes in morphology. ~ Since the original findings, many studies have been carried out on
factors that regulate pigmentation and proliferation of these and other
melanoma cells in culture (for reviews, see Pawelek~l'~2).
i F. Hu and R. R. Cardell, J. Invest. Dermatol. 42, 67 (1964).
H. E. Harding and R. D. Passey, J. Pathol. 33, 417 (1930).
3 y. Yasumura, A. H. Tashjian, and G. H. Sato, Science 154, 1186 (1966).
4 G. E. Moore, D. F. Lehner, Y. Kikuchi, and L. A. Less, Science 137, 986 (1962).
5 R. E. Gerner, H. Kitamura, and G. E. Moore, Oncology 31, 31 (1975).
6 L. A. Quinn, L. K. Woods, S. B. Merrick, M. Arabasz, and G. E. Moore, J. Natl. Cancer
Inst. 59, 301 (1977).
7 The American Type Culture Collection, 12301 Parklawn Drive, Rockville, Maryland.
a The Jackson Laboratory, Bar Harbor, Maine.
g A. M. Cloudman, Science 93, 380 (1941).
lo j. Pawelek, G. Wong, M. Sansone, and J. Morowitz, Yale J. Biol. Med. 46, 430 (1973).
11 j. Pawelek, J. Invest Dermatol. 66, 201 (1976).
12 j. Pawelek, Pigm. Cell 4 (in press).
M~THODS
IN F_MZYMOLOOY,VOL.LVIII
Copyright1979by AcademicPress,Inc.
All rights of reproductionin any form reserved.
ISBN 0-12-181958-2
[51]
MELANOMA CELLS
565
The purpose of this article is to describe techniques for culturing Cloudman $91 in monolayers and clones, and to outline some of the problems
unique to melanin-producing cells that are encountered in normal culturing procedures and in some biochemical analyses.
Cell Culture
Culture procedures are straightforward. For monolayers, cells are
placed into plastic flasks or petri dishes that have been coated for cell
culture (available commercially). The culture medium consists of H a m ' s
F10 medium 13 supplemented with 2% fetal calf serum and 10% horse
serum. Antibiotics can be added if desired, but it is important to first
determine whether a specific antibiotic has an effect on pigmentation. For
example, tetracycline and bacitracin both increase pigment formation
whereas streptomycin and penicillin do not. Fresh culture medium is
added 3 times per week. Subculturing of cells involves removal of the
medium and addition of ethylenediaminetetracetic acid (EDTA) (1 m M or
less) in a Ca- and Mg-free balanced salt solution. After about 10 min at
room temperature, cells detach from the culture vessel and are collected
by centrifugation. The use of trypsin or other proteases should be avoided
in the harvest procedure, particularly if regulatory processes involving
the outer cell surface are being studied.
For culturing cells in clones, we use the "dilute agar colony m e t h o d "
of Chu and Fischer. TM This involves culturing cells in medium supplemented with 0.12% agar; the concentration of horse serum is increased to
15%. Culture vessels can be in a variety of sizes and shapes, but for
convenience we use plastic screw-cap test tubes. The clones grow in
suspension, and it is therefore unnecessary to use vessels that have been
coated for cell culture. Clones are generally visible to the naked eye after
1 week and can be picked with a Pasteur pipette after 2-4 weeks.
Isolation of Amelanotic Mutants

Pigmentation has long been used as a genetic marker as evidenced by
the fact that more than 70 genes for pigmentation have been identified in
the mouse. 15 Simple procedures have been developed for isolating lines of
13R. G. Ham, Exp. Cell Res. 29, 515 (1963).
14M. Y. Chu and G. A. Fischer, Biochem. Pharmacol. 17, 753 (1968).
is W. C. Quevedo, in "Biology of Normal and Abnormal Melanocytes" (T. Kawamura,
T. B. Fitzpatrick, and M. Seiji, eds.), p. 99. Univ. of Tokyo Press, Tokyo, 1971.
566
SPECIFIC CELL LINES
[51]
EFFECTS OF ETHYLMETHANESULFONATE ON THE APPEARANCE OF

AMELANOTIC CLONESa
EMS concentration
(mg/ml culture
medium)
Total
clones
Amelanotic
variants
Relative
survival (%)
0
0.02
0.05
0.10
0.20
0.50
1.0
6110
6400
2292
1846
1438
100
0
0
0
11 (0.5%)
51 (2.7%)
27 (1.9%)
5 (5 %)
0
100
i05
38
31
24
1.7
0
a Cells in monolayer were incubated with ethylmethanesulfonate (EMS) at the concentrations indicated. After 2 hr the EMS was removed, the cells were rinsed several
times with fresh culture medium, and allowed to pass through three generations (6
days). Cells (3 x 104) from each category were then cultured in petri dishes (100-mm
diameter) by the method of Chu and Fischer.~3 After 4 weeks the total number of clones
and the number of amelanotic variants were determined by counting with the aid of a
dissecting microscope.
amelanotic melanoma cells from large populations of isogenic melanotic

cells. 16 In essence, cells are treated with an agent known to induce mutations, cultured for two to three generations so that mutations can be
expressed, and cloned by the "dilute agar colony method" mentioned
above.14 After 3-4 weeks, amelanotic clones are isolated with a Pasteur
pipette and recloned in dilute agar. The effects of ethylmethanesulfonate,
a potent mutagen, on the appearance of amelanotic clones is shown in the
accompanying table. One problem with this technique is that many of the
amelanotic clones prove to be unstable in that they eventually "revert" to
a melanotic phenotype. Although the instability itself may prove to be an
interesting problem, we have generally discarded the unstable clones and
studied lines that remain faithfully amelanotic through many cell generations. It should be noted that, in addition to mutations affecting pigmentation, a number of other mutant phenotypes can be isolated by standard
techniques. 17,18
16 j. Pawelek, J. Sansone, J. Morowitz, G. Moellmann, and E. Godawska, Proc. Natl. Acad.
Sci. U.S.A. 71, 1073 (1974).
17 j. Pawelek, M. Sansone, N. Koch, G. Christie, R. Halaban, J. Hendee, A. B. Lerner, and
J. M. Varga, Proc. Natl. Acad. Sci. U.S.A. 72, 951 (1975).
18 j. Pawelek, R. Halaban, and G. Christie, Nature (London) 258, 539 (1975).
[51]
MELANOMA CELLS
567
Cytotoxicity of Melanin Precursors
Several years ago, Lerner suggested that in the process of melanization pigment cells produce substances that are potentially autotoxic. TM
Later it was demonstrated that pigmented cells are killed when exposed to
excess tyrosine or dopa in the culture medium.l.1. ~ It has been shown
recently that among all the naturally occurring amino acids, only tyrosine,
dopa, and trytophan are toxic to pigmented cells. The toxicity of these
agents is proportional to the amount of tyrosinase activity within the cells.
5,6-Dihydroxyindole is far more toxic than tyrosine, dopa, or tryptophan,
and it is likely that this ~elanin precursor is responsible for cell death. 2
Although these findings may be potentially useful for chemotherapy of
melanoma, they present a problem when culturing melanoma cells. It is
important to choose a culture medium that is low in tyrosine, and it is for
this reason that Ham's F10 medium (tyrosine concentration is 2 rag/l)
was used for the Cloudman cells. Most other media have tyrosine
concentrations that are orders of magnitude, higher than that in Ham's
F10, and cells with high tyrosinase activity usually do poorly in these
media.
Cyclic AMP and the Cell Cycle
Low amounts of cyclic AMP markedly stimulate the proliferative rate

of Cloudman $91 cells, while higher amounts of cyclic AMP inhibit proliferatiOn. TM Thus melanotropin inhibits the growth of cells by raising intracellular levels of cyclic AMP, but insulin inhibits growth of cells by
lowering cyclic AMP levels. The results indicate that the Cloudman $91
cells require an optimal concentration of cyclic AMP in order to pass
through the cell cycle. Whereas cells exposed either to insulin or to
melanotropin cease cycling, those exposed simultaneously to both hormones continue to proliferate as if they were in ordinary culture medium.12 The basis for this phenomenon is not understood, nor is it known
whether other cell types have such cyclic AMP requirements. Serum
factors are known to alter intracellular levels of cyclic AMP. This might
explain why we and many other have encountered variations in the cycling time of isogenic cell lines when different lots of serum are used in the
culture medium.
~9 A. B. Lerner, Am. J. Med. 52, 141 (1971).
20 j. Pawelek and A. B. Lerner, Nature, in press (1979).
568
SPECIFIC CELL LINES
[SI]
Tyrosinase Activity
A simple and convenient assay for tyrosinase activity was developed
by Pomerantz. 21 Tritiated 3,5-tyrosine is available commercially. When
this molecule is oxidized to dopa by tyrosinase, [ZH]H20 is formed and is
separated from unreacted [ZH]tyrosine by the addition of activated charcoal. A second purification of [ZH]H~O consists of passing the material
that does not bind to charcoal through a Dowex-50 column. 22 Tyrosinase
can be assayed in living cells by adding [3H]tyrosine to the culture medium. The cells actively take up the tyrosine and release [ZH]H~O to the
culture medium. Thus by removing small amounts of medium and measuring [3H]H20, one can measure tyrosinase activity in the cells. Experiments of this sort must be interpreted with caution, however, because
[3H]H~O released to the culture medium is not.only a function of tyrosinase
activity but also of the rate of transport of [ZH]tyrosine into the cells, the
preexisting pool of unlabeled tyrosine within the cells, and the amount of
[ZH]tyrosine being diverted from melanin synthesis to protein synthesis.
Nonetheless, it is frequently useful to measure [ZH]H~O produced by
living cells as an assay of tyrosinase activity. To avoid the problems
mentioned above, one should then compare the results in living cells with
direct measurements of tyrosinase activity in cell homogenates. A typical
assay in cell homogenates has been described in detail. 2~
Normalization of Data
The colorimetric method of Lowry e t al. ~ for measuring protein is
used widely for the normalization of data. Unfortunately this technique
should not be employed with crude extracts of melanin-producing cells
because melanin reacts 2-3 times more strongly than bovine serum albumin (fraction V) with the Folin reagent that. is used. It is difficult and
time-consuming to quantitatively separate melanin from protein, and for
purposes of normalizing data we have found it more convenient to estimate DNA content by using Button's modification of the diphenylamine
reaction, 24 or to measure cell number in a Coulter counter.
Isolation of RNA
Melanin interferes with the isolation of RNA because it binds tightly
to RNA molecules. We found this to be a particularly vexing problem
zl S. Pomerantz, Science 164, 838 (1969).
22 A. Korner and J. Pawclck, Nature (London) 267, 444 0977).
2z O. H. Lowry, A. L. Rosenbrough, and R. J. Randall, J. Biol. Chem. 248, 190 (1951).
24 K. Burton, Biochem. J. 62, 315 0956).
[51]
MELANOMA CELLS
569
in the isolation of polyadenylic acid-containing RNA by means of

oligodeoxythymidine-cellulose affinity chromatography. By combining
extraction procedures from several laboratories we were able to separate
melanin from RNA and isolate poly (A)-containing RNA that translated
efficiently in both wheat germ and mouse ascites cell-free protein synthesizing systems.
All procedures are carried out at 4 . Cells are suspended in 4 volumes
of buffer containing 0.1 M Tris HC1 at pH 7.5, 0.15 M KC1, 15 mM
magnesium acetate, 6 mM 2-mercaptoethanol, 0.5% Triton X-100, and
0.1% bentonite. The mixture is homogenized with a loose-fitting Teflon
homogenizer and centrifuged at 30,000 g for 15 min. The supernatant fluid
is layered onto 30% sucrose dissolved in the above buffer (without Triton
X-100 or bentonite) and centrifuged (100,000 g for 5 hr). The pellet that
contains polysomes is dissolved in 100 volumes of a buffer containing 0.01
M Tris HC1 at pH 7.5, 0.15 MKCI, 1.5% sodium lauryl sulfate, and
0.02% pronase K. This mixture is incubated at 20 and triturated until the
solution becomes clear (20-40 min). The mixture is extracted 3 times with
3 volumes each of ether. The aqueous phase, which still contains melanin,
nucleic acid, and some protein, is made 70% with ethanol and stored in
the cold until complete precipitation of the macromolecules (about 2 hr).
The precipitate is dissolved in 5 volumes of 0.4 M NaCI containing 1 mM
ethylenediaminetetraacetic acid (EDTA) at pH 7.4. An equal volume of a
mixture of phenol, chloroform, and isoamyl alcohol (50 : 50 : 2) is added.
The mixture is heated to 70, .agitated on a Vortex Genie mixer (5 min at
top speed), and centrifuged (20,000 g for 15 min); melanin is located as a
tight band at the phenol-aqueous interface. The aqueous phase containing
RNA is removed and saved. The phenol phase and interface are reextracted with an equal volume of 0.4 M NaC1, centrifuged, and the aqueous
phases pooled, extracted 3 times with ether, and precipitated at a final
concentration of 70% ethanol. The precipitate, which is white and free of
melanin, is dissolved with 0.4 M NaC1 containing 1 mM EDTA at pH 7.5.
Poly (A)-containing RNA is isolated by chromatography on oligo (dT)cellulose. About 3% of the RNA is retained by the oligo (dT) and replresents authentic mRNA since it translates efficiently in cell-free systems.
Comments
Cultured mammalian melanoma cells provide excellent material for

studying the regulation of gene expression in eukaryotic cells. The oxidation of tyrosine to melanin is well understood. A single enzyme,
tyrosinase, is thought to control this pathway. A simple and rapid assay
for tyrosinase activity in living cells and in homogenates is available.
570
SPECIFIC CELL LINES
[52]
Melanin in the cells can be detected visually and thus provides an easy
assay for the expression of a differentiated function. Isogenic pigmented
cell lines can be established by clonal techniques, and from these lines
amelanotic variants can be isolated. Since at least some of these variants
should have lesions in the genes controlling melanin biosynthesis, a genetic analysis of this differentiated function should be possible. The
Cloudman melanoma line has been found to respond dramatically to
MSH, and studies with these cells promise to provide much information
concerning the mechanism of action of this hormone. It is obvious that the
factors regulating pigmentation and proliferation of melanoma cells are
complex. The molecules regulating these functions include hormones affecting cyclic AMP levels, the corresponding hormone receptors, adenylate cyclases, phosphodiesterases, cyclic AMP-dependent protein kinases
and their phosphoprotein Substrates, and possibly phosphoprotein phosphatases. Also to be considered are structural and regulatory genes that
control the synthesis of these molecules. Certainly this is an abbreviated
list that should expand rapidly as research continues.
[52] A d r e n o c o r t i c a l Y1 C e l l s
By BERNARD P. SCHIMMER
The Y1 adrenal cell line was developed by Sato and his colleagues I as
one of a series of endocrine cell cultures that retain many of their tissuespecific characteristics. 2 This cell line originated from a transplantable,
adrenocortical tumor of an LAF1 mouse; a it was cloned in 1964 after
adaptation of the tumor to culture conditions by the technique of alternate
passage between culture and animal. 4 The cell line currently is available
from the American Type Culture Collection (No. CCL 79).
The Phenotype and Its Stability
The Y1 cell line has an average doubling time of 30-40 hr, has a plating
efficiency5 of 4-10%, and reaches saturation density at approximately
2.7 l05 cells/era 2. Its morphology is epithelial-like in monolayer culture,
and its karyotype is nearly diploid with a modal number of 39 acrocentric
Y. Y a s u m u r a , V. Buonassisi, and G. Sato, Cancer Res. 26, 529 (1966).
z y. Y a s u m u r a , Am. Zool. 8, 285 (1968).
A. I. Cohen, E. Bloch, and E. Cellozi, Proc. Soc. Exp. Biol. Meal. 95, 304 (1957).
4 V. Buonassissi, G. Sato, and A. I. Cohen, Proc. Natl. Acad. ScL U.S.A. 48, 1184 0962).
5 Plating efficiency is defined as the percentage o f single cells plated which grow into
colonies.

ISBN 0-12-181958-2
[52]
ADRENOCORTICAL Y 1 CELLS
571
or telecentric chromosomes. Y1 cells behave like normal mouse adrenal

cells in several respects: (1) they synthesize A4-3-keto-C~l-steroids, principally 20a-hydroxyprogesterone and l l/3-20a-dihydroxyprogesterone,6
from cholesterol and secrete the products; (2) they increase the rate of
steroidogenesis 4- to 10-fold in response to adrenocorticotropic hormone
(ACTH); ~ and (3) they concentrate cholesterol and ascorbic acid from
their growth medium. The cell lifie, in the presence of maximally effective
concentrations of ACTH, produces steroids at the rate of 0.72 ___0.13
/zg/mg protein per hour (N = 7).
In our hands, the YI line has maintained its karyotype, growth rate,
and rate of steroidogenesis for more than 3 years in continuous culture.
However, the cell line shows considerable clonal variation in maximum
steroidogenic capacity, with six different clonal isolates producing
steroids at an average rate of 0.38 -4--0.23 txg/mg protein per hour. We
have isolated a subclone of the Y1 line and have maintained it in continuous culture for approximately 3 years. Thirteen independent clonal isolates of this subclone produce steroids maximally at an average rate of
0.36 _ 0.02 /zg/mg protein per hour, indicating that clonal variation in
rates of steroidogenesis can be reduced by recloning the Y1 line.
A number of stable variants of the Y1 cell line have been isolated in
our laboratory including clones with altered ACTH-sensitive adenylate
cyclase activity, 7 altered hypoxanthine phosphoribosyl transferase activity, 8 and altered adenosine 3' ,5 '-monophosphate-dependent protein
kinase activity, g
Growth Requirements
Usually, Y1 cells are grown in 82.5% nutrient mixture F10, TM 15%
horse serum, and 2.5% fetal calf serum. The sera used are heat inactivated
at 56 for 30 rain. Penicillin (200 U/ml) and streptomycin (200/~g/ml) may
be included as antibacterial agents. Not all of the components in this
growth medium, however, are required for growth. H A simpler nutrient
mixture, such as Eagle's minimal essential medium, ~2may be substituted
readily for the Fl0; and, in gradually adapted cells, 5% dialyzed horse
serum can replace the usual serum supplement. H Growth of Y1 cells is
R.
7 B.
s B.
a B.
10 R.
11 L.
12 H.
W. Pierson, Jr., Endocrinology 81,693 (1967).

P. Schimmer, J. Cell. Physiol. 74, 115 (1969).
P. Schimmer, J. Tsao, and N. H. Cheung, Nature (London) 269, 162 (1977).
P. Schimmer, J. Tsao, and M. Knapp, Mol. Cell. Endocrinol. 8, 135 (1977).
G. Ham, Exp. Cell Res. 29, 515 (1963).
J. Cuprak and G. H. Sato, Exp. Cell Res. 52, 632 (1968).
Eagle, Science 130, 432 (1959).
572
SPECIFIC CELL LINES
IS2]
arrested in medium containing 0.2% serum and can be reinitiated upon

addition of a growth factor isolated from bovine pituitary glands.13
Y1 cells, when grown as monolayers, require the charged surfaces of
plastic tissue culture dishes; they will not attach as well to glass. The cell
line is anchorage independent and can be cloned in soft agar without
feeder layers or can be grown in suspension culture. In suspension, Y1
cells grow into large aggregates that can be dispersed only after treatment
with a proteolytic enzyme solution.
Methods
All operations are carried out using strict aseptic bacteriological procedures. Growth medium and other reagents are sterilized before use by
filtration through sterile membrane filters (0.2 /zm pore size). Cultures
should be screened regularly for aerobic and anaerobic contaminants ineluding pleuropneumonia-like organisms (PPLO).
Initiation of Stock Cultures. Stock cultures of YI cells are grown as
monolayers in plastic tissue culture flasks (75 cm ~ area). Dispersed cells
(approximately 2 x 106) are added to 15 ml of growth medium, and the
suspension is gently rocked over the growing surface of the culture flask
to effect a uniform distribution of cells. The inoculated flasks are capped
loosely and maintained at 36.5 in a humidified atmosphere of 5% CO2 in
air. The cells attach and begin to spread out on the plastic surface within
2-3 hr. Culture medium is changed every third or fourth day, and the
progress of the culture is monitored visually at these intervals with an
inverted, phase~contrast microscope. Spent growth medium is removed
by aspiration, and fresh medium is added immediately, with care taken to
ensure that the cell monolayer is not disturbed. At the second medium
change, the volume of growth medium is increased to 30 ml per flask.
Cells reach saturation densities of 2 10r cells per bottle after 10-12
days. Cells can be maintained at saturation density for at least I week if
the growth medium is replenished regularly.
Subculture. When the stock culture reaches saturation density, the cells
are subcultured by treatment with a solution of Viokase (Viobin Corp.,
Monticello, Illinois). Growth medium is removed, and the monolayer is
rinsed 14 twice with 2-ml portions of the Viokase solution (0.1% in
phosphate-buffered saline 1~ without Ca 2+ or Mg2+ salts) and incubated at
10 D. Gospodarowicz and H. H. Handley, Endocrinology 97, 102 (1975).
~4 Rinsing is effected by gently rocking the solution over the monolayer surface and then
aspirating the liquid.
15 R. Dulbecco and M. Vogt, J. Exp. Med. 99, 167 (1954).
[52]
ADRENOCORTICAL Y1 CELLS
573
36.5 for 10 min. The residual Viokase solution coating the cells is sufficient to detach the cells from the monolayer. Treatment with Viokase is
complete when gentle tapping of the flask dislodges the cells. The cells are
washed from the flask wall with 10 ml of growth medium and are dispersed by pipetting repeatedly. One milliliter of fully dispersed cells, i.e.,
2 106 cells, is used to inoculate each new stock flask. Since the cells
settle rapidly, they must be resuspended immediately before each transfer. An Erlenmeyer flask and magnetic spin bar can be used to suspend
cells where a large number of uniform transfers are required. The cells are
stirred at a moderate speed over a nonheating stirrer.
The size of the culture vessel and the size of the inoculum can be
altered over a wide range as required by the investigator. From the data
given for plating efficiency, growth rate, and saturation density of the cell
line, the state of the culture can be approximated at any time after plating.
Storage. Y1 cells can be stored frozen in liquid N2 for at least 8 years in
growth medium supplemented with 10% glycerol. Cells, 2 106 in 1 ml of
growth medium containing glycerol, are placed in sealed ampules, packed
in a cardboard container, kept at - 7 0 for 3.5 hr, and transferred quickly
to liquid N2. To recover cells from the frozen state, one vial is thawed
rapidly in a 37 water bath, suspended by pipetting, and divided between
two tissue culture flasks (75 cm 2) containing 15 ml of growth medium.
Cells are ready for subculture after approximately 3 weeks of growth.
Plating of Adrenal Tumors. YI cells also can be grown as tumors and
studied in primary cell culture. Cells, 105--106 in 0.1 ml saline, injected into
the hind leg muscle of an LAF1 mouse will give rise to a tumor in approximately 6-8 weeks. To harvest the tumor, the mouse is killed by cervical
dislocation and cleaned with 70% ethanol. The tumor and associated leg
tissues are dissected free, transferred to a Petri dish, and washed several
times with saline solution. The tumor capsule is cut, and the tumor tissue
is removed and minced with scissors. The fine mince is triturated by
pipetting with 25 ml of Viokase (0.2% solution in nutrient mixture F10) in
5-ml portions, transferred to an Erlenmeyer flask with spin bar, stirred
gently for 20 min at room temperature, and centrifuged in conical centrifuge tubes at about 200 g for 10 min. The pellet is washed once and
resuspended in complete growth medium. Residual large tissue fragments
are allowed to settle, and the dispersed cells are collected for plating.
Cells from one tumor distributed among 40-60 tissue culture plates
(60-mm diameter) will approach confluence after 2 weeks in culture.
Propagation of Y1 cells as tumors minimizes the requirements for maintaining stocks in cell culture and also provides a convenient means of
generating cells in large yields. These advantages, however, are offset to
some degree by the heterogeneity of the tumor material.
574
SPECIFIC
CELL
LINES
[53]
Caution
Although experience to date suggests that there is no biohazard associated with the culture of these mouse adrenal tumor cells, it seems prudent to adopt containment procedures similar to those of the Medical
Research Council of Canada 16 when handling them.
16 "Guidelines for the Handling of Recombinant DNA Molecules and Animal Viruses and
Cells," Cat. No. MR21-1/1977. Minister of Supply Services, Ottawa, Canada, 1977.
[53]Long-Term Culture of Dissociated Sympathetic Neurons
By EDWARD HAWROT and PAUL H. PATTERSON

Long-term primary culture of dissociated neurons, although a relatively new technique, has already proved to be a valuable tool in studies
of neuronal development and function. Under proper conditions neurons
can be grown in culture in virtual isolation, free from the influence of
other cell types. Such cultures offer the potential for systematic manipulation of the fluid medium surrounding the cells. These neuron-alone cultures are particularly well suited to biochemical analysis since the properties of any function under study are directly attributable to the neurons
being cultured, without complicating contamination from other cell types.
Furthermore, such cultures also permit the examination of the specific
effects on neurons of co-culture with a wide variety of other cell types. In
this way, for example, neuronal interaction in culture with physiologically
important target cells can be directly compared to what is seen with the
neuron-alone cultures.
On the other hand, the major problems with primary neuronal culture
are the expense in animals and supplies, the time-consuming dissection,
and the relatively small amounts of material obtained for biochemical
analysis. Neurons do not appear to divide in culture and thus growth
consists of cellular maturation and outgrowth of neuronal processes. Furthermore, since neuron-alone cultures are considerably more fastidious
than explant cultures or cell lines, they can be difficult to reproducibly
maintain over long periods of time. Although these difficulties have been
largely overcome with some types of neurons, extreme care and some
experience with cell culture are helpful.
In fact, the particular growth conditions employed can affect not only
neuronal survival and growth, but may also influence the type of synapses
formed in culture as well as the neurotransmitters produced by cultured
neurons. In dissociated cell culture under appropriate conditions, sympathetic neurons exhibit a normal developmental differentiation and mat-
METHODS IN ENZYMOLOOY, VOL. LVIII

