United *
Environmental PratectmVi
Agency
Tranafd/'
Effects Research
Laboratory
coh Triangle Park NC 2771
v>EPA
•*
',/
jf .•
Environmental Assessment
I •» ,», , » -,, -.'
[Short-Term Tests for
^^cTjfifl'' ,
Car'cinopensr Mufagens and
Other Genotoxic Agents
-------
Cower—Chemically transformed 8a!b/3T3, 1-3 mouse fibroblasts (x20Q, Giemsa stain), showing
borderline between normal and malignant cells, The darker, spindle-shaped transformed cells pile
up anci constitute the foci, Provided by Dr. D. R. Lang, Department of Microbiology, University of
Cincinnati, Cincinnati, Ohio.
-------
TechnologyTransfer EPA-625/9-79-Q03
for
and
July 1979
This report was developed under contract by
Energy Resources Co, Inc.
in cooperation with the
Genetic Toxicology Program
Health Effects Research Laboratory
Research Triangle Park NC 27711
for the
Environmental Research Information Center
Cincinnati OH
-------
Introduction
Short-Term Tests and Hazard Assessment
Short-Term Tests
Phased Testing Strategies
Program Applications
Research Trends
Perspective
2
8
8
14
IS
20
25
Appendix:
How Effects are Measured in Various Short-Term Tests
Representative Short-Term Tests for Genotoxicity
26
21
Glossary
-------
Introduction
OF
COSTS
OF
Short-Term
Long-Term
Short-Term
Lonf-Term
$200-
Per Test
$20,000-
Per Test
4 Days to
26
26 to
3 Years
-------
Introduction
In recent years, federal statutes such as the Toxic Substances
Control Act, the Resource Conservation and Recovery Act,
and the Clean Air and Clean Water Acts have given the U.S.
Environmental Protection Agency IEPA) the responsibility for
regulating the release of toxic chemicals into the environment.
In order to fuffili this function effectively, the EPA must first
determine which of the thousands of chemicals currently in
use or proposed for use are toxic.
Detecting a chemical's ability to cause immediate (or acutel
toxic effects is a relatively straightforward task. Assessing the
long-term (or chronic) toxic effects is much more difficult.
Chronic effects such as cancer, birth defects, and genetic dis-
ease characteristically appear several years or decades after
the initial chemical exposure has occurred, and long-term
studies using live animals must be conducted in order to
detect these latent effects. Such studies are expensive and
time-consuming, and require the use of highly specialized facil-
ities and personnel. A single test for a chemical's earctnogenic-
fty (cancer-causing ability}, for instance, may take as long as 3
years and cost $250,000 or more.
The number of compounds whose chronic toxicrty has not
been determined is overwhelming. Over 50,000 chemicals are
currently In commercial production, and most of them have
never been examined for chronic effects. The world laboratory
capacity for long-term studies has been estimated at only SGQ
compounds per year, not enough to keep up with the 700 to
1,000 new chemicals that are introduced into commerce
annually.
In response to this situation, short-term tests have been devel-
oped to serve as rapid and relatively inexpensive predictors of
a chemical's potential to cause chronic effects, These tests
employ bacteria, yeast, plants, insects, isolated mammalian
cells and whole animals. Short-term tests can detect a
chemical's genotoxicity, that is, its ability to alter a cell's
genetic material fDNA). An increasing amount of evidence
exists to indicate that latent diseases such as cancer, birth
defects, and genetic disease may be initiated by alterations in
the DNA.
OTHER
CHRONIC EFFECTS
Short-term tests enable a large number of chemicals to be
screened for their genotoxic potential at a fraction of the time
and cost required for long-term tests. Results from short-term
tests can be used to make more informed decisions as to
which chemicals should be examined in the limited number of
long-term testing facilities available. Several other promising
uses for short-term test data include;
* Determining which of several alternative chemicals under
development will be the least hazardous to human health.
* Identifying the toxic components of complex environmental
pollutants.
» Monitoring industrial emissions, effluents, and wastes in air,
water, and soil.
* Determining which control technologies are most efficient in
eliminating toxic chemicals.
* Providing interim guidance for using a chemical when no
other data are available.
-------
Introduction
Because of their rapid and inexpensive nature, short-term tests
are extremely valuable in helping the EPA to fulfill its responsi-
bility for identifying and regulating toxic substances. For this
reason, significant efforts are being applied to the research and
development (R&D) of short-term tests. The purpose of this
document is to briefly describe some of EPA's R&D activities in
this area.
The document is organized into five sections. The first section
discusses how short-term tests can contribute to hazard as-
sessment, while the second describes the scientific basis and
techniques of short-term tests, A general strategy for how
short-term tests can be used to detect a chemical's potential
long-term toxieity is outlined in the third section. Some pro-
gram applications of short-term test research are presented in
the fourth section, and the fifth section describes some of the
current research activities. An overall perspective concludes
the document. A glossary of technical terms is provided at the
end of the document along with an appendix of technical
information on specific short-term tests.
-------
Introdyction
HOW DNA ALTERATIONS MAY BE RELATED TO
MUTATiQ'N, CANCER, AND OTHER CHRONIC DISEASES
Alteration in DNA
In Reproductive Ceils
(eggs or sperm)
in Nonreproductive Cells
{somatic cells like skin or organs)
» Cell Death
• Cancer
• Aging, Heart Disease, or Other Illness
» Birth Defects
• Genetic Diseases
-------
Short-Term Tests Assessment
THE THREE MAIN SOURCES OF INFORMATION ON THE CHRONIC EFFECTS OF A CHEMICAL
EPIDEMlGLQGiCAL
SHORT-TERM
TESTS
LQWG-TERM
ANIMAL STUDIES
-------
Short-Term
Hazard assessment is the process of evaluating the human
health threat associated with a chemical. It involves answering
such questions as; Does the chemical have the potential to
cause serious human health effects such as increases in cancer
levels or birth defects? How great an exposure is necessary to
cause an effect? Are any special groups of the population par-
ticularly susceptible to the chemical's effects? Is the health risk
serious enough to require use restrictions or an outright ban on
the chemical? Should more research be done to evaluate the
hazard?
Three principal sources of information can be used to evaluate
the health risk associated with a chemical:
• Human exposure (or epidemiologies!) data.
• Long-term tests ysing various species of animals.
» Short-term tests using microorganisms, plants, insects, and
animals.
In a typical hazard assessment, scientists consider alt the avail-
able data but ascribe different significance or weight to each
set of data depending on the type of information and quality of
the test or study.
