U.S. DEPARTMENT OF COMMERCE
National Technical Information Service
PB-260
SUBSTITUTE CHEMICAL PROGRAM - THE FIRST YEAR
OF PROGRESS, PROCEEDINGS OF A SYMPOSIUM
VOLUME II. TOXICOLOGICAL METHODS AND
GENETIC EFFECTS WORKSHOP
ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON, D, C,
1975
-------
SUBSTITUTE
CHEMICAL
'ROGRAM - THE FIRST YEAR OF PROGRESS
PROCEEDINGS OF A SYMPOSIUM
Volume II
lexicological Methods and Genetic Effects Workshop
UNITED STATES ENVIRONMENTAL PROTECTION AGENCY
OFFICE OF PESTICIDE PROGRAMS AND
OFFICE OF RESEARCH AND DEVELOPMENT
WASHINGTON, D.C.
JULY 30 - AUGUST 1, 1975
J
-------
BIBLIOGRAPHIC DATA
SHEET
1. Ilepor'
-014
3. Recipient's Accession No.
PB-260 415
4. Title and Subtitle
Progress .
Substitute Chemical Program — The First Year of
Proceeding of a Symposium. Volume 31,
Toxicological Methods and Genetic Effects Wbrksht
5. Report Date July 30 _
August 1( 1975
7. Author(s)
8. Performing Organization Repi.
No.
'. Performing Organization Name and Address
U.S. Environmental Protection Agency, Office of Pesticide
Programs and Office of Research and Development, Washington,
D.C. 20460
10. Project/Task/Work Unit No.
11. Contract/Grant No.
12. Sponsoring Organization Name and Address
13. Type of Report & Period
Covered
9
15. Supplementary Notes
J6. Abstracts jhis Volume covers the Toxicological Methods and Genetic Effects Workshop
17. Key »ont* «od Document Analysis. I7o. Descriptor*
toxicological methods
genetic effects research
progress session intro-
duction
inhalation toxicology of
substitute chemicals
Inhalation toxicology
toxicity to mammals of
small particle aerosols of
"cJear^.oJghe^r^sisr.vir.usA
I7b. I
pesticide metabolism
toxicological evaluations
acute LD50 studies
2-AAF as a model .compound
age sensitivity to the
''.chemical carcinogen
2- acetylaminofluorene
benchmark toxicity
pharmacokinetic modeling
residues in human milk
blood lipoproteins, arteries,
cardiac muscle — effects
of pesticides
fetal tissue analysis
in vitro and invivo studies
detection of chemical-induced
mutation
epidemiology *•- cancer mortality
and pesticide use in U.S.
cadodylic acid on prenatal
development of rats and mic»
17«. COSATI Field/Group
18. A»«ilabilay Statement
19.. Security Class (This
Report)
UNCI-AS
NCI-ASSIFIFD
urity Class (Thi
20. Security Class (This
Page
FORM KTIS-3J (Rev. ioos> ENDORSED BY ANSI AND UNESCO.
THIS FORM MAY BE RKI'KODUCKD
USCOMM-OC
-------
SUBSTITUTE
CHEMICAL
'ROGRAM -THE FIRST YEAR OF PROGRESS
PROCEEDINGS OF A SYMPOSIUM
Volume II
Toxicological Methods and Genetic Effects Workshop
SHERATON-FREDERICKSBURC MOTOR INN
FREDERICKSBURC, VIRGINIA
X^JULY 30 - AUGUST 1, 1975 ll
J
-------
The Proceedings of the SUBSTITUTE CHEMICAL PROGRAM— THE
FIRST YEAR OF PROGRESS are published in four volumes. Volume I
contains speeches and discussion from the Plenary Session, the agenda,
and,lists of participants and speakers. Volumes n, III, and IV cover.the
Toxicological Methods and Genetic Effects Workshop, the Ecosystems/
Modeling Workshop, and the Chemical Methods Workshop, respectively.
-------
TABLE OF CONTENTS
Wednesday. July 30. 1976
TOXICOLOGICAL METHODS AND GENETIC
EFFECTS RESEARCH PROGRESS SESSION
INTRODUCTION
Dr. August Curley and
Dr. Lamar B. Dale, Jr 1
INHALATION. TOXICOLOGY OF SUBSTITUTE
CHEMICALS
Dr. James T. Stevens. 3
INHALATION TOXICOLOGY
Dr. Gordon W. Newell 5
STUDIES ON TOXICITY TO MAMMALS OF
SMALL PARTICLE AEROSOLS OF NUCLEAR
POLYHEDROSIS VIRUS (NPV) PESTICIDES
Dr. James T. Stevens 25
METABOLISM OF PESTICIDES
Dr. Ronald L. Baron 27
TOXICOLOGICAL EVALUATIONS
Dr. Ralph I. Freudenthal 37
TOXICOLOGICAL RESEARCH: ACUTE
LD50 STUDIES
Dr. Larry L. Hall 41
2-AAF AS A MODEL COMPOUND
Dr. Thomas J. Haley 49
A STUDY OF AGE SENSITIVITY TO
THE CHEMICAL CARCINOGEN
2- ACETYLAMINOFLUORENE
Dr. David L. Greenman 57
BENCHMARK TOXICITY
Dr. Lawrence Fishbeln 67
PHARMACOKINETIC MODELING OF
SELECT PESTICIDES
Dr. John F. Young 75
Preceding page blank
-------
Thursday. July 31. 1976
PESTICIDE RESIDUES IN HUMAN MILK
Dr. Eldon P. Savage 87
EFFECT OF SUBSTITUTE PESTICIDES
ON HORMONE-DEPENDENT TISSUE
Dr. Sydney A. Shain 93
EFFECTS OF PESTICIDES ON BLOOD
LIPOPROTEINS, ARTERIES, AND CARDIAC
MUSCLE
Dr. Jack E. Wallace 101
FETAL TISSUE ANALYSIS FOR
PESTICIDE RESIDUES
Dr. Irwin Baumel J.07
RESEARCH PROGRESS: CARCINOGENIC AND
TERATOGENIC TESTS INTRODUCTION
Dr. Morris F. Cranmer 109
IN VITRO AND IN VIVO CARCINOGENIC
AND MUTAGENIC SCREEN DEVELOPMENT
Dr. Erling M. Jensen Ill
DROSOPHTLA MUTAGENESE TESTS
Dr. Ruby Allen Valencia 119
IN VITRO AND IN VIVO STUDIES OF
SELECTED PESTICIDES TO EVALUATE
THEIR POTENTIAL AS CHEMICAL MUTAGENS
Dr. Gordon W. Newell 141
UNSCHEDULED DNA SYNTHESIS TESTING
OF SUBSTITUTE PESTICIDES
Dr. Ann D. Mitchell 151
MAMMALIAN SCREENS
Dr. Gordon W. Newell 155
USE OF MUTAGENESIS TEST TO
INDICATE CARCINOGENESIS
Dr. Barry Commoner. 173
CRITERIA FOR COMPARISON OF
TERATOGENESIS PROTOCOLS
Dr. Herbert J. Schumacher 175
vi
-------
FEASIBILITY STUDY: DETECTION OF
CHEMICAL-INDUCED MUTATION BY ASSAY
OF METABOLIC CHARACTERISTICS
Dr. Harvey W. Mohrenweiser 183
FEASIBILITY STUDY: THE USE OF DIPLOID CELL
STRAINS FROM SPECIFIED MOUSE GENOTYPES FOR
USE IN DEVELOPING IN VITRO ASSAY FOR MUTAGENIC
ACTIVITY, INDUCTION, AND ANALYSIS OF GENE
MUTATIONS IN MAMMALIAN CELL LINES
Dr. Daniel A. Casciana , 191
EPIDEMIOLOGY OF PESTICIDES: CANCER MORTALITY
AND PESTICIDE USAGE IN THE UNITED STATES
Dr. William F. Durham 195
EFFECTS OF CACODYLIC ACE) ON THE PRENATAL
DEVELOPMENT OF RATS AND MICE
Dr. Neil Chernoff 197
ASSESSMENT OF SUBTLE AND DELAYED
EFFECTS OF SUBSTITUTE CHEMICALS
Dr. Daniel A. Spyker 205
vtl
-------
TOXICOLOGICAL METHODS AND GENETIC EFFECTS
RESEARCH PROGRESS SESSION
INTRODUCTION
Dr. August Curley*
and
Dr. Lamar B. Dale, Jr. **
The Research and Special Studies as outlined In this progress section are being con-
ducted cooperatively under the Substitute Chemical Program by the Office of Research
and Development and the Office of Pesticide Programs, Criteria and Evaluation Division.
These studies are conducted as:
1. An Intramural research effort by the Office of Research and Development, En-
vironmental Research Center, Health Effects Research Laboratory, Environmental
Toxicology Division, Research Triangle Park, North Carolina (formerly/the National
Environmental Research Center, Pesticides and Toxic Substances Effects Laboratory);
2. As an extramural contractual research effort with the several Individual labora-
tories as Indicated In the program; and
3. Through an Interagency research program at the National Center for Toxicologlcal
Research, Jefferson, Arkansas, jointly sponsored by FDA and EPA.
The major areas of research Included In this scope of the cooperative Substitute
Chemical Program are:
1. Basic short- and long-term toxic effects research which Includes methodology
development and evaluations; and
2. In vivo and jn vitro screening tests to address the problems In carclnogenesis,
mutagenesis, and teratogenesis associated with possible human exposure to these
compounds.
*Office of Research and Development, Environmental Research Center, Health Effects
Research Laboratory, Environmental Toxicology Division, Research Triangle Park
**0ffice of Pesticide Programs, Criteria and Evaluation Division, EPA
-------
This research and special studies effort is the primary responsibility of the Office
of Research and Development and supports the internal review process of the Substitute
Chemical Program, including the Cooperative Liaison Studies effort between other
Governmental agencies, pesticide manufacturers, and academia as accomplished through
integrated planning by the Office of Pesticide Programs/Criteria and Evaluation Division
and Office of Research and Development. This research effort responds to specific
deficiencies in the scientific data base as identified by the review process. Through
these reviews, new toxicity data are generated where required; testing, screening
techniques, and protocols are developed to evaluate certain toxicological aspects of
potential substitutes; and behavioral, pulmonary-hepato metabolic alterations, includ-
ing hormonal effects, are determined.
Through the implementation of these studies, the Substitute Chemical Program ex-
pects to determine the suitability of these chemicals which now or in the future may
serve as replacements for those major or minor use pesticides that are currently under
internal review, or those which have been suspended, granted limited uses, cancelled,
or are in litigation to be suspended or cancelled. In short, the overall Intent of this
program is to assess the human health hazard associated with the use of these com-
pounds in pest control.
This session, which is called the Toxicological Review Session, has been divided
into several subsessions: namely, Inhalation Studies; Acute, Subacute and Chronic
Studies; Epidemiological Residue Studies; Carcinogenesis, Mutagenesls, and Tera-
togenesis Studies. This afternoon we will have research progress reports on the in-
halation studies and the acute, subacute, and chronic studies. Tomorrow Dr. Cranmer,
Director of the National Center for Toxicological Research at Jefferson, Arkansas, will
chair the session on carcinogenesis, mutagenesis, and teratogenesis testing. The
evening session, which is a continuation of the studies in carcinogenesis, mutagenesis,
and teratogenesis, will be chaired by Ms. Jean Pull Lam. We will follow the program
as outlined.
-------
INHALATION TOXICOLOGY
OF SUBSTITUTE CHEMICALS
Dr. James T. Stevens*
NOT AVAILABLE
FOR
PUBLICATION
*National Environmental Research Center, Research Triangle Park, North
Carolina »
3
-------
INHALATION TOXICOLOGY
Dr. Gordon W. Newell*
A thorough Investigation of Insecticides is being conducted by EPA with a view
to determining their effects on health and the environment. From a review of the
1 Iterature and laboratory studies, it is anticipated that alternate chemicals for
pest control might be recommended for primary use In place of others currently
used.
To support these aims the Toxicology Department at Stanford Research Institute
was commissioned by EPA to investigate the acute toxicity of five organophosphate
Insecticides in experimental animals. The modes of application Included the intra-
gastrlc, Intravenous, percutaneous, and Inhalation routes. Of necessity, this
program was divided into five phases: 1) generation of resplrable particles;
2) analytical methods to determine the aerosol concentration In the exposure
chamber; 3) acute toxicity study In rats; 4) gross and histopathological changes
of animal organs; and 5) the analysis of animal tissue for Insecticide content.
The compounds studied In this program were technical-grade pesticides: Azodrin,
Guthion, methyl parathlon, parathlon, and Thimet.
Phase 1 - Generation of Resplrable Particles
This phase had as Its primary aim the generation of a reliable and reproducible
aerosol of the test pesticides. In order to obtain the maximum biological effects,
the mass mean particle size was to be In the range of 0.5 to 3 microns, and consti-
tute approximately 90 percent of the participates. Of the remainder, those particles
less than 0.5 micron and larger than 4.0 microns should be less than 10 percent.
Only under these conditions could maximum lung penetration be achieved.
These aerosols can be generated most effectively by either pneumatic or ultra-
sonic means. Both techniques have been used In this study.
*Department of Toxicology, Stanford Research Institute
* Preceding page blank
-------
Concentrations of the insecticides were to be maximized at 200 mg/m3 and any
desired ratio dilutions were to be controllable through the use of critical flow
orifices and the regulation of air pressure in the upstream manifold.
Both the insecticide concentration and the particle size distribution were to be
monitored. Insecticide concentration is discussed later, while particle size distri-
bution has been monitored by one of three techniques, depending on the vapor
pressure of the candidate: 1) by optical means — particles captured on a non-
wetting surface are photographed and counted with a minimum time delay; 2) by
physical means — surface impaction; and 3) electronically — using a cryogenic
stage within a scanning electron microscope.
With the SRI exposure chambers it is possible to present the test animals,
radially situated, to a down draft of the aerosol, in such a way that all 10 animals
breathe air within a 16-square-inch area. By so doing, variations in aerosol
density due to abnormal turbulent deposits within the air stream are minimized.
Description of Aerosol Exposure Facilities
General Description
The aerosol exposure system provides for the simultaneous exposure of groups
of animals to successively lower concentrations of a test material. It permits
predetermined and independent control of the concentration in the first chamber
and the ratio of concentrations in successive chambers. The system is composed
of five component elements: 1) a generator, 2) a primary dilution chamber,
3) a concentration ratiolng system, 4) the exposure chambers, and 5) a disposal
system. The system is shown in Figure 1.
Initial Dilution
The pesticide is introduced to the dilution chamber in vapor or aerosol form by
an ultrasonic or pneumatic generator. The aerosol concentration in the first chamber
(desired maximum exposure concentration) is controlled by the amount of dilution
air metered into the dilution chamber.
-------
Figure 1
Condensed Schematic Diagram of Toxicant Exposure System
I Secondary Dilution System l^ Compressed Air
M) IM ) ( M
14 l/minj 14 1/min.I 14 l/mln.
Dilution
Air
50-500
l/min.
Dilution
Chamber
Toxicant
Generator
Overflow Air
6 1/min.
39-452 1/min.
/ . Generator Air
Exhaust Air
3-40 l/mln.
>
/ \
f \
f \
f \
s
Metering Exhaust System
59-96
>
l/min.
t
To Atmosphere
98-548 l/mln.
M " Crltioal-now orifice meter
-------
Some of the diluted stream goes to the ratioing system where a portion is drawn
off into the first exposure chamber. The balance of the stream entering the ratioing
system is successively diluted with clean air. A portion of each diluted stream is
passed to an exposure chamber before the next dilution. Thus, each chamber
receives a successively lower toxicant concentration. The ratio of concentration
between successive chambers is determined by the flow rate leaving the ratioing
system relative to that entering and leaving the ratioing system. Thus, the ratio
R can be changed by simply changing the exhaust air rate from the ratioing system,
which is actually done by changing the size of the orifice used to meter the exhaust
air rate.
All flows are metered by critical flow orifices. Secondary dilution rates are
metered by flow from a compressed air main to essentially atmospheric pressure.
Chamber flows and the ratioing system exhaust are metered by flow from substan-
tially atmospheric pressure to high vacuum ( 4. 1/3 atmosphere).
Although the dilution ratio is held constant, a different ratio between the chambers
could be established by a corresponding selection of the respective orifice plates
used for the secondary dilution air supply metering and for metering the flow through
the test chambers.
Any of the gas from the dilution chamber that is not required (or demanded by)
the ratioing system overflows directly to the disposal system through a filter.
The exhaust from the ratioing system vacuum pump also goes to the disposal
system. No general filter is used on the ratioing system, since each of the ratioing
system streams has its own filter. These filters remove the aerosol. Any aerosol
particles escaping the filter or any vapors are destroyed by passing them through a
gas-fired incinerator. The gas rate is set to adjust the outlet Incinerator tempera-
ture from 1400 to 1500° F. The total gas residence time in the incinerator is hi the
.order of one second.
8
-------
All the gases from the entire system are exhausted to the atmosphere by means
of a compressed-air operated ejector located downstream of the Incinerator.
There Is also a provision for cooling the gases leaving the Incinerator by Induced
room air.
Safety Features
System pressure Is controlled by adjusting the compressed air rate to the exhaust
ejector. The absolute value of this pressure Is not critical to the operation of the
ratiolng system. The pressure Is maintained at an arbitrary -0.5 Inch water to
avoid leakage of toxicants into the room. A warning system has also been provided
to give an alarm if the system pressure exceeds -0.2 inch water. This alarm
consists of both a flashing red light and a buzzer. The buzzer can be deactivated
by a switch.
A heating element Is provided in the dilution air feed line, and is provided to
compensate for any heat requirements for evaporating residual solvents from the
aerosol particles emanating from the generator.
Humidity of the dilution air is controlled by saturating a predetermined portion
of the primary dilution air with a dual primary dilution air metering system, an
air-water contacting system, and a cyclone entrainment separator.
The valving in the test system is so arranged that any one test chamber may be
sampled or disconnected from service during a run without influencing conditions
at any other test chamber.
Critical aspects that have been considered in the design are 1) versatility,
convenience, and reliability consistent with reasonable cost; 2) safety of the operator;
3) thoroughness of mixing of air streams; 4) avoidance of deposition of aerosols
in the apparatus; 5) ease of cleaning; 6) short equilibration time (probably less
than one minute); and 7) visibility of animals. The dimensions of the various
components have been based on a detailed quantitative evaluation of all these
-------
factors. In many cases a compromise is necessary. This is particularly true in
the case of aerosols, where thoroughness and rapidity of mixing are not compatible
with avoidance of aerosol deposition on the walls. The ducts and mixing elements
have been designed to achieve what is considered adequate mixing with an acceptable
minimal amount of wall deposition. From test results, it can be inferred that
mixing is at least 95 percent complete and that aerosol deposition is well under
5 percent.
The animal test chambers have been designed to achieve a reasonably uniform
aerosol flow across all animals with a minimum opportunity for loss of aerosol
particles. The animals are confined by wire enclosures so that they are almost
immobile with their noses exposed into the incoming aerosol stream. There is
room for 10 rats per chamber and the ring through which the aerosol flows into
which the rats' noses extend is 5.5 inches in diameter. The total volume of the
animal cages, including the stainless steel approach and discharge cones, is
approximately 30 liters. All animals are visible from the outside at all times,
since the top of the chamber is transparent. At a chamber air flow of 14 liters
per minute, the air velocity past the rats' noses averages 1.5 cm/sec.
Particle size measurements were made of the various pesticides using the
Royco Optical particle counter in conjunction with a dilution system, the develop-
ment of which was sponsored by SHI. Without the diluting system, the aerosol
concentration would have been too high (even from the last animal chamber) for
the Boyco to handle. The data points are based on the actual count obtained with
samples taken from the last animal chamber and diluted to various extents before
reaching the Royco counter. The average curve through these data gives the
size distribution of the aerosol on a number basis. The mass distribution curve
was calculated from this average number distribution curve. With the exception
of Guthion (which gives a somewhat finer aerosol), all the other materials (Thimet,
ethyl parathion, and methyl parathion) were sprayed in neat form and gave
10
-------
essentially the same size distribution. Because Guthion was sprayed as a 20
percent solution in DMSO, it is to be expected that the final average aerosol
1/3
particle size should be about (0.2) or 0.6 that of the other pesticides.
Phase 2 - Sampling of Chamber Contents
Chamber contents are sampled by taking grab samples of the material during
the animals' exposure period. The sampling train consists of a pump pulling air
at approximately one liter per minute through a particulate filter and a solvent
trap. The length of time for sampling depends on both the compound and its
concentration in the chamber.
Studies were conducted to determine that the aerosol concentration did not vary
with time or location within any given chamber, and also to verify that serial
concentration between chambers was within the hypothetical norm.
Phase 3 - Chemical Analysis of Chamber Contents
A gas chromatographic (GC) method for analyzing the pesticides was developed
using a flame photometric detector (FPD). The FPD detector was modified as
suggested by Burgett et al. (J. Chromatog. Sci. 12: 356, 1974) to eliminate
solvent flameout and to increase response linearity. For Azodrin, analyses were
performed under the following conditions:
• Column, 3 percent SE-20 (2 mm x 4') on 80/100 mesh Gas-Chrom Q
o
• Temperature, 170
• Flow rate, 60 ml/min. N0
£t
Quantitatlon was based on the internal standard method with respect to methyl
parathlon. The minimum detectable limit for Azodrin was found to be 20 ng.
Direct air sampling (5 ml) of the Chamber S (highest concentration) was unsuccess-
ful. Therefore, preweighed nucleopore filters were weighed after a timed expo-
sure to the aerosols. The filters were extracted with 1 ml of acetone, and methyl
parathion as an internal std. was added.
11
-------
Phase 4 - Acute Toxicity
Acute toxicity studies were conducted with parathion, methyl parathion, Thimet,
Azodrin, and Guthion, administered by four routes. At least f6ur effective
insecticide concentrations were used in each case.
Table 1
Administration Maximum
Route Vehicle Dosage
Oral Water or propylene glycol 5000 mg/kg
Dermal Water or propylene glycol 2000 mg/kg
Systemic Ethanol, propylene glycol 2000 mg/kg
Inhalation DMSO 200 mg/kg
All studies are carried out on Sprague-Dawley rats weighing between 220 to 250
grams. Ail animals are fasted 24 hours prior to dosing, returned to their.cages
after exposure, and watched closely for pharmacotoxic signs. All rats are held
for 14 days to determine the presence or absence of late deaths. Body weights
are determined at the 1- and 2-week levels. Food and water are freely available
after dosing. Animals that die during the holding period are necropsied, and
sacrificed animals are also necropsied. Tissues are gathered at necropsy from
all animals exposed to the two highest concentrations of insecticide, regardless
of route.
Acute Oral Toxicity: The four dosage levels are calculated on the.fasted weights
of the animals. A glass syringe fitted with a ball tip needle is used for adminis-
tration. After dosing, the animals are returned to the cage for observation.
Percutaneous Toxicity; Twenty-four hours before dosing,: rats have their backs
and-isldes clipped free of hair and are returned to their cage. Dosage-of insecti-
cide is based on the fasted weight and is applied to the backs and sides in a
suitable vehicle; runoff is not tolerated. While dosages vary, the volume added
12
-------
to each animal Is constant. A treated animal is wrapped in an impervious layer
and placed in a restrainer for 1 or 4 hours. Following this exposure, surplus
material is wiped off using cotton moistened with the diluent vehicle and the
rat is returned to its cage.
Acute Systemic Toxicityt Fasted rats are placed in a restrainer and dosed
intravenously through the dilated tail vein. The dosage rate varies, but the total
administered volume remains constant. Treated animals are returned to their
cages for observation.
Acute Inhalation Toxicity; Fasted rats of known weight are restrained in individual
cells, with only the anterior portion of the head exposed. These cells are placed
in the exposure chamber where the animals receive a 1- or 4-hour exposure to the
aerosolized Insecticide. These rats are returned to their cages with minimal
cleaning and held for observation.
Phase 5 - Gross and Microscopic Pathology
Necropsies are conducted on those rats dosed at insecticide levels which are
used in developing the LD .
ou
Histopathological assessments are routinely carried out on the lungs of rats
exposed to the high concentrations of insecticides administered via the inhalation
route. Additional histopathological investigation is performed on all tissues where
a repetitive gross abnormality has been detected.
Phase 6 - Gross and Histopathological Changes
No important gross changes have been observed. Approximately 320 H&E slides
(from 76 or 78 rats) of liver, kidney, lung, brain, and skin were examined micro-
scopically. Lesions in animals In the various treatment groups were recorded
and compared. Since the sizes of the groups were often inadequate and some
groups were not represented, our conclusions are tentative.
13
-------
Table 2: Number of Sampled Tissues
Route of Administration
Oral IV Dermal Total
Guthion 820 10
Parathion 407 11
Methyl parathion 0 16 9 25
Thimet 07 5 12
Azodrin 865 19
Total 20 31 26 77
Moreover, most lesions were of a type that is usual and spontaneous and,
hence, unlikely to be related to treatment. Nonetheless, the overall incidence
of lung lesions in rats treated with Azodrin intravenously and orally was
approximately half that in rats treated with other pesticides by the oral, intra-
venous, or dermal routes.
Lung tissues are being taken and examined from animals undergoing various
inhalation exposures, but we do not believe continuation of histopathologic
evaluation of any of the other tissues (previously examined) is worthwhile.
We have seen no brain lesions, some degree of congestion in many or most
livers and kidneys, and a variety of common spontaneous lesions in lungs.
No lesion resulting from treatment or exposures has been recognized.
Chemical Analysis of Tissues for Insecticide Content:
Body Burden vs. Exposure
The method of Vukovich et al. (J. Ag. & Food Chem. 17: 1190, 1969) was the
procedure initially used for the cleanup and analysis of these organophosphates in
tissues. Initially, a Varian Aerograph 1800 gas chromatograph with an electron
capture detector was applied. Subsequently, a Tracor Model 222 gas chromatograph*
equipped with a flame photometric detector in conjunction with a 5 percent OV-17
14
-------
column 43" in length and 3/16" L D. was found to be more sensitive and was used
in all subsequent analyses. For example, whole blood was spiked with 0.01 ppb
of methyl parathion and yielded a recovery in plasma of 95-106 percent. For
parathion recovery was 88-104 percent.
For the basic extraction and cleanup procedure, 1 ml of plasma is mixed with
20 Ml of cone. HC1 and 1.5 ml of N-hexane, shaken for 20 minutes on a mechanical
shaker, centrifuged for 10 minutes to break the emulsion, and an appropriate
aliquot injected into the GO for measurement.
Improved recovery was accomplished for parathion and methyl parathion with
the use of a Florisil column, The column, half an inch in width, is packed with
Florisil to a depth of 10 cm, to which 1 inch of anhydrous sodium sulfate is added.
Florisil is preconditioned for at least 24 hours at 130 C and then allowed to cool to
room temperature before making up the column. The column is rinsed with 10 ml
of hexane and then the sample in hexane is added. Elution is done with 15 ml of
10 percent diethyl ether in hexane, followed by an equal volume of 50 percent
ether in hexane. The eluants are concentrated on a steam bath and final evaporation
to 0.5 -1.0 ml is done with a gentle stream of nitrogen. The sample is then
injected into the GC. By this technique recoveries of 95 percent methyl parathion,
104 percent of parathion, and 98 percent of Thimet were attained. For elution of
Guthion from the Florisil column, the technique of Zweig and Sherma in using
CHC1_ as the eluant was found to be a satisfactory procedure, with recoveries of
3
94 percent obtained.
Reproducibility of results made from aliquots of the same sample, starting from
the Initial Vukovich extraction step, were excellent. Plasma samples obtained
from rats 1 hour after a 155-minute inhalation exposure to methyl parathion at
g
152 mg/m had the following duplicate values.
15
-------
4A 26 and 23 ppb
4B 18 and 15 ppb
4C 26 and 27 ppb
4D 25 and 24 ppb
Despite the good sensitivity and reproducibility of the analytical methods, there
was great variation in the plasma values between individual rats which received the
same pesticide treatment, regardless of the route of administration. Plasma
concentrations of Thimet in 7 rats given a 1.6 mg/kg Intravenous dose 30 minutes
after treatment ranged from 6 to 30 ppb; a 2.0 mg/kg intravenous dose after 30
minutes was in the range of 7 to 49 ppb; a 2.5 mg/kg dose showed values ranging
from 12 to 46 ppb. A dermal application of 50 mg/kg of parathion after 2 hours
showed plasma concentration ranging from 38 to 80 ppb; after 4 hours the values
in other rats ranged from 40 to 79 ppb; at 6 hours 76 to 102 ppb; and at 24 hours
35 to 56 ppb. Similar variations were seen for Thimet given intravenously and for
parathion given intravenously.
In view of these major variations, it was apparent that body burden could not be
reliably assessed by chemical measurements of blood plasma. Thus, we are now
measuring whole blood cholinesterase inhibition by the Hestrin method as modified
by Fleisher and Pope. Initial results show greater uniformity of response between
animals subjected to identical treatment.
Table 3 will give you a few summarizatlons of data. These are rounded values,
of course, for easy comparison. We tried to put together values by the various
procedures for the five compounds. The percutaneous LD_. value — and these
DU
are for male rats — for Guthion was not available by the time we made the slide.
It is approximately 210 to 200 mg/kg, in that range.
16
-------
Table 3: Organophosphate Pesticide
LD and LCcr. Values in Male Rats
5U 5U
LD50(mg/kg)
LC50(mg/m )
Parathion
Methyl
Parathion
Thimet
Azodrin
Guthion
Oral
14
12
4
35
16
Intravenous
6
9
2
12
18
Percutaneous
49
110
9
220
Inhalation
530
257
63
178
69
And, as you might expect, parathlon and Thimet show the highest degrees of
toxiclty across the various routes of treatment. Table 4 shows the same types of
data for female rats. Again, there are similar types of responses. Parathion
and Thimet in the female rat show increased sensitivity or they produce a higher
level of toxiclty, but this is not consistent for the other three compounds.
Table 4: Organophosphate Pesticide
LD__ and LC_. Values in Female Rats
50 50
LD50(mg/kg)
LC50(mg/m )
Parathion
Methyl
Parathion
Thimet
Azodrin
Guthion
Oral
8
18
1
20
18
Intravenous
4
15
1
9
8
Percutaneous
20
120
4
206
222
Inhalation
200440
423
15
179
50-60
17
-------
We have attempted to interrelate the kinds of data from one type of exposure
to another. la Table 5 we used Guton's formula to convert body weight to respira-
tory rate. This formula uses 2. 1 grams of body weight to the three-fourths power,
and obtains the breathing rate in milliliters per minute. We've used the table as
an approach for this conversion.
Table 5: Conversion
^mg/m (LC5())
= Ymg/rat = Ymg/61 *
XT * /i,
No. of rats/kg
167 x Ymg _ 167 mg = 167 Ymg
167 x 6 I ~ 1000 I 3
m
*Air breathed by a 200-g rat in 1 hour; 100 breaths per minute and 1 ml per breath.
If one knows that a value has an LDKrt of so many milligrams per kilogram, and
ou
knowing the number and weight of rats used, you can determine the milligrams per
rat, which is equivalent to the milligrams of compound Y per 6 liters of air.
The 6 liters is the amount of air breathed by a 200 -gram rat in one hour, assuming
100 breaths per minute, and one milliliter volume per breath.
Therefore, if we convert the 6-liter volume to a cubic meter, or 1, 000 liters, you
multiply upper and lower cases of the formula by 167. We used this approach in the
next several tables. If these data were idealized as, for example, having an LD
oO
of five milligrams per kilo, this would be equivalent to producing an LC Aof 167
uv
milligrams per cubic meter for a 1-hour exposure, or 42 milligrams per cubic
meter for a 4-hour exposure (Table 6).
18
-------
Table 6: Conversion Data
mg/kg^mg/m
(Assume 200-g rat breathes 100 ml/mln.)
mg/kg
1
2
3
4
5
7.5
10
20
1 Hour3
mg/m
33
67
100
133
167
250
334
668
4 Hours
mg/m
8
17
25
33
42
62.5
83
167
Now, with the data that we have obtained, we attempted to make some comparisons.
Using the intravenous route as our basis of comparison against the LC.. inhalation
ou
values, we attempted to develop some factors, at least for the organophosphates.
For the male rats, by these two comparisons, we have a ratio of approximately
0.9 to 2.6, or an average of 1.4 (Table 7).
Table 7: LD.Q Comparisons (g/kg) - Male Rats
Parathion
Methyl
Parathion
Thimet
Azodrin
Guthion
(Base)
Intravenous
6
9
2
12
7.5
Average
F3
2 2/3
0.85
1
1.2
1.4
Inhalation
16
7.7
2
8.5
19
-------
For the female rats, the ratios are a little tighter, an average of about 0.9
(Table 8). In comparing the intravenous values against oral values, the range
factors are greater, ranging from 1.3 to 3.4, and averaging at 2.1 (Table 9).
In the same comparison for female rats, the values are a little tighter, with an
average of 1.7 (Table 10).
Table 8: LDcn Comparisons (g/kg) - Female Rats
50
Parathion
Methyl
Parathion
Thimet
Azodrin
Guthion
Average
Parathion
Methyl
Parathion
Thimet
Azodrin
Guthion
Average
(Base)
Intravenous
4.5
15
1.2
9
7.5
Table 9: LD5Q
(Base)
Intravenous
6
9
2
12
7.5
F
3 Inhalation
1.5 6+1
0.85 12.7
0.4 0.5
0.6 5.5
0*9 6-7.4
0.85
Comparisons (g/kg) - Male Rats
F3 Oral
^•••B MMMMIMM
2 1/3 14
1 1/3 12
2 4
3.4 3.5
2 15.5
2.1
20
-------
Table 10: LD.n Comparisons (g/kg) - Female Rats
ou
Parathion
Methyl
Parathion
Thimet
Azodrin
Guthion
(Base)
Intravenous
4.5
15
1.2
9
7.5
Fl
1.8
1.2
1.2
2.1
2.4
Oral
8
18
1.4
19
18
Average 1.7
In attempting to compare Intravenous against percutaneous, the ratios run all
over the map (Tables 11 and 12). There is no basis for even a crude factor for
comparison for both male and female rats.