ISBN 0-12-181958-2
[53]
DISSOCIATED S Y M P A T H E T I C N E U R O N S
575
uration into adrenergic neurons. 1 Differentiation occurs in the nearly

complete absence of nonneuronal cells. Nevertheless, the sympathetic
neurons at the time of explantation do possess a certain restricted level of
plasticity consistent with their embryologic derivation from the pluripotent neural crest. 1This flexibility permits the experimenter to obtain, from
the same population of sympathetic neurons, cultures that display strikingly cholinergic functions as opposed to the normal adrenergic differentiation. Such a phenomenon was first observed upon co-culture of sympathetically derived neurons with certain nonneuronal cells. 2"~ Neurons
co-cultured with these nonneuronal cells produce 1000-fold more acetylcholine than control neursan-alone cultures. However, cell-to-cell contact
between the neurons and nonneuronal cells is not necessary for this effect.
Medium conditioned by incubation on cultures of appropriate nonneuronal cells, when added to neuron-alone cultures produces an increase
in neuronal cholinergic characteristics with a concomitant decrease in
adrenergic characteristics.4 Similarly, the frequency of nicotinic cholinergic synaptic interactions between the neurons is increased by growth in
conditioned medium. Since differential survival and selection of subclasses of the neuronal population have been ruled out, an inductive phenomenon is strongly suggested. 1 It is also important to note that an in vivo
counterpart to the alterations in differentiated fate of cultured sympathetic
neurons has been observed in embryonic transplantation studies of neural
crest. 1 Clearly the extracellular environment (and choice of culture conditions) is critically important in determining the differentiated fate of sympathetically derived neurons.
Sympathetic Neuron Cultures: Standard Conditions
Cell Preparation
Newborn rats (CD colony), born and shipped the same day, are obtained from Charles River Breeding Laboratories, Inc., Wilmington, Massachusetts. In general, rat pups up to 4 days old can be used for dissociated cultures. The animals are killed by a blow to the head, and the
superior cervical ganglia are removed with dissecting forceps and scissors
under a stream of sterile saline. Decapitation readily exposes the ganglia
for this manipulation. The ganglia are placed in plating medium (see be1 p. H. Patterson, Annu. Rev. Neurosci. 1, 1-17 (1978).
2 p. H. Patterson and L. L. Y. Chun, Proc. Natl. Acad. Sci. U.S.A. 71, 3607 (1974).
3 p. H. O'Lague, K. Obata, P. Claude, E. J. Furshpan, and D. D. Potter, Proc. Natl. Acad.
Sci. U.S.A. 71, 3602 (1974).
4 p. H. Patterson and L. L. Y. Chun, Dev. Biol. 56, 263 (1977).
576
SPECIFIC CELL LINES
[53]
low) and cleaned free of all surrounding tissue. The ganglion cells are then
dissociated mechanically using very fine watchmaker's forceps (#5).
Usually the ganglion sheath is first peeled away with forceps, and the
ganglion cells are then gently teased apart into smaller groups of cells. The
chunks and cells are triturated by several passages through a stainlesssteel hypodermic needle (22 or 23G, 1.5 inch) using a 5-ml disposable
plastic syringe. The resulting suspension is further agitated with a Vortex
mixer for 3-5 min (avoiding any swirling action) and undissociated pieces
removed by sedimentation or filtration. The suspension is either allowed
to settle undisturbed for 5-10 min or filtered, e.g., with a 10/xm Nuclepore filter (Nuclepore Corp., Pleasonton, California), to remove undissociated fragments. The dissociated cells are centrifuged (3-5 min at
800 g), resuspended, and plated in modified culture dishes (see below) by
adding 1 or 2 drops per dish of cells, suspended in plating medium (see
below). In general, the overall yield of neurons in culture ranges from
1-10% based on the number of neurons found in adult rat superior cervical ganglion (approximately 25,000). Normally, 40 to 50 dishes of about
3000 to 5000 viable neurons each are prepared from 40 pups. In addition to
the mechanical procedures described here, sympathetic neurons have
also been dissociated with trypsin treatment) '6
Culture Dishes
Falcon petri dishes (35 ram, #1008) are modified by boring a 1-cm hole
in the bottom of the dish and affixing a polystyrene (Lux Scientific Corp.,
Newbury Park, California; #5407, 25 mm) tissue culture cover slip with
paraffin to form a shallow well. This well protects the cells from movement of the medium, is convenient for isotopic incubations, provides better optics for inverted phase microscopy, and facilitates preparation for
electron microscopy. Prior to plating, a glass ring (1.5-cm diameter microslide ring, #6705-R12, A. H. Thomas, Philadelphia, Pennsylvania) is
placed in each dish surrounding the center well. The ring is added after
the growth medium (see below) and serves to restrict the area to which
the cells can settle to that of the coverslip. After 2 days of culture these
rings are removed.
Culture Surfaces
Several substrata are suitable for attachment and extension of
neuronal processes. Most commonly, a thin, clear film of collagen is used.
S. Varon and C. Raiborn, J. Neurocytol. 1,211 (1972).
6 K. D. McCarthy and L. M. Partlow, Brain Res. 114, 391 (1976).
[53]
DISSOCIATED SYMPATHETIC NEURONS
577
Fresh collagen is prepared monthly as an acid extract of rat tail tendons. 7

Good adhesion and culture longevity are dependent on using a fresh collagen extract that is stored at 0-4 . The clear collagen extract (one adult rat
tail per 250 ml of 0.1% v/v acetic acid) is diluted 1 4 with distilled water
and 3 drops placed onto each cover slip. The dishes are dried in a desiccator for 24 hr at room temperature a~nd sterilized by ultraviolet
irradiation.
Sympathetic neurons also can be grown on three-dimensional hydrated collagen gels. These can be made either by photoreconstitution with
riboflavin, 8 by exposure to ammonia vapors, 9 or by salt reconstitution. TM
Neuronal processes but not somas extend into the gel matrix, firmly anchoring the neurons in place. Gels offer the advantage of obviating the
need for the viscosity-increasing agent, Methocel, which is normally
added to the standard growth medium (see below).
Culture surfaces treated with polylysine or other positively charged
polymers also have been successfully used with nonneuronal as well as
neuronal cell types. H-13 Sympathetic neurons attach and grow well on
polyornithine-treated coverslips over the short term (< 14 days) but do not
survive well in long-term culture (unpublished observations). Better results are obtained if the polymer is covalently linked to an underlying
layer of dried gelatin which is activated by incubation with glutaraldehyde. A solution of swine skin gelatin (7.5 mg/ml) is prepared by heating in a boiling water bath and 2-3 drops are applied to the center well of
the culture dish. The gelatin is then allowed to dry down and form a clear
thin film. Enough glutaraldehyde (2.5% in 0.12 M sodium phosphate buffer, pH 7.4) is then added to the dish to completely cover the gelatin layer.
After about 1 hr at room temperature, the dishe'S are rinsed once with
distilled H20 and the polymer (5 mg/ml) is added to the gelatin layer in a
solution of 0.15 M sodium borate buffer, pH 8.4. Good results have been
obtained using poly-o-lysine (approximate molecular weight 150,000)
which should be less susceptible to proteolytic digestion. After overnight
incubation, the dishes are rinsed several times with distilled HeO,
sterilized, and rinsed several times with medium prior to use.
Sympathetic neurons can also be grown on pre-existing monola3~ers of
r M. B. Bornstein, Lab, Invest. 7, 134 (1958).
s E. B. Masurovsky and E. R. Peterson, Exp. Cell Res. 76, 447 (1973).
9 R. L. Ehrmann and G. O. Gey, J. Natl. Cancer Inst. 16, 1375 (1956).
10 S. D. Hauschka and I. R. Konigsberg, Proc. Natl. Acad. Sci. U.S.A. 55, 119 (1966).
11 E. Yavin and Z. Yavin, J. Cell Biol. 62, 540 (1974).
12 p. C. Letourneau, Dev. Biol. 44, 77 (1975).
1~ W. L. McKeehan and R. G. Ham, J. Cell Biol. 71,727 (1976).
578
SPECIFIC CELL LINES
[53]
nonneuronal cells as substrata: both primary cardiac 14"15 and skeletal

muscle cells TM have been used. In both cases nonneuronal cell proliferation is suppressed prior to neuronal plating by means of T-irradiation
(eCo; 5000 rad in 25-30 sec). In general, the growth medium is changed
immediately after exposure to irradiation. Radiation treatment effectively
blocks cell division while leaving intact such functions as the spontaneous
electrical activity and beating of cardiac myocytes. In some cases the
cultures were irradiated again several days after the neuronal plating to
block overgrowth by the ganglionic nonneuronal cells that are plated with
the neurons.
Culture M e d i a
1. L-15-Air. Of several commercial media tested, L:15 with modified
osmolarity and pH appears to be optimal for growth of sympathetic
neurons.17 One package (14.9 g) of L-15 powder (North American Biologicals, Inc., Miami, Florida) is dissolved in 1080 ml freshly distilled water.
(It is very important that the water be of the highest purity for all media
additions; it can be distilled from permanganate or glass distilled after
purification by charcoal and ion-exchange resins.) The following ingredients are then added: 60 mg imidazole (recrystallized from acetone),
15 mg aspartic acid, 15 mg glutamic acid, 15 mg cystine, 5 mg/3-alanine,
2 mg vitamin B~2, 10 mg inositol, 10 mg choline chloride, 0.5 mg lipoic acid,
0.02 mg biotin, 5 mg p-aminobenzoic acid, 25 mg fumaric acid, and 0.4 mg
coenzyme A. The above additives, excluding imidazole, can be prepared
in advance as a combined 200-fold concentrate and stored at - 2 0 . Phenol
red is added at a concentration of 10 mg/liter. The pH is adjusted to 7.35
with 1 N HC1 and the medium sterilized by filtration and stored at 4.
Filtration is perfor0aed with a Millipore stainless-steel pressure filtration
apparatus (142-ram filter disc holder; Millipore Corp., Bedford, Massachusetts). Nuclepore filters (0.2 /~m, standard membranes) are washed
extensively prior to use, first with boiling distilled water and then with
ambient-temperature distilled water. Thereafter the filter holder and filter
are wrapped in aluminum foil and sterilized by autoclaving. After the
apparatus has cooled to room temperature, the medium is pressure filtered with compressed nitrogen under sterile conditions into clean sterile
t4 E. J. Furshpan, P. R. MacLeish,P. H. O'Lague,and D. D. Potter, Proc. Natl. Acad. Sci.

U.S.A. 73, 4225 (1976).
15p. H. O'Lague, P. R. MacLeish,C. A. Nurse, P. Claude, E. J. Furshpan, and D. D.
Potter, Cold Spring Harbor Syrup. Quant. Biol. 40, 399 (1975).
~6C. A. Nurse and P. H. O'Lague,Proc. Natl. Acad. Sci. U.S.A. 72, 1955(1975).
17R. E. Mainsand P. H. Patterson,J. Cell Biol. 59, 329 (1973).
[53]
579
glass bottles. (Glassware is never cleaned with detergent; if used for proteinaceous solutions, it is discarded after use. Bottles are cleaned by
autoclaving them full of water followed by repeated rinsing while still
hot.)
2. L-15-C02. The medium used both for sympathetic neurons and
nonneuronal cells is prepared by adding 170 ml of 150 mM NaHCOa for
every 850 ml of basal L-15-Air (see above). (The sodium bicarbonate is
sterilized by filtration either before addition to sterile L-15-Air or in combination with the L-I5 medium). The pH of this medium is adjusted by
blowing COz over the solution. In general, most solutions are sterilized by
passage through autoclaved Nuclepore filters (0.2 /zm) washed as described above.
3. Plating Medium. The medium used for ganglion dissection and cell
dissociation consists of 100 ml of basal L-15-Air to which is added glucose
(4 ml of a 30% w/v solution), glutamine (1 ml of a 200 mM solution;
Microbiological Associates, Inc., Walkersville, Maryland), penicillin, and
streptomycin (1 ml of a solution containing 10,000 U of penicillin and 10
mg of streptomycin per milliliter; obtained as lyophilized powder from
Grand Island Biological Co., Grand Island, New York). Plating medium
contains twice the normal glucose concentration so as to increase the
density of the suspending medium and facilitate cell settling during
plating.
4. Complete Growth Media. For routine growth of cells in either
L-15-Air or L-15-CO2, the following additives are combined with 100 ml of
the basic formulation: glutamine, penicillin, and streptomycin (same as
for "Plating Medium"), glucose (2 ml of a 30% w/v solution), fresh vitamin mix (1 ml, see below), adult rat serum (5 ml, see below), Methocel
(0.6 g; see below), and nerve growth factor (NGF; final concentration of
1/~g/ml 7S, see below). Complete media are stored for no longer than 10
days at 4. L-15-Air cultures are incubated in a humidified air atmosphere
at 36 whereas L-15-CO2 cultures are incubated in a 5% CO2 in air
humidified atmosphere, also at 36. The growth medium (2 ml) is changed
every 4 days except for cultures receiving conditioned medium (see below) which receive fresh medium every 2 days. Rat sympathetic neurons
also have been grown in media based on commercial preparations other
than L-15 and with additives different from those described here. 18-2
5. Growth Medium Additfi, es. The fresh vitamin mix is prepared as a
100-fold concentrate and contains I mg of 6,7-dimethyl-5,6,7,8-tetrahydropterine (Calbiochem, La Jolla, California), I00 mg ascorbic acid
18 R. P. Bunge, R. Rees, P. Wood, H. Burton, and C.-P. Ko, Brain Res. 66, 401 (1974).
19 C.-P. Ko, H. Burton, and R. P. Bunge, Brain Res. 117, 437 (1976).
20 K. J. Lazarus, R. A. Bradshaw, N. R. West, and R. P. Bunge,Brain Res. 113, 159 (1976).
580
SPECIFIC CELL LINES
[53]
(Sigma Chemical Co., St. Louis, Missouri), and 10 mg of glutathione

(Sigma) in 20 ml of distilled water. The pH is adjusted to 6.0 with 1 N
NaOH and the solution stored at - 2 & after sterilization by filtration.
Methocel (hydroxymethyl cellulose, standard grade; Dow Chemical
Co., Midland, Michigan) increases the viscosity of the medium thereby
protecting the cells from mechanical stresses and fluid movements in the
culture dish. It does not appear to serve a nutritive function since neurons
can be grown on collagen gels in medium lacking Methocel. After sterilization in the autoclave, the dry Methocel is combined with the culture
medium and the mixture stirred vigorously for 10-24 hr at 4.
Rat serum is necessary for optimal long-term growth of dissociated rat
sympathetic neurons; sera of fetal calf, calf, horse, rabbit, or swine are
not as effective.17 Commercially available rat serum is also less effective
than serum prepared in the laboratory using animals obtained from a
commercially maintained rat colony. Adult rats (CD colony, see above) of
either sex are killed by asphyxiation in a COs chamber and exsanguinated.
The blood is collected on ice, allowed to clot, and centrifuged for 30 min
at 35,000 g and 4 . The serum is stored at 0 for 12-16 hr and centrifuged to
remove remaining cells. Finally, the clot-free serum is filtered by pressure
filtration in the same manner as the culture media. Portions of 5 ml each
are stored at - 2 0 .
Nerve growth factor (NGF) is essential for survival, growth, and maturation of sympathetic neurons from newborn rats. In the absence of this
protein, only the ganglionic nonneuronal cells will survive in L-15-CO2
cultures of 'dissociated sympathetic ganglia. NGF is prepared as the 7S
form .by gel filtration through Sephadex G-10021 followed by DEAEcellulose fractionation. 2z The pooled DEAE-cellulose fractions are concentrated by pressure filtration (Diaflo, UM-10, Amicon Corp.,
Lexington, Massachusetts) to 1 mg/ml and, after sterilization by filtration,
individual portions are stored at - 2 0 . NGF is stable for 6-12 months
under these conditions. A new, rapid procedure for producing the lowmolecular-weight form of NGF is also available. 23
Removal of Nonneuronal Cellsfrom Dissociated Sympathetic Ganglionic

Cultures
By means of a simple manipulation in the composition of the medium,
the growth of ganglionic nonneuronal cells may be effectively controlled.
When dissociated sympathetic ganglia are plated in L-15-Air (see above),
21 V. Bocchini and P. U. Angeletti, Proc. Natl. Acad. Sci. U.S.A. 64, 787 (1969).
2z S. Varon, J. Nomura, and E. M. Shooter, Biochemistry 6, 2202 (1967).
22 W. C. Mobley, A. Schenker, and E. M. Shooter, Biochemistry 15, 5543 (1976).
[53]
581
only the neurons survive; less than 10% of the cells seen by phase microscopy are nonneuronal even after several weeks in culture. 4
When dissociated ganglion cells are plated into L-15-CO2, both
neurons and nonneuronal cells become established in culture. The nonneuronal cells proliferate and form a monolayer within 2-3 weeks. Growth
of nonneuronal cells in L-15-CO2 cultures can be prevented by the use of
antimitotic poisons such as cytosine arabinoside or fluorodeoxyuridine.
Both of these inhibitors reduce the proliferation of ganglionic nonneuronal
cells without producing any reduction in neuronal catecholamine synthesis. Sympathetic cultures are incubated in growth medium containing 10
/zM cytosine arabinoside (Sigma) for 2 periods of 48 hr each, beginning on
days 2 and 6 in culture. Fluorodeoxyuridine and uridine (Sigma) are also
antimitotic when added at 10/~M on the same schedule.
In addition to these methods, nonneuronal cells may be removed at a
stage prior to plating by cell fractionation procedures. One method successfully employed for this propose in cultures of chick sympathetic
neurons involves differential cell adhesiveness and takes advantage of the
fact that, under appropriate conditions, chick ganglionic nonneuronal cells
attach to the substratum sooner and adhere tighter than the neuronal cell
population. ~,6
Co-Culture of Neurons with Nonneuronal Cells
Ganglionic Nonneuronal Cells

As previously mentioned, use of L-15-CO2 growth medium without
antimitotic treatment permits proliferation of the indigenous ganglionic
nonneuronal cell types from dissociated sympathetic ganglia. Among the
nonneuronal elements are glial-like (satellite) cells as well as fibroblastlike cells.
Other Nonneuronal Cells in Mixed Culture

As with ganglionic nonneuronal cells, L-15-CO2 rather than L-15-Air
growth medium is better for growth of many other types of nonneuronal
cells. Sympathetic neurons can be plated either onto pre-plated irradiated
nonneuronal cells as previously discussed (see "Culture Surfaces") or
together with nonneuronal cells.
Conditioned Medium
For the study of the control of the differentiated fate of sympathetic
neurons by nonneuronal cells, conditioned medium (CM) 4 can b~ prepared
582
SPECIFIC CELL LINES
[53]
from cultures of nonneuronal cells in the following manner. Dissociated

primary cells are obtained from minced tissue (e.g., heart, brain, liver,
skeletal muscle) that has been incubated for 20 rain with continuous agitation in phosphate-buffered saline containing 1 mg/ml collagenase (Worthington Biochemical Corp., Freehold, New Jersey). Undissociated pieces
are allowed to settle and are treated with additional collagenase in a
second incubation. The combined dissociated cells are collected by lowspeed centrifugation, washed several times, and a sample counted in a
hemocytometer. In general, 5-10 million cells are plated per 75-cm 2 flask in
L-15-CO2 growth medium containing 10% fetal calf serum (Microbiological Associates) instead of rat serum. Cell lines are similarly maintained
with the same medium.
Prior to preparing conditioned medium, flasks with a dense layer of
nonneuronal cells are incubated for 2-12 hr with L-15-CO2 growth medium containing rat serum. After removal of this wash medium, the medium to be conditioned is added to the flasks. Usually, 20-30 ml of complete L-15-CO~ growth medium containing rat serum, Methocel, etc. (see
above) are added to each flask and incubated for 2 days. The CM is then
harvested and stored at - 2 0 . Before use, the CM is thawed and mixed
with fresh medium to yield the desired final concentration. Neurons receive such mixed media every 2 days.
The acetylcholine inductive effect of CM is species specific with regard to the nonneuronal cells used for conditioning. 4 Mouse and hamster
cell lines as well as mouse and chick primary heart cultures are ineffective
in producing CM capable of acetylcholine induction in rat sympathetic
cultures grown in the presence of rat serum. Thus this additional feature
of co-culture and CM must be kept in mind in planning induction
experiments.
Isolated Sympathetic Neurons Grown in Microculture
For studies requiring the growth of solitary isolated sympathetic
neurons, methods are now available for maintaining and analyzing single
cells in microculture. For most biochemical purposes neurons can be
grown in the segregated wells of Falcon 3034 Terasaki tissue culture
plates. 24 Each plate contains 60 wells with a capacity of 10/~1 each. As
with the standard cultures, growth in microculture wells requires a culture
substratum of dried collagen. Even under standard culture conditions
single neurons appear to be more fastidious than mass cultures, and the
microcultures are not always successful. Although adhesion of single cells
24L. F. Reichardtand P. H. Patterson,Nature (London) 270, 147 (1977).
[53]
583
to the culture surface is especially problematic, the use of collagen gels

instead of dried collagen films appears to improve neuronal survival. Furthermore, single-cell microcultures appear to be more easily maintained
in CM-containing L-15-CO2 growth medium. Growth in CM from chick
heart cells produces predominantly adrenergic neurons while growth in
CM from rat heart cells produces predominantly cholinergic neurons. 24
Single cells in microculture wells also can be grown on pre-plated layers
of nonneuronal cells where proliferation of the nonneuronal elements is
inhibited by T-irradiation.
Alternately, for purposes of electrophysiological, pharmacological,
and morphological examination, single neurons can be grown on separated collagen islands formed on the standard culture coverslips. 14,25 In
general, small droplets of collagen solution are allowed to dry onto nonwetting polystyrene coverslips (Lux, untreated for tissue culture) forming
islands of collagen 300-500 /zm in diameter. Dissociated cardiac cells
(myocytes and fibroblasts) from newborn rats are plated onto such islands
by incubation for about 2 hr. Cells not adhering to the collagen islands are
then washed away with medium changes. After another 1-2 days, proliferation of the heart cells is suppressed by the standard T-irradiation procedure. Dissociated sympathetic neurons suspended at a low density are
plated onto these cardiac cell islands 1-5 days after irradiation. Under
proper conditions, such islands receive only one or a few neurons. These
microcultures are subsequently grown in L-15-CO2 growth medium lacking Methocel and containing either the standard 5% adult rat serum or
10% fetal calf serum.
Phenotypic Characteristics of Sympathetic Cultures
Several parameters can be used to monitor the development of dissociated sympathetic neurons in culture. Overall neuronal growth is measured by the increase in rates of synthesis of protein, lipid, and RNA, 26 or
by the increase in total neuronal protein and lipid. 4 Process outgrowth and
elongation as well as the general culture morphology can be readily followed by inverted phase microscopy. Spontaneous and evoked electrical
activity as well as synaptic interaction can be examined by standard electrophysiological techniques.
Of particular relevance to studies of neuronal differentiation, the sympathetic cultures are able to synthesize and accumulate neurotransmitters
from labeled precursors. These labeled neurotransmitters can be isolated
25 S. C. Landis, Proc. Natl. Acad. Sci. U.S.A. 73, 4220 (1976).
2s R. E. Mains and P. H. Patterson, J. Cell Biol. 59, 361 (1973).
584
SPECIFIC
CELL
[54]
LINES
and identified by high-voltage paper electrophoresis.'7 Furthermore, uptake, storage, and evoked release of neurotransmitters can be assayed, z7
The biosynthetic enzyme, choline acetyltransferase, also can be measured in extracts of sympathetic cultures, and levels of this enzyme are
elevated in cultures induced for cholinergic function. 4
Acknowledgments
Part of the work described here was supported by the American and Massachusetts
Heart Associations, the Helen Hay Whitney Foundation, and the National Institute of
Neurological and CommunicableDiseases and Stroke.
27p. H. Patterson, L. F. Reichardt, and L. L. Y. Chun, Cold Spring Harbor Symp. Quant.
Biol. 40, 389 (1975).'
[54] N e u r o n a l C e l l s f r o m R o d e n t N e o p l a s m s
By DAVID
SCHUBERT
and
WILLIAM
CARLISLE
Over 70 years ago Ross Harrison used cultures of neural epithelium

from amphibian embryos to examine the mechanism of neurite growth.'
Since these initial experiments, advances in tissue culture technology
have permitted the study of many cell types in defined environments that
minimize the multiple interactive events found in vivo. Because of the
extreme complexity of the nervous system there is a clear requirement for
isolating individual component cells for biochemical and electrophysiological studies. Fruitful exploitation of cell culture for the study of the nervous system requires a collection of cells representative of those found in
vivo. In addition, the biochemical and electrophysiological behavior of
these cells should mimic that found in the animal. An assumption underlying all work in cell culture is that i f a function of the in vivo cell is found in
culture, then the underlying biochemical mechanisms are the same as
those employed in vivo. The cells can then be used to study the molecular
basis of these phenomena.
Clonal cell lines that possess functional characteristics of their in vivo
counterparts have been isolated from the nervous system. However, the
variety of cell types from the nervous system available for study as homogeneous (clonal) populations is very limited, and there is a continuing
need for the establishment and characterization of additional cell lines. It
is to this end that we describe our experience in developing cell lines from
the rodent nervous system.
'
R. Harrison, Anat. Rec. 1, 116 (1907).
METHODS IN ENZYMOLOO, VOL. LVIII
Copyright o 1979by Academic Press, Inc.

ISBN 0-12-181958-2
[54]
NEURONAL CELLS FROM RODENT NEOPLASMS
585
Methods for Obtaining Neuronal Cell Lines

Since the normal differentiated nerve cell may not divide, and a dividing cell is needed to obtain cell lines, it follows that procedures are required to select for or induce cell division. Four procedures have been
used successfully to obtain clonal cell lines from the nervous system: (1)
spontaneous tumors, (2) normal embryonic tissue, (3) viral transformation, and (4) chemically induced tumors.
To date, only two spontaneous rodent tumors of neuronal origin have
been adapted to clonal cell culture. These are the C1300 neuroblastoma 2'3
and the PC12 sympathetic ganglion-like cell. 4 Both of these tumors occurred peripherally to the central nervous system and were identified on
the basis of tumor pathology as possible neuroblastoma and pheochromocytoma, respectively. Positive identification of the cell type required the establishment of clonal cell lines from the original tumor.
The successful establishment of clonal electrically excitable cells directly from embryonic mammalian central nervous system (CNS) tissue
has recently been described. 5 Embryonic brain cells from specific areas
were dissociated with trypsin and exposed to antiglia antiserum in the
presence of complement to lyse all cells expressing glial antigens. This
selection technique was repeated several times with the intent of eliminating the more rapidly dividing glial cell population. In addition, loosely
adherent cells with nerve-like morphologies were selectively passaged
away from the more rapidly dividing anchorage-dependent cells by ringing
techniques (described below). Several of the resultant clones expressed
nerve-specific surface antigens and were electrically excitable as defined
by a veratridine stimulated sodium flux assay.
The third alternative successfully used to obtain dividing nerve cells is
by in vitro viral transformation. De Vitry et al, 6 dissociated single cells
from a 14-day embryonic mouse hypothalamus and grew the resultant
cells in culture for 6 days. At this time the cultures were washed and
covered with serum-free medium containing 7 10r plaque-forming units
of SV40. After adsorption for 30 min, the cells were covered with growth
medium and the transformed cells isolated from foci of growing coils by
methods similar to those described below. By selecting cells primarily on
the basis of morphological complexity, e.g., the most "nerve-like" in
2 G. Augusti-Toccoand G. Sato, Proc. Natl. Acad. Sci. U.S.A. 64, 311 (1969).
8 D. Schubert, S. Humphreys,C. Baroni, and M. Cohn, Proc. Natl. Acad. Sci. U.S.A. 64,
316 (1969).
4 L. Greene and A. S. Tischler,Proc. Natl. Acad. Sic. U.S.A. 73, 2424 (1976).
5 K. Bulloch, W. Stallcup, and M. Cohn, Brain Res. 135, 25 (1977).
F. De Vitry, M. Camier, P. Czernichow,P. Benda, P. Cohen, and A. Tixier-Vidal,Proc.
Natl. Acad. Sci. U.S.A. 71, 3375 (1974).
586
SPECIFIC CELL LINES
[54]
appearance, a clone of mouse hypothalamic neurosecretory cells was

obtained which made both neurophysin and vasopressin.
The final technique for obtaining dividing cell populations of nerve and
glia is by the induction of tumors of these cells. The most efficient way of
obtaining brain tumors is by the transplacental induction of tumors with
nitrosoethyl urea (NEU). r Since all rat strains are not equally susceptible
to this carcinogen, BD-IX rats have been used most frequently. 7 Tumors
are induced by injecting NEU, freshly dissolved in saline at a does of 40
mg/kg body weight, intravenously into the femoral vein of female rats at
15 day's gestation. This dose reduces the litter size by approximately
50%. Tumors of the trigeminal nerve and spinal roots can be induced with
NEU using a different strain of rat, Sprague-Dawley, and injecting 18day-pregnant animals intravenously with 20 mg/kg of NEU. 8 After birth
the pups are weaned and placed in separate cages. Between 150-250 days
after birth the majority of the offspring develop symptoms of neurological
disease, including loss of motor control, eyelid closure, and weight. These
rats are then watched carefully and, when they appear to be near death,
are anesthesized with ether, washed with 70% ethanol, exsanguinated,
and carefully autopsied. Most tumor masses are found in the cranial cavity and brain stem. Using the NEU induction scheme described above,
visible tumors are found in about 93% of the BD-IX offspring.
Reagents
Tissue Culture Medium
For both historical and practical reasons all of the culture work on
cell lines derived from NEU-induced CNS brain tumors has employed
modified Eagle's medium 9 containing 10% fetal calf serum. With one exception, the PC12 sympathetic neuron-like cell line, 4 all cell lines that we
have tried will grow in this medium, even if they have been originally
adapted to other types of culture media and serum. This standardization
of culture conditions simplifies medium preparation and serum testing
and, in addition, provides a common culture regime by which the various
cell lines can be compared. Modified Eagle's medium can be purchased in
powdered or liquid form, but we have found that preparing it ourselves is
necessary both to maintain high quality and to provide the flexibility
needed in altering its composition for experimental purposes. (The prepr H. Druckrey, R. Preussmann, S. Ivankovi~, and D. Schmahi, Z. Krebsforsch. 69, 103
(1967).
a A. Hirano, J. Hasson, and H. M. Zimmerman, Lab. Invest. 27, 555 (1972).
a M. Vogt and R. Dulbecco, Proc. Natl. Acad. Sci. U.S.A. 49, 171 (1963).
[54]
587
aration of media is described in this volume [5].) Before each batch of

medium is used it is tested for sterility, by dilution into a nutrient broth
solution that is incubated at 37% and for its ability to support cell growth
and phenotypic expression. We routinely use a skeletal muscle myoblast
cell line to test the medium since its ability to fuse is very sensitive to
culture conditions. Freshly prepared medium is stored in 90-ml aliquots at
4 in the dark for periods of up to 1 month. The cells are grown in a
humidified atmosphere at 36 in 12% CO2 in air.
Serum
There is a tremendous variation in the fetal, calf serum obtained from
different commercial sources and between lots from the same dealer.
Therefore, samples from several lots are tested for their ability to support
the growth and differentiation of different cell types. Once a lot is decided
upon; sufficient serum from that lot is purchased to last from 6-12 months.
It is stored frozen ( - 2 0 ) and dispensed in 10-ml aliquots for use over a
2-week period. The dispensed serum is stored at - 2 0 and only thawed
once immediately before use.
Culture Dishes
Many cell types are anchorage dependent for growth and require a
tissue-culture dish surface. Although most of the nerve cell lines will grow
in plastic petri dishes, where cells attach very poorly, all of the nerve cell
lines are routinely passaged on tissue culture dishes; 60-ram plastic dishes
are used for most purposes.
Viokase (Pancreatin 4XN.F., GIBCO)