At their present state of development, short-term tests are not
considered to be as authoritative as epidemiological data or
long-term whole animal studies. For this reason, short-term
data are not used to provide "definitive" evidence that a
chemical is hazardous. Rather, they are considered to be "sug-
gestive" evidence of a chemical's potential to cause genotoxic
effects,
The official EPA policy regarding the use of short-term tests
for carcinogens has been expressed by the Interagency Regu-
latory Liaison Group (IRLG)1 in their report "Scientific Bases
for Identifying Potential Carcinogens and Estimating Their
Risks";
Short-term tests for chemical carcinogens presently do not, in
the absence of animal bioassay [test] and epidemiology data,
constitute definitive evidence as to whether a substance does
(or does not} pose a carcinogenic hazard to humans. However,
positive responses in these tests are considered suggestive evi-
dence of a carcinogenic haiard,
'The Interageney Regulatory Liaison Group is a consortium of four federal
agencies, the EPA, the Occupational Safety end Health Administration, the
Consumer Product Safety Commission, and the Food and Drug Administra-
tion,
Thusr short-term studies can be used to support existing ani-
mal data, or as a temporary substitute if these data are lacking
In rare instances, short-term tests can call into question ade-
quately conducted iong-term animal studies, but this can occu
if and only if short-term test data are consistently and clearly
positive and long-term findings are negative, tn this ease,
short-term test data are taken as suggestive evidence of haz-
ard until further long-term testing resolves the discrepancy.
Accuracy
A great deal of research is currently being done to determine
how accurate short-term tests are in predicting whether or not
a chemical has the potential to cause human health effects.
The accuracy of short-term tests for carcinogens is usually
determined in reference to animal or human data. The more
frequently the results of a short-term test concur with what is
known about a chemical's carcinogenic potential through long-
term tests or epidemiological data, the more accurate that test
is considered to be.
For mutagenicity {ability to cause mutations) tests, the issue of
accuracy is not so clear-cut. Because of the technical difficul-
ties involved in detecting mutations in the human population,
there are no human data that can be used to validate short-
term test results. At present, the accuracy of a short-term
mutagenicity test must be determined by comparing test
results with the findings from other mutagenicity tests. The
degree of concordance with other mutagenicity findings is
considered to be the best measurement of a test's accuracy.
There are basically two ways a short-term test can give an
inaccurate result. It can indicate that a chemical is genotoxic,
when in fact it is harmless (this is called a false positive result),
or it can indicate that a chemical is harmless when it is actually
genotoxic (a false negative result}. Different short-term tests
vary in their likelihood of making false positive and false nega-
tive errors.
False negatives are of great concern in short-term tests, partic-
ularly when these tests are used as early warning systems (see
section on Phased Testing Strategies). With a false negative, a
toxic chemical may not be examined in iong-term tests before
significant human or environmental exposure occurs. With a
false positive, there is a good chance the error witl be cor-
rected during follow-up testing. In general, false positives
present a cost rather than a public health concern. Ideally, a
short-term test should minimize both false positives and false
negatives.
-------
Short-Term Tests
KEY TO SHORT-TERM TESTING
_
Lower Organisms
c
.2
'.Ł•
a
E
o
M~»
c
CD
C
03
o
>
'x
J3
a
o
O
CTJ
c
TO
®
o
C
Higher Organisms
Assumed site for
carcinogens, rnutagens,
and genotoxic agents.
-------
Short-Term
Scientific Basis
Short-term tests for genotoxtcity look for the abitity of a
chemical to damage the genetic material (ONA) of a cell. Since
DMA controls all the functions of the individual cells that make
up an organism, even a small change in the DNA can have
severe consequences for an organism or its tiff spring. As was
stated in the introduction, an increasing amount of evidence
exists to indicate that latent diseases such as cancer, binh
defects, and genetic disease may be caused by alterations in
DNA.
The key to short-term testing is the fact that the fundamental
structure of DNA is the same in all organisms (see Figure 11.
Thus, a chemical that affects the DNA of a single cell or orga-
nism in a short-term test can theoretically have a similar effect
on the DNA of an exposed human,
Although the fundamental structure of DNA is the same in all
organisms, the amount and complexity of DNA varies accord-
ing to the complexity of the organism. Generally, the more
complex a test organism and its form of DNA organization, the
more likely it is to approximate the human response. Thus,
mammalian cells are better models than bacteria for human
cells.
How They Work
AH short-term tests follow the same basic format. The sub-
stance of interest is applied in some predetermined concentra-
tion to the test system or organism. The substance is then
metabolicaliy activated either within the organism or by the
addition of a special enzyme treatment. After a suitable period
of time to allow the effect to take piace, the test system is
examined for signs of genotoxicity. This may be done in any of
several ways, as indicated in Table 1 at the end of this section.
Metabolic activation is an essential element of short-term test-
ing. As Figure 2 indicates, many chemicals are not toxic them-
selves, but can be converted into a toxic chemical by an
organism's normal chemical conversion processes (metabo-
lism).
Compared to animafs and humans, the microorganisms and
isolated animal ceSIs used in short-term tests have only a lim-
ited capacity for metabolizing chemicals, For this reason,
compounds that are not directly active (these compounds
are sometimes called procarcinogens or promutagens)
cannot be detected in short-term tests unless some form of
metabolism is supplied to "ectivate" the chemical. Fortunatelv
several types of metabolic activation are available. Three
methods are illustrated in Figure 3. Of these three, in vitro
activation is the most commonly used because of its simplicity
and effectiveness.
Lethal Toxicity Determination
Before any test is made of a chemical's biological effects, it is
important to determine the dose or concentration that wilt
allow the test to be performed effectively. Too large a dose ca
kill the test organisms, so that the effect being studied never
has tha chance to appear. Too small a dose can give a misleat
ing impression of safety, since the effective dose may not be
achieved,
Fortunately, relatively straightforward techniques exist for
determining the optimal range of concentrations for testing.
Several different concentrations of a chemical are applied to
the test system and the survival at each dose is determined.
Doses that exhibit some toxicity, but that do not drastically
reduce the cell populations, are generally the ones used for
short-term testing.
-------
Short-Term
10
Types of Biological Activity
Five different types of biological activity related to genotoxieity
can be studied in short-term tests:
* DMA damage and repair,
• Gene mutation.
• Chromosome alterations.
• Cancer-like (oncogenic) cell transformation.
9 Tumor formation
Each of these classes of activity is discussed below.