Table 11: LD Comparisons (g/kg) - Male Rats
ou
Parathion
Methyl
Parathion
Thimet
Azodrin
Guthion
(Base)
Intravenous
6
9
2
12
7.5
F2
8
12
4.5
18
60
Percutaneous
49
110
9
220
455
Average None
21
-------
Table 12: LD Comparisons (g/kg) - Female Rats
ou
Parathion
Methyl
Parathion
Thimet
Azodrin
Guthion
Average
(Base)
Intravenous
4.5
15
1.2
9
7.5
_2
4.5
8
Percutaneous
19.5
120
3 1/4
23
30
3.9
206
220
None
There are certain differences between intravenous and inhalation responses,
assuming that the intravenous approach is the most direct and the least likely
to have complicating interferences. The dose is gradually acquired by the inhala-
tion route. During an extended inhalation exposure, detoxification, metabolism,
elimination of the compound from the body, storage in tissues, etc., can occur.
All of these variables can affect the ratioing between inhalation and intravenous
comparisons.
Of course, if there is lung damage, gaseous exchange can be affected. Thus,
toxic effects may increase or decrease due to lung passage; absorption maybe
incomplete via the lungs, and pulmonary excretion may not occur (Table 13).
Tentative conclusions from those data suggest that the oral LD_rt is approximately
DU
two times the intravenous LD dose. For inhalation, compared to intravenous,
50
it is approximately equivalent, but it can range from about one-half to three-fold.
There are no reliable dermal comparison factors (Table 14).
22
-------
Table 13: Some Differences
Intravenous vs. Inhalation (IN)
• (IN) dose is gradually acquired (1 or 4 hours);
• During (IN) exposure, detoxification, metabolization,
elimination, storage, etc., may occur;
• Lung damage may impair gaseous exchange;
• Toxic effects may increase or decrease due to lung
passage ;
• Absorption may be incomplete via lungs;
• Pulmonary excretion may not occur.
Table 14: Tentative Conclusions
• Oral LDCA ^ 2 x IV LDert
oU 50
• Inhalation LD A = IV LD_rt but ranges from
50 50
0.5 to 3 x IV LD
OU
• Dermal LD.n ^L IV LD__
00 50
All of this information was developed by a large group of people. It included the
contributions of chemists and chemical engineers, pathologists, analytical chemists,
and toxicologists. Our cholinesterase work is continuing, and we are attempting to
correlate biological effects with cholinesterase depression, in relation to the LC or
During the coming year, we will use the inhalation system with a selected number
of pesticides to investigate possible teratological responses. The exposure regimen
will be a fraction of the LC_A, an hour per day during the 7th to 15th day of gestation.
50
The animals will be necropsied the day before delivery, with the usual teratological
examinations of the fetuses.
23
-------
We also plan to work with at least one labeled compound to try to follow absorption
of the compound across the placenta and into the fetus. For reference teratogens,
we will use either chloroform or methylethylketone. Both have shown to produce
various terata by inhalation exposure.
24
-------
STUDIES ON TOXICITY TO MAMMALS
OF SMALL PARTICLE AEROSOLS OF
NUCLEAR POLYHEDROSIS VIRUS (NPV) PESTICIDES
Dr. James T. Stevens*
NOT AVAILABLE
FOR
PUBLICATION
*National Environmental Research Center, Research Triangle Park, North
Carolina
25
-------
METABOLISM OF PESTICIDES
Dr. Ronald L. Baron*
Unlike the first two speakers, Jim Stevens and I are surrogates, talking about
somebody else's work at the Naval Biomedical Research Laboratory, This is an
interagency agreement with the Department of the Army with the work contracted
out to the Naval Biomedical Research Laboratory (NBRL) in Oakland, California.
The background to this particular study is as follows: The Army contracted with
the Naval Biomedical Research Lab for inhalation toxicology studies on pesticides.
As a spin-off from part of their work, Dr. Peter Berteau found some very interesting
results with the organophosphorous pesticide naled, or Dibrom, which is a brominated
derivative of DDVP. His initial studies indicated that the organophosphate naled
has a low toxicity when administered by the oral route. The LD5(j ranged from 160
to 430 milligrams per kilogram in the adult rat.
Studies with this compound at the NBRL have indicated that by inhalation naled
appeared to be quite toxic to the rat, these conclusions being based upon toxicant
retention values calculated by the inhalation of a radioactive tracer. The inhalation
toxicity to the female rat was greater than 25 times that of the oral toxicity. As
you might have seen with Dr. Newell1 s comparative data, there is roughly a two-fold
difference between oral and inhalation toxicity. With naled the 25-fold variation was
quite unusual.
Greater inhalation than oral toxicity values are also evident with the mouse,
although the differences are less marked with the mouse than with the rat, probably
because the rat retained more material in the lung than the mouse. Dr. Berteau
and co-workers calculated that the rat retained 6.4 percent of the inhaled toxicant,
whereas the mouse retained only 3.5 percent.
Pesticides and Toxic Substances Effects Laboratory, Research Triangle Park,
North Carolina
27 Preceding page blank
-------
The EPA is currently supporting a program at NBRL to define some of these
differences, studying a comparison of the metabolic breakdown of naled in rats,
following administration by both the oral and inhalation routes.
Possibilities that can occur to explain the results of the high inhalation toxicity
are:
1. The corrosive action of naled on lung tissue; naled is known to be highly
corrosive to lung, and lung necrosis has been observed following inhalation exposure.
However, the fact that the rats die rapidly, exhibiting marked cholinerglc signs of
poisoning, indicates that this phenomenon is not the whole story. Corrosive action
would also not explain the greater inhalation toxicity to the rat than to the mouse.
2. The possibility of rapid absorption into the blood stream through the lung
alveoli, and rapid absorption will occur with all pesticides following inhalation.
However, in the case of another organophoaphorous compound, chlorpyrifos or
Dursban, only a very slight increase in toxicity results following inhalation by the
rat; there was none in the mouse. Again, this suggests that naled may have a
special property that accounts for its high inhalation toxicity.
3. The difference in the metabolism or metabolism rate to a more toxic inter-
mediate through some mechanism in the lung. In the body, naled can be expected
to be metabolized by two routes: hydrolysis, or enzymatic breakdown to dimethyl
phosphoric acid, bromodichloro acetaldehyde, and hydrogen bromide. This reaction
is rather rapid at alkaline pH.
In the presence of sulphydryl compounds, such as glutathione, naled is rapidly
debromlnated to produce DDVP, or dichlorvos.
Dichlorvos is also an important insecticide. It is considerably more toxic to
mammals than naled. At NBRL the LD5Q has been determined to be between 56 and
80 milligrams per kilogram, orally, in the rat, whereas, with naled it was calcu-
lated to be 160. Therefore, there is a possibility that in the lung, sulphydryl groups
may convert naled into dichlorvos, which might explain the enhanced Inhalation toxicity.
28
-------
Preliminary in vitro work with unlabeled naled has demoastrated the presence
of dichlorvos using thin-layer chromatography following stirring of a solution of
naled with an aqueous solution of cysteine. However, oral Intubation of naled to
a rat and periodic collection of urine has failed to demonstrate either unchanged
naled or dichlorvos in the urine.
These experiments are very preliminary and form the basis for the study that
EPA has just funded. NBRL is currently awaiting the availability of 14 -labeled
C
ethyl naled, the synthesis of which has been contracted, but problems seem to
have arisen in the synthesis.
It is easily demonstrated that naled can be synthesized through the production
of chloral, the chlorination of ethanol. The reaction of chloral with trimethyl
phosphite will produce labeled DDVP, which could be brominated to produce naled.
At this particular time, we are still waiting for the preparation. In the future,
there will be an examination of the metabolism via both oral and inhalation routes
of administration.
As with the previous interagency agreement, this one was started in the latter
?irt of June. Hopefully, by this time next year, we should have results from this
program.
DR. CURLEY: These research studies on inhalation toxicology summarize the
state-of-the-art in our work with chemical pesticides and also the available facil-
ities of EPA at the Research Triangle Park facility. They point out the need for
good inhalation data, as well as the need for well-trained personnel In the area of
inhalation toxicology.
29
-------
We will now open the session for a discussion of these studies. If there are
any questions, we'd like to entertain these questions now.
DR. CRANMER: It seems to me that the purpose of inhalation toxicology, or
inhalation as a route of exposure, as opposed to a feeding study, is to be more
representative of different exposure groups. Now, if we were talking about pesti-
cides and we were talking about residues, the feeding study would be perhaps more
appropriate. We then begin to speak of the types of persons that we could be con-
cerned about being exposed, and two groups immediately come to mind — the person
who elects to use a pesticide around the home and the occupationally exposed indi-
vidual. They would likely be exposed to particulates or aerosols of these com-
pounds. Now, with respect to that, it would seem to me that we should pay an ex-
traordinary amount of attention to explaining not that there's a difference in these
toxicities through these different routes, because I think we all agree that there
would be differences in this, but to explaining this expression of differences on the
basis of two basic groups of phenomena. One would be expression through a dif-
ferent mechanism. Now, it might be a different effect on the lung, or an altered
metabolism, or it indeed could be a difference in the absorption kinetics. And
now, drawing on Dr. Newell's presentation, I would ask if future work, especially
with the anticholinesterase pesticides, would draw on the relative kinetics of inhi-
bition of brain cholinesterase, red cell cholinesterase, and plasma cholinesterase
as indexes of these phenomena, and that with parathion, a compound that was dis-
cussed, the plasma and red cell are inhibited at differential rates at different
levels. I think expressing these things on a relative anticholinesterase potency
might give more insight than searching for body burdens or blood levels with these
very rapidly metabolized types of compounds. And I would just like to ask if
special emphasis is going to be placed on those phenomena in the future?
DR, CURLEY: Yes, that is, I think Dr. Newell or Dr. Baron can answer that.
This is the reason we decided against looking at body burdens and we decided to
30
-------
look at the cholinesterase. I think he summarized this in his summary statement.
He indicated that he would use cholinesterase determinations as an assessment
technique for the toxicity of those compounds via the inhalation route of exposure.
DR. CRANMER: Well, I think it's important to move this up perhaps one more
level of sophistication into cholinesterase and its subgroups: the carboxyl esterases,
in general, which might be inhibited; the plasma cholinesterase; red blood cell
cholinesterase, and then brain cholinesterase. As far as the plasma and red blood
cell, it's easy to get 50 microliter samples, which will give you the answer.
DR. CURLEY: Yes.
DR. CRANMER: And one final question, and that is, since we are looking at very
specific exposure groups now, are we attempting to relate the components used to
generate the aerosols and the types of aerosols that would be generated with the
expected exposures that humans actually have? Dimethylsulphoxide, for instance,
I don't think that's used — is it? — in any of the commercial formulations, and it has
very peculiar absorption kinetics associated with that. And there's the physics which
is very interesting being able to relate the ability to absorb certain size particles
with different velocities in the lung. I don't think that's well enough worked out.
An interspecies comparison would be a very important contribution. What are
the velocities of the aerosol that's available to be trapped? I'd just like to throw
some of those points open for discussion.
DR. CURLEY: I think these questions that Dr. Cranmerasked have been raised
by us, by Dr. Newell, and by the people in inhalation studies at the Naval Biomedical
Research Laboratory. Dr. Cramner raised the question concerning the effect of
physical chemical properties, that is, particle sizes and velocity of these particles.
We've also raised the same questions with Dr. Newell, including the use of various
vehicles and the effects of these vehicles on the toxicity of the compound, and whether
there is a synergistic or potentiation effect by use of, say, DMSO as a vehicle to
generate the aerosol. Hopefully, these questions will be answered.
31
-------
We have given consideration to those studies. Dr. Stevens, I think, has plans
although he did not mention it in his presentation, to institute studies at our laboratory
to assess the physicochemical properties of the particulates or of varying the par-
ticle sizes, etc. These pesticides were generated as aerosols. He used the term
toxicokinetics, or pharmacokinetics, which includes rates of absorption, lipid solu-
bilities, and the effects of particle size and that sort of parameter. I think a great
deal of this work has been done with anesthetics through intratracheal injections
of compounds directly into the respiratory tree. But this kind of work has not been
done via the nasopharyngeal region and, therefore, we must assess that aspect of
pesticide inhalation, including the vehicles in which these compounds are generated
as aerosols. There is a possibility that any of these parameters might contribute to
the toxicity of a compound.
DR. NEWELL: Dr. Cranmer commented that it is desirable to study compounds
in the neat form. Dr. Cranmer, we talked about this earlier. It has been the opinion
of the EPA staff and ourselves that you should obtain the initial information on the
technical product. You could go back, and if it appears feasible, look at the vehicle
or the carrier Itself and also the neat material to see if there is potentlation or inhi-
bition. In this manne r, you could obtain relative toxicological information. Whether
or not we go in that direction is problematical. Perhaps the manufacturers of the
compounds will do this themselves.
DR. ENGLER: I have a question with regard to the presentation on the viral pesti-
cides. Why has the Heliothis virus been chosen to be tested again? There are some
inhalation toxicity studies available on that virus. Maybe they were not conducted
with the same degree of sophistication as these current studies but they are adequate.
I would like to point out or urge therefore that other viruses which may have the poten-
tial of becoming useful control agents are just as important to be studied as the Helio-
this virus, for which we have considerable data on toxicity or infectivity available.
32
-------
I also would like to point out that although we are talking mostly about chemicals
at this conference, we should strike the word toxicity when we consider insect viruses.
\Ve are not concerned about their toxicity but rather their infectivity and other effects
attributable to microorganisms and, of course, serological responses. Therefore,
j?ight from the start, when we include the viruses in the substitute chemicals, we
should be aware that they are not to be treated as chemicals.
DR. CURLEY: Dr. Engler of OPP asked the question, "Why was the nuclear poly-
hedrosis virus selected?" It was selected as an extension of an in-house effort.
Presently, to our knowledge and perhaps to your knowledge, there may be other
Juanufacturers in the country other than Nutrilite in California which is producing
viruses other than the nuclear polyhedrosis viruses. The product Nutrilite produces
is a virus. As Dr. Stevens pointed out, insect parts, fungi, and just about a little
bit of everything has been found, including a lot of ash, indicated by analysis by two
groups of researchers -- one at Scripps and the other at the NBRL in California.
\Ve would appreciate any studies or any data that you could make available to us on
viruses other than those, Dr. Engler.
DR. KAMIENSKI: I'm still trying to evaluate what the significance of these inhala-
tion studies is as they would apply to making a hazard evaluation as to what the
Workers in the field are exposed to. I don't know if there are follow-up studies pro-
vided for this area or not. My question is, are there any inhalation studies that
are being considered using the particle size range that is normally encountered or
being generated by agricultural equipment? In other words, you've identified the
toxicity of some of these materials, and this is with particle size ranges that are
95 percent respirable, as opposed to inhalation studies using the equipment that is
normally used in agricultural sprays which generate a much larger particle size,
\vhere the respirable particles are probably less than 1 percent.
The other question is, can you explain to me the logic of running teratogen studies
Via the subacute inhalation route?
33
-------
DR. CURLEY: I'll attempt to answer the last question. I'll ask Dr. Dale to
answer the first.
DR. DALE: Number one, this study, as I envisioned it, wasn't attempting to
study the hazard of pesticides as exposed to field workers. It was to generate
aerosol to get the most toxic form, so we could compare routes of exposures —
inhalation versus dermal versus oral versus intravenous. And, in this case,
aerosol was used because you get the most respirable particle and the most
toxicity on this route. No, this cannot be extrapolated to field hazard from spray-
ing equipment. Although I wonder if it could not be extrapolated to vapors arising
on reentry problems, because in this case the particles are certainly in the small
range. But to answer your first question, it was an attempt to determine hazard
from spray equipment or from dust.
DR. CRANMER: I was speaking of this in my usual kind of nebulous style a
minute ago, but what I was trying to say was, if one takes a certain particle size
which is consistent with what perhaps a worker would be exposed to, then one
tries to model after that a very small rodent, then the trajectory physics associ-
ated with the impacting of that particle, the absorption of that aerosol by the lungs
or other associated paraphernalia of the inhalation route is quite different. So
what we really have here is a modeling problem, where we really need to base it
on relative capability to absorb or be absorbed by, rather than an exact replica
of particle size. I think we'd fall into some gaping holes if we tried to do it that
way. The physical model though, to my knowledge, is not worked out very well.
Some particular attention will be necessary in that area if we're to be able to
make that bridge.
DR. CURLEY: Yes, that's correct. Now, to your second question, there are
several inconclusive reports in the literature on several of the insecticides, prin-
cipally the organophosphate ester type compound that suggests there may be pos-
sible teratogenic effects. These are reports — they're basically medical reports
34
-------
in medical journals — that indicate housewives, or whoever, or migrant female
field workers were exposed to these compounds and children with certain types of
malformities were born to these mothers. However, these reports are inconclu-
sive and, therefore, they suggested a need to evaluate the possibility of producing
terrata in animals via the inhalation route, because the apparent route of exposure
to these females was the inhalation route. I think that this was the question that
you raised. And therefore Dr. Newell will extend his studies to evaluate that
possibility.
35
-------
TOXICOLOGICAL EVALUATIONS
Dr. Ralph I. Freudenthal*
A number of registered pesticides are presently being reviewed by EPA's Office
of Pesticide Programs. One phase of this review is to evaluate the existing toxicity
data for each pesticide and identify and define areas of missing data. On June 17,
1974, Battelle's Columbus Laboratories was awarded a toxicology contract to help
fill these data gaps by initiating research programs of a high-priority nature.
In order to fulfill the toxicity testing needs of EPA's Office of Pesticide Programs,
we are providing the following capabilities:
1. Acute and subacute feeding studies;
2. Oral and dermal exposure studies in small mammals;
3. Pharmacoklnettc studies in small mammals and in subhuman primates —
this includes measuring the rate of absorption, whole-body distribution, metabolism,
and rate of excretion;
4. Studies to determine the effect of the pesticide on enzyme activity;
5. Multigeneration reproduction studies; and
6. Neurophysiological and behavioral studies.
The general scope of our work is divided Into two major areas:
1. Response to inquiries for data from ongoing review processes; and
2. Response to inquiries for data on compounds identified as impurities in tech-
nical pesticide formulations.
Although the contract was awarded in June 1974, no research tasks were assigned
to BatteUe until February 1975. Because of this time lag between contract award and
task assignment, I cannot present to you the results from any completed study, but
I will present a brief overview of the research tasks presently assigned to our labora-
tory.
*Battelle Memorial Institute, Columbus Laboratories, Columbus, Ohio
37
-------
Task 1. Ninety-Day Toxicity Study of Ethylenebisisothiocyanate Sulfide (EBIS)
It is now well documented that the ethylenebidithiocarbamate (EBDC) fungicides
decompose to form ethylene thiourea (ETU). ETU exhibits goiterogenic as well as
dose-related carcinogenic properties. The EBDC fungicides also degrade to EBIS,
whose toxicity and carcinogen!city are presently unknown. (See Figure 1.)
Figure 1
H -
t >.5
H,C — H — C
2 ^
\
S
H H s
S HC— K_C^
\-s EBDC | \
H2C— |' V-f-S
K H
ETU EBIS
Our assigned task is to perform a 90-day EBIS toxicity study in rats, in which
the test chemical is incorporated into the feed. Unfortunately, there is no commer-
cial source of EBIS. Therefore, a custom synthesis of EBIS is underway at Trans
World Chemicals, with an estimated delivery date of August 1. The study will be
initiated upon receipt of the chemical and its incorporation into the feed.
Since ETU is goiterogenic, 125I uptake studies and 1$ and T4 tests will be per-
formed on the rats. After reviewing the T3 and T4 kits from Ames, Squibb, Oxford,
Abbott, Pharmacia, and Nuclear Medical, we chose the Nuclear Medical kits, our
choice being based on their sensitivity, assay time, and components. In a pilot study
to test the Nuclear Medical kit, we used serum from 10 control rats and from 5 rats
which were treated with methimazole, a thiocarbamate type antithyroid compound.
The results are shown In Table 1.
38
-------
Table 1: Rat Serum Assays
Group T3 + S.D. T.+S.D.
Control 63.12 + 1.16 4.66 + 1.18
Methimazole 63.13 + 2.07 1.95 + 0.84
The data clearly show that the TBG levels were not affected by the drug, whereas
the concentration of thyroid hormone was substantially affected. We are now convinced
that the Nuclear Medical kit is suitable for measuring Tg and T4 levels in rat serum.
Task 2. Determine the Degradation of Selected EBDC Fungicides under Various
Cooking Conditions and on Certain Fresh Vegetables by Measuring Quantitatively
EBDC. ETU. and EBIS
Initially, this study was to be performed in our laboratories. For a number of
reasons, a subcontract was awarded to Technological Resources, Inc., a subsidiary
of the Campbell Soup Company, who will perform the following studies:
1. Develop a quantitative assay for measuring EBIS;
2. Determine residue levels of EBDC, ETU, and EBIS on three crops following
the application of Dithane M-45 and Dithane M-22;
3. Determine the.effects of normal washing and preparation on residual EBDC,
ETU, and EBIS;
4. Determine the effects of food processing procedures (canning and freezing)
on residual EBDC, ETU, and EBIS; and
5. Determine the rate of formation of ETU and EBIS under different cooking
conditions (pH, ionic strength, etc.)
Technological Resources, Inc., is presently working on the EBIS assay. In addi-
tion, they have harvested and processed the spinach grown in New Jersey. Analysis
of the Dithane M-45-treated spinach samples for EBDC and ETU is also in progress.
39
-------
Task 3. Pharmacokinetics of Pentachloroaniline in the Rhesus Monkey
Pentachloroaniline (PCA) is a metabolite of the soil fungicide, pentachloronitro-
benzene (PCNB). It has been identified in the body fat, liver, and feces of a number
of animal species to which PCNB was administered. In addition, PCA has been iso-
lated from cow's milk and from grass grown in PCNB-treated soil.
We have been asked to determine the rate and extent of absorption of PCA by the
oral route, its metabolism, and its rate of excretion. To carry out this study, we
asked California Bionuclear Corporation to custom synthesize 1 mCi 14C-PCA. We
have just received the labeled PCA and will begin this task immediately.
A single dose of ^C-PCA will be orally administered to five male rhesus mon-
keys and then blood, urine, and feces samples will be collected at specified time
points. These biological samples will be analyzed for PCA and its metabolites.
Task 4. Pharmacokinetics of Pentachlorobenzene in the Rat and Rhesus Monkey
Pentachlorobenzene (PCB) is an impurity found in essentially all commercial
preparations of pentachloronitrobenzene fungicides. A recent study of certain halo-
genated benzenes showed that PCB accumulated in the fetus to a greater extent than
did PCNB and the other test compounds.
We have been asked to determine the pharmacokinetics of PCB in the rat after
oral and dermal administration, and also in the rhesus monkey after oral adminis-
tration. California Bionuclear Corporation is presently custom synthesizing 1 mCi
for use in these experiments.
In summary, we are just beginning the experimental work recently assigned to
us as oart of our toxicology contract.
-------
TOXICOLOGICAL RESEARCH: ACUTE LD STUDIES"
Dr. Larry L. Hall**
Pursuant to the Federal Insecticide, Fungicide, and Rodenticide Act, as amended,
and Public Law 93-135, EPA is charged with the responsibility for the evaluation
and testing of alternate (substitute) chemicals such that the substitute is not more
hazardous to man and his environment than the "problem" pesticide which it is to
replace. In order to meet this mandate, our laboratory has initiated a program
of research to provide a data base for the gaps that exist in our knowledge of the
biological activity of these alternate chemicals. Included in this repertory is our
ability to perform a variety of acute, subacute, and chronic studies, including
metabolic, kinetic, and reproductive experiments.
Because of the "man or mouse" difficulty, we use a longitudinal approach such
that all experiments are performed on the Sherman strain rat so that comparative
analysis on pharmacodynamic and pharmacokinetic studies can be performed on
different compounds without the introduction of additional variables such as species
and strain in order to provide more meaningful input into the review process.
This report presents the acute toxlcity study results obtained preparatory to
more extensive studies of these compounds in the Alternate Chemicals list.
Methods
Non-fasted adult (90-120 da.) and weanling (30-34 da.) Sherman rats (COBS)
obtained from the Center for Disease Control colony were used for these studies.
The solutions, formulated just prior to dosing, were of concentrations such that
a (relatively) constant volume (0.005 ml/g) could be administered by gavage. In some
instances (pentachloronitrobenzene) limited solubility and low order of toxicity re-
quired the use of up to 0.02 ml/g. All dosing was performed between 10 and 11 a. m.
* Co-authors are R. Linder, W. Felsenstein, J. Goldstein, and A. Curley
**Chief, Acute and Chronic Studies Section, Environmental Toxicology Division,
Research Triangle Park, North Carolina
41
-------
For the dermal toxicity study, pentachloronitrobenzene (PCNB) was dissolved
in xylene and applied to the skin at a volume of 0.01 ml/kg after clipping the hair.
No attempt was made to prevent oral ingestion.
All compounds used were technical formulations obtained from contract except
for the PCNB used in the suspension studies which was obtained commercially (Table 1.)
Table 1: List of Common and Chemical Names of Compounds Tested
Common Name Chemical Name Purity
Cap tan
Folpet
Guthion
Fenthion
Dicamba
PCNB1
PCNB2
MSMA
DSMA
Cacodylic acid
Use
N-[(Trichloromethyl)thio] -4-cyclo-
hexene-1,2-dlcarboximide
N-(Trichloromethylthio)phthalmide
[azinphos-methyl] O,O-Dimethyl
S-[(4-oxo-l, 2,3-benzotriazin-3(4H)-
yl) methyl] phosphorodithioate
O.O-Dimethyl O-[3-methyl-4-(methyl- Technical
thio)phenyl] phosphorothioate
Technical Fungicide
Technical
Technical
3,6-Dichloro-o-anisic acid
Pentachloronitrobenzene 95%
Pentachloronitrobenzene 99+%
Monosodium methanearsonate
Disodium methanearsonate
Dimethylarsinic acid
Technical
Technical
Technical
Formulation
58.4% active
ingredient
Formulation
80.1% active
ingredient
Formulation
65.6% active
ingredient
Fungicide
Insecticide
Insecticide
Herbicide
Fungicide
Fungicide
Herbicide
Herbicide
Herbicide
Following dosing, the rats were observed frequently for signs of intoxication.
The animals were weighed weekly until termination of the study. Gross autopsies
were performed on dead or moribund animals and at the end of each study.
42
-------
The LD5Q values were calculated by the method of Flnney (1) or Litchfleld and
Wilcoxon (2) when convergence did not occur in the regressional analysis.
A subacute feeding study was undertaken with pentachloronitrobenzene. Formu-
lation for the feeding study (PCNB) was accomplished by preparing a concentrate of
the chemical in powdered ration and serial dilution with additional chow to the levels
of concentrations required. The formulation was prepared at trl-weekly intervals.
Six groups of four female weanling Sherman rats were fed technical-grade (98
percent) PCNB for 101 days. The dietary levels were 0, 50, 100, 200, 400, and
800 ppm. Growth rate, hematological, tissue, and biochemical examinations were
made.
Liver specimens and urine were taken for biochemical studies. Urine was assayed
for amlnolevulinic acid, porphobilinogen, coproporphyrln, and uroporphyrln. liver
was assayed for total porphyrin, microsomal protein and P-450, aryl hydrocarbon
hydroxylase, and N-demethylase.
Selected tissues were taken for hlstopathologlcal study.
Results
MSMA or DMSA
Signs of toxicIty in rats given MSMA or DSMA Included diarrhea, depression,
and dyspnea with most deaths occurring in 2-3 days. The predominant finding at
autopsy was mild to severe irritation of the gastric mucosa. Cecal contents were
mucoid and yellowish. In addition, most rats dying from MSMA had hemorrhagic
adrenals and small spleens.
Cacodvlic Acid
Signs of toxlclty In rats dosed with cacodyllc acid were diarrhea, depression,
and dyspnea with most deaths occurring 3-5 days after dosing. Gross inspection of
dead rats revealed moderate to severe irritation of the gastric mucosa, hemorrhagic
adrenals, and atrophlc spleens. Tissues of animals given a nonlethal dose (1500 mg/kg)
were grossly normal after 14 days.
43
-------
Table 2: Acute Toxiclty of Substitute Chemicals ia the Sherman Hat
Compounds
Route
Sex
Age
LD50 (rogAg) 95%
Confidence Limits
JLD50
Slope of Probit Highest Nonlethal Llterature-TSL*
Htigresslon Line Dose Tostcd3 {rag/kg) (nig/kg) Rat
MSMA [58,4%]
DSMA (80.1%)
Cacodylic acid
[65.6%J
Guthlon
Fenthion
Dicamha
Folpet
Folpet
Captan
Captan
PCNB (99%)
PCNB (99$.)
Suspension
PCNB [95%J
Suspension
irt-%?JB [95%]
Suspension
PCNB [95%]
Suspension
PCNB (95%)
Suspension
0
0
0
0
0
0
0
0
0
0
0
0
0
0
D
D
F
F
M
F
M
M
M
F
M
F
F
F
M
M
M
F
W
W
A
W
W
W
W
W
W
W
W
A
A
W
A
A
1743 [1580-1835]
Ifl95 11507-187R)
1802 [14]
16.3 [Ki.a-5w.ij
284 [244 -338]
328fi [2979-3643]
5000
5000
5000
5000
1359 [1254-1461]
1989 [1890-2101]
2550 [2257-28621
2812 [2G9L-2931]
5000
5000
10.52
10.10
9.16
13.19
5.87
9.45
-
-
-
-
ifi.se
20.36
-
31.25
-
-
*
1200
1200 (1/10)
1000
12.5
-
1650
5000 (1/10)
5000
5000 (1/10)
5000
900
1500
2000
2250
5000
5000
700
1800
1350
-
215
1040
10000
-
480
1650
1 Oral [O]; Dermal [DJ.
2 Weanling {WJ; adult [A].
3 ( ) Mortality at lowest dose.
4 Toxic Substances List (1914).
-------
Guthlon
Weanling female rats showed typical cholinergic signs, i.e., salivation, muscle
fascidilations, lacrimatlon, and dyspnea with onset a few minutes after dosing. Most
deaths occurred within 2-3 hours. Recovery of poisoned animals usually occurred
within 20 hours.
Fenthion
Weanling male rats displayed typical cholinergic signs as described for guthlon.
The onset of signs was observed less than 2 hours after dosing with most deaths
occurring between 24 and 48 hours. Survivors recovered in 48-72 hours.
Dicamba
Weanling males displayed dyspnea and generalized myotontc type responses when
disturbed. These signs occurred within a few minutes after dosing and persisted for
24-48 hours. Most deaths occurred within 24 hours.
Captan and Folpet
The sign of poisoning in rats given folpet or captan was mild to moderate diarrhea
which in most animals persisted 2-4 days after dosing. A few rats exhibited abdominal
distension.
Pentachloronitrobenzene (99 Percent Pure)
Signs of toxicity in rats dosed with PCNB included diarrhea, depression, dyspnea,
weakness, and tremors when agitated. Most mortalities occurred 1-3 days following
treatment. Gross inspection of dead rats revealed hemorrhage into the lumen of the
gastrointestinal tract. Tissues of rats surviving on LDOA doses (1200 mg/kg) were
av
grossly normal.
Pentachloronitrobenzene (Technical)
Oral and dermal LD determinations for 95 percent technical PCNB in Sherman
ou
rats were completed. The material, because of an unexplained solubility problem,
was given as a suspension in peanut oil for oral dosing. Signs of toxicity in orally
dosed animals were labored breathing, depression, and diarrhea, followed by excit-
ability in rats that recovered. Most deaths occurred 2-3 days after dosing. No
definite signs of toxicity were seen in dermally exposed animals.
45
-------
PCNB - Pilot Subacute Study
No signs of toxvctty were observed In the test animals. Growth rates in the
test groups were comparable to those of controls. No change In organ weights
was observed in the test groups. Hematological profiles of the test groups did
not differ from those of the controls.
The only definite changes observed grossly at autopsy were moderate lesions
of both kidneys in one rat at the 200 ppm dietary level. This finding is probably
not dose-related. Histological examination is not complete at this time.
Liver samples were taken from all rats at the termination of the experiment
for biochemical studies. No significant changes were noted in the total liver por-
phyrin or microsomal protein concentration. Urinary aminolevulinic acid, copro-
porphyrin, or uroporphyrin were not changed. The data were suggestive of a
slight increase in porphobilinogen at the higher dietary levels. Liver P-450 and
N-demethylase appeared slightly elevated at the 800 ppm level. Aryl hydrocarbon
hydroxylase appeared elevated at 100 ppm and above.
Discussion
Reasonable agreement exists between our LD values and those recorded in
the literature in spite of the possibility of quite different degrees of purity. The
exceptions were MSMA and captan. Captan, folpet, and dicamba exhibited rela-
tively low orders of acute toxicity.
Relative toxicity probit analysis of the organic arsenicals revealed no signi-
ficant differences between the formulations in spite of a rather wide range of
active ingredients (58.4 - 80.1 percent). The similiar slope values suggest similiar
mechanisms of acute toxicity.
Guthion was found to be 15.5 times more potent than fenthlon and showed a much
steeper dose-response relationship. The reason for this phenomenon is not
understood.
The battery of acute PCNB studies suggests a sex difference with females being
more sensitive than males. However, no significant effect of age was found
in male rats. A difference between suspensions and solutions of the agent is also
suggested.
46
-------
No grossly significant toxicological effects were noted at any of the dose levels
in the PCNB feeding study. The importance of the biochemical changes noted is
unknown at this time. A chronic study is needed to evaluate the significance of
these early changes. In addition, the possibility of the contaminant being the
causative agent needs to be ruled out.
References
1. Finney, D. J.: Statistical Method in Biological Assay. 2nd edition
Hafner Press (1971)
2. Litchfield, J. J. and Wilcoxon, F.: A simplified method of evaluating dose-
effect experiments. J. Pharmacol. Exptl. Therap. 96; 99-133 (1949)
3. The Toxic Substances List, 1974 edition, H. E. Christensen edition. U.S.
DHEW - NIOSH
47
-------
2-AAF AS A MODEL COMPOUND
Dr. Thomas J. Haley*
Following the selection of the chemical 2-acetylaminofluorene, the mouse
strains (BALB/c and C57BL/6) and a thorough review of the literature on the
toxicity of 2-AAF, it became apparent that an acute oral LD determination and
50
a subacute 90-day feeding study were required before the beginning of the ED...
experiment was attempted.