Viokase, a mixture of pancreatic enzymes, is sterilized by passage
through 0.22-/~m filters and stored at - 2 0 as 1-ml aliquots.
Establishment of Cell Lines from Neoplastic Tissue

The tumor is removed aseptically from the animal and rinsed sequentially in three tissue culture dishes containing culture medium. After the
final rinse, a portion of the tumor is removed, minced with scissors, and
injected subcutaneously into syngenic animals in case the initialcells do
not adapt to continuous cell culture; of the approximately 150 neoplasms
we have attempted to adapt to culture, only one did not grow from the
initial explant. A piece smaller than 5 mm in diameter is removed from a
588
SPECIFIC CELL LINES
[54]
nonnercotic area of the tumor, placed in 5 ml of complete medium, and

finely minced with scissors. This procedure liberates many viable free
cells as well as producing small pieces of tissue. The mixture of pieces and
cells is dispensed at different concentrations into at least 24 60-mm tissue
culture dishes. For example, the first four dishes receive 1 drop, the
second four receive 2 drops, and so on. A few dishes receive cells in
suspension only and no pieces of tissue. After about a week, the plates are
carefully examined for areas of growth. If the medium in the dishes containing higher cell densities becomes acid, the cells should be removed by
scraping with a rubber policeman and the cells, tissue pieces, and medium
diluted by a factor of 10 into fresh medium. Some conditioned medium
should be carried along with the cells until they are cloned.
Selection and Cloning of Cells from Primary Cultures
Initial Selection of Cell Type

In our experience, cells grown from all of the primary tumor explants
are morphologically heterogeneous, even if histologically the neoplasm
from which they were derived consists primarily of one cell type. Thus a
choice must be made as to which morphologies are to be cloned. It is
desirable to initiate the selection of the desired cell type as early as possible since, in the case of neuronal tumor explants, flat fibroblast-like cells
tend to quickly dominate the culture. We try to select cells with the most
complex morphology, e.g., with long processes, and also cells that grow
loosely attached to the substratum with round, phase-bright bodies.
The latter growth characteristic is shared by most clonal nerve cell
lines. Such clusters of cells are usually observed in the early explant
cultures associated with a small piece of tissue. To isolate a desired cell
cluster, it is located with an inverted microscope and its location marked.
The culture medium is removed (some should be saved), the dish carefully
washed once with serum-free medium, and a sterile cloning ring (see this
volume [12]) is placed over the area. A solution of 3 parts serum-free
medium and I part Viokase (Gibco), warmed at 37, is placed in the ring.
After 5 rain at 37, the cells are removed from the dish with a sterile
Pasteur pipette and placed into a 35-mm culture dish containing approximately 80% fresh medium containing 10% fetal calf serum and 20% conditioned medium from the original dish. These cells are examined regularly for growth; if cell heterogeneity persists, the above procedure should
be repeated. It should be emphasized that this procedure does not yield
clones, but only enriches for a particular cell type and removes it from the
[54]
589
less desirable cells that will eventually overgrow the primary explant cultures. Many "ringed" cultures should be made, for all do not grow.
Once these "ringed" cultures begin to divide and appear somewhat
homogeneous, they should be transferred at high cell densities (1 : 5 split
of a confluent dish) along with conditioned medium for several transfers.
Assuming the cells are growing, they are ready for cloning.
Cloning
There are several methods for cloning cells (see this volume [12]).
Since the CNS-derived nerve and glial cells tend to grow attached to the
tissue culture dish, the method most frequently employed is that in which
cloning rings are used. Exponentially dividing cultures are dissociated to
single cells with a 1 : 4 dilution of Viokase for 1 hr at 37, and plated
between 102 and 104 cells per 100-mm tissue culture dish. If the plating
efficiency is low, the use of 50% conditioned medium may be helpful.
After 24 hr, well-isolated single cells are located on an inverted microscope and marked from below with a felt-point pen. Between 1 and 3 weeks
later, the plates are examined again for cell growth in the marked areas. If
a small colony has formed and remains isolated from other cells, then it
should be removed with a cloning ring and placed in 35-mm tissue culture
dishes. To be sure of the clonal origin of a cell line this procedure should
be repeated once. At this stage, the clones are grown in a dozen 60-mm
dishes, and the cells are stored in liquid nitrogen to preserve the lines.
Handling of Clonal Cell Lines
Storage and Recovery of Cells

Cells from the late experimental stage of growth are scraped with
rubber policemen, and the cells from each dish are sedimented separately
by centrifugation and taken up in 1 ml of ice-cold medium containing the
following ingredients: 4 parts double-concentration modified Eagle's medium minus sodium bicarbonate; 3 parts water; 2 parts, fetal calf serum;
and 1 part filter-sterilized dimethylsulfoxide. The suspension is sealed in
ice-cold plastic screw-cap vials, frozen at a rate of l/min to a final temperature of - 80, and stored indefinitely in liquid nitrogen. To recover the
frozen cells, vials are removed from the liquid nitrogen and immediately
thawed in a 37 water bath. As soon as the medium is liquid, the suspension is diluted at different final concentrations into six or eight 60-mm
dishes containing modified Eagle's medium and 20% fetal calf serum.
Evidence of cell growth should be visible within a few days.
590
SPECIFIC CELL LINES
[54]
Routine Passage of Cell Lines

To maintain cell lines for extended periods of time in the laboratory, it
is mandatory that the cells be passaged before the culture becomes confluent. This procedure minimizes selection pressure against cell that are
killed at high cell density. The ceils may be passaged into fresh medium by
scraping the monolayer with a rubber policeman and placing a few drops
of the suspension into dishes containing fresh medium. Alternatively,
cells can be dissociated with a 1:4 dilution of Viokase in serum-free
medium for 30 min at 37, removed from the Viokase by centrifugation,
and diluted into fresh medium. The routine passage of dissociated cells is
more time Consuming than by scraping, but is preferable because the cells
grow evenly dispersed over the culture dish.
Shipment of Cells
If cells are to be shipped in the mail, they are placed in sealed tissue
culture bottles filled to the top with medium so that no air bubbles are
present and shipped via regular air mail.
Growing Large Quantities of Cells

With the current technology, there are two alternatives for growing
large quantities, i.e., a gram or more, of anchorage-dependent cellst in
large 150-mm tissue culture dishes or in roller bottles. If the cells grow in
suspension culture, then large quantities can be grown in spinner culture
much like bacteria (see this volume [15-17]). Cells are grown in 150-mm
dishes exactly as in the smaller dishes, and they can be readily harvested
from the dish by scraping with a common soft rubber spatula used in
cooking. When plating cells into a large surface area, as with the 150-mm
dishes and roller bottles, it is necessary to dissociate the cells with Viokase to assure an even distribution on the substratum. To grow cells in
roller bottles, the sterile bottles are gassed for 1 min with a mixture of 12%
COz in air passed through a cotton plug at the end of a sterile plastic
pipette. Then 100 ml of modified Eagle's medium for a 16-inch roller
bottle containing approximately 108 cells are poured into the bottle and
the bottle tightly sealed. It is then placed at 37 on a roller apparatus
turning at low speed. The cells are harvested by washing in situ with the
desired buffer and are scraped off with a soft rubber policeman mounted
on a metal shaft.
AUTHOR INDEX
591
Author I n d e x
Numbers in parentheses are footnote reference numbers and indicate that an author's
work is referred to although his name is not cited in the text.
A
Ashby, J., 297, 302(11)

Astrup, T., 59
Aaronson, S. A., 22, 27(17), 96, 264, 413, Athreya, B., 31
416, 417, 418, 422(5, 6)
Aubin, J. E., 315
Abraham, S., 126
Auerbach, R., 125
Abrahams, P. J., 426, 434(11)
Auersperg, N., 151
Abercrombie, M., 134
August, J. T., 418
Absher, M., 143
Augusti-Tocco, G., 585
Acton, R. T., 211, 212(7), 213(7), 215(7), Austin, P. E., 162, 425
216(6), 217, 218(7), 219(7, 14), 223,224,
Axelrod, D. E., 507
228
Adair, G. M., 309, 317, 318
Adams, D. O., 494, 496(1), 498(1), 503
B
Adams, L. R., 152
Aden, D., 349, 375(17), 376(17), 378
Aftonomos, B., 125
Babinet, C., 97, 105(22)
Agy, P., 60
Bach, R., 507
Alderdice, P. W., 342
Bain, B., 487
Allderdice, P. W., 174, 343
Bajaj, Y. P. S., 367, 484
Alien, R. G., 530, 531(23), 534(23)
Baker, J. R., 274
Allen, T., 266
Baker, R. M., 312, 314(11), 317(11), 353,
Allison, A. C., 265, 271(21), 345,502
359(1)
Allred, L. E., 284
Balazs, E. A., 268
Ames, B. N., 302
Balducci, D., 4, 5(6)
Andersen, V., 124, 130(8)
Ball, G. H., 457
Anderson, C. V., 549
Baltimore, D., 416
Anderson, D., 297, 302(11)
Bancroft, F. C., 375(19), 377(19), 378, 528,
Anderson, E. C., 255
529(3, 14), 530, 532(3), 533(14, 36, 56),
Anderson, L. E., 60, 454
534(3, 14, 56), 535
Anderson, W. F., 507
Banerjee, S. D., 268
Andrews, P. M., 140
Bang, F. B., 271
Andrews, T. M., 147
Barbacid, M., 418
Angeletti, P. U., 580
Barile, M. F., 22, 219, 220(24), 227
Aoki, T., 219, 418
Barker, B. E., 369
Arabasz, M., 564
Barker, J. E., 507
Armelin, H., 102
Baron, S., 296
Armelin, H. A., 94
Baroni, C., 585
Arndt-Jovin, D. J., 246
Barowsky, N. J., 528, 529(53,533(53,534(53,
Arnold, E., 248
535(5)
Arnstein, P., 264, 343
Barski, G., 345, 348(4)
Arrighi, F. E., 325
Barstad, P. A., 211, 213, 216(6), 223, 224
Artzt, K., 97, 105(22)
Bartholomew, J. C., 248
Ash, J. F., 267, 273(42)
Barz, W., 478
592
AUTHOR INDEX
Baserga, R., 248, 250, 259, 260, 286, 287,

288, 290, 552
Basilico, C., 254, 317, 348
Bassin, R. H., 423,424(32)
Bauer, H., 396, 417
Bauer, W., 407
Baxter, J. D., 534
Beam, A. G., 166
Beasley, A. R., 467
Beaudreau, G., 417
Becker, A. J., 256
Becker, B. G., 25, 27(22)
Beckner, S. K., 521, 523(29)
Belkin, M., 125
Bellet, A. J. D., 433
Belliard, G., 360
Benda, P., 97, 585
Benes, E., 149
Benirschke, K., 173
Bennett, D., 97, 105(22)
Bennett, E. L., 148
Bensadoun, A., 146
Bentvelzen, P., 417
Benveniste, R. E., 421, 424(28)
Benya, P. D., 564
Berg, P., 411
Bergrahm, B., 210
Bergsma, D., 339, 342, 347, 348(17)
Berman, L., 33, 175
Bermudez, E., 60, 128
Bemaert, D., 544
Bernfield, M. R., 268
Bernhard, H. P., 346
Bernheim, B. C., 405
Bernheim, J. L., 492
Bernini, L. F., 166
Bernstein, E. H., 399, 403
Bernstein, I. A., 265
Berry, S. F., 365, 367
Bertolotti, R., 347, 353
Berwald, Y., 296
Bhat, U. K. M., 462
Bianchi, N., 287
Biedler, J. L., 264
Biempica, L., 271
Bigley, R., 507, 511(53
Bilello, J. A., 375(20), 377(20), 378
Billingham, R., 265
Binding, H., 364
Binns, A., 485
Birch, J. R., 60
Bissell, D. M., 544

Biswas, D. K., 528, 529(11)
Black, L. M., 149, 456, 462
Black, P. H., 160
Blackbunn, M. J., 147
Blevins, C. S., 423, 424(33)
Bloch, E., 570
Bloon, S. E., 326
Bliimel, E, 146
Bocchini, V., 580
Bodmer, J., 356
Bodmer, W. F., 356
Boettiger, D., 381
Bogtich, S., 264
Bolognesi, D., 396, 417
Bolognesi, D. E, 418
Boltz, R. C., Jr., 132
Bonifas, V. H., 432
Bonnett, H., 360, 361(9)
Bonney, R. J., 544
Boon, T., 97, 105(21)
Boone, C. "W., 29, 197, 267
Boorstein, R., 536, 543(3)
Bootsma, D., 347
Bornstein, M. B., 271,577
Bornstein, P., 267, 273(42)
Botchan, M., 411,425, 426(6)
Bouck, N., 297, 377
Boumenthal, A. B., 464
Bowden-Essien, F., 512
Boyd, G. A., 290
Boynton, A. L., 261,262
Boyse, E. A., 219
Brack, C., 350
Brada, Z., 553, 560(5)
Bradley, T., 162, 163(34)
Bradshaw, R. A., 579
Braendstrup, O., 124, 130(8)
Brand, K. G., 178
Brandt, B., 96
Breslow, R., 506, 510(3)
Briand, P., 151, 152(53)
Bridson, W. E., 533(553, 534(553,535
Brindle, S. A., 115
Brinkley, S., 130
Brookes, P., 297
Brookman, D. H., 130, 513
Brooks, M. A., 455, 456, 512
Brooks, R. F., 251
Brown, A., 203
Brown, A. L., 371,375(5, 19), 377(19), 378
AUTHOR INDEX
Brown, B. L., 60
Brown, J. M., 126
Brown, M. S., 444
Bruce, W. R., 151
Bruchovsky, N., 256
Brunsting, A., 235
Bryant, J. C., 14, 60, 142, 203, 211
Buckingham, M. E., 512
Buckler, C., 296
Buckley, P. A., 518, 519(22), 524(22)
Buell, D. N., 257
Biittner, W., 425, 426(3)
Bui-Dang-Ha, D., 367
Bullard, J. C., 130, 513
Bulloch, K., 585
Bunge, R. P., 579
Buonassisi, V., 97, 528, 570
Burger, M. M., 368
Burges, J., 364
Burk, R. D., 346
Burk, R. R., 265
Burke, D. C., 294
Burstin, S. J., 254
Burton, H., 579
Burton, K., 568
Butcher, G. W., 349, 351(26)
C
Caffier, H., 425, 426(2)
Cahn, M. B., 132
Cahn, R. D., 132, 198
Cailleau, R. M., 264, 266(4)
Caldwell, J. M., 451,457(1)
Calenoff, M. A., 184
Cailahan, R., 417, 421,424(28)
Camier, M., 585
Campbell, R. L., 162, 163(33)
Campenot, R. B., 306, 307
Canary, P., 377
Caplan, A. I., 130, 513
CarboneU, A. W., 252
Cardell, R. R., 564
Cardella, C. J., 265, 271(21)
Carlisle, W., 96
Carlson, P. S., 359, 360, 362(2), 365, 366(2),
367(2)
Carnright, D. V., 309
Caro, L. G., 280
593
Carpenter, G., 108

Casanova, J., 528, 529(9, 10), 533(9, 10),
534(9)
Casey, H. L., 203
Caskey, C. T., 317
Casperson, T., 174
Cassell, G. H., 228
Cassiman, J.-J.,268
Castor, L. N., 134, 149
Catanzaro, P., 266
Cavalieri, R. L., 292
Ceccarino, C., 354, 355(45)
Cellozi, E., 570
Celma, C. L., 405
Cencher, R. L., 264
Chan, T., 352
Chang, C., 267
Chang, H., 259, 260
Chao, J., 457
Chapman, V., 421,424(28)
Chapple, P. J., 44, 78(3)
Charney, J., 178
Chase, A., 315
Chasin, L., 348
Chatigny, M. A., 38
Chatterjee, R. N., 288
Chandhary, S. P. S., 455, 456
Chavin, 197
Chen, L. B., 375(16), 376(16), 377(16), 378
Chen, L. M., 184
Chen, S., 292
Chen, T., 162, 163(33), 354
Chen, T. R., 22, 25
Cheronis, N. D., 151
Cherry, W., 203
Cheung, N. H., 571
Chi, D., 403
Chiarrugi, V. P., 269
Chick, W. L., 183, 184
Chinnadurai, G., 434
Chinnadurai, S., 434
Chin, R., 462
Christie, G., 566, 567(18)
Christman, D. R., 485
Chu, M. Y., 565, 566(14)
Chuman, L. M., 160, 375(18), 376(18),
377(18), 378, 553,554, 559
Chun, L. L. Y., 575, 581(4), 583(4), 584(4)
Chung, E., 271, 273(68)
Cieciura, S. J., 59, 121, 142, 160, 321
Cifonelli, J. A., 274
594
AUTHOR INDEX
Cikes, M., 218

Cintron, R., 98, 528, 529(8), 533(8), 534(8)
Clark, C., 493
Clark, H. F., 467, 475(2)
Clark, J. L., 94
Clarkson, B., 552
Claude, E, 575
Claus, T. H., 544
Clausen, O. P. F., 532
Clingan, D., 251
Clinger, D. I., 38
Cloudman, A. M,, 564
Clyde, W. A., 24
Cocking, E. C., 361,363, 365, 367
Coffman, W. D., 96
Cohen, A. I., 528, 570
Cohen, P., 585
Cohen, R. H., 268
Cohen, S., 108
Cohn, M., 585
Cohn, Z. A., 494, 495(2), 496(2), 497(2),
501(3), 503(6)
Coleman, D. L., 60
Colton, C. K., 184
Compton, A., 523,526(37a)
Condamine, H., 97, 105(22)
Connolly, T. N., 136
Connors, N., 567
Constabel, F., 361,365, 366, 367
Constable, F., 363, 364(21), 366
Contreras, R., 405
Cookson, S., 503
Coombs, R. R. A., 178
Coon, H. (3., 132, 421
Cooper, H. L., 486, 487(1), 492(1), 493
Cooper, J. T., 445, 446(5)
Cooper, N. R., 218
Coriell, L. L., 31, 110, 178, 441
Corsaro, C. M., 315
Cort, A. M., 533(48), 534(48), 535
Corvol, M. T., 561
Cory, J., 460, 461(38a)
Cotran, R. S., 140
Coutts, R. H. A., 364
Covalesky, A., 153
Cox, P. G., 512
Cox, R. M., 211, 212(7), 213(7), 215(7),
216(6), 218(7), 219(7), 223
Cozzarelli, N. R., 251
Cram, L. S., 252
Creagan, P. P., 292
Creagen, R., 352
Crissman, H. A., 149, 240

Cristofalo, V. J., 44, 146, 148
Croce, C. M., 348, 349, 354, 375(17),
376(17), 378
Crosby, W. H., 498
Cross, J. H., 454
Crouse, E. J., 264
Crupton, M. J., 221,223(1)
Cummins, S. E., 361,365
Cunha, G. R., 265
Cunningham, B. A., 492
Cuprak, L. J., 571
Curran, J. F., 544
Czernichow, P., 585
D
Daesch, G. E., 425
Dalton, A. J., 417
Daly, W. F., 149
D'Amato, F., 484
Danna, K. J., 412
Dannies, P. S., 528, 529(12), 532, 533(12, 42,
45, 49, 57), 534(12, 49, 57), 535

Dantchev, D., 391
Darlington, G. J., 346
Darlow, H. M,, 38
Darzynkiewicz, Z., 252
Davidson, R. C., 346
Davidson, R. L., 317, 346, 356
Davidson, S. V., 315
Davies, PL, 265, 271(21)
Davies, P., 265, 271(21)
Davison, A. E., 495, 496(5)
Dawe, C. J., 266
Dawson, K. B., 288
Dawson, R. F., 485
Day, M., 158
Day, P. R., 360
Dean, P. N., 241
Dearing, R. D., 359, 362(2), 365, 366(2),
367(2)
De Asua, L. J., 251
Deaven, L., 251
Deaven, L. L., 343
Decallonne, J. R., 149
Dedrick, R. L., 184, 210, 264
Defendi, V., 268, 502, 503
Deisseroth, A., 346, 507
De La Cruz, C., 446
de Lamater, E. D., 365
AUTHOR INDEX
DeIGiudice, R. A., 22, 24, 25(II),27
DeIl'Orco, R. T., 145
Dempo, K., 541
Denis, K., 351
Deshmukh, K., 561,563(7), 564(7)
Desmond, W., 160, 375(14, 18),376(14, 18),
377(18), 378,554
Dev, V. G., 174, 342
De Vitry, F., 585
Devlin, T., 267
deVries, F. A. J., 426, 434(11)
DeWitt, C. W., 149
Dhanda, V., 463
Dhar, R., 405
Dhermgrongartama, B., 351
Dieckmann, M., 411
Diez, N., 33
di Mayorca, G., 297
Dimmick, R. L., 38
Di Paolo, J. A., 343, 375(21), 377(21), 378
Dirston, W. E., 302
Dittrich, W., 240
Dixon, M. A., 296
Dolbeare, F. A., 243
Donfini, S. F., 343, 455
Donner, L., 355
Donta, S. T., 60
Dorfmann, A., 561,563(8, 9)
Dosik, H., 264
DougaII, P. K., 154
Dougherty, E. P., 140
Douglas, G. R., 343
Douglas, W. H. J., 140
Drake, W. P., 149, 151(31)
Drochmans, P., 267, 544
Druckrey, H., 586
Dubois, P., 97, 105(22)
Dudits, D., 366
Duesberg, P. H., 380
Duff, C., 309, 310(9), 343
Dulak, N. C., 107, 162, 163(31)
Dulbecco, R., 95, 120, 121, 142, 156, 298,
406, 530, 572, 586
Dupr6e, L., 153
Durie, B. G. M., 288
Dutrillaux, B., 325
E
Eagle, H., 59, 86, 95, 136, 144, 145, 153,

155(I, 2), 354, 355(45), 382, 427, 522,
523, 525, 530, 545, 555, 558(12), 571
595
Earle, W. R., 14, 47, 60, 120, 121, 153,

156(13), 188, 209, 203, 211,296
Earley, E., 405
Ebert, P. S., 507
Ecklund, P. E., 174
Edelman, G. M., 492
Edelson, P., 494, 495(2), 496(2), 497(2)
Eder, M., 271
Ediger, R., 373
Eduardo, A. J., 223
Edwards, G. E., 364
Eesterman, I. L., 124, 130(7)
Efstratiadis, A., 411
Ege, T., 347, 359(18)
Ehrensvard, G., 59
Ehrmann, R. L., 577
Eide, P. E., 451,457(1)
Eipper, B. A., 530, 531(19, 20)
Eisen, H., 376, 507
Eisler, W. J., Jr., 150
Eliceiri, G. L., 59
Eliot, J. A., 297
Ellen, K. A. O., 260, 261(27)
Ellis, B. J., 456, 457
Emerson, C. P., Jr., 519, 521, 523(24, 29),
524(24)
Emerman, J. T., 266, 269(32), 271(32),
272(32)
Emery, R., 507
Empson, J., 347
Engel, E., 347
EngeLhardt, D. L., 149
Engvall, E., 267, 273(34)
Ensinger, M. J., 425
Ephrussi, B., 346
Epstein, C. J., 22, 26(14), 27(14)
Eremenko-Volpe, T., 145
Eriksson, T., 360, 361(9), 364
Eflanger, B. F., 326
Erner, Y., 478
Essner, E., 125
Estes, W., 553,560(5)
Euders, J. F., 354
Evans, C. H., 375(21), 377(21), 378
Evans, G. A., 533(50), 534(50), 535
Evans, L. H., 507, 511(5)
Evans, R., 498
Evans, V. S., 14, 31, 60, 188, 197,209, 211,
545
Everett, N. B., 486
Evers, J. L., 200
Everson, L. K., 257
596
AUTHOR INDEX
F
Friedman, L., 264

Friedman, R. M., 293, 294(4), 295(4), 296
Friend, C., 506, 507(1), 508
Freeze, G., 149
Frohman, L. A., 533(55), 534(55), 535
Frye, J., 368
Fujimoto, W. Y., 147, 148
Fujinaga, K., 425, 426(2)
Furshpan, E. J., 575, 578, 583(14)
Furth, J., 528, 530
Furusawa, M., 359, 507
Fairfield, S., 534(65, 66), 535