— ONA Damage and Repair
Whenever a cell's DNA is disturbed or damaged, cell mech-
anisms come into action to repair the damaged parts, DNA
damage and repair tests take advantage of this fact in
searching for evidence of genotoxic effects. These tests
look either for direct evidence of alterations in the DMA, or
for evidence that DMA repair mechanisms are in action. If
the repair mechanisms can be demonstrated to be operating
above the normal level, then damage to the DNA is indi-
cated. DNA damage and repair tests are available using
bacteria, yeast, mammalian cells, and whofe animals.
— Gene Mutations
Gene (or point) mutations are submicroscopic DNA altera-
tions occurring in a single gene1 and leading to an altered
gene product. Most gene products are proteins. Since pro-
teins are involved in at) the chemical reactions taking place
in a cell or organism, the biological consequences of even a
small change can be severe. For instance, a minute change
in an enzyme (a type of proteinl can interfere with a cell's
normal functioning by making a key chemical reaction
impossible.
Gene mutations are most easily detected by looking for
altered gene products, such as enzymes. An enzyme defi-
ciency can manifest itself in any number of ways that can
be conveniently measured. In test systems using bacteria,
yeast, or mammalian cells in culture, a cell's requirement
for a certain nutrient may change, or its tolerance of a
chemical poison may be altered, in short-term tests involv-
ing whole organisms such as fruit flies or plants, specific
changes in the test organism's features (i.e., color or
shape) can be observed as evidence of mutation.
Chromosome Aberrations
Chromosome alterations or aberrations are microscopically
visible disturbances in chromosomes,2 They can include the
loss or gain of entire chromosomes, chromosome breaks,
and faulty assembly processes such as nondisjunctions and
translocations. Chromosomal aberrations are a major cause
of heritable human disease, and their occurrence is often
associated with cancer. They are detected either by search-
ing for microscopically evident alterations or by examining
tissues or organisms for traits known to result from such
alterations. Cells from insects or mammals are frequently
used.
Oncogenic Transformation
Oncogenic transformation is the chemically induced conver-
sion of normal cultured mammalian cells into masignant-like
cells. Whether or not transformed cells are actually malig-
nant lor cancerous! can be ascertained by injecting them
into whole animals to see if they give rise to tumors. Most
frequently, transformed cells are distinguished in culture by
abnormal growth patterns that are visible under the light
microscope.
Tumor Formation
Tumor formation in rodents is a definitive indicator of a
chemical's carcinogenicity. Short-term tests measuring
tumor formation use special strains of mice and rats that
develop tumors especially rapidly — within 10 to 26 weeks
of chemical treatment. (In traditional long-term or lifetime
tests for carcinogenidty, 2 or more years may elapse before
tumors appear.)
In short-term tests for tumor formation, the number of
tumors appearing in treated animals is compared to the
number that have appeared spontaneously in an untreated
control group of animals. A higher number of tumors m the
treated animals indicates potential carcinogenicity. Non-
malignant tumors may be counted when their presence cor-
relates with the later appearance of malignant tumors.
2A form of ONA organization found in higher organisms.
*A portion of the DNA that directs the formation of a single product.
-------
Short-Term
11
THE MECHANISM OF METABOLIC ACTIVATION
Promutagen
or
Procarcinogen
Promutagen
or
Procarcinogen
Promutagen
Procarcinogen
Metabolic
Activation
Derived from
Another
Source
No
Metabolic
Activation
Metabolism
or Other
Whole Animal
Processes
Isolated Cells
Isolated Cells
No
Harmful
Effect
Observed
Harmful
Effect Is
Observed
Modified form {metabolite) of the original
chemical that is now capable of causing cancer
or mutation.
Biological processes that can take place in humans and
other organisms can modify a nonreactive chemical
to make it harmful.
Unless metabolic activation is provided,
isolated cell systems cannot detect
promutagens and procarcinogens.
Fiyuie 2
-------
Short-Term
12
THREE TYPES OF METABOLIC ACTIVATION
FEEDER CELL ACTIVATION
IN VITRO
ACTIVATION
ACTIVATION BY
BODY FLUIDS
Induce enzymes in
animal with inducer
Expose antrnal
to chemical
Collect blood,
urine, feces
FEEDER CELL
that can activate
chemical but
•cannot show
effect
INDICATOR CELL
that cannot activate
chemical but can
show effect
Kill animal and re-
move organ(s)
(liver, kidney, lungs)
Isolate metabolites
of chemical
Break down
organs to cell
fragments
Isolate active
forms
Separate out
enzymes
Chemical
Activation
Occurs
\\. Add to cell
\ \. system
V
C\ enzyme mixture
\i\ to cell system
\>
Figure 3,
-------
Short-Term
13
Table 1
Basic Ways Short-Term Tests Are Done
How to tell a
Systems8
Visual observation of whether growth has occurred under
special conditions
Visual observation of abnormal growth patterns
Microscopic examination for gross changes in the genetic
material
Visual observation for unusual color or shape
Observation in whole animals of birth tosses and unusual
offspring
• DNA damage in microbes
* Gene mutation in microbes arid isolated mammalian celts
• Oncogenic transformation of isolated mammalian cells
* Cytogenetic assays for chromosome alterations
* Unscheduled DNA synthesis using mammalian celis
* Micro-nucleus test with mammalian cells
* Sister ehromatid exchange in mammalian ceils
* Sperm morphology test
8 Trac/escantia — plant gene mutation test
« DNA damage in yeast
* Gene mutation or chromosomal alterations In fruit flies
(Drosophila)
• Dominant lethal test in rodents or fruit flies
* Heritable translocation in rodents
* Specific locus test in mice
BSee appendix for more complete information.
-------
14
PHASED TESTING STRATEGY
Effect Being Tested
1
i
Careinogeiesis
PHASE ONE:
(Detection)
Mutagenesis
Mytagenesis/Carcinogenesis
• Microorganisms (± Activation)'!
Point Mutations
Primary DNA Damage
• Mammalian Ceils
Chromosomal Effects
PHASE TWO
(Verification)
•Mammalian Cells (± Activation)
Gene Mutations
Primary DNA Damage
* Mammalian Celts (± Activation)
Oncopenic Transformation
» Insects and Plants
Gene Mutations
Chromosomal Effects
I oitiation/Promotion
* Rodents
Carcinoqenesis Bioassay
(Skin}
• Rodents
Gene Mutations
Chromosomal Effects
PHASE THREE;
(Risk assessment)
« Rodents
Gene Mutations
Chromosomal Effects
* Rodents and Other Animals
Carcinogenesis Bioassay
D Core Battery Test
* Test Organism
In Vitro (New York. Academic Press, 1977).
