United States
Environmental Protection
Agency
Environmental
Research Laboratory
Gulf Breeze. FL 32561
Research and Development
EPA/600/S3-87/032 Dec. 1987 ' >
Project Summary
Microinjection of Fish
Embryos as a Laboratory
Assay for Chemical Carcinogens
John M. Grizzle and Marshall R. Putnam
During this project, techniques
were developed for microinjection of
chemicals into fish eggs, lesions
were described in fish reared from
injected eggs, and eggs of various
species were compared to determine
which were best suited for use in
carcinogenicity assays. Eggs of the
following species were injected: gulf
killifish, sheepshead minnow, rivulus,
inland silverside, gulf toadfish, and
channel catfish. Chemicals injected
into eggs were diethylnitrosamine,
N-methyl-N'-nitro-N-nitroso-
guanidine (MNNG), aflatoxin B1, and
trichloroethylene.
Quantification of carcinogen
dose in the egg immediately after
Injection indicated that variation of
the dose retained was a major
problem. During the project,
improvements in procedures re-
suited in increased mean percentage
of the dose remaining in the egg, but
variation between eggs remained
high.
The incidence of all lesions, and
especially neoplasms, was low,
probably because of the low doses
retained in the eggs. The most
important lesion found during this
project was a pancreatic acinar cell
carcinoma in a gulf killifish injected
with MNNG.
Gulf killifish, sheepshead min-
now, and rivulus eggs have potential
for use in carcinogenicity assays.
However, improvements in
microinjection techniques are
needed before an assay system
involving injection of test chemicals
into embryos of these species can be
used reliably. Our results suggest
that channel catfish are less
susceptible than other species to the
pathologic effects of the chemicals
that we tested.
This Project Summary was
developed by EPA's Environmental
Research Laboratory, Gulf Breeze, FL,
to announce key findings of the
research project that is fully doc-
umented in a separate report of the
same title (see Project Report
ordering information at back).
Introduction
Several species of fish are
potentially valuable as laboratory animals
in oncology Numerous studies have
indicated that fish develop neoplasms
after exposure to a variety of known
carcinogens. For some fish models, the
neoplastic responses of fish to chemical
carcinogens resemble those of rodents.
Two major advantages of fish as assay
animals are evident from previous
studies. The length of time required for
neoplasms to develop in fish is often less
than 6 months and has been reported as
5 to 8 weeks in some experiments. In
addition, spontaneous neoplasms are
rarely found in non-exposed fish used in
laboratory assays. These features of fish
assay systems indicate that they could
be useful as alternative or com-
plementary tests for carcinogens.
Rainbow trout, Salmo gairdneri,
embryos are highly sensitive to
chemically induced neoplasia, embryo
exposure of Shasta strain rainbow trout
appears to be the most sensitive model
-------
available for hepatocarcinogenicity of
aflatoxin B1 (AFB). Initially, embryo
exposures involved dipping whole eggs
into the test solution, a technique that
worked well with slightly water-soluble
carcinogens such as AFB. However, this
model is not suitable for exposing the
embryos to highly water-insoluble
chemicals.
Microinjection of carcinogens inside
fish eggs ensures exposure of the
embryo to the test chemical. This
technique has been used to induce
neoplasms in rainbow trout and coho
salmon, Oncorhynchus kisutch, exposed
to a variety of carcinogens. Micro-
injection using acetone or dimethyl-
sulfoxide (DMSO) as carriers for test
chemicals allows embryo exposure to
highly water-insoluble compounds such
as benzo[a]pyrene and 7,12-
dimethylbenz(a)anthracene. An additional
advantage of the microinjection method
is the diminutive amount of test chemical
needed.
Although fish embryo exposures
appear to be a promising method to test
new chemicals for carcinogenicity, the
use of salmonids presents limitations to
this model. These limitations include
exacting rearing conditions needed for
these species, long generation time, and
difficulty in obtaining eggs outside of the
short, natural spawning season. The
relative merits of non-salmomd species
for use in embryo exposure models have
not been adequately tested. There are
many fish species that would eliminate
all of the limitations listed above for the
salmonid species.