Farber, R. A., 343
Farr, A. L., 144, 272, 403, 420
Fasman, G. D., 533(44), 534
Fedorko, M. E., 497, 498(10)
Fedoroff, S., 197
Fentiman, I., 265
Ferault, M., 360
Ferchmin, P. A., 148
Ferenczy, F., 367
Ferrone, S., 218, 219
Field, E. O., 288
G
Fiers, W., 405, 426, 434(11)
Fietzek, P. P., 271
Gaarder, A., 493
Fioramonti, M. C., 60
Gagn6, H. I., 126
Fiovelli, G., 166
Gailliani,S., 199, 203
Fischer, A., 59
Gaillard, J., 97, 105(21)
Fischer, G. A., 59, 158, 565, 566(14)
Galfre, G., 349, 351(26)
Fischinger, P. J., 423, 424(32, 33)
Gall, J. G., 326
Fisher, C. L., 418
Gall, M. H., 267
Fitzgerald, P. J., 60
Galletti, P. M., 184
Fitzpatrick, T. B., 565
Gallimore, P. H., 375(16), 376(16), 377(16),
Flandermeyer, R. R., 29, 165, 166, 174, 323,
378, 425, 426(6, 7)
342, 439, 440
Gallo, R., 417
Fleck, A., 146
Gallop, P. M., 271,272(62)
Fleischmajer, R., 265
Gamborg, O. L., 3, 361, 363, 364(21), 365,
Fleissner, E., 418
366, 367
Fleming, E. N., 364
Gardner, M. B., 343
Flint, S. J., 425) 426(7)
Gardner, R. S., 550
Flintoff, W., 315
Gartler, S. M., 166, 439
Fodor, A. R., 203
Gasseling, M. T., 265
Fogh, J., 140
Gaush, C. R., 552
Foley, J. F., 125
Gauss, V., 561,564(2)
Folkman, J., 140
Gautvik, K. M., 532, 533(45, 46, 58, 60-63),
Foster, D. O., 555
534(60-63), 535
Fourier, R. E. K., 359
Gay, S., 271
Fowke, L., 361
Gazzalo, L., 403
Fox, M., 254, 257, 262
Gee, E A., 343
Fraccastoro, A. E, 251
Gefter, M. L., 351, 356
Franklin, H. A., 59, 204, 205(28)
Gelb, L. D., 411
Franks, D., 178
Gelderblom, H., 396
Frearson, E. M., 365, 367
Genereux, D., 60
Freed, J. J., 467
George, D., 367
Freedman, V. H., 160, 297, 348, 368, 371, Geplinski, N., 351
372, 374(8), 375(3-5, 20), 376(3-5), Gerhard, G. F., 417
377(3, 4, 20), 378(3, 4)
Gerland, P. S., 356
Freeman, G., 95,298, 530
Gemer, R. E., 59, 564
French Anderson, W., 346
Geschickter, C. F., 151, 152(50)
Friedman, D. L., 233
Gey, G. O., 96, 121, 142, 166, 203,211,577
AUTHOR INDEX
Gey, M. K., 121, 142, 203, 211
GfeUer, M. D., 265
Giagnoni, G., 530, 531(21)
Giambrone, M. A., 271
Giard, D. J., 192, 264
Giardinello, F., 203
Giardinello, F. E., 203,204(23)
Gibbs, J. E., 288
Gifford, G. E., 198
Gilden, R. V., 405,418
Giles, R. E., 354, 359(51)
Gilman, A. G., 557, 558(15)
Gimbrone, M. A., Jr., 140
Ginsberg, H. S., 425
Giordano, R., 297
Giovanella, B., 373
Giovanella, B. C., 375(15), 376(15), 378
Girardi, A. J., 405
Giuffre, N. A., 115
Glaser, D. A., 310
Gledhill, B. L., 236
Glimelius, K., 360, 361(9), 362, 364
Glover, F. L., 203
Glover, J. S., 418
Godawska, E., 566
GiShde, W., 240
Goetz, I., 33
Goldblatt, H. L., 264
Golde, D. W., 497, 498(7)
Goldman, D., 267
Goidsby, R., 350
Goldschmidt, E., 478
Goldstein, J. L., 444
Goldstein, S., 342, 445, 446(5)
Gonzalez, E., 272
Good, N. E., 136
Good, R. A., 371
Goodhead, D. T., 301
Goodman, H. M., 534
Goodpasture, C., 326
Goodwin, R. H., 451, 460, 461
Gopalakrishnan, T. V., 507
Gordon, S., 496, 503(6)
Gorer, P., 219
Gorham, L. W., 60
Gosch, G., 365, 366, 367
Gospodarowicz, D., 78, 102, 107, 572
Grace, J., 153, 155(10)
Grace, T. D. C., 457, 458(27)
Graessman, A., 412
Graf, T., 396, 403
597
Graffi, A., 417

Graham, A. F., 203
Graham, F. L., 412, 426, 434
Grambow, H. J., 367
Gray, J. W., 241,243, 246
Gray, L. H., 179
Greaves, R. I., 30
Green, H., 59, 97, 265, 347, 348(16), 352
Green, I., 499
Green, M., 417, 425, 426(2, 3), 431, 432,
433(14, 17), 434(4), 435
Greenblatt, M., 297
Greene, A., 31, 32, 33(17)
Greene, A. E., 33, 178
Greene, L., 585, 586(4)
Greene, R. C., 251
Greenwood, F. C., 418
Gregori, C., 544
Grewal, M. S., 174, 342
Griffin, J. E., 445
Grimes, E., 31
Grobstein, C., 125, 265
Gropp, A., 326
Gross, J., 274
Grossberg, S. E., 296
Group6, V., 379
Grout, B. W. W., 364
Gude, W. D., 279, 282(1)
Guguen, C., 544
Guilbert, L. J., 500
Gullino, P. M., 14, 184, 210, 264
Gunn, J. M., 60, 129
Gupta, N. P., 463
Gurner, B. W., 178
Guru, P. Y., 462, 463
Gvaramia, I. A., 174
Gwatkin, R. B. L., 138
H
Haapala, D. K., 423
Habeshaw, J., 495, 496(5)
Haegeman, G., 405
Haguenau, F., 417
Halaban, R., 566, 567(18)
Hales, A., 356
Hall, D. A., 272
Hall, R. F., 30
Halle, W., 60
Ham, R., 153
598
AUTHOR INDEX
Ham, R. G., 44, 49, 50, 52(9), 53(10), 55(1,

2), 59, 60, 61, 75(1, 2, 9), 76(1, 2, 9),
77(2, 18), 78(3, 10, 19), 79(1), 82(10),
83(13, 19), 89(10, 21), 90(1), 91(1, 2, 9,
20), 95, 107(9), 137, 153, 155(4-9), 264,
382, 522, 530, 545, 548, 565,571,577
Hamamoto, S. T., 553,555(6), 556(6), 557(6)
Hamerton, J. L., 343
Hamilton, 264
Hamilton, S., 153, 155(8, 9)
Hamilton, W. G., 52, 53(10), 60, 78(I0),
82(10), 83(13), 89(10), 95, 107(9)
Hammaker, L. E., 544
Hammond,,J. M., 533(55), 534(55), 535
Hammon, S., 153, 155(9)
Hammond, S. L., 51, 52(9), 60, 75(9), 76(9),
91(9)
Hammond, W. R., 125
Hanafnsa, H., 370, 380, 387, 390
Hanafusa, T., 387
Handley, H. H., 572
Hanks, J. H., 121, 142, 151
Hanks, J. W., 120
Hansen, D., 351
Hansen, E., 465(53), 466
Hard, W. L., 552
Harding, H. E., 564
Hardy, F. M., 203
Harper, E., 267
Harris, G., 264
Harris, H., 343, 346, 347
Harris, M., 149
Harrison, P. R., 506
Harrison, R., 584
Harrison, R. W., 534(65, 66), 535
Hartley, J. W., 369, 413,421,422(4), 423
Hartmann, J. X., 365
Hartmann, M., 185
Hashmi, P. S., 343
Hasson, J., 586
Hatanaka, M., 27
Haudenschild, C. C., 140
Haug, E., 533(58), 534(58), 535
Hauschka, S. D., 512, 521, 522(26), 524,
525, 577
Havaranis, A. S., 131
Havell, E. A., 292, 293, 295(3), 296(3)
Hawthorne, P. K., 174, 439
Hay, A. J., 294
Haysahi, I . , 530
Hayashi, H., 371
Hayashi, I., 60, 79, 88(23), 94, 99

Haynes, R. O., 267, 273(35)
Healy, G. M., 60
Heard, P., 162, 163(37)
Heaysman, J. E. M., 134
Heijneker, H. L., 426, 434(11)
Heine, U. l., 417
Heinemann, S., 96
Hellen, A., 493
Hemmelfarb, P., 14
Hendee, J., 566
Hendricks, H. R., 124, 130(7)
Heneen, W. K., 342
Henley, S. L., 216
Hensen, D., 356
Herbert, E., 530, 531(22, 23), 534(23)
Herrmann, H., 131
Hershberg, R. A., 265
Herzenberg, L. A., 235, 350
Hess, D., 360
Heston, W., 417
Heywood, S. M., 131
Higuchi, F., 146
Higucki, K., 60, 82, 85
Hildebrandt, A. C., 154
Hilfer, S. R., 126, 130, 138
Hill, B. T., 147, 148
Hill, D. A., 296
Hill, M., 391
Hillis, W. D., 271
Hillova, J., 391
Hilwig, I., 326
Hink, W. F., 455, 456, 457(16), 461
Hinide, P. M., 528, 529(13), 532, 533(13, 40,
41, 44, 47, 52, 53, 59), 534(13, 47, 52, 53,
59), 535
Hirano, A., 586
Hirohashi, S., 371
Hirsch, J. G., 497, 498(10)
Hirt, B., 406
Ho, C. C., 309, 310(9), 343
Hobika, G. H., 200
Hoch, H., 176
Hi56k, M., 267
Hohman, L. K., 320
Holland, J., 372
Holland, J. G., 507
Holm, D. M., 252
Holmes, R., 60
Holmgren, N. R., 165
Honn, K., 197
AUTHOR INDEX
Hoogasion, A., 30
Hopps, H. E., 22, 24, 25(11), 197, 405
Horine, R. K., 361
Horiwitz, S. B., 289
Horng, C. B., 184, 186(5), 207
Horwitz, A., 561,563(8)
Horwitz, A. L., 561,563(9)
Horwitz, Z. D., 528, 529(10), 533(10),
534(10)
Hosick, H. L., 147
Hoskins, J. M., 151
House, W., 14
Howard, J. C., 349, 351(26)
Howe, S. C., 349, 351(26)
Hoyt, R. F. Jr., 533(51, 54, 60), 534(51, 54,
60), 535
Hozumi, N., 350
Hsie, A. W., 288
Hsu, S. H., 454
Hsu, T. C., 173,220, 325,439
Hu, F., 564
Huang, L. C., 485
Hudson, B. G., 271
Hudson, L., 132
Huebner, R., 417
Huebner, R. J., 413
Hull, R. N., 37, 203
Humason, G. L., 287
Humphrey, R. M., 308
Humphreys, S., 583
Humphries, E. H., 389
Hungerford, D. A., 148
Hunter, W. H., 421
Hunter, W. M., 108, 418
Hutching, S. E., 94
599
J
Jacob, F., 97, 125(21, 22)
Jacobson, B., 268
Jaffe, E. S., 499
Jagannathan, V., 462
Jainchill, J. L., 413, 422(5)
Jakob, H., 97, 105(21)
Jameson, P., 296
Jamieson, C. W., 149
Jarrett, O., 297
Jenkin, H. M., 60, 454
Jensen, D. K., 528, 529(5), 533(5), 534(5),
535(5)
Jensen, F. C., 405
Jensen, M. D., 210
Jett, J., 241
Joftes, D. L., 287
Johansson, C., 174
John, R., 495
Johnsen,'J., 493
Johnson, K. M., 405
Johnson, M. J., 52, 60
Johnson, R., 352
Jones, C., 372
Jones, C. W., 364, 365, 366(30)
Jones, F. A., 124
Jones, G. E., 314
Jones, K. L., 94
Jones, V. E., 223
Jouanneau, J.-P.,482
Jovin, T. M., 246
Joworska, H., 287
Junghans, R., 377
Ignoffo, C. M., 457

Ikawa, Y., 507
Imae, T., 533(44), 534
Inbar, B., 368
Isaacs, R. J., 262
Iscove, N. N., 55, 500
Ishi, Y., 297
Itagaki, A., 434
Ito, M., 364
Ivankovic, S., 586
Iwakata, S., 153, 155(10)
Izawa, S., 136
Kabat, D., 507, 511(5)

Kafatos, F. C., 411
Kahan, B. D., 184
Kahn, R., 567
Kahn, R. H., 150
Kaighn, M. E., 60, 149
Kaji, A., 560
Kaltenbach, J. P., 151
Kaltenbach, M. H., 151
Kamentsky, L. A., 235
Kameya, T., 364, 371
Kanai, R., 364
Kao, F.-T., 309, 317, 321(17), 348, 352
600
AUTHOR INDEX
Kao, K. N., 360, 362, 363, 364(21), 365, 366,

367
Kaplan, J., 178
Kaplan, M. E., 493
Karsten, U., 147, 148
Kartha, K. K., 365, 366, 367
Kasten, F. H., 132
Kato, M., 60
Katsuta, H., 52, 60, 82, 85
Katsuta, H. T., 60
Katz, I., 152
Kauffman, R. S., 425
Kawai, S., 380, 390
Kee, S. G., 411
Keen, L. N., 13
Kehinde, O., 97
Keijzer, W., 347
Keller, W. A., 362
Kelly, D., 140
Kelley, T. J., 351,370, 405, 406(8), 412
Kelly, W. N., 146, 148
Kennedy, D., 267, 274(40)
Kennett, R., 355, 356
Kerkof, P. R., 126
Kerr, H. A., 60
Kersey, J. H., 264
Keuhl, N. M., 350, 356(28)
Kevei, F., 367
Kida, S., 544
Kieler, J., 151, 152(53)
Kikuchi, K., 60
Kikuchi, Y., 564
Kilburn, D. G., 184, 186(4), 207
Kimball, R. F., 288
Kimes, B., 96
Kimes, R., 425, 434(4)
Kimes, R. C., 432, 433(17)
Kimura, G., 434
Kinsley, N., 204, 205(28)
Kirk, J., 274
Kirkpatrick, J. W., 365, 366
Kitamura, H., 564
Kitos, P. A., 60
Kivirikko, K. I., 271
Klagsbrun, M., 561
Klebe, R. J., 267, 273(37), 354
Klein, G., 218, 343
Kleiner, G., 170
Klement, V., 421
Klevecz, R., 251,255
Klinger, H. P., 160, 371,375(5), 377
Klinman, N., 349, 350

Knaian, A., 533(40), 534
Knapp, M., 571
Knazek, R. A., 14, 183, 184, 210, 264
Ko, C.-P., 579
Koch, K., 542
Koch, K. S., 536, 543(3), 544
Kock, N., 566
Kohler, G., 349, 351(25), 357(25)
Kohler, P. O.,'184, 210, 264, 533(55), 534
(55), 535
Kohn, D. E., 411
Koide, T., 371
Koler, D., 507, 511(5)
Kolobow, T., 181
Komaiko, W., 561
Konigsberg, I. R., 511, 512(2), 517, 518,
519(3, 22), 523(3), 524(3, 21-23), 525,
526(3, 23, 43), 577
Konrad, M. W., 310
Kopriwa, B. M., 280, 287
Koprowski, H., 345, 349(3), 354, 356,
375(17), 376(17), 378, 405
Korner, A., 568
Koushanpour, E., 559
Kozoman, F., 400
Kraemer, C., 33
Kraemer, P. M., 268, 269(60)
Krakower, J. M., 418
Kramer, M. J., 423
Kritchevsky, D., 146, 148
Kriz, M., 533(61), 534(61), 535
Kruse, P. F., Jr., 4, 5(3), 13, 59, 137, 145
Kubicek, M. T., 96
Kucera, C. J., 203, 204(23)
Kuchler, R. J., 203
Kuchler, R. S., 14
Kuehl, N. M., 354
Kuhn, K., 271
Kullen, J. W., 204, 205(28)
Kurtti, T. J., 455, 456, 512
Kurz, J. B., 233
Kusano, T., 352
L
Labib, G., 363, 367(19)
Lake, S., 236
Laland, S., 493
Lambert, W. C., 261
AUTHOR INDEX
Landgraf-Leurs, I., 425, 426(2)
Landis, S. C., 583
Lang, W., 286
Langer, R., 561
Lapiere, C. M., 274
Lamer, J., 60, 79, 88(23), 99
Landau, T., 162, 163(36)
Latt, S. A., 317, 326
Lauris, V., 183, 184
Law, L. W., 369
Lawley, K. R., 271
Lazarus, K. J., 579
Leake, C. J., 463
Leblond, C. P., 287
Leder, P., 507
Lee, F. D., 302
Lee, G. F.-Y., 149
Lee, J. C., 467
Lefevre, P. A., 297
Leffert, H. L., 536, 541,542, 543(1-3), 544
Lehner, D. F., 564
Lehr, H., 31
Leibovitz, A., 59, 264, 463
Leibovitz, J., 356
Leighton, J., 264
Leighton, J. L., 553, 560(5)
Leighton, S., 554, 557(10)
Leinerman, J., 266
Leis, J., 401
Leiter, E. H., 60
LeJeune, J., 325
Lengyel, J., 454, 455(8, 13)
Lerner, A. B., 566, 567
Lerner, R. A., 218
Less, L. A., 564
Letourneau, P. C., 577
Levenson, R., 561
Levine, E. M., 22, 25(13), 27(16, 22)
Levine, D. W., 96, 185, 186(7, 8), 192,
193(9), 208
Levine, L., 97, 528, 529(2, 3), 530(2), 532(2,
3), 533(36), 534(3)
Ley, G. O., 60
Ley, K. D., 258
Leyva, A., Jr., 146, 148
Li, C. Y., 498
Li, S. Y., 454
Lieber, M. M., 416, 421
Lieberman, M. W., 297
Liebo, S. P., 31, 33(9)
Lightbody, J., 97
601
Like, A. A., 183, 184

Likely, G, D., 47, 153, 156(13)
Lillehei, C., 561
Lilly, F., 413
Lira, T. W., 444
Lima-de-Faria, A., 287
Lin, C. C., 342
Linder, E., 267, 273(43)
Lindl, P. A., 309, 318(8)
Lindmo, T., 532
Ling, C. T., 60
Ling, N., 530, 531(20)
Ling, V., 315
Linial, M., 391
Liotta, L., 266
Lippman, M., 268
Lippman, M. E., 549
Liskay, R. M., 343
Little, J. B., 534
Littlefield, J., 351,355(37), 359
Liu, H. Z., 364, 365, 366(30)
Livingston, D. M., 418, 421(18)
Loeb, D. H., 192
Loewenstein, W. R., 265
Lofgren, D. J., 317
Long, C., 27,352
Longstaff, E., 297
Louie, M. K., 132
Lowenstein, L., 487
Lowry, O. H., 144, 272, 403, 420, 568
Lubit, B. W., 326
Lubs, H. A., 174
Ludovici, P. P., 165
Lundholm, U., 425
Lwoff, A., 156
Lwoff, M., 157
Lynn, D. E., 462
Lynn, J. D., 211, 212(7), 213(7), 215(7),
216(6), 217, 218(7), 219(7), 223
Lyons, J., 528, 529(11)
Lyons, W. B., 151
M
McCarthy, K. D., 576, 581(6)
McCawley, P., 451
McCoy, T. A., 59
McCullough, E., 162, 163(35)
McCutcheon, J., 412
MacDonald, K. B., 151
602
AUTHOR INDEX
McDougall, J. D., 375(16), 376(16), 377(16),

378
McGardty, G. J., 10, 18, 19(4), 21, 22,
29(1-4), 375, 439
McHale, J. S., 457
MacHattie, A., 425, 434(4)
Mclntosh, A. H., 455
McKeehan, K. A., 50, 52(9), 60, 61, 75(9),
76(9), 91(9, 20), 153, 155(9)
McKeehan, W. L., 44, 49, 50, 52(9), 55(1, 2),
60, 61, 75(9), 76(1, 2, 9), 77(2), 78(19),
79(1), 83(13, 19), 90(1), 91(1, 2, 9, 20),
95, !07(9), 137, 153, 155(8, 9), 264, 577
Mackensen, S,, 96
Mackenzie, I. A., 367
McKenzie, L. S., 467
McKenzie, W. H., 174
Mackey, J. K., 435
McKinney, E. C., 467
McKusick, V, A., 341,345, 348(2)
MacLeish, P. R., 578, 583(14)
McLeod, M. P., 33
McLimans, W. F., 184, 186(5), 199, 202, 203,
204(23, 24), 207, 264
McMeans, E., 454
Macmorine, H. G., 60, 184
McNutt, N. S., 265
Maepherson, I., 160, 297, 302
McQuilken, W. T., 60
Macy, M. L., 12, 166, 167
Maina, D., 533(43), 534
Mains, R. E., 131, 530, 531(19, 20), 578,
580(I'/),583, 584(17)
Malamud, D., 250
Malemud, C. L, 561
Malinin, T. I., 150
Manager, R. A., 379
Maniatis, T,, 411
Mannucci, P. M., 166
Mansukhani, S., 553, 560(5)
Mao, W. H., 454
Maramorosch, K., 3, 455, 462
Marcus, L, 181
Marcus, P. C., 256
Marcus, P. I., 47, 160, 321, 522
Mardiney, M. R., 149, 151(31)
Margu/ies, D., 350, 351,356(28)
Margulies, D. H., 354
Mariage, R., 391
Markarjan, D. S., 174
Marks, E. P., 451, 457(1)
Marks, P. A., 506, 510(3)

Maroudas, N. C., 194
Marrack, J. g., 176
Marsh, W. H., 60
Marshall, R., 343
Martial, L A., 534
Martin, G., 266
Martin, G. M., 356, 357(63)
Martin, H. E., 14
Martin, J. L., 33
Martin, M. A., 411
Martin, S. G., 368
Martin, T. F. J., 528, 529(9), 532, 533(6, 33,
48), 534(6, 33, 48), 535
Martinez-Lopez, G., 149, 456
Martz, E., 267
Marunouchi, T., 455
Mascarenhas, A. F., 462
Mason, W. S., 387
Mastrangelo, L A., 364, 365, 366(30)
Masui, H., 96
Masurovsky, E. B., 577
Mather, J., 94, 96, 97, 102(6, 18), 103, 109(6)
Mathews, M. B., 268, 274
Maurer, W., 286, 289, 290(30)
Maxwell, M., 59
Mayall, B. H., 146
Mayer, A. J., 425
Mayer, K., 152
Mazur, P., 31, 33(9, 10)
Mealey, J. Jr., 162, 163(33)
Meck, R. A., 364, 365, 366(30)
Meek, G. A., 290
Meier, S., 561
Meins, F., 485
Meiss, H. K., 254, 317
Melamed, M. R., 152, 252
Melchers, E, 55
Melchers, G., 362, 363, 367(19)
MeUi, M., 248
Melnick, J. L , 417
Mendelsohn, J., 492
Mendelsolm, M. L., i46, 235, 241
Mened, B., 192
Merchant, D. J., 14; 149, 150, 203
Merrick, S,, 174
Merfick, S. B., 564
Meryman, H. T., 30, 33
Messier, B., 287
Meyer, U. A., 544
MeyneU, G. G., 151
AUTHOR INDEX
Mezger-Freed, L., 467
Michalopoulos, G., 266, 269(31), 270,
271(31), 544
Michaud, T., 567
Michayluk, M. R., 362, 363, 364(21), 365,
367
Michl, J., 499
Mickelson, M. M., 493
Miedeman, E., 145
Migeon, B. R., 315
Mijazawa, M., 219
Miller, D. A., 174, 325, 341,342, 343
Miller, E. J., 271,273(68)
Miller, O. J., 174, 325, 326, 341, 342, 343,
348, 439
Miller, O. L., Jr., 289
Miller, R. A., 367
Mills, S., 264, 278(7)
Milstein, C., 349, 351(25, 26), 357(26)
Minna, J., 346
Mironescu, S., 260, 261(27)
Misfeldt, D. S., 553,555(6), 556(6), 557(6)
Mishra, N., 297
Mitchell, B. F., 257, 258
Mitsuhashi, J., 452, 456, 462
Mix, T. W., 184
Miyake, T., 455
Miyamoto, T., 387
Mizrahi, A., 136
Mobley, W. C., 580
Moehring, J. M., 315
Moehring, T. J., 315
Moellmann, G., 566
Mohandas, T., 341, 343
Mohberg, J., 60
Moklebust, R., 33
Molander, C. W., 28, 29(29)
Montagnier, L., 160, 297
Montes De Oca, F., 166
Moore, G. E., 59, 136, 197, 204, 205, 211,
264, 564
Moran, J. S., 78, 102, 107
Moran, T., 536, 543(3)
Morgan, J. F., 59, 153, 155(12), 382
Morecki, R., 271
Morimoto, H., 148
Morley, C. G. D., 146, 147(17)
Morowitz, J., 564, 566, 567(10)
Morris, P. W., 352
Morrison, S. J., 60
Morrison, S. L., 351
603
Morton, H. J., 59, 60, 86, 153, 155(12), 382

Moscona, A., 138
Moscovici, C., 403
Moscovici, M. G., 403
Mosna, G., 455
Mosselmans, R., 267
Mossman, B. T., 151
Mueller-Eberhard, U., 544
Mulder, C., 426, 434(11)
Muller, P., 561,564(2)
Munson, R. J., 301
Murad, F., 557, 558(15)
Murashige, T., 478, 480, 482, 484, 485
Murphy, D. G., 375, 442
Murphy, H. M., 380
Murray, M., 567
Mustard, J. F., 493
Muto, M., 560
Mount, D. T., 264
Mukherjee, A. S., 288
Mullaney, P. F., 149
Munkelt, B. E., 182, 184(4)
Munro, H. N., 146
Murakami, O., 60
Murnick, J. G., 152
Murphy, D. G., 21
Murphy, W. H., 150
Myers, M. W., 293, 294(4), 295(4)
N
Nadal-Ginard, B., 377
Nagai, K., 371
Nagington, J., 30
Nagle, S. C., Jr., 60
Nairn, R,, 434
Nakano, R., 482
Natali, P. G., 218
Nathan, C., 501
Nathans, D., 370, 405, 406(8), 412
Nathanson, M. A., 130
Nebes, S., 401
Neff, J. M., 354
Neill, W. A., 288
Nell, M. B., 354
Nelson-Rees, W. A., 29, 165, 166, 174, 323,
342, 343,439, 440
Nemeroff, K., 286
Nettleship, A., 296
Neufed, E. F., 444
604
AUTHOR INDEX
Neuman, R. E., 60, 530

Nevo, Z., 561,563
Newbold, R. E., 297
Nias, A. H. W., 254, 257, 262
Nichols, W. W., 21,375, 442
Nicholson, W. E., 534(64), 535
Nicolas, J.-F., 97, 105(21)
Nienhuis, A. W., 346, 507
Niewiarowski, S., 493
Nimni, M. E., 564
Nirenberg, M., 530, 531(21)
Nisalak, A., 405
Nishikawa, K., 94
Nishimura, T., 359
Nishizuka, Y., 266
Nomura, J., 580
Nomura, S., 423, 424(33)
Nomura, T., 371
Norwood, T. H., 356, 357(63)
Nowell, P. C., 148
Nowinski, R. C., 418
Nurse, C. A., 578
Nussbaun, A., 199
Nyiri, L. K., 195
O
Obata, K., 575
O'Brien, S. J., 170
Obrink, B., 267
Oehlenschlager, V., 59
O'Farrell, M. K., 251
O'Gorman, P., 219
Ohasa, S., 96
Ohishi, M., 124, 130(9)
Ohlhaum, D. J., 499
Oie, H. K., 296
Oka, M. S., 149
Okada, Y., 355, 356(53), 359
Okayama, M., 560
O'Lague, P. H., 575, 578, 583(14)
Old, L. J., 219
Oldberg, A., 267
Oldstone, M. B. S., 218
Oliveros, R., 200
Olson, R., 170
O'Malley, K., 356
O'Neill, M. C., 521
Onoue, K., 124, 130(9)
Oroszlan, S., 418
Orth, D. N., 534(64-66), 535

Osborne, B., 350
Osborne, R., 533(43), 534
Ostrowski, K., 288, 289(26)
Owen, A. A., 256
Owen, O. V. H., 211
Owren, P. A., 493
Oyama, V. I., 144, 145, 555, 558(12)
O'Young, S., 184
Ozanne, B., 425, 426(6)
Ozawa, E., 522
Ozzello, L., 97
P
Padilla, S. R., 564
Pagano, J. M., 412
Pahl, K., 365, 366
Palser, H. R., 343
Pan, J., 405
Panayi, G. S., 288
Pannett, C. A., 523, 526(37a)
Panol, G., 184
Pant, U., 462
Papahadjopoulos, D., 55, 359