-------
Testing
1i
Short-term tests are rarely used individually for risk assess-
ment, since no single short-term test is capable of detecting ail
the types of effects that may be caused by a genotoxic chemi-
cal. More often, a group of tests is used, with the particular
testing strategy depending on time considerations, cost, and
the type, scope, and accuracy of information desired.
Recently, EPA scientists have been exploring the advantages
of a phased testing strategy for chemical hazard assessment in
which testing is conducted in one or more distinct stages or
phases. The extent of testing at each phase is determined by
the test results from the previous phase and by the degree of
potential hazard suggested by factors such as production vol-
ume, projected human exposure, and the known toxicity of
related chemicals. By considering this information, limited test-
ing resources are utilized in a manner that provides for the pro-
tection of human health in proportion to the anticipated risk
involved.
In the phased approach, tests are organized into three phases,
Phase One tests principally involve work with microbes and
cost on the order of $2,000 or less. They are less definitive
than Phase Two short-term tests which commonly use mam-
malian celts, insects, and plants and cost about 10 times as
much as Phase One tests. Phase Two tests are used to confirm
the effects detected in Phase One and to characterize more
specifically the nature of the effects (i.e., whether the chemi-
cal is carcinogenic or rnutagenic). Phase Three tests are gener-
ally whole-animal studies with rats or mice. They may cost
$250,000 or more and take several years to complete. Phase
Three tests provide the most authoritative evidence concerning
the degree of risk posed by a chernica! and may be used to
establish acceptable levels for environmental exposure to a
chemical. Because of the vast time and expense involved,
Phase Three tests are best reserved for a limited number of
high priority chemicals,
The specific tests used in each phase vary from chemical to
chemical, but the basic concept is the same for all phased
strategies. Simpler, less expensive tests are first conducted to
gain a preliminary indication of a chemical's toxic potential.
More thorough and expensive testing is performed in succes-
sive phases only if it appears to be merited on the basis of pre-
liminary test results or the degree of anticipated risk.
To ensure thorough testing, a "core battery" of short-term
tests may be performed. The "core battery" is a group of
essential tests that scientists agree must be conducted in order
to determine with reasonable confidence whether or not a
chemical may be mutagentc or carcinogenic.
The core battery includes four types of Phase One and Phase
Two tests:
• Gene mutations in microorganisms and isolated mammalian
cells.
• Chromosome aberrations, preferably in cells from treated
animals.
• ONA damage in mammalian celts,
» Malignant-ltke changes loncogenic transformation! in mam-
malian cells.
Phased testing strategies can take advantage of the fact that
some tests tend to give false positive results while others tend
to give false negatives. False positive-prone tests may be deli-
berately chosen for Phase One so as to maximize the chance
that a hazardous chemical will be detected and sent on for fur-
ther testing. In designing Phase Two, scientists can use tests
that are less likely to give false positives so as to minimize the
chances that a relatively innocuous chemical will be sent on for
expensive Phase Three testing.
FUNCTION OF TESTING PHASES
• Detection of Hazard
Confirmation of Phase One Results
# Delineation of Hazard Type
» Final validation of Hazard
• Quantitative R ssk Assessment I
The EPA has begun to use phased testing strategies on a lim-
ited scale in its research activities. In this context, phased
strategies have proven particularly helpful In analyzing complex
mixtures for their hazardous components. An example of the
EPA's use of phased testing is described under "Diesel
Exhaust" on page 18,
-------
SHORT-TERM TESTING IN SUPPORT OF VARIOUS EPA PROGRAMS
HAZARDOUS
WASTES
TOXIC
SUBSTANCES
Diesel exhaust
emissions
Industrial
effluents
Dnnkmg
water
Organic
chemicals
Ambient air
Energy
technology
effluents
Textile
wastewater
Inorganic
chemicals
Fluidized bed
combustion
emissions
U nconcentrated
source water
Conventional
combustion
emissions
Figure 4,
-------
Program Applications
17
Although the primary purpose of the EPA's Research and
Development programs for short-term tests is to develop,
refine, and apply test systems, chemicals that are of genuine
public health concern are often selected for use in the research
program. This practice enables preliminary toxictty data to be
obtained at the same time that test development is taking
place, A number of EPA regulatory programs have been
assisted in this way {see Figure 4), Two prime examples of
research applied to regulatory program uses are discussed
below.
Pesticides
Several years ago, the EPA initiated a short-term testing pro-
gram on 39 pesticides representing several different chemical
classes (i.e., organophosphates, chlorinated hydrocarbons,
and carbamates). As many as eight different short-term tests
were performed for each compound to determine how these
b«ocidat Oife-kiiling) materials would behave in short-term tests,
Specific issues addressed were:
• Would the highly acute toxic properties of pesticides inter-
fere with the ability of the short-term tests to detect geno-
toxic effects?
» Would every compound show up positive in at least one
test, and would this be a realistic indication that ail the com-
pounds were genotoxic?
* Which short-term tests were most suitable for use with pes
ticide chemicals?
• What was the potential of the pesticides to cause long-term
health effects such as mutations and cancer?
Since a phased tasting strategy was not being used. Phase
One, Two, and Three tests were performed concurrently.
Phase One tests consisted of assays for bacterial DNA damage
and gene mutations, while Phase Two tests looked for unsched-
uled DNA synthesis in mammalian cells and gene mutations
in the fruit fly Drosophila. The Phase Three tests involved
assays for dominant lethal mutations and heritable chromoso-
mal translocations in mice. Due to resource limitations, only 20
compounds could be tested in Drosophila, only 10 in the domi-
nant lethal assay, and onfy one in the heritable transtocation
assay.
As it turned out, the short-term tests were adaptable to these
highly bioetdal materials, Test results for each chemical were
considered as a group, so that a single positive response was
taken to mean that a chemical was a potential threat. Contrary
to what had been feared, pooling the test results in this way
did not make every pesticide a potential threat. Seventeen of
the 39 chemicals were uniformly negative. The pooling of test
results was thus shown to be a practicable method for discrim-
inating between genotoxic and nongenotoxic compounds,
For the 22 compounds that registered at least one positive
response, fol!ow-up testing has been started on a case-by-case
basis. Compounds that gave a positive response in only one of
the tests are being retasted to verify the positive results, and
certain chemicals that were not originally tested in Drosophila
and mice are now being tested in those systems. Other Phase
Two tests are also being used to look for the potential of pesti-
cides to cause oncogentc transformation and chromosomal
effects.
