v°/EPA
United States
Environmental Protection
Agency
Health Effects Research
Laboratory
Research Triangle Park NC 27711
Research and Development
EPA-600/S1-81-017 Apr. 1981
Project Summary
Mutagenesis Screening of
Pesticides Using Drosophila
Ruby Valencia
Drosophila melanogaster males
were exposed by feeding (plus contact
and possibly inhalation). The genetic
test found most sensitive and appro-
priate was the sex-linked recessive
lethal test. For this, males of the
Canton-S wild type stock were ex-
posed. They were mated individually
and brooded to sample the entire
range of germ cell developmental
stages. A very large number of tests
(over 7000) were accumulated for
each compound in two or more repli-
cate experiments. Concurrent negative
controls were done with each, and
positive controls were run occasionally.
Thirty pesticides and seven other
miscellaneous compounds were tested
and four reference mutagens were run
through the system, some of the latter
at a series of "doses" (exposure
concentrations). Table 1 lists them all,
with the results.
Of the 15 pesticides (listed first in
the table) which could be tested at
adequate concentrations, four (captan,
folpet, bromacil, and simazine) were
found to be weak mutagens. One
(cacodylic acid) was questionable but
called negative. The rest of the pes-
ticides were so toxic that only very low
concentrations (0.1-5 ppm) could be
used (usually for a reduced exposure
time), and those are not considered
adequately tested, in view of results
obtained with reference mutagens at
these concentrations.
Two of the miscellaneous compounds
(Tris and PtCU) were found to be
potent mutagens. The rest were nega-
tive.
This Project Summary was devel-
oped by EPA's Health Effects Research
Laboratory. Research Triangle Park,
NC, to announce key findings of the
research project which is fully docu-
mented in a separate report of the
same title (see Project Report ordering
information at back).
Introduction
The mutagenicity tests carried out in
this laboratory using Drosophila were
part of a larger program involving
several test systems in several labora-
tories. It was not known how Drosophila
systems would compare with others in
terms of sensitivity to detect mutagenic-
ity. Neither was it known which genetic
endpoints (in Drosophila) would be most
adequate in a screening test. It was
recognized that pesticides (especially
the insecticides) would pose special
problems for genetic tests with fruit
flies. The Drosophila screening system
itself was therefore undergoing defini-
tion and simplification during the course
of these studies.
The original plan was to test each
compound in a stepwise manner, as
follows:
1. Dominant lethal test (the fastest).
Stop if positive.
2. Chromosomal alteration test (a
one-generation test). Stop if posi-
tive.
3. Sex-linked recessive lethal test.
It became clear, however, that the
first two tests were not highly sensitive,
and were not easy to do, thus, the lethal
test had to be done in all cases. At the
same time, evidence was founding
-------
(especially Vogel and Sobels, 1976) that
the sex-linked recessive lethal test is by
far the most sensitive Drosophila test.
This type of mutation isfrequent, involv-
ing some 800 genes on the X chromo-
some, and varied in nature, including
both "point" mutations and chromo-
somal alterations. We feel that the sex-
linked recessive lethal test alone should
be an adequate probe for mutagenesis.
Many people have been hesitant to
accept results with an insect as relative
to man. It was thought that insect
metabolism was probably too different.
These objectives have been consider-
ably reduced by the discovery that
Drosophila has enzymes which carry
out metabolic activation similar to that
effected by mammalian enzyme extracts
(Vogel and Sobels, 1976).