2-AAF as a suspension in aqueous 10 percent polysorbate-80 was administered
orally to the various mouse strains shown in Figure 1. Symptoms of acute toxicity
were lethargy, sedation, loss of righting reflex, circling, orange-red urine, bladder
concretions, and paralysis with all four limbs rigidly extended. In can be seen that,
generally, the females were more susceptible to the toxic effects of 2-AAF. More-
over, both groups of C57BL/6 female mice showed significant differences from the
males (Table 1). Experiment No. 1 CD-I males showed significantly lesser
toxicity than those in the other experiments ( Table 2). Comparisons of responses
between No, 2 CD-I vs. BALB/cj; C57BL/6Crvs. C57BL/6J or BALB/cj; and C57BL/6J
vs. BALB/cj showed no significant differences, while all other comparisons between
groups were significantly different. Comparison between all-female groups and
No.l CD-I showed significant differences in response (Table 3 ). No significant
difference in response was found when comparison was made between No. 2 CD-I vs.
C57BL/6Cr or BALB/cj female mice. Bladder concretions were present in 7.5 per-
cent of the animals and appeared at 48 hours. Of greatest importance was the dif-
ference in responses between animals of the same strain from different sources.
In the 90-day subacute feeding study, 1200 mice, each of the C57BL/C] and
BALB/cStCrlBR strain, equally divided between the sexes, were used. 2-AAF
concentrations in the diet were 500, 250, 100, 50, and 10 ppm. An unbalanced
statistical design was used with groups of 80, 80, 160, 240, and 240 for the above
concentrations. A control group contained 240 animals. Histopathological examina-
tion was performed on the lungs, heart, aorta, thymus, skeletal muscle, kidney,
*National Center for Toxicological Research, Jefferson, Arkansas
Preceding page Hank
49
-------
Table 1: Oral LD /7 Days of 2-FAA in Several Mouse Strains
50
LD and range*
Expt no. Strain No. mice Sex (g/kg) Slope and range a
1
2
3
4
5
CD-I
CD-I
CD-I
CD-I
C57BL/6Qr
C57BLy#Cr
C5*7BL/6j
C57BL/6J
BALB/cj
BALB/cj
100
100
100
100
100
100
100
100
100
100
M
F
M
F
M
F
M
F
M
F
2.02 (1.89-2.16)
1.87 (1.75-2.00)
1.08 (1.02-1.14)
1.08 (1.02-1.14)
1.22 (1.16-1.28)
1.02 (0.96-1.08)
1.23 (1.17-1.29)
0.81 (0.76-0.86)
1.20 (0.98-1.46)
1.17 (1.08-1.27)
1.19 (1.07-1.32)
1.20 (1.06-1.36)
1.17 (1.06-1.29)
1.17 (1.06-1.29)
1.11 (0.98-1.25)
1.13 (1.0 -1.28)
1,11(0.98-1.25)
1.18 (1.07-1.3 )
1.17 (0.84-1.64)
1.26 (1.09-1.46)
All values at p=0.05.
Table 3: Comparison of LD,n/7 Days Between Females of Different Strains51
Ov
Strain No. 1 CD-I No. 2 CD-I C57BL/6Cr C57BL/6J BALB/cj
No. 1 CD-I
No. 2 CD-I
C57BL/6Cr
C57BL/6J
BALB/cj
—
S
S
S
S
S
—
NS
S
NS
S
NS
—
S
S
S
S
S
—
S
S
NS
S
S
~
a S=slgnificantly different at p=0.05; NS=not significantly different.
Table 2: Comparison of LD50/7 Days Between Males of Different Strains*
Strains No. 1 CD-I No. 2 CD-I C57BL/6Cr C57BL/6J BALB/cj
No. 1 CD-I
No. 2 CD-I
C57BL/6Cr
C57BL/6J
BALB/cj
—
S
S
S
S
S
—
S
S
NS
S
S
—
NS
NS
S
S
NS
—
NS
S
NS
NS
NS
aS=slgnificantly different at p=0.05; NS=not significantly different
50
-------
adrenal, liver, spleen, gallbladder, pancreas, cerebrum, cerebellum, spinal cord,
stomach, colon, ileum, duodenum, lymph node, salivary glands, lacrimal gland,
eye, bar del an gland, thyroid, trachea, esophagus, skin, tongue, sternum, testes,
epididymus, preputlal gland, seminal vesicle, coagulating gland, ovary, uterus,
mammary gland, urinary bladder, prostate, pituitary, and any unidentified mass.
All data were collected and stored in a Modular Computer Systems m/5 mini-
computer. No observable signs of toxicity were produced in either strain or sex
of mice. At 500 ppm, a slight nonsignificant weight retardation occurred in both
strains, with a greater effect on the C57BL/6 males and the BALB/c females.
Figure 1 shows the dose-related bladder hyperplasia In BALB/c mice; It was the same
for the C57BL/6 strain. The 100 to 500 ppm gave a 100 percent response. The
BALB/c males had a bladder metaplasia double that of the females which was also
dose-related. One BALB/c male mouse had a bladder carcinoma at 500 ppm and
another at 250 ppm. Adrenal spindle-cell hyperplasia was seen in both strains.
Males of both strains had urinary bladder concretions.
When one considers the 90-day subacute study, it is apparent that the dosage for
urinary bladder hyperplasia was too high; therefore, a serial sacrifice experiment
was designed to study this response. Also another question arose of whether 2-AAF
exposure had to be continuous or if a 3-month exposure would lead to carcinoma 9
months later. Three separate experiments employing 1680 BALB/c mice, each with
an equal distribution of the sexes, were used. The doses fed were 8, 24, 45, 59,
and 86 ppm, with the mice distributed as follows: 480, 480, 240, 160, and 80
animals per dosage level. The same tissues examined in the other experiment
were examined in these experiments. The sacrifice periods were first experiment -
1, 3, 6, 9, and 13 weeks; second and third experiments - 3, 5, 7, 9, and 12 months.
In the second experiment, exposure to 2-AAF ceased at 3 months and in the third
it was continued for the total 12-month period.
Figure 2 shows the hyperplastic response obtained in the males. Whereas
hyperplasia developed In the females only at the two higher doses and at the 9- and
12-month sacrifice periods, hyperplasia in the male made its first appearance at
51
-------
Figure 1
90 DAY.FSED8MG OF 2-FAA !N DIET
100T
0)
o.
X
Z
•o
"0
o.
c
o
u
k.
0)
o.
114/116
x«
I
I
I
I
!47A.38/39
68/72 39/40
/112/120
I
I
I
I
2#2
-------
en
CO
Figure 2
Incidence of Urinary Bladder Hyperplasia
in BALB/c Strain Mice Fed N-2-Fluorenylacetamide
Sacrifice
Interval,
weeks Sex
1
-------
Figure 4
A plot of 5.00 = —9.54 + 7.04 leg (dose) + 2.65 log
(time) with appropriate confidence limits on dose.
£
a
a.
H
UJ
tf>
o
100
30
70
60
50
40
20
Y(prcbit) = -9.54 + 7.04 log (dose)
+ 2.65 log (time)
4
6 8
WEEKS
10 13
20
Y = a + bX '+ cZ (Sq.l )
Var(Y) = 1/Snw + x2/Snwx2+ 2xz/Snwxz + z2/Snwz2 (Sq. 2)
X=X-X (Eq. 3)
z=Z-Z (Eq.4)
3y replacing Y = Y + bx + cz
2A
where
= Y
A
B
C
Y
2(5C - f2z/Snwxz)
(K2- t^/Snw- t^ z2/Snwz2)
Profait of response
Z =
Log jo (dose)
Log -jo
54
-------
Figure 5
Hyperpiasia
Months
Males 3 S 7 9 12
86 6/8 8/8 7/7 8/8 8/8
59 14/16 13/16 13/14 13/15 12/13
PPM 45 20/24 21/23 19/24 21/24 18/22
24 11/48 7/45 16/48 15/42 17'/44
8 - 1/46 1/48 1/44 6/41
Females
PPM
86
59
45
24
8
5/8 8/8 8/8 8/8
2/16 6/15 7/12 10/15
- 1/23 .4/28 0/25
- - - 1/46
8/8
6/16
Figure 6
Males
86
59
45
24
8
CARCINOMA
Months
5 7 9 12
- - 3/8 4/8
- - - 3/13
- - — 1/22
55
-------
3 weeks and Increased thereafter except at the lowest dose. No hyperplasia was
observed in the control groups. A statistical analysis of this male data (Figure 3),
using a maximum likelihood solution, shows that it is possible to take a fixed response
and using a linear equation to solve for dose for a given time or a time for a given
dose and confidence limits placed on either (Figure 4). The only other pathological
change observed was a spontaneous pericardial fibrosis and mineralization which
occurred in all groups, even in the controls. It was seen in 26.4 percent of the
males and 13.8 percent of the females.
In the second experiment, where feeding 2-AAF ceased at 3 months, hyperplasia
cleared up at 5 months and neither it nor carcinoma was observed throughout the
remaining months of the study. In the third experiment, where 2-AAF was fed for
12 months, urinary bladder hyperplasia was observed In both males and females,
with this response appearing earlier in the former (Figure5). Bladder carcinoma
appeared in the three highest doses of the males but not in the females (Figure6 ).
Both the bladder hyperplasia and neoplasia are dose- and sex-dependent and it appears
that hyperplasia is related to total dose consumed and not to dose rate.
In summary, 2-AAF is a compound with a very low acute oral toxicity. When
administered aubacutely, it produces a dose-related urinary bladder hyperplasia
which is sex-related; males are more susceptible than females. Bladder carcinoma
also was sex-related, occurring only in males. Bladder concretions also occurred
only in males. Adrenal spindle-cell hyperplasia was observed in both BALB/c and
C57BL/6 strain mice. Spontaneous pericardial fibrosis and mineralization were
observed in male and female BALB/c mice.
References
1. Haley, T. J., Dooley, K. L. and Harmon, I. P.: Proc. Soc. Exp. Biol, Med.
143: 1117(1973)
2. Haley, T. J., Schieferstein, G., Harmon, J. R., Dooley, K. L., -Jaques, W. E.,
Frith, C. and Farmer, J. H.: Proc. Soc. Exp. Biol. Med. 146; 648(1974)
3. Haley, T. J,, Schieferstein, G., Jaques, W. E., Farmer, J., Frith, C. and
Sprawls, R. W.: J. Pharm. Sci. 63: 1946 (1974)
56
-------
A STUDY OF AGE SENSITIVITY TO THE CHEMICAL CARCINOGEN
2-ACETYLAMINOFLUORENE*
Dr. David L. Greenman**
The appearance of tumors during chronic administration of a chemical carcinogen
frequently occurs only after an extended period of exposure to the carcinogen. In many
instances it is unclear whether the length of the "latent period" is inherent to the carcin-
ogenic process or is related to factors outside of carcinogenesis per se. For example,
numerous developmental and aging changes are known to occur from the time of birth
until death of any given animal. Such changes range all the way from alterations in the
immunological capacity of the animal to changes in the concentration and/or indueibility
of certain liver enzymes. Conceivably almost any of the physiological or biological
changes that occur with aging could bring about a change in an animal's response to a
given carcinogen. In general, studies of the effect of age on the sensitivity of animals
to chemical carcinogens have compared neonatal animals with older animals but have
neglected to compare animals treated at intermediate times between weaning and old
age.
The major objectives of this study were 1) to detect differences in susceptibility to
induction of urinary bladder tumors by 2-acetylaminofluorene (2-AAF) at three different
post-weanling periods in the life of BALB/c female mice, and 2) to determine whether
progression of these tumors following withdrawal of the carcinogen is an age-dependent
phenomenon.
The BALB/c female mice used in the experiment were speciflc-pathogen-free animals,
cesarean-derived, and produced at the National Center for Toxlcologlcal Research.
They were approximately 4 weeks old at the start of the experiment and were housed
four mice per cage. The animal room was maintained as a clean conventional room
*Co-authors are A. Booth and C. Frith
**National Center for Toxicological Research, Jefferson, Arkansas
57
-------
with entry restricted to those associated with the experiment. A face mask, gloves,
cap, clean coveralls, and clean shoes were required apparel for those who entered.
Animals, soiled bedding samples, and drinking water samples were routinely removed
from the room for diagnostic analysis to detect the entry of pathogens into the colony.
Pelleted autoclavable diet was used during periods when mice were on control feed.
Autoclavable meal was used for some of the control animals and during the 6-month
period of carcinogen treatment for all animals receiving 2-AAF. The carcinogen was
dissolved in ethanol and sprayed onto the meal in a drum mixer. Ethanol was removed
by evacuation and the final concentration of 2-AAF was 500 ppm. This feed was given
to the mice in a feeder especially designed to prevent spillage so that food consumption
could be readily measured. The major endpoints to be examined were the extent of
urinary bladder hyperplasia and the presence of bladder tumors. Approximately 30
other tissues were also examined.
The experimental design (Figure 1) may be divided into three major age periods. In
each period, the mice were fed 2-AAF at a concentration of 500 ppm for exactly 182 days.
This period is represented in Figure 1 by the diagonal shading.
Figure 1
1633
624
AGI SINSITIVITY: IXPRIMINTAL DESIGN
Tr*a»m«n» Sch«dul«
12 T8
Timw (monthe)
58
30
-------
In Period 1, 2-AAF was fed from the start of the experiment for 6 months. In
other words, for this period, treatment was started when the animals were 1 month
old and terminated when they were 7 months old.
In Period 2, mice were fed 2-AAF for 6 to 12 months into the experiment and
they were, therefore, 13 months old when treatment was terminated.
la Period 3, the carcinogen was administered during the experimental period
from 12 to 18 months.
Each of these periods may be further subdivided. One experimental group was
killed immediately at the end of the 6-month feeding period. The remaining animals
were placed on control rations and one of these groups was killed 3 months later;
a second group, 6 months later; and the final group was left on control diet to finish
out the remainder of the lifespan.
Control animals were included to match each of the sacrifice times for the 2-AAF
treated animals. Three groups of controls were given the meal diet for a 6-month
period to correspond with the three periods'of;2-AAF treatment. These control
groups receiving meal are represented by the cross-hatched bars of Figure 1. The
total number of animals on experiment was over 4000.
Data presented here must be considered of a preliminary nature since the experi-
nientis only now at tne 18-month stage and data already in hand are still in the process
of verification and analysis.
The effect of 2-AAF on body weight can be seen very readily in Figure 2. The
rate of weight gain was significantly depressed in weaned animals fed a diet con-
tain111*? 2-AAF. Changing to control feed brought about a rapid increase in body weight
but did not completely restore body weight to the control level. Conversely, animals
given 2-AAF after 6 months on control diet quickly lost weight, then failed to gain or
lose weight during the rest of the treatment period. Weight gain in control animals
is indicated by the open circles. During the first 6--month period these animals
received pelleted feed. They were fed control meal in the specially designed Center
feeder during the second 6-month period. The change in diet form and mode of
did not significantly alter weight gain in these animals.
59
-------
32
30
28
26
22
o
(0
20
Figure 2
EFFECT. OF DIET CHANGE ON BODY WEIGHT
o0o0oooo
.„•••
go
AA'
O A'
V
18|- A*
16-
' = change in diet
0 = control
• = 2-AAF (26—52 weeks)
A= 2-AAF (0-26 weeks)
10
50
20. 30 40
Time- on Experiment (weeks)
Effects, of 2-AAF on weight gain may be at least partially explained by changes in
feed.consumption.,. In all cases :2-AAF decreased total weekly: feed consumption (Table
1). However, this effect of the compound was most striking,inthe.youngest,animals
and became.progressively less,apparent;as animals aged.
Table. 1: Effect .of 2-AAF on Feed.,Consumption in BALB/c Females
Treatment Average Feed C onsumption
(g/mouse/week)...
Control
2-AAF
0-6 mos.
21.2
18.3
6-12 mos.
21.7
19.3
12-18 mos.
21.1
19.9
When,'feed consumption is expressed relative to body weight (Table 2), food
intake is depressed by 2-AAF only in the youngest animals>: Even here it.should
be noted that the amount of food, consumed per:unitbody weight was greatest .inthe
young animals. Thus, the rate of exposure, to 2-AAF when-expressed in terms rela-
tive to body weight was greatest in this group; There was virtually no difference
between;the two older, groups in the rate of their, exposure, to the carcinogen.
60
-------
Table 2: Effect of 2-AAF on Feed Consumption in BALB/c Females
Treatment Average Feed Consumption
(g/g body weight/week)
Control
2-AAF
0-6 mos.
0.93
0.85
6-12 mos.
0.73
0.72
12-18 mos.
0.69
0.71
Outside of the effect on body weight, 500 ppm of 2-AAF had little apparent
toxicity during the period of treatment. Mortality (Table 3) during the treatment
period did not significantly differ from mortality among the control animals. Six
to 12 months after termination of 2-AAF treatment, however, mortality of the
treated animals is significantly greater than that for the controls. This is especially
apparent in the last column of Table 3. To a large degree the higher mortality in
these groups can be accounted for by animals dying with urinary bladder carcinomas.
Table 3: Effect of 2-AAF on Mortality in BALB/c Females
Treatment
Control
2-AAF
2-AAF
2-AAF
The major histopathologic findings of interest in the control animals sacrificed
during the course of the experiment are lung alveolar cell adenomas and uterine polyps
/Figure 3). Each of these lesions is present in about 30 percent of the control popula-
tion at 19 months of age. Both of these lesions also have been seen in animals treated
with 2-AAF. However, the carcinogen neither seems to increase nor decrease the
prevalence of the lesions. To date, of 300 or more control animals that have been
examined, only one had bladder tumor and only 2 or 3 animals have exhibited any
hvperPlasia of the urinary bladder.
61
Time of
Exposure
0-6 mos.
6-12 mos.
12-18 mos.
Percent Dead
6 mos.
1.7
2.4
1.8
1.8
12 mos.
4.7
6.6
2.4
3.0
18 mos.
12.0
34.3
20.8
10.7
-------
Figure 3
PATHOLOGY IN BALB/c FEMALE CONTROLS
at
"3
60
50
&? 40
20
10 -
-•.- =
Uterine Polyp
Lung .Alveolar Cell Adenoma
-'W—>f
8 10 12 14
Approximate Age (Months)
16 18
Virtually all of :the animal? killed immediately after 6 months of exposure to
2-AAF exhibited .hyperplasia .of the urinary bladder (Table 4). This .observation
is uniform for animals of all three age groups examined. On the other hand,
severity of the hyperplastic response is an.age-dependent phenomenon. This is
demonstrated by the ratio of Grade I/Grade 2 hyperplasia for each group. Among
the youngest animals only one of about three has Grade 2, the more severe grade
of hyperplasia. In both groups of older mice, at least 9 of 10:animals have Grade 2
hyperplasia.
Table 4: Bladder Hyperplasia After 6-Month Exposure to 2-AAF
Treatment
Period
0-6 mos.
6-12 mos.
12-18 mos.
Hyperplasia
Prevalence Percent
158/165 96
208/210 99
62/63 98
Grade I/Grade 2
1.71
0.09
0.02
62
-------
Withdrawal of 2-AAF from the diet results in a partial reversal of hyperplasia
(Table 5). Reversal is apparent with respect to the number of animals having
hyperplasia as well as in the severity of hyperplasia. In the younger animals, approxi-
mately half of the population have lost the hyperplasia 3 months after 2-AAF was
removed from the diet. An additional 3 months on control feed has no greater effect.
Reversal of hyperplasia in the older animals also appears to have plateaued 3 months
after removing the carcinogen. While reversal of the severity of hyperplasia is very
striking in this group, a complete loss of the hyperplastic response is seen in only
about 10 percent of this population. It will be very interesting to see in what per-
centage of animals with irreversible hyperplasia, bladder carcinomas eventually
develop.
Table 5: Effect of 2-AAF Withdrawal on Bladder Hyperplasia
Treatment
Period
0-6 mos.
6-12 mos.
Withdrawal
Period
-
3 mos.
6 mos.
-
3 mos.
6 mos.
Hyperplasia
Prevalence Percent
158/165
82/182
97/196
208/210
180/204
99/110
96
45
49
99
88
90
Grade I/Grade 2
1.71
5.15
3.00
0.09
2.60
1.33
The prevalence of transitional cell carcinomas of the bladder after 6 months of
treatment with 2-AAF (Table 6) shows a significant age dependence, both in terms of
percent of the population with tumors and the severity of the grade of carcinoma.
in the middle-aged group during treatment are apparently most sensitive to
carcinogen. This is true when considering only the zero grade carcinomas, but
IS especially obvious when the more severe grade of carcinoma is also Included.
Since only a fraction of the pathology from the oldest aged group has been completed,
we can only tentatively say that these animals appear to be intermediate in their
sensitivity to the carcinogen.
63
-------
Table 6: Bladder Neoplasia After 6-Month Exposure to 2-AAF
Treatment
Period
0-6 mos.
6-12 mos.
12-18 mos.
Transitional Cell Carcinoma
Grade Distribution
Prevalence
6/165
39/210
5/63
Percent
3.6
18.6
7.9
G -0
6/165
28/210
5/63
G - 1
-
11/210
_
Age differences becorii'e even more apparent when examining the progression of
tumors after withdrawal of the carcinogen (Figure 4). Progression in the youngest
animals during the 6-month period after removing the carcinogen is considerably
slower than in the middle-aged group. In the younger group only 20 percent of the
population have bladder carcinomas after the 6-month withdrawal period. This
contrasts sharply with nearly 60 percent prevalence in the middle-aged population
after the same withdrawal period. It is of interest, although perhaps of no signifi-
cance, that the prevalence of carcinomas in both of these groups is virtually iden-*
tical at 12 months. One might conclude from this observation that the latent period
is, indeed, shorter in the older animal.
Figure 4
SO
I 40
a 30
h
o
: 20
o
•9
•
BLADDER TUMOR PROGRESSION AFTER
WITHDRAWAL OF 2-AAF
-a-
= controls
= 2-AAF at 0-6 mo.
= 2-AAF at6-12 mo.
= 2-AAF at 12-18 mo
6 8 10 12
Time on Experiment (Months)
14
16
18
64
-------
In both experimental blocks shown In Table 7, carcinomas appear to progress toward
greater severity with time after removing 2-AAF from the diet. Clearly, the number
of Grade 1 carcinomas increases with time.
Table 7: Change in Bladder Neoplasia After 2-AAF Withdrawal
Treatment Withdrawal Prevalence of Transitional Cell Carcinoma
period Period (Percent of animals killed)
0-6 mos.
3 mos.
6 mos.
6-12 mos.
3 mos.
6 mos.
In conclusion, there are distinct age differences in the sensitivity of BALB/c
female mice to the actions of the bladder carcinogen 2-acetylamlnofluorene. These age-
related differences include differences in the severity and reversibility of bladder hyper-
as wel* ^ differences in the Initiation and progression of bladder carcinomas.
G-0
4
4
11
13
31
34
G-l G-2 G-3 G-4
-
4 - <1 <1
7 1 <1
5
13 <1
23 - 1
QUESTION: You indicated that the latent period for the induction or initiation of the
arcinoma and progression differs with age, but you said that the period was shorter
•n the older animals. Is that correct?
Dp. GREENMAN: It was in the middle-aged group.
QUESTION: Do you have any data, or have you evaluated any data, to show why the
ollnger animals in the other age group — do you attribute this to the immune capacity
f the animals or to the lack of Initiation and progression of the carcinoma in that
e group of animals?
65
-------
DR. GREENMAN: We have no good indication at this point. On the basis of the
difference in response of the two younger groups, one might have suspected differ-
ences in immunological capacity. However, results from the older aged animals
do not tend to support this suspicion. There may be other aging changes that are
involved.
66
-------
BENCHMARK TOXICITY
Dr. Lawrence Flshbeln*
The primary objective of this presentation on benchmark toxicity is to briefly review
&e objectives and subsequent process of decision formulation of the recent National
Center for lexicological Research (NCTR) - EPA agreement, formulated on May 25,
1975, to determine the feasibility of a benchmark approach to the toxicologtcal evalua-
tion of pesticides by relating pesticide common denominator functional group, physical
chemical properties, and spatial configurations to selective lexicological responses.
Three separate tasks will be addressed. Task I stipulates that NCTR will conduct
Preliminary studies to determine the functional groups, spatial configuration, and/or
Physical chemical property categories represented in pesticides which are pertinent to
Predicting common toxicological responses.
Task n is based on results of Task I. NCTR and EPA project officers will select a
^ctional group, spatial configuration, and/or physical chemical property for further
study. This will be extensively studied to determine the suitability of data in the open
ilterature and EPA Registration Petition Files for establishing predictive indexes of
Elective toxicology. A monograph report will be prepared, detailing the state of avail-
at)le information.
Task m deals with benchmark bioassay data. NCTR will conduct targeted bioassays,
to Produce data of a highly comparable character, which are necessary to support the
^ork of Tasks I and n. The categories to be studied and protocol and schemata will be
Jointly agreed to by EPA and NCTR project officers.
The process of decision formulation by NCTR will consist Initially of a critical
of pesticide classification by use category, e.g., insecticide, herbicide,
Center for Toxicological Research, Jefferson, Arkansas
67
-------
or fungicide, chemical classification within the above use categories, and mechanisms
of action.
Tables 1 through 3 depict the major chemical classifications and subcategories within
insecticide, herbicide, and fungicide use categories, respectively. The chemical classes
and moieties range from the relatively simple to the complex and offer fertile ground for
subsequent, more definitive structure lexicological considerations.
Table 1; Classification of Insecticides
I. Chlorinated Hydrocarbons
A. DDT analogs
B. Cyciodiene compounds
C. Benzene hexachloride
II. Organophosphorus Derivatives
A. Pyrophosphates and related compounds
B. Phosphorohalides and cyanides
C. Dialkylaryl phosphates, phosphorothioates and
phosphorodithioates
D. Trialkyl phosphates and thiophosphates
IE. Carbamates
A. Naphthyl carbamates
B. Phenyl carbamates
C. Heterocyclic dimethyl carbamates
D. Heterocyclic methyl carbamates
E. Oximes
Table 2; Herbicides
A. Ureas
B. Triazines
C. Carbamates
D. Bipyridinium derivatives
E. Acylated anilides
F. Hydrazides
G. Chlorophenols
H. Nitrophenols
I. Chlorophenoxy acids and esters
J. Arsenical derivatives
-------
Table 3; Fungicides
A. Dithiocarbamates
B. Ethylene bis-dlthiocarbamates
C. Benzimadazole derivatives
D. Thiophanates
E. Phthalimides
F. Hexachlorobenzene
G. Chlorophenols
H. Chloronitrobenzenes
I. Oxathiins
J. Dinitrophenyl derivatives
K. Quinones
L. Organomercurials
M. Organotins
N. Antibiotics
O. Cationic surfactant
P. Miscellaneous
There are approximately 10 toxicological responses of major importance that we may
choose to evaluate, as shown in Table 4. They are acute toxlcity LE,-0» carcinogenicity,
mutagenicity, teratogeniclty, cholinesterase inhibition, neurological delayed demyelin-
ation, thyroid endocrine, cardiovascular, immunological, hepatotoxic, inhalation tox-
icity, and induction of lung lesions.
Table 4; Toxicological Responses
A. Acute toxicity (LD5o)
B. Carcinogenicity-mutagenicity-teratogenicity
C. Cholinesterase inhibition
D. Neurological (nerve demyelination)
E. Behavioral
F. Inhalation toxicity
G. Hepatotoxic
H. Thyroid-endocrine
I. Cardiovascular
J. Immunological
69
-------
We believe it is important to the initial evaluation of toxlcological response of a
requisite chemical or structural category within a particular use category, that stress
be given to the design of a prototype data collection form which would help collate and
possibly simplify the analysis of mammalian toxicity data.
For evaluation of the acute mammalian toxicity LD , for example, we would stress
50
the 10 salient parameters listed in Table 5. A preliminary assessment of primary and
secondary literature sources indicates that there is an amazing, incredible lack of
uniformity, standards of excellence, etc. in the obtaining of requisite acute oral LD
DU
information.
Table 5; Acute Mammalian Toxicity
LD5Q pose Lethal to 50% of Animals)
Correlate and Tabulate Information Relative to:
A. Species, strain, sex
B. Age and/or weight
C. Purity of test pesticide (how determined), solvent or
vehicle, prep, of solution (ultrasonics)
D. Route of administration and frequency
E. Number of animals used for test and control
F. Duration of observation
G. Method of statistical analysis
H. LD50 value (calculated)
I. Site of action
J. Structural formula, stereochemistry (isomers)
Figures 1 through 3 suggest how structure-activity relationships may be considered
-.v.thia one chemical category — the triazines, for example, of which a number are
significant herbicides. These figures depict the substitution, by various alkyl and
s.ryi. substituent, of one chlorine at a time, in a basic cyanuric chloride structure, the
Backbone of the triazines.
70
-------
Figure 1
Structure-Activity Relationships of Triazine Herbicides:
Substitutions of Chlorine Atoms of Parent Cyanuric Chloride
Cl
N
N
R =
R =
R =
V
R =
= Alkylamine
= Alkylamine
Alkylamine
Dialkylamine
= Dialkylamine
R
R =
Rl =
Alky lam ine or
Alkoxyalky lam ine
Alkoxy alky lam ine
Alky lam ine
Cy anoalkylam Ine
Alkylamine
CycloalkylamLne
Figure 2
Structure-Activity Relationships of Triazine Herbicides:
Substitutions of Chlorine Atoms of Parent Cyanuric Chloride
Class I: Only one chlorine atom Is replaced by one of the
mentioned substituents:
R = 0 - Alkyl
= 0 - Aryl
= S - Alkyl
= S - Aryl
R = NH - Alkyl
Alkyl
= N
Alkyl
= NH - Aryl
71
-------
Figure 3
Structure-Activity Relationships of Triazine Herbicides:
Substitutions of Chlorine Atoms of Parent Cyanuric Chloride
Class HI: All three chlorine atoms are substituted:
R & R = Alkylamines
1 2
R = Alkoxy,
= Alkylthlo,
= Azido,
= Isocyanato,
= Alkylthio, R. = Alkylamine, Rg = Azido
R & R- & R_ = Alkylamines
R & R = Alkylamines
R & R = Alkylamines
1 Z
R., Si, R = Alkylamines
1 £
Analogous considerations could be invoked, for example, for the vinyl phosphate
insecticides, the basic structure of which is shown in Figure 4. The variation in the
group provides a range of not only insecticidal chemical/physical properties, but
obviously, of toxicological properties — the major ones we discussed previously.
We would then, hopefully, correlate the various substltuents wherever the literature
permits1,, with observed toxicological response of choice, for example, acute LDg^
ir;to'tabular summaries including the following information: chemical structure, chemi-
:-ii names, alternate names, and relevant LD information and references.
oU
We are well aware that the number of correlations exist for the correlation of
structure activity with toxicological response. These are, for example, the Hanson
72
-------
Figure 4
Vinyl Phosphate Insecticides
R1© X
2 /
R °
1. R and R are alkyl groups; R , R , and R can be one of a wide
variety of groups including hydrogen, methyl, carboxymethyl, alkyla-
mido, polychloraryl, etc., X is S or 0.
2. The variation in the groups provides the range of insecticidal,
chemical, and physical properties that exists with the class.
3. All currently commercially available members have both R and
R^ either methyl or ethyl and all except one have an oxygen atom as X.
Multiple Parameter Method and the Free-Wilson Technique, both of which Have been
employed with limited success, primarily in the drug industry, in the evaluation for
homologous series of compounds, for specific functions such as antlmalarial or
-rrcinostatic activity.
We don't believe that there exist sufficient analogous data in pesticide areas of
concern at present that would permit the use of these cdrreiative techniques. In
ummary, we believe that the discussed strategy will permit us to better review and
valuate the requisite data bases. This will allow us to more rationally suggest
dditlonal areas of research in order to fill the gaps that our preliminary studies
av indicate and, hence, to assist ultimately in the enhancement of the Substitute
Chemical Program.
73
-------
PHARMACOKINETIC MODELING OF SELECT PESTICIDES
Dr. John F. Young*
The use of pharmokinetic parameters obtained from acute studies to predict
accumulation of chemicals on subacute or chronic exposure has not yet been attempted
in. the field of toxicology. The implication that selected pesticides accumulate in
various biological systems is not based on pharmacokinetic prediction but on single
point determinations. If two animals are from the same chemical environment and
one has a higher chemical residue than the other after serial sacrifice, what does
this really mean? . . . Accumulation? . . . Biological variability? . . . Different absorption
rate constants? . . . Different elimination rate constants? . . . Different distribution
Parameters? . . . Actually, it is very hard to say unequivocally which is taking place.
However, if we take a small group of animals and determine their pharmaco-
tonetic parameters from a single dose, the extent of biological variation can be
Determined, and these pharmacokinetic parameters can be used to predict the tissue
on subacute or chronic exposure to the chemical.
initial pharmacokinetic parameters are obtained by exposing a small group of
a»imals (four to six) to the chemical by intravenous injection. Several dosage levels
are used to determine the effect of the initial amount of the chemical on the pharma-
Co&netic parameters, if any. Blood, urine, and feces levels of the chemical (and
lts metabolites) are determined as a function of time. From these data a model is
tested by simulating the data with a hybrid computer. When the computer has
accurately simulated the actual single animal data, the proper model has been ob-
tained.
^ Figure 1 this plot of amount of chemical vs. time illustrates the computer
Emulation (solid lines) of data (X for amount of chemical in the blood and circles for the
accumulative amount of chemical excreted) and the computer prediction of tissue levels
line).
Division, National Center for Toxicological Research, Jefferson, Arkansas
75 Preceding page Wank
-------
Figui-e 1
* —Amount of choniiccsl in blood
o =: A ecu 11
on
ri
-------
A set of rate constants can be obtained directly from the computer at this
e for each animal individually. A simple model might be as shown in Figure 2.
This is a two-compartment model with first order elimination. The rate constants
^e labeled with the first letter representing the compartment which the chemical
is coming from and the second letter representing the compartment the chemical
is going to. The intravenous information is essential in that it simplifies the model
for the most accurate determination of the pharmacokinettc parameters for a given
m°del. However, very few animals are exposed to pesticides by the intravenous
route. The next step then is to administer the chemical orally or by inhalation, which-
ev«r route has been determined to be the route of environmental exposure. Again
Mood levels and urine and fecal elimination are analytically determined. On sim-
^ating the data with the hybrid computer, the model has now been expanded to include
this absorption.
Flgure 3 shows a two-compartment model with first order elimination and first
°r
-------
oo
Table 1: Estimates of Biological Variation
Absorption Distribution (Return) Elimination
Animal
1
2
3
4
"x"
S.D.