Early results from the pesticide testing program have contrib-
uted to an increased understanding of the sensitivity and
adaptability of short-term tests. The program has also provided
valuable preliminary genotoxicity data which have been used
by the EPA's Office of Pesticide Programs to help assess the
chronic hazards associated with the various pesticides.
Once Phase Two and Phase Three test results for the 39 pesti-
cides are available for verification and validation purposes,
scientists will be able to determine the concordance (or degree
of agreement} between the various short-term tests, as well as
their accuracy in detecting potentially mutagente or carcino-
genic pesticides. This information should enable scientists to
specify which of the tests studied are most suitable for use
with pesticides and other biocidai materials. In addition, Phase
Three test results will provide the authoritative information
needed for pesticide hazard assessment.
-------
Exhaust
The increasing use of diese! engines in automobiles has
prompted EPA concern about the concomitant rise in atmo-
spheric levels of diesel exhaust and the potential health haz-
ards this may create. Diesel exhaust is a complex mixture of
thousands of different chemicals, most of which have never
been identified chemically.
To isolate all the substances in diesel exhaust and examine
each one individually for health effects would be prohibitively
difficult, expensive, and time-consuming. The EPA is therefore
using a phased testing strategy in combination with chemical
analysis to identify which portions (or fractions) of diese)
exhaust are hazardous and should be subjected to further
ehemicaf analysis and testing. Fractions that appear to be rela-
tively nontoxic are being assigned a low priority for further
analysis, so that limited resources will not be devoted to sub-
stances less likely to threaten human health.
The testing strategy for the diese! exhaust program has pro-
ceeded in several steps, Diesel exhaust was first examined
using Phase One core battery tests. One group of tests — the
rtitcrobia! mutagenicity tests — registered positive, indicating
that diesel exhaust may contain mutagenic {and possibly car-
cinogenic) chemicals and therefore should receive further
testing.
To get a better idea of which portions of the exhaust were
potentially hazardous, chemical procedures were used to
divide (or fractionate) the exhaust into several distinct frac-
tions. Each fraction was subjected to Phase One microbiat
mutagenicity tests. The most mutagenic of these fractions
were fractionated further, and the resulting subfractions were
then tested in the rnicrobial mutagentcity tests. Several of
these proved to be mutagenic.
To con%m the activity indicated by the Phase One tests,
Phase Two tests were performed on the positive fractions and
subfractions. When these tests registered positive as welt, it
was decided to perform Phase Three whole animal tests on the
diesel exhaust to further confirm earlier results and to deter-
mine the magnitude of the health threat,
-------
Program Applications 19
Mutagenic subtractions are currently being analyzed to deter-
mine their chemical composition. The pure compounds that
are identified by these procedures will be tested individually in
mutagenicity tests in an effort to pinpoint precisely which
chemicals in diesel exhaust are potentially hazardous.
Although the diesel exhaust program is still in progress, it has
already produced some very important and useful results. St
has demonstrated two important applications of short-term
tests; they can be used with complex mixtures to indicate
whether or not a hazard may exist, and they can also be used
to pinpoint which fractions of the complex mixtures are
responsible for the observed mutagenicity, {These two features
may facilitate hazard assessment of common environ mental
pollutants that are camptex mixtures,} In addition, the results
of the diese! exhaust program have contributed to the EPA's
preliminary assessment of the potential health impacts that
may be associated with the increased use of diesel engines.
-------
Research Trends
THE EVALUATION OF TOXIC AND GENOTOXSC EFFECTS
Effect
Nonspecific Toxicrty
Wutagenicity
Carcinogenicity
^v Measured
^V
Organism ^v
c
,2
8
a
,_
o
1
c
o
o
Risk
\ Assessment
Bacteria
Yeast
Mammalian
Cells
Plants
Insects
Mamrnafs
Humans
.._
Activation Germinal Somatic _
T -in: T i-, DNADamaoe . it;.n P.' n.---!-in Tumor
Capacity and Repair Ger)(J Cjlrornosoma! Qene Chiomosom^r Formation
• • •
• • * o
m m m
9 W IP w w w w
• • o • • • • o
• « o • * o
• • ••••• •••
• • • o • o • •
Effect that may be developed in the future
-------
Trends
21
Research on short-term tests is currently proceeding in several
areas. Existing test data are being compiled and evaluated in
order to document the accuracy and utility of short-term tests.
At the same time, laboratory research is being conducted to
refine and improve the tests, and several important applica-
tions of short-term tests are being explored. Some current
areas of research are described below.
Streamlining a Short-Term
Currently, there is no Phase One test for the ability of a chemi-
cal to cause oncogenic transformation, because the short-term
tests that are presently used to detect this effect take too long
to serve as rapid screening tests. Generally, 6 weeks must
elapse after chemical exposure before the cell system exhibits
the usual signs of transformation (abnormal growth patterns)
that are visibfe under a fight microscope
The possibility of shortening the time for detection of transfor-
mation is currently being explored. Investigators have noticed
that transformed cells exhibit a strikingly different surface from
normal cells when viewed under a scanning electron micro-
scope {SEM} (Figure 5). Research is currently under wav to
investigate how soon these changes occur after chemical
exposure, and whether they are as accurate an indicator of the
transformed state as abnormal growth patterns. If the SEM
can detect transformation quickly and accurately, oncogenic
transformation may become usable as a rapid prescreening test
in the first phase of testing. This would help make Phase One
testing a more comprehensive indicator of potential carcino-
gen icity.
A, Normal Cells (3000x)
B. Transformed Cells (2000k)
C. Single Transformed Cell (4000x)
Figure 5.
-------
Research Trends
Tackling the Sample Size Problem
Each short-term test requires a certain amount of the chemical
in order to adequately test for genotoxic potential. If too small
a concentration is used, the chemical's toxic properties may go
undetected. This sample size requirement can make short-term
testing impossible if the substance to be analyzed is in short
supply. Gas samples are particularly troublesome since the
materials in them are highly diluted, and it is often difficult to
obtain a concentrated sample for analysis. The volume of
material available for testing purposes may be reduced even
more if the sample must be broken down chemically into
smaller and smaller fractions in order to pinpoint the hazardous
components (see "Diesel Exhaust" on page 18).