Paquette, T. L., 530, 531(23), 534(23)
Parder, A., 555
Pardue, M. L., 326
Park, C. R., 544
Parker, M. L., 528, 529(2), 530(2), 532(2)
Parker, R. C., 4, 5(5), 59, 60, 121, 153,
155(11, 12), 382
Parker, R. F., 30, 545, 548(2)
Parks, A. S., 30
Parks, W. P., 264, 417, 418, 421(18)
Parsa, I., 60
Parshad, R., 264, 369
Parson, J. T., 425, 426(2)
Partlow, L. M., 576, 581(6)
Passey, R. D., 564
Pastan, I., 267, 274(38--40), 369
Paterson, B., 521
Pati!, S. R., 174
Patrick, J., 96
Patterson, D., 309
Patterson, M. S., 251
Patterson, M. R., Jr., 4, 141
Patterson, E H., 131,307, 575, 578, 580(17),
581(4), 582, 583(4, 24), 584(4, 17)
Paul, D., 536, 543(1, 2)
AUTHOR INDEX
Paul, J., 4, 151
Pawelek, J., 564, 566, 567(10, 11, 18), 568
Paz, M. A., 271,272(62)
Pearlstein, E., 273
Pearse, A. G., 287
Pearson, P., 325
Pearson, P. L., 343
Peaud-Lenoel, C., 482
Peebles, P. T., 423,424(34)
Peehl, D. M., 90
Pellegrino, A., 218, 219
Pellegrino, M. A., 218, 219
Pelletier, G., 360
Penman, S., 454, 455(8, 13)
Penso, G., 4, 5(6)
Peraino, C., 150
Peralta, P. H., 405
Perdue, S. W., 288
Perez, A. G., 22, 27(15)
Perlman, D., 115, 211
Perper, R. J., 493
Perrone, M. H., 533(59), 534(59), 535
Perry, V. P., 33, 150
Pessac, B., 268
Pestka, S., 292
Petersen, D. F., 255, 343
Peterson, E. R., 577
Peterson, J. A., 347
Peterson, W. D., Jr., 165, 166, 174, 178
Peterson, W. P., 33
Petterson, U., 425, 426(6)
Pham, T. D., 326
Philip, J., 166
Phillips, D. M., 22, 28, 375
Phillips, G. B., 38
Phillips, G. W., 140
Phillips, H. J., 139, 151,218
Phillips, H. M., 267
Phillips, L. A., 423
Pictet, R. L., 265, 352
Pierce, S. L., 287
Piez, K., 525
Pierson, R. W., Jr., 571
Pike, R. M., 38
Pilkis, S. J., 544
Pina, M., 425, 426(2), 431,432, 433(14, 17),
434(4)
Pincus, T., 413
Pirt, S. J., 60
Pitelka, D. R., 126, 266, 269(32), 271(32),
272(32), 553, 555(6), 556(6), 557(6)
605
Pitot, C., 544

Pitot, H. C., 266, 269(31), 270, 271(31)
Pitts, J. D., 265
Platkowska, E., 288, 289(26)
Piichon, M. P., 391
Polacco, J. C., 360
Polge, C., 30
Pollack, R., 297, 348, 371, 375(4), 376(4),
377(4), 378(4)
Pontecorvo, G., 356
Pontecoruo, G., 356
Pont6n, J., 273
Popescu, N. C., 343
Popowski, A., 544
Porter, K., 368
Porter, K. R., 140
Poste, G., 55, 345, 359
Potrykus, I., 365, 366
Potten, C. S., 266
Potter, D. D., 575, 578, 583(14)
Potter, M., 350
Poulson, C. O., 348
Pouyssegur, J., 369
Povlsen, C. O., 276, 370
Powell, A. K., 203
Power, J. B., 361, 365, 367
Powsner, E. E, 33
Preisler, H. D., 506, 507(1), 508(1)
Prescott, D. M., 280, 282, 284(7), 289, 368
Press, J. L., 349, 350
Pretlow, T. G., 132
Preussmann, R., 586
Price, F. M., 5
Price, G., 162, 163(35)
Price-Jones, M. J., 367
Primbsch, E., 289, 290(30)
Pruss, M., 267
Przyblyski, R. J., 288
Puck, T. T., 32, 47, 59, 121, 142, 160, 309,
317, 321(17), 348,352. 516, 548
Pudney, M., 455, 463
Pugh, W. E., 421,422, 423(30)
Puhlendorf, C., 296
Pumper, R. W., 8
Punyashthiti, K., 445
Purchase, I. F. H., 297
Q
Quade, K., 401
Quevedo, W. C., 565
606
AUTHOR INDEX
Quimby, M. C., 135, 467, 475, 477

Quinn, L. A., 564
Quiot, J. M., 457
R
Radlett, P. J., 211

Radloff, R., 407
Ragni, G., 352
Rahman, S. B., 115
Raiborn, C., 576, 581(5)
Rake, G. W., 203, 204(23)
Rand, E., 417
Randall, R. J., 272, 403, 420, 568
Randall, R. S., 144
Rands, E., 22, 27(17)
Rappaport, C., 127
Rasbeed, S., 343
Rattazzi, M. C., 166
Ray, M., 341, 343
Read, S., 266
Reardon, M. E, 352
Rechtoris, C., 455
Reddy, V. B., 405
Rees, R., 579
Regoeczi, E., 493
Reichardt, L. F., 582, 583(24), 584
Reid, L., 107, 155, 160, 162(17), 264, 276
Reid, L. M., 372
Reinert, J., 365, 366, 367
Reinbard, E., 478
Reisfeld, R. A., 218
Rej, R., 146
Rekosh, D. M. K., 433
Remberger, K., 271
Reuben, R. C., 506, 510(3)
Reuveni, O., 478
Reynolds, R. K., 416
Rheinwald, J. G., 265
Rhodes, C., 411
Rhodes, K., 271,273(68)
Richards, A. H., 146
Richards, G. M., 149
Richardson, B. J., 343
Richardson, B. L., 456
Richardson, P. D., 184
Richardson, U. I., 528, 529(15), 530,
532(18), 534(18)
Richter, A., 60, 545
Richters, V., 60
Rifkind, R. A., 506, 510(3)

Rigby, P. W. J., 411
Rinaldini, L. M., 126, 452, 511,512(1)
Rinaldini, L. M. J., 125
Rindler, M. J., 160, 552, 559, 560
Rinfret, A. P., 31
Ringer, S., 120, 121
Ringertz, N. R., 347, 359(18)
Risser, R., 297, 348, 371, 375(4), 376(4),
377(4), 378(4)
Risteli, L., 271
Rizzino, A., 105
Robb, J. A., 156, 157(20)
Robbins, E., 256
Roberts, J. L., 530, 531(22)
Robillard, N., 24
Robinson, A., 59, 121, 142
Robinson, A. J., 433
Robinson, A. R., 166
Robinson, F., 199
Robinson, H., 252
Rod6n, L., 274
Rogentine, G. N., Jr., 257
Rogers, A. W., 280, 285(5), 286(5), 287, 290
Rogiers, R., 405
Roholt, O., 199
Rojkind, M., 271,272
Ron, A., 280
Ronveaur, M. F., 267
Roscoe, D. H., 252
Rose, G. G., 14
Rosenbrough, A. L., 568
Rosenbrough, N. J., 144, 272, 403, 420
Rosenfeld, M. G., 533(50), 534(50), 535
Rosenstreich, D. L., 491
Ross, J., 416, 417(8), 507
Ross, P., 248
Rothblat, G. H., 44
Rounds, D. E., 467
Rous, P., 124
Rouse, H. C., 432
Rouslahti, E., 267, 273(43)
Rowe, W. P., 369, 413, 421,422(4), 423(30)
Rubalcava, B., 536
Rubin, H., 134, 368, 379, 391
Rubin, K., 267
Ruddle, F. H., 292, 341, 345, 346, 348(2),
354, 359(51)
Rudkin, G. T., 148
Rudland, P. S., 251
Ruesink, A. W., 361,362, 364
AUTHOR INDEX
Rugaard, J., 348
Runyan, C., 211, 212(7), 213(7), 215(7),
218(7), 219(7)
Rushton, A. R., 343
Russell, W. C., 433, 434
Russin, D., 149
Russlahti, E., 267, 273(34)
Rutter, W. J., 265, 352
Rutzky, L. P., 184
Rygaard, J., 276, 370
Rymbrandt, D., 266
S
Sabol, S. L., 530, 531(21)
Sachs, L., 162, 163(36), 296, 368
Saier, M. H., Jr., 160, 375(14, 18~, 376(14,
18), 377(18), 378,552, 554, 559, 560
Saigo, K., 455
Sakakura, T., 266
Salmon, S. E., 288
Sambrook, J., 411,425, 426(6)
Samuels, H. H., 98, 528, 529(8-10), 532(810), 533(8-10), 534(8-10)
Sanders, F. K., 151
Sanford, K. K., 5, 60, 153, 154, 156(13, 14),
188, 197, 209, 264, 369, 545
Sansone, J., 566
Sansone, M., 564, 566, 567(10)
Sarama, J., 18, 19(4), 29(4)
Sarangi, F., 315
Sargent, P. A., 314
Sarkar, N. H., 418
Sartorelli, A. C., 59
Sato, G. H., 60, 79, 88(23), 94, 96, 97,
98(14), 99, 102(6, 18), 105, 107, 109(6),
155, 160, 162(17), 264, 267(8), 278(7),
375(14, 18), 376(14, 18), 377(18), 378,
528, 529(2), 530(2), 532(2), 554, 564,
570, 571,585
Sato, K., 359
Sato, T., 271,507
Saunders, J. W., Jr., 265
Savage, R. E., 347
Sawicki, W., 151, 152(53), 288, 289(26)
Sawyer, B. D., 561,563(7), 564(7)
Scaramuzzino, D., 60, 128
Schafer, I. A., 125
Schapira, F., 544
Scharff, M. D., 350, 351, 354, 356(28)
607
Schenk, D. K., 456

Schenker, A., 580
Scher, W., 506, 507(1), 508(1)
Scherer, W. F., 30, 166
Schieder, O.. 367
Schilling, E. L., 14, 60, 203
Schilling, R. L., 211
Schimmer, B. P., 571
Schindler, R., 158
Schleicher, J. B., 210
Schlesinger, D., 267, 274(40)
Schlesinger, R. W., 432
Schiossman, S. F., 132
Schmahi, D., 586
Schmidt, J., 31
Schmidt, M., 371
Schneider, E. L., 22, 26, 27(14, 24), 145, 146
Schneider, I., 452, 455, 462, 464(12, 43), 465
Schneider, J. A., 37
Schneider, L. K., 287
Schonbrunn, A., 528, 529(7), 533(7), 534(7)
Schrek, R., 151
Schubert, D., 96, 585
Schuller, R., 350
Scolnick, E. M., 416, 417(8), 418, 421(18)
Scott, J. F., 251
Seabright, M., 174, 335
Seeberg, P. H., 534
Seecoff, R. L., 512
Seefried, A. V., 60
Seegmiller, J. E., 444
Seglen, P. O., 125, 128
Seifter, S., 267
Seiji, M., 565
Sekhon, G., 356
Sela, B.-A., 492
Sell, S., 536, 541,542, 544
Selpeter, M. M., 280
Semar, J. B., 115
Senyi, A., 493
Setaro, F., 146, 147(17)
Shaeffer, L., 553
Shainberg, A., 522
Shannon, J. E., 12, 166, 167, 170, 188, 209
Shapiro, L. E., 528, 529(10), 533(10),
534(10)
Shapiro, L. J., 444
Sharp, P. A., 425, 426(7)
Sharpless, T., 252
Sherr, C. J., 416, 417, 421,424(28)
Sherton, C. C., 507, 511(5)
608
AUTHOR INDEX
Sherwood, P., 210

Shevach, E. M., 499
Shiba, T., 455
Shimosato, Y., 371
Shin, S., 28, 160, 276, 297, 348, 368, 371,
372, 374(8), 375(3-5, 19, 20), 376(3-5),
377(3, 4, 19, 20), 378(3, 4)
Shooter, E. M., 580
Simpson, S. B., Jr., 512
Singer, R. H., 454, 455(8)
Singh, K. R. P., 454, 462(7)
Shipley, G., 60
Shipley, G. D., 61
Shipman, C., Jr., 127, 138
Shonia, N. V., 174
Short, S. S., 145, 146, 148
Shortman, K., 497
Siciliano, J., 308
Siciliano, M. J., 308
Sigel, M. M., 467
Silverstein, S. C., 499
Siminovitch, L., 203, 308, 309, 310(2), 315,
317
Simpson, E., 350
Simpson, W. F., 165, 174, 175, 178
Sinclair, R., 60
Singh, R. M. M., 136
Singley, A., 197
Skehel, J. J., 294
Skelly, H., 544
Skoog, F., 480
Skyler, J. S., 183
Slesinski, R., 359
Sligh, D. D., 203
Sly, W. S., 356
Smadel, J. E., 405
Smiley, J., 434
Smith, A. N., 30
Smith, B. J., 252
Smith, D. F , 127
Smith, G. L., 107, 162, 163(32)
Smith, H. H., 359, 360, 362(2), 364, 365,
366(2, 30), 367(2)
Smith, J. D., 95, 298, 530
Smith, R. E., 243, 399, 400, 401,403
Smith, S., 126, 561
Smith, T. F., 552
Snary, D., 221, 223(1)
Sohi, S. S., 457
Sokoloff, L., 561
Soller, A., 297
Solt, M. L., 485

Solursh, M., 561
Sonnenschein, C., 532, 534
Southern, E., 411
Sparkes, R. S., 347
Spier, R. E., 184, 186(6), 190(6)
Spinelli, G., 248
Spiro, R. G., 271
Sprandling, A., 454, 455(8, 13)
Stadtler, J., 351,356
Stallcup, W., 585
Stamato, T. D., 320, 321
Stanbridge, E., 21
Stanbridge, E. J., 22, 26, 27(14, 24)
Stanely, P., 309
Stanford, K. K., 47, 60
Stanley,,F., 162, 163(37)
Stanley, R., 528, 529(9, 10), 533(9, 10),
534(9, 10)
Stanley, T. B., 418
Stanners, C. P., 59, 317
Stanton, T. H., 213
Stehlin, J. S., 375(15), 376(15), 378
Steinbach, W. Culp, 96
Steinberg, M. G., 513
Steinberg, M. S., 125, 267
Steinkamp, J. A., 149
Steinmetz, L. L., 236
Stephenson, J. R., 416, 418
Stephenson, N. G., 467
Steplewski, Z., 356
Stetten, G., 317
Stevenson, R., 178
Stewart, G. J., 493
Stiles, C., 160
Stiles, C. D., 160, 375(14, 18, 22), 376(14,
18), 377(18), 378, 554
Stockdale, F. E., 521, 523
Stoker, M., 297, 300
Stoker, M. G. P., 134
Stone, G. E., 289
Storie, B., 310
Stossel, T. P., 496, 498(7)
Stow, N. D., 434
Strand, M., 418
Straus, E., 151
Straus, F. H., 151
Strauss, E., 455, 461
Streibel, M. J., 132
Strohman, R. C., 521
Stuart, A. E., 495, 496(5)
AUTHOR INDEX
Stuhblefield, E., 251,255
Studzinski, G. P., 261
Stulberg, C. S., 33, 165, 166, 174, 175, 178
Stumpf, W. E., 289
Styles, J. A., 297, 302(10, 11)
Subramanian, K. N., 405
Sugano, H., 507
Sugg, M. T., 203
Sulkin, S. E., 38
Sullivan, J. C., 125
Sullivan, P. M., 90
Summers, J., 434
Sumner, M., 162, 163(34)
Sun, A. M., 184
Sun, T., 265
Sun Lin, P., 221
Sussman, P. M., 533(56), 534(56), 535
Sutherland, Win. M., 520
Swanborg, N. K., 166
Sweet, B. H., 457
Sweet, R. G., 235
Sweet, W., 97
Swim, H. E., 30, 545, 548(2)
Sykes, J. A., 140
Syverton, J. T., 166, 178
Szegedi, M., 367
Szybalski, E., 352
Szybalski, W., 351,352
T
Tadokoro, J., 355, 356(53)
Taft, E. B., 251
Takaoka, T., 52, 60, 82, 85
Takebe, I., 363, 367(19)
Takemoto, H., 528
Talavera, A., 348
Tan, Y. H., 292
Tarikas, H., 96
Tashjian, A. H., Jr., 96, 98(14), 264, 267(8),
528,529(2, 3, 5-7, 11-13), 530(2), 532(2,
3), 533(5-7, 12-14, 33, 36, 40-45,
47-49, 51-54, 57, 60, 62, 63), 534(3,
5-7, 12-14, 33, 47--49, 51-54, 56, 57, 60,
62, 63), 535(5), 564
Taub, M., 552, 560
Taylor, J., 265
Tchao, R., 264, 554, 557(10)
Teague, J., 147, 148
Teleki, S., 60
609
Telling, R. C., 211

Temin, H. M., 107, 162, 163(31, 32), 379,
380, 389, 416
Tennant, J. R., 151
Tepperman, K., 513
Terasima, T.,.193, 252, 254, 256
Thayer, P. S., 14
ThilIy,W. G., 96, 185, 186(7,8), 192, 193(9),
208
Thimann, K. V., 361
Thimmappaya, B., 405
Thomas, C. A., Jr.,425, 434(4)
Thomas, J. A., 52
Thomas, W. J., 203, 204(24)
Thompson, E. B., 507, 544, 549
Thompson, L. H., 308, 309, 310(2), 312,
314(11), 317(11), 318(8), 353, 559(I)
Thompson, M. H., 297
Thompson, S. N., 287
Thurston, J. M., 287
Till,J., 162, 163(35)
Till,J. E., 256
Tischler, A. S., 585, 586(4)
Tixier-Vidal,A., 585
Tjio, J. H., 405
Tobey, R. A., 240, 255, 258
Todaro, G. J., 22, 27(17), 37, 96, 264, 413,
416, 417(8), 421, 422(5, 6), 424(28)
Todd, P., 132
Tolmach, L. J., 193, 252, 254, 256
Tolnai, S., 151, 152(54)
Tomita, J. T., 184
Tomlinson, R. H., 179
Tompkins, G. J., 451
Tomkins, G. M., 544
Tonegawa, S., 350
Toole, B. P., 268
Tooze, J., 370, 395
Topp, W., 411
Toth, E. M., 343
Toyoshima, K., 389
Traganos, F., 252
Trauthen, T., 297
Trelstad, R. L., 268, 271
Tremlay, R., 262
Tritsch, G. L., 197
Trujillo, T. T., 252
Tsai, J. S., 98, 528, 529(8, 9), 533(8, 9),
534(8, 9)
Tsao, J., 571
Tupper, J. T., 257, 258
610
AUTHOR INDEX
Turek, L., 355

Turner, J. K., 168
Tushinski, R. J., 528, 529(14), 533(14, 56),
534(14, 56), 535
Turtle, N., 423, 424(32)
Tyler, R. W., 486
Tyrode, M. V., 121,452
Tytell, A. A., 60, 530
Tze, W. J., 184
Voft, M., 586

Vogl, W., 38
Vogt, M., 95, 120, 121, 142, 156, 298, 406,
530, 572
Vogt, P. K., 370, 380, 381,387, 389, 390, 395
Volckaert, G., 405
Volpe, P., 145
von der Mark, H., 561,564(2)
von der Mark, K., 561,564(2)
von H. Owens, O., 203
Vosseller, G. V., 136
U
Uchida, T., 359
Uchimiya, H., 484
Uecker, W., 146
Ungaro, E C., 149, 151(31)
Ussing, H. H., 555, 557(14)
V
Vagi, Y., 136
Vaheri, A. J., 267, 273(43), 278
Vail, W. J., 55, 359
Valentich, J. D., 554, 557(10)
Vanaman, V., 18, 19(4), 29(4)
van der Eb, A. J., 412, 426, 434(11)
Van de Voorde, A., 405
Van Diggelen, O. P., 28, 375
Van Dilla, M. A., 235, 252
Vanha-Perttula, T., 533(553,534(553,535
van Hemert, D., 184, 186(4), 207
Van Herreweghe, J., 405
Van Heuverswyn, H., 405
Vanucchi, S., 269
van Wezel, A. L., 184, 186(1-4), 207
Varga, J. M., 566
Varma, M. G. R., 463
Varnum, D., 31
Varon, S., 576, 580, 581(53
Vas, M. R., 487
Vaughan, F. L., 265
Vaughn, J. L., 135, 451,455, 460, 461(38a)
Vaughn, V. L., 356
Velez, R., 346
Venuta, S., 368
Veri, A., 267
Vesell, E. S., 166
Vi~ek, J., 292, 293, 295(3), 296(3)
Vinograd, J., 407
Virolainen, M., 502
W
Walaas, E., 533(46), 534
Walaas, O., 533(46), 534
Walcott, R. M., 213
Waldren, C. A,, 321
Walicke, P. A., 307
Wallace, J. H., 149, 151
Wallace, R. E., 120, 121
Wallach, D. F. H., 210, 221
Wallin, A., 360, 361(9), 362, 364
Walton, M. J., 60
Waltz, H. K., 209
Waltz, J. K., 188
Wang, D. I. C., 96, 185, 186(7, 8), 192,
193(9), 208
Wang, L.-H., 380
Wang, R. J., 75, 136
Wanson, J. C., 267, 544
Wantanabe, H., 534(64), 535
Warburg, B., 368
Warburton, D., 325, 343
Wamaar, S. O., 426, 434(11)
Warren, R. J., 446
Wartiovaara, J., 267, 273(43)
Wasley, G. D., 495
Wasmuth, J. J., 317
Watabe, H., 544
Watts, R. W. E., 147
Waymouth, C., 31, 44, 60, 78(3), 141, 142,
149, 264, 500
Weber, G., 361,365, 366
Weber, M., 368
Wedum, A. G., 38
Weed-Kastelein, E., 347
Weinstein, D., 125, 146, 541
Weinstein, D. B., 542
Weinstein, R. S., 265
Weiss, M., 347, 348(16)
AUTHOR INDEX
Weiss, M. C., 347
Weiss, R., 210
Weiss, R. A., 370, 387
Weissman, S. M., 405
Wensink, P. C., 425,434(4)
Werb, Z., 496, 503(6)
Werdelin, O., 124, 130(8)
West, N. R., 579
Westermark, B., 273
Westwood, F. R., 297
Wetter, L. R., 3, 366
Weyns, J. C., 149
Whatley, S., 147, 148
Wheeler, T. B., 356
White, M., 553
White, N. K., 521,524(25)
White, P. R., 59
Whiteside, J. P., 184, 186(6), 190(6)
Whitfield, 262
Whittle, W. L., 13, 137
Widholm, J. M., 361
Wiebel, F., 248
Wiepjes, G. J., 129
Wife, R. L., 506, 510(3)
Wigglesworth, N. M., 252
Wigley, C. B., 297
Wigzell, H., 219
Wilkie, N. M., 434
Williams, G. M., 60, 128, 129, 544
Williams, J., 425,426(6)
Williams, J. F., 267, 425
Williams, J. L., 213
Williams, L. J., 375(15), 376(15), 378
Williams, R. H., 147, 148
Williamson, F., 361
Willing, M., 507
Willingham, M., 369
Willis, G. E., 365
Wilson, J. D., 445
Winget, G. D., 136
Winter, W., 136
Wise, K. S., 211,216(6), 223,228
Wiseman, L. L., 125
Wither, L. A., 363
Wittinghorn, D. G., 31, 33(9)
Wittmarm,., 512
Wold, W. S. M., 425, 426(3), 435
Wolf, C. F. W., 182, 184(4)
Wolf, K., 135, 467, 475, 477
Wolfe, S. W., 60
Wollenberger, A., 60, 147, 148
Wong, F. C., 184
611
Wong, G., 564, 567(10)

Wong, J. S., 96, 185, 186(8), 208
Wood, P., 579
Woods, L. K., 564
Woods, M. W., 369
Woroch, E. L., 533(40), 534
Worton, R. G., 309, 310(9), 343
Wright, B. M., 223
Wu, C. H., 271
Wu, R., 94, 102(6), 109(6)
Wyatt, S. S., 457
Wyke, J. A., 391
Y
Yaffe, D., 516
Yaffe, I. R., 512
Yahara, I., 492
Yam, L. T., 498
Yamada, K. M., 267, 274
Yamada, S. S., 267, 274(38, 39)
Yamaizumi, M., 359
Yamamoto, K., 346
Yamane, I., 60
Yamasaki, E., 302
Yang, T. K., 454
Yanishevsky, R., 146
Yasumura, Y., 96, 98(14), 264, 267(8), 528,
529(2), 530(2), 532(2), 564, 570
Yavin, E., 577
Yavin, Z., 577
Yen, P. M., 533(57), 534(57), 535
Yergarian, G., 354
Yip, D. K., 151
Yokoro, K., 528
Yokota, H., 248
Yoshida, F., 146
Yoshimura, M., 560
Young, C. S. H., 425
Ysekaert, M., 405
Yu, L.-Y., 528, 529(14), 533(14, 56), 534(14,
56), 535
Yunker, C. E., 460, 461(38a)
Z
Zain, B. S., 405
Zapol, W., 181
Zaroff, L., 264, 278(7)
612
Zech, L., 174
Zee, T. W., 493
Zeigler, C. J., 356, 357(63)
Zenk, M. H., 478
Zerahn, K., 555, 557(14)
Zerthen, J., 347, 359(18)
AUTHOR INDEX
Ziegler, D. W., 203, 204(24)
Zimmerman, H. M., 586
Zimmerman, M., 267
Zimring, A., 152
Zwemer, R. K., 211, 212(7), 213(7), 215(7),
216(6), 218(7), 219(7, 14), 223, 224
SUBJECT INDEX
613
Subject Index
A
A2 APG medium, 57
Abortive infection, see Infection, abortive
Acetate, in media, 66
Acetoorcein, autoradiography prestain, 286
Acid cleaning, 5
Acridine orange, cell viability, 152
ACTH, see Adrenocorticotropic hormone
Action potential, in culture, 307
Adenine, in media, 54, 66
Adenocarcinoma
cervical, monolayer culture, 133
renal, 376
Adenosine, resistant strains, 314
Adenosine monophosphate, in media, 66
Adenosine triphosphate, in media, 66
Adenovirus
availability, 426
human, 425--435
isolation of DNA, 426
large-scale growth, 431--433
media, 427-431
monolayer culture, 426, 430
purification, 431-433
serotypes, 425
suspension culture, 426, 430
transformation, 370, 426
tumorigenicity, 425
Adenovirus DNA, 433,434
Adhesion
cellular, 369
to vessels, 137
Adhesion protein, 267
preparation, 273,274
Adipocyte, differentiation, 105
Adrenal cell Y1, see Y1 adrenal cell
line
Adrenal cortex carcinoma, 376
Adrenal tumor, plating, 573
Adrenergic neuron, 575
Adrenocortical Y1 cell, see Y1 adrenal cell
line
Adrenocorticotropic hormone
effect on Y~ cell line, 571
pituitary tumor cell, 527
release in GH cells, 534
Aedes aegypti
media, 457, 462

suspension culture, 454
Aedes albopictus
media, 462
Aedes novalbopictus, media, 462
Aedes taeniorhynchus, media, 462
Aedes vexans, media, 457
Aedes w-albus, media, 462
Aerosol, biohazard, 39, 40
African green monkey kidney cell
microcarrier culture, 190
SV40, 404
Agallia constricta, media, 462
Agar assay, soft, 297, 299, 300
Agglutinability, transformed cells, 368, 369
Aging, cell, 30
Aging Cultured Cell Repository, 442
Air circulation, 40, 43
Alanine, in media, 63
Albumin, in media, 88, 500
Alpha-MEM medium, 56
Alveolar macrophage, see Macrophage, alveolar
AMBD 647/3 medium, 59
Amelanotic mutant, see Mutant, amelanotic
a-Amanitin~ resistance, 309, 314
American Type Culture Collection, 119,440,
441
Ames test, 302
Amino acids
essential, in media, 53, 54, 62
in insect media, 455
nonessential, in media, 63
Amino acid derivative, in media, 63
Aminoacyl-tRNA synthetase, mutants, 319
p-Aminobenzoic acid, in media, 64
a-Aminobutyrate, in media, 63
Amoxicillin, in media, 112, 114
AMP, see Adenosine monophosphate
Amphibia
balanced salt solution, osmolarity adjustment, 469
incubation temperature, 471
media, 469
skin culture, 473
614
SUBJECT INDEX
Amphotericin B, in media, 112-I 14, 116