S.D. %
X
KAB
0.316
0.295
0.310
0.280
0.300
0.016
5.4
Accumulation
KBT
0. 189
0.157
0.185
0.146
0.169
0.021
12.4
KTB
0.023
0.018
0.020
0.015
0.019
0.003
17.7
KBE
0.436
0.387
0.400
0.360
0.396
0.032
8.0
-------
Figure 2
Tissue
to
ki
iliminerfion
Figure 3
Tissue
Blood
Elimination
-------
Figure 4
00
o
16
-------
Figure 5
i
o
E
I
B
20
40
Time
60
80
-------
lower set of curves is for 10 doses, the upper curves for 50 doses, and the continued
dashed line is for one animal continued for 100 doses. Even after 100 doses, a plateau
has not been reached. This particular set of data would not even be classified as a
"deep" compartment.
Figure 6 is a plot of amount vs. time showing the blood level, tissue level, and
accumulated excretion of the chemical. The lower primed curves are from a single
dose and the non-primed curves are after multidosing for 10 doses. The same phar-
macokinetic parameters are used for both sets of curves. This is a good illustra-
tion of how a seemingly low level of exposure can very quickly accumulate to a fairly
high level in the biological system.
If we alter our rate constants to favor the accumulation in the tissue to a greater
degree making the tissue represent a deep compartment, the only difference in our
pharmacokinetic parameters (Figure 7) is that the return from the tissue has been
decreased by a factor of 10. From this it can be seen that a higher level is obtained
and maintained for a much longer time period.
The accumulation data on multidosing or multiexposure can be predicted from
the single dose exposure studies. This not only provides a decrease in the cost and
time involved in such a study, but allows for prediction beyond what would be prac-
tical to experimentally determine.
For any new pesticide that is proposed to replace an older established pesticide,
the potential to accumulate must be known. The single dose pharmacokinetics and
prediction are rapid means by which the old and new pesticide may be evaluated on
similar and easily controlled experimental grounds.
82
-------
Figure 6
ot>
10
8
<
B
^
E
kBT
kTB
— 1
Time
-------
Figure 7
•
-
0
= 10
Time 20
30
40
-------
DR. CUR LEY: You indicated that the latent period for the Induction or initiation
of the carcinoma and progression differs with age, but you said that the period
shorter in the older animals. Is that correct?
DR. GREENMAN: It was in the middle-aged group.
DR. CURLEY: Do you attribute this to the immune capacity of the animals or to
the lack of development of enzyme systems, or to this lack of initiation and progres-
sion of the carcinoma in that age group of animals?
DR. GREENMAN: We have no good indication at this point. On the basis of the
difference in response of the two younger groups one might have suspected differences
in immunological capacity. However, results from the older aged animals do not
tend to support this suspicion. There may be other aging changes that are involved.
85
-------
PESTICIDE RESIDUES IN HUMAN MILK
Dr. Eldon P. Savage*
The objective of this study is to estimate the levels of organochlorine pesticides in the
HI Ilk of nursing mothers giving birth in United States hospitals on a nationwide basis. As
most of you know, in 1951 Longjjt jO. (1) reported DDT in 30 of 32 human milk studies
collected in Washington, D. C. The current study is designed to try to analyze milk sam-
ples from 1,600 mothers from across the country so that we get, for the first time,
some national levels of organochlorine pesticides In human milk.
The participating groups in this study are Colorado State University, the University of
South Carolina Medical School, Mississippi State University, and the two State Health
pepartments of Michigan and Utah. All of these groups have participated in the community
studies on pesticides. They all have laboratories that have been in the laboratory quality
control program that is operated by the EPA.
One of the reasons that this study was done Is that we had worked up a design for a
study on the incidence of acute pesticide poisonings In hospitals across the United States,
and that this design is based on the 7,000 plus hospitals that are listed by the American
jjospital Association. A sample of the 783 hospitals that were visited in the acute pesti-
cide study was used as a clustering point where we could obtain a list of nursing mothers.
Ideally* what we would obtain would be a list of all the nursing mothers In the United
States during a specified time period. Obviously, such a list is not available. Approxi-
mately 150 hospitals were randomly selected and these hospitals then became a cluster
point for a list of women's names who have given birth in these hospitals during the past
ar. Then from that list, the number of mothers who are nursing is determined and
are contacted to obtain a sample of milk from a particular group.
Epldemiologic Pesticide Studies Center, Fort Collins, Colorado
87
Preceding page blank
-------
Within each pesticide usage level within the five geographical regions, the general
hospitals with nursery facilities were stratified according to the number of births in the
hospitals for the 1973 calendar year. The sample size selected from a particular
hospital is determined by the number of births in the hospital in 1973. The sample
size range is from two to three samples per hospital up to approximately 100.
The total number of samples that we thought we could handle in this program, in ac-
cord with the amount of monies available, was approximately 1,600. Personnel from
Colorado State University covered the Western half of the country, and people from the
Michigan State Health Department covered the Northeast. Mississippi State University
is covering about five states in the South Central Region and the University of South
Carolina is covering the rest of the Southeast.
At the time that the nursing mother is visited, a sample of approximately 10 ml is
obtained. At the same time that we are obtaining the milk samples, we are also obtain-
ing information regarding age, the collection point, the occupation of the household head,
race characteristics, as well as number of siblings, plus the highest school year com-
pleted. In order to gain more knowledge regarding the pesticide usage patterns of the
nursing mothers sampled, this study will incorporate a pilot study to survey household
pesticide usage. A stratified random sample of 10 percent of the participating nursing
mothers will be drawn subsequent to a preliminary analysis of the results of this study.
As currently conceived, 5 percent of the participants with the highest levels and 5 per-
cent of the participants with the lowest levels of organpchlorine pesticides will be re-
visited by a field epidemiologist to determine their patterns of pesticide usage. A pesti-
cide usage form will be completed on each household.
Another important aspect of this study Is to determine the levels of pesticides in milk
from nursing mothers over time. A subsample of approximately 40 women will be drawn
in the Colorado Center area and samples will be collected at approximately 7 days, 4
.weeks, and 3- to 6-month Intervals to determine If pesticides in milk remain constant,
increase, or decrease over time.
88
-------
Analyses In this study will include the following pesticides and metabolites: dieldrin,
chlordane, heptachlor, and heptachlor epoxide oxychlordane. When sufficient milk is
collected, this milk will be frozen and stored for future analyses for the poly chlorinated
biphenyis (PCB).
^vgllty Control
To check the quality of the survey work performed by the nurse participants, 5 per-
cent of the participants will be checked by telephone. This will serve as a check on
thoroughness of the nurse In obtaining pertinent data from the study participants. This
activity is extremely important since the hospital nurses will receive only the training
the field epidemiologist at the time of the Initial contact with the hospital.
The study got underway in early April with a training session for the field epidemi-
iQgists participating in the study. About 2 weeks later, we had a meeting of the head
hedists from the participating laboratories to discuss analytical methodology and
control. At this time 149 of the hospitals In the United States were contacted,
this includes Alaska and Hawaii. We have completed sampling of 26 percent of the
tai hospitals. Currently, 19 percent of the 1,600 milk samples are In the laboratory
j ia some step in the analytical procedures.
Samples have been positive for DDT, BHC, dieldrin, oxychlordane, andHCB. We
als° looking for mirex.
The design for analysis has been developed, and we are going to be doing cluster
originally, plus chl square, plus some variance analysis.
, doubt very seriously with our time schedule that we will have the follow-up of the
GO mothers on pesticide usage done before early next year.
interesting thing I probably should say about this particular study Is the fact that,
to starting It, we heard all types of stories about the nursing mother in the United
tbat tnls group nursed and that this group didn't nurse, and that we probably
teSt
idn't be able to get samples from some hospitals. And If there's one thing that
11*
89
-------
we've learned from the hospitals that we've contacted to date, it is that there really is
no pattern. Every hospital has been different. In some hospitals we find that practically
every woman who gives birth in that hospital is nursing. We can go to another hospital,
maybe a fairly large hospital where we need 25 samples, and we find very few nursing
mothers.
We have found cooperation to be very good. The one problem that we have had is that
some of the hospitals are very busy, and they may not want to participate. When we en-
counter this, we obtain a list of the women who have given birth, and then by a random
design, we contact women from this list.
We hope to have the field work on sample collection done in the next 90 days. And
this means that the laboratories that are doing the analysis are really going to be busy.
But most of them are geared to where they can run a number' of samples in a short period
of time. And hopefully we'll have this study completed in the next few months.
QUESTION: Were you trying to relate the source of pesticides in the milk to other
exposures ?
DR. SAVAGE: We don't have that, based in this study, with the exception of those
ISO women whom we'll be following up, the ones who have the highest residues and
i:hft ones with the lowest. And we will get some good information on that particular
question.
DB. CRANMER: Since the loss of pesticides from the body fat will be a very major
contributor in this phenomenon, and you are following up these women, are you making
estimates of weight changes or some other index of change in the fat metabolism of
.re women involved?
90
-------
. SAVAGE: We are not, but we are doing the milk on a fat basis.
DR. CRANMER: Are you following the fat metabolism of the individuals?
DR- SAVAGE: Ideally, we would need something of this nature, but logistically
and money-wise, we thought the best we could do was take these 40 women in the
Colorado area, and try to follow them over time.
91
-------
EFFECT OF SUBSTITUTE PESTICIDES ON
HORMONE DEPENDENT TISSUE
Dr. Sydney A. Shain*
Our program was initiated about 3 weeks ago; consequently, I will have no data
to present. What we would like to do this morning, however, is to present some of
the rationale for the design of the study and some of the precedences in the literature
on which the study is based. Basically, our program is a pilot study to determine
whether or not we can develop a predictive model for the chronic effects of pesticide
in steroid hormone-dependent tissue.
The potential of certain pesticides to deleterlously affect reproductive function has
long been recognized. In 1950, Burlington and Lindeman (5) reported that DDT caused
inhibition of both testicular growth and the development of secondary sex characteris-
tics in cockerels. More recently, Dlechmann et al, (6) have observed diminished re-
productive activity in male beagles receiving aldrin, which, in part, appeared to be
the consequence of a suppression of libido. In addition, Esplr et jil. (8) have reported
the occurrence of Impotence among four of a group of five farm workers following
exposure to pesticide. These types of observations imply a possible deleterious
action of pesticide on accessory organs of reproduction, at extra-organ sites, or a
combination of multiple effects at organ and extra-organ sites.
QJ order to appreciate some of the potential multiple sites of pesticide action, it is
necessary to briefly review the mechanism whereby steroid hormones regulate
accessory reproductive organ fimction. A general model for steroid-hormone regula-
tion of cellular function in hormone-dependent tissue has evolved as a consequence of
extensive studies with numerous hormone-dependent tissue models (11, 15). This
general model, as it applies to the regulation of function in male accessory organs °f
reproduction, is schematically represented in Figure 1. It is seen that the testicular
hormone, testosterone, circulates in the extracellular plasma. In the accessory
organs of reproduction, plasma-borne testosterone enters cells by a non-facilitated
Foundation for Research and Education, San Antonio, Texas
93 Preceding page blank
-------
passage across the plasma membrane. Once within the extranuclear space of the
hormone-dependent cells, testosterone is rapidly metabolized to the accessory sex
4.
organ-specific hormone, 5 «-dihydrotestosterone, by the enzyme A -3-ketosteroid-
5 ar-oxidoreductase (4, 14).
Figure 1
Pathways of Testosterone Metabolism in the Rat Prostate
Pathways of prostate androgen metabolism are represented in Figure 2. Under
normal physiologic conditions, 5«-dihydrotestosterone is the principal prostatic
androgen, whereas the other androgens depicted in Figure 2 are minor components
of the total androgen content of the prostate. In the rat ventral prostate, 5»-dihydro-
testqsterone concentration is approximately 1.5 times greater than that of testosterone
(17). A quantitatively similar distribution of androgens is found in normal canine (9)
and human prostate (21). It thus appears that testosterone should be considered to
.be a prehormone with respect to steroid regulation of male accessory sex organ
function.
94
-------
Figure 2
POSTULATED MECHANISM OF ANDROGEN ACTION IN THE PROSTATE
The next step in the sequence of events (Figure 1) involves the highly steroid-
gpecific (18), high-affinity binding of 5 «-dihydrotestosterone by a class of cyto-
plasmic macromolecules, the cytoplasm!c androgen receptors. These intracellular
j^acromolecules are thought to be the component ultimately responsible for regulation
Of cellular function. The binding of 5 «-dihydrotestosterone by the cytoplasmic
receptor, R , to form the steroid-hormone complex permits or facilitates a change
jjj receptor conformation with resultant formation of a neonuclear steroid-receptor
coraPlex» RCTT' As the name ^PUes, the neonuclear steroid-receptor complex is
capable of transport across the nuclear membrane. Once within the nucleus, the
jiuclear steroid-receptor complex binds to highly steroid-receptor complex-specific
chromatin sites and subsequently permits the initiation of transcription of DNA
sequences into RNA copies which are translated into newly synthesized proteins.
Multiple sites of potential pesticide alteration of .androgen regulation of prostate
Auction are aPParent* The model implies that prostate competence for androgen
4 a determined by the intracellular content and distribution of androgen receptor. Conse-
uently, an action of pesticide which results in a change in receptor content or distri-
would indicate a change in prostate competence with respect to potential androgen
95
-------
regulation of tissue function. Such an alteration in prostate competence could be
directly achieved by the binding of pesticide to prostate androgen receptor with the
subsequent formation of active or inactive pesticide-receptor complexes. Precedence
for this mode of pesticide action exists since dieldrin has been demonstrated to diminish
cytoplasmic and nuclear binding of 5 *-dihydrotestosterone in rat ventral prostate
preparations (25) and o,jD^-DDT has been demonstrated to be a specific inhibitor of
5 «-dihydrotestosterone binding to rat ventral prostate cytoplasmic androgen receptor
(26). These in vitro studies are consistent with the results of acute in vivo studies
which have demonstrated a diminished accumulation of testosterone by the mouse
anterior prostate after exposure to jj^pJ-DDT (22) or dieldrin (24).
A second effect of pesticide at the level of the prostate could be achieved by a
change in the capacity of this tissue to metabolize testosterone to 5 «-dihydrotestosterone.
Although a number of acute in vivo exposure studies have failed to reveal a significant
effect of pesticide upon the production of 5 «-dihydrotestosterone by the mouse anterior
prostate (7, 23, 24), the addition of pesticide to in vitro preparations of mouse anterior
prostate has demonstrated parathion reduction (23) and carbaryl augmentation (7) of
5 a-dihydrotestosterone production.
In addition to these possible effects of pesticide at the level of the prostate, it is
clear that pesticide can also alter prostate competency by affecting the extraprostatic
metabolism of testosterone. The absolute dependence of the young, mature prostate
upon peripheral testosterone for the maintenance of normal structure and function is
established beyond doubt (16). Consequently, the well-documented capacity of numerous
pesticides to increase peripheral metabolism of testosterone by hepatic mixed-function
oxidaaes (1, 7, 10, 12, 24) indicates the potential of these compounds to alter prostate
competence. In addition, certain of the organochloride pesticides must be considered
to possess the potential to alter peripheral levels of testosterone through changes in
the pituitary-^gonadal axis .as evidenced by the Ability of dieldrin to increase serum
luteuiizjing ;hormone levels in male rats (2). These potential changes in peripheral
testosterone levels would be expected to alter prostate androgen-receptor content as
evi&suced ,by the observation that orchiectomy of the rat ultimately results in a decrease
in-jjihy prostatic androgen-receptor content (3, 13) and that the known aging-associated
96
-------
diminution in the androgen-receptor content of the rat ventral prostate (19) may be
related to the concomitant aging-associated diminution in plasma testosterone
content (20).
We have attempted to indicate the most prominent known potential sites of action
of various pesticides which could result in a change in the status of accessory organs
of reproduction. Using the prostate as a model for the potential effect of pesticides
upon male accessory organs of reproduction, it appears that the ultimate expression
of the multiple potential effects of pesticide would be a change in the androgen-receptor
status of the prostate, This program attempts to develop the rodent prostate as a
predictive model of the potential effects of pesticides upon accessory organs of repro-
duction and other hormone-dependent tissue. To this end, we will evaluate the effect
of chronic ingestion of pesticides upon prostate androgen-receptor levels and distri-
bution and upon selected biochemical parameters which are specific measures of pros-
tate competence. Furthermore, In order td attempt to distinguish extraprostatlc
effects of pesticide from those which occur at the prostate, we will evaluate plasma-
testosterone levels for correlation with the measures of prostate competence.
Abernathy, C. O., Hodgson, E., and Guthrie, F. E.: Blochem. Pharmacol. 20;
2385 (1971)
Blend, M. J., and Lehment, B. E., In: "Pesticides and the Environment, A
Continuing Controversy, " W. B. Deichmann (ed.), Intercontinental Medical Book
Corporation, New York, 1973, p. 189
Boesel, R. W., and Shaln, S. A., submitted for publication, 1975
3»
Bruchovsky, N., and Wilson, J. D.: J. Blol. Chem. 243;2012 (1968)
Burlington, H., and Llndeman, V. F.: Proc. Soc. Exp. Biol. Med. £4:48 (1950).
o»
Deichmann, W. B., MacDonald, W. E.: Beasley, A. B., and Cubit, D., jfad. Med
6' Surg. 10:10 (1971)
pierlnger, C. S., and Thomas, J. A.: Environ. Res. 7;381 (1974)
1•
Espir, M. L. E., Hall, J. W., Shirreffs, J. G., and Stevens, D. L.s Brit. Med.
&' S. 1:423 (1970)
Gloyna, R. E., Slltteri, P. K., and Wilson, J. D.: J. Clln. Invest. 49jl746 (1970)
9.
Hart, L. C., and Fouts, J. R.: Proc. Soc. Ex&. Biol. Med. 114:388 (1963)
97
-------
11. King, R. J. B., and Mainwarlng, W. I. P.: Steroid-cell Interactions, University
Park Press, Baltimore, 1974
12. Kuntzman, R., Welch, R., and Conney, A. H.: Adv. Enzyme Regul. 4:149 (1965)
13. Mainwaring, W. I. P., and Mangan, F. R.: J. Endocrinol. 59;121 (1973)
14. Moore, R. J., and Wilson, J. P.; Endocrinology 93;581 (1973)
15. O'Malley, B. W., and Means, A. R.:"Receptors for Reproductive Hormones,"
Plenum Press, New York, 1973
16. Price, D.* and Williams-Ashman, H. G., Jn: "Sex and Internal Secretions, "
W. C. Young (ed.)» Vol. 1, 3rd edition, Williams and Wilkins Co., Baltimore,
1861,p.366
17. Robel, P., Corpechot, C., and Baulteu, E. E.: FEES Lett. 33-.218 (1973)
18. Shaln, S. A., and Boesel, R. W.: J. Steroid Biochem. 6:43 (1975)
19. Shain, S. A., Boesel, R. W., and Axelrod, L. R.: Arch. Biochem. Biophys.
167;247 (1975)
20. Shain, S. A., and Wilson, L. M., in preparation (1975)
21. Siitteri, P. K., and Wilson, J. D.: J. Clin. Invest. 49:1737 (1970)
22. Smith, M. T., Thomas, J. A., Smith, C. G., Mawhinney, M. G., and Lloyd,
J. W.: Toxicol. Appl. Pharmacol. 23_:159 (1972)
23. Thomas, J. A., and Schein, L. G.: Toxicol. Appl. Pharmacol. 29_:53 (1974)
24. Thomas, A. J. ,-and Lloyd, J. W., In: "Pesticides and the Environment, A
Continuing Controversy, " W. B. Deichmann (ed.), Intercontinental Medical Book
Corporation, New York, 1973, pi 43
25. Wakeling, A. E., Schmidt, T. J., and Visek, W. J.: Toxicol. Appl. Pharmacol.
25:267 (1973)
26. Wakeling, A. E., and Visek, W. J.: Science 181;659 (1973)
98
-------
DR. CURLEY: In your report, Dr. Shain, I don't think you indicated to us what
compounds would be tested. Could you tell us what these compounds are? What
the route of administration will be? What the period of administration will be?
pR. SHAIN: We will be using chlordane, heptachlor, methoxycolor, parathion,
and carbofuran. The route of administration will be in the diet for a
of 3 months. Each pesticide will be administered to 42 animals and analy-
3 js will be on animals in groups of six.
pR. CURLEY: In your introduction you mentioned three studies, one on DDT,
one on dieldrin. You mentioned one study that indicated impotence in some group
Of test animals. Which study was this ?
DR. SHAIN: The impotence studies were in 1970 by Espir, and they were in a
roup of four or five adult male farm workers who had been exposed repeatedly to
esticides and herbicides. The impotence was reversible upon treatment with
methyl testosterone. The time required for full recovery— of course there's a
ertain amount of psychological recovering necessary— varied from 4 months to
over 12 months.
QUESTION: I think in some in vitro studies with estrogen-dependent tissues,
.. has been shown that there is a heat-dependent conversion of the estrogen receptor
Qin one form, or one conformation, to another. To your knowledge, has this
en shown in androgen-dependent tissues?
pR. SHAIN: The activation step that you're discussing has been demonstrated,
s you said, for the estrogen receptor and in addition for the progesterone and
lucocorticoid receptor. To my knowledge, there is no similar direct evidence
the prostate androgen receptor, although it too is capable of undergoing some
fhe transformations that have been shown for the estrogen, progesterone, and
of w
cOCorticoid receptors. The general theory would indicate that the androgen
«.otor would also undergo this transformation; however, the evidence has not yet
^V
* Presented.
99
-------
QUESTION: Will you attempt to determine turnover rates of the androgen-receptor
complex in this particular study? And also, if you use the Scatchart Plot, will you
be able to tell us anything about the site of the binding of that particular compound
on the receptor complex? You'll be using labeled compounds, I would assume.
Is that correct?
DR. SHAIN: We will not use radioactive pesticides in any of our studies since they
are not available in suitable preparations. But, in order to obtain some direct infor-
mation relevant to the questions which you have asked, we will perform a series of
direct in vitro determinations which are designed to evaluate the mechanism of inter-
action of the pesticides with the receptor proteins. The work of Wakeling has indi-
cated that at least with DDT the interaction may be a noncompetitive type of inhibition.
100
-------
EFFECTS OF PESTICIDES ON BLOOD
LIPOPROTEINS, ARTERIES, AND CARDIAC MUSCLE
Dr. Jack E. Wallace*
This project is under the direction of Dr. Henry C. McGill, Jr., Professor of
Pathology at the University of Texas Health Science Center in San Antonio. The
information we are presenting consists principally of Dr. McGilTs concepts and
approach to this problem.
Because of the widespread use of pesticides and the possible exposure of humans
to these chemicals over long periods of time, it appears desirable to determine
whether or not pesticides in current use, or proposed for future use, affect the
cardiovascular system in small doses over long periods of time. In particular,
it is desirable to know whether these compounds affect physiological systems and
organs in such a way that they may potentiate or accelerate common cardiovascular
diseases to which humans are prone.
Arteriosclerosis and its sequelae are the most common cardiovascular diseases
in humans in the technically developed countries. The process starts in infancy and
develops clinical disease manifestations when individuals reach middle age or later.
In this study, we will use the baboon as the model to approximate those conditions
which occur in man. When the baboon's diet is made to resemble human diets in the
technically developed countries, that is, high in saturated fat and cholesterol, fatty
streaks develop in the arterial wall that closely approximate atherosclerotic lesions
that occur in humans. If that diet is continued over a long period in the baboon,
these fat deposits begin to resemble very closely the lesions that occur in humans.
There are several reasons for using a nonhuman primate for this experiment,
and particularly the baboon. The rat is resistant to dietary cholesterol and does
not develop atherosclerosis on a high-cholesterol, high-saturated fat diet. The
rabbit is just the opposite. That animal is extremely sensitive to dietary fat and
Trhe University of Texas Health Science Center, San Antonio, Texas
101
-------
cholesterol, much more so than the human. The dog must be subjected to thyroid
ablation in addition to being fed an atherogenic diet to develop:atherosclerosis. Some
species of pigeons develop atherosclerosis spontaneously. Swine.>also arexuseful,
but are large and difficult to'handle. Because atherosclerosis appears to involve
many complex interactions, many investigators are turning to nonhuman primates
in order to simulate the human situation as closely as possible. The baboon is
the largest readily available nonhuman primate, and its sensitivity to an atherogenic
diet is very similar to that of the human.
The Southwest Foundation for Research and Education (SFRE), which is.engaged
in a joint effort with the Medical School at San Antonio for this study, is well known
for its baboon colony. Dr. ^McGill, the principal investigator, has.developed
extensive programs of research in atherosclerosis in the recent years involving
currently over 400 baboons at;SFRE. These animals, the .facilities required-to
house them, and the personnel that are skilled in handling them represent a unique
resource.
The baboons are housed in gang cages for breeding and for long-term studies,
and in individual cages for studies that require frequent collecting of: blood, mea-
suring of blood pressure, jsrtly from egg yolk, as in human diets, and the remainder from USP:crystalline
.•;; :--,terol. The dietary fat will be provided by lard at a level;of'40 percent of total
102
-------
calories, identical in total fat to the U.S. diet. This same diet is being used in other
research projects at SFRE to investigate the effects and mechanism of action of other
risk factors on experimental atherosclerosis, such as cigarette smoke inhalation,
blood pressure, social stress, and glucose intolerance.
The specific aim of this study is to determine whether pesticides affect the blood
lipoproteins that are involved in the pathogenesis of atherosclerosis, or whether they
accumulate in the fatty deposits of atherosclerotic lesions of the arterial wall and
influence their course, or whether they affect the metabolism of the myocardium.
There are several fragmentary reports from the biomedical literature from Russia
and Eastern Europe that indicate that they do change the plasma levels of lipoproteins.
These lipoproteins, which carry cholesterol, cholesterol esters, triglycerides, and
phospholipids in plasma in a stable form, are involved in the pathogenesis of ather-
osclerosis because certain ones (particularly low density, or ,Slipoprotein) are
elevated in persons who have higher rates of coronary heart disease. It is possible
that alterations in the concentrations of lipoproteins, or alterations in their nature,
could influence the rate of progression of atherosclerosis. Also, it is possible
that the fat-soluble pesticide compounds and their breakdown products may become
concentrated in the lipids of the atherosclerotic lesion and influence its course.
The pesticides to be tested are chlordane, parathion, carbofuran, and diazinon.
In a preliminary range-finding experiment, we are attempting to find the levels of
•ntake of these four pesticides that will exceed the allowable daily intakes for humans
DV at least an order of magnitude or more and that will still not produce acute toxic
ffects. la the definitive study, we will feed each pesticide at two dose levels to
. baboons, three males and three females. There also will be a control group that
•n receive the same diet without added pesticide.
The animals will be studied under basal conditions for one month while on a stan-
dard stock diet which is low in fat and cholesterol. Thereafter, they will be placed
the high-fat, high-cholesterol diet without pesticides for 3 months, and their re-
nse to the diet will be measured. It is anticipated, on the basis of numerous other
eriments with these animals, that the serum cholesterol will rise to about 250-350
103
-------
mg/dl, which is the range of hyperlipidemia encountered in most humans in the
United States. There will be a wide range of variability in these responses to the
diet, just as in humans. Among the lipoproteins, the low density lipoprotein
(/? lipoprotein) will be the principal one to rise.
After the 4 months of baseline studies (one on stock diet, three on the atherogenic
diet), the pesticides will be added to the diet in doses to be determined by the results
of the range-finding study.
Blood will be drawn at 2-month intervals to determine total serum cholesterol and
triglycerides, and at 4-month intervals additional blood will be drawn to perform
detailed analyses of the lipoproteins. Lipoproteins will be fractionated by a com-
bination of ultracentrifugal separation and gel electrophoresis. Each lipoprotein
fraction will be quantitated with regard to protein content, lipid composition, and
pesticide content. For the chlorinated pesticide analyses, we will use electron
capture and gas-liquid chromatography. For the carbamates and organophosphates,
we will use a thermionic detector.
At autopsy, after 12 months of pesticide feeding, additional samples of blood
will be taken for final lipoprotein analyses, all tissues and major organs will be
examined for changes that may be related to chronic pesticide administration, and
special attention will be given to detailed examination of the arteries to assess the
extent, severity, and characteristics of the atherosclerotic lesions produced.
Portions of the cardiac muscle will be studied for capability of the mitochondrial
tractions to utilize oxygen. We will preserve samples of all organs for possible
analysis for pesticides and pesticide residues.
QUESTION: The animal model you are using, as you mentioned, is a very
expensive animal model, and it's becoming very limited in supply. When you use
inbred baboons, you're even increasing the cost of these studies. We recently did
an in-house study at Battelle involving animal model selection, and we found that
t'ie tapia, which is a pro-simian, forms these arteriosclerotic plaques and lesions
104
-------
very similar to those of the baboon. I was wondering whether or not your group has
looked at any of the pro-simians as a possible alternate animal model.
DR. WALLACE: Feral baboons, not colony bred and reared animals, will be used
in this oroject. Baboons appear to be, at this particular time, the one nonhuman pri-
mate that is readily available, and we have had no difficulty in acquiring these animals.
We do not have any experience with pro-simians, but others are studying them. The
advantage of the baboon is its large size, which makes larger quantities of blood and
tissue available.
DR. SPYKER: I'm not entirely clear on your design. It is going to be six animals
all together, or six for each compound?
DR. WALLACE: Six for each compound at each dose level and a control group.
QUESTION: Are you analyzing the diet for pesticides?
DR. WALLACE: Yes, the diet is periodically analyzed for pesticides.
QUESTION: What proportion of the baboons may be resistant to the development
of atherosclerotic lesions?
DR. WALLACE: There is a wide range of variability in all animal models to
the induction of hyperlipidemia and of atherosclerotic lesions. This is also true
f the human. There will be some animals resistant to both hyperlipidemia and
atherosclerosis on the diet alone. It is important to determine the effects of
esticides on these animals also.
QUESTION: I may have missed your comment in the design, but what are going
be the relative levels of your pesticide?
r\a WALLACE: These will be ascertained after the range-finding study. The
will be low enough not to cause acute toxicity, but larger than the allowable
intake for humans.
DR. FISHBEIN: What is the nature of the pesticides on tests? Are they commer-
formulattons?
WALLACE: These pesticides will be out of commercial stock as used for
,est contr°L
105
-------
FETAL TISSUE ANALYSIS
FOR PESTICIDE RESIDUES
Dr. Irwin Baumel*
NOT AVAILABLE
FOR
PUBLICATION
Effects Branch, Criteria and Evaluation Division, Office of Pesticide
Programs, EPA
107
Preceding page blank
-------
RESEARCH PROGRESS:
CARCINOGENIC AND TERATOGENIC TESTS - INTRODUCTION
Dr. Morris F. Cranmer*
The compound of the week type of phenomena may very well become compound of
the morning and afternoon. As the list grows and grows and grows. For those of
you who are not familiar with the list of agents that are under test by the National
Cancer Institute, I urge you to examine this list. It may provide some very interest-
ing results to you.
There's also an effort underway, to be coordinated by the National Library of
Medicine, to solicit from Government and private laboratories and Industry, lists
of compounds on chronic bioassay. The purpose of this, of course, is to make
this information available at the earliest possible time to anyone who is interested,
and to decrease the duplication of these very costly tests that had more than one
laboratory. I would hope that we would be successful in bringing an international
component into this because, as you all know, there's a great deal of carcinogenic
bioassay work which is done overseas. For instance, the British Industrial Bio-
logical Research Association does a great deal of this type of work.
There was a public meeting, sponsored by the Toxicology and Related Program
Coordinating Committee, on the appropriateness of new mutagenic tests, not only
as indicators of mutagenesis, but as indicators of potential carcinogenesis of a
compound to be examined. This meeting was successful; it did provide a forum
for persons from the Federal Government, universities, and industry to express
their opinions on the utility of these techniques. Dr. Gary Flam is chairing a sub-
r-ommittee which will be preparing a report on the official, if you will, status, as
far as DHEW is concerned, with respect to these specific tests.
Looking at the titles to be discussed today I'm sure some of these topics should
stimulate some very interesting discussion as to the predictiveness, the utility,
and the potential for producing both false positives and false negatives.
National Center for Toxicological Research, Jefferson, Arkansas
109
Preceding page blank
-------
IN VITRO AND IN VIVO CARCINOGENIC
AND MUTAGENIC SCREEN DEVELOPMENT
Dr. Erling M. Jensen*
The studies being conducted in our laboratories are divided into two general
areas of investigation. In vitro carcinogenesis assay development studies are
being conducted in our Washington Laboratory, and an in vivo mutagenesis feasi-
bility study is being conducted in our Worcester, Massachusetts Laboratories.
At the present time, the only absolute methods for determining the carcinogen! -
city of chemical agents is by in vivo bioassays, and even these are sometimes
questioned as to their relevance. These assays are extremely time-consuming
and expensive, and the need for a simple, reliable, inexpensive, and quantitative
assay is obvious. Various microbial assays for mutagenesis or presumptive car-
cinogenesis fulfill these requirements to a certain extent, but the need for an assay
system utilizing mammalian cells still exists.
The ability of chemical agents to transform mammalian cells in culture was
first demonstrated by Berwald and Sachs in 1963, and since then has been confirmed
by many investigators using a variety of cell cultures. Cells which have undergone
transformation in culture exhibit a number of new characteristics, such as a change
in colony morphology, loss of contact inhibition, increased growth rate, greater
resistance to the toxic effects of the transforming agent, loss of anchor dependence
tor growth in soft agar), greater agglutinability by plant lectins, and, of course,
bility to produce tumors in a suitable host.
A number of investigators have proposed the use of this system as an assay for
rcinogenesis.- In each of these assays, colony morphology has been selected as
. criterion or endpoint for neoplastic transformation. The assay is illustrated
in
Figure 1.
"irG&G/Mason Research Institute, Rockville, Maryland
111 Preceding page blank
-------
Figure 1
_In Vitro Carcinogenesis
10-13 day
hamster embryos
Trypsinization
Dispersed cell*
X-ray
5,000 R
I Primary culture
5 X
100-500 Target cells
cells (overlay)
24 hours
Test compound
(several dilutions)
Feeder layer
14 days
Colonies
(transformed and normal)
Hamster embryos are dispersed by trypsinization to prepare primary cultures
which are incubated for several days to eliminate dead cells and erythrocytes. The
cultures are then trypsinized to give single cell suspensions from which feeder
layer cultures are prepared. Since most mammalian cells grow poorly at low popula-
tion densities, i.e., 100-500 cells per plate, one can overcome this problem by add-
ing approximately 50,000 X-irradiated cells. This allows the target cells to replicate
and form colonies without being overgrown by the feeder cells. After the target cells
are added, the test compound is added and the plates are incubated for 9 to 14 days.