Research is currently being conducted to modify short-term
tests so that they will require smaller sample amounts,
Already, the standard Ames test for carcinogens and muta-
gens (see Figure 6} has been modified so that effectively one-
fifth as much sample is required. By using a "well test" proce-
dure, a one* milligram sample can be used to make four or five
different measurements that would each normally require milli-
gram quantities. Scientists are also attempting to develop
innovative chemical fractionation techniques that will allow
larger samples to be derived from complex mixtures.
IkMlS SALMONELLA,-'MI€HOSOMt MUTAGEKIC1TV TtSI
Figute 8.
Developing Human Cell Systems
The most obvious system to use for detecting potential human
carcinogens and mutagens is one that employs human cells.
Many attempts have been made to develop such a system, but
only recently has the work proven successful In 1978, a group
of scientists showed that cell lines derived from human
foreskin could be cultured and used to detect oncogenic trans-
formation. At about the same time, other human cells were
shown to be amenable to studying mutagenic effects. These
advantages prompted the EPA to initiate research into human
cell systems for mutagenesis and oncogenic transformation.
Research is proceeding in several areas. Techniques are being
developed to allow the continued propagation of human cells
under culture conditions, and different typos of human cells
are being tested for their ability to survive in culture. Along
these lines, scientists are exploring the possibility of developing
model systems using cefls from organs that are often the sites
of cancer (i.e., lung, intestine, and prostate). If this effort is
successful, such model systems could provide a more accurate
means of predicting the effects of chemicals on specific sites
in the human body.
Multi-Effect Tests
Most short-term tests that use mammalian cells have been
designed to measure only one type of biological effect. EPA
scientists are currently working to develop mammalian cell sys-
tems that will be able to detect two or more different kinds of
effects. Various mouse and human cells that may be capable
of detecting both mutation and oncogenic transformation are
being explored, and special attention is being focused on a
Chinese hamster ovary (CHO) cell system that may be able
to simultaneously test for four different types of effects (gen-
eral toxicity, gene mutation, chromosomal effects, and DNA
damage),
Development of multi-effect test systems has important ramifi-
cations, Such systems would enable scientists to gain greater
insight into the relationships between the various effects being
tested. By eliminating the differences that may result from the
use of different cell systems, it should be possible to see if
there is some underlying connection between the various
effects that a chemical may have. Enhanced understanding of
the mechanisms underlying the genotoxic effects measured by
multi-test systems could eventually lead to major refinement of
testing strategies and better understanding of the significance
of test results.
-------
Improving Metabolic Activation
Metabolic activation is such an important step in short-term
tests that deficiencies in activation systems are often sus-
pected as the cause of inaccurate test results. EPA scientists
are currently exploring means of improving existing systems
and developing more potent methods of activation. Research
is proceeding in several areas. Existing activation systems are
being biochemically analyzed in order to gain a better under-
standing of the important enzymes involved. At the same time,
scientists are working to develop test systems using organisms
and cells known to have a high metabolic capacity. Such sys-
tems could hopefully provide their own metabolic activation
and thus not require the addition of enzymes from another
source (exogenous activation}.
Researchers are also exploring the possibility of using enzyme
inducers to activate mammalian cells in culture. Currently,
inducers are injected into whole animals to raise their enzyme
levels, and various organs {usually the liver) are then ground
up and applied to test systems. It is hoped that direct applica-
tion of inducers to mammalian cell cultures will raise the cells'
enzyme levels and eliminate the need for exogenous activation.
Such activated mammalian cells could also be used to acti-
vate another cell system by the feeder layer technique (see
Figure 3),
The Gene-Tox Program
Since short-term tests were first devefoped more than 10 years
ago, a significant amount of-information has been generated
concerning their accuracy and reliability. The purpose of the
Gene-Tox' program, which is being directed by the EPA's
Office of Toxic Substances, is to compile all the available infor-
mation on short-term tests and to provide an up-to-date evalu-
ation of their status.
Twenty-seven different short-term tests have been scheduled
for the initial evaluation. For each test, expert scientists drawn
from government, industry, and academia will review and eval-
uate the available information. Answers will be sought to such
questions as: How accurately can the test system detect car-
cinogenic and/or mutagenic chemicals? Is the accuracy greater
with certain classes of chemicals? Can the actual magnitude of
•Derived from the program's official title: An Evaluation of the Current Sfe-
Ms of Sioassays in Genetic Toxicology,
the hazard to human health associated with a chemical be pre-
dicted? in addition to addressing these questions, investigators
will attempt to determine which groupings of tests are most
suitable for specific purposes; for example, which tests would
together most effectively test for a specific type of genetic
damage. This information will be extremely valuable in design-
ing testing strategies that are both accurate and cost-effective.
An important part of the Gene-Tox program will be the devel-
opment of a computerized data management system for the
storage and analysis of the significant data. This will facilitate
the review process by enabling the information to be readily
accessed according to such parameters as chemical, chemical
class, type of test system, or organisms used. The computer-
ized data file will also permit future test data to be added as
they become available, thereby providing an accessible and up-
to-date record of all relevant information on the tests.
In addition to providing a state-of-the-art evaluation of short-
term tests, the Gene-Tox program will also accomplish another
important task. It will hefp to identify aspects of short-term
tests that require further development and validation. Such
information will be of great value in designing and guiding
future research programs.
Plants and Humans
Scientists have recently discovered that plants have enzymes
similar to those found in animals. This intriguing finding has
inspired two related avenues of research. One seeks to exploit
plants as a source of metabolic activation for short-term tests.
The other is concerned with the possibility that chemicals
applied to plants may be converted by plant enzymes to new
chemicals that may be toxic to hurnans ingesting the plants.
So far, this research has produced some exciting, but possibly
disturbing findings. Tissues taken from plants treated with pes-
ticides were ground up and applied directly to the Arnes assay
with no additional metabolic activation. While the pesticides
themselves had shown no mutagenic activity in the Ames
assay (either with or without animal-derived metabolic activa-
tion), the pesticide/plant mixture showed weak positive activ-
ity. Such a finding suggests that plants may have a previously
unrecognized capacity to transform substances into potentially
mutagenic and carcinogenic compounds.
-------
Trends
24
MUTAGEN1CITY OF AMBIENT AIR MEASURED BY TRADESCANTIA IN THE MOBILE MONITORING VEHICLE
5 -
Plants Untreated
Plants Exposed
to Ambient A»r
Average for all
untreated
plants
State
Major locai
pollution
sources
Month
of test
NJ | WV
T
Petroleum & 1 _,
Automotive ] cher"»-ak
OCT i MAR
At,
Steel
APR
LA
Petroleum
MAY
TX
Petroleum
JUL
CA
Automotive
and Petro
chemical
SEP
UT
Copper
OCT
AZ
Clean
Air
DEC
Snu-fct. L A ,"chai«T «'. a , pa. 419 MO n -"IT- '«. -"wi r« S'V1 ' / erm Bioastays in the Frjt notation
•ad Analysis at Cennp/ert trivirotwnanta! Mixture (Health fc!(ests Rp.warch Labontory, U S. EPA,!