The objectives of this project were
(1) to develop techniques needed for
microinjection of chemicals into eggs of
several fish species, (2) to compare eggs
of several fish species to determine
which are best suited for use in
carcinogenicity assays, and (3) to
determine the types of lesions resulting
from exposure of eggs to known
carcinogens. Ultimately this could result
in a method for rapid, reproducible, and
sensitive screening of chemicals relevant
to humans.
Six species of fish were selected for
this project: gulf killifish (Fundulus
grandis), sheepshead minnow
(Cyprinodon variegatus), rivulus (Rivulus
ocellatus), inland silverside (Menidia
beryllina), gulf toadfish (Opsanus beta),
and channel catfish (Ictalurus punctatus).
Chemicals injected into eggs were AFB,
diethylnitrosamine (DEN), N-methyl-
N'-nitro-N-nitrosoguanidine (MNNG),
and trichloroethylene (TCE).
Results
Hatching percentages of some
groups of injected channel catfish and
gulf killifish eggs were nearly as high as
for unmjected eggs. The highest survival
of sheepshead minnow eggs was less
than one-half of the percentage for
uninjected eggs. Eggs that were not
disinfected with antibiotics had a lower
hatching percentage that disinfected
eggs. There was only a small difference
between groups of sheepshead minnow
eggs injected in the perivitelline space or
into the yolk sac. Only two groups of
inland silversides eggs were injected
because hatching percentage of injected
eggs was low [9% for DMSO injected
(control) and 8% for DEN injected].
Hatching percentages of many groups of
eggs injected with test chemicals were
as high or higher than eggs injected with
DMSO or saline. Except for inland
silversides, the lowest hatching
percentages were above 12%.
Long-term survival of test fish was
highly variable among groups of fish, but
survival did not seem related to
treatment. Many of the test fish and
controls died of infectious diseases.
Enteric septicemia of catfish caused by
the bacterium Edwardsiella ictaluri killed
most of the fish in some of the channel
catfish treatments, including the DMSO-
injected controls. Posthatching survival of
injected toadfish was low, apparently
because of fungal infections. These fish
could have been predisposed to this
disease because of injury caused by the
injection. An intestinal protozoan was
noted in sections of sheepshead
minnows.
Quantification of carcinogen dose
retained in the egg immediately after
injection indicated that variation in the
percentage of the dose retained was a
problem in our methodology. Rivulus
eggs injected in the perivitelline space
retained less than 7% of the nominal
dose and those injected in the yolk sac
retained 6 to 27%. One attempt to inject
0.25 ill instead of the usual 0.5 nl dose
did not improve the results. Injection of
sheepshead minnow eggs was the most
successful, 35 to 47% of the dose was
retained. Standard deviations were
approximately equal to the mean for
most injections of labeled compounds.
The syringe system was not the source
of error because the mean dose
delivered was 106% of the nominal dose
with a standard deviation of 6%.
The incidence of all lesions, and
especially neoplasms, was low, probably
because of the low doses retained in the
embryos. Two neoplastic lesions and
another lesion that further study may
determine to be neoplastic were found in
carcinogen-exposed fish. Other lesions
were preneoplastic, teratogenic, toxic
reactions, or pathogen related. Lesions
caused by pathogens were not
considered in this report, although the
occurrence of infectious diseases could
have been influenced by the chemical
exposures.
The most important lesion found
during this study was a pancreatic acinar
cell carcinoma in a gulf killifish
necropsied 139 days after an injection of
2 ng MNNG. This neoplasm had invaded
the posterior liver. The tumor cells
resembled pancreatic acinar cells by
having basophilic cytoplasm except for
the strongly eosmophilic zymogen
granules. The cells tended to form acini
and had a high mitotic index. This is only
the third report of an experimentally
induced neoplasm of the exocrine
pancreas in fish.
Sheepshead minnows injected with
DEN or AFB tended to have increased
numbers of rodlet cells in the mesentery
or mesenteries. Nuclear pleomorphism
was observed in livers of AFB injected ,
sheepshead minnows. Lesions in rivulus'
injected with DEN included hepatic
adenoma, megalocytic hepatosis, nuclear
pleomorphism, thyroid hyperplasia, and a
convoluted retina embedded in the brain.
Nuclear pleomorphism was also
observed in the liver of rivulus injected
with AFB. A mass that has not been
identified yet was found in the posterior
body cavity of a rivulus injected with
MNNG. No lesions related to carcinogen
exposure were observed in the 72
exposed channel catfish examined. Only
one gulf toadfish exposed to a-test
chemical survived until necropsy; this fish
had no lesions.