The colonies can be stained in the culture plates or observed by phase contrast micro-
scopy if one wishes to isolate viable cultures for further study.
Cells which are transformed exhibit a characteristic growth pattern in which the
cells tend to pile up through loss of contact inhibition and grow in a disorganized,
criss-cross pattern. The neoplastic nature of this type of colony has been confirmed
by in vivo transplantation, and it is rarely seen in untreated control cultures.
112
-------
Table 1 shows the results of one experiment with 4-aminobiphenyl in which a
relatively high rate of transformation was observed, even though a good dose response
was not achieved. The cloning efficiency data, based on an initial inoculum of 500
cells per plate, show a slight toxic effect at 1.2 Mg/ml.
Table 1: Transformation of Hamster Embryo Cells by 4-Aminobiphenyl
Compound Average Cloning Transformed/ Transformation
Cone. Colonies/Plate Efficiency Total Rate
Control 52 10%
1.2 Mg/ml 33 7%
0.6 Mg/ml 47 9%
0.3 Mg/ml 53 10%
0
542
7
423
471
28
778
0%
1.65%
0.4%
3.6%
Therefore, this system can be used to demonstrate neoplastic transformation by
chemical agents. However, the assay is extremely sensitive to any variations in the
culture environment, such as the type and age of cells used, culture medium, tempera-
ture, pH, trace nutrients, and method and time of treatment. Many months of labora-
tory experience are required to develop reliable assay results. We feel that we now
have a very clear understanding of this assay system, and in a very few months we
should be ready to apply it to the testing of selected pesticides.
This assay has considerable potential as a qualitative screen for carcinogenesis,
but the standard endpoint of morphological alterations is quite subjective and, because
of the low incidence of transformation, truly quantitative results are difficult to achieve.
Therefore, we are investigating another endpoint based upon the ability of transformed
cells to grow and produce colonies in suspension in soft agar. For reasons which are
unknown, most normal, untransformed cells require a surface of some type to attach
to before mitosis can occur. Most cancer cells and virus-transformed cells, on the
other hand, become "anchor independent" and can grow and produce sizable colonies
113
-------
in soft agar. If this phenomenon is true also for chemically transformed cells, a
very rapid, quantitative, sensitive, and precise assay for carcinogenesis can be
developed, comparable to the microbial mutagenesis assays.
Our studies in this area have been devoted primarily to establishing the relia-
bility of the phenomenon as an indicator of malignancy and to establishing the culture
conditions which are required for such growth. We have examined a number of
freshly isolated tumor cell cultures, established transformed cell lines, and normal
untransformed cultures,
Table 2 shows our results to date. As the data show, growth in soft agar cor-
relates well with the neoplastic or transformed state of the cells, although the per-
centages of cells forming colonies, or colony-forming efficiency, varies considerably
in each line. Two of the human cancer lines have not produced colonies, and we are
now attempting to determine if this is a problem related to cultural conditions or to
the cell line itself. Our goal is to demonstrate that chemically transformed cells
will also grow consistently in soft agar, and these studies are now in progress.
Table 2: Growth of Cell Cultures In Soft Agar
Colony Forming
Cell Cultures Type Efficiency
B-16, Melanoma Cancer 40%
Lewis Lung Carcinoma Cancer 10%
R-35 Mammary Adenocarcinoma Cancer 5%
13762 Mammary Adenocarcinoma Cancer 5%
P-388 Leukemia Cancer 40%
XC, Rat Carcinoma Cancer >90%
HeLa Cervical Carcinoma Cancer 10%
MDA-MB-134, Breast Cancer Cancer 0%
MDA-MB-157, Breast Cancer Cancer 0%
MDA-MB-231, Breast Cancer Cancer 6%
FL, Human Amnion Transformed Cell Line 37%
WI-38 Untransformed Cell line 0%
BALB/3T3, Mouse Embryo Untransformed Cell Line 0%
Primary Hamster Embryo Untransformed Culture 0%
114
-------
The in vivo mutagenesis program being conducted in our Worcester Laboratories
is a feasibility study to determine whether broad phenotypic parameters, such as
behavioral, biochemical, and/or physical characteristics, which are controlled by
multiple genetic loci, can be used for detecting mutagenic effects of chemical agents.
If this is shown to be possible and practical, an assay can be developed with a much
greater probability of detecting mutagenic agents than is now possible with the estab-
lished in vivo specific locus assays.
The experimental plan for this program is illustrated in Figure 2. The assay
requires approximately 14 months for completion. Because of the high level of effort
required in the final testing phase, new compounds are introduced at 2-week intervals.
Briefly, groups of six male C57BL/6J mice are treated with known mutagenic agents
at three dose levels below the maximum tolerated dose. Treatment is administered
daily for a period of 8 weeks to assure maximum exposure of the germ cells to
the mutagen during all phases of spermatogenesis. The treated males are mated,
and subsequent breeding is conducted by the cross-backcross method, which is the
most effective method for accumulation of homozygous genotypes of recessive muta-
tions. In this process, F females are mated back to their F, sires.
& 1
Some testing is conducted on the F progeny for the detection of dominant mutations.
Most of the testing will be conducted on the F generation to detect both dominant and
3
recessive changes.
Three behavioral tests are being employed in this study: shuttling activity, active
avoidance acquisition, and Go, No-Go discrimination. These tests were chosen
because variation in the genotype is known to cause changes in these behaviors,
because they can be automated assuring objective measurements and allowing mass
screening with few technicians, and because they make use of the same testing appa-
ratus and support systems. Each test measures separate behavioral domains which
involve vision, hearing, general arousal, acquisition of new habits, inhibition of
learned habits, and intermodal stimulus discrimination.
115
-------
Figure 2
In Vitro Mutagenesis
Experimental Plan
Test
Compound
#1
#2
:#3
#4
i-»
i-»
(33
n
#2
#3
#4
Months
12345678'
Ti^t-mont- Mate to Maturation of F Mate to Maturation of F_
Treatment Females 1 Females 2
1
1
10 11 12 13 14 15 16 17 If
Cross* Maturation of F3 Testing
* Mate females back to sires.
-------
The C57BL/6J mouse was selected for these studies because it is a standard
strain used in many behavioral genetic analyses, and there already exists a body
of data on its behavioral characteristics which can be used in mutagenic studies.
Base line behavioral studies are being conducted on C57BL/6J and DBA/2 mice
and their reciprocal F hybrids. These studies indicate that this system is capable
of detecting genetic variations or mutagenic effects with as few as 70 mice per test
group.
Other endpoints will be observed as indicators of mutagenic effects, including
general phenotypic abnormalities, such as skeletal deformities, pre-weaning mor-
tality and juvenile death, litter size, anemia at birth, sterility, and alteration in
glucose metabolism.
Ten known mutagenic agents are being examined by this procedure. Table 3
shows the current status of the study and some of the problems which have been
ncountered. These initial studies were conducted at one subtoxic dose level.
TTnfortunately, almost all published toxicity data on these compounds describe single-
Hose treatment regimens with essentially no information concerning continuous treat-
t such as we have employed. There was, therefore, no way of anticipating the
delayed mortality and sterility which we experienced initially with some of the com-
ounds. For example, all of the mice treated with N-nitroso-diethylamine died
after termination of the treatment, some as much as 3 months after treatment.
veral other treatment groups have been unable to breed after either 1 or 2 months.
mpounds which have shown these problems are being retested at several lower
se levels. However, six treatment groups are breeding, several of which have
ached the backcross stage, and final testing results will be available in approxi-
uiately 4 months.
This is, as I mentioned, a feasibility study to determine whether and how consis-
tly mutagens can be detected by this approach. Should it prove reliable, a specific
orocedure utilizing the optimal treatment procedures, endpoint determina^
r
and minimum number of mice will be established for the assay of pesticides
*"
other agents.
117
-------
Table 3: Current Status of In Vivo Mutagenesis Assay
o>
Compound
1. N-Nitrosodiethylamine
2. N-Nitroso-N-methy lure thane
3. 3-Methylcholanthrene
4. Nitrogen Mustard, HCl
5. Methylmethanesulfonate
6. Ethylmethanesulfonate
7. 7, 12-Dimethylbenzanthraeene
8. Streptozotocin
9. Tolazotoluidine
10. 4-Aminobiphenyl
Dose
(mg/kg/day)
3
8
25
10
37.5
25
0.1
0.1
50
30
Route
Gavage
Gavage
Gavage
S.C.
Gavage
Gavage
S.C.
I. P.
Gavage
Gavage
Status
Delayed deaths
F progeny - back crossed
&
Some sterility and deaths in F ,
some F progeny
&
Delayed deaths and sterility
Sterility
Some sterility, some F progeny
back crossed
Delayed deaths
F progeny
F progeny
A
F1 progeny
-------
DROSOPHILA MUTAGENESIS TESTS
Dr. Ruby Allen Valencia*
The WARF Institute contract with EPA calls for screening for germ line
mutations, i.e.,. molecular alterations in genes or structural changes in chromo-
somes, which occur in the germ cells and therefore affect the descendants of the
exposed individual. Our purpose is to look for mutations in a eukaryote, where
there is a digestive mechanism, and other enzyme systems, such as microsomal
enzymes, which could convert the compound in question to a different form, result-
ing in either a loss of genetic effectiveness or the acquisition of mutagenicity.
Our test organism is Drosopbila melanogaster. the fruit fly. Drosophila is
well known as a tool for genetic studies. It has many genetic and economic
advantages and since these are also well known, I will not describe them here.
As a screening organism for environmental mutagens, however, Drosophila used
to be considered to be at a disadvantage by not being a mammal and therefore too
different from man with regard to the physiological barrier between the chemical
and the genes. It has been found, however, that flies do indeed have a microsomal
enzyme system, even though they don't have livers. It has been shown that muta-
tions are obtained with substances known to require enzymatic activation (in the
Salmonella test, for example). This work was done by Vogel and Sobels and is in
press (!)• Thus we now feel on much safer ground with Drosophila testing, and
this is most fortunate, since there are no other eukaryotic systems that offer the
versatility of Drosophila.
Our lab was installed just a year ago and testing was started in the fall. At
the end of the year it was decided that our program would be intensified, and
this involved further expansion, hiring and training of additional technicians, etc.
VVe have, therefore, been in full operation only since about February of 1975. We
have been sent 20 compounds in the course of the year and we have run them all
through our basic pre-determined screening procedures.
-\Visconsin Alumni Research Foundation Institute, Madison, Wisconsin
119
-------
We were lucky that the interest was in looking for effects from low level
concentrations, since many of the compounds are insecticides and are, of course,
lethal to flies. They can be applied only at very low concentrations. The LCso
for some was 0.1-1.0 ppm so we treated at below that concentration. The non-
toxic compounds (most of the fungicides and herbicides) were used at concentra-
tions of 2 to 4 ppm.
Exposure was by feeding. The chemicals are dissolved in DMSO, then fed in
1 percent glucose solution on filter paper discs. Actually, exposure is by both
ingestion and contact, since the flies must rest on the wet filter paper to feed.
Also, the treatment chambers are quite small, so it is difficult for them to
avoid contact. We expose for 48 hours, which insures ingestion, since flies can-
not survive that long without taking liquid.
The original plan involved two genetic schemes, one for sex-linked recessive
lethals and one for chromosomal alterations. These schemes are diagrammed in
Figures 1 and 2. m the sex-linked recessive lethal test we score for any mutation
of any of about 200 genes on the X chromosome, which results in lethality. In
the chromosomal mutation test we score for chromosome breakage, rearrangement,
loss, or non-disjunction.
At mid-year when we intensified the program, we added dominant lethals, since
these data can be obtained much faster and easier in Drosophila than in the mouse,
and they should be just as meaningful. A flow plan was set up, whereby we would
get dominant lethal data in 10 days after treatment (that is for the entire germ
cell development period), chromosome mutation data in 2 1/2 to 3 weeks, and
recessive lethal data in 6 weeks. The dominant lethal test, however, took quite
£ v/hile to develop and meanwhile most of the chemicals had been tested with the
other schemes. It seemed unnecessary to go back and do dominant lethals for
these compounds. Also, there is beginning to be a consensus that dominant lethals
are not a suitable screening system, especially in Drosophila. Vogel and Sobels (1)
120
-------
Figure 1
Genetic Scheme for Detection of Sex-Linked Recessive Lethal
Mutations
pl
Fl
3 FM6^j 00 X
FMi* ' f T
1 FM6 indiv. Q X
CS T
1 CS<2) 6* (treated)
Y
2 FM6 c?c?
Y
Daughters of each
P^ cf kept as a
"family" group
FM6 FM6 X FM6
FM6 CS Y
When no CS O cf are present,
indicating a lethal, make
this mating
CS
Y
Observe cultures for
presence or absence of
these
Observe for confirmation
of lethals
/I) FM6 = First Multiple No. 6, an X-chromosome with a complex of inversions (to
suppress crossing-over) and visible markers (y B w)
/2) Canton S wild-type stock
121
-------
Figure 2
Scheme for Chromosome "Mutations"
vq bw
vq bw ?
(1)
X
cho
+.„. -S
-------
have shown this in comparative studies. Therefore we have not been doing the
dominant lethal test, although we did work on it and got it going.
The chromosome mutation test also did not rate well in the Vogel and Sobels
experiments, as compared to the recessive lethal test. We have applied the test,
however, to all 20 compounds and we have at least preliminary data for all of
them.
The sex-linked recessive lethal test was found (also by Vogel and Sobels) to
be far more sensitive than either of the other two tests. It is capable of indicat-
ing mutagenicity at much lower concentrations of the chemical and it works for
a much wider range of types of compounds. In view of this, I feel that we should
perhaps concentrate our efforts on the recessive lethal test in future testing.
We treat adult males of the appropriate genotype for each type of test, and
these males are then mated and brooded sequentially, so that we sample germ
cells exposed at different stages of development. In each test scheme, individual
treated males are followed, so that we can detect clustering (i.e., the occurrence
of two or more mutations in the germ cells of one male). This procedure allows
conclusions to be drawn from a much smaller sample size than would otherwise
be necessary. I will point out examples later.
We usually treat a batch of flies every other week. When we do, we set up
both kinds of tests and we apply two or three different compounds plus one con-
trol for each test scheme.
The results from the recessive lethal tests are as follows:
Table 1, for Guthion, shows how the data are compiled. In this case, there were
two "runs," each with its parallel control. The table shows the number of treated
_r untreated parental males, the number of FI daughters tested in each sequential
brood, and the number found to carry lethal-bearing X chromosomes. Lethals
lirringin clusters are so indicated. The standard errors are calculated using
123
-------
Table 1: Recessive Lethal Data
to
Guthlon
No.
PI No. FI tests per
Compound
#1
#2
£. 7/17/75
Control #1
Control #2
£ 7/17/75
Males (1) 1
15 0/1201
29 0/1561
44 0/2762
10 0/783
29 2/1272
39 2/2055
2
1/748
0/728
1/1476
0/104
2/661
2/765
Total
stage [1/N1 (2) Fi
3
1/398
0/453
1/851
1/317
0/290
1/607
4 Tests
1/347 2697
4/670 3416
5/1017 6113
1/178 1384
0/610 2837
1/788 4221
Total Lethals % + se
S (3) Cl (4) T
1 1(2) 3 0.111+0.083
40 4 0.117+0.058
70 7 0.115+0.043
20 2 0.144^0.102
40 4 0.141+0.070
60 6 0.142^0.058
(1) No. of treated or control males which produced progeny
(2) No. of F! females which produced progeny (lethal bearing/normal)
(3) Single lethal in progeny of one male
(4) Cluster of lethals in progeny of one male (no. of clusters/size of cluster)
-------
the standard formula when there are no clusters. When clusters are present, we
use a formula devised by H. J. Muller, which corrects for the probability that the
lethals of a cluster are of common origin. The effect is a larger error and avoids
attaching undue significance to an Increased number of lethals.
Table 2 shows the data from all the control runs. It is generally consistent
and the overall frequency (0.18 percent) is within the usual range for this stock
(wild type Canton S). Run No. 3 was a bit high and No. 12, our last run, was
veiy much out of line. I will come back to this run later and discuss it further.
Table 3 shows the sum of all results for each compound. With the standard
errors derived from the Muller formula, none of the frequencies is significantly
greater than the overall control frequency.
Malathion gave the highest overall frequency. The cluster information, how-
ever, shows that 21 of the 31 lethals were found in the progeny of a single male.
These lethals were found in all broods of that male and certainly represent a
cluster arising from a single spontaneous event in a gonial cell. Assuming this
to be the case and counting only 11 total lethals, the frequency becomes 0.16 per-
cent. Even counting the 31 lethals, the standard error derived from the Muller
formula removes the significance of the higher percent. This case shows the
impedance of following individual treated males and recording clusters. Table
4 shows the malathion data in detail. Run No. 1 gave a frequency of 0.963 per-
cent versus a control of 0.144 percent. This would have been read as a strong
positive mutagenic effect in a standard mass-mating type lethal test. The repeat
_jn would have disproved this conclusion, however, thus demonstrating the
importance of repeat runs.
Table 5 shows the bromacil data, to Bun No. 1 bromacil gave an elevated
frequency and there were no clusters. The parallel control, however, was equally
levated and there was only one cluster of two. Run No. 2, however, gave lower
equencies for both treated and control. This is another example of the importance
f p£irallel controls and repeat runs.
125
-------
Table 2: Recessive Lethal Data
10
en
Controls
No. Total
Pi No. F1 tests per staffe fl/N ] (2) FT
Run Males (1) 1 2
1 129
2 56
3 27
4 10
5 18
6 24
7 30
8 27
9 25
10 29
11 10
12 28
Total 413
3 4 Tests
12062
5689
1543
1386
2732
1997
3721
3495
3299
2837
384
3216
42361
Total Lethals
S(3)
14
5
3
2
2
3
6
5
2
4
0
5
51
Cl(4)
0
1(2)
1(2)
0
2(2)
0
0
0
1(3)
0
0
1(3)1(6)
1(7)
4(2)2(3)
1(6)1(7)
T
14
7
5
2
6
3
6
5
5
4
0
21
78
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0
0.
0.
% + se
116
123
324
144
220
150
161
143
152
141
653
184
+ 0.
+ 0.
+ 0.
+ 0.
+ 0.
+ 0.
031
052
171
102
115
087
+ 0.066
+ 0.
064
+ 0.093
+ 0.
+ 0.
+ 0,
065
182
031
(1) No of treated or control males which produced progeny
(2) No. of Fi females which produced progeny (lethal bearing/normal)
(3) Single lethal in progeny of one male
(4) Cluster of lethals in progeny of one male (no. of clusters/size of cluster)
-------
to
-1
Table 3i Recessive Lethal Data
Cumulative to July 18, 1975
Compound
Thlmct
E. parathlon
Bromacll
Captan
Malathlon
Guthion
PCNB
Dursban
Trifluralln
Fenthion
Azodrln
Folpet
MSMA
Oinoseb
Cacodyllc Acid
DSMA
Slmazlna
Dlcamba
iL Controls
No.
Males (1)
146
128
117
46
49
44
38
54
52
55
54
30
29
30
52
52
107
66
413
Total
No. Fj tests per stage (1/N1 (2) FI
1
23/15450
16/11914
18/6198
4/2075
14/2593
0/2762
2/1477
5/2870
5/2827
8/2925
7/3121
1/1876
2/2224
4/2007
2/2032
4/2240
16/3922
9/3034
49/30309
2
_
-
0/641
8/1594
12/1784
1/1476
2/1089
4/1906
9/1531
5/1950
2/2119
1/587
0/409
1/709
5/1978
1/1537
5/999
5/1002
9/3790
3
_
-
0/270
2/664
4/954
1/851
2/518
9/1405
1/1311
1/1376
3/1565
0/443
0/57
0/150
0/1394
2/1666
4/1868
2/1409
7/3438
4
_
-
2/628
4/811
1/1271
5/1017
2/854
3/1451
3/1422
0/1297
2/1461
1/438
0/728
2/766
1/1607
1/1378
0/1680
1/2204
13/4746
Testa
15473
11930
7757
5182
6633
6113
3946
7653
7109
7551
8276
3347
3420
3659
7019
6829
8494
7666
42361
Total Lcthala
S(3j
23
14
20
12
8
7
6
7
6
10
14
3
2
3
6
8
7
5
51
CU41 _T_
0 23
1(2)
0
1(5)
1(2)1(21)
0
1(2)
2(2)1(10)
2(2)1(3)
1(5)
2(2)
0
0
0
2(2)
1(2)
0
4(2)1(3)
1(7)
3(2)2(3)
4(2)2(3)
1(6)1(7)
16
20
17
31
7
8
21
18
14
14
3
2
7
8
8
26
17
78
%+se
0.149
0.134
0.258
0.328
0.467
0.115
0.203
0.274
0.253
0.185
0.167
0.090
O.OS8
0.191
0.114
0.117
0.294
0.222
0.164
+ 0.031
+ 0.035
+ 0.058
+ 0. 117
+ 0.320
+ 0.043
+ 0.080
+ 0.140
+ 0.097
+ 0.056
+ 0.045
+ 0.052
+ 0.041
+ 0.091
+ 0.045
+ 0.041
+ 0.106
+ 0.077
+ 0.031
(1) No of treated or control males which produced progeny
(2) No. of F} females which produced progeny (lethal bearing/normal)
(3) Single lethal In progeny of one male
(4) Cluster of lethals In progeny of one male (no. of clusters/size of cluster)
-------
Table 4: Recessive Lethal Data
to
CO
Malathion
Compound
#1
n
£
Control #1
Control #2
£
No.
?!
Males (1)
16
33
49
10
29
39
No. FT tests per stage fl/Nl (2)
1
12/1040
2/1553
14/2593
0/783
2/1272
2/2055
2 3
11/866 4/399
1/918 0/555
12/1784 4/954
0/104 1/318
2/661 0/290
2/765 1/608
4
0/471
1/800
1/1271
1/179
0/610
1/789
Total
FT Total Lethals
Tests S (3) Cl (4) T
2803 4 1(2)1(21) 27
3830 40 4
6633 8 1(2)1(21) 31
1384 20 2
2837 40 4
4221 60 6
% +se
0.963 +0.749
0.104+0.052
0.467 +0.320
0.144 +0.102
0.141 +0.070
0.142 +0,058
(1) No. of treated or control males which produced progeny
(2) No. of FI females which produced progeny (lethal bearing/normal)
(3) Single lethal in progeny of one male
(4) Cluster of lethals in progeny of one male (no. of clusters/size of cluster)
-------
Table 5: lleeessvve Lethal Data
to
tO
Bromacil
Compound
#1
#2
£
Control #1
Control #2
*
No.
Pl No. Ft tests per stage fl/Nl (2)
Males (1) 1 2 3 4
89 17/5551 -
28 1/647 0/641 0/270 2/628
117 18/6198 0/641 0/270 2/628
27 5/1538 -
24 0/711 2/584 0/273 1/426
51 5/2249 2/584 0/273 1/426
Total
Fj Total Lethals % + se
Tests S (3) Cl (4) T
5568 17 0 17 0.305+0.074
2189 30 3 0.137+0.079
7757 20 0 20 0.258+0.058
1543 3 1(2) 5 0.324+0.171
1997 30 3 0.150+0.087
3540 6 1(2) 8 0.226+0.089
(1) No. of treated or control males which produced progeny
(2) No. of F£ females which produced progeny (lethal bearing/normal)
(3) Single lethal in progeny of one male
(4) Cluster of lethals in progeny of one male (no. of clusters/size of cluster)
-------
Table 6 shows the results of Run No. 12 for simazine, dicamba, and the
parallel control. Although both compounds gave a somewhat elevated lethal
frequency, this is meaningless, since the control rate was also unusually high.
There were several clusters in all three groups. The conclusion is that some
factor other than the pesticide treatments entered into this run. We are very
concerned about this result and wevare investigating possible causes. The
presence of a mutagen in the environment will almost certainly be ruled out and
the cause will probably be found to be genetic events in the stocks. The solution
to this problem could be of considerable interest genetically and also of practical
importance in connection with the avoidance of "false positives."
The overall lethal frequency for captan is not significantly high, but there are
some aspects of the total picture which seem suspicious: Table 7 shows the de-
tailed results. In both runs, the compound gave a lethal frequency of about twice
the control frequency. In addition, as shown in Table 8, in both runs the other
compounds gave frequencies similar to the controls while captan gave a higher
frequency. (Malathion, as pointed out above, gave a frequency of 0.15 percent if
the cluster is counted as one event.) We feel that captan should be investigated
further.
The "chromosome mutations" (breakage, loss, rearrangement, and non-dis-
junction) will be discussed here only briefly, since the analysis of the data is quite
involved and in any case our conclusion at this time is that none of the compounds
has given a definite positive mutagenic result. As in the case of recessive lethals,
each run tests two or more compounds and there is a parallel control.
Table 9 shows how the data are recorded for each run of each compound tested.
The F females and males are counted and exceptional flies are observed by pheno-
type. Clustering is noted and this can involve flies of different phenotypes and
aex as well as the same phenofcype in one sex, i. e., a single event can give rise
to>two phenotypic types.
130
-------
Table 6: Recessive Lethal Data
Run #12
No.
Total
PI No. Ft tests per stage [1/Nj (2) Fi
Simazine
Oicamba
Control
Males (1) 1 234
58
59
28
Tests
6458
7410
3216
Total Lethals
S (3) Cl (4)
7 4(2)1(3)
5 3(2)2(3)
5 1(3)1(6)
1(7)
T
18
17
21
%+se
0.279
0.229
0.653
(1) No. of treated or control males which produced progeny
(2) No. of Fj females which produced progeny (lethal bearing/normal)
(3) Single lethal in progeny of one male
(4) Cluster of lethals in progeny of one male (no. of clusters/size of cluster)
-------
Table 7: Recessive Lethal Data
K)
Captan
No.
Pj No. FT
Compound Males (1) 1
#1 17 0/1024
#2 29 4/1051
46 4/2075
Control #1 10 0/783
Control #2 29 2/1272
39 2/2055
tests per
2
5/971
3/623
8/1594
0/104
2/661
2/765
Total
stage [I/NT (2) Fi Total Lethals % + se
3
1/388
0/297
2/684
1/318
0/290
1/608
4 Tests S (3)
3/395 2787 4
1/416 2395 8
4/811 5182 12
1/179 1384 2
0/610 2837 4
1/789 4221 6
Cl (4) T
1(5) 9 0.323+0.193
0 8 0.334+0.118
1(5) 17 0.328+0.117
0 2 0.144+0.102
0 4 0.141 +0.070
0 6 0.142+0.058
(1) No. of treated or control males which produced progeny
(2) No. of F! females which produced progeny (lethal bearing/normal)
(3) Single lethal in progeny of one male
(4) Cluster of lethals in progeny of one male (no. of clusters/size of cluster)
-------
Table 8: Recessive Lethal Data
co
CO
Two Rims
Captan
Malathion
Gulhion
PCNB
C
Captan
Malathion
Guthion
C
No.
P!
Males (1)
17
16
15
11
10
29
33
29
29
Total
No. F1 tests per stage [1/N] (2) Fj
1.2 3 4 Tests
2787
2803
2697
1793
1386
2395
3830
3416
2837
S(3]
4
4
1
2
2
8
4
4
4
Total Lethals
> Cl (4)
1(5)
T
9
1(2)1(21) 27
1(2)
0
0
0
0
0
0
3
2
2
8
4
4
4
% + se
0.323 +0.193
0.963+0.749
0.111 +0.083
0.112 +0.079
0.144 +0.102
0.334 +0.118
0.104 +0.052
0.117 +0.058
0.141+0.070
(1) No. of treated or control males which produced progeny
(2) No. of Fj females which produced progeny (lethal bearing/normal)
(3) Single lethal in progeny of one male
(4) Cluster of lethals in progeny of one male (no. of clusters/size of cluster)
-------
Compound Guthion (run #1)
Table 9: Chromosome Rearrangement, Loss and Non-Disjunction
Date Feb. 1975
Number Treated Males 35
Brood
1234
No. FX Females 866 584 585 461
Female Exceptions v -d'l y v -d*9 y v -&A
yv-<330 mo^Lc~
^V OC
\J «O
No. F Males 855 550 541 475
S S
Male Exceptions cho B - y cho B -
CTl4 CFl
y cho B
Cfl7
Total Progeny 1723 1137 1128 937
Total Number Exceptions 232 1
Total Frequency Exceptions
Totals
2496
5
2421
3
4925
8
(4 singles
cl (2) - dT 1
cl (2) - CT14
16.24X10"4
134
-------
Table 10 shows the total data to date. The variation in frequency of exceptions is
considerable. Only one compound, however, gave a frequency lower than the overall
control. This was ethyl parathion. The rest gave frequencies higher than the control
frequency although many are only slightly higher. Most of those giving the higher
-4
frequencies had clusters. Notable exceptions are methomyl (12.83 x 10 ) and MSMA
-4 -4
(20.29 x 10 ). In the case of MSMA, the parallel control frequency was 5.55 x 10 .
The sample size is still quite small, however, and is for a single run. We therefore
reserve our conclusion. The sample for methomyl is that considered adequate for this
test, so the result on the surface is somewhat suspicious. A look at the individual
data, however, shows that in both runs the compound and control gave equivalent rates.
The two control values in question are numbers 13 and 14 in Table 11. This table
shows the various control runs with the chromosome mutation tests. It can be seen
that the variation between runs is quite great here too, although none of the groups
gave a frequency as high as some of the pesticides. There is less tendency to cluster-
ing also, but then the sample sizes are generally smaller. Thus, while there is some
suggestion, of a positive effect for some of the compounds, none can be proved at this
poiot.
I would have to say, then, in summary, that under the conditions of these experiments,
none of the 20 pesticides were mutagenic in either the recessive lethal test or the
chromosomal mutation test. I cannot say that all of these compounds are Incapable of
inducing mutations in Drosophila. Perhaps if applied at higher concentration or by
other route or to some other cell type, mutagenicity could be demonstrated. Under
present conditions, however, none were proved positive.
a scientist, I would of course like to try some of the above-mentioned variations.
. difficult to keep to a predetermined, standardized, limited screening procedure,
must in order to be able to screen the desired number of compounds. However,
is no reason that we cannot change our standardized procedure at the appropriate
e if vve have found a better way to screen, a way that would Increase sensitivity or
•litate the work or be morr discriminating.
135
-------
Table 10: Chromosome Rearrangement, Loss and Non-Disjunction
Data Cumulative to July 20, 1975
Compound
Thlmet
Ethyl Parathion
Methyl Parathiou
Captan
Folpet
PCNB
Malathion
Bromacil
Guthion
Azodrin
Fenthion
Dursban
Dlnoseb
Trifluralin
MSMA'.
DSMA
Cacodylic Acid
Dicamba
Sunazine
Methomyl
Monuron
Total Treated
Total Controls
No. Pl
Males
378
235
-
75
29
26
68
118
65
76
84
75
23
68
30
50
43
118
151
111
106
1929
617
No. F1
Progeny
34547
20721
-
12676
4655
4270
11701
7125
9792
15030
13155
12385
3406
12013
3942
8609
5937
14396
15861
20266
15166
245653
69767
No. Freq. Exc.
Exceptions x 10
19, cl (13) = 32
11
-
10
5, ci (2) = 7
5
9, 2 cl (2), cl (3) = 16
6
9, 4 cl (2) = 17
11, 2 cl (2), cl (3), cl (4) = 22
12, cl (2) = 14
11, cl (2) = 13
1, cl (2) = 3
9, cl (2) = 11
8
8
5, cl (4) - 9
10, cl (2) = 12
11, cl (4) = 15
26
15
260
54
9.26
5.31
-
7.89
15.03
11.71
13.67
8.42
17.36
14.64
10.64
10.50
8.81
9.16
20.29
9.29
15.16
8.34
9.46
12.83
9.90
10.58
7.74
136
-------
Table 11: Chromosome Rearrangement, Loss and Non-Disjunction
Date Cumulative to 7-20-75
£j"" •• —
rroug_
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
No. PX
Males
167
100
25
9
29
15
20
41
38
30
32
30
35
23
23
No. FI
Progeny
12581
10387
1245
1485
3232
3604
3527
4365
6906
6005
3217
1753
5846
2208
3406
No.
Exceptions
2
8
1
1
4
2
2
5
3
3, cl (2), cl (4) = 9
cl(2)
2
7
3
1, cl (2) = 3
Freq. Exc.
x 10 "4
1.59
7.70
8.03
6.73
12.38
5.55
5.67
11.45
4.34
14.99
6.22
11.41
11.97
13.59
8.81
617
69767
54
7.74
137
-------
We do try to carry on some method development experiments, and I think that now
that the lab is established and getting past the "growing pains" stage, we will be able
to do more experimenting. There are several variations in exposure procedure that
I want to look at, using known chemical mutagens and doing some comparative experi-
ments. There are some genetic schemes that look good on the drawing board, but
they must be tried in practice. Since we have just received nine more pesticide com-
pounds, there is plenty of work to be done.
Reference
1. Vogel, E. and Sobels, F. H. "The Function of Drosophila in Genetic Toxicology
Testing." (To appear in Chemical Mutagenesis, Ed., A. Hollaender, Plenum
Press.)
DR. FESHBEIN: I realize insolubility is a problem with pesticides. I'm a bit appre-
hensive of the use of DMSO.
DR. VALENCIA: So are we. We are using it in desperation. We didn't know what
to do with these things. We can't get them in solution.
DR. FISHBEIN: You couldn't vouchsafe the stability of some of these materials in
DMSO, that in point of fact you are testing captan, per se, and not some degradation
products after it's been sitting in DMSO. Have you tried any other solvents?
DR. VALENCIA: No, we haven't. This is one of the things that I would like to try.
COMMENT: I just want to remark that of the substances you tested, we have tested
only two in the Salmonella work we've done, because we weren't working on pesticides.
138
-------
Those are PCNB and captan. Captan is clearly positive, and PCNB is negative. And
so the suspicion that you have about captan, I think, will be well worth additional
efforts so that you can build up your statistics.
DR. VALENCIA: One of the bases for my suspicion of captan is simply that it has
come up positive in a lot of different systems. Some negative results in Drosophila
have been published, but I, for one reason or another, have reservations about those
egatives. So it really does need another more intensive look.