BiMIM^^™ftmiM»l«'B)KaBMBB!!Bai«SiBmH^^
!978i
Figure
Monitoring
One exciting short-term test application currently being
explored by the EPA is the use of the Tradescantia plant sys
tern for monitoring on-site air quality. Traditionally, air has
been monitored using chemical methods to determine the con-
centrations of specific known hazardous chemicals. The Tra-
cfescantis plant test may provide an improved means of
measuring overall air quality fay allowing ambient air to be
screened for the presence of mutagenie chemicals.
The concept behind the Tradescantia test is a simple one.
When exposed to gaseous mutagenie chemicals, the stamen
hair cells of the Tradescantia flower mutate from blue to pink.
By counting the stamen hair cells that change color, a measure
of potential mutagenieity can be obtained. Tradescantia is par-
ticularly suited to field testing because it can tolerate a broad
range of conditions and, unlike some other short-term tests, it
does not require special sterile conditions,
In a collaborative research program,1 a mobile laboratory for
exposing Tradsscantia to ambient air has been tested success-
fully in eight different natural and industrial environments,
including sites in New Jersey, California, Texas, Utah, and the
Grand Canyon in Arizona. The mutagenieity findings are pre-
sented in Figure 7. Further research will have to be performed
to gain a better idea of the potential and limitations of the Tra-
ffescantia test, but the results obtained so far suggest that the
test can be used as a sensitive and rapid indicator of ambient
air quality.
'Involving Srookhaven National Laboratory, the National institute of
Environmental Health Sciences, and the Environmental Pfottction Agency.
-------
Perspective
One of the highest priorities of the EPA Research and Develop-
ment Program is the protection of human health through the
identification and control of toxic substances. Short-term tests
have been singled out for intensive research and development
because they are potentially valuable tools for achieving this
goal.
Many types of short-term tests are being developed, but this
document has focused on short-term tests that attempt to
detect a chemical's potential to cause cancer and genetic dis-
ease. These disorders are among the most devastating that a
chemical may cause, and short-term tests in this area have
consequently received great attention. This attention has been
rewarded by highly encouraging results.
Our advanced industrial economy uses and produces thou-
sands of chemicals, most of which did not even exist until the
past several decades. To reduce the threat of these substances
to human health, scientists must be able to discriminate toxic
materials from those that are nontoxic. This is an awesome
task, but not an impossible one if short-term tests fulfill their
early promise for rapid and effective detection of potentially
hazardous chemicals.
-------
Appendix: How Effects are Measured in Various Short-Term Tests
DMA Damage in Bacteria (Pol A test, rec test)
Two strains of bacteria are used that are identical except In
their ability to repair DNA damage; one strain can repair dam-
age while the other cannot. Both strains are exposed to the
test substance, and the extent to which cells are killed is
measured"for each. If the repair-deficient strain has a greater
degree of cell killing, DNA damage is assumed to have
occurred.
DNA Damage in Yeast (MJtotic recombination, mitotic gene
conversion, or mitotrc crossing over)
Special strains of yeast cells are used to test for these effects.
When the cells change color from white to either pink or red,
DNA damaging potential is indicated.
Gene Mutation in Bacteria or Fungi (Ames test, WP2 assay,
yeast assays, and others)
Special strains of bacteria are used which cannot grow without
a nutritional supplement. Certain types of mutations will permit
these bacteria to grow in unsupplemented media, By treating
the cells and then seeing if they can grow in unsupplemented
media, mutagenicity can be measured, Distinguishing mutated
bacteria from nonmutated bacteria is not necessary using this
procedure, since only mutant celts can grow and form visible
colonies.
DNA Damage in Mammalian Cells ^Unscheduled DNA
Synthesis and Sister Chromatid Exchange)
Abnormal distribution of a DNA marker indicates
whether DNA damage has occurred. Ways of detecting this
abnormal distribution include microscopic examination and
photographic and machine measurements.
Gene Mutation in Mammalian Cells (HGPRT, TtOand Na/K-
ATPase assays)
In these systems, mutations that confer resistance to a poison
are measured. Cells are first treated with a test chemical and
then exposed to the poison. Since only mutant cells can sur-
vive and grow, mutagenictty can be measured simply by
observing the extent of growth in the poisonous environment,
Gene Mutation in Plants {Tradescuntia and maize waxy locus)
Mutations in these plants are detected by looking for color
changes in the stamen hairs or pollen grains. In Tradescantia,
mutation causes the stamen hairs to change from blue to pink.
In maize, mutated pollen grains can be detected by the purple
color they acquire when they are treated with iodine.
Chromosomal Effects in Isolated Cells or Whole Organisms
{Cytogenetics assays)
Treated ceils (or ceils from treated organisms} are stained and
then examined under the microscope for various chromosomal
abnormalities. Lost, broken, or disarranged chromosomes indi-
cate genotoxicity.
Oncogenic Transformation (Transformation assays)
When certain types of mammalian cells are treated in vitro
with carcinogens, they undergo cancer-tike transformation. If
these ceils are injected into appropriate experimental animals,
tumors will appear. Most frequently, transformed cells are dis-
tinguished by their unusual growth patterns in culture, such as
abnormal piiing-up and disorientatton of cells.
Micronucleus Test
Animals are treated with a chemical, and their red blood cells
are removed, stained, and examined under the microscope. If
small fragments of the genetic material (micronuclei} are
observed, chromosomal damage is indicated. Normal red
blood ceils will not contain any genetic material or fragments
of genetic material.
DrosQphila melanogaster (Sex-linked recessive lethal test for
gene mutations; nondisjynction and heritable transiocation
assays for chromosomal effects!
Drosophila have a variety of "marker" traits that can be used
to signal whether gene mutations or chromosome disturbances
have occurred. In general, Drosophila tests involve treating
specially "marked" male or female flies with a substance, mat-
ing them, and then observing whether their offspring have cer-
tain specific features, such as unusual eye color or shape.
Depending on the test, genotoxic events can be indicated
either by the presence or the absence of a specific feature in
the offspring.