Amphotericin B methyl ester, in media, 112
Ampicillin, in media, 112, 114, 116
Ampule, shipment, 34
Amylase, tissue culture, 125
Anchorage dependence, 44, 45, 48, 81, 133,
137, 152
Anterior pituitary cell
calf, microcarrier suspension culture, 207,
208
Antheraea eucalypti, 451
media, 457
storage, 453
Antibiotic, see also specific substance
in cell culture, 95, 110-116
antimicrobial spectrum, 112, 113
criteria for usefulness, 110
commercial source, 113-115
cytotoxicity, 111
DNA determination, 147
excipients in, 111
resistance, 19, 115, 116
solubility, 113
stability, 113
sterility, 111
strategy of use, 115, 116
Antibody
antiviral, titration, 419, 420
monoclonal, 350
Antibody labeling, fluorescent, 175, 176
Antidiuretic hormone, 559
Antigen
carcinoembryonic, 184
tumor-specific cell surface, 371
Antigen-antibodyreaction, cell monitoring,
174-178
Antiserum
cell monitoring, protocol, 176-178
fluorescent, 175, 176
preparation, 174, 175, 276, 277
Arbovirus, interferon producing, 293
Arginine, in media, 53, 62
Armigeries subalbatus, media, 462
Ascites hepatoma, dispersion, 125
Ascites tumor, 376
Ascorbic acid, in media, 54, 64
Asparagine, in media, 63
Asparagus, from protoplasts, 367
ATCC, see American Type Culture
Collection
ATP, see Adenosine triphosphate

Atropa, from protoplasts, 367
A t r o p a petunia, 365
Australian emperor gum moth, media, 457
Autoclaving, 6, 7
Autoradiography, 279-292
artifacts, 290-292
cell culture, 281,282
cell synchrony, 249-251
chemography, 291
darkroom, 280, 281
safelight, 281
developer, 285
diffusible substances, 289
double isotope, 287, 288
environmental radiation, 292
exposure, 285
film, 287
fixation, 282
high-speed scintillation, 288, 289
light exposure, 291
photographic processing, 290, 291
poststaining, 286, 287
pressure effects, 291
prestaining, 286
procedure, 240, 281-286
protocol, 250, 251
self-absorption, 289, 290
staining, 285, 286
static discharges, 291
Auxin, in media, 478
Auxotroph, selective media, 352
Avian pineal gland, cell dissociation, 130
Avian retrovirus, 379
8-Azaadenine, resistant strains, 314
8-Azaguanine, hybrid selection, 352
8-Azaguanine-resistant cell line, 376
Azetidinecarboxylic acid, resistant strains,
314
B
Bacitracin, in media, 565
Bacterial contamination,see Contamination,
bacterial
Balanced salt solution, see Salt solution,
balanced
Balb/c cell, effect of hormone, 101, 102
Balb 3T3 cell, 96
Banding
chromosomal, 324
techniques, problems, 337
isopycnic viral, 415
615
SUBJECT INDEX
Barley mesophyll x soybean culture, 365
Basement membrane, 265
collagen, 266
components, in media, 269
in malignancy, 266
reconstituted, rafts, 263-278
Batch growth, cost, 206
B cell, 178, 493
B1e cellline, 564
B1e melanoma M 2 R cell,96
Benzidinc, staining, 510
N6-Benzyladenine, as cytokinin, 478
Benzylpenicillin,see PenicillinG
B H K cell,trypsinization,298
B H K 21 cellline, 377
Bicarbonate
as buffer, 199, 200
in media, 70
Bilirubin, conjugation, 184
Biohazard, 36--43
aerosols, 39, 40
infectious virus, 37, 42, 43
lymphoid cells, 37, 42
Biomphalaria glabrata, media, 466
Biopsy, skin, see Skin biopsy
Biotin, in media, 54, 64
Birch and Pirt's medium, 58
14-Bis-2-(4-methyl-5-phenyloxazolyl)benzene, as scintillator, 288, 289
Biuret, 176
Blister formation, MDCK cells, 553
BME, see Eagle's basal medium
B o m b y x rnori, media, 457
Bone marrow cell
cloning, 163
human, growth factor, 163
mouse, growth factor, 163
Bovine herpes virus, 28
Bovine serum
fetal
freezing, 31
preservation, 31
growth promotion, 20, 21
mycoplasmal testing, 22-27
sterility, 19-21
tests, 20, 21
toxicity, 20
viral contamination, 28, 29
virus testing, 28, 29
Brain cell
cloning, 163
dispersion, 123
Brain tumor cell, growth factor, 163

Brd Urd-resistant cell line, 376
5-Bromodeoxyuridine
as mutagen, 316
resistant strains, 314
Bryan high-titer strain, Rous sarcoma virus,
380
Buffer
digestion, 542, 543
effect on culture, 70, 72, 73
in media, 199, 200
C
Cabbage looper, media, 457
Calciferol, see Vitamin D
Calcium
deprivation, synchrony, 261,262
in media, 68
Calf anterior pituitary cell, suspension culture, 207, 208
Calf serum, 382
Callus culture, 478
establishment, 479, 480
cAMP, see Cyclic adenosine monophosphate
Canine thymus cell, 414
Capillary, artificial, cell culture, 178-184
Capillary culture unit, 178-184
construction, 179-181
histology, 183
occlusion, 182
oxygenation, 183
perfusion, 181-183
seeding, 182
sterilization, 182
use, 183, 184
Carbenicillin, in media, 112, 114
Carbohydrate, in media, 66
Carbon dioxide
contamination, 202
on virus yield, 401
in media, 201,202
Carbon dioxide-free media, 86, 87
Carbon monoxide, as contaminant, 202
Carcinoembryonic antigen, 184
Carcinogen, see also specific substances
chemical, transformation, 370
solution, 298, 299
Carcinogenicity
chemical, 296-302
616
SUBJECT INDEX
Carcinogenicity ( c o n t ' d . )
testing, 301
survival curve, 301
transformant induction, 301
Carcinoma
adrenal cortex, 376
cervical, 376
colon, 184
embryonal
effect of hormone, 105
mouse, 97
Lewis lung, 376
lung, 376
nasopharyngeal, 376
transplantable, 276, 277
Carrot, from protoplast, 367
Carrot x petunia, 365
Cartilage, 560
Cat kidney cell, media for, 58
C-banding, s e e Centromere banding
Ce cell, effect of hormone, 103
Cell, s e e a l s o specific types
anchorage dependence, 44, 45, 48, 81,
133, 137, 152
characteristics, monitoring, 164-178
differentiated, 80, 263-278
culture, 264
media for, 58
DNA content, 148
enucleated, 359
hard-to-grow, media for, 56
intended use, 81
lifespan, 134
metaphase, s e e Metaphase cell
mitotic, s e e Mitotic cell
nontransformed, 45
media for, 56, 58
normal, 45
media for, 85, 90
protein content, 145
pulse-labeled, 244
transformed, 80, 81, 134
effect of lectins, 368
properties, 368
virus-infected
cloning, 155
microcarrier culture, 191, 192
Cell adhesion, s e e Adhesion, cellular
Cell classification, criteria, 44
Cell counting, 143
data expression, 150
direct, 143
liver cells, 543, 544
microcarrier, 188, 189
Cell culture, s e e a l s o Media; specific cell
type
characterization, 164-178
cost, 195,204-206
defined, 263
equipment, s e e Laboratory equipment
effect of buffer, 72, 73
of carbon dioxide, 72, 135
of humidity, 73, 74, 136
of light, 136
of osmolality, 73, 136, 137
of oxygen, 74, 136
of pH, 72, 135
of temperature, 55, 135, 450
facility, s e e Laboratory facility
large-scale, s e e Large-scale culture
management, 116-119
materials, sources, 196
monolayer, 46
packaging, 36
preparative operations, 195-202
shipment, 29-36
regulations, 34-36
sources, 119
sterility testing, 21
Cell cycle
analysis, 233-247
DNA content, 243-246
parameters, 241
Cell cycle phase, 248-262
quantitation, 241
Cell density, 47, 218, 219
measurement, 225
Cell dispersion, 239
fluorescence, 246, 247
mechanical, 130, 131
protocol, 246, 247
shearing, 130, 131
Cell disruption, 223
Cell dissociation
with collagenase, 128, 129
after culture, 129, 130
epithelial cell, 129
of explants, 129, 130
muscle, 513-516
by organ perfusion, 128, 129
with pronase, 129
sorting, 239
SUBJECT INDEX
Cell filter, 514
Cell fixation, 140
Cell fractionation, 222
Cell fusion, 324, 345-359
polyethylene glycol, 353, 356-359
protocol, 353, 354
Sendai virus, 353-356
suppression of function, 347
virus detection, 348, 349
Cell growth
analysis, 223
curves, 75, 76, 215
data evaluation, 150
interactions, 78, 79
kinetics, 133, 214-217, 241, 245
large-scale, 13
measurements, 142-150
capillary culture unit, 183
chemical, 144-149
electronic, 149
hematocrit, 149
microscopy, 149
visual, 142-144
on microcarrier, 187-190
parameters, 214-217
plateau region, 75, 76
population-dependent, 47
quantitation, 141-152
requirements, 51-79
synchronous, 218
Cell harvest, 221-229
Cell hybrid, s e e Hybrid cell
Cell identification, 164-178
karyotyping, 323, 324
Cell line, s e e a l s o specific types
cloning, 155
heteroploid, 323
mammalian, 371
invertebrates, 450-466
mixed, 117
nontumorigenic, 378
permanent, 45
quality control, 439
sources, 132, 439-444
stable, 439-444
suspension culture, 202-206
tumorigenic, 371, 376-378
human, 376
mouse, 377
various species, 377
Cell monitoring, s e e Monitoring
617
Cell number, 218

Cell preparation, 12
Cell preservation, 29-36
Cell recovery, 212
Cell repository material, 12, 440-442
Cell shearing, 223
Cell size, 238
Cell sorting, 239
efficiency, 240
Cell staining, s e e Staining
Cell storage, 12, 29-36,220, 221
Cell surface protein, 267
Cell synchronization, s e e Synchrony
Cell transfer, precautions, 50
Cell transformation, 368-370, s e e a l s o Cell,
transformed
adenovirus, 370
human, 426
chemical carcinogen, 370
criteria, 368
defined, 368
DNA tumor virus, 370
helper virus, 369
malignant, 368
murine lymphoma virus, 369
murine sarcoma virus, 369
polyoma virus, 370
RNA tumor virus, 369
Rous sarcoma virus, 369, 370
SV40, 370
high, 369, 370
low, 369, 370
temperature-sensitive mutant, 370
Cell viability, 131, 141-152, 218
assay, 150
measurement, 150-152
Cell yield, 131
Centrifugal elutriation, 258
Centrifugation, 222
dye-density equilibrium, 408,410
Centrifuge, safety equipment, 40
Centromere banding, 325
Cephalexin, in media, 116
Cephalosporin, in media, 116
Cephalothin, in media, 112, 114, 116
Cervical adenocarcinoma, monolayer culture, 133
Cervical carcinoma, 376
Cesium chloride, DNA extraction, 408, 409
CF-3 cell protein content, 145
Ce glial cell, 96
618
SUBJECT INDEX
Chang cell, protein content, 145

Chelating agent, in media, 70
Chemical carcinogenicity, see Carcinogenicity, chemical
Chemography, 291
Chemostat, mass culture, 205
Chicken
embryo cell
cloning medium, 384
media for, 56, 58
primary culture, 385, 386
secondary culture, 388-390
tissue dispersion, 126
embryo extract, preparation, 526, 527
embryo fibroblast, 59
freezing, 392, 393
preparation, 381-384
storage, 392, 393
embryo heart fibroblast, media for, 90
fibroblast, microcarrier culture, 189
nontransformed cell, media for, 56
Chicken red cell tobacco mesophyll, 365
Chicken serum, 382
in media, 139
Chick helper factor, 387, 388
Chikugunya virus, interferon-producing,
293,294
Chinese hamster cell line, tumorigenicity
testing, 374, 375
Chinese hamster karyotype, 340-342
Chinese hamster ovary cell, 53-55, 86, 377
cell cycle, 242
Colcemid treatment, 255, 333
growth, for mutagenesis, 310
media for, 58, 89, 310
microcarrier culture, 189-191
mutagenesis, 309
strains, 310
Chloramphenicol, in media, 112, 114, 116
Chloride, in media, 68
7-Chlortetracycline, in media, 112, 114
CHO cell, see Chinese hamster ovary cell
Cholesterol, in media, 54, 68
Choline, in media, 54, 68
Chondrocyte
from cartilage, 560
characterization, 563
large-scale preparation, 560-565
media, 562
source, 561
staining, 563
Chondroitin sulfate, 268

chondrocyte culture, 563
Choriocarcinoma, 184
Chorionic gonadotropin, see Gonadotropin,
chorionic
C h o r i s t o n e u r a f u m i f e r a n a , media, 457
Chromium, in media, 69
Chromomycin Aa
fluorescence spectrum, 238
fluorescent dye, 236-238, 240, 242
preparation, 247
Chromosomal localization, 348
SV40, 349
Chromosomal rearrangement, 323
Chromosomal variation, 322, 323
Chromosome
examination, 170-174
identification, 322
loss, 322, 348
microscopy, 336
photography, 336-339
slide preparation, 326-334
Chromosome banding, 324, 325
Chromosome deletion, 323
Chromosome inversion, 323
Chromosome marker, 174
Chromosome number, 323
Chromosome staining, 324-326, 334-336
Chromosome translocation, 323
Chymotrypsin, tissue culture, 125
Cleaning facility, 3-7
Clonal culture, media for, 56, 58, 59
Clonal growth, 46, 83
media for, 56-59, 86, 87, 89, 90
Clone 3T3-L1, effect of hormone, 105, 106
Cloning, 46, 152-164
anchorage-dependent cells, 152
protocol, 158, 159
anchorage-independent, 153
protocol, 159-161
capillary technique, 153
efficiency, 509
HTC cells, 547-549
layer preparation, 161-164
media, 155, 384, see also specific types
melanoma cell, 566
morphology, 139
in multiwell plastic trays, 321, 322
mutant, 321, see also Mutant
neuronal cell, 588,589
plant cells, 482-484
SUBJECT INDEX
protocol, 154, 155
Rous sarcoma virus, 390-392
Cloth, for plating, 320
Cloudman $91 melanoma, 564
CMRL 1066 medium, 57, 88, 155
composition, 62-70
CMRL 1415 medium, 58, 87, 90, 91
composition, 62-70
CMRL 1415-ATMmedium, 58, 87
CMRL 1969 medium, 59, 87, 91
composition, 62-70
Cobalt, in media, 69
Codling moth, media, 457
Coenzyme A, in media, 64
Colcemid
mitotic cells, 255
mitotic spindle inhibitor, 327
slide preparation, 329, 331,343
synchrony, 261
Colchicine
mitotic inhibitor, 256, 327
resistance, 309, 314
Colchium mesophyU soybean culture, 365
Collagen
as culture substrate, 266
floating substrates, 266
purification, 270, 526
as raft, 271
cell separation, 275
preparation, 270
treatment of culture dish, 525, 526
Coilagenase
cell dissociation, 128, 129
crude, 126
in media, 139
tissue culture, 125
types, 126
Colon carcinoma, 184
Colony stimulating activity, 162, 163
Complementation, genetic, 347
protoplast fusion, 367
Concanavallin A, as mitogen, 487, 492
Conditioned medium, 47, 153, 524, 581
Conditioning, mechanism, 47
Confluence, 133
Connective tissue, transformed, 376
Contact feeding, 315
Contact inhibition, 134
Contaminant, see Contamination; specific
type
619
Contamination
bacterial, 18, 19
tests, 20
carry-over, 83
culture, 201
detection, 18-29, 116
evaluation, large-scale, 219, 220
by extraneous cells, 165
fungal, 18, 19
test, 20
microbial, 110
mycoplasmal, 18, 21-28,219, 220, 439
testing, 21-28
toxic compound, 5
viral, 28, 29
Continuous culture, 216
Continuous growth, cost, 206
Coon's medium F12, 86
Copper, in media, 52, 69
Corneal cell, rabbit, 414
Corneal tissue culture, fish, 473
Corn mesophyll x soybean culture, 365
Cotton bollworm, media, 457
Cross feeding, 315
Cross-reacting protein, 177
Cross reaction, immunological, 178
Cyrogenic storage, 31
Cryogenic supplies, 17
Cryopreservation, HTC cells, 551,552
Cryoprotective agent, 12
Crystal violet, microcarrier culture, 189
Culex quinquefasciatus, suspension culture,
454
Culex tritaeniorhynchus, suspension culture,
454
Culiseta inornata, media, 457
Culture flask, 13
Culture media, see Media
Culture plate, see also Plating
collagen-coated, 158
fibrin-coated, 158
microtest, 156
Culture system, 12-15, see also specific
system
Culture vessel, 12-15
Cyclic adenosine monophosphate
measurement, MDCK, 557-559
melanoma cell cycle, 567
Cycloheximide resistant strains, 314
Cyclostome
balanced salt solution osmolarity.adjustment, 469
620
SUBJECT INDEX
Cyclostome (cont'd.)
Cynthia moth, media, 457
Cysteine, in media, 53, 62, 87
Cystine, in media, 53, 62
Cytogenetics, 322
Cytokinin, in media, 478
Cytological hybridization, see Hybridization, cytological
Cytopathogenicity, mixed culture, 421-423
Cytosine arabinoside, as mutagen, 316
D
Darkroom
for autoradiography, 280, 281
safelight, 281
N - D e a c e t y l - N - m e t h y l c o l c h i c i n e , metaphase
preparations, 170
DEAE-cellulose
as microcarrier, 186
preparation, 186
DEAE-Sephadex, as microcarrier, 186, 209
Decontamination, 39
6-Demethyl-7-chlortetracyline, in media,
112, 114
Deoxyadenosine, in media, 66
Deoxycytidin, in media, 67
Deoxyguano[ine, in media, 67
Deoxyribonuclease, tissue culture, 125, 126
Deoxyribonucleic acid
cell content, 148
in cell cycle, 243-246
determination, protocol, 146-149
extraction, 406-409
Hirt procedure, 406--409
injection into cells, 412
isolation, from human adenovirus, 426
linear duplexes, 411
relaxed circles, 411
separation, 411
supercoiled, 411
thymidine incorporation, 251,252
viral, see Viral DNA; specific types
Deoxyribose, in media, 66
Dermatan sulfate, 268
Detergent, in media, 70
Diarrhea virus, 28
Dibutyl cAMP, effect on growth, 245
2,4-Dichlorophenoxyacetic acid, as auxin,
478
Digestion buffer, liver cells, 542, 543
Dihydrostreptomycin, in media, 112, 114,

116
l lfl,20a-Dihydroxyprogesterone, synthesis,
571
6,8-Dihydroxypurine, in media, 67
Dilution plating, techniques, 156-158
Dimethylsulfoxide, 12, 32
effect on mouse erythroleukemia cells,
506
in freezing, 30, 448
2,5-Diphenyloxazole, as scintillator, 288
Diphtheria toxin, resistant strains, 314
Disaggregation, with enzymes, 125, 126
Dispersant solution, 124
Dissociation, enzymic, effect on viability,
138
DME, see Dulbecco's modified Eagle's
medium
DME/FI2 mixed medium, 103, 104, 106
DM medium, 59
120, 57, 88
145, 57, 62-70
160, 57, 88
DNA, see Deoxyribonucleic acid
DNA adenovirus, 433,434
DNA animal virus, 404
DNA-dextran, permissive tests, 412
DNA polyoma, extraction, 406, 407
DNA restriction fragment assay, 434, 435
DNase, see Deoxyribonuclease
DNA simian virus
assay, 411,412
isolation, 405
quantitation, 411,412
radiolabeling, 410, 411
DNA staining
fluorescent, 246
procedure, 240
DNA tumor virus, 425
transformation, 370
Dopa-oxidase, melanocytes, 564
Doxycycline, in media, 112, 114
Drosophila immigrans, media, 464
Drosophila melanogaster
karyotype, 342
media, 464
Drosophila virilis, media, 464
Drug-resistance mutant, see Mutant, drugresistance
Dual-rotary circumfusion, 14
SUBJECT INDEX
621
Eosin, autoradiography poststain, 287

Epidermal growth factor, 78, 100, 101, 103,
109
[125I]-labeled, binding to HeLa cells, 107109
as mitogen, 109
Epithelial cell
culture, 263
dissociation, 129
enrichment, 277
E
freezing, 31
intestinal, protein content, 145
Eagle's basal medium, 84, 85, 91
rat, media for, 57
suspension, 270
composition, 62-70
Eagle's minimum essential medium, 54, 84, Epithelial-mesenchymalinteraction, 265
85, 91, 155, 405
in adult, 266
composition, 62-70
Epithelium, feeder layer, 265
HTC cells, 545, 546
Epstein-Barr virus, 37, 42
for lymphocytes, 486
Erythrocyte
for macrophages, 499
contamination, 492
myoblasts, 522, 523
freezing, 31
Earle's balanced salt solution, 120, 121, 142
lymphocyte separation, 490, 491
Erythroleukemia cell, s e e Mouse erythEastman Kodak AR-10 film, 287
roleukemia cell
Eastman Kodak NTB nuclear emulsion,
250, 279, 280, 283
Erythromycin, in media, 112, 114
EDTA, s e e Ethylenediaminetetraacetic acid Ethanol, in media, 70
Ethidium bromide
Elastase, tissue culture, 125
Elasmobranch
DNA extraction, 408
balanced salt solution osmolarity adjuststaining, 412
Ethylenediaminetetraacetic acid, in media,
ment, 469
70
Electronic cell counter, 149
Ethyl methanesulfonate, as mutagen, 312
Electrophoresis, cell monitoring, 167, 168
Ethyl nitrosourea, as mutagen, 312
Explant
Embryonic cell, s e e a l s o specific types
plant cell, 479
cloning, 155
skin, see Skin explant
Emetine, resistant strains, 314
EM-1 medium, 105
Emulsion
F
choice, 279, 280
dipping, 284, 285
Facility, s e e Laboratory facility
drying, 284, 285
Familial hypercholesterolemia, 444
nuclear, 279, 280
Fatty acid, polyunsaturated, in media, 54, 55
preparation, 283, 284
Feeder cell, Japanese quail, 392
Feeder layer
pressure-sensitive, 291
Endocrine cell, differentiation, 269, 270
epithelium, 265
Endothelial cell, vascular, monolayer, 133
human fibroblast, 358
Enzyme, s e e a l s o specific substance
mesenchymal cell, 270
in cloning, 125
preparation, 278
extract preparation, 167
Fermentor, mass culture, 204, 205
stabilization, 126
Fetal calf serum, 95, 213, s e e a l s o Calf
for tissue dispersion, 125
serum
Enzymic dissociation, effect on viability, 138 Fetuin, in media, 88, 105
Duck embryo fibroblast, media for, 59
Dulbecco's modified Eagle's medium, 58,
86, 87, 91, 95, 97,100, 101,213,423, 499
composition, 62-70
Dulbecco's phosphate-buffered salt solution, 120, 121, 142
Dye-density equilibrium centrifugation, 408,
410
Dye exclusion test, for cell viability, 151
622
SUBJECT INDEX
Fixative solution, metaphase preparation,

Fibroblast
171
chicken embryo, 55, 76, 84
Flow cytometer, 235
freezing, 392, 393
suppliers, 234
heart, media for, 90
Flow cytometry, 233-247
media for, 59
instruments, 233, 234
principles, 233-239
Flow microfluorometry, 149, 252
storage, 392, 393
Flow sorter
duck embryo, media for, 59
schematic, 234
explant, 446--448
suppliers, 23,~
fetal lung, 50
Fluorescein isothiocyanate, 176
freezing, 448, 449
Fluorescence, cell, 233, 234, 236
human, 55, 76
Fluorescent antibody labeling, 175, 176
DNA content, 148
embryo lung, microcarrier culture, 190 Fluorescent staining, s e e Staining,
fluorescent
as feeder cell, 358
Fluorine, in media, 69
foreskin, 51
5-Fluorodeoxyuridine, as mutagen, 316, 317
Folic acid, in media, 54, 64
interferon production, 295
Folic acid deficiency, 53
isolation, 444-450
Folinic acid, in media, 54, 64
media for, 56, 58, 59
Folin reagent, melanoma cell, 568
monolayer culture, 132
Follicle-stimulating hormone, in media, 102
preservation, 31
Follicular cell, rat, effect of hormone, 100,
propagation, 444
101
mouse
Foreskin fibroblast, 51
effect of hormone, 101, 102
Forest tent caterpillar, media, 457
embryo, 90
Freezing, s e e a l s o specific cell type
media for, 58, 59
adjuvants, 30
cell damage, 30
Swiss 3T3, 105, 106
cooling rate, 30, 31
storage, 448, 449
conditions, 30
Fibroblast growth factor, 78, 163
method, 33
in media, 99
protocol, 220, 221
Fibronectin, 267
recovery from, 448
Fibrosarcoma, 376
storage, 448
Ficoll gradient, 257, 258
Freund's adjuvant, 175
Film, autoradiography, 287
Friend cell, s e e Mouse erythroleukemia cell
Filter
Fruit fly
bacteriological, 8
media, 464
cell, 514
high-etticiency particle air, 10, 40
F12/DME mixed medium, 103, 104, 106
sterilizing, 75
F12K medium, 59, 87, 155
Filtration system, supplier, 16, 17
F12 medium, s e e Ham's medium FI2
Fischer's medium, 56, 86
F12M medium, 155
Fish, blood collection, 472
Fuchsin, basic, autoradiography prestain,
Fish cell, 467
286
Fish corneal tissue culture, 473
Fume hood, 5, 10, 40
Fish fin, culture, 473
Fungal contamination, s e e Contamination,
Fish leukocyte, 472
fungal
5A medium, s e e McCoy's medium 5A
Fungizone, in media, 523
Fixation, monolayer cultures, 140
SUBJECT INDEX
G
Galactose, in media, 53, 66

Ganglion cell, 581
Gas contaminant, in bottled gas, 202
G-banding, see Giemsa banding
Gelatin
in media, 525, 526
surface treatment, 577
Genetic complementation, see Complementation, genetic
Genetic disorder, see specific type
Gentamycin, in media,4112, 114, 116, 198
Gey's balanced salt solution, 120, 121, 142
GHAT medium, 352
GH cell, 96, 528, 529
characteristics, 529, 532-534
culture, 530
growth rates, 532
hormones, effect of, 98-100
hormone-producing, 532-534
karyotype, 532
media, 530
in somatic cell hybids, 534
stability, 532
subculture, 532
Giemsa banding, 324, 325, 334-336
method, 326
protocol, 326
Giemsa buffer, 172, 173
Giemsa stain, 172, 173, 285-287, 324, 333,
422
Gimmel factor, 96, 101, 106
Glassware
drying, 6
liver cells, 541
sterilization, 4-7, 10, 11
suppliers, 16
Glioma, 377
Glioma Co cell, rat, 97
effect of hormone, 103-105
media for, 57
Gluconolactone, in media, 66
Glucosamine, in media, 66
Glucose, in media, 53, 66, 198
Glucose-6-phosphate dehydrogenase
assay, 167, 168
cell monitoring, 167, 168
623
Glucose-6-phosphate dehydrogenase isoenzyme, 167, 168