COMMENT: I would prefer to think about the suspicion as being related only to
ur data, and certainly your data indicate that of all the things you've tested, that's
ti.e one that's approaching a significant effect. So I really think you ought to just
lug away at that one. It's obviously very important economically.
. VALENCIA: You just reminded me of one thing that I forgot to point out. You
have noticed that the standard errors were very large in some of the cases.
MS is because in the cases in which we have found clustering, we apply a formula
. jj was developed by Dr. H. J. Muller to take into account that there are muta-
wrhich probably are of common origin. This simply wipes out the significance
never you happen to hit a big cluster.
139
-------
IN VITRO AND IN VIVO STUDIES OF SELECTED
PESTICIDES TO EVALUATE THEIR POTENTIAL
AS CHEMICAL MUTAGENS
Dr. Gordon W. Newell
The Federal Insecticide, Fungicide, and Rodentlcide Act designates EPA as the
Governmental body responsible for the safety of all pesticides used in the United States.
More recently, the Federal Environmental Pesticide Control Act (PL 92-516) identified
EPA's regulatory responsibilities in the area of pesticides to include intra- as well as
interstate commerce.
To be Federally registered, a pesticide must be determined to be not hazardous to
health or to the environment when used according to its labeling restrictions. Thus,
relative to new law, as well as to specific directives included in Public Law 93-135,
1973, EPA is now reviewing the use of alternate chemicals for pest control, including
older registered pesticides.
In the EPA pesticide review process, emphasis is placed on development of scientific
criteria for evaluating the safety of compounds, substituting those pesticides found to
be hazardous. In addition to reviewing and evaluating the pesticide literature, as well
as maintaining liaison with industry and academia, the program includes laboratory
studies to obtain additional data to support the literature review effort. One of these
laboratory programs is directed toward gathering mutagenesis data on a selected number
of compounds. Stanford Research Institute is applying its best efforts in conducting this
study.
Development of methods to evaluate the mutagenlc hazard of chemical compounds has
advanced markedly In the last few years. From undefined empirical tests of a short
time ago, procedures are now available which can detect chromosome breaks and genetic
"Department of Toxicology, Stanford Research Institute
141 Preceding page blank
-------
changes caused by chemical stress. Mutant strains of microorganisms in cell culture
and mammalian fibrobiast cells in tissue culture are effective in vitro systems for re-
liable evaluation of presumptive mutants, while the mammalian dominant lethal test,
using mice or rats, is a recognized test for the assessment of chromosome damage,to
germinal cells.
Today there are many pesticide chemicals in commercial use which have not been
adequately investigated for their mutagenic hazard. With the increasing concern by the
public about possible stress of chemicals to our environment, there is need for such
studies of the widely used pesticides. This project is utilizing test methods which are
appropriate for these evaluations and which are in use by the scientific community.
Objective and Scope
The objective of this program is to investigate the mutagenic potential of selected
pestic ides, us ing _in vitro and jn vivo test procedures.
To investigate the mutagenic potential of selected pesticides, the following three test
systems are used:
• Microbial assays with and without metabolic activation
« Human fibrobiast cell culture assays for DNA repair with and without metabolic
activation
• Dominant lethal tests in mice
In Vitro Mutagenesis Studies
.Microbial Indicators of Mutagenesis
Numerous systems are available for studying chemical mutagenesis, and they have
been described in detail. Each system has specific advantages and disadvantages, but
-certainly no one system can give a true estimate of the total effects on man — our
primary concern. Considerations such as uptake, metabolism, detoxification, acti-
vation, type of exposure, repair mechanisms, and dosages are difficult to evaluate by
any one test system. However, from the standpoint of time and economics, it is un-
realistic to use a wide range of test systems to examine all chemicals to which man may
be.exposed.
142
-------
MLcroblal systems, Including viruses, prokaryotes, and eukaryotes, have been used
for chemical mutagenesis studies. In addition to straight mlcrobial systems, combina-
tion systems allowing for metabolic activation have been developed, such as the host-
mediated assay and the microsomal activation systems.
There are many advantages to using microorganisms since they can be very sophis-
ticated and yet extremely sensitive, simple, and economical. On the other hand, they
are non-mammalian and, therefore, their validity with regard to man has been questioned.
Microbial systems lack the metabolic activation potential of man, and there are questions
related to the repair potential between microorganisms and man. However, many of
these questions can be circumvented by using mammalian metabolic activation systems
and by eliminating repair mechanisms. Furthermore, since the site of mutations Is
DNA and since there is a homogeneity or uniformity with regard to the chemical and
physical composition of all living systems, an adverse effect in one system may repre-
sent a potential threat to other living systems.
Microsomal Activation Systems
The conversion of a compound to an active metabolite Is a requirement in the detection
of certain mutagens or carcinogens. Hepatic microsomal enzymes are very active
metabolically and have been shown to activate various compounds.
Mailing and Chu have used liver microsomes as an activation system with mammalian
ceil cultures or microorganisms as the indicator organism. Chu has reported that the
combined treatment of dimethylnitrosamine (DMNA, a potent carcinogen in rodents
thought to be metabolized via oxidative dealkylation before becoming tumorigenic) with
liver microsomes produces a significant increase in the frequency of azaguanine -
resistant mutant hamster cells in tissue culture, over treatment with DMNA alone.
Similarly, Mailing showed increased mutation frequencies for the reversion of hlstidine
auxotrophs of Salmonella typhlmurlum to prototrophy when cells, liver microsomes,
dimethylnitrosamine were incubated together.
143
-------
Slater e£aL also have shown that a liver microsomal fraction potentiates diethylni-
trosamine and dimethylnitrosamine activity in systems utilizing bacteria deficient in
DNA polymerase; such systems are more sensitive to agents that alter cellular DNA
than is the parent strain from which they were derived.
Ames has incorporated a rat microsomal activation system with his tester strains
of histidine auxotrophs of Salmonella typhimurium in studying the mutagenic effect of
chemicals. In this procedure the microsomes are included in the top agar along with
the bacterial test strain and the chemical being tested. Using the Ames method, we have
activated 4-aminobiphenyl, 2-amino and 2-acetylamtnofl.uorene, aminoanthracene,
amlnopyrene, methyl and dimethyl benzanthracene and benzpyrene to mutagenic metab-
olites.
As part of this contract, we have determined the mutagenicity of 20 proposed sub-
stitute chemicals toward microorganisms (Table 1). Three types of mutagenic ity have
been measured:
1. Qualitative changes in the information encoded in the DNA (Reverse mutation in
Salmonella typhimurium andjS. coll WP2).
2. Chromosomal changes in Saccharimyces cerevisiae.
3. The toxic ity of chemicals as a function of the absence of various DNA repair
systems. These studies were conducted inl^. subtilis and E. coli.
Table 1: EPA Pesticides - Contract No. 68-01-2458,
SRI Project 3493
Folpet Cacodylic acid
Captan MSMA
Methyl parathion DSMA
Ethyl parathion Dursban
Malathion Trifluralin
PCNB FenthFon"
Bromacil Dinoseb
Azodrin Methomyl
Guthlon Slmazine
Thimet Monuron
144
-------
Table 2: Pesticides Showing a Positive Mutagenic Response
in One or More La Vitro Systems
T>esticide_ S. typhimurium S. cerevislae E. coli B. subtQis
Captan + + +
Folpet + + + . -
3romacil + + +
Azodrin - +
Gutbion - +
Cacodylic acid - +
Dursban - +
Dinoseb - +
The following are characteristics of the first type of assay system, whether it is con-
ducted Ln Salmonella or Escherichia. The bacterial strain contains a mutation which
results in a requirement of an amino acid for growth. Thus, in the absence of the exo-
genous addition of the amino acid, the microorganism cannot grow unless the mutation
is specifically reverted. The advantage of this assay system is that only mutants grow.
Thus, the frequency of mutation can be measured by counting those bacteria which can
give rise to colonies.
Two types of mutation are known to be specifically reverted by chemical mutagens.
The first type is called base-pair substitution. A base-pair substitution occurs when a
base, e.g.i adenine, is replaced by one of the other three bases: thymine, guanine, or
vtosine. The second type of mutation is a frame-shift mutation. The Information en-
Oded in the DNA is read (converted into protein) by starting at a fixed point and reading
. ee bases. Thus, we speak of a reading frame of three bases or a triplet, which
ecifies an amino acid. If one (or two) bases are lost or gained, the reading frame
9V
om that point on will be out of phase. Thus, each subsequent triplet will specify a
j-ong amino acid. Polycycllc hydrocarbons can become inserted In the DNA and cause
aine-shift mutations. Chemicals which tend to cause the loss of one base will revert
rations resulting from the addition of one base or the addition of two bases. E. coll
jjJ.UK'*
pg and S. tvphimurium strains TA1535 and TA100 are base-pair substitution mutations.
y are reverted by many alkylatlng chemicals. S. typhimurium strains TA1536,
A 1537, and TA1538 are frame-shift mutants and are reverted by polycyclic aromatic
hydrocarbons.
145
-------
Strain TA100 is derived from TA1535. It contains independently replicating DNA
elements (plasmids) which, by a mechanism that is not currently understood, increase
its sensitivity to mutation by some chemicals.
The second type of mutagenic activity we have measured is mitotic recombination in
£. cerevisiae. Mitotic recombination defines an event in which there is an exchange of
chromosomal material between homologous chromosomes. It implies that these have
been broken, simultaneously or sequentially, in each of the chromosomes. Therefore,
chemicals which increase mitotic recombination may be identified as chemicals which
potentially increase chromosome breakage. Chemicals which cause aberration in the
mechanism involved in the distribution of chromosomes to daughter cells might also be
scored (incorrectly) as mitotic recombinants. In general, chemicals which tend to in-
crease mutation also increase mitotic recombination.
The types of mutagenesis detected by the above mutagenic assay systems are rela-
tively specific, e.g., base-pair substitution or a frame-shift mutation within a specific
gene of the histidine operon in Salmonella, chromosomal exchange between the centro-
mere and a specific gene.
Methods
Bacterial Mutation
The histidine requiring Salmonella typhimurium strains TA100, TA1535, TA1536,
TA1537, and TA1538 are used in all in vitro assays. Strains TA100 and TA1535 are
particularly sensitive to chemicals that cause base-substitution mutations; and strains
TA1536, TA1537, and TA1538 are sensitive to chemicals that cause frame-shift muta-
tions. All tests are performed at five dose levels (1, 5, 10, 50, 100, 500 and 1,000
,ug per plate) and in the presence and absence of a metabolic activation system. In the
a
absence of metabolic activation, approximately 5 x 10 bacteria are added to 2 ml of
top agar (0.6 percent Difco agar, 0.6 percent NaCl, 0.050 mM of L-histidine, O.OSmM
of biotin). The test chemical is added to the top agar, mixed, and then poured onto the
surface of a minimal agar plate, which is composed of Vogel-Bonner E medium, 1.5
percent agar and 2 percent glucose.
146
-------
The metabolic activation system consists of 0. 15 ml of the 9000 x g supernate of
homogenized rat liver, MgCL, KClt glucose-6-phosphate, TPN and sodium phosphate
buffer (pH 7.4); 0.5 ml of this mixture is added directly to the top agar Immediately
before it is poured over the minimal agar plate. The plates are incubated at 37°C for
40 hours. The number of his+ revertant colonies on each plate is then counted.
To determine if the test chemical is toxic to the bacteria, the following modification
of the above technique is used. Approximately 300 bacteria are added to the top agar
and the concentration of L-hlstidine in the top agar is increased to 1 mM. After over-
night incubation, the number of colonies in each plate is determined. A significant
reduction in the number of colonies formed in the presence of the test chemical -when
compared to the controls indicates that the test chemical, or a metabolite of it, is
toxic.
1\/mptic Recombination in Yeast
The ade" heterozygote Saccharomvces cerevtstae D3 is used in all in vitro assay$ .
rfYie D3 strain is incubated for 4 hours at 30°C in phosphate buffer with the test chemt-
cal &t a concentration which provides a 50 percent kill, or at a concentration of 5 per-
cent, whichever is lower. Dilutions of 1:5 and 1:25 of the highest test concentration
are also tested. When present, the metabolic activation system Is the same as; that
uscd In the bacterial assays.
After treatment, the sample is diluted serially in sterile saline and plated on tryptone-
ast agar plates. Plates of a 10 "3 dilution are Incubated for 2 days at 30° C followed by
2 days at 4°C to enhance the development of red pigment, indicative of ade" hompaygbsity.
are exammed f°r the presence of red colonies or red sectors by scanning the
with a dissecting microscope at lOx magnification. Plates of a 10 ~5 dilution are
ctibated for 2 days at 30°C to determine the total number of colony forming units.
a metabolic activation system Is present, the yeast suspension Is Incubated for
i at 37°C with the test chemical. The metabolic activation system is the same as
t used for Salmonella.
147
-------
All experiments are performed in duplicate. Both positive and negative (solvent)
controls are included in each experimental run. Positive reference compounds include
4-0*tolylazo-o-toluldine and MNNG.
Additional test systems are now being evaluated; also attempts are underway to Im-
prove the sensitivity of the yeast assay system.
B.;._3ubtills Assay (Recombination-Deficient Strain)
Kada has reported on several experiments In which mutagenic chemicals have been
tested by the "Rec-assay" (1,2,3). This assay is. based on the differential sensitivity
of bacteria to agents that interact with DNA. The origin of the differential sensitivity
is the Introduction of mutations in one of the pathways for repairing damaged DNA. One
such pathway is the repair of DNA by recombination. This assay system has been used
by Kada-for mutagenic screening of pesticides (2). However, Dr. Kada reported at the
joint United States-Japan Conference on Chemical Carcinogens (December 9-11, 1974)
that he-had not incorporated a metabolic activation system into his procedure.
We are testing five chemicals known to require metabolic activation to develop this
system. Each chemical is tested using the following three procedures to determine which
gives the most sensitive response:
1. The metabolic activation system Is incorporated Into the agar (as In the Salmonella
assays) with the bacteria, and the test chemical is placed on a disc in the center of the
plate.
2. The metabolic activation system is incorporated into the agar and the test chemi-
cal 4s placed in a well in the agar.
3. The metabolic activation system and the chemical are placed in a well in the agar.
The five chemicals we are testing are 1) dlmethylnltrosamlne; 2) /3-naphthylamine
(/3-naphthyl-hydroxylamLne will be used as a positive control); 3) 4-aminobtphenyl;
4) N-2-fluorenylacetamlde; and 5) 3'-methyl-4-dlmethylaminoazobenzene« Dlmethyl-
nitroSamine causes base-pair substitution mutations, /?-naphthylamine causes base-pair
148
-------
substitutions and frame-shift mutations; and the last three chemicals cause only frame-
shift mutations. These experiments also are conducted in the absence of metabolic
activation.
E. coli Polymerase Assay(4)
The principle of this technique is the same as that decribed for B. subtilis, except
that the repair system that has been inactivated is the DNA polymerase gene polA rather
than the recombination genes.
K. coli WP2 Reverse Mutation Assay
The principle of this technique is the same as that for the Salmonella assays. E. coll
^P2 is a tryptophan auxotroph (try") that can be reverted to tryptophan prototrophy by
jxiutagens that cause base-pair substitutions (5,6). The protocols for these experiments
are identical to those described for S. typhlmurium, except that tryptophan Is substi-
tuted for hlstldlne in the top agar.
Development of a More Sensitive Yeast Assay System
A number of carcinogens, e.g., auramine o, have been detected as mutagens in
assays with Su cere vis lae D3 that have not been detected In bacterial assays. Further-
0iore, the assay for mltotlc recombination measures chromosome damage and sub-
sequent mitotlc crossing-over. However, the primary disadvantage of this system
appears to be that the yeast cell wall Is Impermeable. We have observed that many
carcinogens require metabolism to reach active form (procarclnogens such as 2-
ajjthramine and 4-aminobiphenyl) but do not Increase mltotic recombination.
We are approaching the permeability problem In two ways:
1. Testing the feasibility of using glusulase, an enzyme from the snail gut, In order
to Increase the permeability of the yeast cell wall to compounds under test.
2. Cell wall-defective yeast mutants. We have received three strains of yeast from
pr. Clint Ballou of the University of California at Berkeley. Two of the strains are
defective In mannose polysaccharlde biosynthesis. We are comparing these strains
149
-------
with their parent strain for sensitivity to a variety of polycyclic compounds to determine
whether their permeability is increased. If the mutants are more sensitive, we will
obtain or genetically construct adenine heterozygotes to determine whether the sensi-
tivity to mitotic recombination also is Increased. The compounds being tested in this
phase are 4-aminobtphenyl, 4-nitroblphenyl, and 2-anthramine.
References
1. Kada, T.:Mutation Res. 16:165-174 (1972)
2. Mutation Res. 26;243-248 (1974)
3. .; In New Methods in Environmental Chemistry and Toxicology.
F. Coulston, F. Korte, and M. Goto(eds.), International Academic Printing
Go., Tokyo, 1973, pp. 127-133
4. Slater, E. E., Anderson, M. D., and Rosenkranz, H. S.: Cancer Res. 31;
970-973 (1971)
5. Bridges, B. A.: ,Lab. Pract. 21: 413-416 (1972)
6. McCaUa, D. R. and Voutsinos, D.: Mutation Res. 26: 3-16 (1974)
150
-------
UNSCHEDULED DNA SYNTHESIS TESTING OF SUBSTITUTE PESTICIDES
Dr. Ann D. Mitchell*
Unscheduled DNA synthesis (UDS), the incorporation of nucleotides to repair DNA
damage, has been observed in all stages of the cell cycle and occurs in a wide variety
of mammalian cell types. This repair (or misrepair) of DNA damage may be considered
to be an important type of early genetic event leading to cytogenetic aberrations, to
mutagenesis, and/or to carcinogenesis.
Under contract to the National Cancer Institute, we recently developed a new UDS
assay system for prescreening chemical carcinogens. Currently, this system is being
validated through use of a series of chemicals that have shown in animal bioassay tests
to be ultimate carcinogens, procarcinogens, and noncarcinogens. The efficiency and
economy of the new assay system are compatible with the requirements for short-term
prescreening, since the results of each test can be obtained in 2 to 3 days. Moreover,
we have found that the sensitivity and reliability of this method are greater than that
observed in the measurement of DNA repair synthesis by other prescreening approaches.
rf0 date, for approximately 50 chemicals, a greater than 95 percent correlation has
been observed between UDS assay results and the bioassay results. Therefore, we
have used the UDS assay system to test 20 substitute pesticides for the Environmental
protection Agency.
For the UDS assays, we use diploid human fibroblasts (WI-38 cells) contact-inhibited
£n the GI phase of the mitotic cycle by depletion of serum in the medium. Since a low
percentage of cells still may escape synchrony, 10 ~2 M hydroxyurea is added to inhib-
it scheduled (S-phase) DNA synthesis that otherwise would obscure measurement of
fa0 "unscheduled" repair of DNA damage. We have used autoradiography to check the
C0mbination of these two approaches and have found that no greater than 0.2 percent
Oi the nuclei are in S-phase at the time of the assays.
#Biochemical Cytogenetics Program, Biomedical Research Department, Stanford
Research Institute
151
-------
In preparation for the assays, fibroblasts are grown for 2 weeks in small (T-25)
tissue culture flasks. Compounds are then tested in the presence and in the absence
of a metabolic activation system with replicate samples for each concentration of a
series of log dilutions. Appropriate positive and negative controls are tested with each
compound. The positive control, without metabolic activation, is 4-nitroquinoline-N-
oxide; dimethylnitrosamine is used as a positive control with metabolic activation; and
solvent only is the negative control. In the absence of metabolic activation, we expose
the cells to the compounds and to tritiated thymidine (3R-TdR) for incorporation in the
repair of DNA damage for 3 hours. However, an exposure of only one hour is used when
a metabolic activation system is added, because we have found that the liver homogenate
(used as a source of metabolic activation enzymes) also may contain substances toxic
to tissue culture cells. Following the compound exposure, the samples are rinsed
thoroughly and incubated with media containing unlabeled thymidine to remove excess
3H-TdR. The cells then are removed from the flasks, and DNA is extracted using
perchloric acid. An aliquot of the DNA solution is used to measure incorporation of
8H-TdR by scintillation counting. A second aliquot is used to determine the DNA
content by measuring absorbence at 600 HIM following reaction with diphenylamine
reagent. The results can be expressed per unit of DNA and are compared with the
background rate of DNA repair. That is, we calculate the mean, standard deviation,
and standard error for each concentration and then determine the fold increase for
that concentration relative to the negative control. *
We have completed preliminary UDS testing, with and without metabolic activation,
of the 20 substitute pesticides, and these results are presented in Table 1.+ Two
unique observations are that:
1. Positive responses were observed for six compounds. Metabolic activation was
required to obtain positive responses for five compounds; only one substitute pesticide
was positive in the presence and absence of metabolic activation.
* DPM/Mg DNA of treated cells divided by DPM/Vg of negative control.
•*• This testing was completed after the paper was presented and thus has been added.
152
-------
Table 1: Preliminary UDS Results
of Testing 20 Substitute Pesticides
-MA, Without Metabolic Activation
+MA, With Metabolic Activation
UDS Results
Compound -MA +MA
Azodrin + +
Bromacil - +
Cacodylic acid
Captan - +
Dinoseb
DSMA
Dursban
Ethyl parathion -
Fenthion
Folpet - +
Guthion - +
Malathion
Methomyl
Methyl parathion
Monuron - +
MSMA
PCNB
Simazine
Thimet (phorate)
Trifluralin
2. No strong positive responses such as the 20- to 40-fold increases we commonly
^served for positive control compounds were observed for the substitute pesticides.
, all positive responses were approximately two-fold increases. This low
of DNA repair might not have been detected with other, less sensitive approaches.
153
-------
MAMMALIAN SCREENS
Dr. Gordon W. Newell
This part of our presentation describes the work we have done with mammalian
systems and which we are extending. Initially, let me review the dominant lethal pro-
cedure and indicate a few modifications we have made in the procedure over what Is
generally used in the literature; then I will continue with the results and explain what
is coming up in this next year.
Dominant Lethal Procedure
All compounds were supplied by the Environmental Protection Agency.
Although acute toxicity information of some of the compounds was available in the
literature, confirmatory tests were conducted to obtain an LD— under the conditions of
this laboratory. If no data were available, a range-finding dose regimen was initially
conducted, followed by an accurate determination of the oral LD.n.
50
The solubility of each compound is tested with various agents such as water, propylene
glycol, polyethylene glycol, corn oil, trtcaprylln, carboxymethylcellulose, or methylcell-
ulose (Methocel) to determine the most appropriate vehicle for administration.
Based upon single dose acute toxicity data, several dose levels of the test compound
are administered in feed for a period of 2 weeks for the purpose of establishing doses for the
7-week exposure period. Criteria for a maximum tolerated dose Include failure to gain
•weight, development of clinical signs, and mortality. The maximum dosage level selected
jnay produce up to 20 percent weight decrement and mild but transient clinical signs with
oo mortality.
For the dominant lethal study, three dose levels are administered. The highest dose
is the maximum tolerated dose, or percentage thereof, that does not inhibit breeding per-
formance. This will be determined by caging treated males with virgin females and Inspect-
ing the females for the presence of vaginal (mating) plugs. Comparison Is made with vehicle-
treated males. Two approaches are used for the selection of middle and low doses:
Department of Toxicology, Stanford Research Institute, Menlo Park, California
Preceding page Hank
155
-------
1. The intermediate dose is 1/10 of the selected maximum dose with the low
dose 1/100 of the maximum dose, or
2. The middle and low doses are one-half and one-quarter of the highest dose,
respectively.
-Adult male and female ICR/SIM mice are used. Males are proven breeders, while
the females are of virgin stock.
Each experimental group consists of 20 adult male mice. After being fed the compound
in the diet for 7 weeks, each male is mated with two adult virgin female mice for a period
of 7 days. Females are replaced weekly for 8 weeks. Purina Lab Chow and water are
available to the animals at all times.
An unsupplemented group is included in each experiment. Animals in this group are
treated in the same manner as are the test-compound groups.
The positive-reference control group is treated with known mutagen, triethylenemelamine
(TEM), by means of a single intraperitoneal injection of 0.2 mgAg. Breeding and implant
data are obtained throughout the 8-week study.
Pregnant females are sacrificed over a 4-day period starting on the 15th day after the
first day of breeding. This schedule allows for the sacrifice of females between the llth
and 17th days of gestation. A complete autopsy of each female is done to determine
if an incur rent infection is present, since such a condition can induce pre-lmplantation loss
and early fetal deaths. Infected animals are excluded from the experiment.
At the time of sacrifice, each female is scored for early fetal deaths, late fetal deaths,
and living fetuses (all of which provide a total Implant score).
The following parameters indicate an effect in dominant lethal studies: total implants
(live fetuses plus early and late fetal deaths), total dead (early and late fetal deaths),
and dead implants per total Implants. All data are analyzed for statistical significance by
•the Meat.
Results
All 10 pesticides studied by the dominant lethal procedure in ICR mice produced nega-
tive responses. These were parathlon, methyl parathion, captan, Thimet, folpet,
rnalathion, PCNB, bromacll, Azodrin, and guthion.
156
-------
Oral Reference Mutagens
The major reference mutagen for mammalian studies Is trlethylenemelamine (TEM).
It has been shown to be effective in producing dominant lethal effects, and also causes
heritable mutagenlc responses in the translocatlon test. Not only Is it active when
djuinistered intraperltoneally, it is also effective by the oral route when administered
. the drinking water. However, this compound is available only from one source, a
jiarmaceutical manufacturer, and the company has decided to stop production. Other com-
rjound3 are effective mutagens when given by injection, but data are lacking when they are
^ministered per os. Since a major route of pesticide hazard is by ingestion, it was
decided to investigate additional compounds as candidate oral reference mutagens.
jn this portion of the program we are studying the following compounds: cyclophosphamlde,
methanesulfonate (EMS), methyl methanesulfonate (MMS), trls (1-azirldlnyl) phosphlne
(TEPA), trls [l-<2-methyl azlridinyl)] phosphine oxide (METEPA), and trimethyl
. O9phate (TMP). Where necessary, we are testing LE|* a to determine the oral acute
xicity of each material.
Based on this information as well as other data available from the literature regarding
ge levels shown to cause positive mutagenlc responses when the compound Is administered
• raperitoneally, several dose levels are selected for a short-term dominant lethal
ftidy. For these compounds, dose levels are administered In the feed to adult male mice
_ 2 weeks so that a dose response useful for subsequent breeding can be established.
fo* u
. fae end of the 2-week exposure period, the males are caged with virgin females, and the
•**
ajes are examined dally for the presence of vaginal (mating) plugs. At mid-term of
egnancy, the females are sacrificed and examined for total implants, as well as for
rjy and late fetal death. These initial screenings should indicate dosage levels which
_,0t affect breeding performance and will Identify effective dosages for mutagenic activity
tJ
-------
In the more extensive dominant lethal study, the test compound is administered to
adult males at several dosage levels in the diet for 8 weeks. At the end of this time,
treated males are caged with virgin female mice and the females are examined daily for
the presence of vaginal (mating) plugs. Evaluations for mutagenic responses are con-
ducted as indicated above. Results obtained at this phase should indicate dosage levels
that do not affect breeding performance but identify effective dosages for mutagenic
activity by the oral route of administration. As such, the following conditions may exist:
1) one or more dose levels show positive oral mutagenic activity, in which case the
compound is retained for validation in the translocation procedure; and 2) none of the
dosage levels tested indicate mutagenic activity. No further work is done on that compound
Based on positive response results obtained in the dominant lethal studies of candidate
oral mutagens, two dose levels of the compound will be used in the translocation procedure,
A high dietary level is the maximum tolerated dose (MTD) and the low level is one-fourth
of the MTD.
Adult male proven breeder mice, ICR/SIM strain, are fed the test compound at the
selected dietary levels for 8 weeks. We also use an untreated control group as well as a
group treated with the standard reference mutagen, triethylenemalamine (TEM); the TEM
treatment is for 4 weeks, with the compound being added to the drinking water and fresh
solutions prepared daily.
At the end of the treatment period, 25 to 30 males from each group are mated to 2 virgin
females each for 1 week. These females are allowed to deliver their offspring; the
parent males are discarded.
Our experience indicates that this breeding schedule should provide 500 to 600 offspring.
The litters are raised to weaning age, at which time the females are discarded, and the F..
males are raised to maturity (11 weeks of age).
At maturity, 100 F males from each group are mated to 3 virgin females each for
2 weeks. We examine females daily to identify the breeding date (presence of vaginal plugs;
Fourteen days after confirmed mating, the females are sacrificed and examined for total
implants, as well as for early and late fetal deaths. Males bred to females producing
158
-------
litters which do not meet "normal criteria" are considered as suspect translocate
animals. Each of those males is rebred to three additional virgin females. We
evaluate the second breeding in the same manner as the first breeding.
Our criteria for classifying a male as "partially sterile," "sterile," or "non-
breeder" are:
0 Partially sterile male: If all three females are pregnant, each must have
nine or less live implants, with at least one having six or less live implants. If
two of three females are pregnant, both must have nine or less live implants, with
one having six or less live implants. If only one of three females is pregnant, this
female must have six or less live implants.
« Sterile male: None of three females are pregnant — previously identified
by presence of vaginal plug.
» Nonbreeder male: None of three females are pregnant — not previously
identified by presence of vaginal plug.
£Qy Female not fitting one of these criteria is considered normal and discarded.
For each male meeting the criteria of sterile or partially sterile after a second
. reedingi we conduct cytogenetic evaluations. Such evaluations include measure-
ent of testes weight, meiotic metaphase I cell evaluation for translocations, and
paration of photomicrographs for each animal studied.
I will show you where we are at the present time in the evaluation of some of
se candidate oral reference mutagens.
Table 1 shows the results with cyclophosphamide. We are conducting a screen
A as a result you will not see control animals, but we do have a good reference
0.^
ta base. We know that the average implants of a control female are in a range
10. 4» Plus or minus °'2« Tne average dead implants per female are normally
mid about 0. 2 to 0. 4 in our experience. So, if we see an increase of implants,
know that the response is suggesting a mutagenic event.
***
159
-------
Table 1: Preliminary Breeding Data From Male Mice
Treated Orally with Cyclophosphamide
Dose
(mgAg)
Single
125
250
500
Week Pregnant
Average
Implants/Female
Average Dead
Implants/Female
1
2
1
2
1
2
70
60
2Q
40
17
33
9.86
9.67
7.50
11.25
15.00
8.00
5.57
0.50
4.50
4.00
0
1.00
Multiple
37.5
75
1
2
3
1
2
3
38
50
50
60
30
40
11.33
9.25
5.67
4.17
5.33
14.25
3.67
3.00
1.33
2.67
2.33
7.75
These are relatively small numbers. We are looking at single dose data, trying to find
a tolerated level. The multiple system here is five doses, repeated dally, at this level
per kilogram per day.
It is obvious in looking at the repeat data that these dose levels are markedly reducing the
litter size and causing an increase in dead implants. Again, with 75 mg/kg of cyclophos-
phamide, there is a marked reduction of implant frequency, where we would like to see
a normal number of implants but a significant increase in dead implants. This is really
too toxic a level.
Tables 2 and 3 show the same type of response with EMS and MMS. There appear to
be problems of cumulative effects, and I question whether we can use these materials
successfully.
160
-------
Table 2: Preliminary Breeding Data From Male Mice
Treated Orally With EMS
pose % Average Average Dead
Week Pregnant Implants/Female Implants/Female
jingle.
125
250
500
250
1
2
1
2
1
2
1
2
1
2
80
100
60
50
90
70
70
80
0
30
11.38
14.70
10.83
11.20
7.89
11.00
10.57
11.38
-
9.33
0.25
1.00
0.17
0.60
2.56
0.86
4.86
4.50
-
2.00
Table 3: Preliminary Breeding Data From Male Mice
Treated Orally With MMS
Dose
Single
125
250
315
Week
1
2
1
2
1
2
1
2
Pregnant
60
90
20
90
0
31
0
20
Average
Implants/
Female
5.33
10.67
2.50
10.22
—
6.20
.«.
9.50
Average Dead
Implants/Female
2.50
0.67
2.50
2.89
—
2.60
_„
3,00
161
-------
Tables 4 and 5 show similar data for TEPA and METEPA. In Table 6 the
preliminary responses with TMP appear promising and we have hopes that this
compound may eventually prove to be a satisfactory new oral reference mutagen.
Table 4: Preliminary Breeding Data From Male Mice
Treated Orally With TEPA
Dose
(nig/kg)
Single
25
50
125
Multiple
25
50
Week
1
2
1
2
1
2
1
2
3
1
2
3
%
Pregnant
80
80
40
70
10
40
10
50
40
0
20
10
Average
Implants/Female
9.13
13.13
6.00
13.29
10.50
8.13
3.00
6.00
9.00
-
1.50
12.00
Average Dead
Implants/Female
4.63
0.88
4.00
1.86
8.00
2.75
3.00
4.60
6.00
-
1.50
12.00
162
-------
Table 5: Preliminary Breeding Data From Male
Mice Treated Orally With METEPA
Dose %
(mgAg) Week Pregnant
Single
50 1
2
100 1
2
300 1
2
Multiple
31.25 1
2
3
62.5 1
2
3
50
80
60
70
17
25
80
70
67
80
70
30
Average
Implants/ Average Dead
Female Implants/Female
13.6
14.5
7.33
7.86
5.33
5.75
10.5
11.86
9.33
10.25
7.86
2.00
1.60
3.00
5.17
4.00
4.67
4.50
2.50
1.71
1.67
4.63
2.71
0.33
163
-------
Table 6: Preliminary Breeding Data From Male
Mice Treated Orally With TMP
Dose
(mgAg)
Single
2000
2500
3500
Multiple
750
Week
1
2
1
2
1
2
1
2
%
Pregnant
50
70
50
40
17
83
38
63
Average
Implants/Female
10.00
11.29
6.40
7.50
7.00
13.60
7.67
10.60
Average Dead
Implants/Female
0.80
0.86
3.00
0.50
6.00
1.00
6.00
7.60
164
-------
DR. CURLEY: Since the toxicant first comes in contact with epithelial cells
and is absorbed or transported by some transport mechanism and detoxified, or
metabolically altered, what is the feasibility of developing, validating, and utiliz-
ing such cell culture systems which would approximate the normal cellular
exposure to the toxicant in the animal, rather than using fibroblasts for mutagenic
or carcinogenic jn vitro screens?