Conclusions and
Recommendations
Survival of channel catfish and gulf
toadfish embryos injected with
stainless-steel needles (31 gauge) was
satisfactory as was the survival of gulf
killifish, sheepshead minnow, and rivulus
injected with sharpened glass needles
(48 to 112 urn outside diameter).
Mortality of injected inland silverside
embryos was too high for this species to
be useful for this technique. Survival of
sheepshead minnows was similar for
injection of test chemicals into the yolkd
sac or into the perivitelline space. A 5-
minute rinse in a combination of
penicillin, streptomycin, and fungizone
-------
reduced the number of embryos killed by
bacterial and fungal infections. Although
these antimicrobial drugs reduced
mortality of injected embryos, future
research should consider their possible
role as promotors or inhibitors of
neoplastic responses.
Quantification of carcinogen dose
retained in the egg immediately after
injection indicated that variation in the
percentage of the dose retained was a
problem. During this project, im-
provements in procedures resulted in
increased mean percentage of the dose
remaining in the egg, but variation
between eggs remained high. Additional
research is needed to complete
development of techniques for accurate
delivery and adequate retention of
carcinogens in small eggs of fish.
The incidence of lesions, and
especially neoplasms, was probably low
because of the low doses retained in the
embryos The variation in retention of
carcinogen could have resulted in
survival of fish that retained little of the
dose while other fish, injected with the
same nominal dose, died because they
retained a higher percentage. This would
have biased the lesion-incidence data
because it was not possible to determine
which fish retained high doses.
The use of fish embryo exposures
seems promising as a method to induce
pancreatic tumors that could be used to
resolve questions about histogenesis of
exocrme pancreatic carcinomas There
are only two previous reports of chemical
carcinogens inducing pancreatic
neoplasms in fish and both of these
previous studies involved exposure of
young fish Our results support the
hypothesis that the types of neoplasms
developing in carcinogen-exposed fish
depend on the age of the fish when
exposed.
Gulf killifish, sheepshead minnow,
and rivulus eggs have potential for use in
carcinogenicity assays. Eggs of these
species can be easily obtained at any
time of the year, and these species
develop neoplasms after exposure to
carcinogens. However, improvements in
micromjection techniques are needed
before an assay system involving
injection of test chemicals into embryos
of these species can be used reliably.
Gulf toadfish and channel catfish have
relatively large eggs, an advantage for a
micromjection assay, but eggs of these
species are available only during a
relatively short spawning season. In
addition, our results suggest that channel
catfish are less susceptible than other
species to the pathologic effects of the
chemicals that we tested. The biologic
basis for the low susceptibility of channel
catfish to carcinogens should be
investigated.
With some further development,
micromjection of fish eggs appears to be
a promising method for testing chemical
carcinogens. The problem of inconsistent
delivery of test chemicals to fish eggs
must be solved before embryos of
sheepshead minnow, gulf killifish, and
rivulus can be used in a micromjection
test procedure. Reasonable approaches
for resolving this experimental problem
include injection of smaller volumes of
the test chemical, use of smaller
diameter needles, a partial enzymatic
digestion of the chorion of the egg, or
use of other fish species such as brown
bullhead (Ictalurus nebulosus) and black
bullhead (Ictalurus me/as).
John M. Grizzle and Marshall R. Putnam are with Auburn University, Auburn,
AL 36849.
John A. Couch is the EPA Project Officer (see below).
The complete report, entitled "Micromjection of Fish Embryos as a Laboratory
Assay for Chemical Carcinogens," (Order No. PB 88-124 623/AS; Cost:
$12.95; subject to change) will be available only from:
National Technical Information Service
5285 Port Ftoya! Road
Springfield, VA 22161
Telephone: 703-487-4650
The EPA Project Officer can be contacted at:
Environmental Research Laboratory
U.S. Environmental Protection Agency
Sabine Island
Gulf Breeze, FL 32561
-------
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
Official Business
Penalty for Private Use $300
EPA/600/S3-87/032
0000329 PS
U S £ĞMR PROTECTION
REGION 5 LIBRflRT
230 S DEARBORN STREET
CHICAGO IL 60604
------- |