The toxicity of many of the com-
pounds did cause serious problems, and
resulted in a "no test" conclusion for
some of the compounds. For those
Table 1. Summary of Compounds Tested
Concentration (ppm)
Pesticide in feeding solution
Mutagenesis
result
Bromacil
Captan
Folpet
Simazine
Cacodylic acid
Dicamba
DMSA
Methoxychlor
Monuron
MSMA
Quintozene (PCNB)
Trifuluralin
Siduron
Acephate
Carbofuran
Dimethoate
Methomyl
Aspon
Azinphos-methyl (Guthion)
Chlorpyrifos (DursbanJ
Demeton
Dinoseb
Disulfoton
Ethyl parathion
Fenthion
Malathion
Monocrotophos (Azodrin)
Phorate (Thimet)
Trichlorfon
Miscellaneous Compounds
Safrole
1 'Hydroxy Safrole
1 'Hydroxy Safrole
-2,3 epoxide
Direct Black 38
Tris (2,3-dibromopropyl
phosphate)
Benzidine dihydrochloride
PtC/4
Reference Mutagens
EMS (Ethyl methanesulfonate
El (ethylenimine)
TMP (trimethyl phosphate)
TMP (trimethyl phosphate)
DBE (1,2-dibromoethane)
2, 3, 5, 2000
2, 3. 2000
2, 3, 2000
3. 5, 2000
3. 500, WO
3, 2000
3, 1500
1000
4. 1000. 2000
2, 3, 4, 1500, 2000
3. 4, 8. 2000
2, 3, 1000
100
10
10
1. 5, (10)
4. (10)
5
0.25-1.0
0.1
1
0.5, 1.4
1
0.25. 0.5
0.1,0.25
0.25. 0.5
2. 3
0.5-3
1
100
1000
1000
2000
4, 10. 100, 1000
1000
100, 500
2, 4, 10, 100, 200, 400
2, 4, 10. 30, 50
1000
100, 300
5, 10, 50
+
+
-t-
-?
which could be adequately tested, the
Drosophila results compared quite
favorably with the other systems, pick-
ing up four weak mutagens and one
questionable mutagen. All of these
except one were picked by at least one
other system. Two were found positive
in all systems.
Exposure Methods
Initial trials were made of several
other exposure methods (aerosol spray,
contact, injection), but the method
chosen as most appropriate was feeding.
Since flies walk on their food, contact is
always involved; and when vaporization
occurs, the test substance is also inhaled.
The method thus seemed more complete
and exposure more certain, and it better
mimicked human exposure routes.
It was shown several years ago by
Lesid and Backer (1968) that feeding of
mutagens in glucose solution is effec-
tive. They fed flies in vials with a bit of
soaked tissue (such as Chem-Wipes).
We chose to feed in small disposable
petri plates with a disc of glass fiber
filter m the bottom. This permitted easy
observation and counting of dead flies
for toxicity information.
At first, flies were exposed (toxicity
permitting) for 48 hours. Later, to make
ingestion more certain, the time was
extended to 72 hours. Flies cannot
survive for more than 24-36 hours
without drinking, but in 72 hours, they
should be obliged to ingest more. Con-
tact and inhalation were also thus
extended.
Compounds were dissolved, when
possible, in 1 % glucose. During the first
year, DMSO was used for compounds
not water soluble. At that time, a deci-
sion was taken (after warnings voiced at
the Fredericksburg meeting) to avoid
DMSO if possible. As a result, many
compounds were used in suspension
rather than in solution.
The concentrations were chosen by
the following criteria. If toxic at 2000
ppm or below, a concentration giving as
near as possible to 50% mortality at 72
hours was chosen. For very toxic com-
pounds, the duration of exposure was
reduced as well as the concentration.
For non-toxic compounds, 2000 ppm
was chosen as a reasonable upper limit.
Sex-linked Recessive Lethal
Test
Canton-S wild type males were ex-
posed. "CS" stock was used because it
has had a low spontaneous frequency
-------
>ver many years of use. The frequency
iverages about 0.15%, with variations
arely exceeding 0.1% and 0.3%.
Treated males were mated to "FM6"
emales. The X-chromosome of this is
narked with yellow, white and Bar and
:arries a complex of stock inversions.
Heterozygous Fi females were mated
ndividually to FM6 males (brothers or
stock males) and the Fa of each was
observed for the presence or absence of
'+" males. Any culture having no or less
han 5% of the expected number of
males was considered a lethal case and
i/vas confirmed by repeating the test
with four individual heterozygous F2
emales mated to FM6 males.