DR. MITCHELL: I think several laboratories are initiating work on cell
transformation in epithelial cells. Results from these studies might be anticipated
in one and a half or 2 years.
DR. CASCIANO: Dr. Weinstein is developing epithelial cell systems, and several
laboratories are developing rat liver epithelial type cultures, and they're showing
a very good correlation between ability to grow in soft agar and ability to form
carcinomas in the whole animal.
QUESTION: This is a question to Dr. Newell. When you're doing these
studies with high dosages, instead of repeated doses — do these pesticide chemicals
on oncogenesis really transform the cells by selecting somehow the pre-existent
flialignant cells, or schussing on to the oncogenesis virus that carries out this
transformation?
DR. NEWELL: As I understand your question, you are asking about mutagenesis
responses by pesticides in mammalian systems. With mammalian systems, we
not found a positive response yet, so I don't think one can give you a direct
. With the microbiological systems there appears a direct effect on the DNA.
QUESTION: I'd like to address a comment and a question to Dr. Newell. We
H know that there's a multitude of possible microbiological screens, both natural
nd mutant, and it would seem that if we go into a study with a preconceived view
a compound is a carcinogen, mutagen, teratogen, what have you, that by
165
-------
analogy with another compound, with another chemical, by a gut feel, or whatever
criteria seem to be in vogue, we'll never put the question to rest. My question is
how are several negative findings and one or two positive findings ever going to
answer this question. In other words, when is a compound finally going to be
cleared of being a presumptive carcinogen, mutagen, or what have you?
DR. NEWELL: I think this is a very proper question from someone concerned
with the problem of registering compounds and getting them approved. As I com-
mented earlier, one of the main reasons for these large amounts of data developed
with known carcinogens and noncarcinogens has been to develop validity for the
system. The data developed with all these compounds have come from investiga-
tors at universities, NIH, and NIEHS, and organizations like ourselves, all attempt-
ing to validate the systems.
These kinds of information are for reference, and are not intended to intimidate
anyone. Whether or not they are directly meaningful to a pesticide you may be
concerned with is secondary. The information developed for those compounds
is one way to show that the procedures have some validity for what we're attempt-
ing to do.
Some people have suggested that more of our studies should be done with the
compounds under code,.. Some say when you know the compound your results will
be what you want to get. Hopefully, that's not raising a question of integrity, but
I think it's a proper question to ask.
When we look at new chemicals and new structure types, unless we have enough
information about compounds active in mammalian systems (and if you are trying
to compare mammalian results with the in vitro ones), unless you have a reliable
base of reference, none of the information is meaningful. Today, the correlation,
the mutational responses of the in vitro microbiological systems are 80 to 85
percent from studying chemicals known to cause carcinogenesis responses in
animals.
166
-------
You asked if there is a positive in vitro response, will it ever get put to rest.
If one is concerned, one can use other systems for confirmation, i.e., cytogenetic
studies or the translocation test.
Actually we've done a number of translocation studies on drugs and agricultural
chemicals, providing heritable-type experimental data on such products.
DR. COMMONER: I'm glad this question came up, because it's the nitty gritty,
and although I will present considerable data that read on this after lunch, I do
want to make a comment here.
Let's simply empirically say you have a compound that you want to put out in
such a way that people will be exposed to it. You want to be sure that it's not a
carcinogen. Now I define a carcinogen as something that causes cancer in people.
So you want to be very sure that it doesn't cause cancer in people.
At the present time there is literally no way of doing that, short of causing
cancer in people. So if you have industrial exposure data, then you can say some-
thing about it. Now obviously that's true of very few compounds.
What can you do ? You can put it through an animal test. That tells you that it
causes cancer in mice and rats, and it does not necessarily tell you that it causes
cancer in people. However, we know that a lot of things that cause cancer in mice
rats do cause cancer in people, and it's a big warning sign.
the trouble with doing that is that it is fantastically expensive and slow.
last figures we got in Seattle from NCR were an average of $150, 000 a com-
pound and 3 years of work.
There is something else that can be done, and that's where all of the in vitro
^jd other tests come in. Suppose we could give you a fast test that will take 2
jays and cost $500.00 a compound, in which the bacteria stand up on end and
yfcistle Dixie if they're exposed to a carcinogen, and if it's not a carcinogen, they
n't whistle Dixie. I say Dixie because we're here in Virginia.
167
-------
I put that to you in exactly that unreal way. If I can establish with statistical
reliability that every time — or within 90 odd percent —- that organism whistles
Dixie we know that it is seeing a substance which we know from animal tests
causes cancer in mice and rats, and you can test it out with unknowns by then
feeding it to rats and mice, then you have got an empirical test which is, within
a certain statistical statement, reliable. It doesn't matter that it's bacterial,
Drosophila, or, in the case of the Japanese, the eggs of silk worms. It doesn't
matter what it is, and it doesn't matter what the response is, as long as I can
tell you what the correlation is and what the mathematical probabilities are.
All that does for you, then, is give you a much quicker way of screening com-
pounds and finding out whether, if you did test them with animals, they would
cause cancer in animals. Now that is one feature of whatwe're doing here, but
it's not the whole thing.
The other side of it is that the methodology that's used for whistling Dixie hap-
pens to be, in our case, mutations. That has two meanings. One is that it's just
plain whistling Dixie. It doesn't matter whether it's mutations or anything else,
but there is another significance. People may suffer mutations. You have to
worry about that. Therefore, those data have a dual meaning, and the Drosophila
work is an effort to take mutation effect and put it into a complex animal where, for
example, you have whole chromosomes, which bacteria don't , so that you can
begin to get some sense whether those things cause mutations in people, because
they may not, even though they cause mutations in bacteria.
In other words, what's confusing is, there is a multiple motivation in these
things. One is the purely empirical attempt to save money and time, and the work
that I'll describe after lunch is that and nothing else. That's all it is. And it's
enough, it works. So that you now can save the money and time.
It also tells you something about some of the compounds. For example, we've
discovered that captan is positive in our system, and I can tell you the probability
with which that means that if captan is fed to a mouse or rat it will cause cancer.
168
-------
All hard data are going to be stated in probabilities. There are no absolutes in
this thing at all.
And the last thing I want to say is, after this is all done, all you know is that
you've got a substance that, with a certain probability, is likely to cause cancer
in rats and mice. After that there's a big, big problem of figuring out whether it
causes cancer in people, and the good news about that is that I think we're begin-
ning to get a method for going from the one group to the other,
DR. DATTA: Do you think the support we gave you through the chemical program,
which is very meager, will be able to summarize for us what type of guidelines on
mutagenesis tests that we can give to the registrant to bring in data for us, or do
you think that you'll need at least 3-5 years of support from us?
DR. COMMONER: My guess is that within a year it will be possible to turn over
to industrial companies a protocol for running the Salmonella tests they can run
themselves at about the cost that I gave you, which will enable them to screen any
compound or intermediate or pilot plant material. We have now tested the set of
knowns, and what's important now is to start testing mixed unknowns and to develop
some more methodology for that. There is also a little refinement on the system
that has to be done to eliminate a few false negatives.
Once that's done, I think it can be delivered. We have counted 43, 000 plates in
the last year in our laboratory at a cost of $125, 000. That's a hundred compounds,
but we're losing money. No, if it weren't for a few things, to play safe I would
say a compound will cost you $300 to $500, and you can get through the work, if
you want to push it, in 3 days.
I think in a year that can be delivered to any reasonable commercial company.
That's for compound, shall we say preparations. For environmental samples,
's say an extract of soil that has been treated, it is going to take somewhat
r, but again, I think after a year and a half's work on our part, we'll have
Protocol that will enable you to do the similar analysis, fractionate the compounds,
I'll show you how this afternoon, let's say on an extract of animal fat that's
been fed something.
169
-------
DR. CRANMER: I think Dr. Newell had a comment, but I think I would like
to draw on the lead given by the Deputy Administrator last night when he said that
scientists sometimes comment on the law as well as lawyers commenting on
science, and I believe — it's been a while since I read it — but FIFRA, as amended,
said something about proving not just the toxic effect of a compound, but a reason-
able expectation that that effect will be effected on man or the environment under
conditions of use. I believe that's correct, isn't it?
COMMENT: Reasonable adverse effect.
DR. CRANMER: Reasonable adverse effect. So, indeed, I think we all agree
with Dr. Commoner that the utility of your first level of evaluation in the systems
is great. The utility in making that last analysis may be a difficult road, at least
it seems to me it might.
DR. NEWELL: I am attempting to provide you with information and background
about utility and reliability of these procedures. There remain. 15 to 20 percent
inconsistencies of response, and I believe because of this the EPA did not include
these in vitro systems in the proposed registration guidelines.
I believe Dr. Commoner will tell us this afternoon how to overcome some of the
false negatives, but we still have other problems. Seven months ago, when we were
intensively discussing these methodologies with people from various governmental
agencies, one of the emphatic comments made by Dr. John Weisbergerwas, "For
God sakes, don't say this is an absolute method at this time. We need at least a
year, or more for development before people can run with it. "
There is a lot of interest in these systems, and it looks like an interesting
tool, but much needs to be done with it.
COMMENT: I think this is really the problem, though, in terms of this new
situation that's developing where compounds are banned just because they're
presumed to be carcinogens, and this I think relates to one or two positive tests,
170
-------
even though there are many negatives. If there are enough data to suggest that
it's possibly a carcinogen, that seems to be all that's needed.
QUESTION: What specific example of a compound?
DR. COMMONER: Can I make a comment? I just want to pick up one thing.
What you're saying is that some of the in vitro tests work and others don't work.
So why do you seize on the one that works and call it a presumptive carcinogen?
That's why I said whistling Dixie. If there is one method — it doesn't matter what
it is — that you can show mathematically correlates, then the fact that others
don't work doesn't mean a thing. This is simply an empirical correlation.
Now it means a lot with respect to the theoretical questions and so on, and I
think this has to be understood. People say well, this thing works with Salmonella
but it doesn't work with yeast. So which one are we going to believe? The
nuestion is, what is the mathematical correlation of the Salmonella data to the
imal data, and what is the mathematical correlation of the yeast data to the
nimal data? It turns out that the Salmonella correlations are good, the yeast
correlations are bad; therefore you can, if you want to, use the Salmonella data.
don't advise using one with a bad correlation.
at the Japanese-American meeting there was a very interesting report
•yen on AF which is a food additive. Two years ago it was discovered to be positive
e>
the Salmonella test, and the people who did it said this is likely to be a
t-cinogen toward animals and we should take it out of the market. The Govern-
was cautious and said no, let's test it against animals. It was tested
ainst animals, and 2 years later it was found to be positive, and it was taken
&&
$ the market.
j^0w what would the Salmonella test do? It saved 2 years and saved money.
I said, it doesn't matter whether the yeast test doesn't work. If you've got one
£*
works with a certain probability, and you want to save 2 years of human
then you use it.
171
-------
COMMENT: Dr. Commoner, there is one thing that we come across all the
time. It's become very difficult on our part, as a regulatory agency. Say, for
example, two chemicals have a very similar structure, but one is found to be
positive, whereas the other is found to be negative. Now what do you do? Do
you study one for 2 years and not study the other for 2 years ? Or what do you do?
What would you recommend to the company that they should do?
DR. NEWELL: I think it's worth a comment, because Dr. Dale raised a
question about the long-term studies. However, let me emphasize again that
we're talking about a mutagenic response, and that's why I think in this case a
translocation study should be done on captan. In that procedure, you're looking
at a potential heritable effect, and you're looking at germinal cells, not somatic
cells, and tuat is the difference, as Dr. Commoner pointed it out. Don't imply
that captan is necessarily a carcinogen because of these studies. It may not even
be a mutagen.
QUESTION: I'd like to make a direct question to Dr. Newell. This is in regard
to the study where you had done the diet incorporation studies for a period of 7
weeks, followed by mating of 8 weeks. This was incorporating the ten compounds
in the diet. Did you use TEM intravenously as a positive control in these studies?
DR. NEWELL: Not intravenously. We've used TEM both ways. Early in the
game we gave it intraperitoneally and produced a postive response in animals for
about five weeks. We now add TEM to the drinking water. We use only a 4-
week treatment, a period long enough to affect the cells in the postmeiotic stage.
There's no need to treat with TEM for 7 weeks.
QUESTION: Didn't you make the comment that you didn't think this was the
way to go, diet incorporation?
DR. NEWELL: No, I said it was the way to go. And that's the way we're
testing all of these compounds by incorporating in the diet.
172
-------
USE OF MUTAGENESIS TEST
TO INDICATE CARCINOGENESIS
Dr. Barry Commoner*
TO BE PUBLISHED UNDER THE TITLE
"CHEMICAL CARCINOGENS IN THE
ENVIRONMENT" IN IDENTIFICATION AND
ANALYSIS OF ORGANIC POLLUTANTS IN
WATER. L. H. KEITH (ED.), ANN ARBOR
SCIENCE PUBLISHERS, ANN ARBOR,
MICHIGAN (1976),
*Center for the Biology of Natural Systems, Washington University, St. Louis,
Missouri
173
-------
CRITERIA. FOR COMPARISON OF TERATOGENESIS PROTOCOLS
Dr. Herbert J. Schumacher*
In 1966, the FDA and, In 1967, the World Health Organization developed principles for
testing of chemicals for teratogenicity. These recommendations still constitute the basic
framework of experimental teratology testing even though a great many refinements in
testing procedures have been proposed and adopted by numerous investigators.
However, no single test or combination of experimental animal tests yet designed
can predict absolutely that a particular environmental agent constitutes a teratogenic risk
in man. Therefore, doubts have been cast on the relevance and predictive value of terato-
genesis testing, but according to present knowledge, there is no chemical assumed to be
teratogenic in man that has not produced malformation in rodents or lagomorphs.
Furthermore, the predictive value of animal tests can be enhanced by paying attention
to the following criteria which are essential, but are often overlooked in the outcome
of comparative studies designed to safeguard the human population.
During this presentation we will discuss some specific considerations which we feel are
important in experimental teratology design and must be accounted for in the interests of
standardization, if you will, so that protocols can be adequately compared.
First of all, the state of health and standard care of the animal must be carefully con-
trolled. Animals that are used for teratological studies must be healthy and should be
housed under the best possible environmental conditions. A high standard of animal
care must prevail. The animal quarters should provide controlled temperature, humidity,
adequate light, and protection from noise and other interference since these factors are
capable of inducing malformation in one or more species.
Proper caging must be considered; overcrowding, for example, may Increase the
noxious effects of chemicals that act upon the central nervous system. Since the admin-
istration of a chemical may change the food consumption of treated animals, it may
*National Center for Toxicological Research, Jefferson, Arkansas
175
Preceding page blank
-------
be necessary to adjust food intake of control animals correspondingly. Results reported
in the literature indicate that nutritional stress or imbalance may produce malformation
or modified action of other teratogens. Thus diet must be carefully controlled in any
teratological study.
For example, the reduction of about 30 percent of food intake in mice produces effects
on the teratological outcome of the results. Fifty percent food reduction is needed to
produce the same results in rats, and no effect was observed in rabbits. Animal bedding
is another factor to be considered, since bedding fumigated with ethylene oxide may
interfere with teratological studies, in that vitamins and proteins can be destroyed or
transformed by this chemical.
A similar situation may exist with pine bedding or with pesticide residues in bedding or
in feed, since the rate of biotransformation of the test material could be changed following
alteration of microsomal activity of the experimental animal. Mycotoxins, plant toxins,
and naturally occurring steroidal and estrogenic compounds in the feed can also alter
the outcome of teratological tests.
Microorganisms and parasitic infections induce teratogenic effects In experimental
animals. Furthermore, significant differences in litter size, body weigit gain, and water
consumption have been reported on distilled water instead of tap water which is available
ad libitum.
Type and form of feed also affect the litter size and the pup weight. The time of dosing
of animals should be held constant to avoid further variability of the results. The species
of choice is also important in teratogenic screening of chemicals. It should have an
easily determinable time of fertilization with a high rate of fertility. The present state
of knowledge of teratology, the embryology and genetic background, as well as incidence
of spontaneous and inducible malformations, should be considered in the species or
strains to be chosen.
At least two species, one rodent and one lagomorph, should be used, according to the
Agencies, in order to fulfill the general stereotypic requirements. If only such species
176
-------
are used for testing, then age and parity should be known for the following reasons. In
polydacious animals, such as mice, hamsters, rats, and rabbits, declining reproductive
capacity with Increasing age has been demonstrated. Studies indicate that in mice the
highest rate of spontaneous anomalies occurs in the first litter and decreases to a very
low incidence in the fourth Utter.
In rabbits and rats reduced litter size with increasing age was observed, but the incidence
of spontaneous malformation was unaffected by maternal age and parity.
Multispecies tests, as proposed by Wilson, should be used. This proposal has the
advantage of grading the number of animals used from plentiful in the less expensive
rodent species to fewer in the more expensive species. Furthermore, such a procedure
would provide much needed multispecies data and the nonrodent species would not have
the yolk-sac placenta. Generally the routes chosen for teratological testing should
simulate the exposure of man. In the case of oral route, one must determine whether the
compound is absorbed by the species in the study. Oral administration should be by
standardized gavage or with capsules, since the amount of the compound ingested can be
accurately determined and usually insures high levels of absorption.
In some studies administration of the chemical may be in the food or drinking water
on the strict condition that the chemical stability and homogeneity of the feed has been
ascertained. Food and water consumption has to be determined accurately. The use of
intravenous, intramuscular, subcutaneous , and intraperltoneal routes generally is without
nroblems, but interpretation of the results may be difficult when the local tolerance at the
site of injection is poor. In inhalation studies, stress, restraining, and/or hypoxia may
the outcome of results.
finally, during the course of treatments females should not be subjected to any kind
f~rough handling, which in itself could cause disturbances of pregnancies as, for example,
or cleft palates.
•Treatment periods generally cover three periods in mice and rats. For example, there
_ o to 19, 6 to 15, and 7 to 9 days of pregnancy treatment regimens. The experimental
£1*
HifflcU*ties encounterec* ln routine teratogenic testing include some problems Inherent to
action of chemicals. Compounds which destroy the intestinal flora, chemicals with
1*1
177
-------
prolonged or intensive pharmacological activity, and substances causing enzyme induction
or inhibition which will rapidly harm the health of the mother can only be administered
for a short period of time.
It may be useful, or even necessary, to limit the treatment of the pregnant female to
a few days, or even a single day, provided that the whole period of organogenesis is
covered by using an appropriate number of animal groups. Furthermore, the pharmaco-
kinetic characteristics of the compound in animal species or strains should be taken into
account. The selection of adequate doses is important and the following factors must be
taken into account: 1) the purity and the physical chemical characteristics of the specific
chemicals to be administered; 2) the nature of the pharmacological activity exerted by the
compound in the particular species; 3) the dose at which this activity manifests itself;
4) the degree to which man is exposed to the chemical; and 5) pharmacoklnetlcs and metabo-
lism In animal species should be compared with those in man. Such, data could be obtained
from acute studies, preclinlcal trials, or by the surveillance of workers In the factories.
There are distinct species, strain, and even interstrain differences In the response of
embryos to both suspected and documented teratogenic chemicals. These differences
appear1 to be multlfactorlal and are related to the fact that teratogeriesis is not a single
toxic response, but is, rather, the result of a complex interaction between the stage of
development at which a compound is administered, the variability time of the compound for
placental transfer, its cellular localization within the embryo, and Its mechanism of
action In situ.
There is, however, one aspect that has not been consistently evaluated in experimental
animals^which we feel may provide a more meaningful mechanistic interpretation of the
results. A combined analysis of placental transfer kinetics and localization within the
embryo of either chemical or its metabolites should be undertaken. Only when infbrinatibn
from-such studies Is forthcoming will it be possible to relate both the dose administered to
the dam and ;the dose reaching the embryo to the observed biological effect. It Is obvious
that whehfmany other parameters, such as plasma half-life, metabolism, protein binding,
and1 tissue 'distribution in adult animals, are measured, they provide little information to
account for the large differences in offspring susceptibility. However, a combined analysis
178
-------
of these parameters, together with placenta! transfer data, will allow an evaluation of
factors which control exposure time and permit the study of the presumed mechanism
of action of a chemical or metabolite in situ.
Other factors are the toxicity of the chemicals to the mother animal which will ultimately
determine the regimen of treatment. The influence of the carrier substance or the solvent
should not seriously influence absorption, distribution, metabolism, or excretion, or
alter toxic characteristics or chemical properties of the test compound. The particle
size of the test chemical, if it Is insoluble in the chosen vehicle, must be sufficiently
small and well-defined and its size stated because of the role this parameter plays in
absorption.
Factors originating from metabolic interactions between various pesticides, food additives
and the variety of environmental chemicals have to be taken into consideration. These
effects are due to microsomal activity of the liver, placenta, and/or offspring, which
may be stimulated or inhibited. In this way, substances foreign to the body may modify
the metabolism of other chemicals that are administered simultaneously.
With this consideration in mind, it is desirable that the highest dose studies should
exert some systemic effect in the pregnant female. It is important that the highest dose
does not cause excessive embryo lethality, and the fetal loss should not be higher than 50
percent. The lowest dose used should not show any systemic effect in the dam.
Teratogenic effects produced by extremely high doses, compared with human dosage,
are probably relevant only when these doses do not impair the well-being of the mother,
nd if they are not accompanied in a dose-dependent fashion or by other disturbances
Of the pregnancy.
In order to establish statistically a valid dose-response relationship, generally three to
five doses should be chosen. At least two of the treatments scheduled should be utilized
any teratogenic investigation. A series of control animals of equal age should be treated
ith the vehicle under the same condition as the animals receiving the compound under
f-udy. T^6 number of animals per series must be sufficient to enable a statistical evaluatiot
e the results. All doses should be expressed on a mllligram/kllogram/day basis. If
>*T H*^
-------
the compound is given in the diet or drinking water, the dose should then be expressed
on a milligram/kilogram/body weight/day basis. Experience has demonstrated that
screening studies involving 20 to 40 pregnant rodents per group for each dose, and
three to five groups per dose, can provide a good indication of possible effects on
the offspring, provided the studies are properly set up and well conducted.
This would include randomized assignment of animals to control and experimental
groups and randomized assignment of cage positions on the racks. Careful replication
of experimental conditions and the repetition of experiments increase greatly the confi-
dence with which an experiment can be evaluated. For studies in which two or more
methods of examination are employed — e.g., when some litters are delivered by cesar-
ean section and other dams are allowed to rear their offspring for studies on survival,
regeneration, or behavior, or when visceral or skeletal examinations are performed sep-
arately — it is advisable to have at least 20 litters for each method of examination.
For larger species, cats, dogs, pigs, sheep, and primates, the number of animals
should be as large as one can handle in order to obtain reproducible results and to accumu-
late much needed background information on such species. Equivocal results should be
clarified by further experimentation. The investigator may adjust the dosage and/or
dosing period to achieve a more pronounced and recognizable response.
There are different methods for examination of offspring. There is at present no method
that allows all of the fetuses to be examined for visceral and skeletal abnormalities. Fetuses
should be randomly selected for examination and examined by an observer without prior
knowledge of maternal exposure.
In the reviewing of different teratological testing protocols, it becomes apparent that
universally acceptable definitions of malformations and definitions of degrees of fetal
toxicities are needed. The litter or the fetus can be considered in evaluating teratogenic
effects. Both units are of importance. Generally, the litter is the unit for statistical
analysis.
For the interpretation of teratological results, the following statistical methods are
mainly used. The two-way analysis of variance, the regression analysis, and tests of
180
-------
best fit must be used to ascertain significance. Chl-square test, Fisher's exact test,
and the T-test are mainly used in the analysis based on the individual fetuses. Much
more work in statistics is needed to maximally evaluate the teratological results and
to evaluate risks.
Finally, we suggest that only by standardization of teratogenic testing protocols, by use
of the best available and agreed-upon methodology, and by use of well-trained personnel,
can we replace the existing confusion in comparing the results of teratogenic research
between laboratories and different species of animals. However, standardization should
in no way be Interpreted as a hindrance or deterrent to the innovative development of
new and better methodology.
181
-------
FEASIBILITY STUDY: DETECTION OF CHEMICAL-INDUCED
MUTATION BY ASSAY OF METABOLIC CHARACTERISTICS
Dr. Harvey W. Mohrenweiser*
Monitoring undue exposure to potentially mutagenic agents and thus reducing the
Incidence of new mutational events is a key to preventing the suspected negative impact
of mutations on the health and well-being of the population. In order to prevent the
undue exposure, it is essential that systems capable of predicting mutagenic activity
be incorporated into toxicological evaluation protocols. Toward this goal, the
Division of Mutagenic Research at the National Center for Toxicological Research is
attempting to develop an integrated approach to the determination of mutagenic
potential by organizing an extensive battery of assays in a pyramidal array; other
people may call it the tier approach or various other names. As indicated in Figure 1,
this is a theoretical example. There are some obvious exclusions. This was intended
just to make sure you would not read anything into the figure that's not there.
Figure 1
Theoretical Pyramidal Array
V.
IV,
m.
n.
i.
In Vivo Microlesion
A^aav
Drosophila
In Vitro
•n^ ^MtWM0m»
DNA Repair
Structural
Features
i In Vivo Repair
MC w
Culture
Dominant
Lethal Assaj
In Vivo |
Cytogeneticsj
HMA W ~
In Vitro
DNA
Binding
Yeast
Bacterial
Systems
Host-mediated assay
Mammalian cell culture
*National Center for Toxicological Research, Jefferson, Arkansas
183 Preceding page blank
-------
The pyramidal array can be composed of a range of potential test systems with
varying degrees of complexity and utility. These will include such things as simple
test systems, as we heard Dr. Commoner talk about this afternoon, or some of the
things that Dr. Newell talked about this morning. They may be very sensitive to
either limited or broad range agents. It will also include more complex systems
which may be appropriate for estimating the relative degree of risk associated with
normal exposure, such as the Specific Locus Assay, which is the one currently
being utilized in mutagenic assay systems.
As you will notice on the lower portion, we have very simple sorts of tests,
structural features, DNA binding, and some of the bacterial systems. As you go up
one more layer, you observe a more complex system, dominant lethality and the
in vitro and in vivo systems, including the Drosophila system. In no way do I mean
to imply that this is an appropriate organization, but it gives you an idea of our
thoughts on the organization of more complex systems.
Our thoughts are that this integrated array of test systems should serve as a
guide to direct accumulation of the scientific data necessary for valid regulatory
decisions. It is again important to emphasize that this is only an example of the
tests available and indicative of a possible organization.
It should also be pointed out that such important features as comparative metab-
olism, pharmacokinetics, dosimetry, and molecular biology data will be necessary,
particularly if transitions between various tests and extrapolation from the various
tests, particularly at the upper level of the pyramid, are to be meaningful. These
latter areas of comparative metabolism, pharmacokinetics, and dosimetry studies
particularly are areas in which we are now currently developing active research
programs.
We have two programs currently underway, one which I will talk about and the
other which Dr. Casciano will talk about. The one I'm going to talk about is a
feasibility study of detection of chemically induced mutations by assay of metabolic
characteristics. Special emphasis in this area has been placed on the development
184
-------
of in vivo mammalian systems which would be useful for determining the relationship
between level of exposure and induction of microlesions.
We define mLcrolesions as point mutations, small deletions, other sorts of
genetic damage in the mammalian system which are not identifiable by currently
utilized microscopic techniques. Currently the major effort is directed toward
determining the feasibility of utilizing the techniques employed in the study of inborn
errors in metabolism as a method for detecting induced microlesions.
Table 1 is an outline of the approach we are using. Three different technical
approaches are being used to detect altered gene products. The first one monitors
normal enzyme activity, as well as preliminary kinetic analysis, and heat stability
measurements are made on various enzymes from liver, brain, kidney, heart, and
red blood cells.
Table 1: Detection of Metabolic Events by Assessment of
Metabolic Characteristics
I. Enzyme Activity
a) Normal activity
b) Heat stability
c) Kinetic parameters
II. Electrophoretic Assays
a) Isoelectric focusing
b) Isoenzymes
c) Structural proteins
HI. Metabolite Assay
a) Serum and urine chemistry assay for compounds which are
associated with 70 human metabolic errors
185
-------
The second component is electrophoretic studies of enzyme and nonenzyme proteins
These are conducted to look for structural alterations which are not detectable by
changes in activity parameters. In the third, the metabolic status of plasma and
urine is monitored as an indication of the various storage diseases, metabolic
blocks, and other pathway alterations.
At this point we have determined the appropriate assay conditions for apporxi-
mately 40 liver and kidney enzymes. The majority of these enzyme analyses are
performed on a miniature centrifugal fast analyzer (see Clinical Chemistry 20:1003,
1974 for discussion), which was built for us by Oak Ridge National Labs.
The repeatability of this system when using one microliter of a liver supernatant
is such that we have a coefficient variation of about 1.4 to 1.5 percent. If we go to
2 microliters of sample, our coefficient of variation is down to about 0.4 or 0.5
percent. We have one reference cuvette and 16 sample cuvettes, so that in one run
we're able to make 16 enzyme assays. Our turn-around time with some of the liver
enzymes, particularly those like LDH which are fairly high in activity, is about 4
minutes when we're working at 30 degrees centigrade. The total assay volume for
our work is 110 to 140 microliters. With liver we normally work with 1-5 micro-
liters of a one plus six liver homogenate. With tissues like brain, kidney, and heart,
we work with a one plus seven or one plus eight and normally work with 3 or 4-
microliters this time.
This instrument can be utilized for most enzyme assays which can be coupled
to either a product or a substrate within an absorption peak between 290 and 880
millimicron. Approximately 5,000 assays per week can be completed with the
current instrumentation.
At this point we have about 40 assays that we are currently able to do. We are
also working on new enzyme assays, and we would like to be capable of assaying
in the neighborhood of 60 to 70 different enzymes. With the assays being done on
various tissues, when we get fully operational, each mouse should be assayed for
approximately 115 to 130 different enzymes by direct enzymatic analysis.
186
-------
Table 1: Detection of Metabolic Events by Assessment
of Metabolic Characteristics
L Enzyme Activity
a) Normal activity
b) Heat stability
c) Kinetic parameters
II. Electrophoretic Assays
a) Isoelectric focusing
b) Isoenzymes
c) Structural proteins
HI. Metabolite Assay
Serum and urine chemistry assay for compounds which are associated
with 70 human metabolic errors
187
-------
As indicated in Table 1, we are also looking at blood and urine parameters. We
now have normal values on approximately 50 different plasma and urine constituents.
These components are associated with 70 known human metabolic diseases. We're
also currently in the process of developing our electrophoretic studies. It is
estimated that with the techniques currently in hand, this approach of enzyme acti-
vity measurements, electrophoretic mobility, and metabolite assays will allow us
to directly or indirectly examine the gene products for 200 to 300 loci.
Our initial experiment is outlined in Table 2. We are using radiation at this time.
Obviously, when you're attempting to develop a new method and attempting to stan-
dardize your assay conditions, you're best off trying to relate it to the normal
established data which are currently in the literature.
Table 2: Treatment and Mating Scheme for Irradiation Experiment
a) Frt Otreated I x FA O untreated
0 4, 0 + Q
b) FI (y x F. + heterozygotes
c) F also heterozygotes
Lt
d) Tissue taken from F0; urine collected from F for metabolic assays
2 1
We are using C57BL/6 mice, and matlngs were set up for examination of mutations
induced in mature sperm as well as stem cell mutations. Two groups of animals (F_)
were irradiated with either 300 or 600 rads. These treated males were then mated
to untreated females with production of an F population. This F population would
be heterozygous at any loci where mutations occur.
Obviously, we are concerned about heritable events. We mate the FJJS which will
give us an F0 population, and remember, each of these F s is a unique individual,
2 1
so that even though we do go to an F, generation, this F» generation will be hetero-
zygous at any loci or most any loci we want to look at.
Remember also that each F2 individual could be heterozygous at several different
loci. Based upon data derived from studying inborn errors of metabolism in humans
in excess of 95 percent of a sample of about 200 inborn errors of metabolism where
188
-------
we know the enzyme defects, it is possible to detect enzymatic differences between
heterozygotes and either homozygous recessive or homozygous normal individuals.
At this point we have collected urine from approximately 90 FI individuals in
a very preliminary sort of screening; using basically yes and no answers, we have
several which look interesting. We are currently completing the work on quanti-
tative analysis of the urine.
We have looked at five different metabolic errors in mice obtained from Jackson
Labs to prove that we can differentiate by determination of enzyme activity within
small samples between homozygous normal, homozygous abnormal, and hetero-
zygotes. We've also worked out or are in the process of working out our background
data base, using the C57 black mice. Although we are using irradiation, we will in
the near future, if and when we get the techniques worked out, make a selection of
appropriate chemicals. Obviously, these could be pesticides.
Although most of our activity is directed toward Jn, vivo microleslons, as soon
as we have the various background features worked out, we will use this system as
a means of detecting differences based on both structural and functional relationships.
We'll make a direct effort toward quantitating the dose which gets to the germ cells
in these C57 black mice.
189
-------
FEASIBILITY STUDY: THE USE OF DIPLOID CELL STRAINS
FROM SPECIFIED MOUSE GENOTYPES FOR USE IN DEVELOPING
IN VITRO ASSAY FOR MUTAGENIC ACTIVITY, INDUCTION, AND
ANALYSIS OF GENE MUTATIONS IN MAMMALIAN CELL LINES
Dr. Daniel A. Casciano*
As Dr. Mohrenweiser has just described, we are attempting to define an inte-
grated approach to mutagenesis. The objective of the second major program in our.
integrated approach is to develop new and reliable, rapid and economic methods for
estimating the potential mutagenicity of environmental agents utilizing in vitro
somatic cell culture bioassays.
Two approaches are currently being used. We're using specific established cell
systems to quantitate the induction of specific gene mutations as a. method for screening
environmental agents for their potential mutagdnic activity, and we are employing
recently derived, "functional" mammalian cells to detect and quantitate specific gene
mutations. Each of these approaches that I'm going to describe is currently in a
developmental stage.
While no available established cell lines appear to be ideally suited for genetic
studies of mammalian cell culture systems, several existing lines have proven usable
for the isolation of different classes of mutants. The choice of cell line is primarily
by requirement of karyotype stability and high plating efficiency. In addition, since
mutational events occur at low frequency, any variant or genetic marker must
effectively respond to artificial selection to be useful in mutational studies.