Glucose-6-phosphate dehydrogenase stain,
168, 169
Glucuronate, in media, 66
Glucuronolactone, in media, 66
Glutamate, in media, 63
Glutamine, in media, 53, 62, 89
Glutathione, in media, 63
Glycerol, 12
in freezing, 30, 31
Glycine, in media, 53, 63
Glycosaminoglycan, 267, 268
differentiation, 268, 269
preparation, 274
G n o r i m o s c h e m a o p e r c u l e l l a , media, 462
Gonadotropin, chorionic, 184
Goodwin IPL-52 medium, 460
Grace medium, 457-460
Gradient centrifugation, 256-258
Granulocyte
DNA content, 148
growth factor, 163
Growth factor, for cell types, 163
Growth hormone
in media, 107
pituitary hormone cell, 527
production, 183
Growth hormone cell, see GH cell
Growth kinetics, 133, 214-217, 241, 245
Growth parameter, 214-217
Growth phase, 214
Growth promotion test, 20, 21
Guanine, in media, 67
Gypsy moth, media, 457
H
Halle's medium SM-20, 59, 90
Ham's medium F10, 56, 86, 155, 393,522
composition, 62-70
Ham's medium F12, 53, 57, 58, 86, 89, 91,
95, 97, 99, 100, 106, 107, 130, 155, 524
Composition, 62-70
Hanks' balanced salt solution, 120, 121,130,
142
Hansen medium S-301,465, 466
Harding-Passey cell line, 564
Harvesting, see specific type
624
SUBJECT INDEX
HAT medium, 351-353

Heart, dispersion, 123
Heart cell, media for, 59, 90
HeLa cell, 77, 84, 96
chromosome, 174
contamination, 439
media for, 56, 84, 85
microcarrier culture, 189-191
monitoring, 166
monolayer culture, 132, 133
serum-free, 99
HeLa culture x tobacco mesophyll, 365
HeLa-S cell, effect of hormone, 107-109
H e l i o t h i s z e a , media, 456, 457, 460
Helper virus
mouse sarcoma virus, 413
Rous sarcoma virus, 380
transformation, 369
Hematocrit, 149
Hematopoietic precursor cell, growth factor, 163
Hematoxylin, autoradiography poststain,
287
Hemocytometer counting, 188
protocol, 143
Hemoglobin
gene expressions, 507
inducibility, 509, 510
spectrophotometry, 511
staining, 510
HEPA filter, see Filter, high-efficiency particle air
Heparin, 268
Heparin sulfate, 268
Hepatic cell, bilirubin conjugation, 184
Hepatocyte, media, 53
Hepatoma
large-scale growth, 211
rat, 544
Hepatoma tissue culture cell, 544-552
cloning, 547-549
cryopreservation, 551,552
effect of steroids, 544, 545
media, 545,546
composition, 546
propagation, 545
routine culture, 546, 547
serum-free growth, 549, 550
storage, 5 5 1 , 5 5 2
HEPES buffer, in media, 70, 72, 73
H e r p e s v i r u s s i m i a e , 37, 43
Heterokaryon, 345
Heteroploid cell line, see Cell line,
heteroploid
Higuchi's medium, 57
composition, 62-70
Hink medium, 461
Hirt procedure, extraction, 406-409
Histidine, in media, 53, 62
Histidinol, resistant strains, 314
Histone, 49
33258 H/Sechst dye, as sensitizer, 317
Hormone, see also specific substance
effect on embryonal carcinoma cells, 105
on GHa cells, 98-100
on growth, 94-109
on HeLa cells, 100
on HeLa-S cells, 107-109
on melanoma ceils, 102
on mouse fibroblasts, 101, 102
Swiss 3T3, 105, 106
on neuroblastoma, 97, 98
on rabbit intimal cells, 106, 107
on rat follicular cells, 100, 101
on rat glial C6 cells, 103-105
on testicular cells, 103
in serum, 78, 83
Hormone growth factor, 78
Horse serum, 95, 213
in media, 523, 530
HTC cell, see Hepatoma tissue culture cell
Human
adenovirus, see Adenovirus, human
cell
media for, 56, 57
nontransformed, media for, 56
tumorigenic, 376
epidermal keratinocyte, media for, 90
embryonic lung fibroblast, microcarrier
culture, 190
fetal lung cell, protein content, 145
fibroblast
DNA content, 148
as feeder cell, 358
foreskin, microcarrier culture, 190
isolation, 444--450
media for, 56, 58, 59
SUBJECT INDEX
preservation, 31
propagation, 444
karyotype, 340-342
leukemia cell, protein content, 145
leukocyte, cloning, 163
lymphoblast, slide preparation, 328-330
lymphocyte, 486-494
culture conditions, 491,492
DNA content, 148
marrow cell, growth factor, 163
RSTC-2 cell, interferon, 293
Human Genetic Mutant Cell Repository,
441,442
Humidity, effect on culture, 73, 74, 136
Hyaluronate, 268
aggregation, 268
Hyaluronidase, tissue culture, 125
Hybrid
mouse human, 347, 357
somatic, 359, 360
Hybrid cell, 324, 345
Hybrid fusion, plants, 364-366
Hybridization, cytological, 326
Hybrid plant, cell fusion, 364-366
Hybrid selection, procedures, 351-353
Hydrocortisone, in media, 100, 101, 109
Hydroxymethyl cellulose, 580
20ct-Hydroxyprogesterone, synthesis, 571
Hydroxyproline, in media, 53, 63
5-Hydroxytetracycline, in media, 112, 114
Hydroxyurea
synchrony, 259-261
Hygienic practice, 39
Hypercholesterolemia, familial, 444
Hypothalamic peptide, regulatory effects,
527
Hypoxanthine, in media, 54, 67
Hypoxanthine-guaninephosphoribosyl
transferase, hybrid selection, 352
I
IMEM-ZO medium, 58, 87
composition, 62-70
Immunological cell monitoring, 174-178
Immunological cross reaction, 178
IMR-91 cell, protein content, 145
Incubation, mutant isolation, 315, 316
Incubation temperature
poikilotherm vertebrate cells, 470-472
625
invertebrate cell line, 450, 451

Incubation vessel, large-scale, 213, 214
Incubator, 11
supplier, 16
Indicator, in media, 70
3-Indoleacetic acid, as auxin, 478
Induction, 292-296
mutant, s e e Mutant
transformant, 301
Infection
abortive, 404
human, 37, 38
laboratory-acquired, 38
Infectious bovine rhinotracheitis virus, 28
Inhibition, density-dependent, 47, 48
Inoculation
intracerebral, Chikugunya virus, 293
mutant isolation, 315,316
Inositol, in media, 54, 68
Insect cell line, media, 455-466
Insulin
effect on melanoma cells, 567
in media, 70, 78, 99-107, 109
Interferon
assay, 296
formation in microcarrier culture, 192
inducers, 293,294, s e e a l s o specific types
induction, 292-296
production, 292-296
priming, 296
from RSTC-2 cells, 293
superinduction, 295
Intestinal epithelial cell, protein content, 145
Intimal cell
rabbit
aortic, 96, 97
Invertebrate cell, 450--466
culture methods, 451,452
effect of atmosphere, 450, 451
of temperature, 450, 451
media, 454--466
storage, 453
subculture, 451,452
Iodine, in media, 69
Iron, in media, 52, 69
Isoleucine
deprivation, synchrony, 258, 259
in media, 53, 62
N6-Isopentenyladenine, as cytokinin, 478
626
SUBJECT INDEX
Isozyme
cell monitoring, 166-170
extract preparation, 167
staining, 168-170
J
Japanese quail
feeder cell, 392
Rous sarcoma virus, 395
Jensen sarcoma cell, protein content, 145
K
Kanamycin, in media, 112, 114, 116
Karyotyping, 170-174, 220, 322-344
characteristics, 342
CHO cells, 340-342
Drosophila melanogaster, 342
GH cells, 532
human, 340-342
modifications, 333, 334
mouse, 341,342
mutants, 309
problems in, 330
rat, 342
standardization, 339
Y1 adrenal cell line, 571
Katsuta medium (DM, 57, 62-70, 88
KB cell line
infection, 425; 431
Keratinocyte, human epidermal, media for,
9O
A4-3-Keto-C~z-steroid, synthesis, 571
Kidney, dispersion, 123
Kidney cell, 377
cat, media for, 58
monkey
dissociation, 127
SV40, 404
Kidney epithelial cell, 552-560
Kinetics, of growth, 133,214-217, 241, 245
Kinetin, as cytokinin, 478
K12 cell, synchronization, 252-256
L
Laboratory equipment, suppliers, 15-17,
196
Laboratory facility
cleaning, 3-7
design, 4-7
sterilization, 3-7
storage, 3, 7-9
Laboratory operation, 3
Laboratory safety, 36--43
Lactate dehydrogenase
assay, 169, 170
cell monitoring, 168-170
isoenzyme, 169, 170
Lactate dehydrogenase stain, 169
Lamelopodia, transformed cells, 368
Laminar flow system, 10, 40, 154
Large external transformation substance,
see Fibronectin
Large-scale culture, 211-221
adenoviruses, 431-433
chondrocytes, 560-564
cost, 205
equipment, 221-229
growth methods, 194-210
microcarrier, 193
Large-scale growth, see Large-scale culture
L cell
mouse, 77, 84
cloning, 163
medium, 84, 85
Leafhopper, media, 462
Leaf mesophyll cell, 360
Lecithin, in media, 68, 500
Lectin
cell transformation, 368
Leibovitz's medium L-15, 56, 87, 91
composition, 62-70, 463
Lesch-Nyhan syndrome fibroblast, 444
Leucine, in media, 53, 62
Leukemia antigen
murine thymus, 219
Leukemia cell
granuloeytic, DNA content, 148
SUBJECT INDEX
human, protein content, 145
media for, 56
Leukemia virus, see Murine leukemia virus
Leukocyte
cloning, 155
human, 163
isolation, 487-489
poikilotherm, 472
Leukopheresis, 493, 494
Lewis lung carcinoma, 376
L-5 cell, protein content, 145
Light, effect on growth, 136
Light scattering, 238
Lincomycin, in media, 112, 114
Linoleic acid, in media, 54, 68, 90
Lipid, in media, 68
a-Lipoic acid, in media, 54, 64
Lipopolysaccharide, as mitogen, 351
Liposome, 54
Liquid nitrogen
in freezing, 30
storage, 33
Liver
dispersion, 123
rat, perfusion, 128, 129
Liver cell
buffer solutions, 540
cloning, 163
counting, 543, 544
differentiation, 541
digestion buffer, 542, 543
establishment of cell lines, 536-544
isolation, 543
media for, 59, 537, 542,543
plating, 538,539, 543
rat
adult, 539
fetal, 538
media for, 90
Lowry protein assay, 144
interference, 146
L. pomonella, media, 457
LS cell, 89
L251A cell line, 219
Lung, dispersion, 123
Lung carcinoma, 376
Lung cell
human embryo, protein content, 145
fibroblast
fetal, 50
627
human embryo, microcarrier culture,

190
mink, 414
Luteinizing hormone, in media, 106
Luteinizing hormone-releasing hormone,
102
Lymantria dispar, media, 457
Lymph node, lymphocyte preparation, 124
Lymphoblast
human, slide preparation, 328-330
media for, 89
Lymphoblastoid cell, 376
media for, 56
routine, 218
EL4, 223
Lymphocyte
activated, culture conditions, 492
activation, 487
bone marrow-derived, see B cell
cloning, 155
contaminants, 492, 493
human, 486-494
culture conditions, 491,492
DNA content, 148
platelet removal, 490
from leukocytes, 489, 490
preservation, 31
quantitation, 491
separation from erythrocytes, 490, 491
suspensions, 124
thymus-derived, see T cell
yield, from blood, 491
Lymphoid cell, biohazard, 37, 42
Lymphoma, 376
mouse $49, 238
cell cycle phase, 242
DNA content, 244
Lymphosarcoma, mouse MB (T-86157)
Lysine, in media, 53, 62
Lysolecithin, cell fusion, 345
Lysozome transfer, 359
M
MAB 87/3 medium, 57, 88
composition, 62-70
628
SUBJECT INDEX
Macrophage, 492, 494-506

activation, 494
adherence, 497, 498
alveolar, 496, 497
centrifugation, 497
continuous cell lines, 503,504
culture, 506
protocol, 504--506
detachment, 501
effect of serum, 501
in exudates, 502
fixing, 502
growth ~actor, 163
histochemistry, 498
identification, 498
long-term culture, 502
lysis, 502, 503
marker, 499
media, 499-502
morphology, 498, 502
peritoneal, 495, 496
phagocytic uptake, 498, 499
purification, 497, 498
separation, 497,498
sources, 495--497
staining, 502
viability, 501,502
Maden Darby canine kidney cell line, 552560
blister formation, 553
growth, 554
kidney function studies, 553
maintenance, 554
media, 554
as model of distal tubule, 559, 560
response to antidiuretic hormones, 559
t~'ansport studies, 553,554
Maize x oat root, 365
Magnesium, in media, 68
M a l a c o s o m a distria, media, 457
Malignancy, cell fusion, 348
Malignant transformation, 296, 368
M a n d u c a s e x t a , media, 457
Manganese
as inhibitor, 99
in media, 52, 69
Mannitol, as protoplast stabilizer, 360
MAP 954/1 medium, 57, 89
Marburg virus, 37
Mass culture, see Large-scale culture
Mastocytoma, large-scale growth, 211
MB 752/1 medium, 57, 88, 89

composition, 62-70
McCoy's medium 5A, 56, 86, 91,208,423
composition, 62-70
MCDB medium, for specific species, 87
MCDB 104 medium, 59, 87, 91, 155
MCDB 105 medium 59, 76, 87
composition, 62-70
MCDB 202 medium 59, 76, 87, 91
composition, 62-70
MCDB 301 medium 58, 89
composition, 62-70
MCDB 401 medium 59, 87, 90, 91
composition, 62-70
composition, 62-70
composition, 62-70
MDCK cell, see Maden Darby canine kidney
cell line
MD 705/I medium, 57, 88
Media, 44-93, see a l s o specific type
A2 + APG, 57
alpha-MEM, 56
AMBD 647/3, 59
basal (Eagle), see Eagle's basal medium
Birch and Pirt's, 58
chemically defined, 82
choice of, 81, 90-93
CMRL 1066 (Parker), 57, 62-70, 88, 155
CMRL 1415, 58, 62-70, 87, 90, 91
CMRL 1415-ATM, 58, 87
CMRL 1969, 59, 62-70, 87, 91
commercial, 84, 196, 197
human adenovirus, 427
composition
amino acid derivatives, 63
buffers, 70
carbohydrates, 66
chelating agents, 70
detergents, 70
essential amino acids, 62
indicators, 70
inorganic ions, bulk, 68
lipids, 68
nonessential amino acids, 63
nucleic acid derivatives, 66, 67
polymers, 70
proteins, 70
solvent, 70
SUBJECT INDEX
629
trace elements, 69
MCDB 501, 59, 62-70, 87
vitamins, 64, 65
MD 705/1 (Kitos), 57, 88
conditioned, 47, 153,524, 581
minimum essential (Eagle), s e e Eagle's
minimum essential medium
costs, 204-206
Mitsuhashi and Maramorosch, s e e Mitdefined, 88-90
suhashi and Maramorosch medium
development, 91-93
mixture, 97; 99, 102-104, 106
four factors, 98
DM, 59
DM 120, 57, 88
Mohberg and Johnson's, 57
DM 145, 57, 62-70, 88
MPNL65/C, 59
DM 160, 57, 88
Nagle and Brown's, 58, 89
Dulbecco's, s e e Dulbecco's modified
NCTC-109, 57
Eagle's medium
NCTC 135 (Evans), 57, 62-70, 88
EM-I, 105
N16, 56
Fischer's, s e e Fischer's medium
199, 56, 62-70, 87, 90, 91, 129, 155
5A (McCoy), 56, 62-70, 86, 91
Parsa, 90
F10(Ham), s e e Ham's medium F10
pH, 199, 200
F12(Ham), s e e Ham's medium F12
preparation, 7-9, 94-109, 198, 199
F12 + hormones, 57
products, 225-228
F12K, 59, 87, 91, 155
quality control, 453, 454
F12M, 155
RPMI 1640, 56, 62-70, 89, 91, 155,
GHAT, 352
213
Goodwin IPL-52, s e e Goodwin IPL-52
Schneider, s e e Schneider medium
selective hybrid, 351, 352
medium
Grace, s e e Grace medium
serumless, 57, 62-70, 94-109
half-selective hybrid, 352
serum supplements, 122
Hansen S-301, s e e Hansen medium S-301
7C's, 58, 89
HAT, 351-353
Sinclair's, 58
Higuchi's, s e e Higuchi's medium
SM-20 (Halle), 59, 90
Hink, s e e Hink medium
SM-201, 59
history, 84, 85
$77, 546
sterility testing, 21
hormone-supplemented, 94-109
IMEM-ZO, 58, 62-70, 87
storage, 7-9
insect cell, 455-466
supplements, 122
L-15 (Leibovitz), 56, 62-70, 87, 91,463
suppliers, 15, 16
testing, 91
listing, 56-59
liver cells, 537
toxicity, 117
MAB 87/3 (Gorham and Waymouth), 57,
toxicity testing, 91,453,454
virus in, 225-228
62-70, 88
V-605, 56
macrophage, 499-502
maintenance, 87
White's, 56
MAP 954/1 [Waymouth (Donta)], 57, 89
Williams E, 59, 90
Williams G, 59, 62-70, 90
MB 752/1 (Waymouth), 57, 62-70, 88, 89,
90
WO~, 57
McCoy's 5A, s e e McCoy's medium 5A
Wollenberger, 90
MCDB 104, 59, 87, 91, 155
Yamane's, s e e Yamane's medium
Yunker e t a l . , s e e Yunker, Vaughn and
MCDB 105, 59, 62-70, 76, 87
MCDB 202, 59, 62-70, 76, 87, 91
Cory medium
MCDB 301, 58, 62-70, 89
Medium A2 + APG, 57
Medium alpha-MEM, 56
MCDB 401, 59, 62-70, 87, 90, 91
MCDB 411, 57, 62-70, 89
Medium AMBD 647/3, 59
630
SUBJECT INDEX
Medium CMRL 1066, 57, 88, 155

composition, 62-70
Medium CMRL 1415, 58, 87, 90, 91
composition, 62-70
Medium CMRL 1415-ATM, 58, 87
Medium CMRL 1969, 59, 87, 91
composition, 62-70
Medium DM, 59
Medium DM 120, 57, 88
Medium DM 145, 57, 88
composition, 62-70
Medium DM 160, 57, 88
Medium E (Williams), 59, 90
Medium EM-I, 105
Medium FI0 (Ham), 56, 86, 155
Medium F12 (Ham), 53, 57, 58, 86, 89, 91,
155
composition, 62-70
Medium F12K, 59, 87, 91, 155
Medium F12M, 155
Medium 5A (McCoy), 56, 86, 91
composition, 62-70
Medium G (Williams), 59, 90
composition, 62-70
Medium GHAT, 352
Medium HAT, 351-353
Medium IMEM-ZO, 58, 87
composition, 62-70
Medium IPL-52, 460
Medium L-15 (Leibovitz), 56, 91
composition, 62-70, 463
Medium MAB 87/3, 57, 88
composition, 62-70
Medium MAP 954/1, 57, 89
Medium MB 752/1, 57, 88, 89, 90
composition, 62-70
Medium MCDB, for specific species, 87
Medium MCDB 104, 59, 87, 91, 155
Medium MCDB 105, 59, 76, 87
composition, 62-70
Medium MCDB 202, 59, 76, 91
composition, 62-70
Medium MCDB 301, 58, 89
composition, 62-70
Medium MCDB 401, 59, 87, 90, 91
composition, 62-70
composition, 62-70
composition, 62-70
Medium MD 705/1, 57, 88
Medium MPNL 65/C, 59
Medium NCTC 109, 57, 89

Medium NCTC 135, 57, 88
composition, 62-70
Medium N16, 56
Medium 199 (Morgan), 56, 84, 87, 90, 91,
129, 155
composition, 62-70
Medium RPMI 1640, 56, 89, 91, 155, 213
composition, 62-70
Medium 7.C's, 58, 89
Medium SM-20 (Halle), 59
Medium SM-201, 59
Medium $77, 546
Medium V-605, 56
Medium WOs, 56
Melanin
autotoxicity, 567
in melanocytes, 564, 565
precursors, 567
Melanocyte, 564-570
Melanoma, 376, 564-570
amelanotic, 566
cloning, 566
Folin reagent, 568
media, 565
melanin formation, 565
mutants, 565, 566
pigment formation, 565
Melanoma cell line PG19, mouse,
tumorigenicity testing, 374
Melanotropin, cAMP levels, 567
MEM, s e e Eagle's minimum essential
medium
Menadione, s e e Vitamin K
Mesenchymal cell, feeder layer, 270
Mesophyll cell
leaf, 360
pea, 361
Metabolic cooperation, s e e Contact feeding
Metaphase cell
evaluation, 172-174
preparation, 170
staining, 172-174
Methane, as contaminant, 202
Methionine, in media, 53, 62
Methocel, s e e Hydroxymethyl cellulose
Methotrexate, resistance, 309, 314
Methylcellulose, in media, 70
5-Methylcytosine, in media, 67
5-Methyldeoxycytidine, in media, 67
Methyl methanesulfonate, as mutagen, 312
SUBJECT INDEX
as
mutagen, 312
Microcarrier culture, 184-194
cell counting, 188
cell yield, 210
culture initiation, 187
growth cycle, 187-190
harvesting, 190, 191
interferon formation, 192
large-scale, 193
primary cells, 207, 208
seeding, 187, 188
source, 185-187
synthesis, 185-187
toxicity, 184, 185
Microculture, neurons, 582, 583
Microscope, 11, 12
suppliers, 17
Microscopy
chromosome study, 336-339
monolayer cultures, 139, 140
Microsome preparation, 224
Microtest plate, 156
Microvillus, transformed cells, 368
Minimum essential medium, see Eagle's
minimum essential medium
Mink lung cell, 414
Mithramycin, DNA determination, 147
Mitochondria, isolation, 222
Mitogen
dosage, 492
effect on lymphocyte, 486
Mitotic cell
increase, 255
selection, 192, 193
Mitotic detachment, synchronization, 252256
Mitotic index, 233, 327
Mitotic spindle inhibitor, 327
Mitosis, peak activity, 254-256
Mitsuhashi and Maramorosch medium, 454,
462
preparation, 462
Mixed culture
cytopathogenicity, 421-423
neurons and nonneuronal ceils, 581
Mixed media, see Media, mixture
Mohberg and Johnson's medium, 57
Moloney murine leukemia virus, microcarrier culture, 192
Molybdenum, in media, 52, 69
N-Methyl-N'-nitrooN-nitrosoguanidine,
631
Monitoring
antigen-antibodyreaction, 174-178
with antiserum, 176-178
cell characteristics, 164-178
by chromsomal examination, 170-174
by isozymes, 166-170
Monkey kidney cell
dissociation, 127
Monocyte, 492, 494, 495
Monolayer culture
H T C cells,545
media for, 88, 89
microscopic, evaluation, 139
morphology, 139, 140
poikilotherm vertebrate cells, 474-476
properties, 133-135
protocol, 138, 139
substrate for attachment, 137, 138
techniques, 132-140
Moore's medium RPMI 1640, 56, 62-70, 89,
91
Morris hepatoma, 544
Moscona's saline, 138, 139
Mosquito cell
media, 457, 462
Moth cell line, 450
media, 456, 462
Mouse
cell
nontramsformed, media for, 56
tumorigenic, 376
embryo
carcinoma, 97
cooling rate, 31
fibroblast
media for, 58, 59, 90
media for, 87
crythrolcukemia cell, 346, 506-511
cloning, 509
growth conditions, 507-509
hemoglobin induction, 509, 510
media, 508
serum
requirements, 508
testing, 508, 509
fibroblast
632
SUBJECT INDEX
Mouse ( c o n t ' d . )
Swiss 3T3, effect of hormone, 105, 106
karyotype, 341, 342
L cell
cloning, 163
lymphoma cell $49
cell cycle phase, 242
DNA content, 244
lymphosarcoma MB(T-86157), suspension
culture, 203
marrow cell, growth factor, 163
melanoma cell line PG19, tumorigenicity
testing, 374
neuroblastoma cell
media for, 57
C1300, 89
nude, s e e Nude mouse
pancreatic cell, media for, 59
pituitary cell, characteristics, 531
plasmacytoma, cell fusion, 349
sarcoma virus, 413
testicular cell TM4, 96, 97
thymusless, 370
Mouse human hybrid, 347, 357
MPNL 65/C medium, 59
Mucopolysaccharide, 267, 268
Mucopolysaccharide storage disease, 444
Multicell culture, 264
Multiplication stimulating activity, 79, 96,
107, 162, 163
Murashige and Skoog inorganic salt, 480
Murine cl 1 cell, microcarrier culture, 190
Murine leukemia virus, 225, 226
assay, 412-424
enzymic, 416---421
immunological, 416-421
virological, 421-424
cells for propagation, 414
host range, 413
purification, 412-424
SL
- test, 423, 424
source, 414
XC test, 421-423
Murine lymphoblastoid cell, 218
EL4, 223
Murine lymphoma virus, transformation,
369
Murine sarcoma virus, transformation, 369
Murine thymus leukemia antigen, 219
Murine type C virus, 412

Muscle, cell dissociation, 513-516
Mutagen, s e e a l s o specific substance
care in handling, 312, 313
toxicity, 312, 313
Mutagenesis
effect of medium, 310, 311
of serum, 310
sensitizer, 317
giutant
amelanotic, isolation, 565
auxotroph, 316
cell cycle action, 316
cloning, 321
colony isolation, 321
conditionally lethal, 316-320
drug-resistance, 313-316
formation, 308
frequencies, 309
induction, 312, 313
isolation, s e e Mutant isolation
karyotyping, 309
melanoma cell, 566
multistep selections, 316
mutagen use, 312
phenotype, 313
protein synthesis, 317-319
screening, 319, 320
temperature-sensitive, 319
temperature-sensitive transformation, 370
transformation-defective, of RSV, 390
Mutant isolation, 308-322
effect of ploidy, 308
incubation, 315, 316
inoculation, 315, 316
method, 311
protocol, 311
strategy, 311
timing, 311
Mutation
auxotrophy, 309, 316-320
chemical, 312, 313
conditional, temperature-sensitive, 309
enzymes, 309
lectin resistance, 309, 314
recessive, 309
in CHO cells, 309
Mycoplasma, 21
autoradiography, 27
biochemical tests, 25
broth, 22
SUBJECT INDEX
colony, 24
DNA stain, 25
immunofluorescence, 24, 25
infection, testing, 375
isolation from medium, 227
large-scale production, 227
microscopy, 25
pseudo-colony, 24
scanning electron microscopy, 28
in serum, 23
testing, 22-27
uracil uptake, 26, 27
uridine phosphorylase, assay, 25, 26
M y c o p l a s m a hyorhinis, immunofluorescence, 24, 25
Mycoplasmal contamination, see Contamination, mycoplasmal
Mycoplasma ribosomal RNA, 27
Mycostatin, in media, 198
Myeloma, 376
hybrid, 349
Myoblast, skeletal, 377
cell source, 512
culture, 511-527
differential release, 517
differentiation from fibroblasts, 517
establishment of culture, 512
inoculum size, 518-520
media, 520-524
composition, 518-520
conditioned, 524, 525
secondary suspension, 516-518
synchrony, 511,518-520
Myo-inositol, in media, 478
N
NAD, see Nicotinamide adenine dinucleotide
NADP, in media, 64
Nagle and Brown's medium, 58, 89
1-Naphthaleneacetic acid, as auxin, 478
Nasopharyngeal carcinoma, 376
National Institute on Aging Cell Repository,
442
National Science Foundation Cell Culture
Centers, 443
NCTC 109 medium, 57, 89
NCTC 135 medium, 57, 88
composition, 62-70
Neomycin, in media, 112, 114, 116
633
Neonatal spleen cell, fusion, 351