Each treated or control male was
ransferred to new sets of FM6 females
at intervals of 4, 3, 3, and 4 days to
produce 4 broods of progeny. The differ-
ent broods sample germ cells in gradually
earlier stages of development at the
ime of exposure. Mature sperm are
sampled in Brood 1, spermatids in Brood
2, spermatocytes in Brood 3, and sperm-
atogonia in Brood 4. These are not
"clean," but rather "rough" samples,
due to the length of the mating times
and that fact that chemicals may remain
in the body beyond the external exposure
ime.
"Cluster" detection and
handling
Treated and control males were num-
aered, mated, and transferred individu-
ally and the F, daughters of each were
mated as a "family." This is done in
order to detect cases where more than
one lethal is produced by one male.
These cases will be referred to as
"multiples" (as opposed to "singles"
and "nulls"). A multiple can be due to
multiple individual mutations (MIM) or
o a single mutation event in a gonial
cell, which then reduplicates and pro-
duces two or more sperm cells carrying
he same lethal (a "cluster"). The
distinction (if possible) between these
wo circumstances is very important
when attempting to detect low-level
mutagenicity. With potent mutagens,
the contribution of clusters is insignifi-
cant, and mass matmgs of treated and
control males are the rule. When ex-
posure is simple (as with radiation), and
many flies can be exposed, the pro-
cedure is often to test only one daughter
of each treated or control fly, this
avoiding both MIMs and clusters. Ex-
posure is not simple with chemicals,
id other precautions must be taken.
The following procedure was applied
to this data regardless of the recognized
pitfalls. Whenever a multiple was found,
a statistical method (devised by Seymour
Abrahamson) was applied, which yields
the probability (binomial expectation) of
that number of mutations arising inde-
pendently in a single male given (a) the
number of males in that particular
treated or control group which produced
progeny, (b) the average number of FI
females tested per Pi male, and (c) the
mutation frequency.
If the number of lethals in the multiple
greatly exceeded probability, we con-
sidered it a cluster and did not count
these lethals.
In addition, however, it was often
necessary to make some subjective
judgments on particular compounds.
When the lethals in the replicate treated
groups occurred in multiples, while
those in the concurrent control group
did not, this might have indicated
"spotty" exposure.
Number of chromosomes tested
The number of tests needed per
compound of "compound equivalent,"
which would be any variant on a com-
pound, such as dose level, germ cell
stage, or exposure method depends
mostly upon (a) mutagenicity—the more
mutable, the fewer tests needed to
prove it—and (b) the increment in
frequency desired to detect.
When the control rate is 0.15%, and
when it is desired to detect (and prove
statistically) an equal induced frequency
(a doubling of background), then about
8000 treated and 8000 control chromo-
somes should be tested.
Dominant Lethals
Methodology was developed for dom-
inant lethal screening and tests were
performed on simazine and dicamba,
using very low concentrations. Simazine
gave possibly positive results. This test,
however, was abandoned since it was
time-consuming and would probably
rarely help avoid future testing.
Chromosome Loss, Replace-
ment and Non-disjunction
The test is relatively fast, requiring
only one generation, and appears to be
simple. In use, however, very large
numbers are required and interpretation
of results are difficult, due to the several
different types of variants and their
quite different meanings in terms of
cytogenetic events. Since the recessive
lethal test also picks up chromosomal
rearrangements, it was decided that
this separate test was not worth the
time and effort required.
Results and Conclusions
The pesticide results fall into three
categories:
(a) Those which were tested at con-
centrations above 10 ppm for at least 24
hours and which yielded results indicat-
ing a weak mutagenic effect. These
were captan, folpet, bromacil, and
simazine, plus possibly, cacodylic acid.
(b) Those which were presumably
tested adequately (as above) but which
yielded negative results. These were
dicamba through carbofuran in Table 1.
(c) Those which may not have been
adequately tested, since they were
highly toxic, were used at very low
concentrations, and gave negative
results.