The Chinese hamster ovary (CHO) cell line is presently being used as a model
system in our laboratory and in others. This cell line has a stable karyotype, short
generation time, grows in suspension as well as a monolayer culture, and possesses
a high plating efficiency. In addition, CHO cells have already been successfully
used in isolating'drug-resistant and auxotrophic mutants. The genetic marker to be
selected for is resistance to 6-thioguanine. This selective system has the advantage
•"National Center for Toxicological Research, Jefferson, Arkansas
191
Preceding page blank
-------
that the gene conveying this drug resistance is X-linked, so that only one gene copy
is present per cell. Utility of this selective system allows one to determine whether
the variant isolated is a phenotypic or genotypic variant by assaying the gene product,
hypoxanthine guanine phosphoribosyl transferase (HGPRT).
In somatic cell culture it is at present possible to obtain dose-response relation-
ships for mutants which are fairly reproducible. However, we are still a long way
from determining exact mutation frequencies, which are important for estimation of
genetic risk, and from obtaining exact dose-responsiveness, which are necessary
for fundamental research on the process of mutation induction.
Recently a sensitive and rapid system utilizing a specific gene mutation affecting
the HGPRT locus in CHO cells has been developed at the Oak Ridge National Labora-
tories. This system, which measures 6-thioguanine resistance, has been employed
to quantitate various physical and chemical mutagens and carcinogens. Mutation can
be quantitated at mutagen doses, which produces no appreciable cell death, and linear
dose relationship has been obtained. We are presently modifying this system to test
the dose-responsiveness of a variety of known and potential chemical mutagens which
include direct alkylating agents, suspected metabolites of promutagens, and promu-
tagens transformed in appropriate activation systems. Studies are also in progress
to quantitate mutation at an autosomal locus, which codes for adenine phosphoribosyl
transferase. These systems should provide a means to quantitate specific locus
mutations, either at X- or autosomal-linked genes.
In addition to utilizing the CHO cell line as a model system, established cell cul-
tures from a variety of mammalian species, including man, will be adapted for muta-
genesis testing. Success in this endeavor would then be followed by attempts to
correlate the dose-response in various mammalian cell cultures to specific mutagens
with dose-response in human cell cultures. This approach may aid in overcoming
the problem of extrapolation to man of results obtained with animal systems.
The second approach that we are using in the somatic system involves use of
•recently derived "functional'' epithelial mammalian cells to detect and quantitate
specific gene mutations. We recently have been attempting to derive "functional"
192
-------
liver, kidney, and lung cell cultures originating from the mouse and other mammalian
species. At present we are emphasizing isolation of "functional" liver cells. We've
used a variety of tissue preparation methods, and we have had success using the
Bauscher and Schaeffer technique published in the spring of 1974 in In Vitro. Using
3-day and 9-day-old mouse liver as starting material, we have been able to obtain
a proliferating low passage epithelial-like culture, retaining morphology of the cells
originally plated in sufficient quantity for liquid nitrogen storage.
We are attempting to characterize these cells as to their true origin. The criteria
for functionality we are applying are both morphological and biochemical. The morpho-
logical criteria include those shown in Table 1. We are using both operational defi-
nitions utilized in cell culture methodology and histological definitions. Table 2
depicts some of the biochemical characters we are using to identify the hepatic cell.
Table 1: "Functional" Mouse Liver Cells
Criteria for Functionality
A. Morphology and Growth Characteristics:
1. Liver parenchyma! morphology
2. Limited life span
3. Contact inhibition of growth
4. Unable to grow in agar, serum-free medium or in suspension
5. Do not form tumors when Injected into host
Table 2; "Functional" Mouse Liver Cells
B. Biochemical Characteristics:
1. Retention of mixed function oxidases: e.g., aryl hydrocarbon
hydroxylase
2. Glycogen synthesis and storage
3. Tryptophan to nicotinamide adenine
Dinucleotide biosynthetic enzymes
4. Phenyl alanine hydroxylase activity
5. Hormone receptor sites
193
-------
The cultures we have isolated from 3-day-old mice are now in the 30th passage,
which represents around 60 generations, and are retaining epithelial-tike morphology.
At present we can identify at least two epithelial cell types, one of which is the
so-called clear cell defined by Grisham, whose liver function is unknown. This cell
type is presently being utilized in several laboratories for carcinogenesis bioassay.
The other epithelial cell that is present in the culture is tentatively being identified
as a liver hepatocyte based on its morphological characteristics. We are attempting
to clone these two cell types and have recently enriched for each morphology.
We have done some preliminary characterization of the mixed culture, and the
results are as follows: The cultures retain epithelial characteristics until the 30th
passage. They also display contact inhibition of growth. After the 25th passage,
which represents around 50 to 55 generations, the culture is apparently senescing
and displaying a limited 1 ifespan. Glycogen is being synthesized and stored up to
the fifth passage; this was identified phytochemically using the para-aminosalicyllc (PAS)
stain; and constitutive aryl hydrocarbon hydroxylase activity is being retained to at
least the 18th passage. In addition, of those hepatic enzymes we have analyzed, the
cells are displaying an enzyme profile characteristic of the 3-day-old mouse. We
are presently deriving cultures from 9-day-old mice whose enzyme profiles are more
nearly like that of the adult, and we are having relative success with this.
The functions we are most interested in are the mixed function oxidase systems,
which will allow us to use these cultures to analyze mutagenicity of known and potential
chemical mutagens requiring metabolic activation. This may allow us to overcome
the limitations of presently available established somatic cell systems used in muta-
genesis testing.
In addition, we are anticipating tying in with the carcinogenesis programs at the
National Center for Toxicological Research, using accepted somatic cell carcino-
genesis endpoints. Additionally, we are developing a program to study DNA repair
synthesis in these cells when exposed acutely and chronically to known mutagens and
carcinogens, and we hope to correlate the effect of various doses of mutagens with
DNA repair and mutation frequency. Hopefully, once these systems are developed,
we.-can.apply the epithelial cells to toxicological evaluations of a variety of chemicals.
194
-------
EPIDEMIOLOGY OF PESTICIDES:
CANCER MORTALITY AND PESTICIDE USAGE
IN THE UNITED STATES
Dr. William F. Durham*
This is going to be fairly brief because it is just an outline of a contract
which we have quite recently — as a matter of fact, June 27 — negotiated with
the School of Public Health at the University of North Carolina at Chapel Hill.
I hope that next year Dr. Carl Shy, who is the principal investigator for this
contract, will be able to be with us and discuss the actual results.
This contract is designed to investigate the possibility of correlation
between cancer mortality and pesticide usage in the United States. Basically
what the contract is set up to do is conduct an epidemiologic study to determine
whether or not a geographic association can be demonstrated between United
States counties of high pesticide usage and counties in which excess cancer
mortalities have occurred after appropriate adjustments are made for other
known epidemiologic determinants of cancer mortality, such as age composi-
tion of the group, sex, race, socioeconomic level, and rural-urban land use
patterns.
More specifically, the study will investigate whether counties with high
usage of DDT in the 1945-1960 period show any excess in mortality from
cancers of the liver, gastrointestinal system, or hematopoietic system or
total cancers, minus lung cancer. One of the specific things that Dr. Shy
will be doing is to tabulate the cancer mortality for all causes and for specific
relevant causes, adjusted by age, for each county in selected agricultural
regions of the United States, including as a minimum the Southeast and South-
west, where high pesticide usage has been prevalent.
* Environmental Toxicology Division, National Environmental Research Labora-
tory, Research Triangle Park, North Carolina
195
-------
Then to correlate with that, Dr. Shy will obtain the pesticide usage by counties
for this same 1945-1960 period, and what he will use for this will be data available
from the United States Census of Agriculture for the specific counties where these
data are available. ft may very well be that to get certain portions of this informa-
tion, it will be necessary to actually go to the records of the county agricultural
extension agent and find out what was being recommended during these particular
years for pesticide usage on the major crops which are grown in these counties.
Initial analysis will focus on the association of specific classes of pesticide
usage — herbicides, Insecticides, and fungicides — with total cancer, cancer of
the respiratory system, cancer of the bladder, cancer of the biliary passages and
liver, and cancers of the hematopoietic system. Multiple regression analysis,
cluster analysis, and factorial analyses will be utilized in addition to simple cate-
gorical analyses of discrete variables.
Considering the possibility of the 5-, 10-, 15-, or 28-year time lags between
first exposure and death from cancer related to that exposure, the cancer mor-
tality for each sex and race group categorized according to socioeconomic and
urban-rural composition as above will be compared in counties with high DDT
usage versus counties with low DDT usage.
As I said, this contract was just set up on the 27th of June, so this is just an
outline of what is planned.
196
-------
EFFECTS OF CACODYLIC ACID ON THE PRENATAL DEVELOPMENT
OF RATS AND MICE*
Dr. Neil Chernoff**
Cacodylic acid (dimethylarsinic acid) is an arsenic-containing pesticide. It is used
for post-emergence weed control, crown kill of trees, and defoliation of cotton. It is
also used for the control of bark beetles.
To date, there have been no published data on the possible effects of cacodylic acid
on reproduction or teratogenesis. It is known that intravenous injections of sodium
arsenate in hamsters (1.2), intraperitoneal injections of sodium arsenate in rats (3),
and intraperitoneal injections of sodium arsenate (4), or sodium arsenite (5) in mice
produce a variety of fetal malformations when these compounds are administered
during the period of organogenesis. The dietary administration of 100 ppm of arsenic
trioxide was shown to have no effects on reproduction or neonatal growth rate in
rats (6). A case of possible arsenic-induced neonatal death in a human has also been
reported (7).
The experiments reported in this paper were carried out to determine the possible
fetal effects in mice and rats of an organic form of arsenic in the environment.
Materials and Methods
Random-bred albino mice (CD-I stock) and rats (CD stock) were used (8). Day 1
of pregnancy was recorded upon demonstration of sperm in the vaginal smear. Ani-
mals were in rooms maintained at constant temperature (22-26°C) with controlled
lighting (12 h. light) and were fed commercial lab chow and water ad libitum.
Cacodylic acid (9) was administered by gastric intubation during the period of
organogenesis on days 7-16 of gestation. Maternal weight on day 6 was used for
calculation of doses. The compound was administered as a solution in distilled water
#Co-author is Ellen H. Rogers
^^Experimental Biology Division, Neurobiology Branch, EPA
197
-------
with intubation volumes of 0.1 ml in mice and 0.2 ml in rats. The doses admin-
istered were 200,400, and 600 mg/kg/day in the mouse, and 40, 50, 60, 75,
and 100 mg/kg/day in the rat. Distilled water controls were run concurrently.
Mice and rats were killed on days 18 and 21 of gestation, respectively. The
uteruses were removed and weighed. Maternal weight changes during gestation
were calculated as the weight of the intact animal at the time of sacrifice less
the weight of the removed gravid uterus, from which the day 6 weight was sub-
tracted. Live fetuses were weighed and the litter divided into either Bouin's
solution or 65 percent ethyl alcohol. All fetuses were examined for external
defects. After 3 to 5 days, necropsy was performed on those fetuses fixed in
Bouin's solution. Fetuses in ethyl alcohol were subsequently cleared in 1 per-
cent KOH and stained with alizarin red S to facilitate skeletal examination.
Calculations were based on the litter as the experimental unit, and standard
deviation therefore represent the variation of litter means. Comparisons between
treatment and control groups were made using the Mann-Whitney "U" test (10).
Dose-response relationships were tested with Jonckheere's test (10).
Results
The data from the rat studies are summarized in Table 1. The two highest
doses, 100 and 75 mg/kg/day, resulted in maternal mortality of 80 and 33 percent,
respectively. There was little or no maternal mortality at the other dose levels.
The only significant alterations in maternal weight occurred at the 100 and 75 mg/
kg/day doses. No changes in liver/body weight ratios were noted at any dose
levels.
The administration of cacodylic acid resulted in significant (P<0.001) dose-
related reductions in fetal weight, sternal ossification centers, and number of
caudal vertebral ossification centers. Fetuses weighed an average of 3.41 g at
the 75-mg dose as compared to 4.10 g in control litters. Sternal and caudal
198
-------
ossification centers were reduced from averages of 5.6 and 4.6, respectively,
in control Litters, to 4.1 and 3.8 in litters within the 75-mg dose treatment
group. There was increased fetal mortality at all dose levels greater than 40
mg/kg/day.
Micrognathia (small jaw) was noted in litters of treated animals at all dose
levels. The micrognathia ranged from slight to severe shortening of the mandible.
Isolated examples of small lungs whose lobes did not extend ventral to the heart
as in controls were found in the 50, 60, and 75 mg/kg/day doses. A common
alteration in treated fetuses was aberrant rugae on the palates. In control fetuses
the large anterior rugae met in opposition at the midline ridge of the palate. In
many of the treated litters at all dose levels the pattern of these large anterior
rugae was not symmetrical and the individual pairs of rugae did not lie in apposi-
tion.
The data on the mouse studies are summarized in Table 2. Significant dose-
related reductions in maternal weight gain (P<0.01) and liver/body weight ratios
(P<0.001) were noted during the course of the study. Maternal mortality (7 per-
cent) occurred in animals receiving the high dose of 600 mg/kg/day.
Significant dose-related reductions in fetal weight (P
-------
Discussion and Summary
The administration of cacodylic acid to CD rats during the period of organogenesis
resulted in fetal toxicity, as Indicated by dose-related reductions in weight and In
the degree of ossification of both the sternum and caudal vertebrae. Micrognathia
was noted at all dose levels. A number of grossly small lungs were evident at the
50-, 60-, and 75-mg dose levels. This anomaly is difficult to identify strictly on
a morphological basis, principally because many of the fetuses themselves mani-
fested significant reductions in size. Experiments utilizing a quantifiable index such
as a lung/body weight ratio are currently ongoing. The significance of the aberrant
palatal rugae is not known, ft should be noted that no cases of cleft palate were
found during the course of the rat studies, even in litters with one or more fetuses
exhibiting aberrant rugae.
In the mouse, the administration of cacodylic acid at doses of 200-600 mg/kg/day
during the period of organogenesis resulted in a spectrum of fetal toxicity similar
to that found in the rat, but its teratogenic response differed markedly. Dose-related
reductions in the average fetal weight and degree of ossification were noted. Unlike
the rat, however, the only anomaly which was seen in the mouse was cleft palate.
This defect occurred only after the administration of 400 and 600 mg/kg/day dose
levels which also resulted In some degree of maternal toxicity, as evidenced by a
reduction in maternal weight gain during pregnancy.
Observations on the effects of perinatal cacodylic acid administration are continu-
ing, as are studies on the possible postnatal sequelae to prenatal exposure and
studies involving both technical grade and the pure compound.
References
1. Perm, V. H. and Carpenter, S. J.: Malformations induced by sodium arsenate.
J. Reprod. Fert. 17: 199-201 (1968)
2. Perm, V. H., Saxon, A., and Smith, B. M.: The teratogenic profile of sodium
arsenate in the golden hamster. Arch. Environ. Health 22; 557-560 (1971)
200
-------
3. Beaudoin, A. R.: Teratogenicity of sodium arsenate in rats. Teratology
10: 153-158 (1974)
4. Hood, R. D., and Bishop, S. L.: Teratogenic effects of sodium arsenate in
mice. Arch. Environ. Health 24; 62-65 (1972)
5. Hood, R. D.: Effects of sodium arsenite of fetal development. Bull. Environ.
Cent. Toxicol. 7: 216-222 (1972)
6. Kojima, H.: Studies on development pharmacology of arsenite. 2. Effects of
arsenite on pregnancy, nutrition and hard tissue. Folia Pharmacol. Japon.
70: 136-149 (1974) (Abstract)
7. Lugo, G., Cassady, G., and Falmisano, R.: Acute maternal arsenate
intoxication with neonatal death. A. J. Pis. Child. 117; 328-330 (1969)
8. Animals were obtained from the Charles River Breeding Laboratories,
Wilmington, Massachusetts.
9. The cacodylic acid which was used in these experiments was purchased from
ICN Pharmaceuticals. Analysis by the Ansul Chemical Co. indicates that the
sample had approximately 5% NaCL, 1.7% methane arsenic acid, 0.2%
arsenoris ions, 0.7% arsenic ions, and 2.0% unidentified organic compounds.
10. Hollander, M., and Wolfe, D. A.: "Nonparametric Statistical Methods." New
York, John Wiley and Sons, 1973, pp. 71, 120.
201
-------
Table 1: Prenatal Effects of Cacodylic Acid in the Rat
Dose Group (mgAg/day)
Data
Conception
No. inseminated
No. died
% mortality
No. pregnant (term)
Maternal observations
Av weight gain (g)
Av liver/body weight
Fetal observations
Av implants
Av mortality (%)
Av weight (g)
Av no. sternal ossification centers
Av no. caudal ossification centers
o
Small lungs
Q
Micrognathia
Q
Aberrant rugae
Control
28
0
0
23
68.4+20.1
4.8+ 0.4
10. 0+ 3. 0
4
4.10+ 0.29
5.6+ 0.5
4.6+ 0.4
0
0
1(4)
40
15
0
0
11
78.4+17.9
5. 1+ 0. 3
10.2+ 3.2
4
3.79+ 0. 31d
5.5+ 0.5
4.2+ 0.9°
0
2(2)b
10(31)°
50
32
1
3
26
63. 1+26. 7
4. 8+ 0. 3
9. 7+ 2. 1
21C
3.50+ 0.486
5.1+ 1.1°
3. 9+ 0. 9d
3(5)
7(12)b
8(28)6'f
60
11
0
0
9
77. 8+19. 3
5.0+ 0.3
10.0+ 2.5
31 d
3.28+ 0.61e
4.3+ 2.0C
3.4+ 1.5d
2(2)
3(7)C
8(26) 6
75
18
6e
33
8
56. 0+15. 8C
4.9+ 0.5
9.8+ 1.5
18C
3.41+ 0.526
4. 1+ 1. 8C
3.8+ 1.0b
1(1)
4(15)d
-
100
21
I7e
80
2
-9.8C
5.2
8.5
82d
2.406
2.5e
i.oc
0
1(1)
-
a Data given as no. of litters (no. of fetuses).
b Comparing these groups with controls produces P<0.10.
c Statistically different from controls, P<0.05.
d Statistically different from controls, P<0.01.
e Statistically different from controls, P<0. 001.
f Only 9 litters examined for this anomaly.
-------
Table 2: Prenatal Effects of Cacodylic Acid in the Mouse
Data
Conception
No. inseminated
No. died
% mortality
No. pregnant (term)
g Maternal observations
M Av weight gain (g)
Av liver/body weight
Fetal observations
Av implants
Av mortality (%)
Av weight (g)
Av no. sternal ossification centers
Av no. caudal ossification centers
Cleft palate, no. of litters (no. of fetuses)
Control
53
0
0
38
5.0+2.5
7.3 + 0.6
12.0 + 2.2
7
1. 12 + 0. 19
5.7 + 0.6
5.0 + 1.9
0(0)
Dose Group (mg/kg/day)
200 400 600
46 56 55
1 14
2 27
27 38 35
5.2+1.8 3.8+2.4E 3.5+2.
7.4+0.5 6.8+0.7b 6.9+0.
12.Q+2.3 11.4+2.9 11.9+2.
13 12 15
1.06+0.19 0.92+0.20° 0.93+0.
5.6+0.6 4.7 + 1.7b 5. 1 + 1.
4.4 + 2.0 2.7 + 1.8b 3.2 + 2.
0(0) 3(4) 9(18)e
3b
7b
5
19C
4b
oa
a Statistically different from controls, P<.0.05.
b Statistically different from controls, P*:0. 01.
c Statistically different from controls, P<0. 001.
-------
DR. KENAGA: Has this compound been studied in any long-term feeding studies,
like reproduction studies or any other long-term studies before?
DR. CHERNOFF: Not to my knowledge.
DR. KENAGA: Is this the first time that the teratogenicity has been known for
this compound?
DR. CHERNOFF: To my knowledge, yes, this is the first teratogenic study of
this nature using this sort of protocol. I don't know if three-generation studies have
been done and oerhaps not published.
MR. BURNAM: I don't think we have any previous teratology studies on cacodylic
acid or reproduction studies either. It was tested in the bionetics cancer study. It
came out negative there. But I don't think there are any other long-term studies-
with it.
204
-------
ASSESSMENT OF SUBTLE AND DELAYED EFFECTS
OF SUBSTITUTE CHEMICALS
Dr. Daniel A. Spyker*
I will first present some brief background on how we came to propose this program
to the EPA, describe our rationale, and then describe the methods by which we propose
to integrate the conventional teratological approach with maturational, behavioral,
endocrinological, and immunological evaluation of experimental animals over their
entire lifespans. I will then describe what we hope to have in the way of data to present
at next year's conference.
Background
In 1967, we began studying a widely used and very effective fungicide, methylmercury.
In a teratological assessment of this compound we documented a variety of effects, all
of which were dependent on dose, stage of development when exposed, and strain of
animal. None of the mothers of affected offspring showed any signs of toxicity. These
responses for the fungicide, methylmercury, were present at a dosage which caused
no gross detectable abnormalities in the mothers. Further, in doses which were not
teratogenic and in animals which during their early development showed no signs of
gross abnormalities, we found their behavior to be abnormal in the open field. In the
evaluation of 20 treated and control animals, we found differences in 4 of the 8 para-
meters measured, including center latency, number of defecations, number of urina-
tions, and in backing behavior. As these animals grew older, such gross neuro-
muscular deficits became evident as ataxia, loss of righting ability, and what we might
loosely refer to as premature aging. We also found conjunctivitis to develop more
frequently in treated animals. Cultures were positive for pneumococcal bacteria, and
we conducted a preliminary screen on immunologic status. An evaluation of T-cell
function using sheep red blood cells showed no differences between treated and control
animals, but Brucella challenges of the B-cell system, although not different in the young
animals, did show significant differences in response in the mature animals. These
and other data have caused us to feel that the organism is significantly more vulnerable
*Department of Medicine, University of Virginia Medical School, Charlottesville,
Virginia
205
-------
during development than it is at any other time. It seems likely that certain specific
systems (for example, the immunologies! and, certainly, the CNS) maybe at greatest
risk.
Program Objectives
We propose an integrated study to detect deficits that are not readily apparent
and/or are initially latent but are expressed later in life. In addition to testing for
deficits that may be subtle and delayed, we will further enhance the sensitivity of our
evaluation by evaluating the organism at risk, i.e., when exposed during development,
and within that organism evaluating the systems most susceptible to insult, i.e.,
the CNS, endocrine, immune systems, and looking at these systems over the animal's
entire lifespan.
We will use an inbred strain of mice for this study, 129/SvSl. Pesticides tested
include chlordane, heptachlor, parathion, carbofuran, methoxychlor, and Diazinon.
All six compounds will be administered orally and, in addition, the animals will be
exposed to chlordane and parathion by inhalation. The oral route will be used to simu-
late the human situation. Dosage will be based on levels monitored in the environment,
and duration of exposure will be based on knowledge of the compounds' kinetics such
that the fetal environment approximates the steady state situation for the dosages
chosen. For example, the chlorinated hydrocarbons are strongly liquid bound,
bioaccumulate, and require longer exposure than the two organic phosphorus compounds,
We will initially determine an acute LD for non-pregnant adults to compare the
' 50
toxic data for our animals with that reported for each compound in other test animals.
Our approach is summarized in Figure 1 which is a schematic flow of animals through
the lifespan evaluation. This will be replicated for each test compound and twice for
control groups. The numbers represent the number of test animals in each of the 4
tracks, e.g., 36 animals will be evaluated on 3 separate occasions in behavioral
batteries and allowed to live out their normal lifespans. Litter mates will be assigned
to these four tracks as far as possible to minimize the intratrack variation.
Teratologic Examination
Although we expect to be using a dosage which will not cause overt defects, we can
efficaciously collect teratologic data on these test animals. These data will include
206
-------
Figure 1
Schematic Flow of Animals Through the Three-Year Evaluation
This basic design is replicated for each of the eight (I.e., two control and six
pesticide) experimental groups. PHASE ONE is designed to generate enough
tested animals to permit subsequent evaluation over their Ufespan and correlation
of results with early measures.
Litter Mates
End PHASE ONE —
End PHASE TWO— —
Teratology
Evaluation
Maturatlonal
Screen
End PHASE THREE jt
^T
Normal Life Span
Endocrine
Immune
Morphology
Endocrine
Trrn^npft
Morphology
Tumor Screen ]
Endocrine
Immune
Morphology
207
-------
physical characteristics, fetal or neonatal lethality, congenital malformation, growth
retardation, and overt functional deficits.
Maturational Screen
We will attempt to assess maternal-neonatal interaction by assessing maternal
retrieving response, nursing, and cannibalization. Offsprings' physical development
will be monitored by recording the age at eye opening, development of a righting reflex,
and development of forelimb placing. Testes descent and vaginal opening will also
be monitored for correlation with endocrinological evaluation. Birth weight and weaning
weight or weight gain is an important index of general health. Sensory development
will be assessed by observing each offspring's performance on visual cliff, auditory
orienting, and olfactory aversion tasks. Strength and coordination will be tested
using rotarod, plank walk, bar holding, and rod climb.
Behavioral Battery
Three major tasks have been chosen for this evaluation. These individual tests or
tasks are designed to stress an array of different functional systems so as to maximize
the likelihood of detecting behavioral deficits if they are indeed present. Differences,
if uncovered, will be subsequently evaluated in more discriminative testing to isolate
the specific system or process involved. The animal's behavior in a novel environ-
ment is a test which assesses emotionality, activity, locomotion, and exploration.
This is done by placing the mouse on a brightly lit, large, open surface and observing
its activity. This technique is commonly known as open field testing.
Learning, memory, and cognition will be tested in a modified Lashley III maze.
We will use a four baffle maze which control animals typically learn to run flawlessly
in approximately 75 trials. After acquisition, the difficulty of this task may be
altered by simply imposing a time limit on execution of the maze. Time discrimination
may be assessed by only reinforcing certain running times longer than the animal's
previous scores. For example, if the animal can complete the maze in 30 seconds,
his reward can be withheld unless he now executes in between 40 and 50 seconds.
A stressful neuromuscular task will be imposed on these animals in the form of
swimming over a specified time in ice water. This test provides assessment of the
208
-------
animal's coordination, equilibrium (vesttbular system), and response to stress. This
task will be recorded on videotape for objective evaluation. Immediately upon com-
pletion of the task, blood will be drawn for an immunoassay of plasma-corticosterone.
Endocrinological Evaluation
No single measurement is available to provide assessment of the entire pituitary-
adrenal axis; therefore, three tests will be employed in this initial screen to identify
sites of interest for further study. One, plasma-corticosterone concentrations will be
determined by radioimmunoassay, both at rest and after application of the standardized
swimming stress. These differences will serve as a crude measure of overall respon-
siveness of the axis. Two, whole adrenal glands will be incubated with added ACTH
under standardized in^vitro conditions, and steroid production will be assessed as an
indicator of adrenal secretory capacity. Three, liver tissue will be homogenized and
incubated in vitro with added corticosterone. A-ring and side-chain metabolism can
thus be assessed, providing not only a measure of hepatic function per se, but also
an estimate of the biological clearance rate of corticosterone.
Gonadal function testing will consist of determination of the age of vaginal opening
or testes descent, as was previously outlined. Female cyclicity will be checked at
appropriate intervals by vaginal cytology. Gonadal weights will be obtained and tissues
fixed for subsequent immunologic evaluation. Plasma samples will be stored for
subsequent measurement of testosterone or estradiol.
Immunologic Assessment
We propose to evaluate both the general and specific immune status of the mice
exposed to each pesticide. The general immune status will be assessed through
measurement of the circulating immunoglobulins IgG, IgG2, IgA, and IgM. This will be
determined by radial immunodiffusion or specific precipitant techniques, as appro-
priate. Single cell suspension from each spleen will be evaluated to determine numbers
of lymphocytes and/or macrophages. Lymphocyte subpopulations of those carrying
surface immunoglobulins, i.e., B-cells, and those possessing receptors for erythro-
cytes, i.e., T-cells, will also be determined.
Specific immune status assessment will initially focus on the previously described
decrease in antibody response to the Brucella abortus. Challenge via the flagelin
209
-------
molecule in polarized form permits assessment of T-independent response, whereas
the monomer challenge will permit assessment of T-dependent response. Additionally,
we will study serum protein antigens (using sheep red blood cells and bacterial poly-
saccharides). We will evaluate the kinetics of antibody response using the plaque-
forming cell technique, and the cellular immune response by the graft vs. host reaction
using a spleen weight bioassay technique.
Morphologic Evaluation
If CNS dysfunction is detected among the animals to be sacrificed during the first
phase, three brains will be examined for neuropathology under the light microscope.
If no CNS deficits are detected, the brains will be fixed and stored for subsequent
examination, as indicated. Likewise, if altered gonadal indices are detected, then
representative ovaries and testes will be examined for histopathology. Figure 2
summarizes the integrated Phase I effort and evaluation of each pesticide.
Data Acquisition and Analysis
As previously described, data collection will begin at birth but animals are not
uniquely identifiable until one week of age when toe tattoo is possible without altering
maternal-off spring interaction. At that time each animal will be randomly assigned
a unique mouse number which will thereafter be its sole identification. A cross-
reference which will include its treatment history will be maintained, but will be used
only in assigning animals for individual tests or in analysis of results. The principal
instrument of data collection will be individually prepared, encounter form type data
sheets. These forms represent the only necessary hard-copy record in the data
collection process.
For each encounter form a special-purpose interactive data entry program for
the Biotronics computer system will be used. This program requests the operator
to enter each piece of datum in a format which matches the encounter form, checks
each entry at the point for reasonableness, and writes the data onto a magnetic disk
file after all entries have been approved by the operator. This data file will be sub-
sequently manually checked against the encounter form and the file will be rewritten
and regrouped by treatment and sex according to the master key. The resulting
file thus becomes one of the master data files. This procedure is summarized in
210
-------
Figure 2
Integrated Team Effort
The six-way integrated assessment of animals exposed during development is
diagrammed below. Broken lines represent planned and potential interaction
among the several discipline specialities and animal groups involved in this study.
1
Teratology
Examination
Maturatlonal
Screen
Behavioral
Battery
loa Water Swim
Fiaama
Cortlcoaterold
Endoorlnologle
Evaluation
Weigh
Organs
Pathologic
Examination
Immunologlo
Aaaessmeat
I
Long-term
Survivors
211
-------
Figure 3. We may hypothesize that any one of the measured parameters Is unaffected
by the treatment, the "null hypothesis." Replication of the t-test for each parameter
is not legitimate, but, of equal importance, it neglects information inherent in the
simultaneous nature of the observations. In the age-old conflict between the means
and the variances, we propose to tip the balance in favor of the means by use of one
or more multivariable statistical techniques. For each set of data, for example the
open field observations, we will apply multivaried discriminate analysis. This will
provide not only a single measure of the discriminating power of the test, but will
also provide a rank ordering of the contributions of each individual measure to this
discrimination.
Product
As a result of this contract, we hope to provide:
1) Data base, consisting of integrated data on teratology, maturation, behavior,
endocrinology, immunology, and pathology for each compound tested.
2) Conclusions, regarding the quantitative effect of each compound on the developing
organisms over the total lifespan.
3) In addition, this approach will evolve a particular methodology or protocol for
each compound tested. That is, it will pinpoint the particular system or target organ
for each compound and permit subsequent efficacious testing of similar compounds.
It may similarly suggest testing procedures for humans exposed to these or similar
compounds.
212
-------
Figure 3
Data Acquisition and Analysis
Schematic of data flow, to be replicated for each test or assessment, I. e., for
each task.
Special Purpose
Encounter Forma
Random
Allocation
Program
Data Base for
• Longitudinal Analysis
• Overall Evaluation
Modification of
Integrated Study
•F-ratlo, p-vulue
•Individual Parameter Analysis
- Task
x—""N — Computer
Program
Permanent
' Data File
213
-------
QUESTION: Dr. Spyker, I was interested In one technical aspect. Maybe
your wife could answer this better, but in the swim test, which is the test for
performance under stress and the like, you emphasized the aspect of ice cold
water instead of just swimming itself. Have you ever run any comparative tests
putting an animal in, you might say, ambient water versus ice water? Is there
a difference ? Do you get greater stimulation, or what's the reason for using ice
water? It's a detail, but I think it's rather interesting.
i
DR. SPYKER: I agree. We have done both ambient water and ice water swims.
The data reported in Science were obtained in ambient water. I don't have any
large numbers and large means of small variances of P values to report to you.
I'm sure, as your question implied, you believe that there's more stress involved
in swimming in ice cold water. In fact, mice can swim for hours in ambient
water, whereas survival time is under a minute in ice water.
DR. CURLEY: Did you do any histopathology on the harderian glands, the
salivary glands, or did you see any swelling in the necks of these animals that
might be associated with a viral infection and not with treatment with the compound
in question? Or, I may phrase it another way. Viral infections are endemic In
some colonies of rats. Was this the problem with these animals that you treated
with, I guess it was methylmercury ? The question is precautionary.
DR. SPYKER: The question is well founded, I'm sure, and my answer will be
a hedge. We did not do the histopathology or the gross pathology that you ques-
tion. Our defense is the best available under the circumstances. That is that our
data .implied a higher incidence of infection in the treated group than in the control
group, which was housed, you know, within a few feet. The cultures of the eyes
were positive for pneumococcus predominantly. That does in no way preclude a
viral infection. But it was very tidy, or perhaps, better, reassuring that the
.defect we did discover was in the B-cell system, which we normally think of as
more responsive to bacterial, pneumococcal type infections, rather than
214
-------
viral. So that's the only response I have. That's a good recommendation.
take that seriously next year.
DR. BARON: I couldn't quite make out on the flow chart whether some of your
behavioral studies were going to encompass weanling animals, young animals. My
first impression was that your tests started at the age of 12 months or 24 months,
and seemed to proceed thereafter. Are we going to avoid behavioral testing .on
young animals, especially since we know that with some organbphosphorous com-
pounds, some of the maximal inhibition is usually noted within the first 6 to 8
weeks after the animals, weanling animals, are put on test? This would show the
maximal effect on these young animals, which then thereafter proceeds to equalize
itself and the animals lose their sensitivity to immunosuppression. Is your test
limited to the older animals after 12 and 24 months ?
DR. SPYKER: Our experimental design really involves four groups of animals.
One group will receive the complete behavioral battery shortly after weaning while
two other groups will not be tested until years two and three. The final group will
be tested repeatedly, beginning at weaning and again at intervals throughout the
iuespan.
215
------- |