Neoplasm, rodent, 584-590
Neoplastic transformation, 296-302
Nerve growth factor, 580
in media, 102 I
Neumann and Tytell medium, 57
composition, 62-70
Neurite, separation from soma, 302-307
Neuroblastoma cell
C1300, 585
media for, 89
growth, 97, 98
media, 57, 89
rat, 96
Neuron
action potential, 307
adrenergic, see Adrenergic neuron
co-culture with nonneuronal cells, 581
electrical stimulation, 307
plating, 306
sympathetic, see Sympathetic neuron
transmitter function, 307
Neuronal cell
cloning, 588, 589
establishment of cell line, 587, 588
isolation, 585
media, 586, 587
from mouse hypothalamus, 585
from neoplastic tissue, 587, 588
primary culture, 585
recovery, 589
from rodent neoplasm, 584-590
routine passage, 590
selection, 588, 589
serum, 587
storage, 589
transformation, 585
tumor induction, 586
Neuronal culture, 574
Newcastle disease virus, interferon producing, 294
Niacinamide, in media, 54, 65
Nickel, in media, 69
Nicotiana glauca N . langsdorffi
mesophyll, 365
Nicotiana t a b a c u m L., callus culture, 479
Nicotinamide, in media, 64
Nicotinamide adenine dinucleotide, in
media, 64
Nicotinic acid, in media, 64
634
SUBJECT INDEX
NIH-LH, in media, 101, 102, 104

NIH-LH-B9, in media, 106, 107
Nitrate, in media, 68
Nitro blue tetrazolium, isoenzyme staining,
168, 169
Nitrous oxide, as contaminant, 202
Noble agar, 23
Nonneuronal cell
co-culture with neurons, 581
removal, 580, 581
NSF, s e e National Science Foundation
N16 medium, 56
Nuclear emulsion, s e e Emulsion, nuclear
Nucleic acid derivative, in media, 67
Nucleus counting, protocol, 143, 144
Nucleus fusion, 366
Nucleus isolation, 222
Nucleus transfer, 359
Nude mouse
breeding, 372, 373
cell culture, 378, 379
cell fusion, 348
characterized, 370
maintenance, 372, 373
mass propagation, 371,375
nontumorigenic cell line, 378
source, 372, 373
transfer from, 378, 379
transformed cells, 368
tumorigenicity testing, 370-379
Nutrients, 52
low-molecular-weight, 52-55
optimum concentration, 76
Nylon cloth, for plating, 320
Nylon column filtration, 489, 490
Nystatin, in media, 112-114
0
Oat roots x maize, 365
Oleic acid, in media, 54, 68
199 medium, 56, 87, 90, 91, 129, 155
composition, 62-70
Organ culture, defined, 263
Organelle
fractionation, 222
subceliular, 223
Ornithine, in media, 63
Osmolarity
adjustments for lower vertebrates, 469
calculation, 200, 201

determination, 200, 201
Ouabain
as selective agent, 353
Ovarian cell
chinese hamster, s e e Chinese hamster
ovarian cell
rat, 96
Ovarian teratoma, 376
Ovine follicle-stimulating hormone, 102
Ovine growth hormone, in media, 103
Oxygen
contamination, 202
effect on culture, 74
requirement for growth, 136
Oxytocin, cAMP production, 559
P
Pancreas, artificial, 184
Pancreatic cell
media for, 90
mouse, media for, 59
rat, media for, 59
Pancreatic islet tissue, 126
Pancreatin, tissue culture, 125
Pantothenic acid, in media, 54, 64
Papain, tissue culture, 125
Parainfiuenza virus, 28
Paramyxovirus, interferon producing, 293
Parathyroid hormone, in media, 99
Paromomycin, in media, 113, 115, 116
Parsa medium, 90
Pea mesophyil Cell, 361
Pea mesophyll x soybean culture, 365
Pea mesophyll x V i c i a culture, 365
Peking duck, Rous sarcoma virus, 395
Penicillin G, in media, 113, 114, 116
Penicillin V, in media, 113, 114, 116
Peptide, hypothalamic, 527
Peptide growth factor, 79
Perfusion
capillary culture, unit, 181
flow rate, 182
Perfusion pump, 181
Peritoneal exudate cell
collection, 505
stimulation, 505
SUBJECT INDEX
Peritoneal macrophage, see Macrophage,
peritoneal
Petri dish, 13
Petunia
from protoplast, 367
protoplast fusion, 367
Petunia A t r o p a , 365
Petunia x carrot, 365
P e t u n i a hybrida mesophyll P. parodii
mesophyU, 365
pH
effect on culture, 72, 135
of media, 199, 200
for poikilotherm vertebrate cells, 470
Phagocyte, see also Macrophage
uptake by macrophage, 498, 499
Phase-contrast microscopy, monolayer cultures, 139
Phase fraction analysis, 241
Phenazine methosulfate, isoenzyme staining, 168, 169
Phenol red
in media, 70
pH indicator, 121, 122
Phenotype
expression, time, 313
stability, 308
Phenylalanine, in media, 53, 62
Phenylmethylsulfonyl fluoride, protease inhibitor, 273
Phosphate, in media, 68
Photography
choice of film, 338
chromosome study, 336-339
film development, 338
printing, 339
Phytohemagglutinin, as mitogen, 487, 492
Phywe flow cytometer, 236
Pigmentation, genetic marker, 565
Pineal gland
avaian, cell dissociation, 130
rat, cell dissociation, 129, 130
Pipetting technique, 38, 39
Pituitary cell
ACTH-producing, 531
corticotropin-producing, 531
/~-endorphin-producing, 531
enkephalin-producing, 531
growth hormone formation, 528
hormone-producing, 528
mouse, characteristics, 531
635
prolactin formation, 528

rat, characteristics, 529
in somatic cell hybrids, 534
source, 528
Pituitary tumor cell, 377,527-535
hormone-producing, 528
Plant
hybrid fusion, 364-366
from protoplasts, 367
Plant cell, 478-486
cloning, 482-484
explants, 479
plating, 482-484
suspension culture, 481,482
Plant protoplast, see Protoplast
Plasma cell tumor, 349
Plasmacytoma, 376
hybrid, 350
mouse, cell fusion, 349
Plasma membrane, preparation, 224
Plasticware
fibroblast culture, 449
liver cells, 541
macrophage culture, 499, 500
multiwell tray for cloning, 321, 332
suppliers, 16
Platelet
contamination, 492, 493
removal from lymphocytes, 490
Platelet growth factor, 79
Plating
adrenal tumors, 573
liver cells, 538, 539
plant cell, 482-484
Poildlotherm vertebrate cell, see Vertebrate
cell, poikilotherm
Polyamine, in media, 55
Polycarbonate bottle, monolayer culture,
137
Polyester cloth, for plating, 320, 321
Polyethylene glycol
cell fusion, 345, 353, 356--359
as fusing agent, 351
fusion size, 363
in lysozome transfer, 359
in nucleus transfer, 359
protoplast fusion, 362, 363
Polylysine, 49, 78
surface treatment, 577
vessel pretreatment, 137
636
SUBJECT INDEX
Poly-D-lysine, 49
Polymer, in media, 70
Polymyxin B, in media, 113, 115
Polyoma virus, transformation, 370
Poly IC, s e e Polyriboinosinic: polyribocytidylic acid
Polyriboinosinic:polyribocytidylic acid, interferon producing, 294, 295
Polysaccharide, in agar
acidic, 160
sulfated, 160
Population analysis, cell cycle, 243-246
Poststalning, autoradiography, 286, 287
Potassium, in media, 68
Potato tuber moth, media, 462
PPO, s e e 2,5-Diphenyloxazole
Prague strain, Rous sarcoma virus, 380, 394
Prestaining, autoradiography, 286
Priming, interferon production, 296
Pristane, priming, 350
Progesterone, in media, 102, 107
Prolactin
pituitary tumor cell, 527
production, 183
Proline, in media, 53, 63
Pronase, tissue culture, 125, 126, 129
Propidium iodide
fluorescent dye, 236-238, 242
preparation, 247
Protamine, 49
in media, 70
Protease, neutralization, 77
Protein
adhesion, s e e Adhesion protein
cell content, 145
cross-reacting, 177
determination, protocol, 144-146
in media, 70, 88
synthesis, mutant, selection, 317-319
Proteoglycan, 268
differentiation, 268, 269
preparation, 274
Protoplast
formation, 361
from pea, 361
plating, 483
separation, 361
stabilization, 360, 361
tobacco leaf, 362
from V i c i a , 361
wall regeneration, 361
Protoplast fusion, 359-367

with polyethylene glycol, 362, 363
protocol, 362-364
selection, 360, 366, 367
subsequent development, 366, 367
yield, 360
Pseudodiploid cell line, 323
Puck's balanced salt solution, 120, 121, 142
Puck's nutrient component, 522
Pulse-labeled cell, 244
Purine
auxotroph, 309
in media, 54, 66, 67
Putrescine, in media, 55, 63
Pyridoxal, in media, 65
Pyridoxal 5-phosphate, in media, 65
Pyridoxine, in media, 54, 65
Pyrimidine
auxotroph, 309
in media, 67
Pyruvate, in media, 55, 66
Q
Q-banding, s e e Quinacrine banding
Quality control
cell line, 439
media, 453,454
Quasidiploid cell line, 323
Quinacrine banding, 324, 325
Quinacrine stain, 325
R
Rabbit
antimouse antiserum, 276, 277
aortic intimal cell, 96, 97
corneal cell line, 414
embryo cell, media for, 58
intimal cell, effect of hormone, 106, 107
Rabies virus, microcarrier culture, 192
Radioimmunoassay
competitive, 420, 421
viral structural proteins, 417--421
Rapeseed, from protoplasts, 367
Rapeseed mesophyll x soybean culture, 365
Rat
embryo cell, media for, 58
epithelial cell, media for, 57
follicular cell, effect of hormone, 100, 101
glioma Ce, 97
SUBJECT INDEX
effect of hormone, 103-105
media for, 57
hepatoma, 544
karyotype, 342
liver cell
media for, 90
perfusion, 128, 129
plating, 538, 539
neuroblastoma, 96
ovarian cell, 96
pancreatic cell, media for, 59
pineal gland, cell dissociation, 129, 130
pituitary cell, 529
serum, sympathetic neurons, 580
submaxillary gimmel factor, 101
submaxillary gland, extract, 104
R-banding, see Reverse banding
Redox potential, of medium, 74
Refrigeration, 9
Renal adenocarcinoma, 376
Replicate plating
cloth, 320
procedure, 320, 321
Reptile
balanced salt solution, osmolarity adjustment, 469
incubation temperature, 471,472
media, 469
Resistance, multiple loci, 315
Resistant strain, drug concentration, 314,
315
Resting cell, human lymphocyte, 486--494
Retinoic acid, in media, 103
Reverse banding, 325
R h i p i c e p h a l u s appendiculatus, media, 463
Riboflavin, in media, 54, 65
Riboflavin 5-phosphate, in media, 65
Ribonuclease, tissue culture, 125
Ribonucleic acid, isolation, 568,569
Ribose, in media, 66
Ribostamycin, in media, 116
Ricin, resistance, 309
Rinaldini's solution, 452
Ringer's balanced salt solution, 120, 121
RNA, see Ribonucleic acid
RNase, see Ribonuclease
RNA tumor virus
transformation, 369
Rodent neoplasm, 584-590
Roller bottle, 13, 14, 397
637
for fibroblasts, 450

incubation, 185
Melinex spiral, 13
Roller bottle perfusion apparatus, 13
Roller flask, 398
Roller tube, 203
Rose chamber, 14
Rotary shaker, 14
assay, 388-390
Bryan high-titer strain, 380
cloning, 390-392
helper virus, 380
large-scale growth, 393-403
nontransforming, 403
Prague strain, 380, 394
quantitation, 388-390
Schmidt-Ruppin strain, 380, 381, 394
storage, 397
transformation, 369, 370
transformation-defective mutant, 390
yield, 400
RPMI 1640 medium, 56, 89, 91, 155, 213
composition, 62-70
RSTC-2 cell, human, interferon, 293
Sabouraud dextrose broth, 20

Safety cabinet, ventilated, 10, 40-42
Safety equipment, see specific type
Safety guidelines, 42, 43
Safety technique, 36--43
Salmine, in media, 70
Salmon cell, incubation temperature, 471
Salt solution, see also specific type
balanced, 120-122
composition, 121
plant, 480
vertebrate, 468, 469
stock, 198
S a m i a cynthia, media, 457
Sarcoma cell, protein content, 145
Schmidt-Ruppin strain, Rous sarcoma virus,
380, 381, 394
Schneider medium, 464--466
Selenite, in media, 500
638
SUBJECT INDEX
Selenium
deficiency, 74
in media, 52, 69, 86, 89, 90
neuroblastoma, 98
Self-absorption, in autoradiography, 289,
290
Sendai virus
cell fusion, 353-356
in nucleus transfer, 359
storage, 354
Sephadex G50M, as microcarrier, 186
Serine, in media, 63
Sertoli cell, 103
Serum, see also specific type
bovine, see Bovine serum
certification, 197
cloning, 163
commercial production, 197
deprivation, synchrony, 259-261
dialyzed, 82, s e e a l s o Serum protein
effect on growth, 77-79, 94-109
on macrophage, 501
minimal requirements, 91-93
neuronal cell, 587
pretreatment, 95
as trypsin inhibitor, 95, 128
weaning from, 91-93
Serum albumin, in media, 88
Serumless medium, 57
composition, 62-70
Serum protein, 82
minimalization, 49-51
7C's medium, 58, 89
Shearing, 223
effect on cell, 224
Silicon, in media, 69
Silkworm, media, 457
Simian virus 40, see SV40
Sinclair's medium, 58
Sindbis virus, microcarrier culture, 192
Skeletal myoblast, see Myoblast, skeletal
Skeletal muscle, dispersion, 123
Skin biopsy
explants, 446-448
method, 445, 446
Skin epithelium, differentiation, 265
Skin explant, 446--448
Slide preparation, chromosomes, 326-334
S+L
- test, 423, 424
SM-20 medium, 59, 90
SM-201 medium, 59
Snail cell, media, 466

Sociocell culture, 275
Sodium, in media, 68
Sodium nitrate, as fusing agent, 361,362
Solute transport, MDCK cells, 554, 555
Solvent, in media, 70
Soma, separation from neurites, 302-307
Somatomedin, in media, 78, 99, 103
Soybean culture x barley mesophyll, 365
Soybean culture x C o l c h i u m mesophyll, 365
Soybean culture x corn mesophyU, 365
Soybean culture x pea mesophyll, 365
Soybean culture x rapeseed mesophyll, 365
Soybean culture x tobacco mesophyll, 365
Soybean culture x Viola mesophyll, 365
Soybean trypsin inhibitor, 95
Spectrophotometry, of hemoglobin, 511
Spin fiiture culture, 14
Spinner culture, 14, 203
Spinner flask, 46
Spleen
lymphocyte preparation, 124
neonatal, cell fusion, 357
S p o d o p t e r a f r u g i p e r d a , media, 460
Spruce budworm, media, 457
$77 medium, 546
Staining, see also specific types
autoradiography, 285
centromere region, 325
chondrocyte, 563
DNA, 240, 246, 412
fluorescent, 233, 234, 236
chromosome, 325
hemoglobin, 510
isoenzymes, 168-170
macrophage, 502
metaphase cell, 172-174
monolayer cultures, 140
nucleolus organizing region, 326
solutions, 246
Stansted Cell Disrupter, 223
Sterility test, 19-21
Sterilization, 3-7, 10, 11, 75, 182
insect media, 457
Steroidogenesis, YI adrenal cell line, 571
Stock culture, 117-119
defined, 117
Stock salt solution, 198
Storage, see Cell storage; specific cell type
by freezing, see Freezing
SUBJECT INDEX
Storage facility, 3
Streptomycin, in media, 113, 115
Stromal cell, 266
primary culture, 278
Subcell culture, defined, 263
Subculture, 95
GH cells, 532
invertebrate cell line, 451,452
poikilotherm vertebrate cells, 476, 477
YI cell line, 572, 573
Sulfate, in media, 68
Superinduction, interferon, 295
Supernatant reverse transcriptase assay, 416
Suspension culture, 14, 46--49, 81
cell lines, 202-206
chondrocyte, 563
HTC cells, 550, 551
invertebrate cell, 454, 455
media for, 56, 58, 89
operations, 205
plant cell, 481,482
primary cell, 206-210
scaling-up, 211-221
Swiss 3T3 cell, 96, 97
SV40
biological activity, 412
cleavage maps, 412
culture, 405
MEM, 405, 406
preparation, 404--412
transformation, 370
SV40 DNA, 405
assay, 411,412
preparation, 406--410
purified virions, 409, 410
quantitation, 411,412
radiolabeling, 410, 411
Sympathetic ganglion-like PC12 cell, 585
Sympathetic neuron
cell preparation, 575,576
cholinergic activity, 575
co-culture, 575,581
conditioned medium, 581,582
long-term culture, 574-584
media, 578-580
microculture, 582, 583
phenotypic characteristics, 583,584
use of Methocel, 580
Synchronous growth, 218
639
Synchrony, 248-262
calcium deprivation, 261,262
criteria, 249
degree of, 249-257,259
double-thymidine block, 261
evaluation, 249-252
Ficoll gradients, 257, 258
gradient centrifugation, 256-258
hydroxyurea, 259-261
isoleucine deprivation, 258, 259
mitotic detachment, 252-256
myoblast, 511,518-520
physical methods, 256, 257
serum deprivation, 259-261
Syrian hamster karyotype, 342
Syringe adapter, 9
T
Taurine, in media, 63
T ceil, 178, 493
Teleost
balanced salt solution osmolarity adjustment, 469
media, 469
Temperature, effect on cuRure, 55, 135, 450
Teratoma, ovarian, 376
Testicular cell
mouse TM4, 96, 97
Testosterone, in media, 102, 103
Tetracarcinoma, 376
Tetracycline, in media, 113, 115, 116, 565
Tetraphenylboron, as chelator, 126
Thiamin, in media, 54, 65
Thiamin monophosphate, in media, 65
Thiamin pyrophosphate, in media, 65
Thioctic acid, s e e a-Lipoic acid
Thioglycolate broth, 20
6-Thioguanine
hybrid selection, 352
3T3 cell, 376
monolayer, 133
mouse, DNA content, 148
Threonine, in media, 53, 62
Thymidine
incorporation, 251, 252
in media, 67
640
SUBJECT INDEX
Thymidine (cont'd.)
synchrony, 261
tritiated, as mutagen, 316, 317
Thymine, in media, 67
Thymus, lymphocyte preparation, 124
Thymus cell, canine, 414
Thyrotropin-releasing hormone, in media,
99
Thyroxine, in media, 63
Tick cell line, media, 463
Tin, in media, 69
Tissue culture, 119-131
defined, 263
Tissue dispersion, 119-131
mechanical, 130, 131
shearing, 130, 131
solutions, 124
Tissue disruption, 119-131
Tissue dissociation, see Tissue dispersion
Tissue preparation, 122, 123
mincing, 124
Tobacco, from protoplast, 367
Tobacco cell, callus culture, 479
Tobacco hornworm, media, 457
Tobacco leaf, protoplast, 302
Tobacco mesophyll x chicken red cell, 365
Tobacco mesophyll x HeLa culture, 365
Tobacco mesophyll soybean culture, 365
a-Tocopherol phosphate, see Vitamin E
Togavirus, interferon producing, 293
Toluidine blue, chondrocyte staining, 563
Torenia baillonii T. fournieri petals, 365
Toxicity, media, 117, 453, 454
Trace element, in media, 52, 69, 95
Tracheal cell, embryonic bovine, 29
Tranfection-infectivityassay, 434, 435
Transferrin, in media, 95, 99-107, 109, 500
Transformation, see also Cell, transformed;
Cell transformation
neoplastic, 296-302
soft agar assay, 297,299, 300
Triatoma infestans, media, 463
Trichoplusia ni cell, 450
media, 457,460
storage, 453
Triiodothyronine, in media, 99, 101
Trout cell, incubation temperature, 471
Trypan blue, cell viability, 151, 152
Trypsin
contamination, 125
in media, 139
neutralization, 77
Trypsin-EDTA solution, metaphase preparation, 171
Trypsin inhibitor, 95, 128
Trypsinization, 50
adenovirus, 433,434
BHK cells, 298
in cloning, 156
enzymes for, 124-128
myoblasts, 517
poikilotherm vertebrate cell, 475, 476
Trypsinizing flask, 127
Trypticase soy broth, 20
Tryptophan, in media, 53, 62
Tumor, transplantable, 276, 277
Tumorigenic cell line, see Cell line,
tumorigenic
Tumorigenicity
testing
nude mouse, 370-379
protocol, 373-379
threshold, 374
Tumor-specific cell surface antigen, see
Antigen, tumor-specifc-cell surface
antigen
Tween 20, in media, 70
Tween 80, in media, 70
Tylosin, in media, 113, 115, 116
Tyrode's balanced salt solution, 120, 121,
452
Tyrosinase
assay, 568
melanocytes, 564
toxicity, 567
Tyrosine, in media, 53, 62
U
Ultrafiltration, virus, 228
Ultraviolet light, as mutagen, 312
Uracil
in media, 67
uptake, mycoplasmal, 26, 27
Uridine, in media, 67
Uridine phosphorylase, assay, 25, 26
Uridine triphosphate, in media, 67
Urine, cloning, 163
Ussing chamber, 555-557
641
SUBJECT INDEX
V
Valine, in media, 53, 62
Vanadium, in media, 52, 69
Vascular endothelial cell, monolayer, 133
Vasopressin, cAMP production, 559
Velveteen cloth, for plating, 320
Ventilation, laminar air, 10
Vertebrate cell, 466-477
lower, 467
osmolarity adjustments, 469
poikilotherm, 466--477
media, 470
monolayer culture, 474-476
pH, 470
primary culture, 474-476
subculture, 476, 477
temperature requirements, 470--472
Vesicular stomatitis virus, microcarrier culture, 192
Viability, s e e Cell viability
Vicia, protoplast, 361
Vicia culture pea mesophyll, 365
Vicia mesophyll soybean culture, 365
Vinblastine, mitotic inhibitor, 256
Viokase
in media, 532, 572, 573
neuronal cell culture, 587
Viomycin, in media, 113, 115
Viral antigenic determinant, 418
Viral contamination, s e e Contamination,
viral
Viral DNA, assay, 434, 435
Virus, see a l s o specific type
acquisition, 226, 227
banding, isopycnic, 415
detection, cell fusion, 348, 349
DNA, s e e Viral DNA
infectious, biohazard, 37, 42, 43
isolation, 222, 227
in media, 225-228
nontransforming, large-scale growth,
403
preparation, 222
production
effect of serum, 226, 227
mechanization, 400, 401
ultrafiltration, 228
purification, 401-403,414-416
testing, 28, 29
xenotropic, 413,414
Vitamin
fat-soluble, 54, 65
in media, 54, 64, 65
Vitamin A, in media, 65
Vitamin B, in media, 54, 65
Vitamin D, in media, 65
Vitamin E, in media, 65
Vitamin K, in media, 65
V-605 medium, 56
W
Walker carcinosarcoma 256 cell, 86
Water
distillation, 195
preparation, 94
cost of, 195
quality, 8
system, suppliers, 17
treatment, 8
Waymouth's medium MB 752/1, 57, 88, 90
composition, 62-70
White's medium, 56
Williams' medium E, 59, 90
Williams' medium G, 59, 90
composition, 62-70
Wisconsin 38 cell, callus culture, 479
WISH cell, protein content, 145
WI-38 cell
cloning, 155
DNA content
Wollenberger medium, 9
WO5 medium, 57
Working culture, defined, 118
Wrist shaker, 203
X
XC test, 421--423
Xenotropic virus, s e e Virus, xenotropic
Y
Yamane's medium, 57
composition, 62-70
Y chromosome, 174
642
YI adrenal cell line, 570-574
effect of ACTH, 571
initiation of stock, 572
karyotype, 571
media, 571
monolayer growth, 572
stability, 570
steroidogenesis, 571
SUBJECT INDEX
storage, 573
subculture, 572, 573
Yunker, Vaughn and Cory medium, 461
Zinc, in media, 52, 69, 89

Methods in Enzymology Volume 58 Cell Culture

Uploaded by

Copyright:

Available Formats

Methods in Enzymology Volume 58 Cell Culture

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Methods in Enzymology Volume 58 Cell Culture

Uploaded by

Copyright:

Available Formats

What are the three main parts of the book?

What are the three main parts of the book?

What are some of the cell types that are discussed in the book?

What are some of the cell types that are discussed in the book?

Preface

RONALD T. ACTON (17, 18), Department of

ROBERT B. CAMPENOT (25), Department of

Microbiology and the Diabetes Research

Neurobiology, Harvard Medical School,

CONTRIBUTORS TO VOLUME LVIII

IRWIN R. KONIGSBERG(45), Department of

Biology, University of Virginia, Charlottesville, Virginia 22901

CONTRIBUTORS TO VOLUME LVIII

Neurobiology, Harvard Medical School,

tute, National Institutes of Health,

CONTRIBUTORS TO VOLUME LVIII

Training Center, University of Alabama in

partment and Research Institute, The

By WILLIAM H. J. DOUGLAS and ROBERT T. DELL'ORCO

Copyright ~ 1979 by Academic Press, Inc.

capabilities. In contrast, most routine biochemical procedures, such as

ASPECTS OF A TISSUE C U L T U R E LABORATORY

capable of producing ultrapure, reagent-grade water. Such systems are

ASPECTS OF A TISSUE C U L T U R E LABORATORY

III. Media Preparation and Storage Facilities

has been added to media. It is also necessary to eliminate toxic elements

ASPECTS OF A TISSUE C U L T U R E LABORATORY

IV. W o r k A r e a for Aseptic Manipulation of Cell Cultures

Many cell culture manipulations may be routinely conducted on a

ASPECTS OF A TISSUE C U L T U R E LABORATORY

b e c o m e the dominant cell type in the culture. If an investigator is familiar

ASPECTS OF A TISSUE C U L T U R E LABORATORY

ASPECTS OF A TISSUE CULTURE LABORATORY

vivo state. The artificial capillaries are semipermeable cellulose acetate or

Media & Biologicals

HEM Research, Inc.

Life Science Group

Cappel Laboratories, Inc.

26201 Miles Rd.

Media & Biologicals (continued)

Lux Scientific Corp.

Gelman Instrument Co.

ASPECTS OF A TISSUE CULTURE LABORATORY

Filtration Systems (continued)

Bausch & Lomb, Inc.

American Optical Corp.

METHODS IN ENZYMOLOGY, VOL. LVIII

Copyright 1979 by Academic Press, Inc.

and biologicals is generally more difficult to detect because of the low

requested. Bacterial and growth promotion tests are described below.

1. Testing for Bacteria and Fungi before Purchase

2. Testing for Growth Promotion of Cell Cultures

3. Testing Individual Bottles of Serum

B. Cell Culture Media

M a n y techniques are available for detection of m y c o p l a s m a s . T h e s e

M y c o p l a s m a l testing is p e r f o r m e d on bovine serum before purchase.

Base m e d i u m is p r e p a r e d in bulk and stored in 75-ml amounts until

Inoculate a mycoplasma broth tube with 0.5 ml and mycoplasma agar

perchloric acid (PCA). Incubate in 90 water bath for 15 min to hydrolyze

PRESERVATION, STORAGE, AND SHIPMENT