-------
Appendix: Representative Short-Term Tests for Genotoxity
Type of Test
Specific Test
DMA Damage in Microbes
DMA Damage in Mammalian
Cells
Gene Mutation in Bacteria
and Fungi
Gene Mutation in Higher Systems
Chromosomal Effects in
Isolated Cell Systems
Chromosomal Effects in
Whofe Organisms
Oncogenic Transformation
Tymor Formation
Pol A test
rec test
Mitotic recombination, mitotic
crossing over, or mitotic gene
conversion in yeast (03, D4, 05, or
D7 Assays)
Unscheduled DMA Synthesis tUDS)
Sister-chromaticl exchange fSCE)
Ames test
WP2 Assay
Yeast "forward" and "reverse"
assays
Miscellaneous
HGPRT, TK, and Na/K-ATPase
Assays
Sex-linked recessive lethal assay
Piarrt tests
In vitro cytogenetics assays
In vivo cytogenetics
Mtcronucieus test
Nondisjunction assay
Heritable translocation assay
Transformation assays tcfonai or
focus)
Mouse skin tumorigenesis
Mouse pulmonary adenoma
Rat tracheal transplant
Eschsrichia coli
Bacillus subtitis
Saccharomyces cerevisiae or
Schizosaccharomyces pombe
WI-38 strain human cells
or various rodent cells
Various celi lines or animal sources
Salmone/la typhimurium
Escherichia coli
Saccharomyces cerevisiae;
Schizosaccharomyces pombe
Aspergillus nidlulans;
Neurospora crassa
L5178Y mouse lymphoma cells;
Chinese hamster ovary cells (CHO);
Chinese hamster lung cells (V-79)
Drosophila metanogaster
Tradescantia; maize waxy (OCLIS
WI-38 strain human cells;
Chinese hamster ovary cells (CHO)
Various rodent species
Various rodent species
Drosophiia me/anogaster
DrosopMa mefanogaster
Syrian hamster embryo celfs
-------
Glossary
acute effect
Ames assay
bioassay
carcinogenic
chromosome
chromosome aberrations
chronic effect
complex mixtures
(deoxyribonucleic acid)
enzyme
epidemiology
neg
false positive
fraction^'ion
gene
gene mutation
genetic material
genotoxie
germrnal cell
hazard assessment
a health effect of short duration that is
usually reversible
a weli-known short-term test that mea-
sures a chemical's ability to cause muta-
tions in a specially engineered strain of
the bacteria Salmonella typhimurium
a test to determine the effect of a chemi-
cal on a living organism
able to cause cancer
a form of DNA organization found in
higher cells and organisms
changes in the number, shape, or struc-
ture of chromosomes
a prolonged health effect that may
involve irreversible change or damage
a grouping of several different chemicals
a large molecule that contains the
genetic information responsible for ceil
growth, function, and reproduction
a protein that acts as a catalyst to allow
a specific chemical reaction to take place
in a ceil
the science of correlating exposure to a
substance with the appearance of a spe-
cific disease or other effect in a human
population group
a test result which indicates that a
chemical is harmless when it is actually
hazardous
a test result which indicates that a
chemical is hazardous when it is actually
harmless
the process of chemically separating a
complex mixture into a series of simpler
mixtures (fractions)
a portion of DNA that directs the forma-
tion of a single product
a mutation in a single gene
see DNA
able to damage genetic material
a reproductive cell (i.e., sperm, egg)
the evaluation process for determining i"
a substance is hazardous to humans
-------
Glossary
heritable
in vitro
in vivo
indicator system
malignant
metabolic activation
metabolism
metabolite
microbes
mutagenic
mutation
oncogenic
oncogenic transformation
procarcinogen
prorrtutagen
protein
somatic celt
stamen hairs
target cells
toxic
transformed
capable of being passed from one gener-
ation to another
pertains to a procedure that takes place
in an artificial medium (lab equipment)
without the use of live animals. Literally
means "in glassware"
pertains to a biological reaction or test
which occurs within the body of a live
animal
a cell or organism that shows (or indi-
cates) a specific effect
refers to the cancerous cells or tumors
that may grow, proliferate, and eventu-
ally kill the organism
the process whereby an inactive material
is changed into an active one {in the
context of short-term testing, this in-
volves the conversion of a procarcino-
gen to a carcinogen or a promutagen to
a mutagen}
the physical and chemical processes in
an organism which transform chemicals
into simpler or more complex forms
a product of metabolism
microorganisms such as bacteria or yeast
able to cause mutations
a stable change in the genetic material
able to cause tumors
a cancer-like change that can be brought
about in isolated mammalian cells by
chemical treatment
a substance which is converted into a
carcinogen by an organism's metabolic
processes
a substance which is converted into a
mutagen by an organism's metabolic
processes
a large biological molecule essential for
many cell structures and functions
any nonreproductive cell in a multicellu-
lar organism
the part of the flower that produces
poilen
isolated cells or cells within an organism
which react in a specific manner to a
toxic chemical or other stimulus
able to produce an adverse effect
see oncogenic transformation
-------
This project was sponsored by the Environmental Research
Information Center {ERIC}, Cincinnati, and was directed by
Clarence A. demons; Douglas Williams was project officer.
The report was written by Anne Trontell with the assistance
of Jan Connery, both of Energy Resources Co, inc., Cambridge,
Massachusetts.
Michael Waters, Stephen Nesnow, and Joellen Huisingh of
the Health Effects Research Laboratory, Research Triangle
Park, NC{HERL-RTP), provided technical assistance in the
development of the document. Technical reviewers were
Kirtay Campbell, Michael Pereira and John Orthoefer, HERL-
Cincinnatr; James Park, ERIC; Jack Keeve, Health Effects
Division, Washington, DC; and Richard Laska, Technical In-
formation Office, Washington, DC, Photographs in the text
were provided by Ken Muse, North Carolina State University,
Raleigh, NC.
Comments or questions regarding this report should be
addressed to:
Dr. Michael Waters
Genetic Toxicology Program
Health Effects Research Laboratory
Research Triangle Park, NC 27711
Area Code(919)541-2537
This report has been reviewed by the Health Effects
Research Laboratory, U.S. Environmental Protection Agency,
Research Triangle Park, NC, and approved for publication.
Approval does not signify that the contents necessarily
reflect the views and policies of the U.S. Environmental
Protection Agency, nor does mention of trade names or
commercial products constitute endorsement or recom-
mendation for use.
U.S. GOVERNMENT PRINTING OFFICE I979-S33-Z32
-------
------- |