For the four compounds called positive,
the conclusion was based upon the
experiments with 2000 ppm. Captan,
simazine, and bromacil were tested
simultaneously in Run No. 37. Folpet
was tested in Run No. 38. Both runs had
controls with quite low frequencies, but
this appeared to be a true low period for
the stock. Run 37 was done with a
newly prepared stock and the control
was the first of a series giving low
values. In all four cases, there were
several multiples in the treated series,
but none in the controls. These multiples
were:
Captan—5 males with 2,2 males with
3, 1 male with 6
Folpet—4 males with 2,1 male with 3
Bromacil—3 males with 2, 1 male
with 3, 1 male with 4, 1 male with 5
Simazine—4 males with 2, 2 males
with 3.
When the Kastenbaum and Bowman
statistical test (Mut. Res. 9 [1970] 527-
549) is applied to the data for these
compounds (using the sum of Controls
37 and 38 to provide approximately
equivalent numbers of treated and
control tests), the result is as follows.
Captan is significant at the .01 level if
the multiple of 6 is included. Without
the multiple, it is significant at the .05
level.
Bromacil is significant at the .01 level
if the multiples (one of 4 and one of 5)
are included. It barely misses signifi-
cance at .05 without them.
Simazine had no multiples greater
than 3 and is significant at the .01 level.
i US GOVERNMENT PRINTING OFFICE 1981-757-012/7031
-------
Folpet also had no correction factor,
but is significant only at the .05 level.
Using the Chi2 test, which is somewhat
less conservative than the Kastenbaum-
Bowmantest, P=<.02. More importantly
the folpet data shows a peak of mutage-
nicity in Broods 2 and 3. Using these
broods, only, the result is highly signifi-
cant (Chi2 = 15, P = <.01).
In all these cases, the deduction of
multiples of 4, 5, and 6 is questionable,
since there are also several multiples of
2 and 3, while the controls have almost
none.
It is possible that exposure may have
been "spotty"—i.e., some males may
have ingested, contacted, or inhaled
more than others, in which case the
compounds may be more mutagenic
than the averaged results indicate.
Cacodylic acid caused marked sterility
of treated males, especially in Brood 3,
indicating a cytotoxic effect on peri-
meiotic cells. Unfortunately, it is not
possible to know from these experi-
ments whether or not the damage is
genetic. The recessive lethal frequency
obtained in the one adequate experi-
ment (Run 52) was actually 3 times the
concurrent control value. The latter,
however, was exceptionally low, and in
this case it was not part of consistently
low control period. (Controls were, in
fact, somewhat erratic during thattime.)
There were no multiples in Control 52
and only one multiple of 3 in the treated
males. The result was, thus, considered
negative but questionable.
In the tests of miscellaneous com-
pounds, Tris and PtCU were clearly
mutagenic, with peak effects in Brood 2,
indicating an effect primarily on sper-
matids. All others were clearly negative.
The reference mutagens EMS, El, and
TMP were positive when tested at
adequate concentrations. EMS, a very
potent mutagen, surprisingly gave fre-
quencies very near control level wh«
tested at very low concentrations (2 ar
4 ppm). TMP was negative at 100 pp
and 300 ppm. DBE was tested only ,
low concentrations and was negative <
questionable. It is probable that a
exposure technique was inappropria
for this volatile compound, In all thes
cases, however, there were multiple
indicating possible "spotty" exposure
Ruby Valencia is with the WARF Institute, Inc., Madison, I/I// 53706.
Michael D. Waters is the EPA Project Officer (see below).
The complete report, entitled "Mutagenesis Screening of Pesticides Using
Drosophila," (Order No. PB 81-160 848; Cost: $9.50, subject to change) will
be available only from:
National Technical Information Service
5285 Port Royal Road
Springfield, VA 22161
Telephone: 703-487-4650
The EPA Project Officer can be contacted at:
Health Effects Research Laboratory
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
Postage and
Fees Paid
Environmental
Protection
Agency
EPA 335
Official Business
Penalty for Private Use $300
000032*
------- |